A Brief Interview with Laurence BonJour

By Braeden Giaconi

Garden of Ideas editor-in-chief

Note: the interview has been lightly edited for print. 

I was honored to speak with Laurence BonJour, professor emeritus of philosophy at The University of Washington, via email over the last month. Professor BonJour is one of the most eminent epistemologists in recent analytic philosophy. His major work, The Structure of Empirical Knowledge (1985), is, along with Donald Davidson’s “A Coherence Theory of Truth and Knowledge” (1986), one of the two most important coherentist theories of the late twentieth century. Professor BonJour and I spoke about coherentism, foundationalism, Wilfrid Sellars, John McDowell, and the influence of his teacher Richard Rorty. -Braeden Giaconi, Editor-in-Chief

BG:   Your Coherentist theory, given primarily in your book The Structure of Empirical Knowledge, is one of the two most important Coherentist theories of recent epistemology. The other, of course, is Donald Davidson’s, which he gives in the paper “A Coherence Theory of Truth and Knowledge”. Importantly, your Coherentism accommodates experiential input through what you call the Observation Requirement, whereas Davidson denies outright that a belief can have anything but a causal relationship to experience. Given that your and Davidson’s theories are frequently discussed together, could you describe your general views, particularly before you renounced Coherentism, towards Davidsonian Coherentism, and how you view his theory in relation to yours?

LB: Here I am hampered by no longer having a copy of Davidson’s paper readily available. A further problem is that Davidson’s epistemology is thoroughly enmeshed in his views on language and meaning, and his views on truth, in a way that makes it hard to sort out in any brief space. But a simple response is that as long as truth is understood as correspondence to independent reality (as I believe that it must be, though Davidson apparently would not agree), there is no basis for thinking that an internally coherent system of beliefs having merely causal relation to experience is thereby likely to be true—and therefore no reason to think that the beliefs in such a system are epistemically justified. My coherence theory attempted to show how experience could affect such a system in a way that would be relevant to epistemic justification. I no longer believe that my attempt—or probably any of this general sort—succeeds. But Davidson seems to me to make no effort at all in this direction, so that the causal impact of experience becomes simply irrelevant to justification, leaving his theory entirely vulnerable to all of the standard objections to a pure coherence theory (outlined in my book and in many other places by others).

BG:     In 20th century epistemology we witnessed an uncharacteristic turn away from foundationalism, brought on by, among other things, Wilfrid Sellars’s attack on the Myth of the Given in his essay Empiricism and the Philosophy of Mind. From then until perhaps the turn of the century, marked by yours and Richard Fumerton’s edited book Resurrecting Old-Fashioned Foundationalism, Coherentism was very popular among epistemologists. Now, one could say that either there has been something of a return to the old foundationalism, or modifications of it like McDowell’s, or a more mixed environment in which both theories are important. In your view, does this arc of Coherentism over the course of the second half of the 20th century tell us anything about Coherentism, epistemology, or philosophy more generally? What can we learn from what one could call the failure of Coherentism? 

LB: This is a very large question, which I will only take a quick stab at here. Sellars convinced me and many others that classical foundationalism was untenable. In a way, a kind of minimum response to Sellars—abandoning the given, but retaining much of the standard epistemological dialectic (centering around the idea of epistemic justification as internally accessible reasons for thinking that beliefs are true in the correspondence sense)—was a coherence theory of justification. I spent some years trying to make such a view work and eventually decided that it cannot be done. (There were many problems, including doubts about the concept of coherence itself, but the main one being the ones involved in the attempt to give a justificatory role to some experiential element beyond mere coherence, as reflected in the infamous “doxastic presumption”). 

This led me back to foundationalism as the obvious alternative. Here I managed to convince myself that Sellars’s view that there could not be a genuine justificatory relation between non-conceptual experience and conceptual belief was wrong. (I believed and still do that my conceptual belief that there is a triangular patch of bright green in the middle of my visual experience can be justified by the non-conceptual experience of such a patch.) But what seemed to me increasingly implausible was that fine-grained beliefs about experience of this sort in fact play any role in the justification of our ordinary beliefs about the physical world (even if in principle they might have).

So where does this leave us? Coherentism can’t be made to work and foundationalism, even if viable in principle, can’t plausibly account for the justification of our beliefs about the world as we actually hold them. Externalism seems to me entirely unsatisfactory, since it would mean that we in fact have no reasons that we have access to for thinking that our beliefs are true. Perhaps the whole of the standard conception of epistemology is misconceived in some deeper way and should just be given up—but I cannot bring myself to believe that. (If you asked me which way to turn, I would suggest having a look at Bill Talbott’s new book. Talbott’s view doesn’t satisfy all of my epistemological goals, but it comes much closer than externalism or any other view that I know of.)

BG:    In your 1997 Synthese exchange with British philosopher Susan Haack, you write that we must abandon the view, common to Sellars, Davidson, and Rorty, that non-propositional experience cannot have a justificatory relation to our beliefs, since this is both wrong and leads to skepticism. In light of this, what is your view on attempts by, e.g., John McDowell in his 1994 Locke Lectures Mind and World, to solve this problem by positing that experience has only propositional or conceptual content? Your foundationalist writings suggest an answer, but this is an interesting view to have your comments on.

LB: Again, I am handicapped by not having a copy of McDowell available. Here I will just say that McDowell’s view of experiential content seems to me simply and obviously false (even though I may have been briefly tempted by it once or twice in the past). He should just go for a walk in the woods (or really anywhere) and ask how much of his rich experience of colors and shapes and sounds can be conceptually captured by the concepts he has at his disposal. I also think, for reasons at least suggested by my previous comments, that such a view could offer no plausible account of how the propositional beliefs involved in his experience are justified.

BG:    You wrote in a paper that became a chapter in the book Resurrecting Old-Fashioned Foundationalism that you had two main reasons for rejecting coherentism as a possible alternative to classical foundationalism: (1) the lack of a non-skeptical alternative to foundationalism, and (2) that you think there is a way to hold a foundationalist view while avoiding the classic and most important objections. One common objection to coherentism is that it fails to make our beliefs (or, better: our belief systems) responsible to experience. Your coherentism tried to account for this by means of the Observation Requirement, which said that a justified system of empirical belief needs to give high reliability to a “reasonable variety” of what you called “cognitively spontaneous” beliefs. The Observation Requirement is intended to guarantee empirical input to the belief system. This aspect of your view came under criticism from Susan Haack in her book Evidence and Inquiry. Reflecting on your past coherentist view now, how do you view this attempt to make a coherentist view acknowledge the input of experience in a system of beliefs? Is it possible (or feasible) for coherentism to acknowledge experiential input in an earnest sense?

LB: I think I have already spoken to this question to some degree, but let me add a little more. I now think that there is no plausible way within a coherentist view to either identify genuinely observational beliefs or account for their justification in a way that gives their “cognitively spontaneous” status any justificatory role. I now think that whatever plausibility my earlier attempt might have seemed to have in fact derived (illicitly) from intuitions about the nature and occurrence of observations that simply don’t fit in (don’t cohere with) a coherentist view.

BG:     The work of Wilfrid Sellars, in particular his (very long) paper Empiricism and the Philosophy of Mind, has motivated much of the criticism and rejection of the epistemic given since its publication. Your book The Structure of Empirical Knowledge contains a section on Sellars (pp. 114-117), in which you discuss specifically his views on the justification of observational beliefs. For Sellars, as you say, observation beliefs are caused by something external to the belief system but are only justified from within the belief system. This is crucial to a coherentist theory because it tries to meet the challenge of guaranteeing genuine experiential input to belief systems. You mention Sellars later in the book (p.142) in connection with your formulation of the Observation Requirement, saying that Sellars’s discussion of token credibility without type credibility, saying that the needed experiential input for a coherentist view must do something along the lines of what Sellars is suggesting. In light of this, how would you describe the influence of Sellars on your earlier coherentism, especially the Observation Requirement, and your philosophical outlook more generally? Rorty of course was strongly influenced by Sellars, and wrote the introduction to the Harvard University Press edition of Empiricism and the Philosophy of Mind. 

LB: As I think I indicate (but probably not strongly enough), my book was in essence just an attempt to develop and flesh out the sort of epistemological view that Sellars’s discussion seemed to me to suggest. Sellars’s own discussion is often rather fragmentary and elusive, and I have never been at all sure how close the view that I came up with was what he had in mind. But his discussion was surely my main inspiration, though I also read and studied a lot of other coherentist and quasi-coherentist views. It is also probably true that my whole philosophical outlook was once largely shaped by Sellars (I studied and pored over all of his writings), though that was much less true later on. (My rationalist view of the a priori was not importantly influenced by Sellars, though there are a few hints to be found in places in his work.)

BG: Finally, I would like to ask about your graduate education. You studied at Princeton in the 1960s with Richard Rorty, the author of (among many other books and papers) the famous book Philosophy and the Mirror of Nature (1979). One of the most important points of the book is that we should reject epistemology as it has been understood since Descartes. Could you describe the influence that Rorty had on your philosophical outlook, in particular your views in epistemology? Would you say that Rorty played a significant role in shaping how you view the role, purpose, and scope of philosophy generally, in addition to the nature of epistemology?

LB: Rorty’s main influence by far was just that he originally introduced me to Sellars (and, to some extent, encouraged my further Sellarsian investigations). I learned a good deal from Rorty, though more about a range of other things than about Sellars specifically. I am also afraid that I was never able to take Rorty’s increasingly anti-epistemological and more generally anti-philosophical views at all seriously, even for a moment. The further he went in that direction, the further apart we became—intellectually and, I am afraid, also personally.