Why be rational?

Timmo writes again about William James and "forced" choices. (Additional commentary can be found in his earlier post on this topic, and Timmo offers an example of a Jamesian "genuine" choice.)

Why does James write about this topic at all? Why does Timmo bring it up? Why does Timmo bring it up in the context of the Sullivan/Harris debate on religious moderation?

The dominant moral ideology in most of Western civilization is that you can believe anything you please, for any reason, good or bad, or for no reason at all. The U.S. Constitution explicitly grants its residents the specific right to exercise their religion with—at least as regards to belief—no conditions at all.

A legitimate response to Harris would be, "It doesn't matter what you think about my beliefs; It's a free country, I'll believe as I please."

So we must conclude that theists bring up James' and other arguments precisely because political freedom is not enough: It is desirable to give the imprimatur of "rationality" to one's beliefs, or, failing that, to subvert the value of "rationality" so that the charge of irrationality no longer has persuasive force.

For any characterization to have persuasive force, it must at least draw a distinction: There must be some things that are X, and some things that are not X; if everything (or nothing) is X, then X is, by definition, vacuous.

Now it's clear (and, I hope, not controverted by Timmo) that the Evidentialist Principle

(EP): We ought only assent to those propositions which can be established by evidence.[1]

at least draws a distinction: Some statements are supported by evidence, some are not. But James appears to want to escape this particular distinction, presumably because religious statements, statements about God, do not appear to fare well: "I suspect that it is not possible to produce an argument that, on pain of absurdity, rationally forces belief in God."

There are two possible cases: Either James' argument actually does create one or more nontrivial distinctions, or it makes no nontrivial distinctions at all.

There are two "domains" of distinction that James appears to be trying to make. Primarily, he's interested in distinguishing a category of statements that can be and cannot be distinctively evaluated by the Evidentialist Principle. Given that distinction, one expects the next part of the argument (not explicitly referenced in the post) to be how to draw some distinction between those statements that cannot be evaluated by (EP).

Timmo presents James' distinction:

A living option is one in which the alternative hypotheses are alive in the sense that they are intellectually defensible propositions which may be true. A forced option is a dilemma, with no possibility of not choosing (because not choose is functionally the same as one of the options). In a momentous option, the option may never again present itself, or the decision is effectively permanent, or something important hangs on the choice. James calls an option genuine when it is living, momentous, and forced. [emphasis added]

It becomes pertinent, then, to ask whether James does indeed create an actual distinction.

Before we look at James' supposed distinction in detail, we can look at other distinctions, distinctions between those statements which can be usefully evaluated on evidence, and those that cannot.

Popper[2] draws a distinction between theoretical or hypothetical statements and metaphysical statements. (EP) itself is a case in point: To evaluate (EP) according to the evidence, one must first uncritically—that is without regard to the evidence—accept (EP). The logical positivists tried mightily to wriggle out of this vicious circularity, but in vain.

In fact, (EP) itself satisfies James' criteria: It is a living option: in that, at least qualified to apply to the domain of ontology, it might well be true. It is forced; to be agnostic about EP is instrumentally the same as rejecting it. It is momentous in the sense that false ideas about reality tend to be deadly or damaging. Furthermore, as noted above, one cannot use (EP) to evaluate itself.

As Quine notes, one can hold (EP) and still arbitrarily affirm any given statement, "come what may" (i.e. as a standing statement), so long as one is willing to adjust other statements around the evidence. There is no better demonstration of this principle than The Dead Parrot Sketch: "He's pining for the fjords!"

It's also clear that the shopkeepers assertion that the parrot is not actually dead meets all of James' criteria: It is living: it might be true that the parrot is really resting; it is forced: agnosticism entails the same result (a refund) as assent to the parrot's death; and it is momentous: if the parrot is actually dead the shopkeeper will have to give a refund.

It's clear that the Evidentialist Proposition by itself will not do the complete philosophical job ((EP) cannot discuss metaphysics), nor will it even do just the complete ontological job: We cannot, on just the basis of (EP), discuss the ontological statements we will affirm as standing, "come what may". We have to grant James at least that he's identified a true problem, not just a pseudo-problem.

But does James actually solve the problem? As I've hinted above, I think not. For any statement at all—even any prosaic declarative sentence about reality—can be "excluded" from (EP) on the basis of James' criteria. James' criteria may appear to create a distinction, it does not actually do so. Every statement is, or can be, "genuine". His criteria is vacuous.

Attempts to rationally justify religion always end up with this sort of vacuity. Either with James' appearance of rationality, or with Sullivan's explicitly abandonment of rationality in favor of truthiness (the subjective feeling of truth). And this why I myself (and Harris too, I suspect) strongly confront religious moderation. Not because we have any particular beef with the ethics that come from religious moderation (especially liberal religious moderates such as Timmo), but rather because attempts to rationally justify religious moderation end up simply erasing distinctions, undermining a philosophical argument against religious extremists.

Because, of course, religious extremists—because they're fundamentally intolerant of contrary opinion—must insist that their ideology is actually true, that God really does exist, that God really does have definite ideas about how people must live. They must wriggle out from Evidentialism (especially scientific evidentialism[3]) but must still insist that their belief is rational and therefore true.

By providing every possible statement a rhetorical escape hatch from scientific evidentialism, but giving no principled criteria to actually distinguish true religion from false, the religious moderates help shield their extremist brethren from any sort of rational analysis.


[1] As a scientific naturalist, I would qualify (EP) as being applicable only within the domain of ontology: Propositions or statements about existence should be restricted to an evidentiary basis.

[2] Popper does not appear to get the details of the distinction quite right, but he lays a good enough foundation to be considered a true pioneer.

[3] Scientific evidentialism adds Occam's razor as an additional restriction to pure evidentialism.