You are on page 1of 6

THE GREAT

805
INTRODUCTION
proof? Is there both an inductive and a deduc-
tive type of syllogism, or is induction the very
opposite of all forms of reasoning and proof?
It is \vith these last questions that the dis-
cussion of induction begins in the great books,
especially in Aristotle's 01ganon and Bacon's
Novum Organun1, but also in the \vritings of
Descartes and Locke, and in observations on
scientific method by Newton, Harvey, and
Pascal. Though many of the controversies and
problems which become central in the nine-
teenth century do not appear explicitly in the
earlier tradition, they are anticipafed by the
fundamental distinctions and issues which can
be found in the earlier writers.
Bacon's dissatisfaction with Aristotle, for
example, leads him to formulate specific rules
for induction. Going further in the same gen...
eral direction, Mill later develops his elaborate
theory of inductive inference. \Ve move in the
opposite direction ifwe are guided by Aristotle's
distinction between scientific and dialectical in"!
duction and by his way of setting induction off
as the very opposite of reasoning. The question
then arises whether Bacon and Mill are treating
induction in all or in only one of several quite
distinct senses.
As THE CHAPTER on LOGIC indicates, the names
of Aristotle and Bacon are sometimes used as
the symbols of opposed tendencies in logic.
The one is supposed to represent an almost ex-
clusive emphasis on deduction, the other the
primacy and importance of induction. An op-
position between Aristotle and Bacon is also
implied in the current use of such phrases as
"inductive logic" and "deductive logic." These
phrases are sometimes used to suggest that the
inductive or the deductive process can be
favored to the exclusion, or at least the subor'"
dination, of the other. Such understanding of
Chapter 39: INDUCTION
the list of Additional Readings indicates,
the theory of induction falls within the
vince of logic and is part of the logician's
cern with the methods of inference or rea'"
ing employed in the sciences. The great
troversies about induction seem to be of
tively recent origin in the history of logic,
inning perhaps with the argument between
'Uiam WheweU and J. S. Mill over the con...
butions of reason and experience to the
uctive process. Later in the nineteenth
tury and in our own time, writers like John...
and Keynes, Russell and Nicod, who pre...
tdifferent formulations of inductive infer'"
e, call attention to the unsolved problems
h which any theory is left. They underline
assumptions that seem to be unavoidahle in
statement of the formal conditions which
IMate the so-called "inductive leap"-the
p from observed particulars to general
ths, truths having a wider generality than
eparticular .evidences from which they are
awn or on which they are based.
the problem of induction, in anyone's ver'"
11 of it, is the problem of generalization. This
yinvolve psychological questions about how
mind generalizes from experience. But
wever they are answered, the basic logical
estions remain substantially unaltered. By
at criteria is valid distinguished from fal...
ious induction? Can induction be secured
In error by rules of inference? Is induction
ispensable in the development of scientifie
owledge, or is there, as Whewell, for exam'"
, .. suggests, a sharp distinction between the
Uctive and the deductive sciences?
What is the relation of induction to dedue'"
n? Is it the relation of a method of discovery
a method of demonstration or proof? Is it a
Union between two modes of reasoning, both
which can be formulated as processes of
CLIFFORD. "The Unseen Universe," in
tures and ESfq.ys
ROYCE. The Co-hception of Immortality
FISKE. Life Everlasting
POHLE. Eschatology '"
SANTAYANA. Reason in Religion, .CH
HUGEL. Eternal Life
BOSANQUET. The Value and Destiny
vidual
VONIER. The Human Soul and Its Wtth ()t,
Spirits
BRADLEY. Essays on Truth
F ARNELL. Greek Hero Cults and Ideas
The Nature of Existence, CH
SCHELER. Vom Ewigen im A1enschen
LAKE. Imn10rtaZity and the Modern j\tJind
BROAD. The Mind and Its Place in Nature,
FRAZER. The Golden Bough, PART II, CH 2;
PART IV, BK I, CH 5-6; PART V, CH 16;
CH 10-1 I
--. Man, God, and Immortality, PART IV
J. S. HALDANE. The Sciences and
XVIII
WHITEHEAD. Process and Reality, PART V
KIRK. The Vision of God
HOCKING. Thoughts on Death and Life
A. E. TAYLOR. The Christian Elope
taZity
The [lope for Imrnortality
804
LEIBNITZ. Discourse on lvfetaphysics, XXXII-XXXVI
--. Mo1Uldology, par 19-28
J. BUTLER. The Analogy of Religion, PART I, CH I
LAW. An Appeal to All That Doubt . .. the Truths of
the Gospel, CH I
SWEDENBORG. Heaven and Its Wonders and Hell
VOLTAIRE. "Heaven," "Hell," "Hell (Descent into),"
"Purgatory," "Resurrection," in A Philosophical
Dictionary
MENDELSSOHN. Phadon; oder Ueber die Unsterblichkeit
der Seek
LESSING. How the Ancients Represented Death
WORDSWORTH. Intimations of Immortality
COUSIN. Lectures on the True, the Beautiful and the
Good, PART III (16)
HAZLITT. On the Feeling of Immortality in Youth
FEUERBACH. Gedanken iiber Tod und Unsterblichkeit
I. PICHTE. Die der PersonZichkeit und der
individuellen Fortdauer
J. H. NEWMAN. "The Immortality of the Soul," in
Parochial and Plain Sermons
KIERKEGAARD. Concluding Unscientific Postscript,
PP 152 - 158
SCHOPENHAUER. The World as Will and Idea, VOL
III, SUP, CH 41
--. "Immortality: A Dialogue," in Studies in Pes-
silnism
FECHNER. Lift After Death
_ ..-. Religion ofa Scientist
THE GREAT IDEAS CHAPTER 39: INDUCTION 806
the matter usually includes the popular notion
that induction is ahvays reasoning from par-
ticulars to universals deduction ahvSlYs
reasoning from universals to particulars.
