On Kant’s Refutation of Idealism and Descartes
Michael Vossen

I will discuss the extent to which Kant’s refutation of idealism undermines Descartes’ position in this paper. Further, I will ignore the question of what it means to be a transcendental idealist, and the question of how Kant’s refutation undermines Berkley’s position. I will argue that while Kant has given an argument that provides sufficient proof to believe in the existence of objects in space outside of us, his argument appears too strong when combined with one of Descartes’ arguments for the doubtfulness of objects of outer experience. I will then suggest that Kant appears committed to the denial of epistemic closure given his other commitments, and that the reasons for which he denies epistemic closure ought to be made explicit in his refutation.

What is idealism? Kant writes, “Idealism - by which I mean material idealism - is a theory about the existence of objects in space outside us.” Kant characterizes two forms of idealism, one form associated with Descartes’ view, and one form associated with Berkley’s view. Kant calls Descartes’ idealism “unproblematic idealism”, and writes that unproblematic idealism “ … holds that the existence of objects in space outside us is doubtful and indemonstrable (as with Descartes’ view that only one empirical assertion is indubitably certain, namely ‘I am’)”.
Unproblematic idealism is weaker than “dogmatic idealism”, which holds that “…space, along with all the things that couldn’t exist without space, is in itself impossible (as with Berkeley’s view that the things in space are merely imaginary entities).” for “It doesn’t assert that space and its contents are unreal; it merely says that through our immediate experience we can’t prove the existence of anything except ourselves.”

Kant seems somewhat sympathetic to unproblematic idealism, writing that unproblematic idealism is based on a “…sound principle of philosophizing, namely: don’t make your mind up about something for which you don’t have sufficient proof.” Kant’s strategy in refuting the unproblematic idealist is to show that there is sufficient proof to believe in the existence of objects in space outside of us. Kant writes:

“But we can give ‘sufficient proof’ of the reality of space and things
in it·. The proof the idealist demands comes from showing that we have
experience of outer things, rather than merely imagining them; and the
only way to prove this is to show that even our inner experience is possible
only on the assumption of outer experience. ·That should be enough to
refute Descartes, who regards inner experience as indubitable.”

Given this outline, Kant’s refutation of unproblematic idealism assumedly is of the following form:

1. Inner experience is possible only on the (To be proved)
assumption of outer experience.
2. Inner experience is possible. Assumption
\ 3. We have outer experience. 1, 2 Modus Ponens

Here, premise two is either an assumption, or follow from the claim that we have inner experience, and the principle that P implies possibly P. Given that Kant claims, “My consciousness of my own existence and of details about myself proves the existence of objects in space outside me.” I take Kant to be committed the claim that we have inner experience, and to some claim about how actually having inner experience implies inner experience is possible. Further, I take “consciousness of my own existence” to be a form of inner experience.

Kant’s argument for premise one of the previous forms the bulk of his refutation of idealism. Kant argues for premise one as follows. Kant takes it as an assumption that “I am conscious of my own existence as determined in time, ·i.e. I am conscious of myself as being in various states at various times.” Further, from his transcendental aesthetic presumably, Kant has shown that “All knowledge of temporal details presupposes knowledge of something persistent in perception.” From these claims, I assume Kant gets the claim that there is something persistent in perception. For if all knowledge of temporal details presupposes knowledge of something persistent in perception, then if I am conscious of myself as being in various states at various times, then there is something persistent in perception. Since Kant holds that “I am conscious of my own existence as determined in time”, he can conclude that there is something persistent in perception by modus ponens. So there is some persistent thing.

The next part of Kant’s argument appears to have the form of “or elimination”, as Kant denies that the persistent thing is an intuition or a representation, and then goes on to conclude that the persistent thing can only be a thing outside me. Kant writes

But this persistent thing can’t be an intuition in me. For the only
grounds there are in me for any account of my various states are
representations; and as representations they themselves require
a persistent thing distinct from them, in relation to which their
change, and so my existence through the time in which they change,
can be determined.

This is presumably based on Kant’s model of conceptualization, which he has already proved presumably. Given that Kant wants to conclude from the previous claim that “Thus perception of this persistent thing is possible only through a thing outside me and not through the mere representation of a thing outside me” I assume Kant has the hidden premise “The persistent thing is either an intuition in me or a thing outside me or a representation of a thing outside me” as well. So Kant is justified in drawing the conditional conclusion that “…. consequently my sense of the details of my existence in time is possible only through the existence of actual things that I perceive outside me.”, which is just premise one of the sketched argument.

I would like to grant Kant his argument at this point. Kant does reach a conclusion that is incompatible with Descartes’ doubts about the existence of objects of outer experience. What I am tempted to deny is that Kant has thus refuted Descartes. Descartes’ doubts about the existence of objects of outer experience appear as the conclusion of an assumingly valid argument of his. Further, while Kant’s argument contradicts Descartes’ conclusion, I’m not sure that it thereby refutes Descartes’ argument. For if Descartes’ argument leads to a contradiction, then if it is valid, at least one premise is false. Further, Kant does not seem to explain which premise is false. Since both premises of Descartes’ skeptical argument may appear true for good reason, Kant ought to give an explenation for why his theory shows one of these reasons to be bad. So, while Kant has given an argument that provides sufficient proof to believe in the existence of objects in space outside of us, his argument appears too strong when combined with one of Descartes’ arguments for the doubtfulness of objects of outer experience. I will now discuss and motivate Descartes’ argument.

In his Meditations, Descartes reflects that his dream-experiences are sometimes identical in experience to his waking-experiences. This means, for some experiences “...there are never any sure signs by means of which being awake can be distinguished from being asleep.” From these considerations, Descartes would like to conclude “...that physics, astronomy, medicine and all other disciplines which depend on the study of composite things, are doubtful...” A plausible reconstruction of Descartes’ argument towards this conclusion is:

1. I don’t know that I’m not dreaming. Assumption
2. If I don’t know that I’m not dreaming, Assumption
then I don’t know that I have some
outer experience.
\3. I don’t know that I am having some 1, 2 Modus Ponens
outer experience.

Premise one comes from Descartes’ reflection that his dream-experiences are sometimes identical in experience to his waking-experiences. Descartes really only needs us to grant that some non-outer experiences can be identical in experience to outer experience, so evil demons, mad scientists, sickness, or anything else that might produce experiences that are of the same quality as outer experiences will work in place of dreams.

Premise two is based on the principle of epistemic closure, which holds that if a person knows p, and p entails q, then that person knows q. In Descartes’ case, if a person knows that what he sees is a house, and being a house perception is to not be a house delusion, then if a person knows that what he sees is a house, he knows that what he sees is not a house delusion.
Finally, the form of Descartes’ argument is modus ponens. Unless Kant denies this form of inference, Descartes’ argument is valid. Kant uses modus ponens in his own arguments, and thinks logic is basically complete, so it is unlikely that Kant denies this inference.

Kant’s refutation seems to attack the conclusion of Descartes’ argument, and thus generates a contradiction when the arguments are combined. Consider the following outline of how Kant and Descartes arguments combine:

1. I don’t know that I’m not dreaming. Assumption
2. If I don’t know that I’m not dreaming, Assumption
then I don’t know that I have some
outer experience.
3. I don’t know that I have some 1, 2 Modus Ponens
outer experience .
4. Inner experience is possible only on the Refutation
assumption of outer experience
5. I have inner experience. Assumption
6. If I have inner experience, then inner Principle: P’‡P
experience is possible
7. Inner experience is possible. 5, 6 Modus Ponens
8. I have outer experience. 4, 7 Modus Ponens
9. I know that I have some outer experience 8 or Assumption
10. It’s not the case that I don’t know that 9, 2, Modus Tollens
I’m not dreaming.
11. ^ 1, 10 Absurd Introduction

From this contradiction, Kant must deny some assumption, or hold that some inference is invalid. I will assume that Kant will not deny his own argument, so really only premise one, premise two, and the inference from lines eight to nine are up for denial.

Kant would probably not deny the move from lines eight to nine, for his refutation is meant to prove that we have outer experience. Since Kant takes Descartes to doubt that we have outer experience, and refutes him by showing that we have sufficient proof for outer experience, it would be odd and somewhat unexplained to then deny that we do not know that we have outer experience. For then we have outer experience, and we can prove that we have outer experience, yet we do not know that we have some outer experience. An explanation of why proving something is insufficient for knowledge seems needed if Kant takes this route. Further, even if Kant could give an explanation given his epistemic claims, his “proof” of outer objects would then still seem doubtful if it did not provide knowledge, as Descartes doubts that we can know that outer objects exist, not that we can give a non-knowledge providing proof. So Kant probably would not deny the move from lines eight to nine.

Kant does not appear to deny premise one, and even appears to recognize something like this argument when he writes:

“From the fact that the existence of outer things is required for the
possibility of a determinate consciousness of myself, it doesn’t follow
that every intuitive representation of outer things involves their really
existing; for a representation of them may well be the product merely
of the imagination (as in dreams and delusions).”

So Kant does not seem to deny the possibility of being deceived by a dream, which seems to entail that he would not reject premise one of the argument. Kant goes on to write:

“But this doesn’t weaken the thesis I have been defending against
problematic idealism, because such a representation merely reproduces
previous outer perceptions, which I have shown to be possible only
through the actuality of outer objects. That’s all I need. All I have been
trying to prove is that inner experience in general is possible [279] only
through outer experience in general. To show that any individual experience
is veridical rather than imaginary, one has to look into the details of the
case to see how well they fit with the criteria for all real experience.

How does one look into the details of the case to see how well they fit with the criteria for all real experience this? Kant writes:

“To decide which of my given intuitions correspond to actual objects
outside me - i.e. which of them belong to outer sense and not to
imagination - I must go by the rules according to which any experience
(even inner experience) is distinguished from imagination – always
presupposing that there is such a thing as outer experience.”

Since Kant seems committed to premise one and the inference from lines eight to nine. appears committed to the denial of epistemic closure given his other commitments, and that the reasons for which he denies epistemic closure ought to be made explicit in his refutation.



Works Cited.

Descartes, René. Meditations on First Philosophy. Translated and Edited by John Cottingham. Cambridge University Press, 1996.

Kant, Immanuel. Critique of Pure Reason. Translated by Jonathan Bennett, http://www.earlymoderntexts.com/pdfbits/kcpr.html 2007.

Kant. Immanuel. Critique of Pure Reason. Translated by Norman Kemp Smith, St Martin’s press, 1956.