A new emphasis in epistemology is burgeoning, known by the phrase ‘the epistemology of disagreement’. The object of investigation is what we will refer to as the epistemic parity situation, EPS for short. Suppose two people disagree about an important issue, and the disagreement is not due to one party's epistemic negligence or obliviousness to relevant evidence. Nor is the disagreement due to the improper function or the unreliability of a particular cognitive faculty. Here we have an example of EPS which can be formally stated, in the extreme, as a thesis about justification:
The justification S1 has for p is on an epistemic par with the justification S2 has for not-p, where justification is to be understood on either internalist or externalist lines, or on both lines.
The most central issue of the epistemology of disagreement revolves around the following question: if EPS happens, does it weaken one's confidence in, or support for, one's belief? We will see two other central issues: If it weakens one's confidence in, or support for, one's belief, should this reduction be best understood on internalist or externalist lines? If EPS does happen, how often does it occur? The issues treated by the epistemology of disagreement are not absolutely new. In philosophy of religion EPS has been the object of discussion as it relates to religious disagreement. This paper describes recent developments in the epistemology of disagreement both in mainstream epistemology and in religious epistemology. The main objective in the first two sections will be to provide background by bringing key points of contention to the surface in the recent epistemologies of disagreement, both in mainstream epistemology and in religious epistemology. A final section asserts that epistemic externalism in religious epistemology doesn't easily escape the challenge of EPS.
Recent developments in the epistemology of disagreement
We can start by flushing out the most central issue through observing points of contention first between Peter van Inwagen and Richard Feldman. The parity situation for Van Inwagen involves a situation where both epistemic peers are totally aware of all the publicly available evidence. Yet they still do not agree. Van Inwagen often refers to his disagreement about free will with David Lewis as an example of epistemic, peer conflict. For Van Inwagen, the state of EPS doesn't necessarily weaken one's epistemic position, because we often hold beliefs without basing them solely on publicly available evidence, for example in politics and philosophy.Footnote 1 He says he is inclined to think, but not willing to say for sure,Footnote 2 that public evidence and arguments adduced in support of one's belief don't constitute the totality of the justification for one's belief. And he is inclined to think that what justifies the retention of a belief in light of an epistemic peer holding an alternative belief is ‘some sort of incommunicable insight that the others, for all their merits, lack’.Footnote 3 We can call Van Inwagen's position public compatibilism,Footnote 4 where this refers to the compatibility of both a subject's rational belief retention and EPS; ‘public’ refers to the fact that the parity exits on the public level. Van Inwagen's public compatibilism allows him to conclude that both parties can rationally retain incompatible beliefs because of the evidence dimorphism between the public and private realms.
Van Inwagen advocates what we can call the segregation thesis, which can be formally stated as follows:
Something about one peer's circumstances of belief, which is not present in the other peer's, allows the belief to be sheltered from the epistemologically debilitating effects of EPS.
Van Inwagen's private insight not present in his peer's epistemic circumstance – for example, David Lewis's – allows his perspective to be segregated from the debilitating effects of EPS.
Richard Feldman would have to reject Van Inwagen's public compatibilism because the simple appeal to private evidence doesn't work in a true EPS where private evidences are acknowledged. Feldman grants that the peer can appeal to private evidence, like an insight. Feldman assumes that in a good disagreement both know when the opponent appeals to private evidence. The knowledge of the opponent's private evidence, however private it is, is itself evidence which must be discredited for rational belief retention.Footnote 5
A debate between Thomas Kelly and David Christensen revolves around the interpretation of evidence. Kelly thinks one can legitimately retain one's belief in light of EPS because of the distinction between general and specific parity.Footnote 6 Kelly considers an EPS over a particular issue where neither is generally smarter, better at reasoning, or better informed. Then, new evidence is brought in, and the two are asked to judge how strongly that evidence confirms or disconfirms a certain hypothesis. The two disagree. One peer can think the other has ‘misjudged the probative force of the evidence’.Footnote 7 Here we have an asymmetry in the particular situation that can justify one peer rationally to retain her belief even when there is general epistemic parity between the two. As Kelly says, ‘All I need to assume is that on this particular occasion I have done a better job with respect to weighing the evidence and competing considerations that they have’.Footnote 8 And this is similar to two chess players who have general skills on a par even though in each particular match one of them wins.Footnote 9 For Kelly, the disparity on the specific level segregates one from the debilitating effects of parity on the general level. We can call Kelly a general compatibilist; ‘compatibilist’ because he thinks one can rationally retain one's belief even while thinking there is epistemic parity on the general level; ‘general’ because the parity exists on the general level.
