1. Disagreement and Epistemology
Disagreement is philosophically interesting. In particular, it is interesting to those looking to analyze the norms of belief and knowledge. Disagreement is, then, epistemically significant. Disagreements come in many forms—there are disagreements between the more informed and less, and there are disagreements between those with good track records on the issue and those with less-than-good. In these cases, it is relatively uncontroversial that those with more information and people with the better track records are in the better position in the disagreement cases. The matter, then, is that disagreements are occasions for some theorizing about longstanding issues in epistemology: What is it to be well-informed? How does one non-question-beggingly identify who has the better track record? Many cases of disagreement are like those, where one of the opposing subjects is in a better cognitive position than the others. And the issue there is just a matter of articulating what that is and how we determine it. These sorts of cases of disagreement are significant for the study of knowledge, then, but only because they occasion thought on epistemic concepts that don’t have anything directly to do with disagreement.
In contrast, there are disagreements that are epistemologically interesting because they are freestanding from these other concepts. What if there were not any obviously better-placed subject? What if they had roughly the same track record on the issue? What if they are, for all intents and purposes, intellectual equals on the matter? Call such individuals epistemic peers—they are equally informed, of the same intelligence, have no more facility of managing the information than the other, and they are both unbiased on the issue to the same degree. There is nothing we can point to that would justify us (or them) to say one is in a better position than the other. How these sorts of cases should be handled is the problem of peer disagreement.
These cases of peer disagreement are, for sure, less frequent than that of non-peer disagreement. In fact, these sorts of cases are idealized to the vanishing point of equality, but this idealization is performed precisely for the reason of maintaining the distinctness of the problem they pose. In peer disagreement cases, the subjects have a kind of informed symmetry between them, where one does not have an edge, however slight, on the standing evidence or some intellectual capacity. Peer disagreement cases, then, despite being less frequent than their non-peer counterparts, are of considerable interest for epistemologists. The first reason is that, despite the idealizations, intellectual humility demands that one’s defaults are not set on holding that one is always the smartest or most informed person in the room. For sure, there are times when we are the teachers and experts, but among well-informed and intelligent people, this is often only temporary. Good teachers produce students of intellectual quality on a par with their mentors. The second reason why peer disagreement is of interest is that it with the idealizations, it is the disagreement itself that we plumb for significance, not what other epistemically significant factors that may contribute to them. And it turns out that disagreement in these cases yields significant consequences.
The main question in the epistemology of disagreement is how one should believe when faced with peer disagreement. In cases of asymmetric positions, belief should follow the better-placed view. So when one is in the better position, one has a reason to remain steadfast. And when one is in the worse position, one has a reason to change one’s mind and adopt the other’s (better-positioned) view (or at least drastically reduce one’s confidence in one’s own). But in peer cases, the epistemic position is symmetric—neither is better positioned.
There are two classes of answers to the question of how to handle peer disagreement: Steadfast views and Conciliatory views. Steadfast views are that one should stick to one’s guns in peer disagreement, and Conciliatory views are that peer disagreements should weaken one’s confidence—to a degree, or totally.
The two kinds of answers to the peer disagreement challenge have particular appealing features, and they have particular costs. Steadfast views, on the one hand, allow one to trust oneself and prevent others from vetoing how things, despite the disagreements, seem to us. But they risk, all too often, reverting to asymmetric views by holding that one has some particular unique piece of private evidence or that others are just systematically misled in these cases. So they are only apparent peers. So Steadfast views either aren’t solutions to peerhood cases or are just dogmatic.
Conciliatory views, on the other hand, are that peer disagreements provide us with reason to mitigate our confidence. And on the Equal Weight version of the view Jon Matheson defends, our peers are just as likely to be right on the matter as we are. Were we to see the disagreement from a third person perspective, we would say that there is no reason to prefer one subject’s judgment over the other’s. And so, when one is aware that one is party to a peer disagreement, one should “split the difference” with the disagreeing peer—one should adopt the doxastic attitude halfway between the two. Conciliatory views clearly exhibit the ethic of fallibilism and humility one prizes in anti-dogmatic belief. However, Conciliatory views have costs. First, it seems one’s beliefs are not weather vanes, indexed not to evidence, but to the opinions of others. Second, it seems that a form of skepticism looms, as disagreement is widespread. Third, and most metaphilosophically interestingly, it seems the view can’t be held in a debate about peer disagreement with peers—if one is a Conciliationist, then one should stop being one when one meets someone who is committed to the Steadfast view.
2. Matheson on Disagreement
Jon Matheson defends a particular version of Conciliationism in his The Epistemic Significance of Disagreement. Matheson states his view:
Awareness of widespread disagreement on controversial matters in religion, philosophy, politics, and science has the consequence that we are much less epistemically justified in our controversial beliefs and that we often are not epistemically justified in holding them at all. (2)
As a consequence, Matheson is defending a view of peer disagreement that accepts the contingent skeptical consequences of the view. Matheson is not saying that we have no justification in widespread cases or that those conditions are essential to being the kind of cognizers we are. Rather, Matheson’s view is limited to a restricted class of cases of disagreement that may or may not obtain for us in some domain or other. And this view is one that obtains even in cases wherein we think we have excellent reasons for our views, since peer disagreement functions as an undercutting defeater for one’s justification, and “the strength of one’s justification for believing a proposition cannot make it immune to defeat” (101).