But none of these things seems to be true, or
at least not without serious qualification. Nei-
ther Aristotle nor'
or induction to the exclusion of the other. On
the contrary, both appear to insist on the ab-
solute priority of induction, since, according to
them, it.provides deductive reasoning 'Vvith its
ultimate premises. Far from conflicting, induc-
tion and deduction complement each other.
"The consilience of, the results of both these
processes," Mill writes, "each corroborating
and verifying the other, is requisite to give to
any general proposition the, kind, .and degree
of evidence which constitutes scientific proof."
Until principles are established, the deduc-
tion of their implications or consequences can-
not begin. Unless principles, once they are ob
r
tained, are then used in the proof of other
truths, or are othervvise rationally employed,
the purpose of inductive generalization is not
fully realized. In this understanding of the re-
lationship between induction and reasoning,
Aristotle and Bacon do not seem to disagree,
nor does either of them conceive induction as a
process of reasoning from particulars to uni-
versals.
There is no question that the direction of
induction is from particulars; but in the precise
sense in which induction precedes deduction-
the sense in which both Bacon and Aristotle
regard it as the source of axioms-they do not
think it is a process of reasoning or a form of
proof. As for deduction, it is questionable, at
least for Aristotle, whether its direction can be
described as from the universal to the particu-
lar.
Aristotle seldom uses the word "deduction"
as the name for that phase of thought which is
complementary to induction. He speaks rather
of demonstration. Demonstration takes place
through the various forms of reasoning which
he calls "syllogisms." As the chapter on REA-
SONINGexplains, theseare collections ofpremises
each of which yields. a conclusion by valid in-
ference. In the most perfect forms of reasoning,
the conclusion is as universal as its premises,
and though there are syllogisms in which a par-
ticular proposition can be demonstratedft
universal and a particular premise, it isseI
the case that from exclusively universal pre
aparticular conclusi0t:c'an be validly drawn.
statement that deduct:I0n is reasoning from
versals to particulars certainly does not see
fit Aristotle's theory of the syllogism,.
less his conception of scientific demonstra
the aim of 'Vvhich is to prove universal,.
particular, propositions.
"WE LEARN EITHER by induction or by de
stration," Aristotle writes in the PriQrAnal
"Demonstration develops from universa
duction froln particulars." In the Posterior
lyticshe says that the ultimate premiseso
onstration must be primary or basic trut
basic truth is an immediate proposition--...,
is sometimes called a "first principle";g
"axiom." Since in his view "an immediate
osition is one which has. no other propOS!
prior to it," the basic premises .cannQt;
demonstrated.
Whence come these primary premisesi:w
are indispensable to demonstration. but
demonstration cannot establish? Aristo
answer is that "we know the primaryp.
ises by induction.',' In another place.
"it is by intuition that we obtain the pri
premises.' ,
The. word "intuition" indicates anesse1tl.
characteristic of the sort of induction, w
because it is not itself a form of ,reasonin
be prior to all reasoning and must be, in ord
supply the premises from which reasoning
ceeds. Reasoning is discursive. It is a proce
vorving steps. One proposition is drawn
another by the mediation of a third. Intui
in contrast, is immediate. Like an act of se
it apprehends its object at once and dire
When Aristotle speaks of induction as a ki
intuition, he implies, therefore, that it co
in the immediate grasp ofa universal truth.
proposition thus held he calls "immediate"
cisely because it can be known intuitively
in no other way. Intuitive
posed to what may be called "inductive re
ing," consists in seeing the universal in the
ticular. When what is seen is expressed in
form of a proposition,. the universal impliq
the known particulars is made explicit.
ml11CULUll and intuition are" however, .' not
for Aristotle.. In one passage in the
Analytics he syllogistic induc-
,ywhich can. hardly be called "intuitive."
in the Ethics, where hediscusses intuitive
he distinguishes between. two .sorts of
ary truth that can be known. by intuition.
IIltuitive. rCClson," he writes, "is concerned
the ultimates in both directions; for both
first terms and the last Clr:e objects of intui-
reason and nofofargument, and.: the intui;.
reason which is presupposed by
ions grasps the unchangeable . and . first
5, while the: intuitive reason involved in
tical reasoning grasps the last and var:iable
,.j.e., the minor premise. For thesevariahle
s;are the starting-points for the apprehen-
of the end, since the universals are reached
theparticulars; of these thereforewe must
and this perception is intuitive
" n.
his .applies to theoretic as well as practical
ledge. By intuitive reason, it seems, we
pboth the universal principlesoraxioms. and
articular factsiof sense:-perception.As per-
on is intuition on the part of the sensitive
ty, so induction is an intuitive u5eof the
(though Aristotle attributes both to
uitive reason").
hese two forms of intuition are functionally
ed. The induction of universal truths from
is impossible without sense-percep"
,>"for it is sense-perception alone which is
to grasp the particulars." But, according
ristotle, a single isolated. perception does
,give rise to an intuitive. induction.. Re-
d perceptions of things of a certain sort-
culars ofa certain class-are formed by
ory into what he calls "an experience."
use the experience refers, not to a. single
Midual, but to a class ofsimilar.individuals,
ovides the material for the mind's intuitive
f induction.
is theory of the role of experience in. in"
'on is more fully discussed in the chapter
XPERIENCE. For our present purposes, the
point is that the universal, lying in1plicitly
e experience, is ready, as it were, to be ex-
therefrom and made explicit. "Though
of sense-perception is of the particular,
is universal," Aristotle writes. With
807
the help of memory and experience, induction
makes the latent universal manifest.
BACON'S CRITICISIvL of the logic of Aristbtle
seems. to rest on two.. counts: first, he com-
plains ofAristQtle's over:-emphasis on syllogisms,
,vhether they are used dialectically.ordemon-
stratively; and second, he charges Aristotle with
a superficial understanding of induction. One
of the chiefefforts ofthe Novum Organum is:to
correct the latter mistake.
