2. Standard Analytic Epistemology: Throwing stones in glass houses

The descriptive core of SAE is a theory that captures the considered epistemic

judgments of philosophers. Some proponents of SAE believe that

those judgments are best captured by a coherentist theory, others believe

that they are best captured by a foundationalist theory, others believe that

they are best captured by a reliabilist theory, etc. The obvious challenge

for these theories is: How are they to extract normative consequences from

a descriptive theory? While we intend to argue that the prospects for the

theories of SAE overcoming the naturalist challenge are not good, we do

not contend that the theories of SAE cannot overcome it. We recognize

that there may be a way we have not properly identified for proponents of

SAE to respond successfully to the naturalist challenge. However, proponents

of SAE have wielded arguments against naturalistic epistemology

The Troubles with Standard Analytic Epistemology 109

that can be readily adapted to show that their own preferred theories

cannot be normative. We emphasize again that we do not endorse these

arguments. Our point here is that proponents of SAE have for too long

been throwing stones at naturalistic epistemology from glass houses.

Michael Williams argues that the normative nature of epistemology

makes it impossible to fully ‘‘naturalize’’ it:

[Epistemic claims] depend on meeting certain norms or standards which

define, not what you do do, but what you must or ought to do. To characterize

someone’s claim as expressing or not expressing knowledge is to pass

judgment on it. Epistemic judgments are thus a particular kind of valuejudgment.

It is far from obvious that investigations with such a strongly

normative component can be fully ‘naturalized.’ (Williams 2001, 11)

Williams’s argument applies equally to the theories of SAE: As we have

just argued, when it comes to epistemic judgments, the theories of SAE

define what we ‘‘do do’’ not what we ‘‘must or ought to do.’’ They don’t tell

us how to reason or believe; they merely tell us how we do make epistemic

judgments (and by ‘‘we,’’ we mean the tiny fraction of the world’s population

who has studied SAE).

According to Richard Feldman, psychology and philosophy must at

best co-exist because psychology can’t ask or answer the relevant normative

questions:

The original epistemological questions seem to be perfectly good questions,

well worthy of our attention. It is difficult to see, then, why the availability

of this other field of study [psychology], concerning how we reason, is a

suitable replacement for the evaluative questions that are at the heart of

epistemology. (Feldman 2003, 168)

Feldman’s point can be made against the theories of SAE: ‘‘It is difficult to

see, then, why the availability of this other field of study’’ (namely, SAE,

which aims to describe how some people make evaluative judgments) ‘‘is a

suitable replacement for the evaluative questions that are at the heart of

epistemology.’’ The proponent of SAE is replacing normative questions

about how to evaluate reason and belief with descriptive questions about

how proponents of SAE evaluate reason and belief. To suppose that

answers to the descriptive questions are also answers to the normative

questions is to take a big leap.

Lawrence BonJour argues that any epistemology subsumed by psychology

does not have the resources to evaluate, positively or negatively, beliefs about alleged occult phenomena of various sorts, such as astrological

or phrenological beliefs. For just as naturalized epistemology can say nothing

positive about the justification of science or common sense, and is thus

impotent in the face of skepticism, so also it can say nothing distinctively

negative about the justification of these less reputable sorts of belief.

(BonJour 2002, 244)

We can once again turn the tables on the proponent of SAE. A theory that

accurately describes how a certain group of people make certain evaluative

judgments ‘‘can say nothing positive about the justification of science or

common sense, and is thus impotent in the face of skepticism, so also it

can say nothing distinctively negative about the justification of ’’ occult,

astrological, or phrenological beliefs. Of course, such a theory might tell us

how some people evaluate those beliefs. But that’s not the same as actually

evaluating those beliefs.

In a similar vein, Alvin Plantinga notes that Quinean naturalism in

epistemology cannot be normative:

[T]he most extreme version of naturalism in epistemology eschews normativity

altogether, seeking to replace traditional epistemology (with its concern

with justification, rationality, reasonability, and their normative colleagues)

by descriptive psychology; this seems to be Quine’s suggestion. [fn deleted]

(Plantinga 1993, 45)

By now, we’re confident our argumentative strategy is wearing thin, but

here it is anyway: SAE ‘‘eschews normativity altogether, seeking to replace

traditional epistemology (with its concern with justification, rationality, reasonability,

and their normative colleagues) by descriptive psychology’’—

a psychology that describes how certain people make certain sorts of

judgments.

Proponents of SAE have argued for decades that radically naturalistic

theories of epistemology cannot succeed because they cannot be normative.

Our aim has been to turn the tables on proponents of SAE. We distance

ourselves from these arguments. Merely pointing out that a theory faces the

problem of bridging the is-ought divide does not by itself damn that theory.

And this is a good thing for proponents of SAE. We all start our normative

musings with psychology. Proponents of SAE start by describing a certain

group’s epistemological judgments, and we start with what we have called

Ameliorative Psychology. (We argue that the former is probably very bad

psychology, but that is not essential to the case we’re building here.) When

it comes to bridging the is-ought gap, everybody has work to do.

