Inquiring minds want to know, not merely to believe or even to believe truly. They want knowledge of “the facts,” at least the facts in a relevant domain. Epistemology thus investigates and elucidates what inquiring minds want. So, epistemology is valuable to inquiring minds, whatever their domains of interest. A person might settle for true belief and remain lazily indifferent to knowledge, but this would be odd indeed. Inquiring minds seek something better grounded than true belief based just on lucky guesswork, for example. They want true beliefs grounded in adequate evidence, if only to avoid the vicissitudes of ungrounded belief.
Epistemology aims to characterize, among other things, adequate evidence and the way such evidence grounds true beliefs qualifying as knowledge. This aim resists easy satisfaction, owing to the complexities of evidence and knowledge, but this, of course, says nothing against the value of epistemology. The world resists easy explanation pretty much everywhere else, too. We thus should not fault epistemology, or any philosophical discipline, for its due complexities in explaining a theoretically demanding segment of the world.
Knowledge entails justified true belief, and more: more specifically, justified true belief undefeated by truths of the sort at work in Gettier-style counterexamples. We can set aside such counterexamples now, since our focus will not be the analysis of propositional knowledge. The focus will rather be on the kind of objectivity available to inquiring minds. In seeking adequately grounded true beliefs, inquiring minds want at least beliefs representative of how things really are. On what basis, if any, can they reasonably lay claim to such beliefs? This question motivates this chapter.
1. Objectivity and Intelligible Realism
In seeking beliefs representative of “how things really are,” we pursue a kind of objectivity, and thereby tread on controversial philosophical ground. In particular, we invite controversy over realism about what our beliefs represent. Realism comes in many shapes and sizes; so a few distinctions may help.Minimalrealism states that something exists objectively, that is, independently of being conceived of.Ordinaryrealism proposes that the tokens of most ordinary psychological and physical types (specified by ordinary language-use) exist objectively.Scientificrealism holds that the tokens of most scientific types exist objectively.1
We shall examinemoderaterealism, the view that what is represented by at least some of our beliefs is objective, that is, logically and causally independent of someone's conceiving of that thing. (Let's render “conceiving” broadly, to encompass such psychological phenomena as assenting, believing, and perceiving.) For example, your belief that Lake Michigan is wet represents, plainly enough, Lake Michigan's being wet. Lake Michigan's being wet is objective now if and only if it does not now depend, logically or causally, on someone's conceiving of it. Even psychological phenomena, including events of conceiving, can be thus objective, so long as they are independent of someone's conceiving of them. Objectivity, then, derives from conceiving-independence of the sort noted. It can, at least in principle, characterize psychological as well as nonpsychological phenomena. Conceivers themselves, for example, might be conceiving-independent, that is, independent of being conceived of.
Put bluntly, talk of the objectivity of something is just talk of how that thing is independently of what any conceiver takes it to be. Some opponents of talk of objectivity have overlooked an important distinction between (a) the conceiving-dependence of one'sconceivingof something, and (b) the conceiving-dependence ofwhat one's conceiving represents.Your conceiving of Lake Michigan depends on conceiving, but it does not follow that Lake Michigan depends on conceiving. As for ourconceptof Lake Michigan, it evidently does not depend on our concept of conceiving. Conceivably, one could have the concept of Lake Michigan while lacking the concept of conceiving. The concept of conceiving is thus not a necessary ingredient of the content of a concept, even if concepts and events of conceiving depend for their existence on conceivers.
Some philosophers have questioned the intelligibility of any notion of objectivity relying on a concept of “how things really are,” or “how things are independently of being conceived of.” Richard Rorty, for instance, has spoken of “the sterility of attempts to give sense to phrases like ‘the way the world is,’” adding that a notion of the world-in-itself is “completely unspecified and unspecifiable.”2Similarly, in commenting on a Kantian notion of the thing-in-itself, Hilary Putnam has suggested that a concept of conceiving-independent reality “makes no sense.”3They thus suggest that it is unintelligible to talk of objectivity in terms of how things really are, that is, how things are independently of being conceived of.
The claim that the notion of how things really are is unintelligible raises a notable self-referential problem. Let's ask whether that claim is itself a claim about how things really are. Either it is or it isn't. If it is, it is unintelligible given its own claim, as it implies that statements relying on the notion of how things really are qualify as unintelligible. If, alternatively, it is not a claim about how things really are, then we need an explanation of what kind of claim it is. Clearly, it will not help to reply that it is a claim about what we arejustified in claiming, rather than a claim about how things really are. Given that reply, the view suggested by Rorty and Putnam is that we are justified in claiming that the notion of how things really are is unintelligible. The latter view fortifies rather than avoids our self-referential problem, because we can ask whether that view is itself a claim about how things really are, with respect to what we are justified in claiming. If it is, then given the very view in question, we are justified in holding that this view itself is unintelligible. If, alternatively, it is not a claim about how things really are, but only a claim about what we are justified in claiming (about what we are justified in claiming), a troublesome regress of increasing levels of supposed justification threatens. The threatening regress is troublesome if only because we seem not to have (and perhaps do not even understand) the kind of complex iterated justification it requires.4
We should doubt, then, the suggestion that the notion of how things really are is unintelligible. The suggestion of Rorty and Putnam about the notion of how things really are presupposes, on pain of implied unintelligibility or infinite regress, a realist notion of how things really are. Indeed, any truth-valued judgment presupposes a notion of how things really are and thereby illustrates the intelligibility of a notion of objectivity. So, moderate realism, implying that what is represented by at least some of our beliefs is objective, qualifies as intelligible. Semantically, in other words, moderate realism seems beyond reproach. Claims to the contrary have identifiable implications about how things really are, thereby testifying to the intelligibility (but not necessarily the truth) of moderate realism. (Section 4 will assess a more recent attempt by Rorty to eliminate issues central to traditional epistemology.)
The notion of how things really are enables us to introduce a minimal realist definition of truth. Using the term “proposition” to signify the content of a statement, we have this definition: the claim that a proposition,P, is true means that things really are as they are stated to be byP.This realist definition is a minimal correspondence definition in that it makes truth a function just of the statement relation between a proposition and how things really are. In addition, it can accommodate Tarski's necessary condition for any adequate definition of truth, namely, schema T:Xis true if and only ifP(where “P” stands for a declarative sentence, and “X” stands for the name of that sentence). The minimal realist definition of truth has the plausible implication that a person believes that a proposition is true if and only if she believes that the proposition states how things really are. It also supplies an important notion ofobjectivetruth.5
The intelligibility of a position does not, of course, entail its truth. Truth is not so easy to come by. The Ptolemaic model of the universe, for example, is intelligible but nonetheless false. Moderate realism, too, might be intelligible but false. That is, what is represented by a person's beliefs might not in fact be objective, even though it is intelligible that it is objective. In that case, moderate realism about the person's beliefs would be false but nonetheless intelligible. Given that moderate realism is intelligible, a quest for objectivity in what our beliefs represent is similarly intelligible. So, the familiar search in traditional epistemology for beliefs that are objective, in virtue of capturing conceiving-independent reality, is semantically acceptable, even if that search ultimately founders in light of some nonsemantical, epistemological problems. Let's turn to the latter problems.
