Certainty, Objectivity, and the Secure Path of a Science
by existential calvinist on 2006年12月14日 11:23 PM
@ Home / Essays / ESSAY14 (edit, history)
Introduction
Critique of Pure Reason is an imposing book. Given the magnitude of the work, it is easy to lose sight of its overarching objective in the innumerable details that make it up. Accordingly, in order to evaluate the Critique as a whole, it is necessary to return to the beginning to ask what were the conditions that motivated its creation. The goal of the Critique as a whole can be found in the preface to the second edition in which Kant repeatedly refers to “the secure path of a science” and the relationship of the discipline of metaphysics to this path. The means that Kant employs to clear this path is critique and as a corollary, Kant felt compelled, as he is so often quoted, “to annul knowledge in order to make room for faith” (Bxxx). As an adjunct to this process, Kant instituted a “Copernican revolution” for the ideas of philosophy in which he took numerous familiar terms and gave the entirely original interpretations by reversing the grounds on which they are supposed possible, such that we are ill advised to attempt to assess his employment of them until after we have first thoroughly familiarized ourselves with his usage. Among the most important for our purposes are a priori and a posteriori; synthetic and analytic; inner and outer; realism and idealism; and subjective and objective. Until we have understood the meanings of these terms in Kant and how he employed them, we will be wholly unable to assess the success or failure of the Critique at putting metaphysics onto the secure path. Accordingly, this essay will begin by examining the marks of the secure path before enumerating properties of the aforementioned terms and finally assessing the strengths and weaknesses of the project as a whole particularly with regard to the effect of the Copernican revolution on the nature of certainty and subjectivity.
1. The secure path
According to the preface of the second edition, the mark of a science that is on the path is easily distinguished by its result. Should practitioners falter on the outset, retrace their steps continually, or disagree about the method employed, this shows that the science is not on the proper path but merely “groping about” (Bviii). However, of all sciences, is it fair to judge philosophy in general and metaphysics in particular by its results? The security of a proposed path for the science of making pudding can be judged in the eating, but for a philosopher, to be judged by the result is to assume the existence of what is to be shown. Philosophy is a process by which we determine ends, hence judging it from results is problematic at least. Ought we not to admire those philosophers like Parmenides and Zeno who stood by the result of their method even when it appeared to all others to be utter nonsense? Surely, our criticism of them must be that their method was flawed not that their results were absurd. For to judge the Eleatics by their result is to presume their guilt before it has been shown. It seems sensible that the dogmatist would judge the method based on the result, for the dogmatist can say, if you do not arrive at my conclusion then you must be doing it wrong, but it less clear what right the critic has to make a judgement on the basis of the result of inquiry.
Here we can see the first mark of the Copernican revolution. For Kant, we do already know what sort of result we want philosophy to produce for us. Philosophy must produce a picture of the world that allows us to navigate it, because if it merely paralyzed us, it would be worse than worthless. Thus it is appropriate to judge the method by the results. If nothing else, the result of critical philosophy has got to be a world that is remarkably similar to the world as we conceive it, or else that critique will suicidal, because it will undermine the grounds on which we accepted the reasoning process which underpins it. Of course, the result of philosophy need not conform to all of our prior expectations, but if it completely overturns all our basic beliefs about the world then there is no basis left for justifying the method that produced that result. We judge the efficacy of our reason on the basis of its ordinary use, so if all ordinary thought is overturned by the application of logical method then the method overturns itself as well. Kant takes numerous aspects of experience to be self-evidently and irrefutably true. A philosophy that cannot account for the objective validity of these aspects is no philosophy at all.
There is another notable aspect here. Why should it matter if there is continual disagreement about the methods employed? If the proposed path is correct, it is correct, so why should we be concerned if others fail to see its correctness?
This gets at the other key presupposition of the Critique. Since reason is common to all (and this is a key assumption of the Critique), it should at least be theoretically possible to popularize a true path, even if some minority of committed skeptics resist it. Hence Kant says on A820/B848 that for an objective judgment, “its touchstone externally is the possibility of communicating the assent and of finding it to be valid for every human being’s reason.” (The importance of this possibility of universal consent will grow as we consider other aspects of the Critique.) In our thinking, we sometimes take for granted the image of the lone philosopher working out the secrets of existence far away from society, but buried in Kant’s work is the half hidden premise that the social reality of rational beings cannot be itself completely irrational. Thus, for metaphysics to be on a secure path, the assent of the community to appropriateness of the methods employed is vital. If Kant can be criticized on this point, it is only that he did not emphasize it enough, but refers to it only elliptically.
