EPISTEMIC CIRCULARITY AND COMMON SENSE: A REPLY TO REED
When one depends on a belief source in sustaining a belief that that very belief source is trustworthy, then that belief is an epistemically circular belief (EC-belief). A number of philosophers have objected to externalism in epistemology on the grounds that it commits one to thinking EC-beliefs can be justified, something they view as an unhappy consequence for externalism. In my 2004, I defend externalism against this sort of charge by explaining why this consequence needn’t be an unhappy one. In the course of doing so, I appeal to what Thomas Reid calls ‘common sense’—a faculty or belief source by which we know noninferentially such things as that our faculties are trustworthy. In his $$$$, Baron Reed raises what he takes to be serious objections to what I say about both epistemic circularity and common sense. In what follows, I’ll respond to his objections, explaining why I side with Reid against Reed.
I. Epistemic Circularity
In my 2004, I say three things in defense of the possibility of justified epistemically circular beliefs. First, and most importantly, I give an argument for the view that not only externalists but everyone who thinks justified belief is possible is committed to thinking EC-beliefs can be justified. Second, I offer an explanation for why it is so tempting to mistakenly think epistemic circularity is a bad thing: I distinguish two kinds of epistemic circularity—malignant and benign—and explain why philosophers tend to focus on the malignant kind and ignore the benign, which is why they wrongly think all epistemic circularity is a bad thing. Third, in order to counter the impression that the only serious proposal for how we in fact form justified EC-beliefs in the trustworthiness of our faculties is the implausible one according to which we do so inferentially on the basis of epistemically circular track-record arguments, I offer an alternative proposal according to which we relied on something like Reidian common sense mentioned above. Reed offers no criticisms of my first point—that nearly all epistemologists, not merely externalists, are committed to approving of epistemic circularity. Instead, he focuses on certain aspects of my discussion of the second and third points. In this section, I’ll consider what he has to say about my second point. In the next, I’ll discuss his comments on my third point.
In making the distinction between malignant and benign epistemic circularity, I identified two different kinds of context in which one might form a belief about the trustworthiness of a belief source:
Questioned Source Context: a context in which the subject has doubt or is uncertain about the source’s trustworthiness.
Unquestioned Source Context: a context in which the subject doesn’t have doubt or uncertainty about the source’s trustworthiness.
And I said that EC-beliefs are malignant in questioned source contexts—since, because of her doubt or uncertainty about the source, the believer has a defeater for beliefs produced by it—but they are benign in unquestioned source contexts. In his $$$$, Reed objects by pointing out that one also has a defeater for beliefs produced by the source in question if one is in an unquestioned source context in which one doesn’t but should have doubt or uncertainty about the source. His claim is that because I ignore this point in my 2004, my account of benign EC-beliefs given there is flawed.
I agree completely with this criticism. In fact, I had already taken it into account in my revised statement of the distinction between malignant and benign EC-beliefs in chapter 7 of my 2006. There I draw the distinction between the following two kinds of situation in which an EC-belief about a source X or belief B can be formed:
QD-situations: Situations where, prior to the EC-belief’s formation, the subject is or should be seriously questioning or doubting the trustworthiness of X or the reliability of B’s formation.
Non-QD-Situations: Situations where, prior to the EC-belief’s formation, the subject neither is nor should be seriously questioning or doubting the trustworthiness of X or the reliability of B’s formation.
(To seriously question or doubt the trustworthiness or reliability of something is to question or doubt it to the point where one withholds or disbelieves the claim that the thing is trustworthy or reliable.) And I said that an EC-belief is malignant if it’s formed in a QD-situation and benign if it’s formed in a non-QD-situation. As far as I can tell, this amended account of malignant and benign EC-beliefs takes care of Reed’s first objection.
