Disagreement: Stage-Setting and Kornblith’s Skepticism

Suppose David and Nates disagree about the morality of Capital Punishment.  David thinks that CP is unjustifiable, whereas Nates–ever the Kantian–thinks that CP may under certain circumstances be not only justified but morally required.  David and Nates are aware of their disagreement with each other–they’re facing off in the seminar room.  Now consider this question: does the mere fact of their disagreement give them reason to suspend judgment on the morality of CP?

We can certainly imagine contexts in which it would not.  Suppose, for example, that David is thoroughly familiar with the philosophical literature on the morality of CP, as well as the legal history of its uses and abuses, and, finally, all the statistical data concerning deterrence etc.  Nates, on the other hand, is unfamiliar with just about all of that stuff, save for some spirited passages from Kant in which Kant talks about avoiding the “serpent windings of utilitarianism” and the importance of the punishment ‘fitting’ the crime.  In that case, we can probably agree, David should not be too troubled by Nates’s disagreement, since he is clearly in a better epistemic position than Nates vis-a-vis the available arguments and evidence with respect to the moral justification of CP.

Alternatively, suppose David and Nates are in the same epistemic position with respect to the arguments and evidence, but that Nates is much smarter and/or better trained in evaluating philosophical arguments and statistical data than David.  And suppose, further, that both David and Nates are aware of this difference between them.  In this case, again, we can probably agree that not only should Nates not be troubled by David’s disagreement, but that David may even have reason to suspend his judgment in the face of disagreement with someone much smarter and better trained than himself.

But now suppose that neither of these differences between David and Nates obtain.  That is, suppose that David and Nates are in the same epistemic position vis-a-vis the arguments/evidence and that they are equally smart and equally well-trained.  Suppose, that is, that David and Nates are what the literature I’ll be discussing calls epistemic peers, and that David and Nates regard each other as epistemic peers.  Suppose, finally, that David and Nates do not only regard each other as epistemic peers generally, but that they regard each other as such in the specific context of forming judgments about the morality of CP.  In other words, facing off in the seminar room and trying to convince each other of their respective positions, neither believes that in this case the other is simply confused, or biased, or has made a mistake, or is being uncharacteristically dense, argumentative, etc.

In the case I’ve just described–David and Nates are epistemic peers when it comes to forming judgments about the morality of CP–does the mere fact of their disagreement give David and Nates a reason to suspend judgment about the morality of CP?  This is the central question being addressed in the contemporary literature on disagreement, which I’ll be discussing in a series of upcoming posts.  In this initial post, I give a brief overview of an essay by the philosopher Hilary Kornblith, titled “Belief in the Face of Controversy.”  Or perhaps I shouldn’t say ‘overview,’ since in these posts I’ll just be highlighting aspects of the essays I’m reading that struck me as most interesting.

On Kornblith’s view, it is very often (more often than not) rational to suspend judgment about philosophical issues in the face of epistemic peer disagreement (hereafter, EPD).  Here’s an extremely simplified version of his argument (as I understand it, and using considerable constructive license):

1.  EPD (absent certain complicating conditions that I won’t get into) calls for a suspension of judgment in the context of perceptual disagreements, mathematical disagreements, and disagreements within the empirical sciences.

2. Either there is a crucial disanalogy between these areas of inquiry and philosophical inquiry that affects the relevance of EPD, or there is not.

3.  There is not.

4.   Therefore, EPD calls for a suspension of judgment in the context of philosophical inquiry.

One might say: But in the perceptual, mathematical, and scientific contexts judgments are right or wrong, correct or incorrect, and that’s why EPDs put such pressure on judgment.  But philosophical judgments are different–they’re not right or wrong, correct or incorrect, but rather ‘open to interpretation,’ etc.  The trouble with a response like this is that it might seem to embrace a kind of relativism that a lot of philosophers wish to avoid.  Many philosophers really want to think of philosophical inquiry as an inquiry into a kind of objective truth–not the truths revealed by perception, mathematics, or the empirical sciences, but objective nonetheless.  These philosophers will not want to overstate the differences between philosophical inquiry and the paradigmatically objective forms of inquiry.

So why shouldn’t these philosophers simply accept (4)–that is, accept that like in other contexts, EPD concerning philosophical issues calls for a suspension of judgment?  Well, the problem is that unlike in the other contexts EPD in philosophy is so pervasive and persistent.  It’s pretty much a joke that philosophers cannot agree on anything.  So the worry is, as Kornblith puts it, that if EPD in philosophy calls for a suspension of judgment, then “a broad skepticism about philosophical matters threatens” (33).

