1. The descriptive core of the theories of Standard Analytic Epistemology
К оглавлению1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 1617 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33
34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50
51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67
68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84
85 86 87 88 89 90 91 92 93 94
As we noted in chapter 1, proponents of Standard Analytic Epistemology
aim to provide an account of knowledge and epistemic justification. One
of the main success conditions on such an account is the stasis requirement
: The correct account of knowledge or justification will ‘‘leave our
epistemic situation largely unchanged. That is to say, it is expected to turn
out that according to the criteria of justified belief we come to accept, we
know, or are justified in believing, pretty much what we reflectively think
we know or are entitled to believe’’ (Kim 1988, 382). If an account of
justification must satisfy the stasis condition in order to be successful, then
such an account can be successful only if (a) for every belief B, clearly in
the extension of the predicate ‘is justified’ as used by proponents of SAE,
the account yields the result that B is justified, and (b) for every belief N,
clearly not in the extension of the predicate ‘is justified’ as used by proponents
of SAE, the account yields the result that N is not justified. (A
similar condition can be defined for a successful account of knowledge.)
The commitment to epistemic stasis is embodied in the method of
SAE. Philosophers accept or reject an epistemological theory on the basis
of whether it accords with their considered judgments. Gettier’s (1963)
paper is a classic because it describes clear and compelling examples in
which the justified true belief (JTB) account of knowledge is at odds with
our considered judgments about knowledge. Our central worry about SAE
arises from the fact that its epistemic theories are so often rejected solely on
the grounds that they violate our considered epistemic judgments. Why
should we place so much trust in our well-considered judgments? We need
some reason for thinking that our well-considered epistemic judgments
are correct. The Aristotelian Principle provides us with part (but only
The Troubles with Standard Analytic Epistemology 105
part) of the story about how we can test the deliverances of those judgments
under the heat of careful experimentation rather than in the relaxation
of the philosophical salon. Perhaps our armchair judgments will
survive those fires. But before offering the world recommendations about
how we should reason or what we should believe about important matters,
it seems prudent to check.
Suppose God gave us the theory of justification that best satisfied the
stasis requirement. What sort of information would such a theory give us?
What would that theory be about? We can approach this as a problem of
reverse engineering: Given how SAE works, what is its likely output? Some
might note close analogies between epistemology and science. In SAE and
science, articles are written primarily for and by people who have received
distinctive educations and who have a highly specialized set of skills. While
there is not much empirical work on the subject, it seems plausible to
suppose that this education significantly affects the concepts, categories,
and inferential patterns one uses in thinking about the world. So far, so
good. In the natural sciences, however, hypotheses are typically tested
against the world. But in SAE, hypotheses are tested against the wellconsidered
judgments of other (similarly trained) philosophers.
Given how SAE works, it seems doubtful that it is geared to informing
us about how regular folk think about justification. One needn’t be a
sociologist to realize that philosophers as a group are a relatively small and
idiosyncratic sample of folks (Goldman 1999a). Philosophers’ median
education and intelligence are surely well above average. We speculate that
philosophers’ median scores on various MMPI scales (e.g., social alienation,
hypersensitivity, social introversion) might be above average as well.
Of course, proponents of SAE might view this as being all to the good.
They might argue: ‘‘We don’t want to offer a description of the epistemic
practices of common folk. We’re examining the concepts of experts. If you
want to know what the right concept of bird is, ask an ornithologist. If you
want to know what the right concept of justification is, ask an epistemologist.’’
If this is the sort of move a proponent of SAE might make, we
need to ask: What exactly is an epistemologist an expert about?