"There are andean exist,". says, Bacon, "but
two ways of investigating and discovering
truthe The one hurries on rapidly ,from .the
senses and particulars to the most general ax-
ioms, and. from them,as principles,. and .. from
their supposed indisputable truth, .Heduces the
intermediate axioms. This is the way: now, in
lise. The other constructs its axioms from the
senses and particulars, by ascending continually
and gradually, until it finally arrives, at the
most general axioms, which is the.truebut un-
attempted way."
Where Aristotle proposes that onlythe pri..
mary truthsor.first principlesbe established by
induction, while all the others .(whichBacon
calls "intermediate axioms") ,are to be derived
from them by.demonstration, Bacon .urges .a
method of induction which shall, mount grad-
ually froln the least general to the most. uni-
versal propositions. We should not .. "suffer the
understanding to jump and fly from particulars
to remote and most general axioms." We should
"proceed by a true scale and successive steps,
without interruption or breach, fromparticu-
lars to the lesser axioms, thence to the. inter..
mediate (rising one above the other), and lastly,
to the most general."
According to: this theory, induction can in..
tuitively draw more general from less general
truths, as well as the least general truths from
the particulars of perception. It might seem at
first as if there were no place for deduction in
the developme,nt.of science. But Bacon divides
the study of nature into two phases.: "the first
regards the eliciting or creating of axioms from
experiments, the second the deducing or deriv-
ing of new experiments from axioms." Here too
there seelns to be a crucial difference between
Baconand Aristotle. This difference is indicated
by Bacon's emphasis upon experiments both as
THE GREj-\T IDEAS CHAPTER 39: INDUCTION 808
the source of inductive generalization and also
as that which is ultimately derived by deduc-
tion from axioms.
The difference bet\veen experience (which
Aristotle makes the source of induction) and
experiment is more than verbal. "The axioms
now in use," Bacon contends, "are derived from
a scanty handful, as it were, ofexperience, and
a few particulars of frequent occurrence."
There has been too little attention given to neg-
ative instances, that is, of cases which seem to
run counter to the generalization being formed.
"In establishing any true axiom," Bacon in-
sists, "the negative instance is the most power-
fu1."
The chapter on EXPERIENCE dwells on the
difference between ordinary experience and
planned experiments. Where Aristotle seems to
be satisfied with the ordinary experience \vhich
arises from the perceptions of men in the course
of daily life, Bacon thinks it does not suffice.
Because it is haphazard, it fails to collect the
variety of instances, both positive andnega-
tive, upon which genuine and solid inductions
can be founded. Unusual and special experiences
must be sought out, and the effort must be
made to invent experiences \vhichdo not arise
spontaneously. For this, experiment--or. the
production ofexperienc-es-is necessary. Bacon
thinks we must, "by every kind of experiment,
elicit the discovery of causes and true axioms."
Two CONSEQUENCES FOLLOW from the several
differences we have noted between Aristotle's
and Bacon's theories of induction.
In the first place, Aristotle does not seem to
think that induction can be methodically pre-
scribed by logical rules. It is a natural act of
intelligence to draw universals from experi-
ence. Though men may differ in the readiness
of their native wit, the induction of the pri-
mary truths, which are the axioms or first
principles of science, does not require special
genius nor can it be improved or rendered
more certain by following rules. Precisely be-
cause it is intuitive rather than discursive,
induction, unlike reasoning, cannot be regu-
lated by rules of inference such as those which
govern the syllogism.
Without disagreeing that it is intuitive rath-
er than argumentative, Bacon seems to think
that induction requires the practice
detailed and precise method. Not only mus
various ascending stages of induction. be
lated by observance qf an order of genera
but the making of experiments and the co
tion and arrangement of particulars, "fo
tables and coordinations of instances," nIU
governed by a complex set of rules. The tW'e
seven tables of instances, set forth in the see
book of the NotJum Organum, constitute
heart of Bacon's method of induction. This
method "of discovering the sciences," he
serves, "levels men's wits and leaves but I'
of their superiority, since it achieves everyt
by the most certain rules."
In the second place, since genuine indue
depends for Bacon upon ample experiment
belongs primarily to the method of the ex
mental sciences-the physical or natural
ences in which experimentation is poss
Though the first principles or axioms ofa.
metic and geometry may be learned by 111
tion, the method of gradual ascent from exp
ments through intermediate generaliza.tt
does not apply to mathematics. Here we
have the beginning of the notion that
the experimental sciences are primarily In
tive, whereas other sciences, like
are primarily deductive.
But such a division of the sciences does
accord with Aristotle's theory of inducti
He thinks mathematics and metaphysics
quire induction for their foundation no
than physics and in no different way; iEa.
thing, induction is of the greatest importa.
for metaphysics, because all its principles
indemonstrable, whereas some of theprinci
needed in rnathematics and physics can bede
onstrated in metaphysics. Yet no science is
culiarly inductive, just as none stand-syi
special relation to experience. All depend eq
ly lipon experience for the induction of
primary truths on which their demonstrati
rest.
Descartes seems to fall some\vhere betwe
Aristotle and Bacon. He regards ari thmetica
geometry as more certain than the physi
sciences, because mathematics is largely
oped by deduction, whereas the study of nat
depends upon induction from experiments.
this lies the superiority of mathematics. "W
intlerenC(:=s fronl experience are frequently
" Descartes \vrites, "deduction, or the
illation of one thing from another ... can-
eerroneous when performed by an under-
ing that is in the least degree rationa!. "
vertheless, Descartes does not exclude in-
ion as the source of the. axioms of mathe-
or, for. that matter.,
excludes the kind of induction \vhich
suponexperiments. Such axioms as when
Is are taken from equals the remainders are
lor the whole is greater thananyof its parts
products of induction, as may be seen, he
ts out, from the fact that a child can be
ht these general truths only "by sho\ving
examples in particular cases. " Similarly,
metaphysical truth in the proposition I
; therefore, I exist cannot be learned by de-
tion or syllogistic reasoning. The axiOlll that
ink is to exist hasto be by induction
bIll the experience of the individual-that
ess he exists he cannot think. For our nlind
o constituted by nature that general propo-
()l1S are formed out of the knowledge ofpar-
lIlars."