So from the fact of philosophical disagreement, Kornblith has generated an uncomfortable dilemma, with skepticism about philosophical knowledge residing on one horn, and relativism about philosophical matters residing on the other.

Toward the end of his essay, Kornblith says the following:

I do not mean to suggest that we should stop thinking about [philosophical] issues, or that thinking about them, and trying to work out tenable views, is not intellectually respectable.  That is not my view at all….

Alas, he doesn’t say anything else about why that is not in fact the reasonable practical conclusion to draw from his arguments about the epistemic relevance of EPD.

 

This entry was posted in Uncategorized. Bookmark the permalink.

16 Responses to Disagreement: Stage-Setting and Kornblith’s Skepticism

  1. CreatureAdam says:

    Does he mention cases where he thinks it isn’t?

    What is the argument for why being an “epistemic peer” means one party can’t be right about any given claim — or have better supporting reasons than the other one?

    • CreatureAdam says:

      Meant to blockquote this with respect to the earlier comment, but the tagging didn’t work :-s

      “On Kornblith’s view, it is very often (more often than not) rational to suspend judgment about philosophical issues in the face of epistemic peer disagreement”

      Does he mention cases where he thinks it isn’t?

      What is the argument for why being an “epistemic peer” means one party can’t be right about any given claim — or have better supporting reasons than the other one?

  2. David says:

    Thanks for posting questions, Adam!

    You ask: Does Kornblith mention cases where he thinks [EPD on philosophical issues] doesn’t make it rational to suspend judgment?”

    Yes, Kornblith suggests that there may be cases in which you have good reason to think that an epistemic peer with whom you disagree about P is wrong and that you are right—for example, because you have good reason to believe that your epistemic peer has in the specific context a bias that is screwing with his rational judgment concerning P or not P (let’s say, for example, he was the longtime student of an eminent philosopher whose view that P you now claim to have refuted). (But Kornblith warns that it would be irrational to simply assume or imply that epistemic peers who disagree with you must be impaired in some way by virtue of the fact that they disagree with you!)

    –You ask: “What is the argument for why being an “epistemic peer” means one party can’t be right about any given claim — or have better supporting reasons than the other one?” When epistemic peers disagree, it may well be the case that one of them has better supporting reasons for his position. The idea is that part of what makes them epistemic peers is their roughly (or ‘precisely,’ if we idealize) equal ability to assess the strength of supporting reasons, and so although they would both admit that one of them has better supporting reasons, they are (the claim goes) in no better position to say who has the better supporting reasons than they are in a position to say who is right and who is wrong. Does that make sense?

    • CreatureAdam says:

      I’ve not read this stuff of course…so I may be whistling in the dark. My first reaction was to think that a pair of ideal “epistemic peers” would in all cases agree with one another (contrapositive: to the extent they disagree, they cannot be true epistemic peers).

      At a sufficiently granular level, you’d think that there could (and almost certainly would) be all kinds of idiosyncratic differences in bias between two people of roughly similar reasoning ability, and familiarity with arguments in a given literature; so I wonder what it even means to be an epistemic peer with someone, in an argument (maybe you and I could be epistemic peers about something like eating pizza: yet you hate pineapple on yours, but to me it’s the best thing ever, and never the twain shall meet. That seems a bit different from disagreeing about substantial moral questions).

      Maybe you need to democratize it — and allow a joint verdict to be the thing that settles the more substantial questions; two madmen could disagree, and both be so wrong about everything that it wouldn’t even be funny.

      • David says:

        –“I’ve not read this stuff of course…so I may be whistling in the dark.” I’m just one essay ahead of you, so take what I say with a grain of salt!

        –“My first reaction was to think that a pair of ideal “epistemic peers” would in all cases agree with one another (contrapositive: to the extent they disagree, they cannot be true epistemic peers).” Yeah, I see what you mean. But even in a case of disagreement between ideal epistemic peers, it’s possible in any given case of disagreement that one of them has made a mistake (perhaps they’re precisely equally susceptible to making errors in reasoning, qua ideal epistemic peers). In other words, X and Y being ideal epistemic peers doesn’t entail that X and Y have to think all the same thoughts.

        In the pizza case you mention, our disagreement doesn’t seem at all threatening, since we’re both happy to concede that at some level judgments about pizza are judgments of taste. It’s harder for us to do that, I think, when we’re disagreeing about issues in metaphysics or ethics. (Of course, maybe this is just my bias as a philosopher-type.)

        • CreatureAdam says:

          ” X and Y being ideal epistemic peers doesn’t entail that X and Y have to think all the same thoughts.”

          I agree with that. I wasn’t suggesting they would have to think the same thoughts though; it was more a suggestion that ideal epistemic peers would ultimately arrive at the same conclusion, when assessing an argument. Here’s one thought: suppose you and Nates, are ideal epistemic peers with respect to something like modus ponens.

          If you disagreed about it, because one (or both) of you were mistaken in some way about how it works, the source of that disagreement would still be resolvable — and one or other of you would change your view accordingly once the mistake had been understood. To any extent that you both followed the other’s argument but disagreed about what it established, the source of the disagreement would lie in the fact you are not, in whatever sense, ideal epistemic peers.

          Also: I want to say that a disagreement between two parties is resolvable, at least sometimes, by appealing to the good judgment of other relevantly informed individuals. So rather than suspending their judgment upon reaching an impasse, establishing a consensus in a larger peer group might be one way in which two roughly epistemic peers could agree to resolve certain disagreements. Or maybe not!

          • David says:

            Hey Creature,

            You write: “Also: I want to say that a disagreement between two parties is resolvable, at least sometimes, by appealing to the good judgment of other relevantly informed individuals. So rather than suspending their judgment upon reaching an impasse, establishing a consensus in a larger peer group might be one way in which two roughly epistemic peers could agree to resolve certain disagreements. Or maybe not!”

            Yeah, that seems right, and Kornblith talks about contexts in which the numbers count. He thinks that especially in the sciences and the formal areas of philosophy an established consensus about P among a community of experts should carry a lot of epistemic weight. He hastens to add, however, that although such consensuses are often arrived at in the sciences, they are almost never arrived at in philosophy. So this move may not help in the philosophical context.

  3. Nates says:

    Great project, David! For professional reasons, I’ve been thinking about philosophy of science lately, and I wonder if Kuhn might be a useful resource here. So, we could distinguish two kinds of disagreements. First, there are disagreements that stem from a difference in our basic philosophical worldview (or paradigm). In fact, you alluded to this above, when you referred to my Kantianism. If two philosophers disagree over even the most basic things – like the nature of moral goodness, then it’s not surprising that they’ll disagree on particular issues. In these cases, it’s not really a dispute about capital punishment (or whatever) at all. Moreover, in these cases, I don’t see why it’s rational to suspend judgment based on the dispute. If my view on capital punishment follows from my Kantianism, then it would seem absurd to abandon my entire philosophical worldview just because my non-Kantian friend David takes a different stand on capital punishment.
    Of course, we might think that the problem reappears at a higher level: if my epistemic peer is anti-Kantian, then don’t I have good reason to suspend judgment about Kantianism itself? But despite the apparent analogy, my sense is that this problem is actually quite different from the earlier one – again for Kuhnian reasons. Worldview disputes just aren’t adjudicated in the same way – in part because our criteria for what counts as a good philosophical system are themselves part of the system. Here we’re in the land of incommensurability and theoretical revolutions. And one of Kuhn’s great insights is that it’s perfectly rational to stick to one’s paradigm – even in the presence of opposing and incommensurable paradigms – as long as yours is healthy, thriving, making progress, etc.
    Things are different when we have a dispute between two epistemic peers working within the same philosophical tradition. Here Kornblith’s version of the problem seems quite pressing. So I guess my claim is that many (maybe even most?) philosophical disputes don’t fit into Kornblith’s schema. Which means that it might not be so problematic to suspend judgment for the remainder that do.

  4. Josh says:

    Having not read the literature to which you refer, this reply will doubtlessly come across as naive – but –

    Doesn’t it depend on the actual specifics of the arguments being offered? Just bc two interlocutors know that they are well-matched doesn’t mean that on the particular occasion under consideration, equally logical/supported arguments are being offered, does it?

    If the source of the disagreement turns out to be based on logical errors (non sequiturs, false generalizations, etc.) then one person could be right and the other wrong, despite their in-general peer-to-peer status. And those errors in logic could be really below-the-surface types of things too, not just surface-level fallacies.

    If two people disagree, one of them may be wrong, is all I’m saying.

    If the disagreement turns out to be more of the Kuhnian type Nates refers to, then the answer might be different – but – there wouldn’t seem to be a rough and ready way to distinguish those two situations, right?

    If its a matter of consequentialism and deontological intuitions clashing, that might be less resolvable than its a simple case of one person’s view leading to a straightforward self-contradiction, but most disagreements are somewhere in the middle.

