I'm well over half way into BonJour's excellent book, and have read his rejoinders to objections to the tenability of a moderate rationalist epistemology. I won't concern myself much with these (but I believe he's done an excellent job of deflecting the various criticisms, many of which are ill-conceived, like the charge of "dogmatism"), and will note that unlike many modern philosophers (particularly of the neo-analytic tradition), BonJour takes the notion of concepts seriously. In so doing, he quotes an author who acts as though there's a clear divide between our "concepts" and the world of things and objects. BonJour, in characteristically good sense, poses the question "but why should we think of them this way" (my own words)? He notes we can all agree that the possession of a concept of a given thing X suggests at least that we can think of X's, classify things as X's and often recognize X's in appropriate circumstances. However, why should this lead us to think that concepts are in any way dichotomous with the notion of reality? He asks, if not reality, whence do our concepts ultimately derive? My concept of redness derives from my encountering things which instantiate the property; one may object that the claim that "nothing can be red and green all over, in the same respects", pertains solely to my concept of redness, but why should we think that it follows from this that what is represented in mind is merely some subjective entity and not an objective property of something? With no good answer to count in the favour of such a view, we have no reason to do so. I am glad that BonJour gives such serious consideration to the role of concepts, because many seem to merely gloss over it; yet it is vital for any epistemological viewpoint to take their role most seriously, lest one confuse how that role might be filled. BonJour is of course concerned with concepts and their interrelation with the world conceived of as a ding an sich, and given his criticism of Kant, I think he'd agree that the Kantian worldview (particularly the impositionist strain) is hard-pressed to give a good account of our concepts and their relation to reality without lapsing into subjectivism.
BonJour later, in a section focusing on metaphysical objections to rationalism, goes on to consider the nature of various objects of a priori cognition (e.g. numbers) and recounts some alternative ways of explicating concept-formation. Though he seems to favour a sort of Platonism, he gives positive consideration to the Aristotelian notion of forms being instantiated by various things in the real world, as well as concept empiricism (roughly similar, and closer to the Objectivist epistemology, where abstraction from concretes allows the formation of concepts), and claims that it at least can account for the causal role which abstract concepts figure in justifying various a priori claims. The book should thus be of interest to Aristotelians as well as other rationalists of the moderate sort, for its close examination of various oft-neglected areas of philosophy.
Been a while since I've last compiled one of these, and as I'm taking a break from the forum to catch up on some reading, what better time than now? I'll be rather brief this time round - primarily because I've yet to finish the book in question. At any rate, I'd recommend to anyone interested in epistemological issues to pick up Laurence BonJour's excellent In Defense of Pure Reason. As an author he is anything but unclear or pedantic. He writes in remarkably clear, jargon-free prose (to the extent that the subject allows for, anyway), and is of a kind with Hans Hoppe who is himself known for his rigorous, logical method of presenting arguments.
BonJour's purpose of authoring this fine book is a rehabilitation of rationalism against a growing tide of scepticism and the failed doctrines of radical empiricsm. He does so by first undermining the arguments and rationale for moderate forms of empiricism, e.g. the evocation of the analytic-synthetic dichotomy by moderate empiricists in the defence of their doctrine (which allows them to conflate all a priori truths with analytic truths, which they take to be necessary in virtue of their logical structure or often their meaning, thereby allowing them to entertain a weakened notion of the a priori.) BonJour artfully explodes the doctrine of apriority being tantamount to analyticity, by deftly demonstrating that if such a conflation is made, it allows analyticity no useful role to play, and will at any rate rely on a rationalist notion of intuition (i.e. grasping that the logical truths analytic statement depend on, are, in fact, valid.) Even so innocuous a notion as reasoning (e.g. inference from premises to conclusion) is not free from an element of a priori justifiability, meaning the moderate empiricist is hard-pressed to escape the force of BonJour's arguments. Besides, the notion of analyticity as is commonly advanced (containment of a predicate in the subject) cannot even characterize most obviously analytic truths (e.g. "either A exists or A does not"), and where the notion of meaning or contradiction is borrowed upon, one must first assume the justifiability of the logical truths in question. The author demonstrates how empiricists often equivocate between different meanings of the term "analytic" in defence of their doctrine. He also demonstrates the utter vacuity of the notion of the analytic being a matter of linguistic conventions, as he shows that this cuts no philosophical ice as far as justifiability goes (which is what the a priori is, after all: a form of justification.) Most interesting, perhaps, is BonJour's unseating of Kant as arch-rationalist, by showing that his belief in the imposition of stucture by the mind on reality deprives Kant of the ability to argue for a genuinely coherent notion of synthetic apriority, thereby rendering him at best a moderate empiricist. He closes his chapter on moderate empiricism by noting the self-refuting nature of the epistemological doctrine (familiar to anyone conversant in Austrian methodology.)
Present also is a clear, rigorous delineation of the questions at hand: BonJour separates the ontological question of the necessity or contingency of a truth from the epistemological question of its justifiability (i.e. apriority or aposteriority) and from syntactic matters (analyticity vs. syntheticity.) The categories are all too often confused, even amongst the more incautious of the Austrians. I've not arrived at BonJour's criticism of Quine and radical empiricism yet, or his positive account of the a priori, but I'm aware that like Barry Smith, he too advances a fallibilistic notion thereof, and as such classifies his rationalism as "moderate". What he has written is pretty much in line with the writings of the likes of Hoppe, and should be read by anyone interested in advancing a coherent defence of the Austrian method. He neatly evades the pitfalls of Cartesian rationalism and excessive faith in reason, and thereby aligns himself with those more moderate rationalists such as Aristotle (whom he explicitly identifies as such.)
All in all, a book well worth reading. I look forward to seeing his solution to the problem of induction. This is not a work for beginners, but for anyone who's read Mises's, Hoppe's or Martin Hollis's works (or those of Brand Blanshard), this work should bring them up to date with modern arguments in the area of epistemology, exposing them to cutting edge thought on the matter. Henry Veatch's Two Logics should also be read for a similarly sustained attack on modern scientism and radical empiricism.
PS: I'd also like to promote this great article by BonJour, though it's not really germaine to this post.