Berkeleyan and Kantian Idealism: How Do They Differ?

The good bishop, as Kant called him, held that reality is exhausted by "spirits" and their ideas. Thus on Berkeley's scheme everything is either a spiritual substance or mind, whether finite or infinite (God), or else an idea 'in'  a  mind. Ideas are thus modes or  modifications of minds.  As such they do not exist independently of minds. That's what 'in' conveys. If everything is either a mind or an idea in a mind, then bodies are not substances given that a substance is an entity capable of independent existence.  Berkeley's ontology is thus a one (type of) substance ontology. This makes for a contrast with Descartes' dualism of substances, thinking and extended. 

Now the gross facts are not in dispute and no (sane) philosopher is in the business of denying them. So every sane person will agree that there are rocks and trees, tables and turnips. You haven't understood Berkeley if you think that he is an eliminativist about such things. That is why you cannot refute him by kicking a stone.  Anyone who thinks that he can be so refuted is utterly bereft of philosophical aptitude. The question is not whether there are bodies, trees and such; the question is what they are, and what the good bishop is telling us is that they are coherent, cohesive, bundles of ideas. Trees and such exist alright; it's just that their esse est percipi, their being/existence is (identically) their being perceived by some spirit.  

The standard picture assimilates Kant to Berkeley, as I wrote earlier:

P.F. Strawson and H. A. Prichard are exponents of this reading along with many others in the Anglosphere. The standard picture makes of Kant an inconsistent Berkeley who limits knowledge to appearances, these being understood as "mere representations" (blosse Vorstellungen), while at the same time positing an unknowable  realm of things in themselves.  Mere representations are assimilated to Berkeleian ideas so that when Kant states that we know only appearances, what he is telling us is that we know only the contents of our minds.

The standard picture shows a failure to grasp what Kant intends with his transcendental idealism. (Note, however, that whether Kant achieves what he intends is an entirely different question.)  When I taught Kant in the 1980s I used the following three-level schema in order to clarify what Kant means by 'appearance' (Erscheinung) when he is using it in his special transcendentally idealist sense.  There are at least three senses of 'appearance' in Kant. We may call them the manifest, the scientific, and the transcendental. The empirical embraces both the manifest and the scientific and stands opposed to the transcendental. Correspondingly, there are three senses of 'reality,' the manifest, the scientific, and the transcendental.   

Level One: We start with the ordinary 'manifest image' appearance-reality distinction. One day I was hiking Jacob's Crosscut along the base of Superstition Mountain. Off in the brush I espied what appeared to be some big black dogs. In reality, however, they were black bears as a closer look revealed.  This is a familiar sort of case. An initial appearance is shown to be a perceptual mistake, one correctable and in this case corrected by further perception.  The initial, non-veridical appearance was not nothing, but its 'reality' was merely intra-mental, a momentary private datum not amenable to public verification, or even ongoing private verification.  It was a mere seeming or semblance, an instance of what Kant calls Schein and distinguishes from Erscheinung.  Kantian appearances are not private mental data. 

Let 'A1' denote an appearance at Level One, and 'R1' a reality or real thing at Level One. An A1 may or may not be veridical. If I jump back from what I take to be a snake but is in reality a tree root, then the A1 is non-veridical. But when I see a tree root and my partner confirms that what I saw was a tree root, then my A1 and his numerically different A1 are veridical.  So an A1 need not be illusory.  Every A1 purports to be of or about an R1, but the purport does not always 'pan out.'

At A 45 = B63, Kant gives his rainbow example. He tells us that a rainbow may be called a mere appearance and the rain the thing in itself.  This is an example of the Level One appearance-reality distinction. In that same obscure passage,  the careful reader can discern the Level  Three appearance-reality distinction.  For he tells us that the rain drops, together with such primary qualities as shape,  are themselves appearances of a "transcendental object" that "remains unknown to us."  It follows that the rainbow is an appearance of an appearance. The empirical object (rain water refracting sunlight) that is the reality behind the rainbow is itself an appearance of something that does not appear to us as it is in itself.

Level Two.  We now wheel the primary versus secondary quality distinction onto the field. An R1 at Level One has both primary and secondary qualities.  The tree I see when I look out my window has both primary and secondary qualities. To mention just two of its primary qualities, it has a size and a shape. To mention just one of its secondary qualities it is green in color.  At Level Two, R1 is stripped of its secondary qualities, and left with its primary qualities alone. We are now operating within the 'scientific image.'  What was R1 at Level One is now A2 at Level Two.  The real extra-mental tree of Level One is now taken to be an appearance of a deeper reality R2 at Level Two.  Thus:

A1 -------------------> R1 

                                       (R1 = A2) --------------------> R2

                                              

A1 is a representation 'in' the mind of a psychophysical being, a human animal for example. The arrows stand for the representing relation. There is difference between the two relations depicted, but I cannot go into this now. What A1 represents (or presents, stellt vor) is an empirical object R1 endowed with primary and secondary qualities. The secondary qualities are perceived at the object even though, at Level Two, they are understood to be merely relational properties of R2 due to the affection (causal impact) of the thing R2 upon the sensory receptors of the psychophysical subject.  Thus R2 in itself is not colored, etc.  But R2 is in space and possesses a location, a size, a shape, a volume, etc. It is either at rest or in motion which implies the possibility of translation and rotation, etc. which motions bring  objective time into the picture.  

