II
Differing ideas of or views about education must
agree with much of what has been said, particularly
with the general outlines of the analysis given of the
idea of education and with the statement of the pre-
suppositions of any educational enterprise. They may
include different views about the ranges of X, Y, W,
and Z to be built into the idea or definition of educa-
tion. However, even if they agree completely about
conceptual matters, they may and do still differ about
substantive issues. In fact, as Eliot sees, it is precisely
these further substantive questions that have been and
are the historically and practically most important
ones. These substantive questions, which remain open
on any plausible definition of education, roughly stated,
are: (1) Are the postulates of education true? Are the
excellences cultivatible by education? Need they be
so cultivated? (2) What dispositions are desirable and
to be fostered by education? What dispositions are
excellences? (3) By what means or in what ways should
education (educators) seek to foster these desirable
dispositions? (4) Who is to be educated? How should
educational opportunity be distributed? (5) Who should
educate?
Actually each of these questions is a family or group
of questions. They are, moreover, interrelated and
hence cannot be answered in entire independence of
one another, e.g., (1) and (2), (2) and (4), (3) and (4),
and (4) and (5). In what follows, however, we shall
have to keep them somewhat separate. It should also
be noted that the last four questions are normative,
since they ask what should be done, or what is desir-
able, while the first is not.
The main point for our purposes now, however, is
the fact that theories and philosophies of education
arise as answers to these substantive questions and,
apart from conceptual or definitional preliminaries like
the above, consist of and are distinguished by their
answers to them. Before we discuss the questions and
the issues involved in answering them, we must stop
to look at such substantive theories and philosophies,
to see what they are like, what they include, and how
they are or should be put together; this is the second
main part of our task—to analyze the kind of idea of
education referred to earlier in (2b).
A theory of education, then, is a set of answers to
the above five questions. Since it includes answers to
the last four it will be normative, saying what educa-
tion should be like, not just descriptive, explanatory,
or predictive, as a psychological theory of learning or
child development would be. In J. S. Mill's language
(A System of Logic [1843], Book VI, Chs. V, XII),
education is not a science, but an art. It may, however,
and no doubt should, make use of such scientific
theories of development and learning as a basis for
some of its normative conclusions; in fact, Mill thought
educational theories should rest their normative “pre-
cepts” entirely on such premisses as psychology alone
can provide, except for the one basic normative prem-
iss supplied by ethics, which for him was the princi-
ple of utility.
What is usually called a philosophy of education is
a theory in this normative sense, but not every such
theory is properly called a philosophy. For a theory
of education might simply assume, without argument,
that the dispositions to be promoted and the methods
to be used are those regarded as desirable by the society
or individual the education is to serve, and then it can
be called a philosophy of education only by extreme
courtesy. It is better regarded as a minimal theory of
education, reserving the title of philosophy of educa-
tion for fuller theories that provide a reasoned justifi-
cation for their answers to normative questions about
education.
Every theory of education in our sense will, then,
assume an affirmative answer to the first question,
though it may do so dogmatically, without discussion.
That is, it assumes that the acquisition of desirable
dispositions is not wholly a matter of nature, luck,
divine gift, or choice, but is in part or to some extent
amenable to educational programming. If it seeks to
defend these assumptions, it must list these dispositions,
analyze them, and show that the claims made in the
assumptions are true. In other words, it must establish
certain facts about human nature and about the world.
To do this it may appeal to science, to metaphysics,
or to theology—different thinkers will have different
views about what is to be appealed to, views that will
depend on their general philosophical orientations.
What means, methods, or practices education is to
make use of—e.g., just what the teacher is to do in
the classroom—will appear in answer to question (3).
Even a minimal theory of education may try to give
a reasoned reply to this question by seeking to justify
its recommendations. How then may a precept about
the method of teaching something be justified? Suppose
one maintains, as the Greeks did, that in order to foster
the moral virtues we should use music of certain sorts,
at least during a certain stage in a child's life (a belief
that was for the most part given up in the Hellenistic
Period, though parents even in the twentieth century
sometimes wonder about the possible moral effects of
some new combinations oof soound that some of their
children listen to). To justify this claim one must use
an argument something like this: (a) Education should
cultivate moral virtue. (b) The hearing of such and such
kinds of music is conducive to moral virtue. (c) There-
fore education should make use of music of those kinds.
Or suppose we use the dictum that, no matter what
disposition is being fostered, learning is by doing. Then
our reasoning must be along these lines: (a) Education
should foster an understanding of music. (b) Any dispo-
sition is more effectively fostered if some relevant
“doing” on the part of the student is arranged for. (c)
Therefore education should include learning to sing or
play an instrument.
Thus, in order to justify any normative conclusion
in answer to question (3), whether this is specific or
general, one must make use of a normative premiss
like (a) in these examples, which says something about
dispositions to be promoted, and of a factual premiss
like (b) which says that a certain method or practice
is necessary, sufficient, or at least helpful for the pro-
motion of those dispositions.
Two things about premisses like (b) should be noted.