David Christensen's thoughts on the matter of EPS pose a deep challenge to Kelly's. Christensen challenges Kelly's advocacy of the segregation thesis, defined above, in at least two ways, one of which parallels Feldman's views. Christensen reasons that if each has general skills on a par, then one should think one's confidence in one's assessment of the probative force of the evidence is weakened. He offers an example of two people separately calculating for a group of friends after a restaurant meal how much each individual owes when the cheque is divided up equally.Footnote 10 The two both feel the other has equal skills for this process, and they come up with two different numbers. Each should take the peer's different result as evidence of a mistake. Generalizing from similar examples, Christensen thinks it doesn't make sense to think, if one truly takes the opponent as an epistemic peer, any interpretation one has shelters one from the debilitating effects of EPS. Christensen is an incompatibilist. Another reason Christensen is against Kelly's segregation thesis involves the feebleness of epistemic abilities. The fact that we often make mistakes should give us more pause when confronted by a peer who thinks differently.
Kelly's analogy of a chess match may seem a great counter to Christensen. Just as one of two grandmasters can win an individual game, so too one can have the right assessment of the probative force of the evidence in the particular situation even though generally there is epistemic parity among the peers. But, if the grandmasters are truly peers, it is not always clear whether winning any individual game is due to luck or slightly better skill. Maybe the opponent got distracted for a split second such that she didn't see the debilitating consequences of moving her castle. And the other capitalized on the accident which would have been avoided had a cell phone not gone off. Chess matches consist of a series of games in order to eliminate the possibility that any individual won by luck. So too one peer may have the right assessment in an individual case of probative force, but the parity would make the peer unsure whether the assessment is due to better skill, an accident, or a mistake.
The epistemology of religious disagreement addressed in philosophy of religion
The issues of EPS are not something new in philosophy of religion. For recent developments, one of the best places to start is William Alston's clear formulation of EPS, and the reactions to it of Philip Quinn and Alvin Plantinga. Alston describes an EPS resulting from religious diversity in what he calls ‘the worst-case scenario’.Footnote 11 In such a scenario Alston says there are no neutral and non-circular reasons for retaining one's own perspective in light of the alternatives. Because Quinn believes alternative religious perspectives can give extremely compelling arguments of their own, and because alternative religions can point to equally compelling religious experiences to back up their perspectives, Quinn thinks what Alston calls the worst-case scenario just is a common scenario when people take the epistemic challenge of religious diversity seriously.Footnote 12
It will be helpful in the discussion of the second, central issue of EPS to look at a debate between Quinn and Plantinga. Their fight over how to assess the epistemic challenge of religious disagreement revolves around the issue of whether or not the rational retention of one's belief requires, when presented with substantial reasons for thinking an alternative to one's religious belief is true, that one consciously give better reasons for one's own belief over the alternatives. Quinn says yes, Plantinga, no.Footnote 13 Basic beliefs, for Plantinga, get their positive epistemic status not from the good reasons or better evidence that we bring to bear for them, not from independent reasons, nor from anything else that we are necessarily aware of. Rather, positive epistemic status comes from the belief being formed by a properly functioning faculty. A belief can have warrant even though one has no way of consciously justifying it, cannot prove it true, and cannot display better evidence or reasons to support it.Footnote 14 Plantinga rejects the internalist demand to reassess one's views in the presence of religious alternatives.