Level Three.  At this level we arrive at the phenomenon or appearance in the specifically Kantian sense. Space and time (and thus all primary qualities) are now stripped from R2 and made out to be a priori forms (or schematizations of such forms), forms that characterize the standpoint of an ectypal intellect, one whose sole mode of intuition (Anschauung) is sensible and thus receptive unlike the intellectual and thus non-sensible mode of intuition of the archetypal intellect whose intuition is creative of its objects.  What exactly this standpoint of the ectypal is is a vexing question. We can say this much with assurance: it is nothing internal to the mind of a psychophysical being such as a human animal, nor is it necessarily dependent on the existence of psychophysical beings.  Extending the above diagram:

 

(R1 = A2) --------------------> R2

                                                    (R2 = A3) -----------------------> R3 (negative noumenon)

(R2 = A3) is an intersubjective object.  It is the objective correlate of the epistemic standpoint of an ectypal intellect.  Nature for Kant is the sum-total of all such phenomena as intersubjective objects. The objectivity of R3, by contrast, is not intersubjective but absolute as befits the objective correlate of the absolute mind of the archetypal intellect, "which all men call God," to adapt a phrase from Aquinas.

The above schema leaves us with a lot of thorny questions.  One such concerns double affection (Erich Adickes). Do both R3 and R2 cause sensations in psychophysical beings?

The main point, however, it is that no one who understands what Kant is trying to do could possibly assimilate his idealism to Berkeley's. There is much more to be said.

Comments

[oz the clever ostrich]

>Off in the brush I espied what appeared to be some big black dogs.

Which were black bears that you thought to be big dogs. This I understand.

>An initial appearance is shown to be a perceptual mistake

What does the noun phrase ‘An initial appearance’ refer to? To the bears? But then you say that this appearance, i.e. the bears, was a perceptual mistake. Or are you referring to something else?

> The initial, non-veridical appearance was not nothing

Clearly the bears were something.

> but its 'reality' was merely intra-mental

Surely the bears were extra mental?

I do not follow any of this. The things that appeared to be black dogs were in fact bears. The appearances WERE the bears.

I have been re-reading Prichard (his collection of essays on perception, and his commentary on Kant, which is excellent). He pays careful attention to the grammar of ‘appearing’ and its cognates.

[oz the clever ostrich]

>At Level Two, R1 is stripped of its secondary qualities, and left with its primary qualities alone.

But the tree is green. How can you possibly strip the greenness out of the tree?

Or are you saying that the colour is a sensation? But then you face the difficult problem I identified earlier , which you did not address. For colour is extended in space, and has a form, therefore if colour is a sensation, sensation is also extended in space, and also has a form.

Then, turning to your Level 3, how can the space that the colour-sensations occupy be 'inter subjective'? For the colour patch with its spatial form that I see, cannot be the same in number as the colour patch that F sees. Each of us must have our own separate space, spaces which can never be spatially connected, for no point in one can ever be contiguous with any point in the other.

[oz the clever ostrich]

Prichard: “After all there can be only one space, and if colours are in space they are in this one space which is at the same time the space in which bodies are, if there are bodies.”

[oz the puzzled ostrich]

“Off in the brush I espied what appeared to be some big black dogs”

In fact you espied some black bears, which appeared at first to be dogs. So you took some bears to be some dogs. Does that mean we have a three place predicate: took(you, bears, dogs)? No, because that implies the existence of some dogs, as well as some bears. But there were no dogs, only you and the bears. It’s not that you thought of some dogs that they were bears, rather, you thought of some bears, that they were dogs. My puzzle is then how you get to the object you call ‘A1’, the appearance. Are you saying that you thought, of some appearance , that it was a group of dogs? But what then of the veridical case, where you got closer and you now correctly took the bears to be bears? Were you still thinking of some appearance, which you now took to be bears? Or does the appearance vanish in the veridical case, i.e. there are only appearances in the illusory case?

It would help me a lot if you could answer these questions, because if you don’t, I have no way of understanding what you are referring to by ‘A1’, ‘R1’, ‘R2’ etc.