W. V. O. Quine: “Epistemology Naturalized”
Epistemology is concerned with the foundations of science [and thus] includes the study of the foundations of mathematics as one of its departments. [225.1]
For Quine, epistemology has been traditionally conceived as
a foundational enterprise,
attempting to find refutations for the skeptic and reasons to believe our most
powerful theories of the world (that is, for Quine, just as for the Logical
Positivists of the
1. conceptual studies (meaning), concerned with clarifying concepts by defining them
2. doctrinal studies (truth), concerned with establishing laws by proving them, some on the basis of others.
These two goals are related in that we clarify mathematics by parsing it into logic, and “the clearer
these terms are, the likelier it is that the truths couched in them will be
obviously true, or derivable from obvious
truths”. BUT:
This particular outcome is in fact denied us, however, since
mathematics reduces only to set theory and not to logic proper. Such reduction still enhances clarity, but
only because of the interrelations that emerge and not because of the end terms
of the analysis are clearer than others.
As for the end truths, the axioms of set theory, these have less
obviousness and certainty to recommend them than do most of the mathematical
theorems that we would derive from them. [225.2]
A parallel project is to “reduce” science to talk of sense experience (in line with the empiricism of Hume.
Quine thinks that we are no further along in doctrinal studies than we were in Hume’s day, BUT that some progress has been made in conceptual studies with two “resorts”:
1. the resort to contextual definition
2. the resort to set theory
The idea of contextual definition. This is the recognition that contra logical atomists like the earlier Wittgenstein, the unit of meaning is not the word, but the sentence. That is, the meaning of particular words (or terms) is dependent on their context in sentences, and they do not have any meaning outside of contexts. (An application of this notion is Russell’s Theory of Descriptions: for Russell, definite descriptions do not have meaning/reference by themselves, but are “incomplete symbols” dependent on their context for meaning. This is why there doesn’t have to be a King of France for “The King of France is bald” to be meaningful.) The idea of contextual definition allows us to more freedom in translation: there doesn’t have to be a one-to-one (or many) correspondence of terms of science to terms of sense experience:
One could undertake to explain talk of bodies in terms of talk of impressions by translating one’s whole sentences about bodies into whole sentences about impressions, without equating the bodies themselves to anything at all. [226.2]
It is possible to translate mathematics into logic plus set theory, so that “resort” (to a richer, albeit still extensional, ontology, allows conceptual progress. But what about doctrinal progress?
But the mere fact that a sentence is couched in terms of observation, logic, and set theory does not mean that it can be proved from observation sentences by logic and set theory. [227.1]
Because: generalizations in science translate into generalizations about experience, and generalizations are infinite, and therefore unprovable by any amount of experience.
Thus, the empiricist project of giving a foundation to science in sense experience fails. However,
Two cardinal tenets of empiricism remained unassailable… One is that whatever evidence there is for science is sensory evidence. The
other…is that all inculcation of meanings of words must rest ultimately on
sensory evidence. [227.2]
Logical positivists defended both:
1. scientism: the view that scientific knowledge serves as a paradigm for all knowledge
2. empiricism: the view that all meaning and knowledge has its origin in experience
This latter is part of his rejection of intensional entities like “meanings” or “propositions”.
Quine agreed, and applied these doctrines to a critique of
“first philosophy.” Recall that Quine
rejects a distinction between analytic and synthetic statements, arguing that
there are none of the former because all sentences are
in principle revisable. This is part of
his holism: the view that the unit of
meaning is not individual words, or even sentences, but language as a
whole. The fact that there are no unrevisable statements means that we cannot cordon off a
section of certainties from which we can assess our other beliefs. (In Descartes’ terms, there is no cogito to use as an indubitable foundation for our beliefs.) Such certainties would comprise a “first
philosophy” that we could use to assess disciplines like science. Instead, Quine insists that we cannot stand
outside of science and judge its effectiveness – science is all we have. We operate from within a scientific view of
the world, and any doubts about science come from within science itself. (This
is why Quine is so fond of Neurath’s analogy of being
on board a ship in the middle of the sea [231.1], forced to rebuild the ship
using the materials at hand. There is no
“dock” of first philosophy to go into to fix the problem, and we have to keep
most of the boat intact – that is, not question most assumptions of science –
while we attempt to repair the broken parts.)
This applies to all areas of philosophy, including
epistemology. Thus the canons of science
cannot be questioned except from within
science. This is “epistemology
naturalized,” that is, brought under the heading of natural science.
The “foundational” project of epistemology is to translate
statements of whatever it is we have doubts about to more certain
statements. But if translation is
impossible (because of indeterminacy)
then this cannot be achieved, so why do epistemology?
What is epistemology, on this view? It is part of (scientific) psychology: we study the relation between the “meager input and the torrential output” that physical human subjects exhibit. This can be either the study of how individuals move from an understanding of simple observation sentences to the understanding of more complex sentences, or the study of the relation between physical input (retinal disturbances, etc.) and beliefs (as objects of neuropsychology rather than “mental” states).
What about the skeptic? One skeptical argument is that it is logically possible at any time that one’s present belief be false. Quine won’t allow this because of the analytical notion of “logical possibility”.
Alternatively, the skeptic could point to the gap between appearance and reality and charge that the world could be different from the way we perceive it. Quine’s response is to say that the world is what we perceive: the only criterion of reality is that which science provides.
Is the skeptic refuted?
Not necessarily. If the skeptic
uses the canons of science to undermine science, that would be successful
skepticism (and a reduction ad absurdum
of science)
Observation sentences: occasion sentences which all observers assent to in the same circumstances [232.1]. Also:
A sentence is an observation sentence if all verdicts on it
depend on present sensory stimulation and on no stored information beyond what
goes into understanding the sentence. [231.2]