Although he wrote influential works on a large number of topics, both inside and outside philosophy, Kant's fame rests largely with his treatment of metaphysics in his monumental Critique of Pure Reason (1781; second edition, 1787). As Kant was well aware, it was one of the most difficult works of philosophy ever written. In an attempt to aid the serious investigator in its comprehension, he published our text, Prolegomena to any Future Metaphysics, in 1783. The Preface to that work places Kant's approach to metaphysics in an historical context.
Kant learned his philosophy in the German university system, which was dominated by the thought of Christian Wolff, himself a follower of Leibniz. Wolffian metaphysics was laid out in the form of a comprehensive system of concepts at the highest possible level of abstraction. Kant himself worked within the framework of that system in his early years as a philosopher, but eventually he renounced it. His "dogmatic slumber" was awakened, he recounted, by his recollection of the arguments of David Hume. Kant believed that if Hume was right, metaphysics is impossible. As he was unwilling to surrender to Hume's skeptical arguments, Kant undertook to place metaphysics on a new footing.
The old basis of metaphysics was twofold, embodied in Leibniz's principles of contradiction and of sufficient reason. The principle of contradiction states that what is contradictory is false, and that the opposite of the false is the true. Leibniz construed this principle as establishing what is possible and what is impossible. In Humean terms, the principle governs relations of ideas only. The principle of sufficient reason states that every fact, existent, or truth has a reason which brings it about and prevents things from being otherwise. For Leibniz, this principle applies universally to what is contingent (possibly true and possibly false). However, in most cases we are ignorant of the specific sufficient reason for whatever we are considering. Hume would describe this principle as (purportedly) applying to matters of fact. [Click here to jump to the corresponding section of Leibniz's "Monadology."]
Kant granted that metaphysics must make use of general principles like that of sufficient reason, but he took Hume to have discovered that such principles have never been adequately justified. Leibniz held that it is known to hold a priori (independently of experience), and Wolff even tried to derive it from the principle of contradiction. But Hume correctly stated that the opposite of every matter of fact is possible, and he concluded from this this that any justification of a claim to a necessary connection could only be based on experience. But experience is as little capable of establishing a necessary connection as is a priori reasoning. At best, experience only builds up our expectations, yielding what Kant called a merely subjective necessity.
Kant agreed with Hume that necessary connections cannot be established on the basis of experience: if they can be established at all, it could only be a priori. But he also agreed that such connections cannot be based on the principle of contradiction. Therefore, if necessary connections are to be established at all, if metaphysics is to be possible, there must be some way to establish necessary connections a priori yet without relying exclusively on the principle of contradiction. This was the task Kant undertook in the Critique of Pure Reason.
There is one other approach to the skeptical problem set by Hume. Perhaps metaphysics should be abandoned altogether in favor of common sense. Hume himself seemed ambivalent between embracing skepticism and laughing it off as revealing the whimsical nature of reason. "Nature is always too strong for principle. And though a [skeptic] may throw himself or others into a momentary amazement and confusion by his profound reasonings; the first and most trivial event in life will put to flight all his doubts and scruples, and leave him the same, in every point of action and speculation, with the philosophers of every other sect, or with those who never concerned themselves in any philosophical research" (An Enquiry concerning Human Understanding, Section XII, Part II).
This was, indeed, the approach of many of Hume's successors (especially in Scotland), a group ridiculed by Kant. Here is what Thomas Reid, who is now taken quite seriously, had to say about metaphysics. "Sensible men, who will never be skeptics in matters of common life, are apt to treat with sovereign contempt everything that hath been said, or is to be said upon this subject. It is metaphysic, they say: who minds it? Let scholastic sophisters entangle themselves in their own cobwebs; I am resolved to take my own existence, and the existence of other things, upon trust; and to believe that snow is cold, and honey sweet, whatever they may say to the contrary. He must either be a fool, or want to make a fool of me, that would reason me out of my reason and my senses" (An Inquiry into the Human Mind, on the Principles of Common Sense, Chapter I, Section VIII). This, Kant solemnly intoned, is the death of metaphysics.