    To paraphrase a sports cliche – that’s why we have the debates. Just bc the argumentative situation is in some ways balanced doesn’t mean that it is in all ways irreconcilable.

  5. David says:

    Thanks for posting, Nates! It’s been a long time since I read Kuhn, but you do a great job of outlining a Kuhnian response to the Disagreement Problem. I’m not sure yet how effective I find that response, so let me put a follow-up question to you.

    Suppose that you are a Kantian and I am a Utilitarian. In other words, you believe that moral goodness consists in a good (rational) will, and I think that moral goodness consists in well-being/absence of pain & suffering. This is a fundamental moral disagreement that will of course underwrite a number of other, more specific disagreements about moral matters. I’m supposing, then, that you would want to say that you (Kantian) and me (Utilitarian) are proponents of different moral ‘worldviews.’

    And you say: “Worldview disputes just aren’t adjudicated in the same way [as disputes internal to a worldview]– in part because our criteria for what counts as a good philosophical system are themselves part of the system. Here we’re in the land of incommensurability and theoretical revolutions.”

    So how does this help soften the Disagreement Problem. I’m supposing the argument goes something like this:

    1) There is no rational pressure to suspend judgment on the correctness of worldview W in the face of competing worldviews not-W.
    2) MANY philosophical disagreements, on reflection, boil down to conflicts between competing worldviews.
    3) Thus, with respect to MANY philosophical disagreements, there is no rational pressure to suspend judgment on the correctness of one’s position.

    So here’s why I’m not loving this response to the Disagreement Problem (DP). First Worry: we seem to be avoiding the DP by embracing a kind of parochialism. That is, neither you (the Kantian) nor me (the Utilitarian) need to abandon or revise our views, but we should probably recognize that in cases where our disagreements reflect or are rooted in our basic moral commitments, we really have nothing profitable to say to each other. Kantians can carry on doing moral philosophy with other Kantians, and Utilitarians with other Utilitarians, and Rational Egoists with other Rational Egoists, and so on, but there’s not much to be said for moral philosophical argument and debate between proponents of these different camps. Is (i) an implication of the Kuhnian response, and, if so, is it (ii) an implication you’re comfortable with?

    Second Worry: How, exactly, does Kuhnian incommensurability differ from relativism? What’s the difference, ultimately, between saying that worldviews are ‘incommensurable’ and saying that worldviews are neither true nor false, correct nor incorrect, but rather systems in which concepts like ‘truth’ and ‘correctness’ receive definite meaning–that is, relative? What’s the difference, in other words, between Kuhn on worldviews and Rorty’s perspectivalism? It won’t help to appeal here to notions like ‘thriving,’ ‘success,’ ‘health,’ etc., for these are all basically normative notions, and, anyway, will be defined relative to what our aims/goals are.

    Third Worry: Setting my first two worries aside, you say that it’s rational to maintain one’s most fundamental commitments in the light of disagreement so long as one’s commitments define a worldview that is healthy, thriving, making progress, etc. But which philosophical ‘worldviews’ can confidently claim to be ‘making progress’? I mean, as Kornblith mentions at one point in his article, there are formal areas of philosophical inquiry–logic, decision theory, etc.–in which progress has clearly been made. But has the Kantian worldview ‘made progress’? Or the Utilitarian worldview?

    I guess I can sum up my worries about the Kuhnian response as follows: at the end of the day, am I only able to hold on to my commitments by divesting those commitments of any claim to commanding the assent of all rational epistemic agents?

  6. David says:

    Hey Josh! So I can only respond to your questions as I think Kornblith might, as my views on these issues are unformed…so that being said….

    You ask: “Just bc two interlocutors know that they are well-matched doesn’t mean that on the particular occasion under consideration, equally logical/supported arguments are being offered, does it?”

    No, of course not. Suppose God is perfectly rational and watching our debate unfold with a wry smile; He knows full well that you’re in fact making the stronger argument, and that I have, unbeknownst to either of us, made a subtle logical error. But God’s POV is not one either of us have access to. We have only our own lights to go by, and, by our own lights, you’re no more likely to be making a logical error than I, since we’re epistemic peers. We might shake our heads and say to each other–“One of us must be making a mistake here!” But if we both think our position is the stronger one, AND we both agree that we are roughly equally able to assess the cogency of philosophical arguments, where does that leave us? What REASON could I have in that situation to assume that you are the one making the mistake? Why shouldn’t I become agnostic pending further investigation?

