Goldman Education and Social Epistemology
Select -----
HOME
1992 Contents
1993 Contents
1994 Contents
1995 Contents
1996 Contents
1997 Contents
1998 Contents
1999 Contents
2000 Contents
2001 Contents
2002 Contents
2003 Contents
Author Index
PHILOSOPHY OF EDUCATION 1995
Education and Social Epistemology
Alvin I. Goldman
University of Arizona
TRUTH, TEACHING, AND
EXPERTISE
At a recent conference I was described as a defender of epistemological
"family values." This took me a bit by surprise, since
I once was considered, and in many quarters am still considered,
a radical epistemologist. However, times are changing. In any
case, I am a defender of the tradition insofar as I remain
unmoved by the tides of postmodernism and social constructivism
that are trying to wash away all vestiges of truth and objectivity.
I believe in truth -- "absolute" truth as it is sometimes
called -- and I believe that a great variety of human endeavors
are dedicated, quite properly and understandably, to the discovery
and dissemination of truths.
Two motives drive truth seeking: simple curiosity and practical
advantage. The first is illustrated by the popular fascination
with dinosaurs and their extinction. People want to know why the
dinosaurs became extinct, although this knowledge would serve
no practical end in most cases. Moreover, they want to know the
truth, that is, what really happened, not simply what is generally
believed (so truth must not be equated with consensual belief.)
The desire for truth also can have a prudential rationale. If
a child has a nasty accident on a trip and needs immediate attention,
the parents want a true answer to the question, "Where is
the nearest emergency room?" Believing the truth is usually
(though not invariably) a helpful means to achieving practical
ends, such as prompt medical attention.
The interest in believing truths is amply demonstrated by the
universal linguistic practice of asking questions. The standard
aim of asking a question is to learn the true answer from the
interlocutor. There are exceptions to this pattern. Teachers direct
questions to students even when they (the teachers) already know
the answers. Survey researchers ask questions of respondents simply
to learn the latter's opinions, correct or incorrect. But the
normal purpose of asking a question is to learn the true answer.
This is why we direct questions, wherever possible, to people
we regard as authoritative or knowledgeable, that is, people in
possession of the truth. I don't ask a random person on the street
whether my department has a meeting scheduled for Friday; I call
the department secretary, who knows about such matters.
Interest in true belief -- or "knowledge," as I shall
call it, using this term in a weak sense -- is not confined to
individuals. Many social institutions also have an interest in
knowledge. Science aims to discover new knowledge; the law seeks
the truth about who violated certain statutes, or who committed
a tort, so that justice may be done. Finally, the fundamental
aim of education, that is, of schooling systems at all levels,
is to provide students with knowledge and to develop intellectual
skills that improve their knowledge-acquiring abilities. This,
at any rate, is the traditional image, and I know of no good reason
to abandon it. I do not claim that factual knowledge and knowledge-acquiring
skills are the sole ends of education; but they comprise,
on my view, its most pervasive and characteristic aims.
Perhaps many people would agree with my emphasis on truth for
a small sector of education, for example, mathematics and science.
Who would urge the teaching of false mathematics? But what about
the rest of the curriculum? Well, even history should aim at teaching
truths. Which historical truths should be taught is a difficult
matter, but I do not think we should teach historical falsehoods,
nor misleading historical theses, where a misleading thesis
is one that is itself true but invites inferences to further conclusions
that are false.
Several objections can readily be anticipated here. First, some
people deny the existence of objective truth altogether; or they
deny that there is truth in certain subject matters, so my principles
cannot apply there. Second, it may be observed that many truths
are too complicated, or require too many qualifications, to inflict
on young children. Surely it is permissible to simplify even at
the cost of inaccuracy. Third, it may often be preferable to let
students learn truths on their own rather than have teachers (or
textbooks) present those truths. Fourth, who is to decide what
is true and therefore what should be taught? How should schools
and teachers proceed when there are divergent opinions in the
local or professional community?