But Reed immediately proposes another objection: he claims that all of us should doubt the trustworthiness of our faculties. Assuming he means that we should seriously doubt this, in the sense defined above, his claim is that all of us are in QD-situations with respect to the belief that our faculties are reliable. The implication is that even if non-QD-situations (and justified EC-beliefs formed in such situations) are possible, they aren’t actual for us.
This doesn’t count against my claim that EC-beliefs can be justified in non-QD-situations (which are possible situations). So it doesn’t affect my main conclusion that epistemic circularity needn’t be a bad thing. It is, nevertheless, of interest. In my 2006 I considered and responded to some ways one might argue that non-QD-situations are impossible. But Reed is claiming merely that they aren’t actual. What does he have to say in support of that claim?
He begins by noting that everyone is aware of skeptical scenarios—at the very least, we’re aware of vivid coherent dreams (either from experiencing them or hearing about them). And he asserts that in virtue of this awareness, we should realize that there is no way for us to tell whether our current experiences are veridical rather than dreams. This should lead us, he thinks, to doubt the reliability of our faculties. To those anti-skeptics who protest that it just isn’t true that, because of our awareness of vivid dreams, we should have such doubts, Reed asks:
[W]hy not? If there is a reason why we should not have those doubts, let’s hear it. If such a reason could be provided, of course, the common sense approach wouldn’t have been needed in the first place. If, on the other hand, there is no reason why we shouldn’t have skeptical doubts, then the refusal to come to grips with skepticism is entirely ungrounded. Common sense, if this is what it amounts to, is not a way of doing philosophy but a way of not doing it.
So much for Reed’s defense of his objection. What can be said in response?
Suppose S is in the circumstances Reed mentions (i.e., she’s aware of vivid coherent dreams) but she has no doubts about the reliability of her faculties. And suppose that, when asked to account for that lack of doubt, she says her lack of doubt isn’t based on any reasons. Does this show any epistemic failure on her part? Well, that depends on what is required for a rational or sensible absence of doubt in those circumstances. Does it require that she have reasons? Or is it enough that her absence of doubt satisfies some externalist conditions such as that it is produced by her cognitive faculties functioning properly in the right environment? Externalists will think that some sort of externalist condition is sufficient. As they see it, rationality—for absence of doubt as well as for belief—doesn’t always require reasons.
Does it follow, as Reed suggests, that because an externalist offers this sort of answer to the “why not?” question, we don’t need the common sense approach in the first place? I don’t see how. Perhaps when Reed asks why it’s false that we should have doubts, he is assuming that a subject should have doubts unless she is aware of a reason not to. Perhaps noting that externalist conditions are satisfied (ones whose satisfaction makes it false that the subject should have doubts) doesn’t satisfy Reed because he wants reasons cited, reasons that the subject in question is aware of and relies on. I suspect Reed is thinking that if the subject could cite those reasons, then she’d be aware of reasons for not doubting her faculties—maybe even reasons for thinking they are in fact reliable—and wouldn’t need a noninferential commonsense belief in their reliability “in the first place.” But the thought that rationality requires such reasons won’t seem plausible to externalists. It seems, therefore, that in supporting his claim that we all should doubt the reliability of our faculties, Reed is simply taking for granted the falsity of externalism. But if that’s what’s behind his claim that we all should have doubts about the reliability of our faculties, then his objection will have no appeal to externalists unless he provides us with a good objection to externalism.
Is Reed right that giving the above response is a way of not doing philosophy? Compare the remarks above with the case of an externalist who forms memory beliefs and, when questioned about them, says they are noninferential and they’re justified in virtue of being reliably formed (not in virtue of her being aware that they’re reliably formed). Is that a way of not doing philosophy? Well, we can grant that an externalist isn’t doing philosophy just in virtue of forming memory beliefs about, say, what she ate for breakfast yesterday. Remembering needn’t and typically doesn’t involve philosophizing. But explaining why memory beliefs are justified is doing philosophy. Likewise, not having doubts about her faculties isn’t a way of philosophizing. But noting that such absence of doubt, even when it occurs unaccompanied by reasons for it, can be rational so long as it satisfies certain externalist conditions is a way of doing philosophy.
II. Common Sense
Reed next turns his sights on my Reidian account of how we know by common sense that our faculties are reliable. Reid thinks of common sense as a belief source that produces noninferential beliefs in what he calls ‘first principles,’ including the first principle that our faculties are trustworthy. According to Reid, the way the faculty of common sense typically operates is as follows: we consider the denial of a first principle and experience the emotion of ridicule. On the basis of this experience, we reject that denial and form a noninferential belief in the first principle itself.
Reed has serious doubts about the epistemic relevance of the emotion of ridicule. This is because he takes the paradigm instances of the emotion of ridicule to be cases involving wishful thinking (as when the gambler scoffs at the warnings that he’s almost certainly mistaken in thinking his ship is about to come in) or prejudice (as when the homophobe flippantly dismisses claims that restrictions on same-sex marriage are unconstitutional). This gives us good reason, Reed thinks, to conclude that there is no faculty of common sense that, when employed, relies on the emotion of ridicule as the experiential ground of noninferentially justified beliefs.
By way of response, we should notice first that the Reidian account of how we know our faculties are reliable isn’t required for the success of my defense of epistemic circularity. So the main point of my 2004 is untouched by the complaints in the previous paragraph. Secondly, it’s worth mentioning that, because I didn’t have the space, in the 2004 paper Reed attacks, to offer much of a defense of the Reidian account, I pointed to a 2002 paper of mine where I had provided a more extended defense of it. Reed’s representation of what I said on the topic would have been more accurate and fair if he had discussed (or at least mentioned) my 2002 defense of Reid’s views. Finally, it’s worth emphasizing that the Reidian position I described includes the following two components: (a) we often form noninferentially justified beliefs in the reliability of our faculties and (b) such beliefs are based on an experience—the emotion of ridicule. But as I said in my 2002 and repeated in my 2004, the point that I rely on is not (b) but (a). I don’t think (b) is implausible but I realize some people do. Fortunately, one can sensibly endorse (a) without endorsing (b). For example, you might think that a person’s noninferentially justified belief in the reliability of her faculties is based on some experience other than the emotion of ridicule; or you might think that it is justified despite the fact that it isn’t based on anything.
However, although I don’t need to defend Reid’s appeal to the emotion of ridicule, I’d like to examine Reed’s claim that the paradigm instances of it involve cases of wishful thinking and prejudice, which show that it has no evidential value. It will be helpful to look at some other examples. Consider the suggestion that whenever you leave your office, and no human is monitoring what goes on in it, it is magically filled with colorful but quiet birds which are magically removed before anyone comes in again. Perhaps a not-so-evil demon does this, being sure to magically return the room to its pre-bird state before anyone enters. Upon considering that suggestion, you no doubt find it ridiculous. You of course don’t believe it but you go further: you believe it is false. And, I assume, you have no doubts at all about its falsity. I think it wouldn’t be inappropriate to call the initial attitude or feeling you had toward that suggestion ‘the emotion of ridicule’. But it doesn’t matter what you call it. What matters is that that sort of attitude or feeling seems to be the basis for your belief that the suggestion is false. And it seems perfectly rational for you to hold that belief on that basis. Similar remarks apply to our reactions to other claims alleging such prankster-like demonic activity: maybe physical objects in the world are made invisible whenever no one is looking at them or videotaping them, maybe the contents of diet cola cans are replaced with regular cola until just before they’re opened, etc. Similar remarks also apply to our reactions to claims such as people are never responsible for their behavior or the thoughts of which I’m conscious are someone else’s not mine or I have existed for only few seconds. It is examples like those proposed in this paragraph, not Reed’s examples of prejudice and wishful thinking, that involve paradigm instances of what Reid would call ‘the emotion of ridicule’. This point is something I made clear in my 2004 when I noted that Reid doesn’t want to classify as ‘common sense beliefs’ beliefs that are peculiar to our own culture or upbringing. On the contrary, his intention is to include only propositions that most everyone believes and knows noninferentially—things that are immediately accepted by sane persons once considered and understood. Obviously, the gambler’s belief and the homophobe’s belief don’t fall into that category.
Notice that it’s plausible, even if not entirely uncontroversial, to think that in these other cases I’ve identified in the previous paragraph (to contrast with the prejudice and wishful thinking cases), the emotion of ridicule does have evidential value. Of course it’s true that there’s no necessary connection between experiencing the emotion of ridicule—directed against the denial of a first principle—and that principle’s truth. But externalists recognize that this lack of a necessary connection between belief grounds and the truth of the beliefs based on them runs rampant in our belief formation. That’s how it goes for sensory experience grounding perceptual belief, for memory seemings grounding memory belief, and for a priori seemings grounding a priori belief. Yet the fact that those belief grounds are only contingently indicative of the truth of the beliefs based on them doesn’t force the externalist to think they aren’t appropriate grounds. What matters, according to the externalist, is that the beliefs in question satisfy the relevant externalist conditions on justification or knowledge.
That’s all I’ll say in defense of the emotion of ridicule. But, as I said earlier, it is only part (a) of the Reidian position that is essential to the point I want to make in appealing to that position. I recognize that even part (a) of that Reidian position might, for some epistemologists, be difficult to accept. So let me lay out some of the considerations that lead me to take seriously (a)—the claim that we often form noninferentially justified beliefs in the reliability of our belief sources. First, when I carefully consider skeptical challenges of the Matrix and Cartesian demon sort, I find myself agreeing with Reid that the sane response is to think such scenarios aren’t actual and that, on the contrary, my perceptual faculties are reliable. And in the same way that it seems sane, this belief that my faculties are reliable also seems to me to be an epistemically justified belief. Second, when I consider how such beliefs (about the reliability of my faculties) are formed, they seem (introspectively) to be formed noninferentially. Third, I don’t seem to have any adequate inferential basis from which I’m inclined to infer that my faculties are reliable. Fourth, based on considerations parallel to the first three, I have on other occasions concluded that I have noninferentially justified perceptual, memory, and a priori beliefs about my immediate environment, my immediate past, and simple mathematical and logical matters: my perceptual, memory, and a priori beliefs seem to me to be sane and justified; they seem to be formed noninferentially; and I typically don’t have any adequate inferential basis for them. From this I conclude that they are noninferentially justified. This way of arriving at such conclusions seems typical for particularists when doing epistemology. So why not conclude the same thing with respect to my belief that my faculties are reliable? Why not conclude, on a similar basis, that it too is a noninferentially justified belief?
These four considerations give me a prima facie reason to take seriously the Reidian account. But in order for this prima facie reason to count as a good all-things-considered reason, the Reidian position must be able to weather objections. I have the space here to deal only briefly with two of them.
First, it’s true, as Reed points out, that it’s mysterious how we can be reliable noninferential detectors of such complicated contingent facts as that our faculties are reliable. But it’s also mysterious how we can be reliable noninferential detectors of necessary facts. (What’s the causal chain leading from them to our true beliefs about them?) Yet we needn’t conclude from this that we can’t have noninferentially justified beliefs in necessary truths. The same goes for our noninferential beliefs in the reliability of our faculties.
Second, at the very end of his paper, Reed claims that “common sense opens the door to every breed of dogmatist—no matter how silly or strange”. This is a familiar sort of charge. It has been applied to those who, like Reid when he appeals to the faculty of common sense, claim that humans have cognitive faculties the existence of which isn’t universally acknowledged. But it has also been applied more widely to philosophers just in virtue of their being externalists. It’s true that permitting noninferentially justified beliefs in the reliability of our faculties—in the way externalists do if such a belief satisfies their proposed conditions on justification—makes it difficult to prevent clever defenders of silly and strange beliefs (such as telepathic beliefs or belief in the Great Pumpkin) from offering parallel externalist defenses of their own positions. But, as I argue in my 2006, a similar worry can be applied with equal force to all but the least plausible internalist views. Moreover, as I also argue in my 2006, there is reason to think that this worry, as troublesome as it seems at first, needn’t cause the externalist any difficulty.
I don’t pretend that what I’ve said in this second section in support of the Reidian common sense position settles matters definitively in favor of it. But it is sufficient, I think, to show that it is a respectable view not so easily dismissed as Reed seems to think.
________. 2004. “Epistemic Circularity: Malignant and Benign.” Philosophy and Phenomenological Research 69: 709-27.
________. 2006. Justification without Awareness: A Defense
of Epistemic Externalism.
Roderick. 1982. The
Foundations of Knowing.
Richard. 1995. Metaepistemology
________. 2001. “Plantinga, Warrant, and Christian Belief.” Philosophi Christi 3: 341-51.
Michael. 1990. Atheism:
A Philosophical Justification.
Alvin. 1983. “Reason and Belief in God” in Alvin
Plantinga and Nicholas Wolterstorff (eds.) Faith
and Rationality. Notre Dame, IN:
________. 2000. Warranted Christian Belief.
Michael. 2002. World
Without Design: The Ontological Consequences of Naturalism.
Reed, Baron. $$$$. “Epistemic Circularity Squared? Skepticism about Common Sense.” Philosophy and Phenomenological Research $$: $$$-$$.
Thomas. 2002 . Essays
on the Intellectual Powers of Man, Derek Brookes (ed.).
Sosa, Ernest. 1997. “Reflective Knowledge in the Best Circles.” The Journal of Philosophy 94: 410-30.
Vogel, Jonathan. 2000. “Reliabilism Leveled.” The Journal of Philosophy 97: 602-23.
 This takes for granted that there isn’t overdetermining belief sustenance.
 See Fumerton 1995, ch. 6 and Vogel 2000.
 In my 2004, section III.A I explain why I interpret Reid as holding the sort of position on epistemic circularity and common sense that I endorse.
 I have since developed an improved version of that argument, noting that if it fails, so does the Fumerton-Vogel objection to externalism based on epistemic circularity. I have also given a different argument for the same conclusion (that epistemic circularity needn’t be a bad thing). See Bergmann 2006, ch. 7 for both arguments.
 Bergmann 2004, 718-19.
 Understandably, Reed was unaware of this, since my 2006 was still being copy-edited when his $$$$, already accepted for publication, was brought to my attention by the editor of this journal.
 ‘QD’ is for ‘questioning’ and ‘doubting’.
 Reed $$$$, ms. p. 8.
 Reed makes it clear ($$$$, ms. p. 8) that he thinks these doubts we should have constitute normative defeaters for the belief that our faculties are reliable, which (according to the definition of ‘normative defeater’ he employs) implies that we should withhold or disbelieve the claim that our faculties are reliable. This suggests that he thinks we should seriously doubt, in the sense I defined above, the trustworthiness of our faculties.
 See section 4.2.2 of my 2006, ch. 7.
 Note that he needs to say something in support of it because this claim is, at first glance, highly implausible. Can it really be that engineers, scientists, doctors, mechanics, children—everyone around me—should doubt the reliability of their faculties? If that were true, how could they appropriately escape that situation? It seems they couldn’t appropriately escape it by relying on faculties they should be doubting. But if not on them, on what? And if that is in fact their situation, is it really appropriate for me to trust anything they tell me? Is it appropriate for me to believe they even exist if I should be doubting my faculties? These considerations make it difficult to take Reid’s claim seriously unless it is defended by a very strong argument.
 Reed puts the words I quote here in the mouth of the skeptic. But it is Reed who has claimed that all of us should have such doubts. So I’ll take it as his own reply since he offers nothing else.
 Reed $$$$, ms. pp. 8-9.
 By ‘externalist conditions’, I mean only that they are conditions that an externalist would view as sufficient for the status in question—in this case, for it not being the case that one should have doubts.
 In note 17 of his $$$$, Reed says that whether we should have doubts about the reliability of our faculties depends on whether we have evidence for the belief that our faculties are reliable. But externalists will think it depends on whether our belief in the reliability of our faculties (or our not having doubts about their reliability) satisfies certain specified externalist conditions.
 As I noted earlier in this paper, the reason I discussed Reid’s views on common sense in my 2004 was to counter the impression that the only serious proposal for how we in fact form justified EC-beliefs in the trustworthiness of our faculties is the implausible one according to which we do so inferentially on the basis of epistemically circular track-record arguments.
 See my 2004 for further discussion.
 Reed $$$$, ms. p. 12.
 See my 2004, 724, n. 35 where I refer to my 2002, sect. VIII.
 See my 2002, 87 and my 2004, 724, n. 34.
 For an example of someone endorsing the latter view, see Rea 2002, 1-7 & 180-81.
 We can add details that make it impossible for us to determine (on the basis of our perceptual faculties and scientific instruments) whether the birds have been there. If a demon is arranging for this to happen, these additions to the story should be unproblematic.
 See my 2004, 722, n. 29.
 Reid is aware of the problem of distinguishing things genuinely known via common sense from things that merely seem to be first principles and he proposes some criteria for telling the difference (Reid 2002 , 452-67). I don’t mean to say he’s got the criteria exactly right. Nor do I (or Reid) mean to suggest that one can’t noninferentially know a first principle without first checking to see if it satisfies these criteria. My point here is just that the problem of including too many things in the category of first principles is one Reid was aware of. And he was quite interested in not including among first principles things known via prejudice or wishful thinking. This is made perfectly clear when he says (Reid 2002 , 460) that we need to take care that our judgments about what counts as a first principle haven’t been perverted by the common causes of error to which humans are prone. These causes of error are examined at some length by Reid when he discusses with approval Bacon’s catalogue of errors in human thinking that are due to bias, passions, etc. (Reid 2002 , 527-41).
 And again, that point isn’t itself essential to the main burden of my 2004—which is to defend externalism by arguing that it’s possible to have justified epistemically circular beliefs.
 It’s probably worth noting that a significant consideration in the background is my conviction that internalism is hopeless and that some form of externalism is true. See my 2006 for objections to internalism and a defense of externalism.
 See Chisholm 1982, ch. 5 for an account of the particularist approach.
 Of course one could say that the particularist approach is misguided in all its manifestations. But then the target isn’t really my Reidian account of the faculty of common sense but instead a widely employed approach to epistemological analysis.
 But see my 2004, section III.B and my 2002, section VIII.
 For example, this sort of complaint has been lodged against Plantinga’s claim that belief in God can be noninferentially justified if it is produced by a faculty John Calvin calls the sensus divinitatus. See Michael Martin (1990, 266-78) for one development of this objection. For responses by Plantinga, see his 1983, 74-78 and his 2000, 342-53 where he explains why his views needn’t open the door to permitting belief in the Great Pumpkin or voodoo.
 Richard Fumerton argues (2001, 343-44) that if externalism is true, the door is thereby opened to Plantingian claims to noninferentially justified Christian belief (which, according to many—see the previous footnote—opens the door even further to claims of noninferentially justified belief in the Great Pumpkin). Ernest Sosa responds to a charge of this sort leveled against his claim that perceptual belief can be epistemically appropriate in virtue of satisfying externalist conditions. See his 1997, section VI where he explains why his views needn’t open the door to claims of justified beliefs produced by gazing into a crystal ball.
 Bergmann 2006, ch. 8, section 2.
 Bergmann 2006, ch 8, section 3.3.
 Thanks to Jeffrey Brower, Trenton Merricks, Michael Rea, and Chris Tucker for comments on earlier drafts.