If we take the Aristotelian Principle to heart, if we believe that good
reasoning tends to lead to better outcomes than bad reasoning, then we
might wonder whether SAE tells us about how a wide variety of people
who lead flourishing lives—people in a wide range of stations, in different
cultures, in different times—have reasoned about important matters. But
this is a deeply empirical matter, and one that the conservative method of
SAE does not seem especially well designed to illuminate. If, however, epistemologists themselves tend to lead particularly successful lives, then
perhaps providing people with their epistemic autobiographies would be
useful. It is not obvious, however, that when socioeconomic factors are
controlled for, epistemologists as a group lead more or less meaningful or
flourishing lives than other folk.
So what is SAE geared to tell us about? We suggest that it tells us
about the reflective epistemic judgments of a group of idiosyncratic people
who have been trained to use highly specialized epistemic concepts and
patterns of thought. By ‘highly specialized’ we mean that people who have
not received the relevant training would find at least some of those concepts
and patterns of thought strange, foreign, or unfamiliar. The conservative
goals and methods of SAE are suited to the task of providing
an account of the considered epistemic judgments of (mostly) well-off
Westerners with Ph.D.’s in Philosophy. This is a thoroughly descriptive
endeavor. Such an account aims to describe the clear application conditions
of an expression as it is used by a particular group of people. It is an
open question whether SAE also provides knowledge of normative matters.
But this possibility should not hide the heretofore unrecognized
naturalistic essence of SAE.
A particularly dramatic way to see that the core of SAE is a descriptive
theory of analytic epistemologists’ own epistemic judgments is to consider
how SAE might be different if it were conducted by a very different group
of people. In a fascinating study, Weinberg, Nichols, and Stich (2001)
found that people in different cultural and socioeconomic groups make
significantly different epistemic judgments. A group of Western subjects
and non-Western subjects were given the following Gettier-style example:
Bob has a friend, Jill, who has driven a Buick for many years. Bob therefore
thinks that Jill drives an American car. He is not aware, however, that her
Buick has recently been stolen, and he is also not aware that Jill has replaced
it with a Pontiac, which is a different kind of American car. Does Bob really
know that Jill drives an American car, or does he only believe it?
REALLY KNOWS ONLY BELIEVES
A large majority of Western subjects gave the answer sanctioned by SAE
(‘‘only believes’’), but a majority of East Asians and a majority of subjects
from India gave the opposite answer (‘‘really knows’’) (2001, 443).
Weinberg et al. considered an anti-reliabilist type of example in which
a reasoner, as a result of being hit on the head with a rock, unwittingly
acquires a reliable belief-forming mechanism for determining ambient
The Troubles with Standard Analytic Epistemology 107
temperature. They found significant differences in the judgments of
Western and East Asian subjects. A majority of both groups of subjects,
however, thought that the reasoner did not have knowledge of the ambient
temperature (2001, 439–440).
Another fascinating finding involved giving high socioeconomic status
(SES) subjects and low SES subjects the following case:
It’s clear that smoking cigarettes increases the likelihood of getting cancer.
However, there is now a great deal of evidence that just using nicotine by
itself without smoking (for instance, by taking a nicotine pill) does not
increase the likelihood of getting cancer. Jim knows about this evidence and
as a result, he believes that using nicotine does not increase the likelihood of
getting cancer. It is possible that the tobacco companies dishonestly made
up and publicized this evidence that using nicotine does not increase the
likelihood of cancer, and that the evidence is really false and misleading.
Now, the tobacco companies did not actually make up this evidence, but
Jim is not aware of this fact. Does Jim really know that using nicotine
doesn’t increase the likelihood of getting cancer, or does he only believe it?
REALLY KNOWS ONLY BELIEVES
There were statistically significant differences between high and low SES
subjects. Low SES subjects were evenly divided on whether Jim really
knows or only believes that using nicotine doesn’t increase the likelihood
of getting cancer. High SES subjects were much more likely to say that Jim
only believes it (82%) (2001, 447–48).