Starting with the first objection, I regret that I cannot here
address a global skepticism or nihilism about truth. I plan to
address this topic in a book I am currently writing,1 but it cannot
be satisfactorily treated in a short conference paper that has
other issues on its agenda. Suffice it to say that I find global
critiques of truth based on postmodernist, social constructivist,
or relativist themes unpersuasive. Let me turn, then, to restricted
skepticism about truth. I grant that there may be domains lacking
in truth values, and my theses would have no direct application
to those domains. But notice that in any domain we may distinguish
primary judgments from secondary judgments. To illustrate,
a primary judgment in the aesthetics of music might be: "Beethoven's
Eroica is greater than Mozart's 40th Symphony." A
secondary judgment in this area would be: "Some music lovers
think (or say) that the Eroica is greater
than Mozart's 40th for reasons A, B, and C." Even if primary
judgments in this area lack truth values, secondary judgments
clearly have them, and it is plausible to expect teachers to aim
at teaching some of these true secondary judgments. A similar
point might be made in ethics. Even if it is conceded that primary
statements of an ethical sort lack truth values, there are truth-valuable
secondary statements that may well be worth teaching; and the
true ones are to be preferred to the false.
Moving to the second objection -- the need to simplify at the
cost of inaccuracy -- I completely concede the point. Often it
is simplifications or approximations of the truth that should
be given to young children; let my proposals be modified accordingly.
Turning to the third objection, it is of course sometimes preferable
to let students learn things on their own rather than instruct
them didactically. That is why education aims to teach skills,
not just facts. But the desired skills or methods of self learning
should be truth-conducive methods: techniques or skills that facilitate
the identification of truth and the rejection of error. The most
basic and universal skills that preoccupy education are the three
R's, and these can best be viewed as means to knowledge acquisition.
Arithmetic competence enables a child to correctly determine (or
know) whether she is receiving proper credit or change in financial
transactions. Reading and writing skills enhance communicative
competence, promote the receipt and transmission of relevant information,
and thereby advance the knowledge prospects of the learner and
the wider community. Another important skill is the ability to
participate constructively in discussion and debate, which also,
I believe, can be rationalized in terms of collective discovery
and mutual persuasion of truth.2
The fourth objection concerns the integrity and viability of expertise.
Expertise is an important topic in social epistemology, which
is the context in which I approach it.3 Let us define an expert
as someone who has true answers to questions in the domain of
expertise, or who has the capacity to readily acquire true answers
when questions are raised. Thus, an opera expert is someone who
can correctly answer questions about opera (without consulting
a reference book), and an expert on automobile engines or kidneys
is someone who can correctly determine why a particular token
of the type is malfunctioning and what treatment would correct
the malfunction. Expertise can be understood either in a comparative
or an absolute sense. Someone is comparatively expert if
her question-answering power ranks high compared with others;
absolutely expert if her power ranks high in absolute terms. The
definition of expertise, then, is relatively unproblematic.
The tougher question about expertise is whether it can be recognized
or identified. Many people claim to be experts, but how
should a community decide whether they are? When, if ever, should
a community defer to the expertise or authority of educators (or
the writers of textbooks) in deciding what is true? How can a
community decide who is an expert unless they know the relevant
truths themselves, in which case there is no ground for deference?
Finally, when does a teacher have an epistemic right to regard
herself as sufficiently expert to present her opinions as truths?
Mathematics instructors and texts often just present "truths,"
or teach certain techniques (e.g., long division, or square-root
derivation) without typically proving their soundness; and this
goes on in many other fields as well. What makes it epistemically
appropriate, if and when it is? And can it be appropriate even
when members of a local or professional community disagree?
I can only touch on this topic briefly, but the main point is
this. It is sometimes possible for expertise to be demonstrated
to novices, and when this happens deference to expertise is defensible
and it is reasonable for an expert to deploy her expertise. Expertise
can be demonstrated by what I call "truth-revealing situations."
Weather forecasters predict the next day's weather, and novices
can check on whether they get it right when the weather is "revealed"
to all. Similarly for auto mechanics and medical diagnosticians.