Another debate between Quinn and Plantinga over what Plantinga calls ‘intrinsic defeater-defeaters’ will help us think about the third key question of the epistemology of disagreement: if EPS does happen, how often? Quinn's thought is that the kind of religious phenomena Plantinga thinks decisively promote belief-retention in times of the most difficult challenges are rare. When challenged to defend his exclusivism, Plantinga refers to the phenomena he calls intrinsic defeater-defeaters. These are beliefs whose non-inferential conditions of occurrence are so vivid and compelling that they defeat even the most difficult challenges to one's belief.Footnote 15 Carrying this intuition over into the religious realm, Plantinga believes Christians can, in the face of religious alternatives, have intrinsic defeater-defeaters with enough non-inferential warrant to allow them to hold firm to the beliefs produced in them trusting them to be produced by a properly functioning process, unless the proper function of the process is in question. Quinn thinks intrinsic defeater-defeaters are present in extreme cases (like Moses experiencing the burning bush), but they aren't nearly as prevalent as Plantinga thinks.Footnote 16
It will be helpful to follow a debate within religious epistemology which parallels, but also diverges from, the Kelly/Christensen disagreement. David Basinger for the most part rejects the segregation thesis, described above, because of what has become known as Basinger's rule, namely, that ‘if a religious exclusivist wants to maximize truth and avoid error, she is under a prima facie obligation to attempt to resolve significant epistemic peer conflict’.Footnote 17 In other words, when confronted by a sincere and knowledgeable peer who disagrees one is under a prima facie, epistemic obligation to attempt to resolve such conflict through a comparative assessment of one's epistemic position in relation to the other, if one is interested in maximizing truth and avoiding error.Footnote 18 In the face of peer conflict one must engage positive, and not just negative, apologetics, where ‘positive’ refers to the presentation, and evaluation, of evidence, and ‘negative’ refers to simply refuting challenges.Footnote 19 Basinger gives a version of the reassessment demand.
Another reason for reassessment parallels Christensen's view of the fallibility of our knowing. Basinger points out how people often change their views on important matters even though the earlier beliefs were considered ‘rock-bottom’, to use Jerome Gellman's phrase for foundational beliefs accepted without argument.Footnote 20 We do this because peer conflict gets us to look back on our rock-bottom beliefs, and often we see they are not as well founded as we thought. Basinger's rule often leads to extreme challenges for many because, once the reassessment demand is engaged, many find there is no non-question begging way of showing one's view superior.Footnote 21
Jerome Gellman says that in the face of peer religious conflict it is perfectly legitimate that one not engage in reconsidering the rationality of one's belief, that one's beliefs are not lowered in justification as a result of unresolved peer religious conflict, and that it is perfectly legitimate that one's rock-bottom beliefs stay rationally unreflective.Footnote 22 No prima facie reassessment obligation exists. There is absolutely no need to engage in reflection about the justification of one's belief given the sheer presence of religious diversity. Considerations about the fundamental nature of our most basic beliefs lead Gellman to this pro-segregation stance. He asserts a Wittgensteinian point that ‘a person cannot subject all of his beliefs to critical scrutiny. In order to begin to engage in reflective scrutiny a person must leave some beliefs as unreflective’.Footnote 23 Also, if the standard for retaining one's beliefs is to produce non-question-begging consensus, then no belief retention would be justified, since there simply is no rational consensus on how to weigh criteria of rationality.Footnote 24 Third, he asserts it is completely rational to start believing based on the training and enculturation one receives growing up.Footnote 25
Because of these considerations he affirms the legitimacy of the following principle of conservatism his grandmother taught him, ‘If the wheel does not squeak, don't oil it’.Footnote 26 Gellman interprets his grandmother to be saying something like the following: If there are no grounds for losing confidence (squeaking) in a belief, one is rationally entitled not to have to subject the belief to reflective scrutiny. A belief ‘squeaks’ when one's confidence in the belief is challenged. Like the wheel that doesn't need to be attended to with oil, the belief that doesn't ‘squeak’ doesn't require any attention, that is, legitimately remains retained without reflection and assessment.
Listening to our grandmothers
Externalists are not off the hook, given the difficulties pertaining to the epistemology of religious disagreement. They may think that one only falls into problems when internalist demands are required, such as the demand for rational reassessment. Externalists see themselves as fortunate they avoid what they see as the quagmire, and scepticism, people get into with reassessment demands. Externalism combined with doxastic conservatism provides a challenge to anti-segregation theses; but, I will claim, not one that is often effective. Belief retention in light of alternatives is a pretty important thing. Life would be very difficult if we didn't conserve beliefs in light of opposition. At every moment we would be changing our minds or re-evaluating our positions. On the other hand, hardly any externalist thinks epistemic conservatism is a license to ignore completely challenging new information. Both these concerns lead to what some call the principle of doxastic conservatism, which can be stated as follows in a way consistent with the principle of conservatism of Gellman's grandmother:
If there are no good grounds for questioning the proper function or reliability of the belief's specific formation process, then the belief should be retained.