2. Grounding Realism: From Intelligibility to Acceptability
Begging questions
Moderate realists want, of course, to maintain the truth and the (epistemically) rational acceptability of moderate realism, not just its intelligibility. Skepticism about moderate realism offers a dangerous obstacle. The central skeptical challenge is that moderate realists should deliver non-questionbegging epistemic support for moderate realism. In delivering such support, moderate realists will not beg relevant questions against their skeptical challengers. They will not simply assume any point needing defense in light of relevant skeptical questions. Otherwise, argument may become altogether superfluous in the exchange between realists and skeptics. A troublesome kind of epistemic arbitrariness would then threaten.
Given the permissibility of begging relevant questions, we can supportanyarbitrary position we like. We need only beg questions in a way that favors our preferred position. In that case, we would have intolerable epistemic arbitrariness in philosophical exchange. Cogent argument would be ultimately pointless, and doxastic caprice would threaten. Even though all arguments rest on premises, an argument's premises need not, of course, be questionbegging in a context of inquiry. So, the skeptical challenge at hand does not oppose argument in general. It opposes only arguments that beg relevant questions in a context of inquiry.
The skeptical challenge does not imply that moderate realism cannot be true or even that it is not true. So, the challenge does not require idealism of any sort. The main skeptical issue concerns not directly the objective truth or falsity of moderate realism but rather the kind of epistemic support available for it. Skeptics can acknowledge, if only for the sake of argument, that our representing any objective thing depends on such cognitively relevant processes as perception, introspection, memory, and testimony. The problem is that such processes apparently fail to yield non-questionbegging support for their own reliability or for the truth of the beliefs they deliver. Skeptics have asked, quite intelligibly, whether the processes in question create or at least decisively influence what is thereby represented. If they do, they risk being an unreliable means, even a completely unsuccessful means, to objectivity. In that case, our cognitive processes would function as producers or distorters rather than sound representers of their deliverances.
Skeptics, among others, doubt that we can avail ourselves of a position independent of our cognitively relevant processes to assess, in a non-questionbegging manner, either their general reliability in representing how things really are or the truth of the beliefs they yield. Skeptics, again among others, suspect that this is the unavoidable human cognitive predicament. Moderate realists seeking to avoid questionbegging reasoning will need either to offer a way out of this predicament or to show that the predicament is ultimately illusory.
William Alston has correctly noted that a threat of circularity, of the kind under scrutiny, “puts no limits whatever on the beliefs that can be justified, nor does it limit what can be known.”6The skeptical problem at hand does not concern one's merely having justifying evidence for one's beliefs or even one's having true beliefs based on evidence resistant to Gettier-style problems. So, skeptics opposing circularity need not claim that we do not have justified belief or knowledge. Their challenge rather concerns one's producing, or at least having,non-questionbegging epistemic supportfor the truth of one's beliefs, given certain immediately relevant questions raised by skeptics. Since this challenge bears significantly, as noted previously, on the value of cogent argument over epistemic arbitrariness in philosophical inquiry, we do well to take the challenge seriously. We can pursue this challenge without digressing to controversy over the exact conditions for evidence, justification, or knowledge.
Skeptical questions
Consider the following familiar bases for beliefs entailing an objective world: (a) ordinary psychological processes, including perception, introspection, and memory, (b) suitable coherence among beliefs, (c) predictive success, and (d) widespread acceptance by one's community. Let's call such bases “grounding conditions” for beliefs entailing an objective world and, in turn, for moderate realism. Grounding conditions, according to many antiskeptics, can yield adequate epistemic support for beliefs entailing an external world and thus for moderate realism. We need to determine what kind of epistemic support is thereby yielded.
An essential feature of epistemic support for a proposition is that it provides atruth-indicatorfor that proposition. A truth-indicator can, of course, be fallible and probabilistic. Our epistemic support does not necessarily guarantee, or otherwise deliver, actual truth. For current purposes, we need not digress to debates over internalism and externalism concerning whether, and if so how, epistemic support must be accessible to a believer.7
In challenging moderate realism, skeptics inquire whether the available grounding conditions yield non-questionbegging epistemic support for such realism. They begin with a simple question: what non-questionbegging epistemic support have we to claim that moderate realism is objectively true? Realists typically reply by invoking some grounding condition for a belief entailing an objective world, often a member of the aforementioned options (a)–(d). Such a grounding condition, according to typical realists, yields the needed epistemic support for their moderate realism.
Skeptics will challenge any antiskeptical use of a grounding condition, as follows: what non-questionbegging epistemic support have we to claim that your preferred grounding condition is actually indicative of what is objectively the case? In other words, what non-questionbegging epistemic support have we to claim that a proposition (for instance, the proposition that mind-independent objects exist) warranted by your grounding condition is actually true? Realists who reply by invoking their preferred grounding conditions will beg this question against their skeptical challengers. The adequacy of those grounding conditions is being questioned now by skeptics, and begging their question will fail to deliver non-questionbegging support for moderate realism. Skeptics ban questionbegging reasoning, not fallibilism (the view that a well-grounded contingent belief can be false) or inductive inference in epistemic support. In addition, their ban on questionbegging reasoning remains neutral on issues about the exact conditions for epistemic justification and knowledge. So, skeptics are not relying on a special, self-serving account of justification, knowledge, or epistemic support.
Skeptics do not demand that realists argue for moderate realism without using premises. That would be absurd. They rather demand non-questionbegging epistemic support for moderate realism and for the common assumption that a preferred grounding condition is actually indicative of what is objectively the case. So, realists may offer premises in arguments whose conclusion is the truth of moderate realism. What they may not offer is a questionbegging premise given the skeptical challenge at hand. Such a premise would automatically fail to convince in a context of inquiry where it is itself under question, explicitly or implicitly. An argument using such a premise will be at best ineffective in the relevant context. Questionbegging thus makes argument superfluous.
The aforementioned human cognitive predicament suggests that realists will not meet the skeptical challenge. Epistemically significant access to anything by us apparently depends on such grounding conditions as perception, introspection, memory, testimony, and common sense. Since skeptics have questioned the reliability of such conditions in delivering objective truths, an appeal to such conditions will fail to yield the demanded non-questionbegging support. Whethereverykind of cognitive being, including God, must beg questions against skeptics will remain open in this chapter. Evenifcognition of every sort entails questionbegging against skeptics, owing to the nature of cognition itself, skeptics will be unmoved. They do not find the epistemic arbitrariness of questionbegging more acceptable in virtue of its allegedly being integral to cognition of any sort.