The diagnosis and treatment of a malfunctioning engine or kidney
can often be checked by seeing whether the system or organ functions
properly after treatment. Success or failure of the treatment
can often be detected even by a novice, and can be used to calibrate
the diagnostician's expertise. In this fashion, non-experts can
assess expertise. When expertise is thereby established, it seems
reasonable to defer to experts (unless they have ulterior reasons
to deceive or misrepresent their knowledge, as repair persons
often do). Not all domains admit of "truth-revealing"
situations, and then the possibility of consensus is dim. In such
cases, claims to expertise cannot be honored in quite the same
way. That is when teachers should move in the direction of teaching
"secondary" statements rather than their own personally
accepted "primary" statements in the subject matter.
For example, we would expect a high school teacher to teach primary
truths about the formal structure of government where expertise
can be established. But on issues of normative politics, for example,
which side is right in a territorial dispute, one would expect
teachers not to press their own views on the rights and wrongs
of the matter ("primary" statements), but to emphasize
"secondary" truths, that is, how each party to the dispute
defends its territorial claim.
VERITISTIC EPISTEMOLOGY AND
MULTICULTURALISM
The brand of epistemology I am advocating might be called veritistic
epistemology because of its heavy emphasis on truth. This epistemology
might initially seem committed to a certain position in the contemporary
debate over the curriculum. It might seem to side necessarily
with "essentialism" -- espousal of a core curriculum
-- as opposed to "multiculturalism."4 Am I not, after
all, just espousing the "tyranny of Truth" (with a capital
"T"), which is the heart of essentialism? Not at all.
This association (between veritistic epistemology and essentialism)
is by no means necessary, and the appearance of such a connection
must be corrected.
The spirit of essentialism is succinctly expressed in the following
argument from Robert Maynard Hutchins: "Education implies
teaching. Teaching implies knowledge. Knowledge is truth. The
truth is everywhere the same. Hence education should be everywhere
the same."5 Now the first
several premises of this argument,
a few quibbles aside, strike me as true. I accept that education
implies teaching, that teaching (at least in large part) is the
conveying of knowledge, and that knowledge is truth (more precisely,
knowledge entails truth). What about the fourth premise,
that truth is everywhere the same? This raises some technical
issues about propositions, but let us restrict discussion to propositions
devoid of any indexical or demonstrative elements like "I,"
"you," "here," "now," etc. Then
I would agree that for any specified proposition P, its truth
value is the same at all times and places. (People's beliefs
about a proposition's truth value, of course, may vary over time,
but that does not entail that the truth itself varies over time.)
Since I accept all four premises, am I committed to the conclusion,
therefore, that education should be everywhere the same? No, because
the conclusion does not follow from the premises. The reason is
simple. There are many truths; although each of these is true
at all times and places, it does not follow that each should be
taught at all times and places. Ignoring our earlier qualifications
about simplification and approximation, we may say that being
true is a necessary condition for being taught but not
a sufficient condition. That leaves open the possibility
of teaching different truths at different times and places.
Hutchins's conclusion might follow from the premises if the premises
are taken to imply that there is a single truth, or a single totality
of truths, that should be taught everywhere. Certainly one possible
reading of the fourth premise, "The truth is everywhere
the same," is that there is a single totality of truths.
But even if we grant the premise, so understood, Hutchins would
need a further premise to the effect that this totality should
be taught everywhere, and that is dubious in the extreme. On a
more plausible interpretation of the original premises, they do
imply that only truths should be taught. But this is compatible
with the idea that the particular subset of truths to be taught
may be relativized to locale, culture, and context.
Some essentialists, no doubt, maintain that the truths of morality
and human nature are found in certain classical works of the European
tradition, and perhaps only in those works. If you hold that position,
and you hold that the truths on those topics should be taught
everywhere, then you get the doctrine that a certain canon should
comprise the curriculum. But this doctrine does not follow from
the objectivity of truth, or from objectivity conjoined with the
view that education should teach truth. These are compatible with
the idea that truths are found in works from many different traditions
so that no particular tradition should monopolize the curriculum.
