Synthetic A Priori Knowledge

Transcription

1 Synthetic A Priori Knowledge In the Introduction to the Critique, Kant tells us that his task will be to explain the possibility of synthetic a priori knowledge. Before we can talk about why this task is philosophically important, we have to explain the terminology. What we have here are two terms, synthetic and a priori, both of which must be understood in contrast to its opposite. So, we have two distinctions to clarify, that between analytic and synthetic, and that between a priori and a posteriori. In Kant s terminology, analytic and synthetic describe different kinds of judgments. Judgments, for Kant, are simply statements, or assertions. But before we can look at what Kant actually says, we have to note that Kant was trained in Aristotelian logic. Logic has changed some since Kant's day, and so we need to do a little "translating" in order to make what he says fully intelligible. In the logic Kant took for granted, all statements were understood to have a subject/predicate form, and so his discussion of the difference between analytic and synthetic statements (again, he calls these judgments ) took this for granted. Here is what he says (this is from the Kemp-Smith translation, which I have a digital copy I can cut and paste out of): In all judgments in which the relation of a subject to the predicate is thought (I take into consideration affirmative judgments only, the subsequent application to negative judgments being easily made), this relation is possible in two different ways. Either the predicate B belongs to the subject A, as something which is (covertly) contained in this concept A; or B lies entirely outside the concept A, although it does indeed stand in connection with it. In the one case I entitle the judgment analytic, in the other synthetic. Analytic judgments (affirmative ones) are therefore those in which the connection of the predicate with the subject is thought through identity; those in which this connection is thought without identity should be entitled synthetic. The former, as adding nothing through the predicate to the concept of the subject, but merely breaking it up into those constituent concepts that have all along been thought in it, although confusedly, can also be entitled explicative ( judgments of clarification ). The latter, on the other hand, add to the concept of the subject a predicate which has not been in any wise thought in it, and which no analysis could possibly extract from it; and they may therefore be entitled ampliative ( judgments of amplification ). If I say, for

2 instance, 'All bodies are extended', this is an analytic judgment. For I do not require to go beyond the concept which I connect with 'body' in order to find extension as bound up with it. To meet with this predicate, I have merely to analyze the concept, that is, to become conscious to myself of the manifold which I always think in that concept. The judgment is therefore analytic. But when I say, 'All bodies are heavy', the predicate is something quite different from anything that I think in the mere concept of body in general; and the addition of such a predicate therefore yields a synthetic judgment. [A7/B11] So, analytic statements really tell us nothing new about the world: they simply assert of some thing a predicate that is already contained in the subject as part of its meaning. They are analytic in the sense that the predicate can be found by analyzing the meaning of the subject. They merely explicate or clarify information that is already contained in our concepts of things. All bodies are extended is analytic because (Kant believes) part of the definition of a body is that it occupies space (which is what extended means here). This statement clarifies what we already understand a body to be, but does not tell us anything more about them. Synthetic statements, on the other hand, assert more of a thing than is already contained in the definition of that thing. They synthesize one concept with another, and tell us that they are found together in some specific thing. They are ampliative in that they convey substantive information about a thing, information that is not contained in the very concept of that thing. All bodies are heavy is synthetic because having weight is not part of the definition of occupying space. While we cannot conceive of an unextended body (any more than we can conceive of a married bachelor), we can conceive of a spatial object that has no weight. Perhaps it is false that there are any weightless bodies. But the claim that there are is not contradictory. This last point provides us with a way of understanding the analytic/synthetic distinction that is not limited by the logic Kant inherits from Aristotle: The negations of synthetic statements are not contradictions, while the negations of analytic statements are. This captures what Kant says (indeed, he often uses just this terminology) without committing us to any particular understanding of the logical form of statements. And it helps bring to the fore something of the philosophical questions we are approaching. I know that an analytic statement is true (and must be true) because to deny it would be to affirm a contradiction. But the same is not true for synthetic statements. To know that a synthetic statement is true (to be justified in my belief that it

