Wordtrade LogoWordtrade.com


Review Essays of Academic, Professional & Technical Books in the Humanities & Sciences


The Epistemology of Keith Lehrer edited by Erik J. Olsson (Philosophical Studies Series, V. 95: Kluwer Academic) Keith Lehrer is one of the leading proponents of a coherence theory of knowledge that seeks to explain what it means to know in a characteristically human way. Central to his account are the pivotal role played by a principle of self-trust and his insistence that a sound epistemology must ultimately be ecumenical in nature, combining elements of internalism and externalism. Given such views the collaborative work of philosophical critics is crucial to appreciate the consequences and assess the tenability of internalism and the coherence theory of Keith Lehrer. The present book is an extensive, self-contained, up-to-date study of Lehrer's epistemological work. Covering all major aspects, it contains original contributions by some of the most distinguished specialists in epistemology, outgoing from the latest, significantly revised version of Lehrer's theory. All basic ideas are explained in an introductory chapter. Lehrer's extensive replies in a final chapter give unique access to his current epistemological thinking.

The authors who have contributed to this book were asked to take a closer look at Lehrer's epistemology from their own different perspectives. Many are, of course, critical; that is in the nature of the game. But the reader will also find several constructive attempts to defend or improve on Lehrer's theory. In the final essay, Lehrer gives his replies. All articles appear in this volume for the first time.

 After a survey Introduction by the Editor which outlines the main outlines of Lehrer’ views the contributing essays are group under five topics.

1.  The first topic, Externalism vs. Internalism, Ernest Sosa, "Epistemology: Does it Depend on Independence?" attempts to clarify and question the way coherentism undermines supervenience arguments but possibly may also undermine its own rationale by only using supportive arguments for independence without supervenience.

In "Why Not Reliabilism?" John Greco raises the question why isn't agent reliability enough? Or better, why does knowledge require, in addition to agent reliability, further conditions regarding coherence? In pursuing this question, Greco considers Lehrer's three most important arguments for thinking that agent reliability is not enough, and that therefore something further is required, such as coherence. Greco argues that none of Lehrer's arguments against a non-coherentist version of reliabilism is persuasive.

Knowledge has a functional role in science and in practical reasoning, and this functional role requires abilities involving articulation, reason-giving, and defense against objections. Hence the grounds of knowledge must be in an appropriate way available to the knower, as they are in a coherence theory.

Externalism in general, and reliabilism in particular, are open to "the opacity objection." More specifically, they are open to counter-examples such as that of Mr. Truetemp, in which a) the person in question is reliable in what he accepts, but b) this reliability is opaque to him. Contra reliabilism, such persons lack knowledge. However, coherence removes opacity and thereby issues in knowledge.

Greco argues something like this: Depending on how we interpret Lehrer's conditions for coherence, we must say either a) that Mr. Truetemp satisfies those conditions and therefore knows, or b) that in ordinary cases of perception, people do not satisfy those conditions and therefore do not know. If (a), then the counter-examples employed by the opacity objection fail to distinguish Lehrer's coherentism from non-coherentist versions of reliabilism. If (b), then Lehrer's coherentism has unacceptable skeptical results.

Gettier problems show that true reliable acceptance is not sufficient for knowledge. However, coherence requires further conditions, and these work so as to solve Gettier problems. Greco argues that a non-coherentist version of agent reliabilism can adopt a strategy very similar to Lehrer's so that his arguments are not persuasive.

Jonathan L. Kvanvig in "Justification and Proper Basing,"  believes that there is more to be said on behalf of Lehrer's view that a causal account of the basing relation is defective than has been appreciated. Causality is ubiquitous in nearly all of our experience of the world, but it is not conceptually involved in the concepts of knowledge or justification. In particular, Lehrer is right that the basing relation is not a species of causal relation. Kvanvig attempts to show why.