Furthermore, even if a particular tradition contains more
truths than other traditions, it may be an important truth for
students to learn that there are many traditions.6 In a community
as diverse as America, for example, it may be particularly important
to teach about such diversity. But the existence of diverse races,
genders, cultures, and ethnicities, and the range of distinctive
values and perspectives that typically accompany these diverse
identities, are themselves facts or truths. So educational multiculturalism
need not stand in conflict with veritistic epistemology.
I just said that it may be particularly important to teach diversity.
But what makes one truth more "important" to teach than
another? Importance, I think, is a function of interests, but
different types of interested parties and types of interests may
be relevant. Let me start with the former. Two types of parties
we can identify are individual believers (or learners), on the
one hand, and the social systems or institutions of which they
are a part, on the other. Although the interests of students are
certainly relevant to the question of what should be taught, the
interests of society as a whole should also be considered (certainly
in the case of primary and secondary education). The situation
is analogous to that of a criminal trial. Disinterested or unconscientious
jurors may not care a whit whether they get the truth about the
guilt of the defendant. But the judicial system as an institution
certainly does have an interest in the rendering of a true verdict.
Similarly, society may have an interest in its children learning
certain truths, even if the children themselves are not terribly
interested in those truths. Society's interest should not be ignored,
just as the judicial system's interest should not be ignored.
Turning to the definition of "interest," let me focus
on the learner's interests. There are three relevant senses or
types of "interest." One measure of a question's interest
is whether the learner finds it interesting, that is, has
an aroused curiosity or concern about the question's answer. Such
concern can arise from intrinsic fascination or from recognition
of the potential practical value of knowing a correct answer.
A second measure of interest is dispositional rather than occurrent.
Many questions would be interesting to a person if he/she
only considered them. A third sense is more broadly dispositional:
what would interest the learner if she knew certain things
she does not currently know. Certain types of knowledge might
be objectively in a student's interest, however unappreciative
the student may be of this at the moment.
Returning to multiculturalism, an argument for it might be based
on several of the foregoing factors. First, it may be in society's
interest for students to have knowledge of the diversity of their
world. Second, such knowledge may be in the students' interest
(in the third sense of "interest"), whether they realize
it now or not. Third, tailoring or adjusting curricula to the
cultures of different student bodies may well be warranted by
the obvious fact that material from one's own culture (or gender,
or ethnicity, etc.) is more likely to be interesting (in the first
sense of 'interest'). Better learning takes place when there is
active interest, and good learning of one subject often has beneficial
consequences for other learning. Thus, even a veritistic approach
to education offers many possible rationales for multiculturalism.
SOCIAL EPISTEMOLOGY, PEDAGOGY,
AND ARGUMENTATION
I have characterized my form of epistemology as "veritistic"
epistemology, but the title of my paper makes reference to social
epistemology, and I have not fully explained what that is. I think
of individual and social epistemology as two sectors of the subject.
Individual epistemology studies intellectual activities of single
cognitive agents in abstraction from others in order to see how
modes of belief formation promote or impede knowledge acquisition.
Social epistemology studies the social or interactive practices
of multiple agents in order to see how their interactions encourage
or obstruct knowledge acquisition. Two categories of social practices
may be highlighted here. First, there are practices of speech
in which a speaker tries to inform or persuade an audience, often
supporting his claims with reasons or argumentation. A second
category of social practices are the inferential practices of
hearers who try to decide how much to trust what speakers say,
assessing their credibility on the topic in question and their
competence compared with other speakers and possible knowledge
sources. Educational theory is obviously concerned with an appraisal
of activities of both sorts. Which speech practices should be
expected of teachers, and which inferential and learning practices
of students ought to be expected or encouraged?
A number of recent writers on the philosophy of education have
stressed the role of reasons in teaching, including Israel
Scheffler, Harvey Siegel, and Kenneth Strike. Here is a representative
passage by Scheffler, quoted approvingly by Siegel:
To teach…is at some points at least to submit
oneself to
the understanding and independent judgment of the pupil, to his
demand for reasons, to his sense of what constitutes an adequate
explanation. To teach someone that such and such is the case is
not merely to try to get him to believe it: deception, for example,
is not a method or mode of teaching. Teaching involves further
that, if we try to get the student to believe that such and such
is the case, we try also to get him to believe it for reasons
that, within the limits of his capacity to grasp, are our
reasons.7
Siegel endorses this idea and expands upon it in terms of an ideal
of critical thinking.8 He writes,
for example:
We want to get students to be able to think critically, and that
means, in part, getting them to understand what the rules of assessment
and criteria of evaluation of claims are. We want our students
to learn, for example, the evidential criteria underlying our
judgments that some piece of evidence supports claim X,
but that another piece does not support claim Y.9
I am in broad sympathy with the position of Scheffler and Siegel,
but I would like to base it on a deeper foundation, and also take
some (limited) exception to their theses.