OM THE FOREGOING we can gather that dif-
nt theories of induction may be, in large
t, theories about different kinds of induc-
n. Common to induction of every sort is the
Hon of the mind from particulars, appre-
ded by sense, to general proposi tions or uni-
sal notiolls. But the character of the induc-
n,or its conditions and method, may differ
ording to the precise character of its source:
whether it arises from ordinary
ce or from planned experiments; and (2)
ether it is based upon a single experiment or
on an enumeration of instances. There re-
ins the most radical distinction in type of
uction: (3) whether it is intuitiveor discur-
e..,-accomplished by an act of immediate in-
ht or by a process of reasoning from prenlises
aeonclusion.
these three divisions cross one another to
e extent. Descartes, for example, seelllS to
rd the complete enumeration of a series of
nected facts as a \vay of drawing a general
elusion about their connection. That he has
reasoning rather than intuitive in-
in lllind, \ve learn from his statenlent
809
that "by adequate enumeration or induction is
meant that method by \vhichwe attain surer
conclusions than by any other type of proof,
\\rith the exception of simple intuition."
Pascal seems to be making the same point
\vhen he says that "in all matters \\7hose proof
is by experiment and not by demonstration, no
can be made except by .the
general enumeration of all the parts and all the
different cases." Bacon, on the other hand,al-
ways thinks of induction as intuitivegenerali-
zation, and therefore maintains that "induction
\vhich proceeds by simple enumeration is puer-
ile, leads to uncertain conclusions, and is ex-
posed to danger from one contradictory in-
stance."
The elaborate procedure which Bacon pro-
poses for collating instances stresses, not com-
pleteness of enumeration, but an examination
of their relation to one another and, in the light
thereof, an interpretation of their significance.
Mill's four or five methods of induction bear a
close resemblance to Bacon's more nunlerous
tables of instances; but Mill's methods are at-
tempts to fornlulate the rules of inference for
inductive reasoning, whereas Bacon's rules are
rules, not of reasoning, but of tabulating the
particulars fronl \vhich intuitive generalizations
can be formed.
On Mill's vie\v of induction, it may be ques-
tioned whether induction from an exhaustive
enumeration is induction at all, for it seems to
result in a summary of the facts enumerated
rather than a generalization from particulars.
Where there is no inductive leap, there is no
induction. Where the inductive leap does occur,
however, it seems easier to understand it as an
intuitive act-a seeing of the universal in the
particular-rather than as a process of reason-
ing. Each of Mill's methods requires a rule of
inference which is itself a universal proposition.
His critics hav:e asked, ,Whence come these uni-
versal propositions about the relations of cause
and effect or about the order and uniformity of
nature? They point out that he cannot answer
that these propositions are themselves conclu-
sions of inductive reasoning without begging
the question.
SUCH CRITICISM of inductive reasoning does
not seem to apply to Aristotle's conception of
mHEGREAT .IDEAS 811
PAGE
812
taken for granted. ,A.ristotle offers this example
of dialectical induction: "Supposing the skilled
pilot,if the most effective, and like,vise the
skilled charioteer, then, in general, the skilled
man is the best. at his' particular' task." In its
rhetorical form, no more than asingle example
may be used, as ,when the orator :generalizes
that honesty is the best policy from the story
of a particular' individual' .,vho .was re-
"'larded for his virtue.
In both forms, the inductive generalization
is at best probable; and it is more ;or lessproba-
hIe according to the soundness of the supposi-
tions or the examples from which it originates
be tested only by extending the enumera-
tionpf particulars. Bpt ifan inducti()n is.merely
probable in the first place, it can only be made
more probable, itean never be made certain,
by multiplying cases ot: by increasing their vari-
ety.
Aristotle's theory of dialectical induction
thus seems to have a bearing on the probability
of inductIon from limited experiments (or fronl
a. single experiment whose perfection is not as-
sured) and bfinduction from: the frequency or
variety of observed instances. other point
to be noted is that Bacon's basic rule of gradual
ascent from particular cases through less gen-
eral to more general propositions seems to be
relevant to dialectical induction, but not, on
Aristotle's view, to that kind of 'induction
which produces the axioms or principles of
sCience.
CHAPTER, 39: INDUCTION
OUTLINE OF TOPICS
The theory of induction: generalization from particulars
lao Induction and intuition: their relation to reasoning or demonstration
lb. Inductive reasoning: the issue concerning inductive and deductive proof
on. Analogy, he says, "leads us to expect
any cause the same events, which we have
rved to result from similar causes. Where.
causes are entirely similar, the analogy is
ect, and the inference 'draWn' froIr!it is
rded as certain' and,conclusive.. ..... But
re the objects have not so exact a similarity,
analogy is less perfect, and the inference is
conclusive; though still it has some force, in
ortion to the degree ofsimilarity and 're-
'blance." The absence of perfect similarity
ume's second reason for the inconclusive-
or uncertainty ofinductive generalizations.
he contrary supposition-that one case can
erfectly representative of an infinite num-
of similar cases-may explain wHy Aristotle
s to think that induction is able to produce
primary ttuths or principles of science. wi th
rtitude which gives certainty to all the dem-
trations founded on these axioms. Another
lanation of Aristotle's view. may be found
is distinction between scientific and dialec-
I induction. He regards the former as based
the .kind of common experience which, un-
even the best experiment, admits of no ex-
ions. In contrast, dialectical induction, or
still weaker form of induction which he
s"rhetorical," is based on an enumeration
ases (\vhich may not be complete) orupon a
Ie example (which provides no safeguard
inst possible exceptions).
