What is synthesis? It is, Kant says, a "running through and holding together" of a manifold of sensuous intuition in one concept. But what on earth does it mean to run through and hold together a manifold of intuition? How does that running through and holding together create the unity of consciousness which, Kant claims, is the starting point of all philosophical analysis? And what is the connection between the bringing into being of a unity of consciousness and the applicability a prior to experience of the Law of Cause and Effect on which the distinction between the empirically real and the empirically ideal is grounded?
These were the questions with which I wrestled in the summer of 1956 when, as a twenty-two year old graduate student, I worked to finish writing my doctoral dissertation. My dissertation was a comparative study of the role of something I called "mental activity" in Hume's Treatise of Human Nature and Kant's Critique of Pure Reason, and oddly enough, inasmuch as Kant's great work is generally understood as an answer to Hume, my success in answering the questions about synthesis and reconstructing Kant's answer to Hume was due in part to ideas I found in Hume's text itself.
This is a very long story, one that I have unfolded in detail in Kant's Theory of Mental Activity, so I am going to cut to the chase in this Introduction and recount my conclusions, without relating the complex process I went through to arrive at them. I cannot urge you too strongly first to read Kant's text itself, and then to read my Commentary, if you really want to understand what is going on in the Critique.
The key to the answer to the question What is synthesis? turns out to be a single sentence in the so-called Subjective Deduction. "All knowledge demands a concept," Kant writes, though that concept may, indeed, be quite imperfect or obscure." Then he continues:
"But a concept is always, as regards its form, something universal which serves as a rule." [A 106. Notice that there is no corresponding passage in B, since Kant completely deleted the entire Subjective Deduction from the Second Edition.]
A concept is always, as regards its form, something universal which serves as a rule. What is it about the notion of a rule that is the key to an understanding of synthesis, and through that understanding, of the entire Critique? Perhaps, it occurred to me fifty-five years ago, if I unpack the notion of a rule, I will be able to reconstruct Kant's central argument.
Now, a rule is a set of instructions for doing something [or for making something]. What are the distinctive characteristics of a rule-guided or rule-governed activity of any sort? [Never mind for the moment about the obscure mental activity called synthesis. We shall return to it soon enough.] Let us choose a simple example of a rule-guided activity, and see whether we can tease out its significant features. As an homage to my older son, Patrick, the chess Grandmaster, let me choose the activity of playing chess.
Suppose you are showing a friend around the local rec center, and he wants to know what is happening. You tell him that there is a chess tournament in progress. He has no idea what chess is, and asks you a series of questions in an attempt to understand what he is looking at. "Are all of these people playing?" he asks [there are fifteen people in the room.] "No," you reply, "only those two sitting at the table over there." "So the other people are not part of this thing called 'chess"?" "No, they are just watching the game." 'That young woman sitting at the table just drank some water from a glass. Is that part of chess?" "No, she is simply thirsty." "The man sitting there with her just scratched his nose. Is that part of chess?"
And so it goes. As a proud father who spent many long hours attending chess tournaments at which my prodigy son played, I can tell you that quite a long time can go by before either of the people "playing chess" does something that is actually part of the game. The point, of course, is that in order to sort out everything one observes into "events or objects that are part of chess" and "everything else," you need to know the rules of the game. Those rules tell you three things, all of which are essential to our understanding of the role of the Categories in synthesis.
First, the rules of chess tell us just which events or objects belong to the game of chess. Now, these events and objects are a disparate collection -- some small pieces of wood or ivory or plastic, shaped and colored in certain ways; a square piece of wood or cardboard or plastic with an array of different colored squares on one side; perhaps a little double clock, controlled by two plungers on its top side; the activity of picking up one of the little shaped pieces and moving it from one of the squares to another; and even, in some cases, the utterance of certain words, in certain special circumstances, such as "check" or "mate." Each of these objects or activities can be considered as a unity, and all of them together constitute a considerable number. And as Hume so elegantly points out, "betwixt unity and number there can be no medium." But this number, or multiplicity, or diversity, or manifold of unitary events and objects acquire a unity by virtue of their common relationship to a single rule, namely the rule that defines the game of chess.
The rules of chess also tell us which moves are in accord with those rules and which are not, which objects are a part of the game and which are not. As every parent knows, it is a major moment in the development of the child when she first grasps the concept of a game. Until that moment, there is no point in saying, "No, no. You can't put that piece there." She simply will not understand what it means for a something to be, or not to be, in accordance with the rules of the game. Notice that, for our purposes, we are not here interested in the distinction between a good and a bad move, but simply in the distinction between a move that is in conformity with the rules and an action that is not so in conformity, and hence is not, in the strict sense, a move at all. [Irrelevant aside: Herein lies the core of the debate between H. L. A. Hart and Lon Fuller about the nature of law.]
Finally, a rule tells you when an activity starts and when it has been completed. To change the example, at big tennis tournaments, the two players usually come out and hit the ball back and forth, warming up, before the match begins. If you do not know the rules of tennis, you may mistakenly think that those volleys are part of the game, since each of them may in fact look exactly like a volley that, as we say, "counts." Similarly, at tournaments like Forest Hills and Wimbledon, the winner often will bat a ball into the stands after the last point has been played, but those of us who understand the rules know that that is not a an errant swing that costs the player a point.
By now, you will perhaps have begun to suspect that the categories are going to be rules for conducting the activity of synthesis. Tomorrow, we shall find out just what this act of synthesis is, and in precisely what way it is guided by the rules that Kant calls Categories.