Reasons-giving, I suggest, should be viewed as argumentation.
To give reasons for believing a certain proposition is to treat
that proposition as a conclusion of an argument of which the reasons
are premises. Now formal logic studies the deductive and/or inductive
relations among propositions or sentences abstractly considered.
But formal logic does not exhaust the subject of argumentation,
where argumentation is construed as a complex speech act in which
a speaker defends a thesis to an audience by appeal to reasons
or premises. In a previous paper, I have claimed that there are
tacitly accepted rules governing the practice of argumentation,
rules that go beyond those of formal logic or the theory of evidential
relations.10 For example, I
suggest the following rules of good
argumentation:
(1) a speaker should assert a conclusion only if she believes
it;
(2) a speaker should assert a premise only if she believes it;
(3) a speaker should assert a premise only if she is justified
in believing it; and
(4) a speaker should affirm a conclusion on the basis of stated
premises only if (a) those premises strongly support the conclusion,
(b) she believes that they strongly support it, and (c) she is
justified in believing that they strongly support it.11
Now the theory of reasons and critical thinking advanced by Scheffler
and Siegel makes little reference to the aims of true belief and
error avoidance. But I suggest that the rationale for the rules
of good argumentation is that they promote (or are thought to
promote) these veritistic goals. For example, the first two rules
instruct a speaker not to assert things that are false by her
lights, because what is false by her lights may well be
false; and assertion of such utterances is apt to induce false
beliefs in the audience. The justification requirement in rules
(3) and (4c) may be rationalized in similar terms. In particular,
on a reliabilist approach to epistemic justification of
the sort I have defended, justified beliefs are ones produced
by belief-forming processes with high truth-ratios.12 So justified
beliefs are likely to be true; and confining oneself to premises
and support relations that one is justified in believing will
conduce to the assertion of true conclusions, and hence to the
production of true beliefs on the part of hearers who accept those
conclusions. The aim of reasons-giving in the sense of proper
argumentation, then, has its foundation in the aim of producing
true belief and error avoidance. To the extent that teachers comply
with the principles of good argumentation, they can also be expected
to serve the educational goal of advancing their students' knowledge.
So reasons-giving, as thus far considered, is not a distinct
goal from truth, but a means to that end. There is also a special
reason for teachers to display the qualities of good argumentation,
namely, that teachers are models and exemplars of speaking and
thinking.13 By displaying good
argumentative practice under the
rules I have sketched, teachers show what counts as good evidence
and good argumentative speech, and through this exposure students
may come to internalize the criteria of good evidence and the
skills of good (internal) inference and good (public) argumentation.
The latter are among the truth-promoting skills that an educational
system should hope to instill in students.
Until now, the rules of good argumentation we have considered
pertain only to the speaker and her state of mind. Shouldn't there
also be rules that bring the audience into the picture? Shouldn't
the content of a good argument be sensitive to the intended audience,
and isn't this particularly relevant to teachers as arguers or
reasons givers? This point is at least partly appreciated by Kenneth
Strike who writes:
Propositions that are objective evidence for some
claim must be
subjectively seen as evidence by the student….A proposition
or a phenomenon is only evidence for a claim in relation to a
set of concepts that interpret it….The suggestion that evidence
is relative to the student's current concepts indicates a need
on the part of the teacher to know what the student's current
concepts are.14
I would say that a proposition is evidentially relevant for a
hearer not only in relation to the hearer's concepts (as
Strike says) but also in relation to the hearer's prior beliefs
and capacities for appreciating (deductive and inductive) support
relations. I would formulate this in terms of an additional
rule of good argumentation:
(5) A speaker addressing a particular audience should restrict
her premises to statements that the audience is (or would be)
justified in believing, and should restrict herself to a support
relationship between premises and conclusion that the audience
is capable of recognizing or appreciating.15
Although this rule applies to all speakers and audiences, we are
interested in its application to teachers and students. The rule
implies that a teacher must always take into account what students
already believe or don't believe, since this determines whether
the students would be justified in believing certain possible
premises. In other words, the permissibility of using certain
statements as premises depends on the students' prior informational
states. As rule (5) implies, a good pedagogue should also take
into account what inferential relations the students are capable
of appreciating. It isn't enough that the support relation is
in fact strong; the audience, in our case the students, should
be capable of appreciating the strength of the relation.