nits'dialectical form, the inductive argu-
t proceeds from a number of particulars
2. The conditions or sources of induction: melnory, experience, experiment
3. The products of induction: definitions, axioms, principles, laws
The use of induction in argument
4a. Dialectical induction: securing,assumptions for disputation
4b. H.hetorical induction: inference from example in the process of persuasion 814
The role of induction in the developfllent of science:. the methods of experimental and
enumerative induction
stranger to them; but. yet, as j have said,
is but judgment and opinion, not know:l
and certainty. This way ofgetting and im}J.?:
ou?: knowledge in only by expeJf.
and history, which is all that the "veakness
faculties in this state of mediocrity .. ca
tain to, makes me suspect," Locke conch.!
"that natural philosophy is not capable oj'b
made a.science."
Hume offers two reasons for the inconclu
ness and uncertainty which he thinks
all our generalizations or inductions fr0111
perience. The first calls attention to the
that, unlike mathematical reasoning, infere
from experience in the realm of physical'
ters depend on the number of cases obser
"The conclusions which [reason] draws fr
considering one circle," he says, "are thesaltl
would form upon surveying all the circles in
universe. But no man,having seen only
body move, after being impelled by anot
could infer that every other body willltl
after a likeimpulse."
The principle "which. determineshiltl
form such a conclusion" is, according to Bu
"Custom or Habit"; and precisely because
ductive generalization is an effect of cust
rather than of reasoning in the strict sense,
strength of the induction-or the force of c
tom-varies with the number of cases fr
which it arises. "After the constant conjunct'
of two objects-heat and flame" for insta
weight and solidity-we are determined by
tom alone to expect the one from the app
ance of the other. hypothesis," Bu
maintains, "seems ... the only one which
plains the difficulty, why we draw, frC)
thousand instances, an inference which We'
not able to draw from one instance, that is in
respect different from them. Reason is '
capable of any such variation."
Since all the relevantcases can never be
haustively observed, the inference from a
tomary conjunction must always remain
certain, no matter how high a probability
derives from the multiplication of like
stances. To this first point, concerning the
pendence of the probability of generalizat'
from experience upon the frequency of the
served instances, Hume adds a second. p
about the similarity of the cases under ob
810
it, for with him it is not, as ,vith Mill, distinct
in form from the syllogism. It is simply a dis-
tinct type of syllogism, which consists in reason-
ing from effect to cause rather thanfrom'cause
to effect. Nor does the observation in-
ductive inference cannot be more thanproba-
hIe apply to what Aristotle means by an induc-
tive syllogism.
The certainty or probability of non-syllo-
gistic induction depends on the source of the
inference-whether it derives from a single
specially constructed experiment. ot from an
enumeration of particular instances, with' or
without a statistical calculation based on their
frequency. The conception of a perfect experi-
ment implies that the operation of a universal
lavv can be exhibited in a single case. It is almost
as if the controlling aimof the experiment .were
to make the universal manifest in the particular.
Newton's experiments on .reflection and re-
fraction seem to be of this sort. From them
certain laws of optics are directly induced, even
as, according to ,Aristotle and Descartes, the
axioms of mathematics or metaphysics
directly induced from simple experiences, avail-
able to a child or familiar to all men. Yet New-
ton does not think that the inductive establish-
ment ofsuch laws is as certain as demonstration.
The analytic method, he writes, "consists in
making experiments and observationsand in
drawing general conclusions from them by in-
duction. And although the arguing from exper-
iments and observations by induction, be no
demonstration of general conclusions; yet it is
the best way of arguing which the nature of
things admits of, and may be looked upon as so
much stronger, by how much the induction is
more general. If no exception occur from phe-
nomena, the conclusion may be pronounced
generally; but if at any time afterwards any ex-
ception shall occur fromexperin1ents, it may
then begin to be pronounced with such excep-
tions as occur."
