Saturday, February 24, 2018

Cognitivism and Moral / Philosophical Peer Intransigence

Richard Rowland's forthcoming Analysis paper on 'The Intelligibility of Moral Intransigence' presents a curious argument against moral cognitivism.  It goes roughly as follows:

P1. Beliefs track perceived evidence.
P2. Perceived peer disagreement is perceived evidence.
Hence C1. Peer intransigent judgments are not beliefs.
P3. Moral peer intransigence is intelligible: moral judgments can be peer intransigent.
Hence C2: Moral judgments are not beliefs.

The argument seems to prove too much, insofar as one could just as well replace 'moral' with 'philosophical' in P3, but non-cognitivism about all philosophy seems pretty absurd.

So, where does it go wrong?  I suggest P2.  As per my previous post, peer disagreement is not necessarily epistemically relevant.  A peer's disagreement is only evidence against your current view if it is evidence that you've made a mistake by your own lights.  But in many (philosophical, but especially moral) cases it is instead merely evidence that your peer's bedrock assumptions differ from yours.  In such cases, intransigence is not only intelligible, I argue, but also perfectly reasonable.  The mere fact that some people have different priors or bedrock assumptions from yourself is not in itself any kind of evidence against your priors or bedrock assumptions.  So P2 is false.  Perceived peer disagreement is only perceived evidence when further conditions are met (as they are in the standard cases -- arithmetic disputes when calculating the dinner bill, etc. -- that Rowlands uses to motivate the premise).

[Terminological variant: If one defines an 'epistemic peer' narrowly as someone you regard as generally just as likely as yourself to reach the truth on questions in this domain then we should instead reject the assumption that peer intransigence is intelligible.  In the proposed cases, the agent should reject the claim that the disputant is their epistemic peer in this strict sense. Even if they grant that they are equally procedurally rational, that is insufficient.  After all, it is incoherent to think that different priors or fundamental epistemic methods are just as likely to be correct as your own -- see the last section of Elga's 'How to Disagree about How to Disagree'.]

1 comment:

  1. Hi Richard,

    While I don't find Rowland's argument persuasive, I don't think your reply fully works, either, for the following reason: granting that your argument regarding the epistemic weight of peer disagreement is correct, the fact is that some other philosophers disagree with you on that, and consider peer-disagreement to be evidence against one's moral beliefs (or other philosophical beliefs). If the rest of Rowland's argument were successful, then this would imply that peer intransigent moral judgments made by people who do take peer disagreement to be evidence, are not beliefs. But if those are not beliefs, that would seem to be considerable evidence (though perhaps not conclusive) that moral judgments in general are not beliefs. Of course, as you point out, one can run this argument for philosophical disagreement in general (at least, with regard to philosophers who disagree with you on the epistemic status of peer disagreement), and the result is pretty absurd, so something is surely wrong.

    I think one problem (there might be more problems) might be intelligibility. While it's intelligible that a person persists in her belief on the face of peer disagreement while believe that peer disagreement is evidence, it's difficult (at least for me) to comprehend that she might not update her probabilistic assessment at all (she might persist in her believe, but assigning lower probability to it than before).

    Still, there are some ways I think I can make sense of something like that. For example, let's say Bob believes, while calmly reflecting on the matter, that peer disagreement is evidence, and that Alice is his peer. But when he thinks about one particular moral claim C1 made by Alice, he's outraged, and not updating his credences at all - but condemning Alice. Yet, in that case, it seems to me that implicitly, Bob is either downgrading Alice from peer to not peer, or perhaps coming to believe that Alice is lying and actually does not believe C1, or something along those lines. There are other scenarios that make it intelligible, but so far, I find that one or more of Rowland's hypotheses does not hold in the scenarios I find intelligible (e.g., a person does not consider the other person a peer, or at least not at that moment, when thinking about that particular belief, etc.).


Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)