Notice this principle of conservatism isn't a license to retain every belief. The antecedent is affirmed only if no grounds come to light to question the proper function or reliability of the belief's specific formation process. Even though externalists don't think anything that we are necessarily aware of justifies our beliefs, still most externalists recognize that the beliefs we retain have to hold up to challenges.Footnote 27 This principle of conservatism doesn't engage in any activities offensive to externalists, since there is no need to know the reliability of the specific formation process. All that is required is that one be able to rout challenges to reliability. One only needs to engage in negative apologetics where one defensively defuses attacks on the reliability, rather than in positive apologetics where one must both bring to mind one's reasons for believing and reassess them in comparison to the opponent's reasons. One can defuse attacks on the reliability of one's belief-forming process without having to know its reliability, just like a good defence lawyer can defuse attacks on his client's character without having to prove his client's character admirable.
We can formulate a conservatism defeater that will tell us when it is appropriate to deny the antecedent of doxastic conservatism:
If subject A can think of no way in which A is generally better epistemicly situated for a belief than B – who holds an alternative belief – in terms of externalist criteria like reliability, then there are good grounds for A questioning the reliability of her belief's specific formation process.
This defeater shouldn't offend any externalist sensibilities. No demand is made that the externalist prove her belief more reliably formed, or that one bring to mind for evaluation all the reasons or processes contributing to belief formation. I can't think of any externalist who would say there would be no grounds for questioning the reliability of a belief-forming process, if two indistinguishable processes produce incompatible results. For example, radically different readings of two of the same electric thermometers, one mine, certainly give me grounds for questioning the reliability of my thermometer. In this case one doesn't need to know the reliability of all the processes that took place in the two thermometers in order to question the reliability of my thermometer. So too one doesn't need to know during EPS the reliability of the processes that produced one's belief in order to question the reliability of one's belief-forming process.
The fact that there are two indistinguishable processes that produce incompatible results makes one question reliability. Here we have an attack, based on the introduction of an indistinguishable separate process, on the reliability of one's own belief-forming process, an outside attack which must be discredited for negative apologetics to succeed. The defence lawyer practising negative apologetics must attempt to discredit a witness who discloses a testimony conflicting with that of her client, since if the two testimonies are assessed on a par, then the client's testimony falls into question.
There are at least two ways of dismissing the conservatism defeater. A detour method accepts it and goes around it claiming its antecedent not necessarily fulfilled. Here we will mention three detour strategies. Plantinga uses this strategy when he asserts there really is no good grounds for thinking the opponent's faculties are functioning just as well as mine. He acknowledges, for the sake of argument, internalist parity, but refuses to admit externalist parity, since he thinks the Christian can easily refer to the reliable sense of the divine faculty as providing the needed symmetry breaking. The sense of the divine is a ‘disposition or set of dispositions to form theistic beliefs in various circumstances’.Footnote 28 Plantinga thinks the sense of the divine (sensus divinitatis) is generally functional especially in those who honestly and intensely seek after the truth, and who are moral. He says,
According to the extended model, we human beings typically have at least some knowledge of God, and some grasp of what is required of us; this is so even in the state of sin and even apart from regeneration. The condition of sin involves damage to the sensus divinitatis, but not obliteration; it remains partially functional in most of us. We therefore typically have some grasp of God's presence and properties and demands, but this knowledge is covered over, impeded, suppressed.Footnote 29
Secondly, one could go with Gellman and see rock-bottom beliefs as largely unrelated to faculties, whether functioning reliably or not.Footnote 30 The conservatism defeater is often discussed in terms of faculties. Yet, for Gellman, religious beliefs are primarily formed by absorbing beliefs one is trained into as part of one's cultural and family contexts, not by some faculty. No need to take up the gauntlet thrown down by the faculty-centred conservatism defeater. Thirdly, one could use a strategy Alston and GellmanFootnote 31 share in common, namely, the view that, since there are often no common criteria for determining who is right in religious disagreement, it makes no sense to lose confidence in one's own view.
But detour strategies are unappealing for many who rigorously take religious diversity seriously. My heart and mind tell me positions like Plantinga's, Alston's, and Gellman's simply don't take the intelligence and rigorous truth-seeking motives – not to mention moral fortitude – of alternative religious practitioners seriously enough. The Buddhist, Muslim, and Hindu friends I have search so honestly and intensely after the truth, are so careful about attending to the evidence, have criteria for evaluating evidence so similar to mine, are so good at explaining how people who disagree with them have fallen into error, and are so morally upright, that I simply cannot think that a generally reliable faculty is functioning improperly in them such that they come up with wrong beliefs; I cannot think that they have been inattentive or slothful.
I and many others find ourselves in a similar position as Van Inwagen found himself in the presence of David Lewis. My friends' backgrounds are similar to mine; they read many of the same journals, and attend the same conferences. I cannot say what epistemic skill I have that they lack. With respect to any of the general cognitive and environmental conditions I am aware of for reliably forming true beliefs, I don't find that I have a clear advantage over my religious peers. Sometimes they are wrong. But, sometimes I am wrong too. Christensen and Basinger highlight, respectively, the feebleness and fickleness of our individual efforts to know. To recapitulate: if there were a widely available process for knowing God generally functioning properly, then the epistemic peers from other religious traditions are so sincere, rigorously truth-seeking, open, and moral that they would find and acknowledge it. But they don't, and so this leads me and many others to think there is no built in faculty for knowing God generally functioning properly. There may indeed be a slight general difference in the epistemic situations that differentiates the two peers' beliefs. But the difference is not able to be resolved by the individuals.Footnote 32 I personally think the Fall was much more severe than Plantinga's estimate.
The Alston/Gellman strategy of advocating belief retention based on the incommensurability of evaluation criteria among religions isn't appealing when considering disagreements with common evaluation criteria. The Buddhists I know, many of them Christian-raised, believe they would be open to experiencing the sense of the divine or the creator God, if they existed. And we can agree on the conditions for determining whether something like the sense of the divine is working. My Muslim friends and I can agree on the conditions that need to be fulfilled for one to believe rationally that Jesus Christ died on a cross. It is just that I do, and they don't, think the conditions are fulfilled. Surely some religious disagreements can be dismissed as due to incommensurable criteria for determining who is right and who is wrong; but just as surely there are many important religious disagreements that can't.
Gellman should have more hesitation, in light of the epistemic challenge of religious diversity, about declaring that religious belief-formation is mostly a matter of enculturation and inferences from the absorbed beliefs. He believes the process of taking over beliefs through enculturation completely rational. But, presumably other religious practitioners go through a similar process of enculturation, and they come up with radically different conclusions. We have already seen how there are often many similarities between one's belief formation process and the others’. And now we add even more parity when we think of the very similar enculturation processes among different religious practitioners. Gellman would have use believe comparative reassessment, or even questioning reliability, is superfluous for the religious believer who has, due to enculturation, right from the start maintained the truth of her rock-bottom beliefs. ‘If one already has the truth the attempt to find the truth would be superfluous’.Footnote 33 Unless we can find some formal epistemic difference between the Hindu and Christian enculturations, we appear arbitrary in treating like processes differently.
Gellman's ‘superfluous’ stance seems as strange to me and others as would the following situation. John goes to the supermarket for the shampoo his parents taught him is the best for his particular hair. Next to the shampoo he always gets, usually without thinking, are two others that claim they are the best for his particular hair type. He glances at one of them, which says a particular chemical in the one he has always bought has been shown in studies to be significantly associated with some unpleasant problem that develops later in life. In every way he can immediately think of the alternative seems legitimate. It is put out by a reputable company, has undergone the required testing, and has raving testimonials from satisfied parents just like John's. It now would seem strange for John to ignore the alternative thinking it superfluous to reassess his shampoo choice on account of the fact that he has already assumed he has the best shampoo. Using Gellman's grandmother's analogy, the wheel seems to be squeaking at this point, and in need of oil.