Skeptics are notorious for questioning our grounding conditionson the whole, that is, as a group. So, we cannot satisfy the skeptical challenge by invoking one preferred grounding condition to support another preferred grounding condition. The relevant challenge iscomprehensive, bearing on grounding conditions, or supposed cognitive sources,in general.Any answer we give to such a comprehensive challenge will apparently rely on at least one of the grounding conditions being challenged. So, questionbegging seems to be our fate in competition with skeptics. Such, apparently, is the human cognitive predicament.
3. Realist Replies to Skepticism
Practical rationality
William Alston has replied to skeptical challenges to realism based on the threat of circularity, as follows:
[It is] rational and proper to engage in our customary doxastic practices without having, or even being able to have, any positive noncircular reasons for supposing them to be reliable… [T]hese familiar practices [are] autonomous, acceptable on their own, just as such, without being grounded on anything external. Since any attempt to show one of these practices to be reliable will, in effect, assume the reliability of some other of our familiar practices there is no appeal beyond those practices.8
This reply has two important parts. First, it concedes that the skeptical challenge cannot be met, as we are unable to deliver noncircular (or, non-questionbegging) epistemic support for the cognitively relevant sources of our beliefs. So, the reply offers no threat to the main skeptical challenge. Second, the reply invokes a kind of rationality compatible with the unavoidable presence of epistemic circularity. What, however, does such rationality involve and what is its epistemic significance?
Alston's proposed rationality is “practical.” It rests on the following rhetorical question offered by Alston: “Since we cannot take a step in intellectual endeavors without engaging in some doxastic practice(s) or other, what reasonable alternative is there to practicing the ones with which we are intimately familiar?”9Claiming that there is no such alternative, Alston follows Thomas Reid “in taking all our established doxastic practices to be acceptable as such, as innocent until proven guilty.”10This talk of “established” doxastic practices is just talk of “customary” doxastic practices; it does not connote a special epistemic status. The core of Alston's practical rationality for doxastic practices is thus twofold: the thesis that there is no alternative to our customary doxastic practices and the thesis that such practices are innocent until proven guilty.
Skeptics can challenge the suggestion that there is no alternative to our customary doxastic practices, especially if our customary intellectual endeavors are optional.Givenour customary intellectual endeavors (such as categorizing, explaining, and predicting), we perhaps have no easy alternative to our customary doxastic practices. Still, skeptics can challenge the epistemic significance of our customary intellectual endeavors. They can ask, quite intelligibly, whether those endeavors deliver accurate information about an objective world. In addition, they can intelligibly ask whether any non-questionbegging epistemic support indicates that practical rationality, of the sort recommended by Alston, yields reliability or objective truth in our resulting beliefs. If one concedes the lack of such support, then one's position is compatible with, and no challenge to, skepticism of the variety at hand. If, alternatively, one lays claim to such non-questionbegging support and offers a relevant grounding condition, then skeptics can challenge that grounding condition in the way indicated previously. The aforementioned human cognitive predicament offers no encouragement here.
Alston evidently agrees about the ineffectiveness of practical rationality in challenging shrewd skeptics. He concedes:
We have shown, at most, that engaging in SP [= sense perceptual practice] enjoys apracticalrationality; it is a reasonable thing to do, given our aims and our situation. But then it is only that same practical rationality that carries over, via the commitment relation, to the judgment that SP is reliable. We have not shown that it is rational in anepistemicsense that SP is reliable, where the latter involves showing that it is at least probably true that SP is reliable…. We have not shown the reliability attribution to be rational in a truth-conducive sense of rationality, one that itself is subject to a reliability constraint.11
So, we cannot invoke Alston's practical rationality to meet the skeptical challenge under examination. Skeptics will naturally ask what non-questionbegging epistemic support we have to suppose that practical rationality is actually indicative of objective truth. It seems doubtful that an answer favorable to moderate realism will avoid begging relevant questions against such skeptics.
An appeal to practical rationality for antiskeptical purposes resembles pragmatic defenses of realism. The latter defenses aim to rebut skeptical challenges on the basis of a belief's overall utility relative to set purposes, whether theoretical or practical. Skeptics will be unmoved, however, owing to the significance of this question: what non-questionbegging epistemic support have we to claim that a belief's overall pragmatic utility (of whatever degree or kind one prefers) is actually indicative of what is objectively the case? Resources for an adequate answer in favor of realism seem not to be readily available. It would not serve realists at all to claim that it is pragmatically useful to believe that pragmatic utility is indicative of objective truth. That move would only invite the skeptical challenge at a higher level, where it remains unanswered. Practical considerations, then, seem not to meet the skeptical challenge. As a result, they leave moderate realism inadequately defended.
Inference to best explanation
Many philosophers have tried to defend realism on the basis of explanatory considerations, specifically on the basis of an inference to best explanation.12A crucial assumption is that moderate realism is part of the best explanation of our overall perceptual experiences and thereby resists skeptical challenges. One might supplement this assumption with the claim that our perceptual inputs apparently are, at least for the most part, involuntary, unlike certain cases of active imagination. Typical perceptual inputs seem not to arise subjectively from our volitional activity.
However realists develop an argument from best explanation, skeptics can identify two problems. First, evenifwe have no indication in our experience of a subjective source of our object-like perceptual inputs, we likewise lack a non-questionbegging indication in our experience of an objective source of those inputs. In particular, we lack non-questionbegging epistemic support for the assumption that the absence of an indication of a subjective volitional source yields an indication of an objective source. Second, inference to be best explanation invites the same kind of challenge facing the grounding conditions noted above. In particular, what non-questionbegging epistemic support have we to claim that an inference to the best explanation (however elaborated) is actually indicative of how things really are, or of what is objectively the case?
The considerations raised in connection with the previous grounding conditions suggest, owing to the human cognitive predicament, that the support demanded by skeptics is not actually forthcoming. We apparently lack an independent standpoint for delivering the required support. At least, realists have not shown otherwise. Inference to the best explanation may be central to “justification” as commonly understood, but we have no reason to think that it delivers an adequate answer to the skeptical challenge.
Science, skepticism, and realism
Many realists hold that our best science lends support to their realism. For example, Michael Devitt has appealed to our best science to challenge skepticism, as follows:
The thorough-going sceptic sets the standards of knowledge (or rational belief) too high for them ever to be achieved. Our best science shows us this. It shows us, for example, that if knowledge is to be gathered we must eliminate implausible hypotheses without being able, ultimately, to justify that elimination. It shows us that there is always an (empirical) possibility of error with any (normal) knowledge claim. Standards that our best science shows cannot be met short of instantaneous solipsism – a doctrine that is literally incredible – should be ignored. Scepticism is simply uninteresting: it throws the baby out with the bath water.13
Contrary to Devitt's suggestion, the skeptical challenge entertained in this chapter does not set excessively high standards for knowledge or justified belief. It grants realists their preferred standards for knowledge and justified belief. It even grants that people can have knowledge and justified belief by those standards. The central issue concerns the availability of non-questionbegging epistemic support for moderate realism. Realists cannot simply assume, by way of convincing reply to skeptics, that they actually possess such epistemic support. Even though moderate realism is nonepistemic in what it claims, proponents must, in a context of inquiry with skeptics, deliver the required non-questionbegging epistemic support for their realism or make a concession to the skeptical challengers.