2. A priori and a posteriori
There is a difference between the two editions of the Critique concerning the a priori. In the introduction to the first edition, Kant speaks of the a priori as just that which commands necessity (A2), but in the second edition he gives a much fuller treatment. Reflecting on just the term a priori, he explains that we can know a priori that digging up the foundations of house will destroy it, in the sense that we know this result will occur prior to carrying the experiment out (B2). However, this is not the sense of a priori used by philosophers in most cases. We know prior to digging up the foundations that the house will collapse, but we know this only because of our other experiences — those of gravity, the structural integrity of building materials, etc. There is another sense of a priori that is favored by the early modern philosophers. Kant calls this sense “pure.” Members of this class of a priori cognitions “occur absolutely independently of all experience” (B2) and “nothing empirical whatsoever is mixed with them.”
Here it is worth stopping and asking, “Is there any sense at all in speaking of cognitions apart from the empirical experience of them?” For whether such cognitions apart from all experience are possible or not, it is quite clear that we never see them. Humans, as beings in time, cannot have a non-temporal cognition without drastically altering their mode of being. Even if I can work out, say, that 5 + 7 always equals 12 and always will, there is no usual sense in which I can work this relationship out without myself existing in time and using thoughts that have some basis in an experiential relationship with the world. Hence, a truly a priori cognition is impossible apart from some transmundane experience which is not experienced in time. We might here posit that a priori truths are those truths that remain true when contemplated by some sort of non-temporal mind, but since we have no regular access to non-temporal minds, whatever speculation we have about the kinds of truths they will accept is without foundation. Yet, many of the early modern philosophers strongly felt not only that there must be something special about the a priori that makes it not only possible but also that the a priori is the only thing worthy of being the basis for a philosophy. This is strange in a way. In philosophy, we like to say that it is more certain that 5 + 7 = 12 than that the sun will rise tomorrow, but in fact if tomorrow a mathematician published a proof that overturned arithmetic, we would be merely bemused; if the sun did not rise, we would be thoroughly overwhelmed by shock and horror. Thus, there is at least one sense in which the everyday is more certain than the a priori.
Kant explicitly acknowledges this seeming difficulty of the a priori by positing that, “even though all our cognition starts with experience, that does not mean that all of it arises from experience” (B1). How is this possible? “[I]t might well be that even our experiential cognition is composite, consisting of what we receive through impressions and what our own cognitive power supplies from itself” (B1). So, the a priori is governed by facts about the nature of how our cognitive powers work that are prior to experience, and not by whether some sort of non-temporal mind would assert them. Next we find that Kant returns to the definition given in the first edition, but with a fuller justification behind it. He asserts first that a priori propositions are those propositions “such that in thinking it we think at the same time its necessity” and second that “experience never provides its judgments with true or strict universality” (B4). Thus, a priori and a posteriori are not merely philosophical categories for propositions, but experiential properties of propositions that we are made to feel by the very nature of our cognitive faculties, but a certain class of propositions (those based wholly on empirical observation) fail to generate the requisite certainty. If we inquire as to the conditions that make possible certainty, Kant reminds us on A822/B850 that we will find, “I cannot assert anything — i.e., pronounce it as a judgment that is necessarily valid for everyone — unless it is something that produces conviction.” Without conviction, we cannot feel either certainty or necessity. However, we do feel necessity, so it must have its basis on a particular kind of uniquely powerful conviction with a basis in the nature of our mental faculties themselves.
The committed skeptic may reply to all this that the feeling of necessity which we have is non-veridical. We may subjectively feel that a proposition is necessary, but that does not make it so. Kant’s objection to this is not clear from the text, but we may reconstruct it as being something like this: just as a well formed eye cannot help but see the redness of a red thing under the conditions that allow for its experience, so too, a rational mind cannot help but feel the necessity of certain propositions. Further, just as the experience of redness shows indubitably there is such a property as redness (though it is not necessarily proof of the redness of the particular object being experienced), so too an experience of the necessity of a proposition shows that there is such a thing as necessity (though the proposition in question may be wrongly inferred to posses it). Repeated observations under varying conditions by well functioning sight organs allow us to verify the redness of an object, and repeated experience of the necessity of a proposition shows us its quality of being a priori. If the skeptic fails to see the necessity of certain propositions, there is a possibility that as the colorblind person possesses defective eyes, the skeptic possesses a defective sense of reason. However, it is more likely that the skeptic is only playing ignorant and sees full well the necessity of a priori propositions but does not wish to assent to them for some other motivation. Note however, there is a sense in which even the skeptic cannot be wholly blind to necessity, for reason itself is universal for Kant.