Matheson has two discussions of disagreement, the first is restricted to what he calls idealized cases and the second to everyday disagreements. In idealized cases, there are only two parties, they are peers, they’ve shared all their evidence, and there are no others involved in the disagreement. In everyday cases, it’s harder to identify whether they are peers or not, but it’s not clear, either, who’s in the better epistemic position. Further, in everyday cases, information-sharing is incomplete and the disagreements are widespread and over long periods of time. And so there may be many parties over generations.
In both idealized and everyday cases, Matheson is willing to accept the skeptical consequences of his Equal Weight View. In fact, in the case of groups disagreeing, Matheson is willing to extend the relevant subjects of the disagreement not only to the dead, but also to the evidentially counterfactual:
If you have good reason to believe that someone with the same first-order evidence . . . could very well have disagreed with you, then you have good reason to be less confident. (133, emphasis added)
And so that a contrary view could very well have been formulated and held is a relevant consideration in moderating our degrees of commitment. As a consequence, Matheson “recommend(s) being skeptical about controversial propositions” (135), where controversiality doesn’t just mean that there is well-informed disagreement, but also that there very well could have been.
These skeptical consequences might be found surprising and objectionable. One, perhaps, lacks the courage of one’s convictions, that to have convictions is to stand for them, and even dig in one’s heels and swim against the current. To this, Matheson reminds us that “after all, humility is an intellectual virtue as well” (140). Further, Matheson reminds us that the view is not a global or radical skeptical view, but one that is contingent on what’s controversial, and the skepticism is limited only to those domains. Finally, Matheson even concedes that, though the view runs that one may not have good epistemic reasons for one’s views and actions in light of the disagreement, “this does not entail that there are other reasons (pragmatic, moral, religious, etc.) to hold a controversial belief” (144). And further, there may be simple instrumental reasons why the belief can be held. And so, though skeptical conclusions follow, prohibitions on believing or acting are not necessarily part of the outcome. In fact, Matheson closes his book with an argument for Moral Caution—namely, that given the skeptical epistemic consequences of disagreement, many of our decisions in moral matters are made under conditions of ignorance. Consequently, we should opt for the view or plan of action that avoids the clearest morally problematic possibilities (163).
It is apt that a book on disagreement is a cause for critical discussion. Our four respondents, Diego Machuca, Amber Carlson, Chad Boghosian, and Nathan Ballantyne, have a number of points of contention.
As noted earlier, Matheson’s view is a kind of contingent skepticism, and Diego Machuca is quick to note Matheson’s being in a distinct minority of those working on disagreement. Very few think skepticism is a plausible view. In fact, observing that a view yields skeptical outcomes, all too regularly, is taken as a reductio. Machuca, like Matheson, believes skepticism is not just plausible, but probably right. But Machuca thinks that stronger skeptical consequences follow from Matheson’s view than Matheson allows. Matheson regularly notes that he “softens the blow” of the skeptical challenges by restricting them to very controversial views, or by allowing non-epistemic reasons to count in favor a views in some cases, and by appealing to others in the disagreement with “epistemic elections” (153). But Machuca’s challenge is that in cases of deep disagreement, controversiality is in the eye of the beholder, that disagreement (if a defeater) can cut down even the most high-quality of reasons, and, finally, that the pragmatic and other reasons aren’t very plausible. Matheson’s mitigated skepticism, Machuca holds, should be more powerful.
Amber Carlson’s reply proceeds along two lines. First, Carlson is skeptical about the idealized notion of the epistemic peer. It is not clear there are any real epistemic peers. Moreover, if there are not any (or are vanishingly few for any of us), Carlson asks whether invoking them should have any significant bearing on how we manage our real-life disagreements. Second, Carlson is skeptical about whether it is so difficult to determine who is in the better epistemic position with many real-life disagreements. With many cases, minimal investigation can turn up evidence of asymmetry. As a consequence, Carlson has doubts about whether our defaults should always be set on taking every disagreement as symmetric.
Chad Bogosian takes particular issue both with Matheson’s Equal Weight view and its skeptical consequences. Bogosian notes that in everyday cases, there is incomplete access to evidence on both sides—one side may have relevant experiences and saw what they saw, while others may just have very good testimony. When there is conflict, Bogosian holds, it is right that those with the experience believe what they believe and those with the testimony believe in accord with it. The consequence, then, is that there are reasons to be steadfast in many everyday cases. Further, Bogosian notes that other epistemic goals are on the docket in disagreement cases, particularly, the stability and coherence of one’s beliefs. These, Bogosian holds, can be defeaters for the potential defeaters posed by peer disagreement, because diachronic considerations of contributing to one’s future character can override other synchronic epistemic goals. Given these other considerations, Bogosian argues, Matheson’s skepticism (however mitigated it may be) need not follow.
Nathan Ballantyne is, for the most part, sympathetic with Matheson’s aspiration of connecting epistemology in the idealized vein with practically guiding our commitments in controversial areas. However, Ballantyne urges caution when we turn from lessons in ideal cases to everyday advice. First, Ballantyne notes that the bridge between theoretical principles in the ideal and our following them in the hustle-and-bustle of our lives is thin. There are too many who’ve posed very impressive theoretical programs, but who’ve failed to follow them in their lives. Second, what connection there is between the two needs to be explicit—how intellectual principles can guide us must be properly theorized in order for us to understand how philosophy yields action-guiding views.