Devitt calls his approach to epistemology “naturalized,” because it recommends that “the epistemic relation between humans and the world itself becomes the object of scientific study.”14He explains:
Naturalized epistemology takes science and hence its posits pretty much for granted. And an obvious starting assumption is … that these posits exist objectively and independently of the mental. So it approaches epistemology from a Realist standpoint…15
Devitt claims that “once the sceptical problematic is abandoned in favour of a naturalized epistemology, arguments against Realism lack all cogency,” adding that “Realism alone explains ‘the regularities in our experience.’”16The latter point suggests a return to inference to best explanation (on which see the previous section of this chapter).
Two considerations raise problems. First, we have already raised doubts about the ability of explanatory considerations to answer the skeptical challenge. Even if realism has singular explanatory value, and thus achieves a kind of explanatory justification, it does not thereby gain non-questionbegging epistemic support. Hence, the skeptical challenge remains unanswered. Second, contrary to Devitt's suggestion, naturalized epistemology does not automatically block the cogency of arguments against realism. The skeptical challenge persists so long as (a) questionbegging is inappropriate in a context of debate with skeptics and (b) realism lacks non-questionbegging epistemic support. This challenge, we have seen, does not rest on excessively high standards for knowledge or justified belief. We can illustrate the second point in connection with W. V. Quine's naturalism about epistemology.
Quine maintains that skeptical doubt is “an offshoot of science” in that such doubt arises from simple physical science about the reality of bodies.17Simple physical science, Quine holds, is crucial to our distinction between reality and illusion. This view prompts Quine's suggestion that epistemology is “an enterprise within natural science.”18Quine has motivated this suggestion in a manner favorable to realism, as follows:
I also expressed, at the beginning, my unswerving belief in external things – people, nerve endings, sticks, stones. This I reaffirm. I believe also, if less firmly, in atoms and electrons and in classes. Now how is all this robust realism to be reconciled with the barren scene that I have just been depicting? The answer is naturalism: the recognition that it is within science itself, and not in some prior philosophy, that reality is to be identified and described.19
Evidently, Quine's naturalism relies on a principle of credulity regarding the reliability of our best science. This principle entails that we may reasonably assume the reliability of our best science in describing objective reality. Can such a principle disarm the skeptical challenge?
We may grant, if only for the sake of argument, that “simple” physical science is crucial to our havingthe distinctionbetween what is real and what is illusory. Even so, our physical science, whether simple or complex, might still fail to deliver objectively true beliefs concerning what is real. The history of physical science bears witness to this. As a result, it is intelligible (but not, of course, thereby true) that most of the theses of our physical science are not objectively true. Principles of interpretive charity, including those offered by Quine and Donald Davidson, offer no serious challenge here, given a distinction between the conditions for the actual content of a belief and the conditions forascribinga belief's content.20
The naturalist principle of credulity entails that we may reasonably assume the reliability of our best science in describing objective reality. Philosophers have rarely noted that this principle is not itself a hypothesis of our best science. Even if some scientists have adopted a principle of credulity, this principle is not a scientific hypothesis. It is rather a philosophical principle that goes beyond natural science proper. Witness the absence in natural science of a means of testing the principle of credulity. The crucial point, however, is that the principle of credulity does not enjoy non-questionbegging epistemic support of the kind demanded by skeptics.
The main question is: what non-questionbegging epistemic support have we for the claim that a belief satisfying the principle of credulity is objectively true? Letscientific rationalitybe the kind of rationality offered by belief-formation in accord with the principle of credulity. The question then becomes: what non-questionbegging epistemic support have we for the claim that scientific rationality is indicative of what is objectively the case? An appeal to our best science will be questionbegging in this connection, even if our best science delivers simplicity, comprehensiveness, and predictive success in belief-formation. Skeptics are equal-opportunity challengers of preferred grounding conditions. So, the sciences get no special respect in debates with skeptics. Naturalized epistemology thus fails to answer the skeptical challenge.
Perhaps, contrary to the available evidence, Quine does not actually hold a principle of credulity for our best science. In any case, this would not rescue him from the skeptical challenge, because questions about the reliability of our best science persist. Apart from a principle of credulity, Quine's naturalism slides dangerously toward skepticism. It would then lack a basis for its reliance on the reliability of prediction regarding sensory checkpoints as a guideline in the formation of theories. The naturalist would then be indistinguishable from the skeptic, at least relative to the skeptical challenge under examination.21
Circularity and intellectual satisfaction
Perhaps we have made too much of the skeptical challenge and thereby overlooked what really matters in an epistemology. Richard Foley has suggested as much, as follows:
Inquiry requires a leap of intellectual faith, and the need for such faith cannot be eliminated by further inquiry, whether it be empirical or philosophical or whatever… . Since we can never have non-question-begging assurances that our way of viewing things is correct, we can never have assurances that there is no point to further inquiry. The absolute knowledge of the Hegelian system, which requires the knowing mind to be wholly adequate to its objects and to know with utter certainty that it is thus, is not a possibility for us…. For us there can be no such final resting spot.22
Many skeptics would welcome this attitude toward inquiry, given its concession that non-questionbegging assurances for the correctness of our beliefs are unavailable. The basis of this concession is evidently Foley's observation that “if a proposed method of inquiry is fundamental, then it cannot help but be used in its own defense if it is to be defended at all.” Foley adds that “some questions deserve to be begged, [and that] questions about the reliability of our fundamental methods of inquiry are just such questions.”23
Two questions arise. First, what can support an inference from the observation that wecannotavoid questionbegging regarding a fundamental method of inquiry to the view that some questionsdeserveto be begged? (I am not attributing this dubious inference to Foley; rather, I am considering how one might develop his remarks on questionbegging.) This question concerns epistemic support of a kind relevant to debates with skeptics. It is not about pragmatic expediency, but even if it were, skeptics could follow up with questions about the epistemic support for assumptions about what beliefs have pragmatic expediency. Skeptics will properly doubt that our inability to avoid begging skeptical questionsentitlesus as truth-seekers to beg those questions. Begging skeptical questions may obviouslyhinderour goal of acquiring truth and avoiding error (especially in cases where it fosters epistemic arbitrariness), and it will automatically leave skeptics unanswered. Our position will then seem epistemically arbitrary relative to intelligible skeptical questions. Second, what reason have we to link a demand for non-questionbegging epistemic support with a quest forcertainty? The skeptical challenge under consideration has no conflict with the common assumption that the epistemic support for contingent propositions is typically fallible and defeasible and fails to yield epistemic certainty. We thus should separate a demand for non-questionbegging epistemic support from a demand for certainty.