In the first place, even if they are simply assumed or
borrowed from common sense or tradition, they are
empirical statements that may in principle be verified
by empirical observation and scientific testing, and any
theory that seeks to justify them must appeal to expe-
rience or to some empirical science. In the second
place, they may be of different kinds depending on
whether they assert that a certain practice is necessary,
sufficient, or neither necessary nor sufficient but still
helpful, for the fostering of the disposition referred to
in premiss (a), or simply that it is more effective in
doing so than other methods are; and the conclusion
in (c) must be understood differently, depending on
which of these claims they make, though the argument
may in each case be read as establishing that the prac-
tice in question has some value or desirability.
Arguments like those illustrated do not, however,
establish that the practices they defend ought to be
employed unless they show the practices to be neces-
sary. Take the following argument: (a) Education
should foster citizenship. (b) Indoctrination is con-
ducive to citizenship. (c) Therefore education should
include indoctrination. Even if one accepts its prem-
isses one may reject the conclusion because one does
not regard citizenship as having top priority; but, even
if one gives citizenship first place, one may reject it
because one regards indoctrination as morally wrong.
Of course, if one believes that citizenship must be given
first place in education, and that indoctrination is nec-
essary for promoting citizenship, then one must con
clude that indoctrination should be used. But then one
will not regard its use as morally objectionable. This
example shows that ethical considerations are impor-
tant in connection with question (3) as well as scientific
ones, since methods must be shown to be morally
unobjectionable as well as effective or helpful in pro-
ducing desirable dispositions before we can consider
them justified.
Still, except when ethical premisses by themselves
dictate something about educational methods, e.g., that
educators should not use lies (except in cases in which
lying is morally excusable, if there are any), the justifi-
cation of answers to question (3) will include a premiss
like (a) in our examples that presupposes an answer
to question (2), plus, of course, a factual premiss like
(b). In this sense, (2) is the central normative question
in any theory of education, and the central part of any
such theory is a list and description of the dispositions
to be fostered by education. How then is one to justify
saying that a certain disposition (which is not simply
a matter of nature, luck, divine gift, or choice) should
be cultivated by education? From what has been said,
it follows that one must show that the disposition is
desirable on some ground and that it is not morally
objectionable. In order to show that it is not morally
reprehensible he must, of course, appeal to some
ethical premiss about what is or is not morally wrong,
bad, or vicious, and at least sometimes also to a factual
premiss. For example, to show that a liking for the
kinds of music Plato and Aristotle banned from educa-
tion is not morally bad, one would have to use a prem-
iss telling us what moral virtues we should have and
a factual one to the effect that a liking for those kinds
of music does not conflict with the acquisition of those
virtues.
In order to show that it is desirable to foster a certain
morally innocuous disposition by education, one must,
again, use premisses of two kinds, namely, ethical or
other value premisses stating more ultimate aims or
principles of education, and factual ones stating that
the disposition in question is necessary, sufficient, or
at least helpful in relation to them. For example, one
might accept, as many would, the three aims of educa-
tion discussed by Eliot (p. 69): 1. To prepare a child
to make a living (for a vocation). 2. To equip him to
be a good citizen. 3. To develop his powers and so
enable him to enjoy a good life. Then to show that
education should foster a certain disposition one would
show that its acquisition or possession is required by
or at least conducive to one of these ends (and not
inconsistent with a more important end). The argument
would have this form: (a) Education should promote
such and such an end (or principle). (b) Disposition
W is conducive to this end. (c) Therefore education
should foster W. Here (a) is a normative or value
premiss; it belongs to one's ethical or value theory,
more specifically, to one's political or social philoso-
phy. Political or social philosophy is thus shown to be
of crucial importance in the theory of education. As
Aristotle said, it is
politike
that ordains which of the sciences are to exist in states,
and what branches of knowledge the different classes of
citizens are to learn, and up to what point...
(Ethics I, 2).
Then (b) is a factual premiss, saying that a certain
disposition is necessary, sufficient, or at least helpful
in achieving a certain end (or living by a certain prin-
ciple); it will usually be of a kind that depends on
experience and science for its verification, but in some
theories of education it might come from metaphysics
or theology.
Thus, answers to question (3) depend on answers to
question (2)—which give us the “proximate” aims of
education—and answers to question (2) depend on
answers to a more basic question which give us the
more “ultimate” aims or principles of education,
factual premisses appearing in both cases. How then
are answers to this more basic question, statements
about the more ultimate aims of education, to be justi-
fied? An educational theorist might stop at this point
and refer us to a philosopher or theologian, but, if he
is offering us a full-fledged philosophy of education,
he will try to justify his statement in (a) of our last
example. Then, again, he must appeal to premisses of
two kinds: first, a still more basic normative or value
premiss, and, second, a still more basic factual one.
There is no one form his reasoning must take, but he
will make use of premisses like the following: We ought
always to do what will bring about the greatest general
balance of good over evil (the principle of utility);
Society ought to be just; Pleasure is the end of life;
Contemplating the heavens with understanding is good
in itself; Belief in Jesus Christ is necessary for salvation;
Making a living (having a vocation) is necessary both
for life and for the good life; This life is all there is.
