Kant’s Critique of Descartes (and time, immediacy, and the external world)

Yesterday I wrote on my favorite philosopher from the Modern tradition, Immanuel Kant, and troubles for his theory – today I want to briefly touch on his critique of the man who began said tradition, Rene Descartes, in order to put Kant’s theory in context.

Kant sets his sights on Descartes’ rationalistic theory, but which Kant terms ‘idealistic’, which holds that the existence of external objects are doubtful and indemonstrable unless sufficient evidence can be given for them through reasoning. Kant calls this view ‘problematic idealism’. It is worth making clear that Kant opposes this position because it entails that the existence of objects are not known immediately through perception, but only through reasoned reflection. Problematic idealism, then, is the view that knowledge of external objects is reasoned from the indubitable existence of inner sense. Kant posits that problematic idealism is erroneous because external objects are necessary for the experience of inner sense, such that, inner sense cannot be used to prove the existence of external objects as the existence of inner sense presupposes the existence of external objects.

Problematic idealism is false, for Kant, because in order to have the sense of self necessary for reasoning to the existence of external objects, one must perceive their self as existing in time, and thus, know that they exist through time by the reference point of an external object. More specifically, Kant argues against problematic idealism by positing that the agent is aware that she is in time. In order to be aware that she is in time she must have a reference point, that is, something that endures through time, in order to know that her self endures through time as well. The reference point cannot be anything about her self, as this would be nothing more than her assuming that she exists in time. On this matter Kant notes that, “this permanent [reference point] cannot, however, be something in me, since it is only through this permanent that my existence in time can itself be determined” (Critique of Pure Reason, 245). In other words, in order to know that you are in time, the reference point cannot be yourself, as this would not be informative of whether you are in time. Thus, the reference point must be something other than the self, such that it must be an external object.

Yet if this is the case, then there must be real objects out in the world – not just representations of objects – which are known immediately in interacting with them, in order for one to have a concept of self. In Kant’s own words, “the consciousness of my existence is at the same time an immediate consciousness of the existence of other things outside me” (Critique of Pure Reason, 245). Therefore, problematic idealism, or Cartesian rationalism, is false because it denies that we immediately know that external objects exist in order to have inner sense.


, , , , , , ,

  1. #1 by Nemo on March 16, 2014 - 5:13 pm


    I came across your blog when browsing WP posts on Descartes.

    I’m not clear on Kant’s conception of time, having not read Critique of Pure Reason. It seems that Kant simply assumed our conception of time, whether right or wrong, as an a priori principle, but it is questionable to me. We may exist and perceive in time, but it is not evident that we are aware of time itself.

    For instance, when we observe a distant star, they appear to exist at the same moment of our observation, but in fact, the light rays from the star were issued millions of years before they reached the Earth. So we perceive something that existed millions of years ago as something current; It is possible that the “I” that one perceives in “I think” is also from the immediate past, but because one is unaware of time, he perceives “I” as existing in the present, and so concludes “I exist”, not “existed”, Even so, the “I” who perceives does “exist” in the present moment, regardless of time.

    “this permanent [reference point] cannot, however, be something in me, since it is only through this permanent that my existence in time can itself be determined”

    It cannot be anything external and corporeal either, because none of the corporeal things are permanent, but change over time. In fact, it is by observing changes and motion in external objects, with ourselves as reference point, that we become aware of time.

    • #2 by ausomeawestin on March 16, 2014 - 5:56 pm

      Well thank you for stopping by, your insightful comments, and linking to me on your own blog! First off, I think you are right to be tentative in granting Kant his thesis on the a priority of time. His supporting premises are a tad question-begging, and their plausibility is given by the overall success of his project, such that the project is somewhat cyclical, but no more than a coherentist epistemology can allow.

      But Kant’s point is not that that the subject perceives that time is passing provides certainty of existence; it is that because the self perceives that external objects persist through time the self recognizes itself as existing in time with those objects, in immediate interaction with them. The knowledge of self is not deduced or mediated by external objects, it is merely given by experiencing self with external objects.

      In other words, the real emphasis in Kant’s critique of Descartes is in the immediateness of knowing of the existence of the self and external world all at once, not in being aware that the subject is in time, as you suggest, which I do not think is what Kant had in mind.

      • #3 by Nemo on March 16, 2014 - 7:05 pm

        IOW, the “self” we perceive is a “phenomenon”, not “noumenon”. Kant posits that one experiences “self” only through the senses, but not direct self-reflection and self-awareness, as Descartes asserts, because, to Kant, the “noumenal” self is unknowable. Is that correct?

        My question is, then, how do we know “freedom” as something belonging to the “noumenal’ self, and not a “phenomenal” construct?

      • #4 by ausomeawestin on March 16, 2014 - 8:20 pm

        That is correct!

        If I’m not mistaken, Kant thought that we cannot know at all whether “freedom” obtains for the “noumenal” self. He thought that because arguments could be made both for the existence of or the lack of freedom for the noumenal self, further argumentation would inevitably lead to logical contradictions or antinomies, showing that we cannot apply logic to the noumenal realm because logic is made possible by human cognitive experience, such that logic is only applicable to human experience. But Kant thought the “phenomenal construct” is quite useful so long as we think of it only as being about the phenomenal realm, where there may be objective truths about human intersubjectivity in virtue of the shared human cognitive experience.

      • #5 by Nemo on March 16, 2014 - 8:55 pm

        If Kant were consistent, he would concede that reason itself, and pretty much everything else, belongs to the phenomenal world, not the noumenal. Therefore, his whole argument on ethics, which concerns the noumenal self, is forfeit.

      • #6 by ausomeawestin on March 17, 2014 - 6:18 pm

        Perhaps. It’s been awhile since I’ve read the metaphysics of morals so I’ll have to take your word that his argument rests on the noumenal self.

      • #7 by Nemo on March 17, 2014 - 7:50 pm

        Kant cannot conceive a universal moral law (Categorical Imperative) if the moral law is dependent on sensory experiences, which vary a great deal among different peoples, so he posits that the noumenal self has freedom that is independent of the senses, and this freedom is the reason that the universal moral law exists.

        The problem is, if the noumenal self is unknowable, where does he get the idea of moral law and freedom pertaining to it? How can we even exercise this freedom that belongs to the unknowable?

      • #8 by ausomeawestin on March 18, 2014 - 11:20 am

        Oh I see, you’re saying that the noumenal self comes into play in the kingdom of ends and the third formulation of the categorical imperative.

        I’ve always read Kant as arguing that practical reason gives us the categorial imperative in it’s three formulations because of synthetic a priori knowledge of the structures of reasoning in virtue of the categories as they apply to experience — such that the categorical imperative never goes beyond the phenomenal realm when advancing claims about the kingdom of ends. Moreover, I took the freedom he spoke of to be inspired by Rousseau (as do many Kantian scholars) in the contractarian tradition of positing that one is free when one follows a law given by oneself by reason. Nothing deeply metaphysical there about the noumenal self, to my understanding, but rather an interesting defense of deontology with a critique of the hypothetical imperatives of consequentialism.

      • #9 by Nemo on March 18, 2014 - 7:38 pm

        one is free when one follows a law given by oneself by reason

        That’s my reading as well. But I don’t think the “reason” referred to here belongs to the phenomenal realm. The application of the CI never goes beyond the phenomenal realm, as you say, but the source of it is in the noumenal, because if the reason is also in the phenomenal realm, it would be the same as consequentialism.

        There are some many types of “reason” with Kant, I find it rather confusing, I have to admit. 🙂

      • #10 by ausomeawestin on March 19, 2014 - 7:02 pm

        “If the reason is also in the phenomenal realm, it would be the same as consequentialism”.

        Interesting, could you elaborate on why that would be so?

        In distinguishing categorical from hypothetical imperatives (hypothetical imperatives are consequentialist moral reasons), Kant was not arguing that categorical reasons are from or apply to the noumenal realm and hypothetical reasons are from and apply to the phenomenal realm. Rather, his point was that our epistemic intuitions tell us that moral imperatives must be categorical if they are to be made true by pure practical reason, that is the moral reasoning that structures our experience. So what distinguishes Kant’s view from consequentialism, here at least, is the logical structure of moral reasons, not where moral reasons originate. It is precisely because moral reasons and facts are not contingent on the situation and the whims of the actor that Kant concludes that moral imperatives must be categorial in virtue of reason made possible by the human cognitive structuring the phenomenal realm.

      • #11 by Nemo on March 19, 2014 - 7:18 pm

        moral reasons and facts are not contingent on the situation and the whims of the actor

        Why not? What exactly is “pure practical reason”?

      • #12 by ausomeawestin on March 19, 2014 - 7:31 pm

        I should have been more careful there; of course the morality of an action depends on the brute facts of the situation. What Kant denied was that moral reasons are conditional on desires, as Hume argued, as this would entail that moral reasons are arbitrary and not objectivity right or wrong by the light of reason.

        I shutter at the thought of saying exactly what pure practical reason is, but I suppose it could be stated as a theory of reason that proposes what actions are intrinsically good — such that pure practical reason is the categorical imperative.

      • #13 by Nemo on March 19, 2014 - 8:04 pm

        What’s not clear to me is how Kant’s ethics would be different from consequentialism, IF reason is contingent on the situation. How is it different from saying “the end justifies the means”, which has some reason behind it? Hence the question: What exactly is “pure practical reason”?

      • #14 by ausomeawestin on March 20, 2014 - 3:25 pm

        Well the reasoning itself isn’t contingent on the situation, it’s just applied to the situation to assess the maxim of the action — whether that action could be willed by a good will, which is perhaps a better way of thinking of pure practical reason. The good will is ruled by reason so it is bound by rules of logic, first and foremost for Kant the law of non-contradiction, and logic is not contingent on the situation, so likewise reason is not contingent on the situation.

        Think of the difference between Kantianism and consequentialism this way:
        Consequentialism entails this sort of moral theorizing:

        “If it causes more pleasure than pain to lie in this situation, then I ought to lie.”

        The moral conclusion follows as a consequent from the antecedent facts (as a conditional/hypothetical imperative).

        For Kantianism the moral conclusion does not follow from an antecedent claim. Moral conclusions are categorial in there being without exception as they are ruled purely by dictates of logic, which are without exception.

        In the case of lying above, suppose that the facts suddenly changed and it would cause more pain than pleasure to lie. The consequent changes as well such that then it would be wrong to lie. Kant thought that morality, as being grounded in human reason, could not change based on the circumstances like this as moral laws must be as stringent as the rules of logic, as the moral law is derived from logic.

      • #15 by Nemo on March 20, 2014 - 5:22 pm

        whether that action could be willed by a good will, which is perhaps a better way of thinking of pure practical reason.

        People can use reason and logic to justify all sorts things that are not “good”, because logic is content-free — it doesn’t tell us what is good. You have to define “good” first, before you can apply logic, and how do you define “good” without an “end” in mind?

      • #16 by ausomeawestin on March 21, 2014 - 3:47 pm

        Good points. I agree, but Kant doesn’t.

      • #17 by Nemo on March 21, 2014 - 4:22 pm

        What’s his “reason” for disagreeing?

        I find it ironic that “you shall not lie” is a categorical imperative.

        The command to not lie presupposes that the person knows the truth, but if the noumenal realm is unknowable, as Kant posits, nobody really knows the truth, which means everybody necessarily lies.

      • #18 by ausomeawestin on March 22, 2014 - 5:40 pm

        Kant’s reason for disagreeing is the belief that a fully-rational person would not perform immoral actions, because the categorical is a rational principle — this sort of fully rational principle is put forth by ideal-observer theories as well.

        Kant thinks there is truth, it is just limited to the intersubjective realm of human experience, so no one is lying in the sense that you suggest. But what is more, “lying” refers to the purposeful deceiving of another person. We are not all constantly attempting to deceive each other through knowing that there is a separation of the noumenal and phenomenal realms and our concepts only refer to the phenomenal realm, so it would be a mistake, even on your misinterpretation of Kant’s view, to maintain that we all necessarily lie.

      • #19 by Nemo on March 22, 2014 - 6:48 pm

        Plato also believes that nobody commits injustice voluntarily, but does it out of ignorance nevertheless. To command all people to act according to rational principle is like commanding newborn babies to excel in quantum mechanics. IOW, it’s very unreasonable.

        One would think that “Intersubjective experiences” are not universal, because people’s experiences are vastly different. How does Kant deal with conflicting experiences?

        “lying” refers to the purposeful deceiving of another person

        I don’t think Kant would agree with that definition of “lying”. The purpose or intent of the person is irrelevant, because it’s focus is on the consequence of the action, which Kant disregards.

        Moreover, what if people lie for a good purpose, not to harm but to protect?

      • #20 by ausomeawestin on March 23, 2014 - 12:50 pm

        Kant was working in the Platonistic tradition in responding to Hume, who argued the necessary and sufficient conditions for motivation are a means-end belief and a desire for the end — all motivation requires desires in some for or other. Kant intended to challenge Hume’s idea and posit that beliefs could be motivating by themselves. As you point to, Kant did not adequately meet his mark and crafted an argument about rationality instead. Against Hume’s rejection of motivational judgment internalism Kant argued for reasons internalism — he argued that beliefs were intrinsically motivating rather than necessarily motivating. What you reject as implausible is motivational judgment internalism, and I agree with you, it it implausible. But Kant argued for something different, and I think it is quite plausible.