This is only a reconstructed attempt at a reply to the skeptic, but it is still worth noting the Copernican reversals undertaken. Necessity is usually taken to be something inhering in the proposition, but here we have deduced its existence based on our subjective experience of it. In the typical Kantian manner, we have asked, what are the conditions for the very possibility of our experiencing a sense of necessity, and supposed that there can be none other than the actual existence of true external necessity, the a priori. Given this justification, it should not be so surprising that Kant will later restrict the application of the a priori to that which can, in principle, be experienced. “[T]he complete cognition of appearances — which, after all, is what all a priori principles must amount to — is merely our possible experience” (A181/B223). The a priori is a catalog of what is possible in experience made prior to a specific experience, hence since the basis of the a priori is our subjective perception of an objective necessity, it follows naturally that for the things of which we can have no experience even in principle, we are without the grounds to determine whether it is “a priori” or not, for we neither have nor can have a prior experience of the necessity of the content of the proposition. This conclusion will have great impact when we consider realism and idealism in Kant.
3. Synthetic and analytic
For Kant, a synthetic judgment is one in which a connection is made between subject A and predicate B, when “B, though connected with concept A, lies quite outside it” (B11). To understand what this means, let us take a recognitional view of concepts. On this view, the concept of dog is a mental mechanism that when given some input an agent can render a judgment, either “this object is a dog” or “this object is not a dog.” (In practice, our mental concepts are more likely to give probabilities — ”might be a dog” and “is almost certainly a dog” — than concrete verdicts, but we can lay that aside for our discussion here.)
If we examine the proposition “all dogs are animals,” there are two possible ways by which we can vouch for its truth. First, we can examine the rules which underlie the mechanism of recognition to see what it is possible for our concept to accept or reject. Naturally, if we examine our concept of dog, we find that a test for the property, “is an animal?” is a part of the process of recognizing an object as a dog, thus it cannot be the case that something satisfies the property “is a dog” without first satisfying “is an animal.” Thus the proposition that all dogs are animals is correct. Second, we can attempt to arrive at some input that satisfies “is a dog” but does not satisfy “is an animal” by imaginatively generating various objects then testing them against our concepts to see if we can ever satisfy the first condition of the proposition without satisfying the second. This empirical method of testing the range of our concepts is often employed in practice, since the rules that are underlie our concepts are not always available to introspection although the concept itself is usable. Unfortunately, we cannot be have complete confidence in the result of this method unless we are confident that we have tested all possible classes of input to the concept.
Next, if we analyze the proposition, “all dogs weigh less than one ton,” there is nothing in our rule for identifying “is dog” that connects it with “weighs less than one ton.” Furthermore, when we try the second test, our imagination can readily provide us with an image of what such a dog would look like. It is only when we examine the image of a one ton dog produced by the second test in light of the concept “is possible” that our experience tells us that we will never find such a dog in experience (barring the work of a mad scientist).
So, the first proposition Kant will call a priori analytic (since we can know the truth of the proposition prior to experience by analyzing the concept) and the second proposition a posteriori synthetic (since it is only by adding experience to our concept that we can deem a one ton dog impossible). Note that those like Frege who feel that the identity of concepts is nothing other than two concepts having identical ranges of applicability will typically overlook the first method for determining the relationship of concepts in a proposition in favor of the second.
If we examine the proposition, “5 + 7 = 12,” it is less clear how to classify it. On the one hand, if we examine our compound concept of “5 + 7,” it is clear that our imagination can produce no single number other than 12 that is equivalent to it. On the other hand, there is nothing about the recognition of “5,” “plus,” or “7″ which necessarily entails 12. If we examine the concept of “plus,” it tells us how to recognize that addition has been done, and by introspecting that concept we can derive how actually to do addition, but recognizing how to do addition is not the same thing as doing it. The joint concept of “5 + 7″ gives us the rule for finding its equivalent, but it does not carry it out in advance. If it did, a number of bizarre consequences would follow. The first is that I could not say that I fully grasped the concept of “12,345 + 54,321″ unless I already had “66,666″ first, since the concept entails its consequence. But if this were so, then so soon as I am given “66,666″ I should be able to decompose it into “12,345 + 54,321,” just as I am able to transpose “an unmarried man” and “a bachelor” with perfect ease once I am familiar with those concepts. (Giving counterexamples of cases where we do not immediately recognize the applicability of certain concepts syllogistically will not help here, since such counterexamples are precisely the ones supposed to be secretly synthetic.) In addition to “12,345 + 54,321,” “66,666″ is also equal to “66,665 + 1,” “66,664 + 2,” and so on. If my having the concept of “54,321 + 12,345″ gives me its result in itself, then having the concept of “66,666″ should give me in itself all of its infinite relations with other numbers. However, if this is the case, then the rule behind the recognition of any number is an infinitely long list of possible relations which that number can have with other numbers. This seems to be patently absurd, for it is clear that my mind is too limited to contain such a list, hence I must not have the concept of any number at all. But if I do not have the concept of any numbers whatsoever, how is it that I am able to recognize the necessity of certain mathematical relationships at all? On this basis, we must say that the concept of “5 + 7″ does not give us its result; it merely gives us the means for finding its result in our a priori intuition. We call the grounds for finding the answer “a priori” because it gives a necessary result, and “intuition” because it is something non-conceptual in objective thought.