Todd Stewart in "Lehrer on Knowledge and Causation" takes up the same situation, arguing that Lehrer's theory of knowledge as it stands remains silent about the relationship of causation to justification. “Indeed, it seems possible to easily accommodate the view that causation is necessary for justification by simply understanding reasons claims as always being partly causal. Causation would enter the picture here because it is explicitly represented in reasons claims, and improper causation can make such reasons claims false. Of course, Lehrer's theory is also compatible with the view that justifying reasons need not be causal. Whether the incompleteness of Lehrer's theory is a virtue or a vice, and both the flexibility and difficulty of assessment this incompleteness reveals” is left up in the air.

Lastly Volker Halbach in "Can We Grasp Consistency?" attempts to show that the arguments proposed to prove that consistency is as hardly accessible as purely external facts are not conclusive. Of course, con­sistency is not accessible, if `accessible' is understood in a very strong way, for general procedures for determining whether an arbitrary given set of sentences is consistent has not been formalized. But this kind of accessibility is not required for most epistemological purposes. For those purposes, we only need to establish the consistency of our own belief system or the consistency of our belief system with another belief. Which the bedrock of Lehrer’s coherentism. Halbach asserts these discussed arguments are not conclusive. He does not claim that consistency actually is accessible in a relevant sense. Church's Theorem and Gödel's Second Theorem makes it hard to assert a general account of how justification may be obtained, if consistency is to be used as a partial criterion for justification.

2. Coherence and Personal Justification  section begins with Glenn Ross’s: "Reasonable Acceptance and the Lottery Paradox: The Case for a More Credulous Consistency." The skeptical and the inconsistency solutions to this paradox presuppose a principle of symmetry: that if I have equally good reason to accept of any ticket that it will lose as I have to accept that any other ticket will lose, I should adopt the same attitude uniformly. I should either accept of each ticket that it will lose or not accept of any ticket that it will lose. Lehrer accepts the principle of symmetry and argues that considerations of consistency show it is not reasonable to accept that any lottery ticket will lose. He then proceeds to show how he can get this result from his theory of rational acceptance. Lehrer analyzes personal justification in terms of coherence in an acceptance system. Whether a statement is reasonable to accept depends upon what statements compete with it.

Ross proposes that we can slip through the horns by adopting a consistently credulous approach: one can rationally accept that one's ticket will lose, while not accepting that of many of the others, even though one is no less confident that they too will lose.

There are good theoretical reasons to reject the Symmetry Principle. We accept in order to gain truth and avoid error. Basing acceptance on purely statistical grounds can be epistemically reasonable when the prospects for gain are great and the risks of loss small. A concomitant commitment to coherence provides the epistemological basis for our avoiding recognized inconsistency. Thus, Ross speak well of a position of credulous consistency as the most plausible of the three resolutions of the lottery paradox for reasonable acceptance. The lesson of the lottery is that we can sometimes have no better reason to accept, but reason enough.

Charles B. Cross’ "Relational Coherence and Cumulative Reasoning" begins by formalizing slightly simplified version of Lehrer's theory of relational coherence as a species of inductive rea­soning. As an exercise in logic, the hypothesis that relational coherence is a species of inductive reasoning appears to be a success. Nothing could be more natural in the context of Lehrer's theory of knowledge than an account of inductive reasoning in terms of relational coher­ence.

In "Lehrer Meets Ranking Theory" Wolfgang Spohn claims that ranking theory is the only existing theory suited for underpinning Keith Lehrer's account of knowledge and justification. However, the result of defining Lehrer's primitive notions in terms of ranking theory can disappoint. Justified acceptance has, depending on the interpretation, either an unintelligible structure or reduction to mere acceptance, and in the latter interpretation, knowledge will be reduced to true belief. Of course, this result requires a discussion of who would be disappointed.

Carl G. Wagner’s "Two Dogmas of Probabilism" rounds off section two, with a discussion of dogmatic restrictions on the representation of uncertain judgment, or on the way in which such judgment is revised, undermine the goal of faithfully representing the evidence regarding the state of the world.

While Bayesian dogmatism has begun to yield to other principled methods of probability revision, the dogma of precision is still dominant in epistemic accounts. One source of resistance to working with non-additive upper and lower probabilities is the fear that such measures must necessarily be mathematically intractable. This greatly exaggerates the true state of affairs. There is a useful theory of upper and lower expectation, as well as a generalization of probability kinematics that provides new evidence. There is a theory of consensus for upper and lower probabilities that is remarkably similar to that in Lehrer and Wagner.

3. Trustworthiness section begins with a contribution by James Van Cleve, "Lehrer, Reid, and the First of All Principles."  Lehrer’s groundbreaking work on the philosophy of Thomas Reid has revitalized Reid's philosophy. Cleve raises certain questions concerning Lehrer’s interpretation of Reid's epistemology, especially with Lehrer's view that one among Reid's principles of common sense as a linchpin of an indispensably important metaprinciple.

In "Self-Trust and the Reasonableness of Acceptance," G. J. Mattey asserts that Lehrer's doctrine of reasonableness is based solely on acceptance, which leaves him open to a charge of broad circularity, a charge avoided by foundationalist accounts of reasonableness. It is only through a relation of mutual support that acceptances can make one another reasonable. Lehrer singles out a special acceptance, that I am trustworthy in what I accept, as playing a key role in providing that support. Mattey argues that acceptance of the principle makes the mere acceptance of a piece of information, including itself, reasonable to some extent, though in an entirely generic way. It does so in the context of the acceptance system as a whole.

The principle of trustworthiness might also make itself reasonable by applying to itself directly, in which case it seems to be foundational and potentially to avoid the problem of broad circularity. But this direct application is narrowly circular and so holds no advantage in this respect over the indirect application. Because a direct application explains nothing that is not explained by the indirect approach, and indeed omits what ought to be included in any explanation of reasonableness, there is no reason to concede anything to the foundationalist. The essential ingredients in the explanation of reasonableness are to be found in the acceptance system as a whole, as is consonant with Lehrer's coherence approach to justification. The narrowly circular application of the principle of trustworthiness to itself is an aberration.

Richard N. Manning in: "The Dialectic Illusion of a Vicious Bootstrap," also finds self-trust problematic by thinking of a principle of general self-trust as not a member of one's acceptance system, but a kind of hand-stamper at the door, whose effect is to mark contents as accepted, thereby making them available in truth directed reasoning and justification. Self-trust, then, operates, not as a principle of detachment of an acceptance operator, but to distinguish which among those contents we can think are to be exploited in epistemic inquiry. The way Lehrer conceives of self-­trust—as a principle which is special, but special among acceptances to be exploited in bootstrapping—forces him to attempt give accounts of justification on two levels, the first subjective and doxastic, the second objective and veridical. The express purpose behind the objective accounts is in each case the avoidance of Gettier and Gettier related counterexamples. The lesson of Gettier and other cases may be that a certain kind of skepticism, the skepticism which claims that in any given case our justified empirical beliefs may fall short of knowledge, is unavoidable. The gap between personal, doxastic justification and veridical epistemic warrant will seem to need closing, by hook or by crook or by bootstrap, only if we fail to separate these two threats. And we will fail to separate them only if we think that the only way to ensure our grip on reality is through the identification of a justificatory procedure that will give us the ability to weed out the false from the true in our particular beliefs. The coherentist who feels the need for bootstraps shares at least that much with the foundational Cartesian: the urge for a principle the satisfaction of which by any given case of believing will show that belief to be true. Surely all the cooperation the world can provide will not make you knowledgeable if you lack the epistemic virtues embodied in the concept(s) of justification. And in this same way we can answer the question which motivated Lehrer's bootstrap in the first place, of how justification, in the light of these results, is epistemic at all. Justification is epistemic because of the intrinsically epistemic nature of what gets justified and does the justifying.

4. Undefeated Justification and the Gettier Problem introduced in . Manning’s essay is given a closer look by Hans Rott’s "Lehrer's Dynamic Theory of Knowledge." Rott uses the concepts and ideas borrowed from belief revision the­ory to elaborate Lehrer's dynamic theory of knowledge. One of the most relevant distinctions for belief revision is that between foundationalist and coherentist approaches in epistemology. The foundationalist's and the coherentist's views seem potentially complementary. Rott refers to his worked out framework that characterizes the two fundamen­tally different perspectives on the process of belief revision. In the current theories of belief change, belief bases are not supposed to carry reductive connotations. Basic beliefs are distinguished from derived beliefs only by the fact that they are somehow `given', either explicitly or as things that are taken for granted. Givenness is not at all supposed to imply indefeasibility here.