Foley aims to liberate epistemology by linking it with “egocentric rationality,” as follows:
The prerequisite of egocentric rationality is not truth or reliability, not even in the long run; it is, rather, the absence of any internal motivation for either retraction or supplementation of our beliefs. Egocentric rationality requires that we have beliefs that are to our own deep intellectual satisfaction – beliefs that do not merely satisfy us in a superficial way but that would do so even with the deepest reflection. To be ego-centrically rational is thus to be invulnerable to a certain kind of self-condemnation. It is to have beliefs that in our role as truth-seekers we wouldn't criticize ourselves for having even if we were to be deeply reflective.24
Egocentric rationality thus preserves our “role as truth-seekers,” and thereby avoids collapse into merely pragmatic rationality. At the same time, it excludes truth as a “prerequisite,” and thereby allows for fallibilism concerning rational belief. So far, so good.
The problem is that many of us truth-seekers, skeptics included, are not actually satisfied by questionbegging in the presence of skeptical challenges. In addition, this dissatisfaction endures even upon our deepest reflection, owing to the apparent epistemic arbitrariness in the relevant questionbegging against skeptics. Acknowledging this, we must deny that egocentric rationality liberates us from the skeptical challenge haunting us throughout this chapter. Foley himself does not pretend that egocentric rationality automatically saves a person from the threat of skeptical challenges. Rather, he properly reminds us that “many intellectually undesirable characteristics are compatible with being rational.”25This is doubtless correct, but the skeptical challenge remains unanswered, and nonskeptics are still open to a charge of epistemically arbitrary questionbegging in the face of skeptical challenges. Skeptics will treat deep intellectual satisfaction as just another preferred grounding condition, and ask what non-questionbegging epistemic support we have to claim that it is actually indicative of how things really are. It is doubtful, then, that egocentric rationality will save moderate realism from the skeptical challenge.
4. Skepticism: Useless or Inconsistent?
Skepticism and eliminative pragmatism
Richard Rorty currently proposes that we reorient epistemology in a way that banishes questions about objectivity as “useless.” Perhaps we can avoid the skeptical challenge of this chapter by simply eliminating traditional epistemological questions as altogether useless, evenifthey are intelligible.
Rorty proposes that any relevant constraint on inquiry will come from the remarks of fellow-inquirers in “our community,” and his characterization of the members of our community is avowedly Darwinian. We humans are “animals with special organs and abilities,” according to Rorty, but “they no more put us in arepresentationalrelation to an intrinsic nature of things than do the anteater's snout or the bower-bird's skill at weaving.”26Rorty proposes, in addition, that “we should see what happens if (in Jean-Paul Sartre's words) ‘we attempt to draw the full conclusion from a consistently atheist position’, a position in which such phrases as ‘the nature of human life’ no longer distract us from the absence of a God's-eye view.”27Rorty recommends that we “think of our relation to the rest of the universe in purely causal, as opposed to representationalist, terms,” on the ground that Darwin, Dewey, and Davidson have made such a view promising.28
Rorty now confesses that many of his earlier criticisms of traditional philosophical distinctions and themes were misplaced:
I should not have spoken of “unreal” or “confused” philosophical distinctions, but rather of distinctions whose employment has proved to lead nowhere, proved to be more trouble than they were worth. For pragmatists like Putnam and me, the question should always be, “What use it is?” rather than “Is it real?” or “Is it confused?” Criticism of other philosophers' distinctions and problematics should charge relative inutility, rather than “meaninglessness” or “illusion” or “incoherence.”29
Rorty thus holds thatutilityrather than reality, or even evident reality, sets the standard for philosophical assessment. This fits well with Rorty's antiepistemological, antirepresentationalist pragmatism.
Rorty's main criticism now of various traditional distinctions pertinent to epistemology (for example, subject-object, scheme-content, and reality-appearance distinctions) is that they lack adequate utility. More generally, Rorty faults traditional epistemology (roughly, epistemology in the tradition of Plato'sTheaetetus) for assuming that truth is something we can and should try to discover. He remarks:
To say that we should drop the idea of truth as out there waiting to be discovered is not to say that we have discovered that, out there, there is no truth. It is to say that our purposes would be served best by ceasing to see truth as a deep matter, as a topic of philosophical interest, or “true” as a term which repays “analysis.” “The nature of truth” is an unprofitable topic, resembling in this respect “the nature of man” and “the nature of God,” and differing from “the nature of the positron,” and “the nature of Oedipal fixation.”30
Rorty urges that we replace the vocabulary of traditional epistemology with a new pragmatist vocabulary enabling us to “stop doing” traditional epistemology, including philosophical inquiry about skepticism. Rorty thus offers an epistemo-logical analogue of eliminative materialism, the view that common psychological vocabulary should be replaced with a “better,” materialist vocabulary.
Let's call Rorty's antiepistemological, antirepresentationalist vieweliminative pragmatism. Its main negative thesis is that the vocabulary, problems, and goals of traditional epistemology are unprofitable and in need of replacement by pragmatist successors. Its main positive thesis is that “all philosophy should do is compare and contrast cultural traditions,” that is, provide “a study of the comparative advantages and disadvantages of the various ways of talking which our race has invented.”31Eliminative pragmatism endorses the philosophical dispensability of concerns about how the world really is, and recommends the central philosophical importance of what is profitable, advantageous, or useful. Rorty emphasizesusefulness, not majority opinion. As he says, “I do not recall that I have ever, even at my worst, spoken of either warrant or truth being determined bymajorityvote,”32
A number of questions arise: Usefulfor whom? Usefulin what manner? Usefulto what extentandfor how long? Since Rorty does not pursue these questions, his eliminative pragmatism resists precise formulation. The view raises, however, a more serious problem. Consider the implication of eliminative pragmatism that a claim is acceptable to us if and only if it is useful to us. (Let's allow Rorty to pick whatever notion ofusefulhe finds useful.) This approach to acceptability is externalist in that it requires neither that we be aware of nor that we be able to access the usefulness of what is acceptable to us. Mere usefulness, rather than accessible usefulness, determines acceptability. Some philosophers, favoring internalism, hold that whatever confers rational acceptability for us must be accessible to us. This is not our problem now, however.
The problem is that if usefulness determines acceptability in the manner implied by eliminative pragmatism, then a view will be acceptable to us if and onlyif it is true, and thus factually the case, that the view is useful to us. The appeal to usefulness, then, entails something about matters of fact, and actual truth, regarding usefulness. Call thisthe factuality requirementon eliminative pragmatism. The factuality requirement, although neglected by Rorty, manifests that eliminative pragmatists do not – and, I suggest, cannot – avoid considerations about the real nature of things, about how things really are. Such implied minimal realism resists easy escape, even among pragmatists.