3 is true), I need something more than an understanding of the meanings of the words involved. What Hume and other empiricists tell us is that this something more can and must, in the end, be some sort of sense experience. To understand the question Kant is asking, we must now look at the distinction between a priori and a posteriori knowledge. While analytic and synthetic refer to different kinds of statements, a priori and a posteriori refer to different kinds of knowledge. The key issues, as we shall see, is not how we get some belief, but rather in how we know that it is true. The issue is one of justification, not of origin. Consider how Kant introduces these terms: There can be no doubt that all our knowledge begins with experience. For how should our faculty of knowledge be awakened into action did not objects affecting our senses partly of themselves produce representations, partly arouse the activity of our understanding to compare these representations, and, by combining or separating them, work up the raw material of the sensible impressions into that knowledge of objects which is entitled experience? In the order of time, therefore, we have no knowledge antecedent to experience, and with experience all our knowledge begins. But though all our knowledge begins with experience, it does not follow that it all arises out of experience. For it may well be that even our empirical knowledge is made up of what we receive through impressions and of what our own faculty of knowledge (sensible impressions serving merely as the occasion) supplies from itself. If our faculty of knowledge makes any such addition, it may be that we are not in a position to distinguish it from the raw material, until with long practice of attention we have become skilled in separating it. This, then, is a question which at least calls for closer examination, and does not allow of any off-hand answer: -- whether there is any knowledge that is thus independent of experience and even of all impressions of the senses. Such knowledge is entitled a priori, and distinguished from the empirical, which has its sources a posteriori, that is, in experience. The expression 'a priori' does not, however, indicate with sufficient precision the full meaning of our question.. In what follows, therefore, we shall understand by a priori knowledge, not knowledge independent of this or that experience, but knowledge absolutely independent of all experience. Opposed to it is empirical knowledge, which is knowledge possible only a posteriori, that is, through experience. [B1-3] I think the first thing we must note here is that Kant begins the entire discussion by claiming that all knowledge begins with experience.. In the order of time, we have no

4 knowledge antecedent to experience. So, whatever he means by a priori, he is not talking about anything like innate knowledge. Once again, the key question here is not how we come to have a certain belief (or how we acquire the concepts that compose that belief), but the manner in which can justify our claim to know that this belief is true. Some claims, Kant is telling us, can be known to be true independently of any information we receive through the senses. This constitutes a priori knowledge. Other claims require justification through appeal to what we have learned through sense experience. This is the nature of a posteriori knowledge. The paradigm examples of a priori knowledge involve mathematical statements. Consider = 4. I know that this statement is true, but my knowledge is not dependent upon sense experience. Yes, I acquired this belief at some point in my life because I heard what some teacher was telling me or I saw what was written in some book. But that is not how I know that it is true. (I have read and been taught in school many things I later learned to be false!) I know that 2 +2 = 4 by (using a phrase from Descartes) the pure light of reason. There are many arithmetical truths I know to be true that I have simply never directly experienced (I m pretty sure I ve never collected 8 groups of 8 objects and counted the resulting collection to confirm that I got a total of 64), and, were my sense experience ever to (apparently) contradict some arithmetical truth (suppose I counted only 63), I would question the veracity of my senses rather than my knowledge of the arithmetical truth. So, I have a priori knowledge of some statement when the justification for calling this knowledge does not rely upon sense experience. When the justification depends upon information from the senses, we have a posteriori (or, as Kant calls it in the above passage, empirical ) knowledge. I began my discussion of a priori knowledge with a discussion of mathematical statements (because these are both familiar, and, as we ll see, integral to Kant s discussion), but we already had before us an entire class of statements that Kant (and most others) say that we know a priori: and these are analytic statements. Analytic statements, we saw, don t say anything substantive about the world, and so could not be disconfirmed by our sense experience of that world. So, for example, if there is a bachelor in the next room, I know, without needing to consult my senses, that this individual is male. Whether or not this individual has red hair, on

5 the other hand, is not something I can know independently of experience. So my knowledge of some individual bachelor that he is male is a priori, while my knowledge of this same individual that he has red hair is a posteriori. So, again, Kant thinks that analytic statements are all known a priori. (If someone told me, for example, that they needed to conduct a test or perform a survey to determine whether nor not all bachelors were male, I would infer either that they did not understand the words, or that they were seeking a government grant!) The question is whether or not a priori knowledge is limited to analytic statements. This is, in effect, what Hume said: what we can know on the basis of reason alone are only statements whose negations are contradictory. All other statements can be known only on the basis of sense experience. This is the core of Hume s empiricism. But Kant thinks there are some synthetic statements that we know a priori. And he thinks that even Hume (in the end) would have to accept that we do, indeed, have some synthetic a priori knowledge. For, as we will see next, Kant claims that the mathematical statements we considered above as paradigm cases of a priori knowledge are, in fact, all synthetic statements. So, unless we are willing to call all of arithmetic into question (something Kant did not think Hume would be willing to do), we will need to explain how, in the face of Hume s arguments to the contrary, synthetic a priori knowledge is possible after all. Hume didn t use Kant s terminology, but he did effectively say that we can have a priori knowledge only of a limited class of statements--statements whose negations are contradictions. All other kinds of statements can be known only on the basis of sense experience. The problem is that sense experience is insufficient for justifying many of the claims that philosophers (among others) have been wont to make. Hume s explicit target is traditional metaphysics, as practiced by (what we now call) rationalist philosophers. Metaphysics, as a discipline, seems to be defined as a set of substantive claims (i.e., synthetic statements) that are purportedly known by reason alone, and not on the basis of sense experience. Hume s conclusion is that all such work is mere sophistry, and that it should be committed to the flames. But Hume s attack also extends to all of what we might call laws of nature --to physics, as well as to metaphysics. His direct target is often All events have a cause, or Causes necessitate their effects. But all events includes events that have never and will never be

6 experienced, and necessitation is never directly observed. So, no amount of experience could justify the claim that all events, even the unexperienced ones, have a cause. And when I observe any specific cause/effect relationship (such as one billiard ball striking another), all I actually see is one event followed by another: I never see that the seconded event couldn t not have happened. All I see is that it did happen. And, of course, this same type of criticism applies to all of what we call laws of nature. These laws always describe universal and/or necessary connections. But no amount of experience could justify claims about things (or connections between things) that have never and will never be experienced. So, Hume s criticism not only commits to the flames discussions of how many angels can dance on the head of a pin, but whether or not all bodies attract one another according to the inverse square of their distance. So, if Hume is right, then there is no such thing as synthetic a priori knowledge. Yet Kant seems to beg Hume s question by posing his challenge as explaining how such knowledge is possible. He does this because he thinks that Hume himself would have been forced to concede the existence of synthetic a priori knowledge, if only he had recognized that arithmetical statements were synthetic. Hume claimed that we know arithmetical truths on the basis of reason alone, independently of sense experience, but he didn t recognize, Kant thinks, that such statements (or at least many of them) are synthetic rather than analytic. Kant claims that arithmetical statements are synthetic. Much of his discussion of this relies heavily on his understanding of intuition, which is a topic that isn t really explained until later in the Critique. There is lots of secondary literature on what Kant has to say here, and much of it takes a dim view of Kant s arguments. But I think that, as interesting as all of this literature can be, it misses the crucial point: all Kant needs to establish here is that the negations of arithmetical truths are not (themselves) contradictions. There is more to the story than I want to discuss here, but it has been understood, since at least the beginning of the 20th century, that, like geometry, arithmetic rests upon a set of basic truths, axioms, from which other truths can be derived. But these axioms, like the postulates of geometry, are substantive claims whose negations are not contradictions. In a word, the axioms of arithmetic are synthetic, at least given the understanding of analytic and synthetic that I have adopted. So if Hume s attack does

7 not extend to arithmetic and geometry, then there are indeed examples of synthetic a priori knowledge. Kant s project is to explain how this is possible. And his answer, as we will see, is that we can a priori knowledge of synthetic statements only when these statements reflect fundamental truths about the conditions of the possibility of our conscious experience of objects. With this, the quest to understand synthetic a priori knowledge leads to an attempt to understand the very essence of consciousness of objects.