(Teach Yourself Books) A. A. Luce, Arthur Aston Luce - Teach Yourself Logic To Think More Clearly-Association Press (1959)
(Teach Yourself Books) A. A. Luce, Arthur Aston Luce - Teach Yourself Logic To Think More Clearly-Association Press (1959)
(Teach Yourself Books) A. A. Luce, Arthur Aston Luce - Teach Yourself Logic To Think More Clearly-Association Press (1959)
SCHOOL OF THEOLOGY
_ AT CLAREMONT
by
A. A. LUCE
M.C., D.D., Litt.D.
Trinity College, Dublin
Theology Library
SCHOOL OF TH EOL_OGY
AT We) REMONT
lifornia
II. TERMS : . ° . . . 12
III. PROPOSITIONS . . . . ° ° 34
Subject, Predicate, and Copula. Propositions and
Sentences. Setting out the Proposition in Logical
Form. Symbols. Quality and Equipollence.
Quantity of Propositions. The Four Types. Dis-
tribution of Terms. Cofnplex and Compound Cate-
gorical Propositions. Hypothetical Propositions.
Disjunctive Propositions. Modal Propositions.
THE SYLLOGISM . . . . 83
Its Structure and Technical terms. The General
Rules. The Figures. The Special Rules. The
Moods.
CONTENTS
CHAP, PAGE
REDUCTION AND PRINCIPLES OF REASONING . Iol
INDEX 205
CHAPTER I
Historical
Questions on Chapter I
1. What does Logic study ? What are the practical aims of
the discipline ? What benefits may be expected from a course
of Logic P
2. Why is it called Formal Logic?
3. What is formalism ?
4. ‘My dear Watson, the conclusion you draw about the
case from yout own observations is va/id, but it is not frue.’
If you found that remark in Sherlock Holmes’ Memoirs, what
inference would you draw about his education? What would
the remark mean?
5. What does Logic owe to Aristotle ? Give an account
of his life and writings? What was his Organon?
6. What is Natural Logic?
CHAPTER II
TERMS
Discourse begins when something is ‘ proposed ’, that is,
set forth in terms, about something or someone. The
Proposition is the unit of discourse, and the constituents
of propositions are Terms. Propositions lead on to
Inference. Connected Propositions yield syllogisms and
other forms of inference. Terms, Propositions, and
-Syllogisms are the main elements of Aristotelian Formal
Logic. We shall treat them in that order, but some
anticipation here and there is unavoidable.
In a proposition of the simplest type there are two
terms, the Subject and the Predicate. ‘ Silence is golden.’
That is a proposition, believed by the speaker to be true,
or at least proposed for acceptance in discourse as true.
Silence is its Subject ; golden is its Predicate ; golden is predi-
cated (affirmed) of Silence. ‘The logical subject corresponds
roughly to the grammatical subject ; it is that about which
the statement is made. The logical predicate has no close
equivalent in grammar; but it has one sure mark; it is
that which is affirmed or denied about the subject. Main
verbs (not verbs in relative clauses, etc.) belong to the
predicate, and sometimes constitute it. The Copula (is,
are, etc.) is discussed below (p. 34).
There are no terms, properly speaking, apart from
propositions. The word wolf by itself, said or written,
is not, strictly, a term; it proposes nothing. In ‘ The
wolf is coming ’, wo/fis aterm. Of course, wolf by itself
and other nouns and adjectives in isolation are often called
12
TERMS 13
terms, because they could be used as terms. Potentially
they are terms ; they cou/d be used as subject or predicate
of a proposition, even when they are not actually being
so used. Particles, prepositions, conjunctions, etc.—+the,
a, in, of, but, when, etc.—are not terms in the ordinary way.
We cannot go far in discourse without them; but we do
not speak of them as a rule, nor predicate them of subjects.
A term may consist of one word, or of several words
linked. Each of the following words or groups of words
is one term, when subject or predicate of a proposition,
or considered as such—wolf, big bad wolf, man, the
human race, the sons of Adam, the daughters of Eve,
ships that pass in the night, Shakespeare, the Swan of
Avon, the last word of the finest sonnet of the Swan of
Avon. A term in its verbal aspect is defined as, A word or
combination of words which can stand by itself as the subject
or predicate of a proposition.
But terms have another aspect which is expressed by,
and is to some extent represented by, the word or com-
bination of words. The real wolf bites; the word wolf
does not bite, though it expresses and to some extent
represents the wolf that bites. A term im its real aspect is
defined as, Whatever we can think of and speak of as the
subject or predicate of a proposition.
This double meaning of the word erm is accepted in life
and sanctioned by usage, and we must accept it in Logic.
It has its advantages and its disadvantages, and the former
outweigh the latter. The context usually makes it quite
clear which meaning is intended; for instance, when we
speak of Concrete Terms as persons or things, we are
thinking and speaking of real terms. When we style
Socrates a Concrete Term, we do not mean the eight-lettered
name, Socrates, but the real Socrates who had a snub-nose
14 LOGIC
and who drank the hemlock cheerfully. Some logicians
try to get round the difficulty by distinguishing verbal
terms as ‘names’ and real terms as ‘concepts’. This
distinction eases the situation to some extent, but on the
whole it raises more problems than it solves. It is better
to keep the one word ¢erm with its two meanings, or two
shades of meaning, and interpret it according to its context.
The word ¢erm comes from the Latin serminus, a limit
or boundary; for the terms of a proposition limit the
movement of the thought, as Euston and Paddington
limit the movement of the trains.
Definition
Definition is a difficult art, but a necessary duty in honest
discourse, especially when unfamiliar or ambiguous terms
1 The question is of minor importance. ‘Those who wish to pursue
it should read J. S. Mill, