Quine taught at Harvard until 1942. During World War II, his skill in
languages and his gift for logic were put to use translating and analyzing decoded
messages from German submarines. In 1946, Quine returned to Harvard and
teaching. Except for sabbaticals and visiting professorships (and traveling in
more than 100 countries), Quine remained a professor at Harvard until his retire-
ment in 1978. He spent the next twenty-two years writing and travelling until his
death in 2000.
In his autobiography, Quine explains that he has always despised conceptual con-
straints and has “been at pains to blur the boundaries between natural science,
mathematics, and philosophy.” Nowhere is this impatience with divisions more
apparent than in Quine’s critique of the analytic-synthetic distinction.
Ever since Hume first distinguished between “relations of ideas” and “matters of
fact,” philosophers have divided all propositions into two categories: analytic and
synthetic. Yet in his famous paper “Two Dogmas of Empiricism,” reprinted here
(complete), Quine argues on the basis of various tests of meaning, synonymity, def-
inition, and semantics that such a division has never been clearly made and that
there is no compelling reason for believing such a separation can be made.
For example, the foundational propositions of logic and mathematics have been
held to be true by convention independent of any matters of fact. Thus, the propo-
sition (1) “No unmarried man is married” is true regardless of the various interpre-
tations of “man” and “married.” But there are also semantic propositions, such as
(2) “No bachelor is married,” which are claimed to be equally analytic. As Quine
points out, it is common to believe that this latter (supposed) analytic proposition
can be made into a truth of logic by “putting synonyms for synonyms; thus (2) can
be turned into (1) by putting ‘unmarried man’ for its synonym ‘bachelor’.” But
what is our criterion of synonymity and how is it any more clear than our criterion
of “analyticity”? We could say that propositions (1) and (2) are synonymous if the
term “bachelor” in (2) is definedas “unmarried man.” But what is the basis for
such a definition? Does analyticity depend on the empirical observations of a lexi-
cographer or the authority of a dictionary? We could say that these propositions are
synonymous if proposition (2) is “necessarily true.” However, this would be beg-
ging the question; to say a proposition is “necessary” is to say it is analytic. Quine
concludes that analyticity as a clearly distinguished notion is “an unempirical
dogma of empiricists, a metaphysical article of faith.”
Having critiqued the analytic-synthetic distinction, Quine next turns his atten-
tion to another empirical dogma: the verification theory of meaning and the
reductionism that it presupposes. According to this theory of meaning, the mean-
ing of any statement is the “method of empirically confirming or inferring it.” But
such a theory implies that each meaningful statement can be reduced to an equiv-
alent statement that contains only references to immediate experience. Logical
positivists such as A.J. Ayer sought to develop such a reductionistic language. But
according to Quine, Ayer and the others are relying on the nonexistent analytic-
synthetic distinction to separate the “linguistic components” from the “factual
components” in any individual statement. Furthermore,nostatement, taken by
itself, could be confirmed or discredited by sensory experience because there
could always be a further experience that would overturn the original statement.
1190 WILLARDVANORMANQUINE
