The End of Faith Again

Continuing my reading of Sam Harris’s The End of Faith (which I have posted on here and here.)

In the last post I discussed the empiricist epistemology proposed by Harris. This plays itself out in a subtle way in his arguments in Ch. 2. Harris proposes that the relationship between truth and evidence is this:

“Because” suggests a causal connection between a propositions’ being true and a person’s believing that it is. This explains the value we generally place on evidence: because evidence is simply an account of the causal linkage between state of the world and our beliefs about them…

The moment we admit that our beliefs are attempts to represent states of the world, we see that they must stand in the right relation to the world to be valid… As long as a person maintains that his beliefs represent an actual state of the world (visible or invisible; spiritual or mundane), he must believe that his beliefs are a consequence of the way the world is. This, by definition, leaves him vulnerable to new evidence. Indeed, if there were no conceivable change in the world that could get a person to question his religious beliefs, this would prove that his beliefs were not predicated upon his taking any state of the world into account. He could not claim, therefore, to be representing the world at all.

Harris’s understanding of the relationship between evidence and belief is inspired by David Hume. It also rules out any possibility of making an empirical argument for the existence of God. This is because God is not in the world like all other beings; God transcends the world as its Creator and ground. God underwrites all possible states of the world; the very existence of “states of the world” implies God as their source. So there can’t be a state of the world that refutes the existence of God. Nonetheless, contra Harris, God can still be empirically deduced from the world.

To see this, consider a more trivial question to which we might apply the Humean epistemology: Does the game of Monopoly have a creator? Applying Harris’s criteria, we require that any belief about Monopoly must represent an actual state of the game. Our belief about Monopoly must be a consequence of the way Monopoly is. If there were no conceivable change in Monopoly that could get a person to question that it has a creator, this would prove that his belief is not predicated on taking any state of Monopoly into account and is therefore unreasonable.

What sort of changes in Monopoly would lead us to believe that it didn’t have a creator? We may imagine all sorts of rules changes like having the pieces move backwards instead of forwards, having the owner of the property pay rent to the tenant rather than vice-versa, playing with three dice instead of two, etc., etc. If we think about it, we see that rule changes of this sort imply a creator as much as the normal rules do; in fact, it isn’t so much the particular content of the rules but the fact that there are rules at all that leads us to believe that Monopoly has a creator. Similarly, suppose we modified the board. We might change the number of squares, or change the squares to circles, have the squares run through the center of the board, double the number of properties, etc. Again, we see that it isn’t so much any particular configuration of the board that indicates a creator, but the fact that we have a board at all.

The only state of affairs of Monopoly that would lead us to conclude that it didn’t have a creator is if the rules were utter chaos and the board made no sense at all. In other words, we would think it didn’t have a creator only if “Monopoly” didn’t name an actual game but merely a region of chaos. It is the fact that Monopoly makes sense as a game that grounds our belief that it has a creator, not the particular way in which it makes sense. And the fact that Monopoly makes sense is an empirical fact. It is something we learn through the actual experience of Monopoly.

Similarly, it isn’t any particular state of the world or the particular way in which the world makes sense that indicates that it has a Creator, but the fact that the world makes sense at all. Why does the world make sense? It is not self-evident that it must. In fact, there are some philosophical traditions that think the world doesn't make sense, that our sense that it does is really an illusion; the Eastern thought of which Harris is enamored tends in this direction. The standard rational atheist response, however, is to say that the world “just does” make sense and leave it at that. Why the world makes sense is not a question "worth worrying about." But this avoids the question rather than answers it. As I said in the last post, it is to decline to philosophize rather than provide a philosophical answer. Just as Monopoly points to an intelligence as its ground, so does the world point to an intelligence as its ground – the God of Aristotle that is self-thinking, subsistent thought. In Christian terms, the word points to the Word.

One of the strains of modern atheism realizes this and in order to deny God, it denies the ultimate rationality of the world. This is the atheism of folks like Sartre and the deconstructionists. Sam Harris wants to retain rationality but deny God, a very difficult tightrope to walk. He’s got to deny the transcendent reach of reason (in simple terms, reasoning about the world as a whole) while retaining the mundane use of reason (reasoning about the relationship of parts of the world to each other). The difficulty is that the mundane use of reason naturally leads into its transcendent use. Harris wants us to search out and investigate “regularities” in the world that we experience (see the last post), but not to ask about the foundation of those regularities. We are supposed to take those foundations for granted. The regularities “just are.” But it is natural to the human mind not to settle for something that “just is.” It is in our nature to ask why. Any philosophy, like Harris’s atheism, that demands that we stop asking the question rather than answering it will never be philosophically satisfying because it constitutes a denial of our very nature.

Even in his footnote, Harris can barely avoid using the transcendent form of reason in discussing its mundane use. Some regularities betray causal connections between phenomena, others do not and are dismissed as “mere correlation.” What is the difference between regularities that manifest lawful connections and regularities that are mere juxtapositions in space and time? Harris demurs from answering this question, perhaps because its implications might lead him to where he doesn’t want to go. The distinction he seeks between regularities obviously cannot itself be a regularity. It must be something that transcends regularities so that regularities can be classified in its terms. Already we see that our reason must do more than merely notice regularities in experience and draw conclusions about them. It must know transcendent principles in which it can make sense of regularities, including distinguishing authentic lawful connections from mere correlations. These transcendent principles must characterize experience as a whole; in other words, they don’t represent any particular state of the world but ground the very experience of the world itself. We invoke transcendent principles when we use words like substance, unity, change, material, and ideal, among many others (many being yet another one).

The use of transcendent principles in our thinking is so necessary that there is no point in trying to deny them, as Hume attempted to do. A better tactic for the atheist is not to deny the transcendent use of reason but to claim that it is prescriptive rather than descriptive. In other words, our transcendent reason does not draw its basic principles from experience but rather reads them into experience. This is the approach of Immanuel Kant. Aristotle taught that man learns the meaning of substance, unity, change, etc. from his encounter with being in experience; Kant says rather that man imposes his forms of substance, unity, etc. on experience. Such principles transcend experience for both Aristotle and Kant; the difference is that for Aristotle they transcend experience objectively and for Kant they transcend experience subjectively. Obviously, a metaphysical argument for the existence of God is not possible on the Kantian view, since any conclusion we drew would only refer to our own mind and its forms rather than objective reality.

I side with Aristotle rather than Kant on the origin and reach of transcendent reason, but I will not engage that argument here. The point is that, on any reading, simple-minded empiricism of the Humean type is unsustainable. Harris has sent his forces to defend Omaha Beach when Normandy has already been overrun; the battle has long since moved to the Siegfried Line.