AUTONOMY, TRUST, AND
TESTIMONY-BASED BELIEF
An unrestricted form of the "reasons" thesis says that
everything a teacher asserts must be backed up by reasons. But
that is obviously too strong; a speaker's reasons must come to
an end somewhere, namely, wherever her assertions are otherwise
undefended premises. Rule (5), however, says that the premises
of a good specimen of argumentation must be statements that the
audience is, or would be, justified in believing. If this is right,
justification for believing undefended premises must have a different
source, not the current argument of the speaker. One possibility
is that the hearer has prior independent information that justifies
him in accepting the speaker's premises. But isn't there another
possibility? Can't a hearer be justified in believing what a
speaker asserts simply because she asserts it? Here we
should refocus our discussion away from speech practices of speakers
to belief practices of hearers, but the issue also bears on speech
practices and educational practices generally, because it raises
the issue of what students should be expected or encouraged to
believe on the basis of teachers' assertions. Many writers on
education stress the need to respect the student's autonomy.
Students, like all people, have a prima facie right and responsibility
to be self-governing, and in the epistemic sphere this seems to
mean that they have the right and responsibility to make belief
decisions for themselves. Now in a sense, this is trivial. There
is clearly a sense in which everyone necessarily makes
their own belief decisions. How can one person literally make
a belief decision for another? What is presumably meant by a thesis
of autonomy, then, has something to do with the rejection of trust.
Strong autonomy would say that nobody should ever trust another
in the sense of accepting what they say simply because they
say so. If a hearer is justified in believing P because some
speaker asserts P, it must be because the hearer has reasons to
trust the speaker. Such trust has to be earned; it cannot come
automatically. So teachers are not entitled to expect students
to accept what they say simply because they say it.
This thesis has a nice liberal-sounding air to it; and it may
be right. But recent discussions in the epistemology of testimony
-- a branch of social epistemology, as I would categorize it --
create much room for doubt. Let us briefly review three historical
positions on the epistemology of testimony, those of Locke, Hume,
and Reid. Locke took the strictest position on intellectual self-reliance
claiming that we should not trust the faculties of others. He
expressed doubts about granting even derivative authority to the
opinions of others, that is, authority based on prior determination
of the speaker's reliability.16
Unlike Locke, Hume emphasized
the usefulness of derivative authority. He appreciated the extent
to which we rely on the opinions of others, but also insisted
that we should rely on these opinions only to the degree that
we have observational, non-testimonial reasons for thinking that
they are reliable. Thomas Reid took a rather different position.
He held that the testimony of others, or at least their sincere
testimony, is prima facie credible, even if we do not have an
independent check on the testifier's reliability. Reid thought
that if our natural attitudes of trust, both in ourselves and
in others, were not reasonable, the inevitable result would be
skepticism. He therefore placed testimonial justification on an
equal footing with perception and memory as a "first principle."
This first principle, Reid held, is founded in certain innate
dispositions: veracity, which disposes us to tell the truth, and
credulity, which disposes us to believe what is said. For Reid,
then, the child's default tendency to believe what he/she is told
is epistemically in order, not something to be purged by an acid
bath of autonomy.17
A Reidian position has been endorsed by a number of recent writers
on testimony, especially Tyler Burge, Richard Foley, and Alvin
Plantinga.18 I find Foley's
formulation of this position particularly
congenial. Foley distinguishes epistemic egoism and non-egoism.