Pragmatists might counter the factuality requirement with the view that we need be concerned only with what it is useful for us (rather than true) to believe regarding what is useful for us. Such a reply may appear to advance pragmatism, but it is ultimately defective, if not useless. It generates a dilemma of either an unanswered important question or an implausible regress. The unanswered important question is: must it betrue, in a particular case, that it is useful for us to believe something regarding what is useful for us? The implausible regress stems from applying the present reply to each level of questioning about whether it must be true that usefulness obtains. The result is an endless regress of increasingly complex claims about what is useful for us. Rorty himself has not embraced such a regress, and this is to his credit. The same threat of regress troubles any appeal to mere agreement about what is useful for us. If an actual fact about agreement has no place, we shall be left with a regress concerning agreement. As section 1 noted, the inevitability of making a claim regarding what is factual bears onanyassertion, not just on the assertions of pragmatism.
Barring an implausible endless regress, the factuality requirement is unavoidable. It follows that eliminative pragmatism makes acceptability a function of how things really are, at least with respect to actual usefulness for us. Given the factuality requirement, we can easily raise traditional epistemological questions, including skeptical questions, about what is in fact useful. We can ask, for example, whether it istruethat a particular view is useful for us, whether we have overwhelmingevidencethat this is true, and even whether we have non-questionbegging epistemic support that this is true. Traditional epistemology can thrive, then, even on eliminative pragmatism. Indeed, eliminative pragmatism invites the questions central to traditional epistemology, including skeptical questions. There is no problem here, then, for traditional epistemology or for the skeptical challenge of this chapter.
What about the status of eliminative pragmatism itself? Is it supposed to offer atrueclaim about acceptability? Does it aim to characterize thereal natureof acceptability, how acceptabilityreally is? Let's assume that it does. It then offers a characterization illicit by its own standard. In that case, it runs afoul of its own assumption that we should eliminate from philosophy concerns about how things really are. As a result, eliminative pragmatism faces a troublesome kind of self-defeat: it does what it says should not be done. It thereby violates its own normative standard for theories. Stable theories, in philosophy and elsewhere, avoid any such defect of self-defeat.33In keeping with the previous remarks, we can also raise traditional skeptical questions about eliminative pragmatism itself. For example, is the claim of eliminative pragmatism about acceptability epistemically supportable in a non-questionbegging manner?
If eliminative pragmatism does not offer, or even aim to offer, a characterization of the real nature of acceptability of belief, then why should we bother with it, if we aim to characterize acceptability of belief? Given the latter aim, we should not bother with it, as it is then irrelevant,uselessto our purpose. Considerations of usefulness, ever significant to pragmatism, can thus count against eliminative pragmatism itself. Pragmatists have overlooked this potential difficulty for their pragmatism. If Rorty's pragmatism is indeed antirepresentational, it will have difficulty accommodating any talk of acceptability of belief, since belief is irredeemably representational. Perhaps, then, his pragmatism is altogether irrelevant to issues of rational belief or judgment.
We now have a simple dilemma for Rorty's antiepistemological position: either eliminative pragmatism is self-defeating or it is irrelevant to typical epistemo-logists seeking an account of epistemic acceptability or justifiability. This dilemma resists easy answers, and it indicates that eliminative pragmatism fails to challenge traditional epistemology and traditional skeptical worries. Many of us, in any case, do not find a self-defeating theory useful, given our theoretical aims concerning epistemic acceptability. So, by the very standards of Rorty's eliminative pragmatism, self-defeat is troublesome for us.
What exactly does Rorty mean in speaking of his eliminative pragmatism as “better” than traditional approaches to epistemology? He remarks:
Let me just grant that, in some suitably broad sense, Idowant to substitute new concepts for old. I want to recommend explaining “better” (in the context “better standards of warranted assertability”) as “will come to seem better tous”… Nor can I see what “us” can mean here except: us educated, sophisticated, tolerant, wet liberals, the people who are always willing to hear the other side, to think out all the implications, etc. – the sort of people, in short, who both Putnam and I hope, at our best, to be.34
If Rorty seeks “better standards of warranted assertability,” he apparently will be committed to representational entities, such as beliefs. Only representational entities can have warranted assertability, at least as standardly characterized. Perhaps, then, Rorty's pragmatism is not antirepresentational after all. Even so, Rorty's characterization of “better” is defective, and perhaps even lacking in utility. It explains “better” in terms of “seemsbetterto us,” thereby using “better” (the very term needing elucidation) in the proposed explanans. In addition, it explains “us” in terms of the sort of people Rorty and others hope, at theirbest, to be, thereby relying on an obscure notion of bestness. One can, of course, be willing to “hear the other side” but then irrationally dismiss it. Certainly more is required of a group whose “seeming better” will determine whatisbetter in standards of warranted assertibility. Epistemological betterness is much too easy to come by, given Rorty's weak social strictures. As a result, Rorty's standard for betterness is largely useless in resolving epistemological disputes. In addition, since there is no necessary connection between what “will come to seem better to us” and what is “actually useful” to us, Rorty's characterization marks a departure from pragmatism, contrary to its avowed aim.
Rorty apparently deems his pragmatism impervious to the sort of self-defeat identified previously. He claims: “My strategy for escaping the self-referential difficulties into which ‘the Relativist’ keeps getting himself is to move everything over from epistemology and metaphysics to cultural politics, … to suggestions about what we should try.”35This move does not, however, escape the problem of self-defeat, because claims about what weshould tryare directly analogous to claims about what isuseful for us.Rorty's attempted “pragmatic justification of [eliminative] pragmatism” generates the troublesome dilemma noted previously, unless some odd, hitherto unexplained approach to assertion is at work. Rorty favors a disquotationalist approach to truth, but he has not offered a theory of assertion that blocks the relevance of epistemological issues in light of the aforementioned dilemma for his pragmatism.
A pragmatist might recommend that we simply eliminate any talk of acceptable belief. Indeed, it is perhaps surprising that Rorty does not propose this, given (a) that he is an antirepresentationalist and (b) that belief is inherently representational. On this proposal, however, Rorty would be barred from offering his own pragmatism as acceptable and traditional epistemology as unacceptable. Rorty would then be left with a kind of evaluative nihilism. Anteaters perhaps do not evaluate in normative terms, but we humans do, and only a desperate theorist would suggest otherwise. In sum, then, Rorty's eliminative pragmatism does not improve on traditional epistemology, but rather suffers from self-defeat. It offers, in the end, no real difficulty for the skeptical challenge of this chapter.
Is skepticism inconsistent?
William Lycan has charged that a common skeptical demand is “contradictory.” So, we need to ask whether this charge bears on the skeptical challenge under examination. Lycan has identified the general skeptical worry “that if the canons [of justification] are ultimate and cannot themselves be justified, then it seems that they are epistemicallyarbitrary, they are the rules that human beings happen to use, perhaps even the rules that human beings are built to use, but that does nothing to justify them in the normative sense appropriate to epistemology.”36Note that Lycan has thus far identified a worry concerningjustification, not questionbegging reasoning.
Lycan replies to the skeptical worry by rejecting an underlying contradictory demand:
On pain of circularity or regress, we know that some epistemic methods or procedures (whether explanatory methods or others) are going to be fundamental; so if a theorist is claiming to have discovered some such fundamental epistemic method, it is a fortiori inappropriate to respond by demanding a justification of it, in the sense of a deduction of it from some more fundamental principle – indeed, it is contradictory. Basic epistemic norms, like moral norms (and logical norms), are justified not by being deduced from more fundamental norms (an obvious impossibility) but by their ability to sort specific, individual normative intuitions and other relevant data into the right barrels in an economical and illuminating way. The present skeptical observation is tautologous, and the attendant demand is contradictory.37
Clearly, skeptics would be guilty of contradiction if they granted the existence of fundamental epistemic norms and demanded that those norms be deduced from more fundamental norms. The skeptical demand of this chapter, however, is that realists not beg questions against skeptics who have raised intelligible questions about the truth of moderate realism and the adequacy of grounding conditions for such realism regarding their being indicative of objective reality. The latter demand does not entail the contradiction just noted. Specifically, it does not entail that fundamental epistemic norms be deduced from more fundamental norms.
Lycan does consider a skeptic who (a) grants that beliefs warranted by particular methods are useful and conducive to success in goal-seeking but (b) asks the following question: what independent reason have we for thinking that theycorrespond to reality?”38Lycan mentions the observation that we cannot vacate our first-person perspective in order to compare our beliefs with objective reality, but he finds that “this observation seems pointless.”39In keeping with the latter remark, Lycan identifies his explanationist epistemology with pragmatism, on the ground that explanatory virtues are pragmatic virtues making beliefs useful.40So, he may be suggesting a response to skeptical challenges that is similar to Rorty's aforementioned pragmatism. In particular, Lycan may share Rorty's view that the appropriate reply to some skeptical questions is just: those questions are not useful; they are pointless.
We have already seen some deficiencies in pragmatist attempts to disarm skeptical challenges. The central point now is that the skeptical challenge of this chapter may be very important, or “useful,” in illuminating the limitations of human reasoning in debates with shrewd skeptics. More specifically, the challenge may illustrate that even though we have a principled opposition to questionbegging, owing to its attendant epistemic arbitrariness, we cannot avoid questionbegging in the face of some intelligible skeptical questions.
The fact that we cannot meet the skeptical challenge of this chapter does not show, or in any way indicate, that the challenge is useless, or pointless. On the contrary, the resulting lesson is very important for our understanding the limitations of human reasoning, especially in the context of debates with shrewd skeptics. What antiskeptics must explain is why questionbegging in the face of the skeptical challenge does not foster epistemic arbitrariness of the sort found, and generally opposed, in paradigmatic cases of circular reasoning. This is the key explanatory challenge raised by skepticism. Neither Lycan nor any other antiskeptic has delivered the needed explanation. Consequently, the skeptical challenge remains unscathed and directly relevant to moderate realism. Although the challenge focuses on circularity in defenses of moderate realism, it bears on any kind of first-order or higher-order support, or evidence, relevant to beliefs entailing moderate realism. Circularity arises in the presence of skeptical questions about the truth of beliefs based on one's preferred first-order or higher-order support, or evidence.
Aside from the views of Lycan and Rorty, a common antiskeptical ploy aims to show that skeptics themselves must assume a certain epistemic status for their premises and conclusions and thereby contradict their official position. (Such a result would concern whatskeptics must assume, not skepticism itself as a thesis.) Sometimes this ploy comes with the charge that skepticism is self-referentially inconsistent or otherwise self-defeating. The truth of this charge would not by itself substantiate moderate realism, but it would disarm a serious challenge for philosophers aiming to defend realism. Perhaps the skeptical challenge of this chapter suffers from a kind of self-defeat, owing to some of its implications.
Two questions arise. First, must skeptics assume that their challenge to realism is non-questionbegging either relative to their own questions or relative to questions raised by realists? Second, is the skeptical challenge of this chapter itself questionbegging in a way that undermines its epistemic significance? Even a question can be questionbegging in virtue of its implications.
Let's distinguish two families of skeptics: skeptics who simply ask troublesome antirealist questions (such as those of this chapter concerning non-questionbegging epistemic support) and skeptics who offer antirealist theses and arguments. Call the formerinterrogativeskeptics and the latterdeclarativeskeptics. Declarative skeptics demanding non-questionbegging epistemic support from moderate realists offer intelligible theses and even arguments. So, we can ask whether declarative skeptics must presume that their theses are themselves non-questionbegging and, in addition, rest on non-questionbegging support. This question prompts another: must presumefor what? A likely answer is: must presume for the skeptical challenge to succeed. Skeptics can, of course, offer theses just for the sake of argument, in order to elicit a troublesome result from an antiskeptical position.
Regarding this chapter's skeptical challenge, interrogative and declarative skeptics will succeed if and only if we have, in the wake of their antirealist challenge, no salient indication of non-questionbegging epistemic support for moderate realism. So, there is no requirement of a presumed or actual epistemic status inimical to skepticism. Is this, however, a principle of credulity for skepticism and therefore a double standard at work in the contest between skeptics and realists? Have we arbitrarily and unjustly shifted the epistemic burden away from skeptics to realists? Many realists will doubtless complain thus.
An interrogative skeptic, who issues challenging antirealist questions without offering theses or arguments, needs no principle of credulity for skepticism. Lacking a salient indication of non-questionbegging epistemic support for moderate realism, interrogative skeptics will leave their potential audience to draw the appropriate lesson from the unanswered skeptical challenge. Moderate realists, in contrast, offer a controversial ontological thesis about the objective world and thereby accrue a burden of epistemic support inapplicable to the interrogative skeptic. Moderate realists inherit a burden of epistemic support from their moderate realism. Interrogative skeptics have no corresponding disputed ontological position. Hence, they have no corresponding burden of epistemic support. Such asymmetry between the two camps makes all the epistemic difference in the world.
What about declarative skeptics? Aren't they promoting epistemic doubt while enjoying benefits thereby forbidden? Both declarative and interrogative skeptics may benefit from the fact that their skeptical challenge should not itself be a matter of dispute given our proper opposition to questionbegging in philosophica exchange. This opposition is proper owing to the threat of epistemic arbitrariness posed by questionbegging. Antiskeptics have not explained why questionbegging against the skeptical challenge does not condone epistemic arbitrariness of the sort found, and generally opposed, in typical cases of circular reasoning. We have already identified this key explanatory challenge raised by skeptics, and we can now see its role in answering the question about burden-shifting.