The possibility that there is considerable variation (not only across
cultures but also within cultures) in people’s epistemic judgments makes
it plausible to believe that we learn about the epistemic judgments of
an idiosyncratic group of people when we do SAE. This is as descriptive a
fact as there could possibly be. Indeed, it suggests that SAE is actually an
odd kind of cultural anthropology: building theories that describe how
privileged (mostly) Westerners with Ph.D.s in Philosophy engage in epistemic
assessment. Weinberg, Nichols, and Stich call this endeavor ‘‘ethnoepistemology’’
(454). If SAE is but anthropology, it is unclear on what
grounds its proponents can reasonably make universal normative claims
about the nature, origin, and limits of human knowledge. To make universal
claims—to claim SAE is more lofty than anthropology—has the
uncomfortable feel of brute cultural imperialism.
Now it may turn out that there is less diversity in people’s epistemic
judgments than the fascinating studies by Weinberg, Nichols, and Stich
108 Epistemology and the Psychology of Human Judgment
suggest. But even if the diversity findings collapse, this, by itself, won’t
help show that SAE is normative. Even if SAE describes a universal practice,
rather than a culturally situated one, that doesn’t make SAE normative.
The diversity findings simply allow us to put our point dramatically:
SAE appears to describe the idiosyncratic epistemic practices of a particular
group of people. But even if the world came to be populated only by
analytic epistemologists, the central endeavor of SAE would still be essentially
descriptive.
As a descriptive attempt to capture the epistemic judgments of philosophers,
we have powerful reasons to think that the methods of psychology
are superior to those of SAE. Insofar as the core of the theories
of SAE is descriptive, they are very likely to be bad descriptive theories.
This point is not essential to our argument. But it is worth noting that
psychologists develop models of our concepts all the time. These models
mimic our categorization judgments. (These models can mimic concepts
with ‘‘fuzzy boundaries’’ and indeterminate instances.) If philosophers
want an account that mimics their epistemological judgments, all we would
need is a psychologist who is willing to model our judgments (e.g., Smith
and Medin 1981, Keil 1989). Indeed, if philosophers really want to begin
their epistemological musings with a descriptive core that accurately accounts
for their judgments about knowledge or justification, they could
save a lot of time, energy, and expense by employing a few psychology
graduate students.
As we noted in chapter 1, proponents of Standard Analytic Epistemology
aim to provide an account of knowledge and epistemic justification. One
of the main success conditions on such an account is the stasis requirement
: The correct account of knowledge or justification will ‘‘leave our
epistemic situation largely unchanged. That is to say, it is expected to turn
out that according to the criteria of justified belief we come to accept, we
know, or are justified in believing, pretty much what we reflectively think
we know or are entitled to believe’’ (Kim 1988, 382). If an account of
justification must satisfy the stasis condition in order to be successful, then
such an account can be successful only if (a) for every belief B, clearly in
the extension of the predicate ‘is justified’ as used by proponents of SAE,
the account yields the result that B is justified, and (b) for every belief N,
clearly not in the extension of the predicate ‘is justified’ as used by proponents
of SAE, the account yields the result that N is not justified. (A
similar condition can be defined for a successful account of knowledge.)
The commitment to epistemic stasis is embodied in the method of
SAE. Philosophers accept or reject an epistemological theory on the basis
of whether it accords with their considered judgments. Gettier’s (1963)
paper is a classic because it describes clear and compelling examples in
which the justified true belief (JTB) account of knowledge is at odds with
our considered judgments about knowledge. Our central worry about SAE
arises from the fact that its epistemic theories are so often rejected solely on
the grounds that they violate our considered epistemic judgments. Why
should we place so much trust in our well-considered judgments? We need
some reason for thinking that our well-considered epistemic judgments
are correct. The Aristotelian Principle provides us with part (but only
The Troubles with Standard Analytic Epistemology 105
part) of the story about how we can test the deliverances of those judgments
under the heat of careful experimentation rather than in the relaxation
of the philosophical salon. Perhaps our armchair judgments will
survive those fires. But before offering the world recommendations about
how we should reason or what we should believe about important matters,
it seems prudent to check.
