Christopher Michael Cloos

In “Evidence, Experience, and Externalism” Jack Lyons proposes a new twist on the Sellarsian dilemma. This dilemma typically argues against standard foundationalism. It claims non-doxastic experiential states cannot justify basic beliefs. This is thought to count in favor of standard coherentism, which is committed to the idea that only beliefs can justify beliefs (i.e., doxasticism). Lyons recasts the Sellarsian dilemma so that it does not count in favor of coherentism (doxasticism); instead, it counts in favor of non-evidentialism and, as a result, it counts in favor of externalism. Lyons’ argument might be reconstructed as follows:

1. Evidential justification is distinct from non-evidential justification.
   • Df: Evidential justifiers (e.g., other justified beliefs) are capable of justifying beliefs by serving as evidence for those beliefs, whereas non-evidential justifiers (e.g., reliability, coherence) are things in virtue of which  beliefs are justified (i.e., they are not evidence for the beliefs but that on which justification supervenes, for example).
2. The Sellarsian dilemma must be restricted to evidential justification; otherwise, it is a nonstarter.
3. The distinction between evidential and non-evidential justifiers makes it possible to hold separate the Belief Principle and the Grounds Principle.  
   • Df: The Belief Principle claims that only beliefs can evidentially justify beliefs, whereas the Grounds Principle claims that all justified beliefs have grounds (i.e., evidential justifiers).
4. Doxasticism is the conjunction of the Belief Principle and the Grounds Principle.
5. When the Sellarsian dilemma is restricted to evidential justification (as premise 2 indicates it must be), it is only an argument for the Belief Principle.
6. Thus, the Sellarsian dilemma is not an argument for coherentism (doxasticism); such an argument requires an argument for the Grounds Principle, which the dilemma does not provide.
7. However, doxasticism can be rejected on independent grounds (assumption not argued for, assumed from the literature).
8. Rejecting doxasticism for independent reasons results in rejecting the Belief Principle and the Grounds Principle.
9. Thus, in rejecting doxasticism, the Sellarsian argument for the Belief Principle turns out to be an argument for rejecting the Grounds Principle.
10. Therefore, the Sellarsian argument is an argument for non-evidentialism (i.e., not all beliefs must have evidential justifiers as grounds), and beliefs can be justified by factors external to the agent (i.e., externalism is true).

In laying bare Lyons’ argument we have already accomplished something because it’s easy to get lost in the mounting of distinctions, positions, and counter-positions in his paper. For a defense of each premise I recommend reading the whole paper, but for the remainder of this post I’ll focus on one aspect of the argument. This aspect might undermine the entire argument.

Lyons responds to an objection to his argument. The objection is that he’s working with too narrow of an understanding of evidence and this begs the question against non-doxastic evidentialist theories. Notice that the argument above is restricted to a brand of doxastic evidentialism. Perhaps a different understanding of evidence, one that included, for example, physical evidence like a murder weapon, would not commit one to doxasticism. As a result, it might be possible to satisfy evidential justification without being led into the Sellarsian dilemma. Lyons responds to this charge without argument, by citing philosophical tradition. Put candidly: if you’ve got a problem with my usage of evidence, it’s their problem not mine! The people he’s referring to are Feldman and Conee, Haack, and Alston. Lyons says that it’s their conception of evidence that implies the Belief principle, and he’s merely borrowing this conception for the sake of argument. Even if we grant Lyons that his use of the notion of evidence is not question-begging against his opponents, it ‘s still problematic for his argument.

There’s solid motivation to move away from a doxastic notion of evidence. If only beliefs can serve as evidential justifiers, even for the sake of argument, then the notion of justification hinges on defending and substantiating a doxastic notion of evidence. Recently, Ram Neta (in “What Evidence Do You Have?”) has provided extensive counterexamples to a doxastic notion of evidence as cashed out in its various theoretical forms (e.g., reliability, JTB, E = K, coherence, deontology, and so on).

As it turns out, Lyons makes a double-mistake. First, he fails to argue against doxastic theories, and merely cites that the literature supports the idea that “doxasticism is a lost cause,” thereby motivating premise 7 without argument. Second, he bases his argument on doxasticism because it is a notion endorsed by mentalism and other internalist-based epistemologies. Without arguing against a non-doxastic understanding of evidence his argument is something of a straw-man. That is, he set-up an easy opponent to knock down. To sustain his argument Lyons must discharge the burden in the literature which points away from a doxastic conception of evidence in the first place, a burden that undermines the initial premises (assumptions) in his argument unless it can be discharged.