        The sort of intersubjective experiences Kant has in mind are those of the experiences of space, time and causality. I think we can agree that it is the norm for humans to experience these, and that those who don’t lack those experiences due to neurobiological processes that differ from the norm.

        I suppose I should have been more careful with my words. By “purposeful” I mean “intended purpose” which is to say “intention”. So, lying refers to the intention of deceiving someone. The focus is on the intention, not the consequences, which is consistent with deontology.

        What of a good intention for a lie? An act being a lie is a moral reason against that action, but perhaps there might be other moral reasons that out weight the duty to not lie. Kant wouldn’t admit that, but I think, as a pluralist, that there are moral features that might override the moral wrong of lying. There was something morally wrong in James Clapper lying about the NSA not collecting data on millions of Americans, even if it was to “protect”.

      • #21 by Nemo on March 23, 2014 - 3:52 pm

        I don’t think Kant is working within the Platonist tradition. Firstly, Plato believes the substance of things (the noumenal) can be comprehended by the intellect, not through sensory experiences. Descartes is much more in line with Plato than Kant in that regard. Secondly, Plato also believes that ethics is derived from rational principles, as rational as the laws that govern nature, and both can be comprehended by the intellect. From a Platonist point of view, a violation of ethics is no different from a violation of the laws of nature, and the offender suffers the consequences of his own action.

        In contrast, since Kant denies that the nature of things can be comprehended by our rational faculty, there is really no “rational” ground for his ethics. He cannot argue that anything is “intrinsically motivating”, because that presupposes knowledge of the intrinsic nature of things, which he denies.

        Your focus on intent may be consistent with deontology, but I doubt that is what Kant has in mind. Discussion of intent takes us into the realm of psychology, but Kant disdains mixing psychology and philosophy, and makes it clear in no uncertain terms.

      • #22 by ausomeawestin on March 24, 2014 - 4:38 pm

        I suppose I should have been more clear that I was saying that Kant is inspired by Plato’s marriage of cognitivism and internalism for moral motivation, I had assumed that since the conversation had drifted to the topic of moral motivation it was implied that Kant is working in the Platonistic tradition when it comes to moral motivation. I certainly can appreciate the differences between Kant’s and Plato’s epistemological theories.

        The intrinsically motivating feature that I speak of is the belief itself, not the state of affairs that the belief tries to fit (in the sense of directions of fit). So Kant can account for rationality without beliefs being about noumenal entities because the rationality of a proposition is from it’s logical form — whether or not a belief is contradictory, which is revealed in Kant’s universalizability test, revealing the internal consistency of the maxim. I see no reason why we must say that beliefs must be about noumenal entities in order for them to be rational and thus motivating.

        Kant’s moral project is to derive the logical moral reasons that underlie the goodness of the the good will, which is good because it does its duty for the sake of doing its duty — intention is central to Kant’s project, many scholars posit that an essential component of the maxim to be tested against the universalizability test is a stated intention.

      • #23 by Nemo on March 24, 2014 - 6:27 pm

        whether or not a belief is contradictory, which is revealed in Kant’s universalizability test, revealing the internal consistency of the maxim.

        But the universality test doesn’t determine whether a belief is self-contradictory. A lie is contradictory only because it contradicts the truth, which has nothing to do with the person’s intention.

        the the good will, which is good because it does its duty for the sake of doing its duty

        Where doe Kant derive the notion of duty?

        many scholars posit that an essential component of the maxim to be tested against the universalizability test is a stated intention

        Could you elaborate? I’m trying hard to understand. 🙂

      • #24 by ausomeawestin on March 27, 2014 - 11:25 am

        No, that’s not the sense of contradiction that Kant has in mind when he says that lying is wrong. If lying is right then it can be willed to be a universal law of natural (first formulation of the categorical imperative). But if lying was a universal law of nature then everyone would lie, and as no one will believe anyones word anymore you won’t be able to lie. “The problem is that this hypothetical world would impossibly contain the contradiction that everyone makes false promises but false promises cannot be made” (Joshua Glasgow, “Kant’s Principle of Universal Law”). The universaliziability test is without a doubt an assessment of the contradictory nature of a moral belief.

        Kant derives his idea of duty by noting that if two persons attempt to protect a superior officer by jumping in front of a bullet, and the bullet strikes one person, the other person is not the worse morally. Both are of equal moral character. What this reveals, for Kant, is that the results of an action are not morally significant, but rather the intent or purpose for which that action was done matters morally. Both men had the same intention, so both did the right thing. Kant was attempting to discover what makes the good will good, and thus uncovering that it is the intention behind an action and not the result of an action that makes a good will good, Kant uses the word ‘duty’ to refer to the state of an action being good because of the intention behind it.