It will be useful at this point to explain how Kant conceived of metaphysics. The Woffian definition, given by Baumgarten, was this: metaphysics is the science of the first principles of human cognition. As a science, its pronouncements should be certain; yet metaphysics has proved to be an endless battleground of conflicting points of view. It should also be systematic; yet it frequently appears as a loose collection of miscellaneous principles. Given these shortcomings, Kant declared that metaphysics is far from being the "queen of the sciences" that she pretends to be.
Leibniz had given metaphysics a pair of first principles: of contradiction and sufficient reason. These he regarded as absolutely fundamental. Moreover, he thought they were fruitful enough that they yield a complete system of the world. Unfortunately, this system is not accessible to the human mind; only the mind of God is capable of using the first principles to their fullest extent. As principles of human cognition they are of limited value.
Kant's metaphysics places particular emphasis on the relation of its principles to human cognition. This meant for Kant that they have their origin in the human mind, independently of experience. The principles of metaphysics are, then, a priori. And it is here that Kant found Hume's investigations to threaten the possibility of metaphysics itself. For they imply that any alleged principle of sufficient reason is based on experience, in which objects of one kind are found to be associated with objects of another. The only contribution of the human mind is its disposition to expect an object of the second kind when an object of the first kind is present.
Because of the chronic shortcomings of metaphysical investigation, made acute by the arguments of Hume, Kant set out to place metaphysics on a new footing. Hume had actually taken the first step toward this end, by correctly understanding the logical structure of causal judgments. The proposition that A causes B is not derived from the mere concept of an A. "When we look about us towards external objects, and consider the operation of causes, we are never able, in a single instance, to discover any power or necessary connnection; any quality, which binds the effect to the cause, and renders the one an infallible consequence of the other. We only find, that the one does actually, in fact follow the other" (Enquiry, Section VII, Part I). The connection is accomplished by an act of the mind: "When we say, therefore, that one object is connected with another, we mean only that they have acquired a connection in our thought, and give rise to this inference, by which they become proof of each other's existence: a conclusion which is somewhat extraordinary, but which seems founded on sufficient evidence" (Enquiry, Section VII, Part II). In Kant's terminology, causal judgments are synthetic, involving an act of the mind which connects the cause and the effect.
Hume's characterization of causal judgments was directly at odds with that of Leibniz, for whom they are (at least ideally) analytic. Wolff had gone so far as to try to prove that the principle of sufficient reason follows from the principle of contradiction. Thus Kant confronted what he thought were two opposite errors: Hume correctly construed causal judgments as synthetic but incorrectly concluded that they are therefore empirical. Leibniz correctely construed the principle of sufficient reason as a priori but mis-classified them as analytic. Kant thought that his lasting contribution to metaphysics was to recognize an overlooked class of judgments, which are at once synthetic and made independently of experience.
Oddly enough, Kant arrived at this insight originally through his investigation of the foundation of mathematics. As we shall see, he found reasons to think that mathematical judgments are synthetic, and he was convinced that they are necessary truths. The mind can arrive at necessary truths, he maintained, only a priori: as Hume had recognized, experience can never yield knowledge of necessity.
Leibniz had contended that, "When a truth is necessary, its reason can be found by analysis, resolving it into more simple ideas and truts, until we come to those that are primary" ("Monadology," Section 33). He gave an example of this by proving that 2 + 2 = 4. Here is a reconstruction of that proof:
Kant's re-interpretation of this kind of argument is that the definitions with which it begins are not analyses of the concepts in question. That four is the sum of three and one is a synthetic truth: four is synthesized from a given three units by the addition of one unit. The units which are combined to form the larger numbers are "intuited," rather than conceptualized. That is, they are given individuals (such as fingers on a hand or dots on a page) rather than abstract or general concepts. This view will be developed in more detail shortly, but in the meantime we can only observe that insofar as this synthesis is said to be a priori, the intuition involved will have to be something quite special.
For a more detailed account of the distinction between analytic and synthetic, a priori and a posteriori, judgments, see my lecture notes for Philosophy 175 .