If Kant were to stop there (B14, “Mathematical judgments are one and all synthetic”), it would be controversial enough, but he goes further. These synthetic a priori propositions are not only found in math. They also are the fundamental principles of the natural sciences (B17) and moreover make up the whole purpose of metaphysics as a discipline (B18). Here again we see the mark of the Copernican revolution in Kant’s thought. For Kant, metaphysics is nothing other than the search for those things which must be added to a proposition to get an a priori synthetic judgment. On the one hand, this deduction follows naturally from the previous discussion of the a priori. The a priori is that which gives us a sense of its necessity. To the extent that we can feel there to be some necessity in experience, it must be derivative of some a priori basis, but at the same time, it must be synthetic if it is going to reveal original information to us. On the other hand, we should be intimately aware of the extent to which this supposition rests on Kant’s assumption that experience is mostly veridical. If you were to tell Kant that because we are actually victims of Descartes’ evil genius or brains in a vat or Zhuangzi’s dreaming butterfly or whatever else ordinary life is not real, Kant may entertain the premise, but he will not accept the conclusion. For whatever else lies beyond our experience, our experiences themselves are actual experiences, and it is not possible to show that they are anything else. Therefore, if something must a priori be the case in order for our experience to exist then that thing is the case, and by Kant’s reckoning, synthetic a priori truths are such necessary things.
4. Inner and outer
In Kant’s time, a specter was haunting Europe — the specter of skepticism. Ever since Descartes’ famous Meditations, philosophers were keen to banish the kind of radical skepticism he raised only methodologically. Descartes himself of course thought he had shown how to disprove the doubts he had raised, but his proofs were not to the satisfaction of his successors. Descartes began with his famous cogito ergo sum as the refutation of skepticism, but others like Berkley and Hume wondered if it would ever be possible to move much past this feeble starting point to the existence of a truly external world. Into this debate, Kant once again brings a Copernican method. Before I think therefore I am, claims Kant, I experience the outer therefore I think.
In the second edition of the Critique, Kant spells out this commitment explicitly the theorem of his “Refutation of Idealism,” “The mere, but empirically determined, consciousness of my own existence proves the existence of objects in space outside me” (B275). He goes on to comment, “we have proved outer experience is in fact direct, and that only by means of it can there be inner experience — i.e., not consciousness of our own existence, but yet determination of that existence in time. …[C]onsequently, inner experience is itself only indirect and is possible only through outer experience” (B277) “For if even outer sense were merely imagined, this would annul our very power of intuition which is to be determined by the imagination” (B227n).
Unfortunately, for all the promise of the result, the actual proof is opaque and somewhat dubious. Kant claims that in order for us to have inner sense, we must be able to make time determinations. (This is one of the central claims of the model of the intellect given in the Critique, though it is not perfectly clear that it is necessarily true.) In order to do this, there must be something permanent in perception against which to make such determinations. (Again, the necessity of this claim is a bit suspect, but the claim is not implausible either.) This permanence cannot come from inner sense, because time determinations can only come as a result of this permanent thing, and inner sense requires time determination for its existence. Hence, concludes Kant, “perception of this permanent something is possible only through a thing outside me and not through mere presentation of a thing outside me” (B275). Here, the reader is left a bit confused. What is it that Kant is proposing keeps the imagination from generating the supposedly permanent in perception? Surely, we dream every night. Why does there need to be something really outside of me and not just presentation of something?