        When we aim to see whether a moral proposition/belief is indeed moral we see whether it is universalizable to decide whether it could be a universal natural law, or in other words, a command of reason. To decide whether an action could be universalizable we also need to consider the intention for that action to see how that action would be implemented when universalized. Consider the difference between “I will drive in the emergency lane when there is traffic to get to work more quickly” and “I will drive in the emergency lane when there is traffic to get to the hospital more quickly”. The intentions to driving in the emergency lane are very different here: to get to work, and to get to the hospital. Making a universal law of the first example results in if whenever someone wanted to get to work more quickly then they could drive in the emergency lane, resulting in everyone driving in the emergency lane during traffic such that the emergency lane would no longer be an emergency lane, there would no longer be a faster moving lane if everyone else was in it too. So that action, because its intention, is not universalizable because if everyone acted in that way you would not get to work faster. By contrast, making a universal law of driving in the emergency lane to get to the hospital does not lead to contradiction, because most people on the highway at any given time are not driving to the hospital, so to universalize that intention would not lead to traffic in the emergency lane, and the emergency lane continues to serve its purpose of being used for emergencies.

      • #25 by Nemo on March 24, 2014 - 6:50 pm

        If I understand it correctly, duty is performed by the noumenal self for the noumenal self.

        To perform a duty for the phenomenal self is like attending to a painted image of a sick patient, while leaving the patient himself unattended. Obviously, you have to have knowledge of the noumenal self to be able to serve him in any meaningful way.

      • #26 by Nemo on March 27, 2014 - 6:00 pm

        First of all, thank you for patiently bearing with me and explaining your position. I really appreciate that, though I don’t think Kan’t ethics is logically consistent or defensible. 🙂

        The problem is that this hypothetical world would impossibly contain the contradiction that everyone makes false promises but false promises cannot be made

        That is just downright sloppy logic.

        The CI demands that whatever one person does, all the people do the same. (It would be ridiculous to demand that a person always does the same thing all the time.) It doesn’t demand that a person must always lie, for the person only lies when he sees fit or when he can benefit from his lies. If the universality rule is applied, it simply means that all the people lie sometimes, but not other times, which is exactly the case in real life. There is no impossibility nor contradiction. The same applies to your emergency lane example.

        Intention is not enough without a rational understanding of the good. Suppose somebody accidentally stumbles into a make-shift operating room, and sees a person being cut open alive. With good intention, he would try to stop the operation, and endanger the other’s life by his interruption.

        Having said that, I agree that Kant’s notion of duty is commendable, if it is based on sound reason. The right intention, in this case, is to perform the duty always, even if the person must suffer painful consequences. The good example is the death and trial of Socrates, who refused the chance to escape from prison and avoid the death penalty, because he reasoned that it would be unjust to break the law.

      • #27 by ausomeawestin on April 3, 2014 - 3:09 pm

        It’s been my pleasure, I do like to discuss the nuances of theory. But I do want to note that I am not a Kantian — this dialogue has not been a defense of MY view, only Kant’s, or rather, my reading of Kant’s view.

  2. #28 by Larry on March 23, 2014 - 11:07 am

    Hello again. Having read this post on Kant and your other interesting one from the day before, I’m wondering about your conclusion: “there must be real objects out in the world – not just representations of objects – which are known immediately in interacting with them, in order for one to have a concept of self”.

    As I understand Kant (which I often don’t), his argument against “problematic idealism” relies on facts about our representations. In order to have a concept of self, we must have representations of external objects. In other words, we must experience phenomena as being separate from ourselves. When you conclude that “there must be real objects out in the world – not just representations of objects”, are you merely drawing the distinction between two kinds of phenomena — the ones that are experienced as “outside me” and the ones that are experienced as “inside me”?

    That is apparently what Kant is doing, although in ordinary language (even ordinary philosophical language), “real objects out in the world — not just representations of objects”, would usually refer to mind-independent stuff, things-in-themselves or noumena. You’re not referring to that kind of stuff by “real objects out in the world”, right?

    • #29 by ausomeawestin on March 23, 2014 - 1:38 pm

      Good catch, I can see that my phrasing there was misleading. My intention was to emphasize the claim “I am immediately aware that there are objects external to the self, and they are not just possibly illusionary representations until proven to exist through deduction”.

  1. “Meditations on First Philosophy” by René Descartes « Books On Trial

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: