Last November I took in interest in the difference between philosophical truth and probabilistic certainty. Reading John Henry Newman’s Grammar of Assent has recalled me to this theme, and to the problem of how the common-sense man (and the Christian) can find a middle way between the probabilist and the philosopher.
Socrates, in the Phaedo, says that to practice philosophy is to prepare for death. This means, most obviously, that the philosopher seeks freedom from worldly attachments and entrance into eternity. But it means, also, that he seeks a truth that he would die for. Athens gives Socrates the option to live, and even to remain in Athens, if he will just stop stirring up trouble. He refuses because he would rather die than cease proclaiming “Thou must know thyself.”
The philosopher seeks a truth that must be spoken–a truth about which it would be better to die than to remain silent. Such a truth is needed because only such a truth can be held with complete sincerity. After all, I cannot lessen the strength of my assertion by speaking less loudly. If I am unwilling to shout it from the rooftop, how can I whisper it in my innermost being? As Newman insists, assent is assent: it does not come in degrees.
So take the most obvious truth imaginable: “This is a table,” said of the table before me. Would I insist on it with a gun to my head? Probably not: I know my senses are fallible, and if someone puts a gun to my head it may well be because I need to be snapped out of a delusion. Before the belief was tested, I had no doubts, but now I know that I do. How can I justify saying something I would not stake my life on?
To this the probabilist says: fine, we’re not certain, so we can’t say “This is a table.” But we still have our uncertain opinion: “It is probably the case that this is a table.” No longer have I said anything about the table that I would refuse to die for.
But–will I die for this attribution of probability? Surely not. If I think of probability as something “out in the world,” then I can be no more sure of this probability than of the table itself; and if I think of probability as something “in my mind,” then I cannot be at all sure of it–after all, I expect it to change as I learn more about the world. Probabilities offer no relief. Still we must speak, and our opinions are the last thing we should speak about: we know we cannot trust them.
Of course the probabilist never claimed to be doing philosophy in the Socratic sense. He sees words as tools: we do not hold to our words, but use them. We assign our beliefs probabilities because beliefs are a sophisticated way of playing the odds, of trying to get ahead in the game of life. If I want to use my belief that “This is a table” is 99% likely to be “true,” and the bookie thinks it 99.9% likely, then I should play the odds and bet on “false.” So, who is the bookie here? On what event am I wagering, and for what prize, and with whom? And what is my stake?
Say I’m betting on whether there’s gold in them there hills, for possession of the gold, against everyone else who might at some point come to possess it, if it exists, and who might get rich at my expense, if it does not and yet I finance a search for it. If the potential gain justifies the expenditure, then I stake the cost of the expedition; if not, then I stake the lost opportunity to do so. This is the game of life; Nature is my bookie.
Similarly, in my “What is a table” scenario, I’m betting on whether this is a table, for my life, against a madman with a gun. I must weigh the worth of my word against that of my life: either he shoots me, or I say “I’m not sure whether this is a table.” But who is the bookie here?
For the probabilist, the answer is the same as before: Nature. Nature always sets the house rules. This neither philosopher nor common-sense man can accept. Whether or not I believe something ought to have nothing to do with whether or not a gun is to my head. I can’t just use certain words as a tool to preserve my life. In Stanley Cavell’s phrase, I ought to mean what I say.
Yet all three–probabilist, philosopher, man of common-sense–would do as the madman commanded. And none would be lying.
The probabilist, because for him there is no such thing as honesty; words are tools, which he uses, in this case, to preserve his life. There are no words the probabilist will die for.
The philosopher, because–as this thought experiment has taught him!–“I’m not sure,” and “Know thyself,” and the like, are the only things he can say honestly. The philosopher will die for these words, and no others.
The common-sense man, because if someone really held a gun to his head, he would become uncertain. He would stake his life on “This is a table,” but only if no one took him up on the bet. He would do so, for example, if he needed a table and nothing but a table would do, and would do so without the slightest doubt that this was indeed a table. But if someone threatened to kill him over saying “I’m not sure,” wouldn’t this, in itself, be reason to doubt? Who would make such a threat without a reason? Until you know he’s mad, his demand gives reason to not be sure; once you know he’s mad, it doesn’t matter–no sounds you make will count as talking to him.
It’s hard to be a martyr for the external world. So what will the common-sense man die for? So far we know too little about him; only, really, that he is ordinary. An ordinary Roman will die for “Rome is great.” An ordinary Christian will die for “Christ is Lord.” Newman thinks any decent man will die for “My mother is not a liar.” What these have in common, I suppose, is that they’re ethics, not physics. They’re matters of fact, but not only matters of fact. They’re not probabilistic stakes, but they’re also quite different from the philosopher’s wager.