Before wrapping up the exposition of this central line of argument in the Critique by stating the full-scale argument in quasi-formal form, let me touch on one matter that was passed over. It actually relates in a way to the section of the Critique called "The Schematism of the Pure Concepts of Understanding," though Kant actually thinks he is doing something else in that section. Strictly speaking, the rule which Kant calls the Category of Causality and Dependence is not a rule for reproducing the manifold of perceptions in Imagination. If it were [as may have occurred to some of you], then we would know the laws of nature a priori, and not require experiment or observation to arrive at them. Kant does not really believe that. Rather, they are what we might call second-order rules, or rules for forming rules.
Think of it this way. The rules of chess, of monopoly, of bridge, and of mahjongg are rules that define a game and determine how many players there are, what the legal rules are, when someone has won or lost, and so forth. Now imagine someone who gets a job at a company that makes and sells board games. She is told to create games with the following characteristics: They must be games for two players, suitable for children seven to ten, playable in under one hour, and [very important] absolutely requiring the patented game package made by the company to be played. There are [the president of the company hopes] many games that can be invented obeying these rules, but he can know a priori that they will all re two-person games and will all require his company's game set.
In something like the same way, all the possible causal laws that can be formulated on the basis of observation and experiment conform a priori to the requirement that one event follow upon another necessarily in conformity with a general rule. Incidentally, for you Hume fans out there, exactly the same distinction between rules and rules for formulating rules crops up in the Treatise in the form of the distinction between dispositions and propensities to form dispositions. [See my paper, "Hume's Theory of mental Activity," on box.net.]
But, you may ask if you are really paying attention, how can we know, in advance of experience, that we will actually encounter, in our sense experience, patterns of what Hume calls constant conjunction of resembling instances on the basis of which causal laws can be formulated? Kant calls this question "the affinity of the manifold," and he devotes a good deal of attention to trying to show that his doctrine of pure intuition offers an answer to it. This is one of the many things in the Critique that I am going to pass over, in order to keep this Introduction within manageable limits. The time has come to pull all of this together and state
And now to the final form of the argument of the Deduction, folding in the materials from the Second Analogy. Here it is:
To Prove: There is an objective order of events, and everything in it which happens, that is, begins to be, presupposes something upon which it follows according to a rule. [Notice that this is as proof only of the Law of Cause and Effect, not of the other things Kant claims, somewhat unsuccessfully, to have proved as well.]
1. All the contents of my consciousness are bound up in a unity. [Premise]
2. The only way to introduce synthetic unity into a manifold of contents of consciousness is by reproducing them in imagination according to a rule.
3. The defining mark of objectivity is necessity of connection. [I have not talked about this, but it should be clear from what I have said.]
4. Synthesis, i.e., reproduction in imagination according to a rule, confers necessity of connection on a manifold. [This was the point of the house/boat example.]
5. If all the contents of my consciousness are bound up in a unity, then they have, qua representations, an objective order. [2, 3, 4, substitution and conversion.]
6. The contents of my consciousness have, qua representations, an objective order,
which is to say
there is an objective order of events. [1, 5, modus ponens]
7. The form of inner sense is time, and therefore all the representations of my consciousness, considered simply qua mental contents, must be arranged in a temporal order. [additional premise]
8. But since these representations must be reproduced in imagination according to a rule before they can be admitted to the unity of consciousness [step 3], they must have a rule-determined time-order which is the order of their reproduction. [from 7]
9. Thus, any mental content, in order to be treated as a representation with objective reference, must be reproduced in a temporal sequence of representations according to a rule. [3, 4, 8]
which is to say
Everything which happens, that is, begins to be, presupposes something upon which it follows (temporally) according to a rule, and, there is an objective order of happenings (events).
Q. E. D.
We now reached the end of the connected line of exposition by means of which I hoped to explain in simple language the central argument of the Critique. There is a vast amount of fascinating and important philosophy remaining in the work on which I have not touched: the Refutation of Idealism; the analysis and refutation of the proofs for the existence of God, including the best known of all, the refutation of the Ontological Argument; the Antinomies, with their Kantian version of Zeno's Paradoxes, and much, much more. As I hope I have made clear, my goal has been to encourage all of you to read -- or to re-read -- the Critique, perhaps finding it easier to master as a consequence of this Introduction.
Before I close this series of posts, do a bit of editing, and place the entire text on box.net, where it will live forever in cyberspace, I would like to discuss one fundamental problem for Kant's ethical theory that grows directly out of one of the very last things I have been talking about, viz. the relation of the empirical self to the world of physical objects. It is worth noting that very few commentators on the philosophy of Kant write about both his epistemological theories and his ethical theories. When I was a student, the only scholar with whom I was familiar who and written full length books on both aspects of Kant's philosophy was the Englishman H. J. Paton. For a long time -- conceivably it is still true, I have not checked -- I was the only American commentator to do so.
The question of the relationship between the Critique of Pure Reason and Kant's ethical theory [most notably, the Groundwork of the Metaphysic of Morals] is important because Kant wrote his ethical theory without taking into account the deeper position at which he had arrived in the portions of the First Critique that I have been explaining. In particular, he did not. when writing his ethical theory, take into account his new teaching that concepts are rules for the synthesis of a manifold of sensibility. Nor did he keep before his mind the radically subjectivist implications of the theory of synthesis put forward in the Subjective Deduction. Now, one might of course argue that since Kant omitted the doctrines of the Subjective Deduction from the Second Edition, we may conclude that they are not really a part of his considered mature teaching. But I think that would be a mistake, and perhaps I ought to take a moment to explain why.
To put it simply, Kant set out to answer Hume's scepticism about the knowledge claims of Newtonian physics. The core of his response is the theory of a priori synthesis according to the Categories. But that theory is hopelessly metaphorical and incomplete without an explanation of the nature of synthesis and its connection to the categories, and the Subjective Deduction is the only place in all of Kant's writings where he gets past the metaphors and actually tells us what this activity of synthesis is. As we have seen, if we take that account seriously, we can actually reconstruct a coherent argument that takes us from his premise, "The 'I think' can attach to all my representations" [i.e., the unity of consciousness] to his conclusion, that Everything which happens, that is, begins to be, presupposes something upon which it follows (temporally) according to a rule, and, there is an objective order of happenings (events). So I have, in reading Kant, made the deliberate interpretative choice of taking seriously the teaching of the First Critique, regardless of the problems that poses. As I indicated on the very first day of this seemingly endless series of posts, engaging with a philosopher like Kant [if indeed there really is anyone else "like Kant"] requires such choices, and the only thing one can do is to make them, make it clear what they are, and hope to find readers or listeners willing to make the same choices, at least provisionally to see where they lead. I think this is what literary critics call a "strong reading" of a poet.
Kant is, as everyone knows, a deontologist in ethics. Indeed, one might say he is the deontologist in ethics. So no one would mistake him for a situationalist, let alone a utilitarian. Now Kant thinks that the Moral Law [which we fallible creatures experience as a Categorical Imperative] requires that I treat persons as ends in themselves, not merely as means to my ends. [Never mind that that is actually an incoherent, albeit admittedly a deeply moving, notion. Anyone who wants to see what I have to say about that and the rest of Kant's ethical theory can look up and read The Autonomy of Reason: A Commentary on Kant's Groundwork of the Metaphysic of Morals.] So whatever you may think of the details of Kant's ethical theory, and a lot of ink has been spilled on that topic, one thing is surely clear. Kant thinks that I encounter persons. If I cannot encounter persons, then the injunction to treat them as ends in themselves is vacuous. So are the obligations to be truthful, to injunction against lying, the obligation to keep my promises [to whom am I making a promise if not to another person?], and so on.
Just who are these persons I am assumed to be encountering, and to whom I owe this, that, and the other obligation? Things now get rather complicated. Let me start with what you might suppose is a somewhat easier question. Who am I, ethically speaking? Remember that for Kant, the I we are talking about when we talk ethics has to be capable of rational freedom, or what Kant calls the exercise of Practical Reason. Pretty clearly, that I is not the empirical self [a.k.a. Robert Paul Wolff, born 27 December 1933 and due to die on some as yet undisclosed date.] That I is governed by the laws of physics in just the way that every other phenomenal object in the realm of Appearances is. That I's behavior in the realm of appearance is as predictable, if only we had enough information, as the behavior of that old standby of 18th century epistemology, billiard balls.
To cut to the chase, and squeeze into one sentence about two chapters worth of textual explication, the empirical self, determined like all other phenomena by the laws of nature, is the appearance in the realm of phenomena of the I of Practical Reason, the I bound by the Moral Law, the Noumenal Self, the Self in Itself, the Self that is also the locus of the Forms of Intuition and Conception, the Self that engages in the activity of synthesis, the Self whose Transcendental Unity of Apperception appears in the phenomenal world as the unity of consciousness.
But this poses an enormous problem, of which Kant appears not have been aware. Tomorrow we shall explore that problem.