This is not the end of the matter however, for as puzzling as Kant’s proof may be, there is still a glimmer of something that we can hope to salvage from it. In a note, Kant instructs us that we must distinguish between “the mere receptivity of an outer intuition from the spontaneity that characterizes all imagining” (B227n). Here, Kant has something. Even dreams, the nightmare of all would-be empirical realists, exhibit something of the spontaneity of the imagination that Kant claims distinguishes the inner and the outer. Forgetting for a moment Kant’s larger ambitions for the inner sense as the realm of time determinations and the outer sense as the realm of the spatial, there is something very plausible about distinguishing between the part of our experience that directly responds to our will and the part that does so only through the mediation of the body in the world. Using such a principle, we can distinguish with a fair degree of reliability between what is inner and what is outer. In a dream, a stray thought of one’s friend brings that person into the room. In outer reality, things continue in their predetermined course unless we exert an effort to change them. Our outer senses impress upon us a kind of alienness to our will. Our thoughts however arise as nothing other than the product of our will. Though in certain cases, we may find our own thoughts repugnant, we never find them to lack the quality of being ours. We can imagine willing our arm to move but being distraught to see it remain slack, but we cannot imagine willing a determinate thought to come and not having it come. To the degree that we cannot present a thought, we have not willed it (though we may will the thought of a thought without getting at the deeper object). Our memory may fail us or we may be unable to follow a complicated proposition, but to will a specific thought is always already to think it. Thus, while we may find the outer world in agreement with our disposition at times, we never fail to distinguish it as something other than us, since it cannot have the intimacy that our inner sense possesses.
In the preceding paragraph, we ventured that there are reliable means of distinguishing the inner and outer sense in ordinary experience, but this does not yet convince us of the necessity of outer objects as truly outer. Perhaps outer objects are different from inner objects but somehow still inside of us. Here we enter the heart of the Copernican method. Kant has been saying since the very first that there is a positive sense in which even outer objects are within us and could not be otherwise. In order to have an experience of an object, that object must be presented to us, and a presentation is nothing other than the replacement of some external object with an internal representation of it. So, it is not possible that the outer world should completely outside of us if we are to have some relationship with it. Thus, what Kant was trying to show in his refutation was not that the outer sense contains nothing of the inner, but that it contains enough of something truly outer that it is resistant to the effects of our will on it. Furthermore, without this outer imposition on us, there would be no basis for our inner sense to work at all. The inner sense is endlessly productive of images, but those images are always in reaction to others. Without some initial outer prompt, there would be no source for the machinations of the inner sense at all. Yet, what exactly is the true outer that is completely outside and gives rise to the partially outside that is in us will be dealt with in greater detail in the next section.
In terms of the specter of skepticism, it is true that we cannot say that the outer world must take a definite shape even in our absence. We might be on the Truman Show or in the Matrix. However, this is hardly the most challenging fact about the outer world. Even accepting it completely as presented by sensibility and science, the earth could pass away tomorrow for some unforeseen but wholly scientific reason. Thus, the specter of skepticism is about a lack of knowledge in a more specific way than just the possibility of our being radically vulnerable to the vicissitudes of fate. It is a specific accusation that even things which we are absolutely confident will continue to existence are subject to possible refutation. But this will not hold, for there is no sort of evidence which can show us that we were never actually distinguishing between inner and outer sense. For though we may have been confused about them in specific instances, our classification of the inner and outer was not merely arbitrary, but dependent on the nature of our relationship to the senses themselves. One sense was self-evidently free, and one was imposed. In this regard, our having outer sense is indubitable.
5. Realism and idealism
Even if we grant Kant the distinction between inner and outer sense, this does not yet commit us to realism over idealism. It may be that outer objects are distinguishable from inner ones but lack the properties that realists assert they have. The exact definition of realism is a matter of dispute, but for our purposes, let us adopt four tests of realism. First, is an object distinct from the perception of it? Next, is an object distinct from its properties? Further, is it possible to observe the same object at different times? Finally, can different people observe the same object? It is difficult to classify Kant under these tests, because in each case, he will answer, “Yes, but…”
Kant categorizes his view as transcendental idealism/empirical realism and contrasts it with all preceding views, which he calls transcendental realism. Explaining the difference on A369–70 he writes,
By transcendental idealism of all appearances I mean the doctrinal system whereby we regard them, one and all, as mere presentations and not as things in themselves… the transcendental realist conceives outer appearance (if their actuality is granted) as things in themselves that exist independently of us and of our sensibility, and that would therefore be outside us even according to pure concepts of understanding. It is, in fact, this transcendental realist who afterwards plays the empirical idealist. Having wrongly presupposed that if objects of the senses are to be external then they must have their existence in themselves, ie. even apart from the senses, he then finds from this point of view all of our presentations of the senses are insufficient to make the actuality of these objects certain.
The transcendental idealist, on the other hand, can be an empirical realist or, as he is called, a dualist… Hence matter is for him only a kind of presentations (intuition), called external; they are called external not as referring to objects in themselves external, but because they refer perceptions to the space… although the space itself is in us.
In other words, the tragedy of all prior philosophy had been that philosophers supposed that if there were real objects, such things would have to exist independently of the mind, however since the mind cannot know anything that is completely independent from it, it cannot know these real objects. As usual, Kant reverses the normal pattern of thinking. For Kant, to be an objectively real thing is to depend on reason and its a priori categories. As such, through the application of universal reason, we can give a valid description of the unity which reason is compelled to impute to appearances.
Unfortunately, these real syntheses of appearances cannot be an object in itself (a noumenon). The synthesis is only a phenomenon, though no less real for it. About the noumena which give rise to phenomena, Kant is insistent that we can only think them, never cognize or know them. Thus, while Descartes said in his Meditations, “[G]reat things are also to be hoped for if I can find just one thing, however slight, that is certain and unshaken,” Kant has found a great number of things that are certain and unshaken, but nevertheless, knowledge of certain great things — the noumena — is not to be hoped for, since knowledge of the real depends on conformity to the categories produced by examination of the a priori synthetic conditions of experience, and the categories cannot apply to things outside of their appearance to our minds. “The sole character of actuality is… the perception that provides the material for the concept” (A225/B273), thus we cannot speak about the actuality of those objects which cannot appear in perception.
Applying his usual Copernican method, Kant has once again reversed the normal pattern of inferences. A phenomenal object is distinct from any specific perception of it but is nevertheless nothing but the unity of the possible perceptions of it that could be synthesized from it using the categories. A phenomenal object is not its properties or even their totality, but neither is it distinct from what gives them their unity in the mind. Phenomenal objects are the same throughout time, but time is (transcendentally) a mere ideal imposed on nature by our inner sense. Phenomenal objects are the same for different observers, but only because all intelligent observers have the same a priori categories in their minds. On the noumenal side, not only can we not answer the tests of realism in the affirmative, we cannot even ask the questions, because each of them supposes the categories, which the noumena are beyond. The noumena are the outer objects that are truly outer, thus there is nothing that we can say about them. As Kant says, “if we accept external objects as things in themselves, then it is absolutely impossible to comprehend how we could arrive at the cognition of their actuality outside of us, since we rely merely on the presentation that is in us” (A378).
Kant gives evidence for his dualistic system from two sides. In the “Transcendental Aesthetic” and the “Transcendental Analytic,” he argues for his system from the requirements of experience as such. For Kant, time and space must be ideal in order for us to have a priori synthetic knowledge of them, and we must have this knowledge in order for experience to be possible at all. Since time and space are transcendentally ideal, it follows naturally that objects in themselves must be free from the conditions of time and space, and thus independent of our ability to know them. In the “Transcendental Dialectic,” he argues for his dualistic system from the impossibilities of knowing a unitary system. The assumption that objects are things in themselves leads to absurd outcomes like the antinomies. The transcendental illusion is our persistent tendency to believe that we can apply our a priori categories to objects that are beyond the limits of experience, though this is quite impossible, according to Kant.
It is important to see why Kant calls himself the only true empirical realist, for this is at the heart of his Copernican method. Many philosophers had been called empiricists before Kant, but to the extent that they denied we can have direct knowledge of the real in experience, Kant thinks that they were actually idealists. Remember that from the start, we have emphasized that Kant finds our experience to be essentially veridical, if occasionally mistaken in particulars. This is why Kant considers himself the lone empirical realist. In order for our empirical experience to be correct, it have a real unity that is outside of us, and yet the things which we experience must have a connection to us in order for us to have knowledge of them. Thus we have Kant’s startling proposition that man makes nature. This is not to say that we control nature with our minds or anything of the sort. It is just that so long as our minds follow reason, we have no choice but to present the outside world as we do, and presenting in this way is what it means to be real. Again, it is the nature of the outer sense that it is a receptivity to presentations without regard to our will. Nevertheless, it is the uniformity of the object in regards to its presentation by our reason that shows the phenomenal object to be real. So, there could be no phenomenal objects without someone to experience them, for they “have no existence with an intrinsic basis” (A491/B519). However, about the noumenal objects that presumably support our phenomenal ones, we are not even able to say that they exist for doing so would employ a category, which we are not entitled to do.