In the Prolegomena, Kant criticized Hume for having regarded mathematical judgments as analytic. Had he recognized that they are synthetic, he would have known that some synthetic judgments can be made a priori. This, Kant speculated, would have driven Hume to reconsider his claim that the synthetic judgments of cause and effect could only be made a posteriori. Rather than being shipwrecked by skepticism, Hume could have been the first to establish the true foundations of metaphysics.
There is a real obstacle, however, in conceiving how synthetic judgments are possible a priori. If the human cognitive faculty is the origin of a given judgment, then the validity of the judgment would seem to be limited to what lies in the compass of the mind itself. In other words, how could we ascribe objective validity to a judgment a priori? Locke and Hume had held that mathematical judgments are about what Hume called "relations of ideas," and that they need not hold of any object in the external world. If there are real triangles, then the relations of ideas pertaining to ideal triangles would hold for them. But there need not be anything which conforms to our mathematical ideas.
Kant, on the other hand, claimed that mathematical judgments are objectively valid: that they hold for the objects of human experience. For example, geometrical demonstrations concern spatial objects, and we represent real things as in space. Thus, proofs in geometry apply to objects insofar as we represent them the way they do, objects as they appear to us. And appearances, Kant maintained, are the only objects of human knowledge. Thus we can make a priori judgments about the spatial properties of appearances, and those appearances are the things which of which we have sense experience.
Judgments a priori have objective validity because they apply to a restricted range of objects, what Kant called appearances. Geometrical judgments in particular apply to appearances because space is the form whereby we represent physical objects. As mentioned above, Kant called our representations of individual things "intuitions," and his doctrine can therefore be restated as the view that space is the form of sensible intuition. Our intuition is sensible because there is a material given to it, i.e., something which is not the product of any human cognitive faculty. An appearance, then, has both a form and a given matter: both the mind and something mind-independent contribute to sensible intuition.
Space is not the only form of human sensible intuition. We also represent objects as having a location in time, and time is a second form of intuition. In fact, time is more comprehensive than space, since we can represent at least one object as in time but not in space, namely, our own minds. More will be made of this point later. In the meantime, it should be noted that the a priori science of time is arithmetic, which involves the successive addition of unit to unit.
The claim that space and time are forms of the faculty of human sensible intuition has important consequences. Specifically, all the properties which we assign to the objects of experience depend on space and time. Kant considered the two fundamental properties of physical objects to be extension (occupation of a volume of space) and impenetrability (exclusion of anything else from the volume of space occupied). Both of these properties are directly dependent upon space, and without them they would not be possible. Therefore, insofar as space is ideal, bodies are ideal as well. Kant called this view the "transcendental ideality of appearances." To Kant's critics, the view was tantamount to reducing physical objects to mere illusion.
For a more detailed account of Kant's arguments concerning space, see my notes for Philosophy 175 .
Not surprisingly, Kant objected strenuously to the claim that he regarded our sense-experience as illusory. Moreover, he resented the association of his view with that of George Berkeley. In the words of the first review of his Critique of Pure Reason, Kant had offered a "freshened up" version of Berkeleyan idealism.
At first glance, it seems that Kant's critics might have been right. In the Prolegomena, Kant defended himself against the charge of idealism in a most peculiar way. He noted that everyone from Locke to the present had accepted the ancient view that colors, sounds, etc. are qualities which are not really in bodies, but are only ways in which we represent them through sense. If this treatment of the secondary qualities does not impugn the existence of bodies, why should a similar treatment of primary qualities? After all, Kant had never denied bodies exist, only that they have, in themselves, apart from all human representation, spatial and temporal properties.
But at this point, one is reminded of Berkeley's argument: strip an object of all its qualities, primary as well as secondary, and what is there left? A bare X, which is as much as to say, nothing at all. This thought must have occurred naturally to Kant's critics. The problem was exacerbated by a further consideration raised by Berkeley. If the primary qualities are mind-dependent, then we cannot ascribe to the bodies themselves the activity of causing the sensations we have of them. For Berkeley, of course, this causality is assigned to God. Yet Kant wanted to continue to hold that the material of sensible intuition is a causal factor. "I grant by all means that there are bodies without us, that is, things which, though quite unknown to us as to what they are in themselves, we yet know by the representations which their influence on our sensibility procures us, and which we call bodies" (Prolegomena, First Part, Remark II). Thus, as an early critic (F. H. Jacobi) put it, Kant cannot live without things in themselves, since he needs them as material of sensible intuition, but he cannot live with them, since they are unknowable.