Such premisses contain no explicit reference to educa-
tion, and hence do not belong specifically to the phi-
losophy of education but to other branches of philoso-
phy, to science, or to theology. One may, of course,
seek to justify them in turn by appeal to more basic
premisses, until one finally comes to one's most basic
ethical or value premisses and one's most basic beliefs
about man and the universe.
To illustrate what has been said, one relatively com-
plete line of argument in education might proceed as
follows: (a) Other things being equal, what is good in
itself should be pursued and promoted. (b) Contem-
plating the heavens with understanding is good in itself.
(c) Therefore the contemplative understanding of the
heavens should be pursued and promoted. (d)This en-
tails acquiring and fostering the knowledge of astron-
omy (the disposition called a knowledge of astronomy).
(e) This can be done by education and by education
alone. (f) Therefore education should foster a knowl-
edge of astronomy (other things being equal). (g) In
order to do this it is necessary, among other things,
to initiate people into the use of the telescope. (h)
Therefore education should initiate the young into the
use of the telescope. Granting the premisses, this is,
as far as it goes, a good argument for its conclusions
in (c), (f), and (h).
It is not final, however, for the acquisition of a
mastery of astronomy might be incompatible with that
of more important dispositions. But arguments to show
this would have a somewhat similar structure, and so
this example can be used as a basis for a number of
points. (1) Both factual and normative premisses are
necessary to answer normative questions about educa-
tion. (2) Among the factual premisses must be some
empirical or scientific ones, e.g., (e), (g), and possibly
(d). (3) Epistemological premisses are neither necessary
nor sufficient to establish educational conclusions, as
so many twentieth-century writers on the philosophy
of education seem to assume. (4) Specifically religious,
theological, or metaphysical premisses are also neither
necessary nor sufficient, as Eliot and many others
allege. (5) The philosophy of education is not autono-
mous, for it depends on premisses from other fields.
(6) What is basic and central in the philosophy of
education is such normative inquiries as ethics, value
theory, and social philosophy, as is shown by the role
of premisses like (a) and (b) and conclusions like (c)
and (f). (7) Philosophers of education might content
themselves with establishing conclusions like (c) and
(f), leaving more practical steps like (g) and (h) to
educational scientists and practitioners, but they have
usually attempted to supply such steps too.
Four points should be added. (8) Eliot and others
who hold that a philosophy of education must ulti-
mately rest on religious or theological premisses assume
that the final premisses one appeals to are religious
or theological just because they are normative, because
they are about the nature of man and the universe,
or because they are ultimate. But to say that they must
therefore be religious or theological and not just ethical,
axiological, philosophical, or scientific is to make them
religious or theological simply by a kind of baptism.
For then atheism, naturalism, secularism, cynicism,
hedonism, perhaps even skepticism, all become forms
of religion without undergoing any conversion and
without relaxing their opposition to theism or to what
usually counts as religious or theological belief; and
nothing is gained but a Pyrrhic victory. (9) One may
insist that specifically theistic beliefs must be brought
to bear on educational arguments like the above, e.g.,
in connection with premisses (b), (d), or (e), but this
is not obvious and it is not logically necessary; in fact,
one can agree to this only if one already shares such
theistic beliefs. (10) It remains true that religious,
epistemological, and metaphysical premisses may, so
far as the logic of the matter goes, be relevant to the
justification of educational conclusions. If one believes,
as Thomas Merton did, that the whole work of man
in this life is to find God, one may and, indeed, must
use this belief as the basis of one's philosophy of educa-
tion. That epistemological premisses may be
relevant
even though they are neither
necessary or
sufficient
is shown by one of Cardinal Newman's arguments
(1852) for teaching theology in universities: (a) A uni-
versity should teach knowledge. (b) Theology is a form
of knowledge. (c) Therefore a university should teach
theology (
The Idea of a University [1959], Ch. II). Here
(b) is an epistemological claim. Incidentally, it should
be noticed that neither of Newman's premisses is
specifically religious or theological. (11) Thus, a full
normative philosophy of education will contain the
following kinds of statements, in addition to definitions,
distinctions, and other bits of analysis: 1. Normative
premisses like (a) and (b) in the longer of our last two
examples. 2. Factual premisses like (d), (e), and (g),
including at least some empirical or scientific ones.
3. Normative conclusions answering questions (2) and
(3) like (c), (f), and (h). It may include epistemological,
metaphysical, or religious premisses, though it need
not; if so, they will belong under the second heading
(unless they are normative).
This brings us to questions (4) and (5) on our earlier
list. Answers to these questions are somewhat inter-
woven with answers to questions (2) and (3), as has
been observed, but it is clear that in general they too
will depend on premisses of the two kinds already
distinguished, normative and factual, and that political
and social philosophy in particular will play an impor-
tant part in establishing them. Among the factual
premisses there will be empirical or scientific judg-
ments, for example, about the capacities, needs, and
responses of different groups of children, or about the
effectiveness of different sorts of teacher training.