The epistemic egoist grants no fundamental authority to others,
just as the ethical egoist grants no fundamental value to the
happiness of others. Epistemic egoists can grant derivative authority
to others, but only on the basis of having personally established
their reliability. Epistemic non-egoists, by analogy with ethical
altruists, are prepared to grant others fundamental intellectual
authority. In ethical theory, it has been debated whether egoism
is a consistent position, the negative side holding that it is
inconsistent to assign value to one's own happiness, but not to
other people's happiness, despite their similarity to oneself.
In a similar spirit, Foley argues that if I grant fundamental
intellectual authority to myself, in order to be consistent, I
must grant it to others, because it is reasonable for me to think
that their intellectual faculties and environment are broadly
similar to my own. Foley goes on to say that when my own opinions
conflict with those of a testifier, the prima facie authority
of his testimony may be defeated or overridden by my opinions
(especially when I take myself to have expertise on the subject
in question). Thus, it isn't always appropriate to place
final trust in other people's say-so. Nonetheless, as a default
position, trust in others is warranted even when one has no independent
grounds for certifying their reliability.
Several other recent writers have pressed the impossibility of
verifying the reliability of testimony by non-testimonial means.
C. A. J. Coady argues that it is practically impossible for an
individual to personally check on more than a tiny percentage
of testimonial reports, so the basis for an induction is too slim
to provide much justification.19
John Hardwig has pointed out
that trust is an essential part of science, where collaboration
is often required among multiple scientists and no specialist
knows enough about the other specialties.20 To take a recent example
reported in Science, mathematical group theory has an Enormous
Theorem (as it is affectionately called), describing the taxonomy
of simple groups, the proof of which runs an estimated 15,000
pages spread over upwards of a thousand separate papers written
by hundreds of researchers. The proof of the Enormous Theorem
has so many pieces that even the experts who produced it rely
on one another for assurance that the pieces fit together.21 If
even expert mathematicians rely on trust, why shouldn't students,
especially young students, be epistemically permitted to exercise
trust in their teachers?
Can the practice of trust be rationalized on veritistic grounds?
It might be. If Reid is right that people have innate dispositions
toward veracity and credulity, and if they are sufficiently competent,
then trust may be a truth-conducive practice. For young children
to decline to trust their elders would consign them to massive
ignorance. The situation may be compared to language learning.
Cognitive studies of language in the Chomskyan tradition indicate
that young children have innate tendencies to lean toward certain
hypotheses about the language corpora they encounter. This may
be a bias, if you wish, but it is a bias that enables them to
learn correctly the grammars of languages they actually encounter.
Innate credulity might have similar properties.
Of course, as one grows older, one can do better than exercise
unqualified credulity. (And indeed it seems plausible that if
there is a credulity "module" at all, it ossifies as
one leaves childhood, just as the language learning module becomes
dysfunctional in adolescence.) So I do not mean to downplay the
value of critical thinking. On the other hand, radical autonomism
may well go too far, epistemically speaking, in disparaging the
propriety of trust. We should not erect an epistemic standard
for education that is excessively high. This is one of many issues
in which educational theory and social epistemology have overlapping
interests.22
 For a response
to this essay, see Phillips.
1. The book in preparation is
tentatively called Knowledge
in a Social World.
2. My picture of intellectual
discussion and debate diverges sharply
from that of social constructivists, who view these processes
as "negotiating," i.e., creating, the truth.
I dispute the suggestion that truth, in general, is created by
discussion or debate. No doubt, some discussions enact new social
policies, and thereby create new policy facts. But intellectual
(as opposed to practical) discussion is generally aimed at forming
beliefs about antecedently existing truths or facts, not at creating
new truths or facts.
3. I discuss it in similar terms in
"Epistemic Paternalism:
Communication Control in Law and Society," The Journal
of Philosophy 88 (1991): 113-131, sect. viii.
4. The label "essentialism"
is used by Amy Gutmann in
her editorial introduction to Charles Taylor et al, Multiculturalism
and the "Politics of Recognition" (Princeton, N.J.:
Princeton University Press, 1992), 13.