In Part One of the Prolegomena, Kant had sought to establish that pure mathematics is possible because its objects depend on space and time, which in turn are supplied by human sensibility. Sensibility is a faculty of intuition: its objects are given individuals. It is distinctive of human cognition that we bring objects under general concepts when we make judgments about them. This is the work of a second faculty, the understanding, which is treated in the Second Part of the Prolegomena.
The question asked in Part Two is how pure natural science is possible. This is indeed a significant question, since a "pure" science would be one whose concepts and principles are products of the understanding itself. The law of inertia ("Every body continues in its state of rest, or of uniform motion in a right line, unless it is compelled to change that state by forces impressed upon it," as Newton put it) contains the empirical concepts of motion. Although it is a principle of great generality, it is not quite pure. On the other hand, it presupposes an even more general law, which states that no change takes place without there being a cause.
The lawfulness of nature was, for Kant, bound up with the very notion of nature itself, and natural science seeks to discover the natural laws. But this formal description of nature as lawful must be supplemented by a material description of nature as the sum total of objects of experience. Natural science, then, must be confined to experience and not attempt to reach beyond it. Its principles are immanent rather than transcendent. So too will be the principles contributed by the understanding itself.
The concept of cause and effect is a pure product of the human understanding (a "category"), and its scope at least seems to extend beyond objects of human experience, e.g., to God's creation (as in Leibniz). Yet Kant was careful to note that such a priori concepts cannot be applied to things in themselves. The reason is that such an application would be synthetic. Mere analysis of concepts reveals nothing about the existence of things and their properties. But synthetic judgements about things in themselves would have to be derived from experience, which conflicts with the a priori character of the concepts. At any rate, Hume had debunked the notion that a concept such as causality can be derived from experience. So the categories apply only to objects of experience, not to things in themselves.
With this description of a pure science of nature in hand, we must again ask how it is possible. How can the understanding supply principles which serve to produce the unified whole we call "nature?" As Hume observed, such principles cannot be derived from perception, since we can observe only conjunction, not connection. Kant went further and held that "judgments of perception" are subjective, reporting nothing more than the perceptions of an observer. I might observe that whenever the sun shines on a stone, it grows warm. But this embodies only my own observation, and does not allow me to ascribe any connection to the objects themselves.
In order to obtain "objective validity" in my judgments, I must go beyond my subjective standpoint and ascribe properties to the objects themselves. Thus I might say that the sun causes the stone to become warm. In this case, I have "superadded" the concept of causality to the judgment of perception, converting it into a judgment of experience. Hume would be unimpressed by this conversion, attributing it only to a feeling of expectation that the stone will be warm the next time the sun shines on it. So how does Kant justify his claim that the superaddition of the concept of causality gives the judgment of experience "objective validity?"
The detailed answer to this question must await our consideration of Kant's justification of the causal principle (every event has a cause) as such. In the meantime, there remains a general question as to what right we have to apply a priori concepts to objects of perception. The question demands a "transcendental deduction" of the applicability of the categories. This is the task Kant referred to in the Prolegomena as "the most difficult task ever undertaken in the service of metaphysics" (Preface). (For a detailed analysis of the argument of the Deduction, see the lecture notes for my Kant class.)
Kant's answer to the question of the applicablity of the categories is very complex. In its simplest terms, it consists in the claim that the application of a priori concepts is what makes experience possible. The very notion of objectivity presupposes them. They are the source of lawfulness, of coherence, which is the hallmark of experience. Locke and Berkeley had appealed to coherence as the basis for our distinguishing from reality and illusion, but they had looked to the real things themselves for the source of coherence. Kant located that source in the human understanding, on the grounds that it cannot be found in perception, but is required even for perception to be possible through the construction of objects by the "transcendental imagination."
