The Inferences That Never Were

The occasion for this paper was the publication of Richard J. Bernstein’s The Pragmatic Turn and a conference the New York Pragmatist Forum held on his book.

The main part of this essay is an account of Peirce’s theory of perception and how, on his view, perceptions justify beliefs.  I argue that Peirce’s theory—understood, of course, in the special way that I understand it—diagnoses and cures the problems with both (i) the Sellarsian view that there are no non-inferential beliefs and the notion of a given is a myth and (ii) the Davidsonian view that perception only makes a causal, and not a rational, contribution to knowledge.  The cure, I contend, lies in Peirce’s semiotics and the dual function of perceptual judgments.

On the one hand, perceptual judgments are interpretations (or “interpretants”) of our percepts.  This is the first function of perceptual judgments.  As interpretations of our percepts, they are indeed merely caused by those percepts, much as a barometer’s reading is an interpretation of the current atmospheric pressure.

On the other hand, the very same perceptual judgments are also the first (or foundational, if you will) premises of our reasonsings.  Peirce is, after all, an empiricist.  This is the second function of our perceptual judgments: to serve as the first premises of our reasonings.  To continue with the barometer analogy, a barometer reading is an interpretation of the current atmospheric pressure.  However, in conjunction with previous readings of the barometer, one can infer whether it is going to rain (viz. if the atmospheric pressure is falling, rain is likely).  In such a case, the barometer reading becomes a premise in an inference.  The conclusion of the inference is itself an interpretation of those prior barometer readings.

To return to the Sellarsian and Davidsonian views, the Sellarsian view is incorrect, for there are non-inferential beliefs.  On Peirce’s view, inference is a self-controlled process of reasoning.  However, the process by which we form perceptual judgments qua interpretations is not self-controlled.  Hence, perceptual judgments are non-inferential.  Nevertheless, those perceptions are the result of a (very complex) semiotic process, viz. the process whereby a percept is uncontrollably interpreted in a perceptual judgment.  Also, those perceptual judgments are beliefs.  Such non-inferential beliefs, I contend, are givens.  In short, the Sellarsian error lies in a failure to appreciate that there are non-self-controlled semiotic processes that yield beliefs even though those beliefs are not assigned a role in a self-controlled semiotic process (i.e. an inference).  (I should note: Peirce once thought those non-self-controlled semiotic processes were inferences but later revised his position.  Hence, these are the “inferences that never were” really inferences.)

Also, the Davidsonian view is incorrect.  Even though percepts as interpreted in perceptual judgments make no rational contribution to knowledge, the very same perceptual judgment has a second function, viz. to serve as a first premise in our reasonings.  In fulfilling that function, our perceptual judgments do make a rational contribution to knowledge.  In short, the Davidsonian error lies in a failure to appreciate that the very same perceptual judgment can have a dual-function both as an interpretant of a percept and as a first premise in our reasonings.  In other words, there is no need to take refuge in coherentism.

Also in this paper are: (a) a discussion of Peirce’s Some Consequences of Four Incapacities in relation to Cartesianism; (b) some clarifications of claims made in Bernstein’s book; (c) comments on how my view of Peirce’s theory of perception differs from Sandra Rosenthal’s view; and (d) brief comments on McDowell’s Mind and World in relation to the issues raised in the paper.