Michael Bergmann: Justification without Awareness: A Defense of Epistemic Externalism.

Murphy, Peter

Michael Bergmann

Justification without Awareness: A Defense of Epistemic Externalism.

New York: Oxford University Press 2006.

Pp. 272.

US$70.00 (cloth ISBN-13: 978-0-19-927574-8); US$35.00 (paper ISBN-13: 978-0-19-956242-8).

This book significantly raises the standard of rigor that future discussions of the internalism/externalism debate will have to meet. Besides its rigor, it is also impressive in scope, as Bergmann provides interesting discussions of connected topics like epistemic deontology, justification defeaters, epistemic circularity, skepticism, and the new evil demon problem.

In the first of the book's two parts, 'Against Internalism', Bergmann argues against two important motivations for internalism, and also against Mentalism. One motivation comes from Laurence Bonjour's Norman case. According to Bonjour, Norman's reliable clairvoyant beliefs about the president's location are unjustified because he is unaware of anything that these beliefs have going for them; from his perspective, Norman's beliefs are no different from a hunch or an arbitrary conviction; hence, from his perspective, it is an accident that these beliefs are true. According to Bonjour, what Norman lacks is awareness that his beliefs possess anything that contributes to their justification. Bergmann levels a dilemma at the awareness requirement that internalists like Bonjour want to impose. Is it strong in the sense that Norman must conceive of some justification-contributor as being relevant to the truth or justification of his belief? Or is it weak in the sense that Norman must be aware of something, but not necessarily of anything that contributes to the justification of his beliefs? If it is strong, Bergmann argues, such conceiving had itself better be justified. This triggers a regress: that such a conceiving is justified will require that Norman conceive of something that contributes to that awareness' justification; but that second conceiving had better be justified too; but it will be only be if the believer conceives of something that contributes to its being justified, etc. The problem with this regress is its complexity: soon enough, the required contents get too complicated for us to entertain, and an implausible form of skepticism ensues. The alternative is to say that some other form of awareness is required. But, Bergmann argues, this is just to abandon the alleged lesson of the Norman case. For any other required awareness will be compatible with it being, from Norman's perspective, an accident that his beliefs are true. Chapter 1 spells out this dilemma; Chapter 2 argues against attempts to circumvent it, attempts made by Richard Fumerton, Laurence Bonjour, Evan Fales, and one inspired by Timothy McGrew's work.

Chapter 4 takes up the second motivation for internalism, deontologism. This is the view that justification is to be analyzed in terms of notions like duty, blame, obligation, and responsibility. Bergmann looks at attempts to show that one or other deontological analysis entails internalism. After ar guing that these attempts fail, he offers an intriguing explanation of why it appears as if deontology entails internalism: the plausible view that there is an awareness requirement on being epistemically blameworthy appears to entail, but really does not entail, that there is an awareness requirement on being epistemically blameless.

Though Bergmann helps to clarify what views of epistemic justification count as internalist, and which ones don't, his own way of dividing up the territory is peculiar. It leaves the view that he dubs 'mentalism' as neither internalist nor externalist. This view, the focus of Chapter 3, says that justification is entirely determined by one's mental states and the basing relation. This view is not internalist, according to Bergmann, because it does not impose an awareness requirement; nor is it externalist since it requires that the inputs into our belief-forming processes be mental states--something externalist views deny.

The second half of the book details Bergmann's own externalist view of epistemic justification. It imposes a no-defeater and a proper function condition on justified belief. The official version says that S's belief B is justified iff (i) S does not take B to be defeated, and (ii) B is produced by faculties that are (a) functioning properly, (h) truth-aimed, and (c) reliable in the environments for which they were 'designed' (133). Chapter 5 argues for the teleologically laden condition, (ii), by cases and by the account that it provides of the connection between justification and truth. Chapter 6 covers the no-defeater condition, (i). This condition departs from the most popular no-defeater condition by restricting justification defeaters to believed justification defeaters; consequently things that one does not believe, but should believe, are not defeaters. It also departs from the standard view by including all believed justification defeaters as defeaters; consequently, even an unjustified belief that some other belief is defeated counts as a genuine defeater. Bergmann ably defends both of these unorthodox claims.

Chapters 7 and 8 round out the discussion with two objections to externalism. To the objection that externalism approves of epistemic circularity, Bergmann offers a pair of arguments that the relevant kind of epistemic circularity is benign; he also offers an important diagnosis of why all epistemically circular arguments appear problematic; and he suggests that we have noninferential knowledge of the reliability of some belief forming processes. To the objection that externalism delivers an unsatisfying response to skeptical arguments that use demon scenarios, Bergmann argues that the most plausible versions of internalism offer responses to skepticism that are similarly unsatisfying.

This is a very good book. It will help readers get a better handle on how different aspects of the internalism/externalism debate relate to one another. More importantly, it may well constitute the most compelling case yet for a position in this debate. All serious epistemologists should study it.

Peter Murphy

University of Indianapolis