Another way to look at the relation of the understanding to experience is in terms of the fundamental activity of the understanding, which is to make judgments, whether of perception or of experience. In so doing, it unites concepts and intuitions, concepts and other concepts, or judgments and judgments. Every judgment has a logical form, and Kant believed that he could enumerate the forms of judgments into a table of twelve, arranged in four groups of three, and related systematically to one another. From these judgment forms he developed his table of categories, with the same structure. (This is called the "metphysical deduction" of the categories, which shows their origin rather than their applicablity.) Finally, with each category or group of categories there is associated a principle of the understanding.
For example, the judgment of perception, "when the sun shines on the stone, the stone becomes warm," is hypothetical in form. From the "If ... then ___" form of judgment we get the pure concept of cause and effect, whose principle is that every change in time takes place in accordance with laws of nature. The judgment, "the stone becomes warm" is of the categorical form, in which a property is predicated of a subject. From this form we get the pure concept of substance and accident. A substance is that which remains the same while its properties change. The principle of substance is that every change is the change of the properties of a substance (which rules out the coming-to-be or passing away of any substance).
Categories are the production of human subjects, and in that sense they are "subjective." However, Kant was willing to redefine "objectivity" as what is valid for consciousness in general, so insofar as the categories are intersubjective, they are objective. The ability of the understanding to generate concepts a priori from its forms of judging cannot be explained, any more than the ability of sensibility to generate space and times can be explained. As Kant noted in Section 36, to attempt to do so would require use of the very faculties which are the objects of investigation. Nonetheless, Kant claimed that neither the representations of space and time nor the categories are derived from experience. If they were so derived, Hume would be right in his claim that they could not yield any necessary connections.
Kant thought that where Hume went astray was in assuming that our perceptions represent things in themselves. If we try to apply concepts generated by the understanding to things in themselves, we would have no reason to think that objects must conform to them. In that case, we could only derive our concepts of objects from perception itself, thus giving up any claim to necessity. The solution is to recognize that objects of perception are only appearances, whose forms (space and time) are the a priori production of the human faculty of sensibility. In that case, it is but a short step to allow that they are also subject to concepts produced a priori by the human faculty of understanding.
With this essential qualification in mind, we can look a little more deeply at how Kant tried to justify the causal principle, which states that every change in an object of experience is the consequence of a law of nature. At times, the claim looks to depend entirely on Kant's use of the term 'experience.' "Experience consists in the synthetic connection of appearances (perceptions) in consciousness, so far as this connection is necessary" (Prolegomena, Section 22). In the Critique of Pure Reason, Kant advanced a more substantive argument, which we cannot pursue here. (For an account of this argument, click here ). Essentially, his claim is that experience at the very least requires that we be able to determine the order in which events occur in time, and that this "time-determination" is impossible without conformity to causal law.
Let us now step back and take a broader view of Kant's project of finding principles of "pure natural science" based on pure concepts of the understanding. A prerequisite of any science is that it advance systematically, rather than piecemeal. Kant criticized previous attempts to produce categorial systems, beginning with Aristotle's, as being haphazard in their approach, even if they did hit upon some or all of the categories. He attempted to produce a system primarily through a process of division, yielding four groups of categories.
The first division is between those "mathematical" categories which concern the way in which objects of perception must be represented in general, and those "dynamical" categories which concern whether and how they exist. The mathematical categories are broken down into those of quantity and those of quality. The former are directly related to space and time, and their principle is that every object of perception is an extensive magnitude, taking up a stretch of time and a volume of space (if it is physical). Thus we know that objects of "outer sense" are extended not by perception, as with Locke, Berkeley, and Hume, but a priori, insofar as space is the form of outer sense.
Space and time are not filled uniformly, but variably, according to Kant. The quantity of gravitational force in the vicinity of the sun is enormous, but in the space between Jupiter and Saturn, it is relatively slight. The categories of quality reflect this variability in the filling of space and time, yielding the principle that they are filled to a degree or "intensive magnitude" which approaches nothing at one extreme and full reality at the other. In making this claim about nature, Kant was siding with philosophers like Leibniz who opposed the possibility of void or empty space. These philosophers (primarily Newton and his follower Locke) held that a space is more or less filled according to the number of discret atoms occupying it. Leibniz had pointed out that there can be perceptions so "small" (petit) that we are entirely unaware of them, so lack of consciousness of an intensive magnitude does not indicate its absence, but only a diminished perception. Thus experiments in which the air is evacuated from a chamber do not demonstrate the existence of a void. In general, Kant believed that nature allows no "leaps" or discontinuities.
The continuity of nature is also reflected in the dynamical categories, which are divided into those of relation and those of modality. The relational categories are substance/accident, cause/effect, agent/patient. In each case, the corresponding principle is one of continuity. As mentioned previously, Kant held that the only change that can occur is a change in the state of an existing thing. Thus there are no discontinuities of existence in nature, no new things coming to be and no existing things passing away. Further, all change is bound by laws of nature, which precludes the discontinuity that would result if change were random.
Finally, all objects stand in a relation of community to one another. The order of the physical universe depends on the interaction of every substance with every other substance. Again, there are no breaks. The gravitation of the sun becomes attenuated by the square of the distance to its center, but no matter how far away another body is, it is acted upon by that gravitation, with gradually decreasing intensive magnitude. (This view is once again reminiscent of that of Leibniz, with the exception that for Leibniz bodies do not really act on one another.)
The realm of nature includes not only extended impenetrable substances (bodies) but also at least one thinking substance (myself). I have a faculty of inner sense which allows me to intuit my "empirical" self as an appearance existing in time. My mental states are intergrated with the rest of nature, in the sense that it is subject to causal laws which involve bodies. I burn my hand and feel pain; I open my eyes and see the sunrise; I desire a snack and head toward the pantry. Kant maintained that all the actions of the empirical self are subject to causal law, which obviously poses a threat to human freedom. We shall return to that topic shortly.
The question of freedom is one of the most vital of those issues which motivates metaphysics. The others are whether I am immortal and whether God exists. Kant held that metaphysics concerns three "Ideas" of objects which lie beyond the pale of experience. Here Kant claimed an advance over previous metaphysical systems which did not separate categories which apply only to experience and Ideas which apply only to what lies beyond experience. These "transcendent" objects (if indeed there are such things) are the soul, the world-whole and God.
To understand the Idea of the human soul, we begin with the category of substance and accident, which applies to the empirical self. My empirical self is that substance whose states are mental states. The self is not a state or property of some other subject, but instead is the logically simple subject to which my mental states belong. However, I know myself only through my mental states, so my empirical self is really an unknown subject of these states. (This claim is very much like Locke's description of mental substance.)
Kant believed that the reference of one's mental states to an unknown subject is destructive of materialism. As we have seen, bodies for Kant are appearances in space and time. As such, they are not suitable for service as the unknown subject that underlies the appearance of mental states, which themselves are not in space at all. As Kant put it in the first edition of the Critique of Pure Reason, "Neither the transcendental object which underlies outer appearances nor that which underlies inner intuition, is in itself either matter or a thinking being, but a ground (to us unknown) of the appearances which supply to us the emprical concept of the former as well of of the latter mode of existence" ("Paralogisms of Pure Reason"). (In the second edition of the Critique, published after the Prolegomena, Kant used a different argument, claiming that the soul is simple but nothing in space is simple.)
To call the soul a substance is to invoke a category that applies to an object of experience, the empirical self. The Idea of the soul, however, is that of an object which has an absolute existence which is independent of its being represented through its temporal states. Thus the defining characteristic of a substance is its permanence, its persistence through change. My thoughts change continually, but I remain the same individual throughout this change. Permanence is a property dependent entirely on time, and apart from time it is meaningless. Time exists only in the context of experience, so it is empty to call the soul apart from experience a "permanent" object.
This, of course, has never stopped philosophers from attempting to prove that the soul is immortal. Kant refers in the Prolegomena to "a very specious argument, which infers the nature of the soul from this supposed cognition of the substance of our thinking being" (Section 46). The argument can take several forms, but the following is perhaps the most accessible.