Suppose God gave us the theory of justification that best satisfied the
stasis requirement. What sort of information would such a theory give us?
What would that theory be about? We can approach this as a problem of
reverse engineering: Given how SAE works, what is its likely output? Some
might note close analogies between epistemology and science. In SAE and
science, articles are written primarily for and by people who have received
distinctive educations and who have a highly specialized set of skills. While
there is not much empirical work on the subject, it seems plausible to
suppose that this education significantly affects the concepts, categories,
and inferential patterns one uses in thinking about the world. So far, so
good. In the natural sciences, however, hypotheses are typically tested
against the world. But in SAE, hypotheses are tested against the wellconsidered
judgments of other (similarly trained) philosophers.
Given how SAE works, it seems doubtful that it is geared to informing
us about how regular folk think about justification. One needn’t be a
sociologist to realize that philosophers as a group are a relatively small and
idiosyncratic sample of folks (Goldman 1999a). Philosophers’ median
education and intelligence are surely well above average. We speculate that
philosophers’ median scores on various MMPI scales (e.g., social alienation,
hypersensitivity, social introversion) might be above average as well.
Of course, proponents of SAE might view this as being all to the good.
They might argue: ‘‘We don’t want to offer a description of the epistemic
practices of common folk. We’re examining the concepts of experts. If you
want to know what the right concept of bird is, ask an ornithologist. If you
want to know what the right concept of justification is, ask an epistemologist.’’
If this is the sort of move a proponent of SAE might make, we
need to ask: What exactly is an epistemologist an expert about?
If we take the Aristotelian Principle to heart, if we believe that good
reasoning tends to lead to better outcomes than bad reasoning, then we
might wonder whether SAE tells us about how a wide variety of people
who lead flourishing lives—people in a wide range of stations, in different
cultures, in different times—have reasoned about important matters. But
this is a deeply empirical matter, and one that the conservative method of
SAE does not seem especially well designed to illuminate. If, however, epistemologists themselves tend to lead particularly successful lives, then
perhaps providing people with their epistemic autobiographies would be
useful. It is not obvious, however, that when socioeconomic factors are
controlled for, epistemologists as a group lead more or less meaningful or
flourishing lives than other folk.
So what is SAE geared to tell us about? We suggest that it tells us
about the reflective epistemic judgments of a group of idiosyncratic people
who have been trained to use highly specialized epistemic concepts and
patterns of thought. By ‘highly specialized’ we mean that people who have
not received the relevant training would find at least some of those concepts
and patterns of thought strange, foreign, or unfamiliar. The conservative
goals and methods of SAE are suited to the task of providing
an account of the considered epistemic judgments of (mostly) well-off
Westerners with Ph.D.’s in Philosophy. This is a thoroughly descriptive
endeavor. Such an account aims to describe the clear application conditions
of an expression as it is used by a particular group of people. It is an
open question whether SAE also provides knowledge of normative matters.
But this possibility should not hide the heretofore unrecognized
naturalistic essence of SAE.
A particularly dramatic way to see that the core of SAE is a descriptive
theory of analytic epistemologists’ own epistemic judgments is to consider
how SAE might be different if it were conducted by a very different group
of people. In a fascinating study, Weinberg, Nichols, and Stich (2001)
found that people in different cultural and socioeconomic groups make
significantly different epistemic judgments. A group of Western subjects
and non-Western subjects were given the following Gettier-style example:
Bob has a friend, Jill, who has driven a Buick for many years. Bob therefore
thinks that Jill drives an American car. He is not aware, however, that her
Buick has recently been stolen, and he is also not aware that Jill has replaced
it with a Pontiac, which is a different kind of American car. Does Bob really
know that Jill drives an American car, or does he only believe it?
REALLY KNOWS ONLY BELIEVES
A large majority of Western subjects gave the answer sanctioned by SAE
(‘‘only believes’’), but a majority of East Asians and a majority of subjects
from India gave the opposite answer (‘‘really knows’’) (2001, 443).
Weinberg et al. considered an anti-reliabilist type of example in which
a reasoner, as a result of being hit on the head with a rock, unwittingly
acquires a reliable belief-forming mechanism for determining ambient
The Troubles with Standard Analytic Epistemology 107
temperature. They found significant differences in the judgments of
Western and East Asian subjects. A majority of both groups of subjects,
however, thought that the reasoner did not have knowledge of the ambient
temperature (2001, 439–440).
Another fascinating finding involved giving high socioeconomic status
(SES) subjects and low SES subjects the following case:
It’s clear that smoking cigarettes increases the likelihood of getting cancer.
However, there is now a great deal of evidence that just using nicotine by
itself without smoking (for instance, by taking a nicotine pill) does not
increase the likelihood of getting cancer. Jim knows about this evidence and
as a result, he believes that using nicotine does not increase the likelihood of
getting cancer. It is possible that the tobacco companies dishonestly made
up and publicized this evidence that using nicotine does not increase the
likelihood of cancer, and that the evidence is really false and misleading.
Now, the tobacco companies did not actually make up this evidence, but
Jim is not aware of this fact. Does Jim really know that using nicotine
doesn’t increase the likelihood of getting cancer, or does he only believe it?
REALLY KNOWS ONLY BELIEVES
There were statistically significant differences between high and low SES
subjects. Low SES subjects were evenly divided on whether Jim really
knows or only believes that using nicotine doesn’t increase the likelihood
of getting cancer. High SES subjects were much more likely to say that Jim
only believes it (82%) (2001, 447–48).
The possibility that there is considerable variation (not only across
cultures but also within cultures) in people’s epistemic judgments makes
it plausible to believe that we learn about the epistemic judgments of
an idiosyncratic group of people when we do SAE. This is as descriptive a
fact as there could possibly be. Indeed, it suggests that SAE is actually an
odd kind of cultural anthropology: building theories that describe how
privileged (mostly) Westerners with Ph.D.s in Philosophy engage in epistemic
assessment. Weinberg, Nichols, and Stich call this endeavor ‘‘ethnoepistemology’’
(454). If SAE is but anthropology, it is unclear on what
grounds its proponents can reasonably make universal normative claims
about the nature, origin, and limits of human knowledge. To make universal
claims—to claim SAE is more lofty than anthropology—has the
uncomfortable feel of brute cultural imperialism.
Now it may turn out that there is less diversity in people’s epistemic
judgments than the fascinating studies by Weinberg, Nichols, and Stich
108 Epistemology and the Psychology of Human Judgment
suggest. But even if the diversity findings collapse, this, by itself, won’t
help show that SAE is normative. Even if SAE describes a universal practice,
rather than a culturally situated one, that doesn’t make SAE normative.
The diversity findings simply allow us to put our point dramatically:
SAE appears to describe the idiosyncratic epistemic practices of a particular
group of people. But even if the world came to be populated only by
analytic epistemologists, the central endeavor of SAE would still be essentially
descriptive.
As a descriptive attempt to capture the epistemic judgments of philosophers,
we have powerful reasons to think that the methods of psychology
are superior to those of SAE. Insofar as the core of the theories
of SAE is descriptive, they are very likely to be bad descriptive theories.
This point is not essential to our argument. But it is worth noting that
psychologists develop models of our concepts all the time. These models
mimic our categorization judgments. (These models can mimic concepts
with ‘‘fuzzy boundaries’’ and indeterminate instances.) If philosophers
want an account that mimics their epistemological judgments, all we would
need is a psychologist who is willing to model our judgments (e.g., Smith
and Medin 1981, Keil 1989). Indeed, if philosophers really want to begin
their epistemological musings with a descriptive core that accurately accounts
for their judgments about knowledge or justification, they could
save a lot of time, energy, and expense by employing a few psychology
graduate students.