GREEK PHILOSOPHY - II.
Aristotle's Theory of Knowledge
We begin the exposition of Aristotle's
philosophy with an account of what Aristotle himself regarded as introductory to
philosophy proper, viz., his theory of knowledge, of its
sources and method.
Kinds of Knowledge
There are in knowledge three fundamental differences that
Aristotle takes cognizance of in his theory of knowledge: differences as regards
the object, method, and source of knowledge. Knowledge may have for its object
causes (or first principles) or phenomena. Its method may be apodictic
(demonstrative) or dialectic ("probable"). Its source may be sense or reason.
Scientific, or Philosophical Knowledge
Scientific, or philosophical,
knowledge (έπιστήμη) is knowledge, the subject of which is causes
(άρχαί), the method, demonstration (άπόδειξις), and the source, reason (νοϋς).
In the knowledge of causes is involved the knowledge of whatever else can be
shown demonstratively to flow
from them ; and a theory of scientific knowledge is an account of the source
from, or faculty by, which we get the knowledge of causes, and of the method of demonstration. Now it was natural, both from the previous history of
speculation and the character of the problem itself, that Aristotle should
consider the latter part of the problem of scientific knowledge first, which he
does, particularly in the Prior Analytics.
Considered as regards method, knowledge is scientific, or
demonstrative, when it is
derived from certain, or necessary, premises by a certain, or necessary, process
The Syllogism (Deductive)
Now the process in
which, certain things being assumed as true, a certain other thing obtains,
necessarily and because of the things assumed, is called by Aristotle the
Syllogism(1). The syllogism is, therefore, the central point in the method of
demonstration expounded by Aristotle; it was regarded, and rightly so, as his
own discovery(2). The syllogism (συλλογισμός) consists of three propositions :
two premises (προτάσεις) and a conclusion (συμπέρασμα) ; and has three
terms (όροι), the major or larger term (μεϊζον
άκρον), the minor or smaller
term (έλαττον άκρον), and the middle (μέσον), which may be larger in compass
than either of the others or between them. The middle term is so called because
it is the mean, or uniting term, in the syllogism(3).—The members of the
syllogism are propositions. Propositions may be
either affirmative or negative, universal, particular, or indefinite. There are
four sorts of propositions that may enter into the syllogism : the universal
affirmative, the universal negative, the particular affirmative, the particular
negative. As regards the relations of these,—the universal affirmative and
particular negative are opposed as contradictories (άντιφατικώς
άντικείσθαι); so are the universal negative and particular affirmative. The
universal affirmative and the universal negative are contrarily opposed (έναντίως
άντικεϊσθαι). Of contradictory opposites, if one be true, the other is false:
contraries may both be false(4).—Now the rules of the syllogism given by
Aristotle are, that in every syllogism there must be one affirmative premise,
there must be one universal premise, and terms must not be treated as universal
in the conclusion which are not so in the premises(5).—Sylogisms differ in kind and
scientific value according to the relative compass and the position of the middle
term. A syllogism in which the minor term is "in the whole middle" (i.e., is the
subject of a proposition of which the middle is predicate) and the middle term
is "in the whole major" (i.e., is the subject of a proposition in which the major
is predicate) is termed a syllogism of the "first figure" (πρώτον
σχήμα). In a
syllogism of this figure the middle term lies "between" the extremes. A syllogism
in which both major and minor terms are "in [less than] the whole middle (i.e.,
are subjects of propositions in each of which the middle term is a predicate),
is a syllogism of
the second figure (δεύτερον σχήμα). A syllogism in which the major and minor
terms are each greater than the middle (i.e., are predicates of propositions in
each of which the middle term is subject) is a syllogism of the third figure
(The "fourth figure" of modern text-books was not recognized by
Aristotle; it did not spring out of his conception of the syllogism.)