Becauseit.lTIUst depend on inductive gener-
alizations from experience which, in .his view,
can never be certain, Locke'doubts that physics
can ever become a science. "1 deny not," he
writes, "that a man, accustomed to rational and
regular eXperilTIents, shall be able to see further
into the nature of bodies and guess righter at
their yet unknown properties, than one that is a
THE GREi\.T IDEAS
REFERENCES
813
42 KANT: Pure Reason, 5a-13d; 45b-46a; 66d-93c
/ Judgement, 562d-563b
45 FARADAY: Researches in Electricity, 659a
46 HEGEL: Philosophy of History, PART IV,
361a-b
54 FREUD: Instincts, 412a
3. The products of induction: definitions, ax-
ioms,principles, laws
8 ARISTOTLE: Prior Ana/ytics, BK II, CH 23 90a-c
/ Posterior Ana/ytics, BK I, CH 3 [72b25.....;30] 99c;
CH 18 111b-c; CH 31 [87b39-888I7] 120a-c; BK
II, CH 2 [90a24-30] 123b-c; CH 7 [92834__br]
126b; CH 19 136a-137a,c / Topics, BKI, GH 12
148d; CH IS [IOSb7-12] 152d; BK VIII, CH I
[I55b35-I5687] 211d-212a; [I56bIO-18] 212e.. d
/ Generation. and Corruption, BK I, CH 2 [3 I 6
a
5-S] 411c
9 ARISTOTLE: Ethics," BK I, CH 4 [I095a30_b8J
340c; CH 7 [Io9sa34-b3J 343d; UK VI, CH 3
[II39b25-34] 388c; CH 6 389d; CH II [1143825-
b13] 392d-393a
19 AQUINAS: Summa Theologica, PART I, Q 79,
A 12, ANS 425c-426b
28 GILBERT: Loadstone, PREF, 1a-c
28 GALILEO: Two New Sciences, THIRD DAY,
200a-b; 207d-208a
28 HARVEY: On Animal Generation, 333d-335a
esp 334c-d
30 BACON: Advancement ofLearning, 34b; 57b-d;
96d-97a /Novum Organum, BK I, APH 17-25
10Ba-d; APH 103=-106 127d-128c; BK II, APH
1-10 137a-140d
31 DESCARTES: Objections and 123a-b;
167c-d
34 NEWTON: Principles, BK III, RULE III-IV 270b-
271b / Optics, BK III, 543a-b
35 LOCKE: Human Understanding, BK I, CH III,
SECT 25 120c-d
35 BERKELEY: Human Knowledge, SECT 107
433d-434a
35 HUME: Human Understanding, SECT I, DIV 2
451b-c; DIV 9 454c-455a; SECT IV, DIV 26
460b-c
42 KANT: Pure Reason, 5a-13d; 66d-93c; 110a
43 MILL: Utilitarianism, 475b,d [fn I]
46 HEGEL: Philosophy of History, PART IV,
361a-b
51 TOLSTOY: War and Peace, EPILOGUE II, 690b
54 FREUD: Instincts, 412a-b
4. The use of induction in argument
4a. Dialectical induction: securing assump
tions for disputation
8 ARISTOTLE: Topics, BK I, CH 12 148d; CH IS
[I08b7-I2] 152d; BK VIII, CH I [15sbI6-I56a7]
211b-212a; [I56bIO-IS] 212c-d; CH 2 [rS7aI9-
38] 213b-d; CH 8 [I6083S-bl] 217d; CH 14 [1648
12-17] 222d
9 ARISTOTLE: Rhetoric, BK I, CH 2 [I3S6a36-b26]
596a-b
CHAPTER 39: INDUCTION
Understanding, SECT III, DIV
NEWTON: Principles, BK III, RULE IV 271b /
Optics, BK III, 543a-b
LoCKE: Human Understanding, BK IV, CH XII,
.EcT 10 361b-e
flUME: Human Understanding, SECT IV, DIV
.z6460b-e; SECT IX, DIV 82 487b-e; SECT XII,
DIV 131- 132 508d-509d passim
:K3ANT: Pure Reason, 45b-46a
MILL: Utilitarianisnz, 445a-447b passim;
47Sb,d [fn I] passim
ARADAY: Researches in Electricity, 659a
tOLSTOY: W-ar and Peace, EPILOGUE II, 690b
JAMES: Psychology, 674a-675b esp 675b
e. conditions or sources of induction:
memory, experience, experiment
RISTOTLE: Prior Analytics, BK I, CH 30 [46a
18-28] 64a; BK II, CH 23 [68
b
IS'-29] 90b-e /
'Osterior Analytics, BK I, CH I [7181-S] 97a;
CH 3 [72b2S-33] 9ge'; CH 18 111b-e; CH 31
[87b39-8S817] 120a-e; BK II, CH 2 [9824-3]
23b-e; CH 7 [9
2B
34-
bl
] 126b; CH 19 [99
b20
-
IOObS] 136a-d / Topics, BK I, CH 12 148d; CH
8 [I08b7-12] 152d; BK II, CH 7 [113
a
31-33]
S8d; BK VIII, CH I [IS,Sb3S-IS687] 211d-212a;
IS6
b
IO-IS] 212e-d / Physics, BK I, CH I 259a-b;
HS [191824-34] 267a-b / Heavens, BK III, CH
[30686-181397b-c / Generqtion and Corrup-
"on, BK I, CH 2 [31685-8] 411c / Metaphysics,
K I, Cll I 499a-500b; CH 9 1992b24-99381]
511a-b
f\.R.ISTOTLE: Ethics, BK I, CH 4 [109S830-..b8]
CH 7 [I09S834-bS] 343d-344a; BKVI,
pH S [1142aI2-I9] 391b; CH II [II43832-b6]
92d-393a / Rhetoric, BK II, CH 20 [1393a22-
3948S] 640d-641d
;\QUINAS: Sumlna Theologica, PART I-II, Q51,
A I, ANS 12b-13e
GILBERT: Loadstone, PREF, 1a-e
GALILEO: Two New Sciences, THIRD DAY,
200a-b; 207d-208a
HARVEY: Motion of the Heart, 267b,d-268d;
285e-d / Circulation ofthe Blood, 322d-323d;
/ On Animal Generation, 331b-335e;
383.d; 473a
ACON: Advancement of Learning, 16a;34b;
7b-d; 59c / Novum Organum, BK I, APH 17
08a; APH 19108b; APH 22 108e; APH 25 108d;
69 116a-b; APH 14-105 128a-c;BK II,
APH II-IS 140d-149a
J:)ESCARTES: Rules, II, 2d-3a; VII, 10e-12a /
Discourse, PART VI, 61d-62c / Objections and
Replies, 167c-d
$pINOZA: Ethics, PART II, PROP 40, SCHOL 1-2
87b-388b
ASCAL: Vacuum, 358a-b
Principles, BK III, RULE III-IV 270b-
/ Optics, BK III, 543a-b
Human Knowledge, SECT 107
31 SPINOZA: Ethics, PART II, PROP
388a-b
33 PASCAL: Pensees, 1-5 171a-173a
35 HUME: Human Understanding,
451 b-e; DIV 9 454e-455a; SECT
460b-e
lb. Inductive reasoning: the issue
inductive and deductive proof
8 ARISTOTLE: Prior Analytics, BK II, CH
/ Posterior Analyties, BK I, CH I
CH 3 [72b2S-3.3] 9ge; CH IS 111b-c;
[92a34-,bl] 126b / Topics, BK I, CH
152d / Physics, BK VIII, CH I
9 ARISTOTLE: Ethics, BK I, CH 4
340e; CH 7 [Io9sa35-b3] 343d; BK
388b-c / Rhetoric, BK I, CH 2
596a-b
28 GALILEO:
252a-b
28 HARVEY: Motion ofthe
30 BACON: Advancement of
57b-58b; 61d; 96d-97a / NOVU111
105a-195d esp BK I, APH
APH 69 116a-b, APH 13-106 1"-11
APH 10-52 140e-195d
31 DESCARTES: Rules, II, 2d-3a; VII,
17b-18b; XII, 24a-b / Discourse,
62c / Objections and Replies,
To find the passages cited, use the numbers in heavy type, which are the volume
numbers the passages referred to. 4 HOMER: Iliad, BKH [26S-2S3]
number 4 IS the number of the volume In the set; 'the number 12d indicates that
sage is in section d of page 12.