A traditional argument for the immortality of the soul is that mortality depends on dissolution of parts and the soul has no parts. But how does one establish the simplicity of the soul? One might appeal to the simplicity of the representation of myself. Every act of my thinking X is referred to a single "I" as in, "I think X." Thus there is a logical subject which is unitary or one. Finally, the unity of the "I" indicates the simplicity of the thinking substance. It is at this last step that the argument falters. If we understand the unity of the subject as logical rather than metaphysical, then the proposition that I am one is true, but lacking in any metaphysical consequences. The problem with the argument is that it contains an equivocation on the word 'unitary,' and because of this it is a paralogism or specious argument. (For more details, see the lecture notes for my Kant class.)
Another specious argument attributed by Kant to transcendental psychology does not concern the nature of the soul so much as the relation of the soul to the body. If the soul is a thing in itself and bodies are things in themselves, then it is impossible to be sure that bodies exist. The existence of bodies could be known only through an inference from effect (perceptions) to bodies as their cause. But insofar as there could be other causes, there is always room for doubt about whether there really are bodies. For example, our perceptions which we refer to bodies may only be dreams. A "material" idealist for this reason refuses to allow the existence of bodies.
If bodies are only appearances, however, they can be known because their reality depends on their being intuited in space, where space is the form of outer intuition. To say that bodies exist "outside us" just means that they exist in space, and therefore they are dependent for their form on the human faculty of representation. Kant calls his own idealism "formal" for this reason, and he claims that his formal idealism abolishes material idealism. The question as to whether bodies are dreams can be decided on the basis of the coherence of our perceptions. But if bodies were things in themselves, then the fact that our perceptions are coherent implies nothing about whether they refer to real things.
Moreover, we know bodies to exist just as surely as we know ourselves to exist, since both bodies and the empirical self are appearances. We know bodies in space through outer sense and ourselves in time through inner sense. Descartes and others had mistakenly assumed that we know ourselves better than we know bodies, and so they could question whether bodies exist while remaining quite certain of their own existence. Kant put our selves as beings in time and bodies as beings in space on a par, such that one cannot question the existence of bodies without bringing into question the existence of the self.
Descartes's alleged idealism was "problematic," in that it questioned the cogency of the inference from sense experience to the existence of bodies. Berkeley, on the other hand, was a "dogmatic" idealist, who was interpreted by Kant in the First Part, Remark III, as having been "mystical" and "visionary," turning mere representations into things. In the Appendix to the Prolegomena, Kant made a more detailed charge against Berkeley. Berkeley's idealism is mystical because it seeks the truth only in objects of the understanding. Bodies are illusory for Berkeley, because his view lacks the resources to distinguish the true from the illusory in what comes before the senses. Specifically, Berkeley took space and time to be empirical representations, not representations a priori. Experience does not reveal any necessary truths, so that Berkeley could not distinguish between what is subject to necessary laws of nature (bodies) from what is the product of our imagination.
We move now from the metaphysics of the soul to that of the physical universe. What distinguishes scientific thinking from metaphysical thinking that the latter seeks to go beyond what can be reached through experience and make pronouncements about the totality of the universe, the cosmos. Kant believed that there is general tendency in rational cosmology to divide into two opposing camps. Both sides allow that in cosmology we begin with what is limited and try to discover what is unconditioned.
For example, bodies limit one another by occupying the adjoining space. But does this process come to an end, or is it repeated infinitely? Is the universe as a whole finite or infinite? The view that it is finite has the advantage that reason finds a resting point in moving from one limited body to another. Bodies at the end of the universe are not limited by anything else. But the view that it is infinite has the advantange that there is no discontinuity: each body is the same as every other, in that each one is limited by another body.
This opposition sets up a set of "antinomies" or conflicts of reason with itself. Each side argues that the opposing position is untenable. On the side of the "thesis," the argument is that if we assume that every body is bounded by yet another body, the totality of the universe will be unattainable, which cannot be allowed by metaphysics. On the side of the "antithesis," the argument is that if we assume that some body is bounded by no other body, then we would be granting that it is bounded by an empty space. But an empty space is not an object which can come before experience, in which case the totality of the universe would again be something which could never be attained.