6. The grounds for objectivity
All of the preceding has been propaedeutic to a discussion of the subjective ground of objectivity in Kant. As Kant says near the end of the Critique (A820/B848),
Assent is an event in our understanding that may rest on objective bases but that also requires subjective causes in the mind of the person who is judging. If the assent is valid for everyone, provided he has reason, then its basis is sufficient objectively…
Thus, all judgments of objectivity must begin with the subjective conviction of an agent of the judgment’s capability for universal assent. To be objective is to be capable of universal assent, and for this to be the case, all intelligent beings must have the same means by which to render judgments. Thus, reason must be universal in order for anything to be objectively real. Since we know that anyone who is making a judgment must be an existing and experiencing person in order to do so, we are justified in using the veracity of our experience to drive the discovery of the a priori concepts underpinning existence to which all beings capable of making a judgment must be subjected using universal reason. At the basis of our subjective sense of the veracity of experience is the objective universality of the categories, which ensures that all intelligent beings will create the same phenomenal objects when given the same sense data, thus allowing phenomenal objects to be objective rather than subjective, since all minds will present them identically. Hence the objective universality of the categories must in turn rest on the objectivity of reason itself, because reason is the one fixed thing in reference to which we make all of our objective measurements of the a priori synthetic, the categories, and so on. For Kant, we can be deceived and we can be mistaken, but pure reason cannot be false, and it in turns shows us the many ways in which even subjective experience can relate to objectively existing objects, i.e. objects for which there will be universal assent, not objects as something outside of us entirely.
The skeptic will protest that all of Kant’s system rests on the assumption that our pure reason is universal in its scope, but our only evidence for the objectivity of reason is its use in our merely subjective experience, hence we do not actually have a secure foundation for pure reason, let alone its self critique. Kant deals directly with this objections on B167–8:
Someone might want to propose… that the categories are… subjective predispositions for thinking that are implanted in us [and given to us] simultaneously with our existence; and that they are so arranged by our originator that their use harmonizes exactly with the laws of nature governing the course of experience… If such a middle course were proposed, the following would decide against it… the categories would in that case lack the necessity which belongs essentially to the concept of them. For the concept of a cause, e.g., which asserts the necessity of a result under a presupposed condition, would be false if it rested only on an arbitrary subjective necessity, implanted in us, to link certain empirical presentations according to such a rule of relation. I could then not say that the effect is connected with the cause in the object (i.e., connected with it necessarily), but could say only that I am so equipped that I cannot think this presentation otherwise than as thus connected. And this is just what the skeptic most longs [to hear].
Now, as good post-Darwinists we are obliged to postulate that it is the case that natural selection arranged our mental faculties such that they harmonize with our surroundings only to the degree necessary for our survival, and thus they do lack the kind of intrinsic necessity that Kant feels is required for the refutation of the skeptic. Does this mean that Kant’s objection is reduced to, “Skeptics cannot be right. If the categories are not objective, skeptics are right. Therefore, the categories are objective.”? All along, Kant has been supposing that experience is the basis for our understanding of the phenomenal world, but along with this, he has also fetishized reason as wholly a priori and outside of experience, thus allowing us to find the basis for experience and those objectively real elements within it. Now, if human reason is subject to the contingency of the evolution of the brain, does this remove from us the ability to reliably evaluate the necessity of our objects of cognition, since our minds are not connected to anything?
Here Rorty, the great skeptical specter haunting contemporary philosophy, will answer in the affirmative. The opinion of the post-modernists is that in the absence of universal perspective there is no way to resolve disputes objectively, thus all disagreements must be settled in the end by appeal to the individual biases of the discoursers. If group A tends to settle disputes by so-called “rational” methods, but group B settles them by some other method, group A has no right to tell group B that its method is inferior. Instead they must discourse with group B by forging a common vernacular that all participants can accept without prejudicially declaring their own method to hold the mantle of “a priori” or “objective.”
I propose that there is a response to the post-modernists and general skeptics based on Kant, but that Kant himself missed it. Yes, it would be impossible for us to suppose that we are in possession of universal, objective reason if reason had to be self-evidently a thing in itself because reason in itself could never be given in experience. However, it is possible that reason can still be objective in its applicability to the phenomenal world, if reason is taken as reason in appearance. I propose that by following Kant’s own Copernican method, we find understanding as it exists in appearance, but to suppose that we can find understanding that is outside of appearances is to neglect the rest of the Critique. The soul, the beginning of the world, causality through freedom, and God are all not to be found in this life because they are noumenal. Reason divorced from experience also cannot be found. With objects in the world, we have no experience of them except as presentations, but nevertheless posit their reality as objectively existing presentations called phenomenal objects that are independent of our individual minds, but dependent on our collective reason as intelligent beings. By this same process, we can argue that just as experience obliges us to suppose that objects are external to us but subject to our presentation of them, so too our reason as experienced is nothing other than a phenomenal mechanism of reasoning, which, though not independent of minds as such, is nevertheless not identical to the subjective reasoning of an individual human mind either.