PAGE SECTIONS: When the text is in one column, the letters a andb
upp.era.ndlower halves of the page. Forexan1ple, in53 JAMES : Psychology, 116a-119b, P,lSS::l9"P
begIns In the upper half of page 116 and ends in the lo\ver half of page 119. When
printed in two columns, the letters a and b refer to the upper and lower halves of the
hand side of the page, the letters c and d to the upper and lower halves of the
the page. For example, in 7 PLATO: Symposium, 163b-164c, the passage begins
of the left-hand side of page 163 and ends in the upper half of the right-hand
AUTHOR'S DIVISIONS: One or more of the main divisions of a work (such as PART,
SECT) are sometimes included in the reference; line numbers, in brackets, are given in
tain cases; e.g., Iliad, BK II [265-2S3] 12d.
BIBLE REFERENCES : The references are to book, chapter, and verse. When the King
and Douay versions differ in titleof books or in the numbering of chapters or verses, the
James version is cited first and the Douay, indicated by a (D), follows; e.g., OLD
MENT: Nehemiah, 7:45-(D) II Esdras, 7:46.
SYMBOLS: The abbreviation "esp" calls the reader's attention to one or more eSf)eCIalJlv
relevant parts of a whole reference; "passim" signifies that the topic is discussed ......... ......
tendy rather than continuously in the work or passage cited.
For additional infonnation concerning the style of the references, see the Explanation
Reference Style; for general guidance in the use of The Great Ideas, consult the
812
1. The theory of induction: generalization from
particulars
1a. Induction and intuition: their relation to
reasoning or demonstration
8 ARISTOTLE: Posterior Analytics, BK I, CH 3
[72bIS-24] 99b-c; CH 23 [84b3I-SSal] 115d-
116a; CH 31 [S7b39-ssaI7] 120a-c; CH 33 [SSb
30-S9aI] 121b-c; BK II, CH 2 [90R24-30]
123b-c; CH 19 136a-137a,e / Topics, BK I,
CH 12 148d; BK VIII, CH I [I5Sb35-I56a7]
211d-212a; [IS6
b
IO-I8] 212e-d / Metaphysics,
BK I, CH 9 [992b24-993aI] 511a-b
9 ARISTOTLE: Ethics, BK I, CH 7 [I09sa34-b3]
343d; BK VI, CH .3 388b-c; CH 6 389d; CH S
[II42a23-3I] 391b-c; CH II 392e-393b esp
[I I43a32-b6] 392d-393a / Rhetoric, BK II, CH 20
[I393a25-26] 641a
28 HARVEY: On Animal Generation, 332a-335c
esp 333d-334d
30 BACON: Advancement of Learning, 43d-44c;
5ge; 61d; 96d-97a / Not/urn Organum, 105a-
195d esp BK I, APH 11--26 107d-108d, APH 69
116a-b, APH 105 128b-e, BK II, APH I'-10 137a-
140d, APH 15-16 149a-b, APH 20-22 150d-
153c, APH 52 194c-195d
31 DESCARTES: Rules, III 3b-5a esp 4a-b; VII,
10e-12a; IX 14d-15d; XI, 17bd / Objections
and Replies, 123a-b; 167c-d
81S CHAPTER 39: INDUCTION
ADDI.TIONAL READINGS
I.
II.
ILL. A System of Logic, BK III-IV
Listed below are \vorks not included in Great Books ofthe Western .W.or1d, .but relevant to the
idea and topics \vith \vhich this chapter deals. These \vorks are diVided into two groups:
I. Works by authors represented in this collection.
II. Works by authors not represented in this collection.
For the date, place, and other facts concerning the publication of the works cited, consult
the Bibliography of A.dditional Readings which follows the last chapter of The Great Ideas.