In "Some Remarks on the Definition of Lehrer's Ultrasystem" Gordian Haas  asserts that, according to the analysis of knowledge proposed by Lehrer, knowledge equals undefeated justified acceptance. Undefeated justification is then spelled out as coherence with every element of the ultrasystem, as Lehrer calls it. The question arises how this key-notion should be defined. Two definitions of the ultrasystem which have been proposed by Lehrer are investigated by Haas. He argues that both definitions are flawed and offers instead an alternative formal proof that is a third and simpler way to define the ultrasystem.

Jacob Rosenthal, "On Lehrer's Solution to the Gettier Problem," claims that, a true belief of a subject is knowledge iff the subject has a justification for the belief that remains a justification when in the subject's acceptance system all false beliefs are replaced with the corresponding true ones. It is not required that every justification of the subject has this property — a belief may be justified in many different ways, and it is no harm when some of them are faulty. But at least one possible justification, a justification that the subject could use if asked, must be able to survive the mentioned strong correction of the acceptance system. Then, and only then, is the true belief knowledge. In comparison with Lehrer-type proposals, this proposal for solving the Gettier problem has the advantage of involving just two belief systems, whereas the former have the advantage of using merely the concept of a justified belief (relative to a system of beliefs) and not the more demanding concept of (sufficient) reasons for a belief (relative to a system of beliefs).

Rosenthal  offers two proposals for solving the Gettier problem: first, the modified Lehrer proposal, and second, the one just mentioned. Rosenthal is not sure whether they are equivalent. But they could both be satisfactory solutions to the Gettier problem and yet not be equivalent, as long as they agree in all clear cases. There are borderline cases of belief in which one does not know whether to call the belief in question knowledge, because the intuitions are unclear or divided. No proposed criterion can be dismissed just because it decides a borderline case in this or that way. As long as it gets the clear cases right, it may count as a solution of the Gettier problem, and so there may be many nonequivalent solutions. But Rosenthal fears that sooner or later a clear example will come up for which the two proposals considered here fail, as was the fate of so many of their predecessors.

5.  The section on Skepticism begins with John W. Bende’s essay entitles. "Skepticism, Justification and the Trustworthiness Argument." Lehrer sees that an evaluation system dereft of principles of trustworthiness and reliability will not produce coherence that supports justification or knowledge. He, therefore, requires them as part of one's personal evaluation system and believes they carry through as truths to the ultrasystem. Knowledge requires this. But, for all his talk of loops—loops of trustworthiness, and loops of explanation—we have been hard-pressed to see how those loops are not circles in the epistemically vicious sense in which no justification is generated. The most "direct" argument for the justification of my acceptance that I am trustworthy is supposed, by Lehrer, to be the "loop of trustworthiness," but this argument is found wanting. Lehrer believes that there is good inductive support for principle, but very little in the way of a detailed argument for this has been provided. We are, once again, left with the old Pyrrhonian, and neo-Humean worry that Lehrer has begged the question against the skeptic. Skepticism is acknowledged by Lehrer as a serious and coherent threat to knowledge. This paper has been an attempt to show that Lehrer's recent theory lacks the theoretical agility to loop around the skeptic and lead us to knowledge and meta-knowledge.

In "Coherence, Knowledge and Skepticism," Peter Klein proposes that we need not "answer" or "neutralize" (in Lehrer's sense of those expressions) the academic skeptic's challenge to our acceptances. We can show that the arguments employed by such a skeptic do not succeed. Further, we can coherently not accept the proposition (T) namely, I am trustworthy (worthy of my own trust) in what I accept with the objective of accepting something just in case it is true. However abstaining from accepting T does not necessarily lead to embracing academic skepticism. Such skeptics accept that T is false. The third alternative—the pyrrhonian alternative —is to withhold accepting T. That would bring with it withholding any proposition whose defense against objections depended upon accepting T. It seems to Klein that such a view is satisfyingly coherent and, indeed, better fits Lehrer's own preference system than the model of justification he has suggested.