In settling disputes, the post-modernists claim that different systems are incommensurable, since each system has its own subjective values which cannot be disputed, and there are no system external values to which to appeal when comparing systems. However, in actuality, we are remarkably adept at communication between groups and cultures. Even in the so-called clash of civilizations, we are always able to understand what the values cherished by other civilizations are and why they do so, even if we do not share those commitments. Thus, in the honored tradition of the first Critique, we are compelled to ask, “What are the conditions for the very possibility of understanding someone else’s viewpoint?” One condition of communicability is the existence of a sharable nature of reason. Thus, we can refine our view of reason the same way we refine our view of phenomenal objects, morality, beauty, or anything else. A color blind person cannot deny the existence of red, when informed of the sensations of those with different sense organs. A non-scientist cannot deny the existence of atoms, when informed of the evidence for their detection by scientific apparatus. Similarly, a non-logician is compelled to assent when presented with the truth tables of logic and the evidence of experience. In the case of red and the case of atoms, our assent is triggered by the obvious conformity of the concepts to the intuitive evidence, in spite of the fact that these concepts owe their existence partly to the nature of the mind itself. Similarly, in the case of truth tables, that assent may be compelled is no disproof of the supposition that their necessity is also partially compelled by the nature of the human mind. Thus, everything we have knowledge of — including the nature of knowledge itself — must be a chimera, composed in equal parts of our own nature and something outside of our nature about which we cannot speak.
Kant instituted a “Copernican” revolution in which, as we have seen, he shifted the basis for basis for realism from transcendental realism to mere empirical realism within a dualist system. Much as the real Copernican revolution changed the center of the solar system from the earth to the sun, Kant changed the center of the metaphysics from the externally existing object to the outer object in the context of a space created by the mind. Going further, we can urge Kant on to an “Einsteinian” revolution. In Einstein’s revolution, we went from thinking that there can be an objective viewpoint from which to observe the universe as it really is, to understanding that any understanding of the universe depends partly on the view point of its observer — though this by no means implies any view is correct! Similarly, future scholars with an affinity for Kant would be well served to recognize that there is no fixed perspective from which we can apprehend reason in itself, but by compositing different views of reasons, we will be able to construct reason as it is given in appearances for us.
7. Conclusion
Having examined Kant’s terminology and its justifications, we are at last in a position to assess the success of his enterprise. Kant’s goal was to put metaphysics onto the secure path of the science, which in part means judging post-Critique metaphysics on its ability to create consensus and make progress. In his system, the a priori is deemed possible on the basis of our experience of necessity rather than the possibility non-experential judgments. The existence of synthetic a priori truths mean that we have some hope of rooting out the categories that underlie our experience. Inner and outer sense can be distinguished by the relative receptivity or activity of the intuition, and we can further postulate that outer objects are real to the extent that they are renderings of intuition according to the categories of experience. The great weakness left unaddressed by Kant is that the objectivity of outer senses is dependent on the objectivity of reason, but if reason must something in itself and outside of human capacities, then there is no greater possibility for us to grasp it than we could hope to grasp any other external objects. Instead I propose that those scholars who wish to contribute to “turning this path into a highway” concentrate their efforts on showing the conditions which make the communicability of reason possible at all.
To the extent that we are to judge Kant by his result, history shows that his work was quite significant. Since the Critique, certain pernicious bits of dogma have become entirely untenable, and philosophy has progressed into new areas. Both the Continentals and the Analytics acknowledge Kant as a great forebear. However, we must also note the irony that what the secure path of the science most reveals is the insecurity of certain solid foundations. Certainty and necessity are no longer what they were. In bringing them within the grasp of human reason, we must also give up the hope of making noumenal inferences that satisfy the demands of speculative reason and instead rest objective existence merely on the possibility of universal assent to an object in appearances. Further, I propose that we must use reason with the knowledge that while the objects of our experience and reason cannot be merely the arbitrary fiat of the individual, neither are they presented without their own viewpoint. This is then the key insight that we can get from Kant, though it perhaps escaped its author: objectivity is not a quality of supposing that we could ever stand without ourselves or even our own cultures and judge them from the void, but it is rather the quality of having brought our intuitions, experiences, viewpoints, and cultures all under the judgment of critique.