The Troubles with Standard Analytic Epistemology 111

The descriptive core of SAE is a theory that captures the considered epistemic

judgments of philosophers. Some proponents of SAE believe that

those judgments are best captured by a coherentist theory, others believe

that they are best captured by a foundationalist theory, others believe that

they are best captured by a reliabilist theory, etc. The obvious challenge

for these theories is: How are they to extract normative consequences from

a descriptive theory? While we intend to argue that the prospects for the

theories of SAE overcoming the naturalist challenge are not good, we do

not contend that the theories of SAE cannot overcome it. We recognize

that there may be a way we have not properly identified for proponents of

SAE to respond successfully to the naturalist challenge. However, proponents

of SAE have wielded arguments against naturalistic epistemology

The Troubles with Standard Analytic Epistemology 109

that can be readily adapted to show that their own preferred theories

cannot be normative. We emphasize again that we do not endorse these

arguments. Our point here is that proponents of SAE have for too long

been throwing stones at naturalistic epistemology from glass houses.

Michael Williams argues that the normative nature of epistemology

makes it impossible to fully ‘‘naturalize’’ it:

[Epistemic claims] depend on meeting certain norms or standards which

define, not what you do do, but what you must or ought to do. To characterize

someone’s claim as expressing or not expressing knowledge is to pass

judgment on it. Epistemic judgments are thus a particular kind of valuejudgment.

It is far from obvious that investigations with such a strongly

normative component can be fully ‘naturalized.’ (Williams 2001, 11)

Williams’s argument applies equally to the theories of SAE: As we have

just argued, when it comes to epistemic judgments, the theories of SAE

define what we ‘‘do do’’ not what we ‘‘must or ought to do.’’ They don’t tell

us how to reason or believe; they merely tell us how we do make epistemic

judgments (and by ‘‘we,’’ we mean the tiny fraction of the world’s population

who has studied SAE).

According to Richard Feldman, psychology and philosophy must at

best co-exist because psychology can’t ask or answer the relevant normative

questions:

The original epistemological questions seem to be perfectly good questions,

well worthy of our attention. It is difficult to see, then, why the availability

of this other field of study [psychology], concerning how we reason, is a

suitable replacement for the evaluative questions that are at the heart of

epistemology. (Feldman 2003, 168)

Feldman’s point can be made against the theories of SAE: ‘‘It is difficult to

see, then, why the availability of this other field of study’’ (namely, SAE,

which aims to describe how some people make evaluative judgments) ‘‘is a

suitable replacement for the evaluative questions that are at the heart of

epistemology.’’ The proponent of SAE is replacing normative questions

about how to evaluate reason and belief with descriptive questions about

how proponents of SAE evaluate reason and belief. To suppose that

answers to the descriptive questions are also answers to the normative

questions is to take a big leap.

Lawrence BonJour argues that any epistemology subsumed by psychology

does not have the resources to evaluate, positively or negatively, beliefs about alleged occult phenomena of various sorts, such as astrological

or phrenological beliefs. For just as naturalized epistemology can say nothing

positive about the justification of science or common sense, and is thus

impotent in the face of skepticism, so also it can say nothing distinctively

negative about the justification of these less reputable sorts of belief.

(BonJour 2002, 244)

We can once again turn the tables on the proponent of SAE. A theory that

accurately describes how a certain group of people make certain evaluative

judgments ‘‘can say nothing positive about the justification of science or

common sense, and is thus impotent in the face of skepticism, so also it

can say nothing distinctively negative about the justification of ’’ occult,

astrological, or phrenological beliefs. Of course, such a theory might tell us

how some people evaluate those beliefs. But that’s not the same as actually

evaluating those beliefs.

In a similar vein, Alvin Plantinga notes that Quinean naturalism in

epistemology cannot be normative:

[T]he most extreme version of naturalism in epistemology eschews normativity

altogether, seeking to replace traditional epistemology (with its concern

with justification, rationality, reasonability, and their normative colleagues)

by descriptive psychology; this seems to be Quine’s suggestion. [fn deleted]

(Plantinga 1993, 45)

By now, we’re confident our argumentative strategy is wearing thin, but

here it is anyway: SAE ‘‘eschews normativity altogether, seeking to replace

traditional epistemology (with its concern with justification, rationality, reasonability,

and their normative colleagues) by descriptive psychology’’—

a psychology that describes how certain people make certain sorts of

judgments.

Proponents of SAE have argued for decades that radically naturalistic

theories of epistemology cannot succeed because they cannot be normative.

Our aim has been to turn the tables on proponents of SAE. We distance

ourselves from these arguments. Merely pointing out that a theory faces the

problem of bridging the is-ought divide does not by itself damn that theory.

And this is a good thing for proponents of SAE. We all start our normative

musings with psychology. Proponents of SAE start by describing a certain

group’s epistemological judgments, and we start with what we have called

Ameliorative Psychology. (We argue that the former is probably very bad

psychology, but that is not essential to the case we’re building here.) When

it comes to bridging the is-ought gap, everybody has work to do.

The Troubles with Standard Analytic Epistemology 111