David A. Truncellito’s "The Ultrasystem and the Conditional Fallacy,"  attempts to defend Lehrer's account of justification and knowledge against Shope's conditional fallacy objection, which is also close to Klein’s objections. Perhaps Lehrer's description of the ultrasystem is not adequately perspicuous— for instance, he might describe the members of the acceptance system in terms of their content (that is. as 'p' rather than as 'S accepts that p'), and he might be more clear that the ultrasystem is distinct from, albeit related to, the acceptance system—which has thus led some readers to overlook the fact that utrasystem is a new system, and a theoretical system at that, rather than an actual modification to the acceptance system.

“Coherence, Circularity and Consistency: Lehrer Replies summarizes and concludes the volume. Keith Lehrer’s responses by unifying the critiques and admitting alternation of his ideas are influenced by his more perspicuous critics. The volume on  the whole offes critical entry to Leher’s epistemology  with crucial areas highlighted without overwhelming in technical detail.

 The Oxford Handbook of Epistemology edited by Paul K. Moser  (Oxford Handbooks in Philosophy: Oxford University Press) The Oxford Handbook of Epistemology contains 19 previously unpublished chapters by today's leading figures in the field. These chapters function not only as a survey of key areas, but as original scholarship on a range of vital topics. Written accessibly for advanced undergraduates, graduate students, and professional philosophers, the Handbook explains the main ideas and problems of contemporary epistemology while avoiding overly technical detail.

Epistemology, the theory of knowledge, is at the center of mainstream philosophical efforts. The Oxford Handbook of Epistemology contains nineteen previously unpublished chapters on the theory of knowledge by today's leading figures in the field. These chapters function not only as a survey of key areas but also as original and interesting scholarship on vital topics currently of great interest. Written accessibly for advanced undergraduates, graduate students, and professional philosophers, the Handbook explains the main ideas and problems of contemporary epistemology while avoiding overly technical detail. see Series

Norms, Naturalism and Epistemology: The Case for Science Without Norms by Jonathan Knowles (Palgrave MacMillan) Jonathan Knowles argues against theories that seek to provide specific norms for the formation of belief on the basis of empirical sources: the project of naturalized epistemology. He argues that such norms are either not genuinely normative for belief, or are not required for optimal belief formation. An exhaustive classification of such theories is motivated and each variety is discussed in turn. He distinguishes naturalized epistemology from the less committal idea of naturalism, which provides a sense in which we can achieve epistemic normativity without norms.

Scientists draw conclusions, based on experimentation and on reasoning about and within particular scientific theories, and in doing so they are following certain practices. It has to be said that these practices constitute a normative component of science, if only because one broad and obvious sense of “norm” applies whenever practices of any sort provide any sort of guidance.

Scientists draw conclusions, based on experimentation and on reasoning about and within particular scientific theories, and in doing so they are following certain practices. It has to be said that these practices constitute a normative component of science, if only because one broad and obvious sense of “norm” applies whenever practices of any sort provide any sort of guidance. 

Jonathan Knowles, as I read him, has no problem with any of this, despite the perhaps overly dramatic subtitle of his new book Norms, Naturalism and Epistemology: The Case for Science without Norms. (Palgrave Macmillan, 2003) His phrase “science without norms” (which he refers to as a slogan) is not meant to be taken as strictly literal or exceptionless. (See page 145, for example.) Knowles is making the case, not for the silly view that science has no normative practices, but for a naturalistic reconception of epistemology and the philosophy of science, one which jettisons the traditional notion of norms as comprising an a priori philosophical foundation for science that is separate from scientific theory itself. 

Anyone familiar with the work of the late and revered philosopher Willard Van Orman Quine will recognize that just such a reconception of epistemology and the philosophy of science is a project that Quine was engaged in. Indeed, the seminal essay for this was his “Epistemology Naturalized.” In a nutshell, a “naturalized” epistemology of science would be one that justifies scientific theory using only the resources of natural science itself. Others, most notably Nelson Goodman and Thomas Kuhn, can be read as sharing in this project, as well. The particular contribution Knowles makes is in terms of exploring the abandonment of the philosophical pretension of separate norms, and the implications of naturalism for the deflation of philosophical pretensions somewhat more generally. 

For this, he draws upon Quine, Goodman and Kuhn, as well as upon others, including Hilary Kornbluth, Larry Laudan and Edward Stein. While at times critical of the views of these philosophers, he clearly rides the wave of the overall project of naturalizing epistemology.  

Nevertheless, he rejects the pretensions of naturalized epistemology to provide better answers to at least some of the questions of traditional epistemology. Questions about the norms of science, as those questions have traditionally been conceived in epistemology and the philosophy of science, are to be rejected, on his view. Rather than seeking to provide better answers to these questions, the outcome of naturalism for Knowles is to dissolve the questions in Wittgensteinian fashion.  

This book is likely to be of interest to, and useful for, those familiar with the analytic tradition of naturalized epistemology and concerned about the relationships, at the most abstract level, between substantive scientific theory and the practices which guide the revision and development of science. Like other work related to the grand project of naturalism, it presents an intriguing view of the scientist as fundamentally autonomous. What is perhaps most distinctive here is the extent to which Knowles believes this epistemological independence goes. The scientist, according to Knowles, is in no need of aid from the philosopher or anyone else in the legitimating of scientific practice.

Epistemic Justification: Internalism Vs. Externalism, Foundations Vs. Virtues by Ernest Sosa, Laurence Bonjour (Great Debates in Philosophy: Blackwell) Epistemology is the theory of episteme, of knowledge. Ever since Plato it has been thought that one knows only if one's belief hits the mark of truth and does so with adequate justification. The issues debated by Laurence Bonjour and Ernest Sosa concern mostly the nature and conditions of such epistemic justification, and its place in our understanding of human knowledge.

Bonjour defends a traditional, internalist epistemology, according to which epistemic justification derives from the subject's (1) taking what is given to his conscious awareness, and (2) accepting claims or steps of reasoning on an a priori basis. Rejecting the emphasis of epistemol­ogy on the concept of knowledge, he is mainly interested in the ques­tion of whether we have or could have good reasons to believe in an external world of the sort that we normally take ourselves to inhabit, and in the question of what could possibly constitute such reasons. His answer to the latter question is internalist and foundationalist, in that it takes the justification for claims about the external world to begin from apperceptions of present states of consciousness (mainly sensory consciousness) and to proceed from there on the basis of (allegedly) a priori reasoning, specifically an argument that the truth of our beliefs about the external world constitutes the best explanation of our sensory experience.

Bonjour also rejects recent proposals according to which justifica­tion can derive from contingent factors external to the consciousness of the believer: factors involving how that belief is caused, or how well it tracks the facts, or how reliably it is formed. While he grants some lesser epistemic status to beliefs that do satisfy such external require­ments of causation, tracking, or reliability, he insists that the more important issues for epistemology, and certainly the more prominent and important issues in the tradition, are the questions that he wishes to address, concerning the internally accessible reasons that one might have for one's beliefs about the world around us.

Sosa had in earlier work (as Bonjour points out) drawn a similar dis­tinction, between animal knowledge and reflective knowledge, so on the issue of whether there are two importantly different kinds or levels of epistemic assessment they are in agreement. But there is still a rel­evant difference in focus and emphasis. Sosa is interested in understanding the conditions of animal knowledge, and not only those of reflective knowledge. His cognitive virtues account of animal knowledge is reliabilist. About such knowledge his views are thus in line with contemporary externalism. In distinguishing between animal and reflective knowledge, however, and in requiring reliability for animal knowledge his views agree surprisingly with Descartes's. Recall the passage early in Meditation Three where the cogito is said to derive its high epistemic standing from its clarity and distinctness, which, we are told, it could not possibly do if clarity and distinctness were to the slightest degree unreliable (and could ever lead us to a false belief). Sosa likewise takes reliability to be necessary in a source of justifica­tion, but of course not sufficient.

Sosa rejects the sort of internalist foundationalism favored by Bonjour, while agreeing to put aside issues of knowledge and its conditions, in order to focus on epistemic, rational, justification. He agrees that a belief's having a reliable source is not enough to render it justified. The source must be a cognitive virtue seated in the subject. This already yields a kind of internalism. Moreover, the source must operate fundamentally through the promptings of experience, through either introspective or perceptual belief formation. Reflective justification goes beyond such unreflective rational justification in requiring a coherent epistemic perspective that underwrites the belief thus justi­fied. What the externalist virtue theorist will add, in sharp disagree­ment with any kind of internalism, including Bonjour's, is that there is no way to delineate what a cognitive virtue is in general, if we prescind from all contingent relations that such belief formation might bear to our external environment. In understanding rationality, having a reason, being reasonable, and the like, as these notions apply to empirical beliefs, we must make proper allowance for such external factors.

Skepticism and the Veil of Perception by Michael Huemer (Studies in Epistemology and Cognitive Theory: Rowman & Littlefield) Since Descartes, one of the central questions of Western philosophy has been that of how we know that the objects we seem to perceive are real. Philosophical skeptics claim that we know no such thing. Representationalists claim that we can gain such knowledge only by inference, by showing that the hypothesis of a real world is the best explanation for the kind of sensations and mental images we experience. Both accept the doctrine of a 'veil of perception': that perception can only give us direct awareness of images or representations of objects, not the external objects themselves. In contrast, Huemer develops a theory of perceptual awareness in which perception gives us direct awareness of real objects, not mental representations, and we have non-inferential knowledge of the properties of these objects. Further, Huemer confronts the four main arguments for philosophical skepticism, showing that they are powerless against this kind of theory of perceptual knowledge.
Author Summary: Skepticism and the Veil of Perception concerns one of the central problems in Western philosophy, the nature of our knowledge of the external world (the world outside the mind). Philosophers who study the nature of knowledge have been beset since ancient times by the arguments of philosophical skeptics. Skepticism, in the sense in which I am using the term here, is the position that it is impossible to know anything about the external world at all. The idea of "the veil of perception" is the idea that all one is ever immediately aware of is one's perceptual experiences, or representations of external objects, not the objects themselves‑an idea which is often supported by the claim that our perception "distorts" the objects we perceive, rather than presenting them as they really are (perceptual illusions and perspective phenomena are often cited in this connection). Thus, our processes of perception are like a "veil" standing between us and the real world, preventing us from ever really perceiving (objective) reality. Evidently, the doctrine of a veil of perception is closely related to skepticism; indeed, I believe the two ideas feed off one another and will stand or fall together‑but that is something needing further discussion later.

I oppose both of those ideas. In this book, I explain how we are aware of, and have knowledge of, the character of the external world. I defend a doctrine sometimes called "naive realism," which holds that perception gives us direct awareness of the external world and that it enables us to know (some of) what that world is like. Besides giving a positive theory of perceptual awareness, l show the flaws in the major arguments that have been used to motivate skepticism…

What have we learned from our encounter with skepticism? For one thing, we learned that skepticism is false‑you really do know that you have hands and that this book you are reading exists, in case you were wondering. But more importantly, we saw the power of direct realism, and in particular of the version of direct realism presented in chapters IV‑V, in accounting for this knowledge. We have seen that indirect realist theories are vulnerable to skeptical assault where my version of direct realism is not. That I take to be a strong argument in favor of my theory of perception.

Perceptual knowledge‑knowledge of the characteristics of the things we ordinarily perceive‑is often thought of (rightly, I think) as a central and paradigmatic kind of knowledge, and it forms a natural starting place for the theory of knowledge. Thus, Descartes began his great epistemological excursion by observing, "Whatever I have up till now accepted as most true I have acquired either from the senses or through the senses."'6 Of course, Descartes was later to go on, in true rationalist style, to argue that in fact the knowledge gained by the senses was derivative, and inferior to purely rational knowledge. Like me, Descartes saw the refutation of skepticism as an occasion to make a larger point about the structure of human knowledge. The point that I make, however, is quite different from, and almost diametrically opposite to Descartes's. I think our discussion of the refutation of skepticism has shown that perceptual knowledge is foundational, and that the knowledge gained by the senses is exactly on a par with the knowledge gained by pure reason‑both are justified in the same way, under a single principle of foundational justification. I think that Descartes, Locke, and the other representationalists failed to refute skepticism, and that their failure is instructive. It reveals a fundamental mistake that they made: they mistook a cognitive process for a special, internal object of cognition.

Most philosophers who have made this mistake have not gone on to embrace skepticism‑I think they had a little too much sense for that, and could not quite stomach a conclusion that was so directly at odds with their thought and practice in everyday life. Even so, I believe that such a mistake must have a profound effect on one's thinking, the more profound the more consistent one is. One such effect I have hinted at earlier; if one holds premises that in fact imply either global or external‑world skepticism, even if one does not embrace those forms of skepticism, one may easily be moved towards less radical forms of skepticism, skepticism concerning more controversial claims. The reason is that here one's common sense (such as it is) would not serve as the check preventing one from drawing out and embracing the logical consequences of one's mistaken assumptions. This means that such a fundamental epistemological mistake as we have seen can produce very widespread errors about what sorts of things we are justified or unjustified in believing.

A second effect of the mistake, I suspect, is to produce a kind of discomfort, the more so the more philosophically minded one is, with the whole of one's beliefs about the external world, and perhaps the whole of one's beliefs about everything. I say this because the arguments indirect realists have been able to produce for the existence of external objects generally seem disappointingly weak at best, and it is open to question whether they have any cogency at all. The indirect realist who is honest with himself can never quite be sure that he is not being irrational in holding beliefs about the physical world, for all that he may find it unavoidable to do so. Hume, indeed, happily embraced this conclusion, being both a skeptic and an indirect realist. Which is to say: he believed in external objects, but he also believed there was no logical reason to do so. He thought that this illustrated the general conclusions, (a) that reason is impotent to produce knowledge and (b) that human beings are irrational. He thought this was a valuable lesson for people to learn, too, for purposes of putting us in our place:

But could such dogmatical reasoners become sensible of the strange infirmities of human understanding, even in its most perfect state, and when most accurate and cautious in its determinations, such a reflection would naturally inspire them with more modesty and reserve, and diminish their fond opinion of themselves .... [I]f any of the learned be inclined . . . to haughtiness and obstinacy, a small tincture of Pyrrhonism might abate their pride, by showing them that the few advantages which they may have`attained over their fellows are but inconsiderable, if compared with the universal perplexity and confusion which is inherent in human nature."

I quote this passage because, although Hume's enormous influence on modem philosophy goes unquestioned by virtually all contemporary philosophers, and although many of Hume's ideas remain under lively discussion, seldom do scholars discuss what Hume himself thought was ultimately the good of his philosophical reflections. I suspect that Hume was right about the psychological effects of his philosophical reflections, though not about the desirability of those effects, nor about the truth of his doctrines.

Theory of Knowledge by Keith Lehrer (Westview) introduces students to the major traditional and contemporary accounts of knowing. Beginning with the traditional definition of knowledge as justified true belief, Lehrer explores the truth, belief, and justification conditions on the way to a thorough examination of foundation theories of knowledge, the work of Platinga, externalism and naturalized epistemologies, internalism and modern coherence theories, contextualism, and recent reliabilist and causal theories. Lehrer gives all views careful examination and concludes that external factors must be matched by appropriate internal factors to yield knowledge. This match of internal and external factors follows from Lehrer's new coherence theory of undefeated justification. In addition to doing justice to the living epistemological traditions, the text smoothly integrates several new lines that will interest scholars. Also, a feature of special interest is Lehrer's concept of a justification game.

This second edition of Theory of Knowledge is a thoroughly revised and updated version that contains several completely new chapters. Written by a well-known scholar and contributor to modern epistemology, this text is distinguished by clarity of structure, accessible writing, and an elegant mix of traditional material, contemporary ideas, and well-motivated innovation.

Headline 3

insert content here