Now the first figure is the only one that gives universal conclusions;
the second figure giving only negative conclusions, and the third only
"particular" conclusions. It is also the only figure that yields
naturally and directly in the conclusion all that is contained in the
premises and no more. We can sometimes derive a universal conclusion
from the premises of a syllogism of the second and third figures, but
this can be done only indirectly; hence these figures are "imperfect,"
the first alone being "perfect."(7)—The hypothetical syllogism (συλογισμός
έξ ύποθέσεως) is a syllogism in
which, there being a certain condition, a certain proposition
obtains. The latter, however, may obtain when the former does not ; but
if the latter does not obtain, the former does not(8).
Definition and the Predicables
Now the conclusion of a syllogism the
premises of which are true, states, if the conclusion be a universal affirmative
and is correctly drawn, a scientific truth, and is virtually the expression of
the essence (ούσία), or nature, of some real
existence ; hence is a definition.
The knowledge of the essence of a thing embraces a knowledge of the common and
characteristic attributes of the class, or genus, to which it belongs and of
the specific attribute that renders the thing an individual representative of
the class. In other words, the definition (όρος) contains the expression of the
union of the genus (γένος) and the differentia (διάφορα) of the thing defined.
Other attributes (not necessary to definition, however) are the property (ϊδιον),
which, though essential, is not a mark of distinction, and the accident
(συμβεβηκός), which may or may not belong to the subject defined.
Essence, or substance (ούσία), is one of the ten aspects,
according to Aristotle, under which things in general must be viewed. Substance
is whatever is the subject of attributes, e.g., man, Socrates ; and it is either
an individual, a species, or a genus. Substance in the primary sense is the
individual, species and genus being only secondary substance(10). The remaining nine
aspects are quantity (πόσος = how many?), quality (ποιός = of what kind?),
relation (πρός πι), place (ποϋ), time (ποτέ), position (κεϊσθαι),
condition (έχειν), action (ποιεϊν), passivity (πάσχειν). Of these ten—termed categories—substance is principal; all others imply it. These are
everywhere employed by Aristotle. (The idea of a table of categories may have
been suggested to him by the Pythagorean table (see
Phylosophy, Theories not purely Pythagorean).) These categories
were not "deduced" in any manner from a higher conception by Aristotle, but were
taken empirically, as suggested, perhaps, by the fundamental forms, or "parts,"
of speech in the Greek language.
Now causes (in the knowledge of which or of what can be
syllogistically shown to flow from them scientific knowledge consists), though visible to
the eye of reason, are not known to us immediately. Knowable things
are of two kinds: those which are prior for us (πρός
ύμάς πρότερον) and those
which are prior by nature (φύσει πρότερον).
Of the latter-named kind are
causes, or first principles(11). Our knowledge of causes, or what is prior by
nature, has its beginning in our knowledge of things as they are for us. The
(syllogistic) process by which we reach those firm universal propositions which
state the essence, or are the definitions, of causes is induction (έπαγωγή),
which is the " passing from particulars to universals," and is the inverse of
deduction, which is the passing from universals to particulars(12). Induction, like
deduction, is syllogistic: for in induction we unite by inference the middle
term to one of the extremes (major and minor terms) by means of the other. Thus,
if B is a "middle" to A and C, we can prove by means of C that A may be
predicated of B. For example, let the deductive syllogism be, "B is A, C is B, therefore
C is A"; then the inductive syllogism is, "C is A, C is B,
therefore B is A". The conclusion of the inductive syllogism corresponds to the
major premise of the deductive. The inductive syllogism is a syllogism of the
third figure, and strictly speaking, its conclusion is not universal
but particular. We may, however, assume it to be universal if we know that C
and B are inter-convertible and that "B is C" holds. (The syllogism is then practically a syllogism of the first figure.) A real induction presupposes a
knowledge of all the individuals of a class. "No particular kinds of Induction
are formulated by Aristotle, but he has noticed incidentally the principle of
most of the 'Experimental Methods', and in particular that of the method of
concomitant variations"(13). The premises of the inductive syllogism are not truths
of reason, corresponding to first principles, but perceptions of sense. But
sense as such gives knowledge only of the particular, and we can by induction
reach universals only on the hypothesis that there is a common and permanent
nature in the many. The idea of a common permanent nature originates in a higher
faculty than sense. Animals have the faculty of sense-perception, but not all
animals have the power of "retaining one certain thing in the soul" and of
forming universal notions. Permanency and universality presuppose reason.