5. Robert Maynard Hutchins, The
Higher Learning in America
(New Haven: Yale University Press, 1936), 66.
6. This point is emphasized, for
example, by Susan Wolf in her
comment on the essay by Charles Taylor, in Multiculturalism
and "The Politics of Recognition."
7. Israel Scheffler, The Language
of Education (Springfield,
IL: Charles Thomas, 1960), 57.
8. Harvey Siegel, Educating
Reason: Rationality, Critical Thinking,
and Education (New York: Routledge, 1988).
9. Ibid., 44-45.
10. Alvin I. Goldman,
"Argumentation and Social Epistemology,"
The Journal of Philosophy 91 (1994): 27-49.
11. Ibid., 34.
12. See my Epistemology and
Cognition (Cambridge, Mass.:
Harvard University Press, 1986), chaps. 4-5, and Liaisons:
Philosophy Meets the Cognitive and Social Sciences (Cambridge,
Mass.: MIT Press, 1992), esp. chaps. 6, 7, 9.
13. Compare Kenneth Strike,
Liberty and Learning (St. Martin's
Press, 1982), partly reprinted in A Professor's Duties,
ed. Peter Markie (Lanham, Md.: Rowman and Littlefield, 1994),
106.
14. Ibid., 106.
15. This rule is suggested in my
"Argumentation and Interpersonal
Justification," in Proceedings of the 3rd International
ISSA Conference on Argumentation, ed. Frans H. van Eemeren,
Rob Grootendorst, J. Anthony Blair, and Charles A. Willard (Amsterdam,
forthcoming). It was partly inspired by Richard Feldman's person-relative
analysis of good argumentation, in "Good Arguments,"
in Socializing Epistemology: The Social Dimensions of Knowledge,
ed. Frederick F. Schmitt (Lanham, Md.: Rowman and Littlefield,
1994).
16. "For, I think, we may as
rationally hope to see with
other Mens Eyes, as to know by other Mens Understandings. So much
as we our selves consider and comprehend of Truth and Reason,
so much we possess of real and true Knowledge. The floating of
other Mens Opinions in our brains makes us not one jot the more
knowing, though they happen to be true. What in them was science,
is in us but Opiniatretry….Such borrowed Wealth, like Fairy-money,
though it were Gold in the hand from he received it, will be but
Leaves and Dust when it comes to use." An Essay Concerning
Human Understanding,, ed. A. C. Fraser (New York: Dover, 1959),
Book I, chap. iii, para. 23.
17. This is my formulation, not
Reid's. Reid's discussion may
be found in Thomas Reid, Inquiry into the Human Mind, ed.
Timothy Duggan (Chicago: University of Chicago Press, 1970), chap.
6.
18. Tyler Burge, "Content
Preservation," The Philosophical
Review 102 (1993): 457-88; Richard Foley, "Egoism in
Epistemology," in Socializing Epistemology: The Social
Dimensions of Knowledge, ed. Frederick F. Schmitt (Lanham,
Md.: Rowman and Littlefield, 1994); and Alvin Plantinga, Warrant
and Proper Function (New York: Oxford University Press, 1993),
chap. 4. For useful discussion, also see C. A. J. Coady, Testimony:
A Philosophical Study (Oxford: Oxford University Press, 1992).
19. Coady, Testimony, chap.
4.
20. Hardwig, "Epistemic
Dependence," The Journal
of Philosophy 82 (1985): 335-49.
21. "At Math Meetings, Enormous
Theorem Eclipses Fermat,"
Science 275 (1995): 794-95.
22. I wish to thank Harvey Siegel
for the invitation to present
this paper to the Philosophy of Education Society, and for helpful
bibliographical suggestions. Thanks also to my commentator, D.
C. Phillips, for valuable comments, and to several people, especially
Patrick Suppes and Sophie Haratounian-Gordon, who raised particularly
pertinent questions during the discussion period.
©1996-2004 PHILOSOPHY OF EDUCATION SOCIETYALL RIGHTS RESERVED
|
|