The solution to the antinomy in this case is to give up the idea of a totality of bodies. This is what is prescribed by transcendental idealism, the doctrine that bodies are nothing more than appearances represented as in space. Under this assumption, both the thesis and antithesis turn out to be false. The world is not infinite, because an infinite totality cannot be reached through experience (here the thesis argument is correct) and it is not finite because an empty space is not an object of experience (here the antithesis argument is correct). But the conclusion to be drawn is that the world is indefinite in its extension. There is always more to experience, but there is not an infinite whole that can be reached. This conclusion can also be applied to the second "mathematical" antinomy concerning the divisibility of bodies. Division is also a process that can be carried on indefinitely only. (For more on the mathematical antinomies, click here.)
A second class of antinomies is "dynamical," having to do with the conditions under which something comes to exist. Every event is limited by being subject to a prior cause. The thesis argument claims that if we assume that there is no end to the chain of causality, reason will never be satisfied because the totality of the chain of causes will be unattainable. The antithesis argument claims that if we assume an uncaused cause, we destroy the continuity of experience, undermining the a priori principle of the understanding that every event has a cause.
The solution to this conflict is to allow the possibility that both sides are right, so long as they are concerned with different ways of representing the universe. As apperances, all events are limited by other events as their causes. The antithesis argument is correct when applied to appearances: a discontinuity there is not allowed. But considered as things in themselves, what appears as subject to causal laws may be an uncaused cause. The principles of the understanding do not extend to things in themselves, so there is room for what Kant called "transcendental freedom." He believed that there is a causality different from physical causality. Human agents act according to reasons: they (sometimes) do what they ought to do. The same human action which as appearance is subject to physical causality may at the same time be the product of a rational will, which is a thing in itself and therefore not subject to any other causality. In this way, Kant thought he had shown how ethics, which depends on transcendental freedom, is possible. (For more on the dynamical antinomies, click here.)
The final Idea of reason is that of God, an object lying beyond all experience whatsoever. Philosophers have through the ages engaged in rational theology, the attempt to prove the existence of God on the basis of rational argument alone. Revealed theology, which depends on testimony, is outside the bounds of philosophy. Within rational theology, Kant distinguished two very different approaches. A "theist" attributes to God characteristics which are analogous to human reason and will, while a "deist" operates with a purely abstract conception of God.
Rational theism is usually based on the observation of nature. The order of the natural world cannot be understood, it is argued, unless it is the product of an intelligent design brought about by a powerful will. Although Kant regarded the argument from design as the most natural argument for God's existence, he recognized that it is severely limited. Hume had pointed out that there are many ways of accounting for the order of the universe, most of which are quite inadequate to any recognizable conception of God. At the end of his Dialogues Concerning Natural Religion, one of Hume's characters stated the only permissible conclusion to be drawn from order in the universe. "The cause or causes of order in the universe probably bear some remote analogy to human intelligence."
Not only is the analogy between humanly designed objects and the order of nature too remote to support a conception of God, it also depends on the premise that God is the cause (author) of nature as well as its designer. Thus it relies on a purely conceptual description of God as a necessary being, the existence of the universe being contingent. But even this "cosmological" conception of God is inadequate, because we need to find an object which could exist necessarily. Here theology becomes even more abstract, pronouncing God to be the "most real being," or "the being of beings." The ontological argument states that such a being exists necessarily. Kant was famous for his criticism of the ontological argument, but it did not find its way into the Prolegomena.
Given Kant's rejection of the arguments from design, the contingency of the universe, and the existence of the most real being, it would seem that Kant would have to throw religious belief back on revelation. But he thought he could salvage rational theology after all. First, he held that there is a successful theistic argument for God's existence, as a necessary condition for morality. Second, he noted that even if the analogy between human beings and God does not allow us to attribute reason and will to God objectively, we can still consider the world as if it were created by an intelligent being. There is no harm in so doing, and in fact it can be quite useful in the organization of our knowledge. (For more on Kant's treatment of rational theology, click here.)
End of lectures on Kant.