Rather than detour around the conservatism defeater, a second way of dealing with it faces it head on rejecting it on grounds that the implication it promotes is incorrect. Just because there is general epistemic parity doesn't mean this should negatively influence one to doubt the reliability of one's belief-forming practices in specific situations. For a number of things can segregate one from the debilitating effects of the conservatism defeater. We can also call these segregation strategies intrinsic defeater-defeaters. We have seen different versions of the segregation thesis, van Inwagen's incommunicable insight, Kelly's assessment of probative force, and Gellman's rock-bottom beliefs. The varieties of segregation theses, defined above, are designed to break the implication from EPS to mistrusting the reliability in specific cases.
The problem with implication-braking, segregation strategies is that more often than not the attempt at segregation is frustrated by the rigorous assessment of general parity. Christensen and Basinger have said something similar. But I will say this in a way that self-consciously avoids offending externalist sensibilities. Nearly every time someone feels compelled to refer to an exclusive source of reliable belief formation in a specific case in which a peer vehemently disagrees, there is insecurity as to whether the compelling reference to the special source (whether an insight, assessment of probative force, or feeling) is a mistake or an accident. Unless I can see that there is some general, epistemic asymmetry I tend to think my compulsion might be a mistake or an accident. I either think that it is an extreme accident that I alone have the compulsion, since my peer is apparently equally well situated; or I think my compulsion is a mistake, since my peer has a very good story about how it is that I came falsely to have this compulsion. I can't rule out these options, given the general parity. I have no better reason to think the error is my peer's rather than mine.
In a chess match between grandmasters so equally skilled that it is extremely difficult to say who is better, it is not sure whether an individual win is due to a mistake, luck, or a slight advantage in skill. Perhaps grandmaster A made a poor move that B didn't anticipate just because it was such a bad move, and the mistake ironically gave A an advantage in the particular game. Or perhaps grandmaster A just got lucky in a particular game. Or perhaps grandmaster A won because she has slightly better skills overall. More than one game has to be played to determine who has slightly better skills. The same happens in religious disagreements among peers. The point is, whether mistake, luck, or slightly better skills, general epistemic parity is sufficient grounds for questioning the compulsion to believe one's view is more likely to be right. Nearly every attempt at segregation, whether due to an insight, assessment of probative force, or feeling, is frustrated by the general parity.
Having said this, a qualification needs to be made. The infrequent and intense vividness of some ad hoc, belief-forming processes can serve as intrinsic defeater-defeaters of the conservatism defeater. Sometimes the belief wasn't formed by osmosis or some other relatively passive mechanism such as enculturation. Sometimes the circumstance of belief formation can be so unusual, meaningful, and intensely vivid that its vividness lives on and decisively informs one later in peer conflict. The experiences are so vivid because they are unusually formed. Because the intrinsic vividness can serve to defeat the conservatism defeater, we can call this the unusual, vivid, intrinsic defeater-defeater, UVIDD, or UV for short. Here the vividness cannot be due to a generally reliable faculty, since my peer would discover it if it were.
Let me give an example of relatively passive versus active/vivid belief acquisition. Most of my students, when I press them why they believe that Jesus Christ died on a cross say something like the following: I believe because that is the way I was raised. This is a passively attained belief; it is something they have taken on and have assumed without thinking too much. However, sometimes I encounter students who say they believe Jesus Christ died on a cross because they had an intense experience of the crucified Christ. They can describe the time and the situation in which they had this compelling belief.
The conservatism defeater says general externalist parity means most can't decide whether they are lucky or mistaken during EPS. But the UV gives one a slight edge in peer conflict for being able to both acknowledge general EPS and think one is lucky rather than mistaken. One has access to a vivid ad hoc situation one's peer doesn't. Notice that this slight edge in the conflict doesn't destroy the general parity one has with the peer because the luck is ad hoc. One peer got the vivid experience not by possessing some better skill, but rather by the grace of Allah, or some other religious entity. If I have a UV my opponent cannot easily convince me that I didn't have the vivid experience, or that I am interpreting it improperly, since the belief comes from a specific, unusual process he was not graced with. I wouldn't expect him to be able to take my unusual experience as seriously as his own experiences. If my peer were so graced, he would likely have my belief, given that we are both just as good at evaluating our experiences. The UV segregates one from the debilitating effects of EPS.
The UV strategy proposed here is different from other segregation strategies we have seen. There is no reason why the UV could not be communicated. I have students that sit in my office and describe their intense and vivid experiences of the crucified Christ. Nor is the UV like Kelly's better assessment of probative force. In Kelly's view there is a common body of evidence weighted differently. In the UV the experience is not generally accessible to my peer. It must be stressed that UVs are rare. More often than not when people take religious diversity seriously they find they have beliefs without the vividness needed to overcome the conservatism defeater.
Notice also that the appeal to a UV need not necessarily offend externalist sensibilities. There is no attempt to evaluate the reliability of the vividness, to bring to mind all the reasons why one should accept the vividness at face value, or to evaluate the processes that lead to the vividness. The UV is simply used to give one's own belief a slight edge in an otherwise symmetric situation, like the defence lawyer who cannot discredit an alternative testimony, and so focuses on showing his client's witness based on a vivid experience. It must be stressed that the UV is not absolute. Under some conditions even the best UVs are blocked, like a good sunscreen blocks UV waves. The person who says they just saw the Virgin Mary after dropping acid probably ought not be trusted.
Obviously, many problems are raised with the location of externalist resistance to segregation defeaters in UVs. For one, this seems to open up the possibility of a relativism where people with many different beliefs refer to special lucky situations of grace to legitimate their beliefs. In this possibility we see a major difference between philosophical and religious epistemologies of disagreement, since the former seems to be able to appeal to more sources of the luck than the latter. I don't know of any philosophical epistemologists who would refer to the ad hoc influence of Allah or Shiva. But any appeal to a special, lucky situation of grace-forming belief is limited by the requirement of general epistemic parity.
A second challenge to this approach asserts enculturation is much more extensively and pervasively at work in UVs than my distinction between vividly and passively attained beliefs suggests. Even vividly attained beliefs show the influence of enculturation. And if they are so enmeshed in a tradition it would seem circular to use them as a basis for retention when engaged in religious disagreement with an epistemic peer from another tradition, that is, when engaged in religious EPS. As an anonymous reviewer of this essay reported in an e-mail, ‘Christians … normally have vivid Christ experiences, and not non-Christians’. Vividly formed, experientially based beliefs seem to loose their efficacy when thinking about the pervasiveness of enculturation, efficacy for belief retention in light of EPS.
To defend the position on the efficacy – even though they are infrequent – of UV intrinsic defeater-defeaters as distinct from passively attained beliefs, consider an example of the difference between the two. Bill feels comforted when thinking about Christ redeeming him. And when he contemplates where this feeling comes from he can't think of any specific experience. He thinks he took over this belief from trusting the environment in which he was raised. Now consider someone, Mary, who has very vivid experiences of Christ. Mary vividly remembers specific experiences when she felt enveloped by the redemptive power of Christ. She experiences this as something coming over her not due to her own power. Mary, like Teresa of Ávila, argues that it is not just her enculturation as a Christian that makes these vivid experiences possible. While Bill's feeling about the redeeming Christ comes from trusting the environment in which he was raised, Mary's comes both from initially trusting her social environment but also from UV experiences which seem to confirm the initial trust.
Though Bill and Mary participated in very similar enculturation, they would respond differently to someone who says, ‘You only have these Christ feelings because you were raised in a Christian environment’. Mary, and not Bill, can say she has vivid experiences that seem to confirm the initial trust in her enculturation. Mary may be only deceiving herself into thinking that her experiences corroborate her enculturation when they are really just deep manifestations of it. But, prima facie, her UV experiences seem to confirm, rather than simply manifest, her enculturation – not the other way around. It is this prima facie efficacy of the UV that can serve to decisively help one retain one's belief during EPS, even though UV experiences are infrequent.
As stated earlier, intrinsic defeater-defeaters are never absolute. Our prima facie, UV seemings may turn out false. Even when it seems a UV confirms one's traditional perspective, it may be the case that one is just deceived by the traditional context. My father used to believe African Americans are mentally inferior, and he believed he had specific and vivid experiences to confirm his view. But, now he thinks that African Americans are just as mentally capable as other races, and he attributes his earlier belief along with the seeming corroboration to the corrupt influence of his social environment. Similarly, while I seem to see a table in front of me, someone may have earlier slipped a hallucinatory drug in my soda such that the table is an illusion. But, as pointed out earlier, externalists using the principle of conservatism have a certain, though not absolute, resistance to the challenges of sceptical alternatives. Since what justifies a belief for the externalist isn't something that one is necessarily aware of being reliably formed, externalists don't have to prima facie know the reliability of their belief forming processes in the face of alternative perspectives. We would be in very bad shape – so externalists often criticize internalists – if justification required us to know the reliability of all our belief forming practices.
At the same time conservative externalists are not totally immune from challenges. Considerations can come along that defeat one's UV. Someone can bring evidence before me that a drug was slipped in my soda such that I would not be able to distinguish between a world where there is a table in front of me and one where I was only hallucinating such a thing. And my father changed his mind because of evidence brought before him that African Americans are just as mentally capable. Earlier we specified one mechanism by which externalist conservatism is often defeated. But, unless alternative considerations are compelling the externalist using the principle of conservatism and UVs believes she is epistemically responsible prima facie in retaining her beliefs even while knowing that there are some situations where she would not be, and even while not being able to prove that her UV is not simply a product, rather than a confirmation, of the social conditioning. We don't insist in our daily lives that people be able to rule out all the alternatives, like that we are in a computer-simulated world similar to the one in ‘Matrix’. So why should we insist people do this in their religious lives?
A third challenge to this approach is that it simply pushes the problem back just one more notch. Say we have two peers who both appeal to UVs for supporting incompatible beliefs during general EPS. Call this second-level EPS symmetry, where first-level EPS symmetry consists of the EPS where there are no UVs involved. For an example of this second-level symmetry, take a Christian who, in EPS with a Muslim, appeals to a UV for thinking Jesus died on a cross and a Muslim who appeals to a UV that Jesus didn't. It would seem one should treat indistinguishable processes similarly, and this means one cannot again favor luck over a mistake. In this difficult challenge two things can be said. First, since the appeal to UV in EPS situations is infrequent, the meeting of two people who make such appeals that support incompatible beliefs would be rare indeed. Second, it is more difficult for the person in second-level symmetry to think her opponent, with his respective UV, is in just as good a specific position to know the truth.
I want to say the tremendous epistemic inertia of one's own UV in rare second-level symmetry situations can legitimately compel one slightly more to think one is lucky rather than mistaken, especially given that one is not sure any doxastic change would be any more likely to be right. For example, the Christian in a rare second-level, UV symmetry situation with a Muslim has inertia for thinking she is more likely right about Jesus dying on the cross than her Muslim opponent who also claims a UV driving his belief that he didn't. I realize the appeal to ‘epistemic inertia’ isn't very convincing. But, the person with a UV in second-level EPS symmetry would have an excruciatingly difficult time turning away from the decisive vividness of her UV in order to think her opponent's UV qualifies him to have just as much purchase on the ability to get at the truth of the matter. Epistemic inertia may not sound very convincing on paper, but in real life it plays a big role in belief retention in EPS religious disagreements with and without UVs.
Much of what has been said about externalist conservatism can be read as a careful application of the wisdom of Gellman's grandmother. Recall that she said, ‘If the wheel does not squeak, don't oil it’. The absence of squeaking is a sufficient reason for not subjecting a belief to scrutiny. My also wise grandmother said something similar to Gellman's: ‘Don't mess with success’. In light of the fear of scepticism that externalists drill into us, we would be wise to listen to our grandmothers. Gellman's grandmother tells us one pathway to retaining our beliefs, a pathway only available if there are no grounds for challenging confidence in a belief. Though this pathway is available for those with UVs in EPS, I have argued this pathway is not generally available.
Gellman's grandmother tells us what to do if it doesn't squeak, not what we should do if it does. Before I mowed my grandmother's lawn, she frequently said, ‘If the wheel squeaks, then oil it’. She never elevated this conditional to a general epistemic principle. But there is no reason why we couldn't take this to mean the following: If one loses confidence in a belief, then there are grounds for questioning it. We have seen that EPS usually causes one to lose confidence in one's belief. And this frequent loss of confidence means that doxastic conservatism usually cannot be used as a way of retaining one's beliefs. Externalists are not off the hook given the difficulties pertaining to the epistemology of religious disagreement.Footnote 34