  7. Carolyn Richardson says:

    Here’s an interesting similarity between the disagreement problem and the philosophical thesis that there are incommensurable worldviews: both must be formulated from the first-person standpoint. (This is what David is saying in his reply to Josh, when he points out that the difficulty arises because we must operate by our own lights.)
    The disagreement problem comes up when I (1) think that P, (2) think that some other person thinks that not P, and (3) think that that other person is equally liable to be right about whether P.
    The incommensurability thesis—the thesis that there are incommensurable worldviews—must likewise be formulated from within some worldview. If it were to be formulated from a third-person point of view *about* two other world-views, the very point of view from which the thesis is asserted would provide the terms for translation between the pair of worldviews (for the one making the assertion has recognized both *as* worldviews from within some umbrella worldview). When someone defending the incommensurability thesis wants to present evidence for it, she must (I think, anyway!) say something like this: “Those phenomena over there cannot be made sense of by us. Therefore, there are incommensurable worldviews.” And this is what leaves Kuhn open to what Davidson calls the “short line” of reply to the incommensurability thesis: “Nothing, it may be said, could count as evidence that some form of activity could not be interpreted in our language that was not at the same time evidence that that form of activity was not speech behaviour” (“On the Very Idea of a Conceptual Scheme” 185) So elegant.
    I don’t know what to make of this similarity, but there it is.

    • David says:

      –“The disagreement problem comes up when I (1) think that P, (2) think that some other person thinks that not P, and (3) think that that other person is equally liable to be right about whether P.”

      That’s a really clear and concise way of putting it, Carolyn. Thanks! And I think what makes the problem so interesting is that (1)-(3) describes my situation with respect to most of my beliefs–the important ones, anyway.

  8. Nates says:

    David, I agree that parochialism is a danger here. What we don’t want to say is that philosophers from different traditions should just ignore one another. And, in fact, it’s just false — it’s clearly often quite profitable to engage with people from different traditions. So the Kuhnian response ought to accommodate this fact. I can think of two ways this might work.
    First, there will be certain common beliefs held by different traditions. So, for example, utilitarians and Kantians might each derive, from their separate starting points, a social contract defense of a certain political institution. (Think of this as being like a mathematical theorem, derived independently from different assumptions.) From there, they can work together (to some extent) in exploring the further philosophical implications.
    Second, it’s useful to engage with other traditions because the strengths of their systems can reveal the weaknesses of one’s own (and vice versa).
    Of course, this second point leads to your next worry. Is this notion of a thriving tradition too weak to avoid the specter (sorry Josh!) of Rortian relativism? Again, I agree that this is a worry. Kuhn’s own response is to say that there are certain basic norms that cut across all traditions: accuracy, consistency, simplicity, etc. This gives us enough to objectively assess the value of a tradition (in the sense that one is obligated to prefer a theory that better satisfies these basic norms), even if it’s not enough to say that one is clearly true or false. I’m not sure if Kuhn has managed to thread the needle here, but I’m sympathetic to the approach.
    On the big assumption that the above works, we still have your third worry: can we really say that philosophical traditions make progress? I think that’s something I’d like to talk about in a separate post, so I’ll hold off for now. But, yes, something needs to be said here.

    • David says:

      “…can we really say that philosophical traditions make progress? I think that’s something I’d like to talk about in a separate post.”

      Looking forward to it! It’s a really interesting question…

  9. Nates says:

    Carolyn, I love that paper by Davidson! But my sense is that it doesn’t quite have the right target in mind. Kuhn always resisted the idea that incommensurability meant a kind of uninterpretability or inaccessibility of one world view from within another – even though some of his metaphorical talk of living in other worlds seemed to suggest this! Here it’s useful to recall the geometric origin of the metaphor: the idea that between, say, the radius and circumference of a circle there is no common measure: no unit that each line contains an exact number of. This doesn’t mean that we can’t meaningfully compare the two lines: in fact, we can define the constant pi that captures their relation as precisely as we want to.
    What’s the non-metaphorical version of this point? I think it’s supposed to be something like this: we can understand other traditions (perhaps by inhabiting them), and we can even assess their value (perhaps appealing to the meta-norms I mentioned above in my response to David). But what we can’t do is decisively assess their truth or falsity, because the criteria for doing so are (to some extent) tradition-dependent.
    Even as I write this, I realize that it all sounds a little wishy-washy. But this is where Kuhn was trying to get to in his post-Structure work. And I like the idea of applying this to our situation in philosophy. Obviously, much more would still need to be worked out!

Leave a Reply

Your email address will not be published. Required fields are marked *