PEARSON. The Grammar of Science, CH 3
BOSANQUET. Logic, VOL II, CH 3-5
--a Science and .. 4
J. C. WILSON. Statement and Inference, PART IV
POINCARE. Science and Method, BK II, CH 3-4
CASSIRER. Substance and Function, PART II, CH 5
N. R. CAMPBELL. Physics; the Elements, CH 4
W.E. JOHNSON. Logic, PART II, CH 8-11; PART III,
CH 2
J. M. I(EYNES., A Treatise on Probability, PART
III
NICOD. "The Logical Problem of Induction," in
Foundations of Geometry and. Induction
Probability, Statistics, an.d Truth
M.R. COHEN. Reason and Nature, BK I, CH 3(3)
JEFFREYS. Scientific Inference
MEYERSON. Du cheminement de la pensee
DEWEY. Logic, the Theory of Inquiry,cH 21
B. RUSSELL. The Problems of Philosophy, CH 6
-.-. Introduction to Mathematical Philosophy,
CH 3
--. Human Knowledge, Its Scope and Limits,
PART V, CH 7; PART VI, cn 2-3
DEMUS. On Methods of Inference
OF SAINT THOMAS. Cursus Philosophicus Tho-
icus, Ars Logica, PART I, BK III, CH 2; Q 8
PART II, Q 5
ELY. Elements of Logic, BK IV,CH I
Y. Logic, PART IV
ELL. The Philosophy ofthe Inductive Sciences,
II, BK XI, CH 1-,6; BK XIII
On the Philosophy ofDiscovery
RT. Logic, PART III, CH 5
The Principles ofScience, CH 7, I I; BK IV, esp
2
. Logic, BK I, CH 3 (A)
'EY. The Principles ofLogic, BKII, PART II, CH 3
EIRCE. Collected Papers, VOL II, par 619-644,
755/9
1
. Principles of E1npirical or Inductive Logic,
-IS, 17, 24
28 GILBERT: Loadstone, PREF, la-c
28 GALILEO: Two Net-lJ Sciences,
200a-b; 207d-208a
28 HARVEY: Motion of the Heart, 280c'
/ Circulation ofJhe Blood, 324c-d /'
mal Generation, 332a-335c esp 334c-d'
473a '
30 BACON: Advancenlent of Learning, loa;
42a-c; 56b-58c; 96d-97a / Novum(;)/!j
10Sa-195d esp BK I, APH 11-26:
APH 69 116a-b, APH 14-106 128a-c
APH 1-:10 137a-140d, APH 15-16 149a
20-21 150d;.153b, APH 52. 194c-195d
31 DESCARTES: Rules, II, 2d-3a; VII 10
Discourse, PART VI, 61d-62c /Objectz
Replies, 167c-d
33 PASCAL: Vacuum, 358a-h / Arithmetz
angle, 451 b-452a; 458b-459b; 464a-4
34 NEWTON: Principles, BK III, RULE IB-I
271b; GENERAL SCHOL, 371b-372a /
BK III, 543a-b
35 BERKELEY: Human Knowledge, SEO'f
433d-434a
35 HUME: Human Understanding, SECT I,
454c-455a; SECT III, DIV
DIV 26 460b-c
42 KANT: Pure ReasQn, 5a-13d; 45b-46ar;
85d esp 72c-74b, 82a-b / Intro.:Metaeh,
Morals, 387a-b / fudgement,,562d:-563
43 MILL: Utilitarianism, 445d-446b;
[fn I]
45, FARADAY: Researches in Electricity, 65
51 TOLSTOY: War and Peace, .EPILO
690b
53 JAMES: Psychology, 385a-b; 617b;8p
54 FREUD: Instincts, 412a-b
THE,' GREAT IDEAS 814
(4. The use of induction in argument.)
4b. Rhetorical induction: inferen.ce from exam-
pie in the process of persuasion
8 ARISTOTLE: Prior Analytics, BK II, CH24 90c-
91a / Topics, BK VIII, CH I
CH 8 [I60a35-bI] 217d / Metaphysics, BK II,
CH 3 [995a6-8] 513c
9 ARISTOTLE: Rhetoric,BK I, CH:2 [I356a36-
135
8a
3] 596a-597d; CH9 [I368a29-31] 611b-c;
BK II, CH 20 640d:-641d; CH 23 [I398a32-bI8]
646d-647a; eH 25 BK III,
CH 17[1417b35-'-1418a3] 672b
30 BACON: Advancement of Learning, 58c-59a
42 KANT: Prej. Metaphysical Elements' of Ethics,
376c-d
'S.The role of iriduction in the development of
science: the methods of experimental
and enumerative induction
8 ARISTOTLE: Prior Analytics,BK II, CH 23 90a-c
/Posterior Analytics, BK I, CH 3 [72b25-33] 99c;
CH 18 Illb-c; BK II, CH 19. 136a-137a,c /
Physics, BK I, CH 2. [I85aI3-14] 259d; BK V,
CH I [224b28'-30] 304d; BK VIII, CHI [252.
a
23-
b
5
] 336a-b / Generation and Corruption, Bl{ I,
CH 2 [316a5-I4] 411c-d / Metaphysics, BK I,
CH 9 [992b30-993aI] 511b; BK VI, CH I [I025
b
1-16] 547b; BK XI, CH 7 [1064
a
4-9] 592b;
BK XIII, CH 4 [I078b28-30] GlOb
9 ARISTOTLE: Generation.ofAnimals, BK V,
CH 8 [788bI0-2I] '330e / Ethics, ;BK' I,' CH 4
[I095a3o_b8] 340c; CH 7 [1098a34-b3] 343d;
BK VI, CH.3
10 GALEN: Natural Faculties, BK III, ,CH 1-2
199a-200a .esp 199c-d
For: Other discussions of induction as an intuitive act of generalization, see JUDGMENT 8a; KN
EDGE 6C(2); PRINCIPLE 3a(I),3a(3)-3b; REASONING Sb( I); SCIENCE Sd.
Other treatments of inductive reasoning and its relation to deductive reasoning,
ING 4c, 6c; SCIENCE Sd; and for parallel distinctions in modes of argument, see Ex;
ENCE 2d; REASONING Sb(3), Sb(S)'
Discussions dealing with the sources or conditions of induction, generalization, or ails
tion, see EXPERIENCE 2b; IDEA 2g;, MEMORY AND IMAGINATION 3c, 6C(I); SENSE
UNIVERSAL AND PARTICULAR 4c.
Induction as the source ofprinciples, axioms, or scientific laws, see PRINCIPLE 3b; SCI
4d; UNIVERSAL AND PARTICULAR 4.
Other treatments of dialectical and rhetorical induction, see DIALECTIC 2b, 3b; RH
4C( I); and for their contrast with dialectical and rhetorical reasoning, see DIALEC
REASONING sc-Sd; RHETORIC 4C(2).
The role of induction in the experimental sciences, see EXPERIENCE sa; REASON:IN
SCIENCE 4d , 5d.
CROSS-REFERENCES

You might also like