Lisa Warenski on Naturalism, Fallibilism and the A Priori

Lisa Warenski is a naturalist. She's committed to the “scientific method”. She characterizes that method by its “ongoing responsiveness of theory to evidence”.

How does the a priori fit into this?

Many apriorists would argue that “some class of claims [are] immune to empirical evidence”. Thus, such a position “would be in conflict with the naturalist’s commitment to scientific method”.

Fallibilism has been an important part of naturalist and empiricist philosophy - especially since C.S. Peirce. One would think that a naturalist and fallibilist would be against the a priori in all its forms. That isn't so in Warenski’s case. She says that

“[t]his fallibilist notion of the a priori accords with the naturalist’s commitment to scientific methodology in that it allows for a priori-justified claims to be sensitive to further conceptual developments and the expansion of evidence”.

Thus not only is Warenski a fallibilist, she also applies her fallibilism to the a priori itself; rather than simply rejecting it. She says that

“[g]iven the overthrow of Euclid’s parallel postulate and the discovery of the set-theoretic paradoxes, an apriorist has reason to be fallibilist about a priori claims independently of considerations of naturalism”.

Warenski divides her aprioristic fallibilism into two parts:

1. One might allow that we could be mistaken in thinking that an a priori claim is true.
2. One may be wrong in thinking that an a priori claim’s justification-conditions are actually a priori.

Warenski also accepts empirical defeasibility for a priori beliefs/claims. That empirical defeasibility applies to the two distinctions mentioned above:

1. The ‘truth’ of an a priori statement/belief can be defeated on empirical grounds.
2. The a priori warrant of a statement/belief can be defeated on empirical grounds.

There's more to this.

An a priori statement or belief can also be “revised on a priori grounds”. Thus it would be a priori defeasible rather than empirically defeasible. Warenski cites some interesting examples of the a priori defeasibility of the a priori. She says that

“[t]he need to revise naive set theory was recognized on purely conceptual grounds, and non-Euclidean geometries were developed prior to the discovery that space was non-Euclidean, so arguably the grounds for revision of an a priori claim could be purely conceptual”.

Thus “conceptual revision” amounts to a priori revision, according to Warenski. We can now say that “[i]f an a priori claim were to be revised on purely conceptual grounds, the revision would not undermine its a priori status”.

We can revise the a priori with the a priori. Not only that: a priori justified beliefs/claims may also come to be empirically supported!... Though not so quick! If an a priori justified belief/claim does come to be empirically supported, then Warenski argues that we can conclude with two points:

1. The belief/claim, or truth, wouldn't be defeated.
2. The a priori justification of the belief/claim above would be defeated.

Warenski offers us another interesting example. She writes:

“… Gauss doubted the a priori status of Euclid’s parallel postulate, understood as a claim about physical space; however, he thought it to be empirically corroborated.”

Here again there are two things to conclude from the quote above:

1. The parallel postulate itself wasn't undermined.
2. The a priori warrant for the parallel postulate was undermined.

A Priori Fallibilism & Naturalism

Despite this fusion of naturalism and apriorism (if that’s what it really is), Warenski happily accepts that a naturalist account of the a priori will prove to be, at best, problematic. She gives two primary reasons for this:

1. “The range of different kinds of beliefs, inference rules or claims that we might classify as a priori may not admit of a single positive characterization.”
2. “Different standards apply to different forms of positive epistemic appraisal: the conditions that must be met for entitlement or epistemically-blameless reasoning are weaker than what must be met for justification sufficient for knowledge.”

It's still clear that Warenski’s principle target isn't radical rationalism; but (extreme?) Quinean naturalism. Putting her case intuitively, she says that

“[i]t is hard to think of convincing cases of empirical observations that should count as evidence against elementary principles such as the law of identity or modus ponens”.

However, the rationalist shouldn't think that Warenski is a completely compliant rationalist.For example, she says that “given the peculiarities of quantum mechanics, perhaps there are such possible observations”.

A Priori Fallibilism & Revisability

Warenski is more explicit about apriorist fallibilism. She says that

“[a] fallibilist about a priori justification thinks that an a priori claim is empirically indefeasible, but he allows that we don’t know that future conceptual developments will never come to reveal empirical evidence against it”.

That characterization may sound self-contradictory at first sight. She's saying that a priori claim is indefeasible as it stands today; though it may defeasible according to how things stands in the future.

This isn't unlike Rudolph Carnap’s position on analyticity. Thus we can distinguish the following alternatives:

1. An analytic statement. Full stop.
2. Analytic-in-L

And in the case of fallibilist apriorism:

2. An a priori belief or justification. Full stop.
3. An a priori belief or justification at time t. (Or, perhaps, a priori-in-L.)

Warenski’s is clearly aware of one possible riposte to her position:

1. If P is known to be true,
2. then P is true.
3. If P is known to be true,
4. then it follows that P won't come to be undermined.
5. Therefore the fallibilist can't claim that both P is known to be true and that it's not known that P won't come to be undermined in the future.

If the a priori is revisable. Then logic, after all, is revisable. And then we're back to Quine.

Reference

Warenski, Lisa. (2008) ‘Naturalism, Fallibilism, and the A Priori’