§87. Dogmas of Empiricism

The twentieth century was impatient to discredit experience. Paul de Man denounced its appeal as naive about the constitutive function of language. “Instead of containing or reflecting experience, language constitutes it.” The specious “effect of reality” exuded by the rhetoric of experience “is associ­ated not with an immersion in reality but with an excess of the symbolic that has been patiently invented and articulated.” For historian Joan Wallach Scott, “We need to attend to the historical process that, through discourse, positions subjects and produces their experience.” Experience is nothing original or foundational, no more than a language game we play with others, and as Richard Rorty says, “What counts as an accurate report of experience is a matter of what a community will let you get away with.”¹

Experience is denounced as ideology.

At mid-century, W V. Quine refuted two so-called dogmas of empiri­cism. He means logical empiricism, Carnap, not Epicurus or Condillac. One of these dogmas is that propositions divide into those true by the meaning of their terms—so-called analytic truths, e.g., “A mare is a horse”; and syn­thetic, empirical propositions, true because of how the world is, e.g., “Ice

Empiricisms.

DOI: 10.1093/oso/9780197508930.001.0001. floats.” Quine finds no satisfactory explanation of the difference and declares the distinction untenable, it being impossible to distinguish between change in the meaning of words and the discovery of new facts. The second dogma holds that any meaningful statement can be translated into terms referring solely to the sensory given. That was implied by Russell's principle of ac­quaintance, and is the gist of Carnap's quip that “gravity” is an abbreviation. This too Quine finds insupportable. The organizing role supposedly fulfilled by analytic sentences is shared by sentences generally, and empirical content, supposedly the preserve of the synthetic and sensory, diffuses throughout the system. Two dogmas down and empiricism is reeling, one empiricism anyway.³

To Quine's dogmatic duo Donald Davidson adds dogma number three, which is to assume that experience endows concepts, beliefs, or sentences with semantic or intentional content. Davidson's dogma dogs Quine's doc­trine too. Quine “dogmatically” assumes a relation between the whole body of sentences constituting scientific theory and something outside, something perceived and experienced, something altogether different from sentences and constraining their truth-value. Science has a touchstone in something natural and physical, because fitting the totality of experience “is what makes scientific method partly empirical rather than solely a quest for internal co­herence.” Quine unapologetically adheres to theorematic empiricism's oldest theme, that sensation is ultimate evidence. “Science itself tells us that our in­formation about the world is limited to the irritations of our surfaces.” The senses are gates, the only way in.⁴

Quine denounced the dogmas of an empiricism he wants to reconstruct more consistently, as we see when he balances dogmas refuted with princi­ples affirmed, one being that “whatever evidence there is for science is sen­sory evidence,” the other that “all inculcation of meanings of words must rest ultimately on sensory evidence.” These principles ensure that emperia is ulti­mate evidence, but for Davidson that is Quine's mistake.

Davidson and James are antipodes in philosophy. James eliminates nom­inalism for a more consistent empiricism, and Davidson eliminates empir­icism for a more consistent nominalism. He found unexpected wealth in the austere technicalities of Alfred Tarski's semantic theory of truth. Before Tarski it seemed inevitable to think of truth in terms of relations extending beyond language, relating language to something else. Davidson showed that with Tarski truth becomes an artifact of the internal, purely formal logic of language. The only way to explain what it is for a sentence to be true is to show how it relates to an infinite set of other sentences. The account never strays from the relation of signs to signs. The result is a concept of truth as rigorous as a logician might ask for, yet supports no onto-logical interpretation—truth without a truth-maker. “Nothing” Davidson says, “no thing, makes sentences and theories true: not experience, not surface irritations, not the world, can make a sentence true.... The sentence ‘My skin is warm' is true if and only if my skin is warm. Here there is no reference to a fact, a world, an experience, or a piece of evidence.”⁶

The argument consummates the nominalist project of Abelard and Ockham, eliminating the last vestige of the idea that the significant use of language depends on something that is natural, unconventional, more than just more signs.