The skeptical challenge of this chapter does not introduce an epistemic standard foreign to moderate realists. It rather identifies a skeptical implication of an epistemic standard already (and properly) accepted by typical moderate realists. The antecedent role of this standard among realists is manifested in their attitude toward ordinary cases of circular reasoning, particularly the epistemic arbitrariness present in such cases. The skeptical challenge of this chapter is really nothing more than a call for epistemic consistency in the views of moderate realists. They properly oppose questionbegging elsewhere, but have no principled basis for making an exception in the case of the skeptical challenge. Skeptics exploit this consideration in their challenge concerning the absence of non-questionbegging epistemic support for moderate realism.
The skeptical challenge of this chapter would be ineffective if it were itself questionbegging against moderate realists. In that case, realists could plausibly challenge it as being epistemically arbitrary in the way that circular reasoning is. So, declarative skeptics do well not to challenge realists with questionbegging theses, and interrogative skeptics do well not to challenge realists with questions that are questionbegging in virtue of their implications. The skeptical challenge of this chapter illustrates that skeptics need not engage in such questionbegging. Of course, skeptics need not start with the assumption that there actually are moderate realists. They can run their story on the assumption of imagined moderate realists. The outcome would be the same: the skeptical challenge remains unchallenged and, therefore, the threat of epistemic arbitrariness persists.
5. Conclusion
Suppose that we cannot answer the skeptical challenge to deliver non-questionbegging epistemic support for moderate realism. What follows? First, realists cannot refute, or even defend themselves against, a notorious protagonist in the drama of Western epistemology, the shrewd skeptic. Second, human reasoning is more limited, less resourceful, than many epistemologists have assumed. Third, we need some way to bracket the skeptical challenge if we aim to proceed with constructive epistemology.41We cannot infer, however, that realism is false, unjustified, or not known to be true. Non-questionbegging epistemic support might not be a prerequisite for justification or knowledge. At least, nobody has shown that it is. Even so, we do well to take the skeptical challenge seriously, in order to identify the limitations of human reasoning. An important result will be increased wisdom and epistemic humility as well.
Thanks to Paul Abela, John Greco, and David Yandell for very helpful comments.
1For further taxonomy for realism, see Michael Devitt,Realism and Truth(Oxford: Blackwell, 1984) and William Alston,A Realist Conception of Truth(Ithaca: Cornell University Press, 1996).
2Richard Rorty,Contingency, Irony, and Solidarity(New York: Cambridge University Press, 1989), p. 20, andConsequences of Pragmatism(Minneapolis: University of Minnesota Press, 1982), p. 15.
3Hilary Putnam,The Many Faces of Realism(La Salle, IL: Open Court, 1987), p. 36.
4For further details on this problem, see Paul Moser, “A Dilemma for Internal Realism,”Philosophical Studies59 (1990): 101–6.
5For elaboration and defense of this definition, see Paul Moser,Knowledge and Evidence(New York: Cambridge University Press, 1989), ch. 1; cf. Alston,A Realist Conception of Truth.
6William Alston, “Epistemic Circularity,” in Alston,Epistemic Justification(Ithaca: Cornell University Press, 1989), p. 349.
7On such debates, see Moser,Knowledge and Evidence, ch. 2; Alston,Epistemic Justification, chs. 8 and9; and Richard Fumerton,Metaepistemology and Skepticism(Lanham, MD: Rowman and Littlefield, 1995), Chapter 3, Chapter 4, Chapter 5, Chapter 6, Chapter 7.
8William Alston,The Reliability of Sense Perception(Ithaca: Cornell University Press, 1993), p. 124; cf. p. 125.
9Ibid., pp. 125–6.
10Ibid., p. 129.
11Ibid., p. 133.
12Some recent attempts can be found in James Cornman,Skepticism, Justification, and Explanation(Dordrecht: Reidel, 1980); Alan Goldman,Empirical Knowledge(Berkeley: University of California Press, 1988), chapter 9, chapter 13; and Moser,Knowledge and Evidence, Chapter 2, Chapter 3, Chapter 4.
13Devitt,Realism and Truth, p. 63.
14Ibid., p. 64.
15Ibid.
16Ibid., p. 68.
17W. V. O. Quine, “The Nature of Natural Knowledge,” in S. Guttenplan, ed.,Mind and Language(Oxford: Clarendon Press, 1975), pp. 67–8.
18Ibid., p. 68.
19W. V. O. Quine, “Things and Their Place in Theories,” in Quine,Theories and Things(Cambridge, MA: Harvard University Press, 1981), p. 21.
20See W. V. O. Quine,Word and Object(Cambridge, MA: MIT Press, 1960) and Donald Davidson, “A Coherence Theory of Truth and Knowledge,” in Dieter Henrich, ed.,Kant oder Hegel(Stuttgart: Klett-Cotta, 1983). On the ineffectiveness of principles of charity against skepticism, see Paul Moser,Philosophy After Objectivity(New York: Oxford University Press, 1993), pp. 228–37, and Richard Foley,Working Without a Net(New York: Oxford University Press, 1993), pp. 67–73.
21For more detail on this challenge to naturalism, as well as a serious self-referential problem for Quine's epistemological naturalism, see Paul Moser and David Yandell, “Against Naturalizing Rationality,”Protossocioloy8/9 (1996): 81–96.
22Foley,Working Without a Net, p. 78.
23Ibid., p. 77; ibid.
24Ibid., pp. 78–9.
25Ibid., pp. 80–1.
26Richard Rorty, “Putnam and the Relativist Menace,”The Journal of Philosophy90 (1993): 449.
27Ibid., p. 449.
28Ibid. Cf. Richard Rorty, “Antirepresentationalism, Ethnocentrism, and Liberalism,” in Rorty,Objectivity, Relativism, and Truth: Philosophical Papers, Volume 1(New York: Cambridge University Press, 1991), p. 7. For critical discussion of Rorty's current antirepresentationalist approach to realism, see Paul Moser, “Beyond Realism and Idealism,”Philosophia23 (1994): 271–88.
29Rorty, “Putnam and the Relativist Menace,” p. 445.
30Rorty,Contingency, Irony, and Solidarity, p. 8.
31Rorty,Consequences of Pragmatism, pp. xxxvii, xl.
32Rorty, “Putnam and the Relativist Menace,” p. 454.
33For more detail on self-defeat, and an application to naturalized epistemology, see Moser and Yandell, “Against Naturalizing Rationality.”
34Rorty, “Putnam and the Relativist Menace,” pp. 455, 451–2.
35Ibid., p. 457.
36William Lycan,Judgement and Justification(New York: Cambridge University Press, 1988), p. 134.
37Ibid., pp. 135–6.
38Ibid., pp. 136–7.
39Ibid., p. 137.
40Ibid., p. 134.
41See Moser,Philosophy After Objectivity, for details on the latter point.