Aristotle, however, allows himself to say that sense-perception introduces, or
"informs" (έμποιεϊ), the universal(14). (He compares the manner in which the
universal unconsciously grows out of the particular of sense to the way in which
soldiers in battle are caused to fly by the perception of one, and then another,
and so on, fleeing.) In so far, however, as there results from the inductive
syllogism something that is not given in sense-perception as such (and it is by
induction only that we reach the first universals that
are the foundation of science), induction does not prove (άποδείκνυσιν) anything (it
is not άπόδειξις) though it
does show (δηλοϊ) something.
Probable Proof ; Dialectical and Rhetorical Method
The foregoing is, in outline, Aristotle's account of scientific method as
employed particularly in speculative, or theoretical, philosophy. He recognizes
and gives a full analysis of another sort of method, which is only quiasi-scientific and
finds place especially in Practical Philosophy, —ethics, politics, etc. Here "dialectical," or probable, reasoning is employed.
In practical affairs it
generally suffices if we have premises that possess only a high degree of
probability, and if our conclusions have, not absolute validity, but a fair
warrant in the premises. In such matters it is not always easy or even possible
to arrive at absolutely correct definitions, and it is not always necessary that
all steps in our processes of reasoning should be stated, that everything should
be proved, even plausibly; indeed, it is better that many things be taken for
granted, that many things be left to the natural bent of mankind towards truth
and justice. This is the case particularly in rhetorical argumentation; in
dialectical reasoning logical method prevails though the premises may be only
plausible(15). In rhetorical reasonings the enthymeme, a quasi-syllogism, having
one premise, and example, by which we argue from a particular to a particular (through
assumed universal), may be employed instead of the complete syllogism and
(1) Prior Analytics, I. I. The reader cannot do
better here than to follow the account of Aristotle's logic given in
Outlines of the Philosophy of Aristotle.
(2) Soph. Elench, ch. 33.
(3) Prior Analytics, Bk. I. chs. 25, 1, 4, 5, 32, 26, 6 (see
Wallace's Outlines, pp. 37-39).
(4)On Interp. 6; Cat. 10; Pr. Analyt. I. ch. 2,
etc. (Wallace, pp. 29-31).
(5) Prior Analytics, Bk. I. ch. 24 (Wlllace, p. 40).
(6) Prior Analytics, Bk. I. chs. 5, 24, 26, 32, 56 (Wallace, pp.
(7) Ibid., Bk. I. chs. 1, 5, 7, 23, 24, etc. (Wallace, pp.
(8) Ibid., Bk. II. 4 (Wallace, pp. 41-42).
(9) Categories, 4, 7, 8; Topics, Bk.. I. 9; Metaphysics,
Bk. VI. 1, 7 (Wallace, pp. 25-27).
(10) Aristotle's doctrine of substance, as we shall see in what follows,
appears inconsistent or at least undeveloped (see Metaphysics,
Bk. VII. ch. 7). The above view is the earliest, and the one that seems
to harmonize best with the theory of categories, in which it occurs.
(11) Posterior Analytics, Bk. I ch. 2.
(12) Topics, Bk. I. 12, 18; Prior Analytics, Bk. II. 23 (Walllace,
(13) Wallace, p. 43; see Prior Analytics, Bk. II. ch. 23.
(14) See below, pp. 144-146.
(15) It is a part of Aristotle's catholicity of temper that he shows
some fondness for this quasi-scientific method. Absolute truth,
he repeatedly says, is not in all cases within the reach of human
powers, and it is often-times necessary and best to be content with less
than that. If we deny that this is in any sense a philosophical view, we
must throw away his works on Ethics and Rhetoric. See below, pp. 148 and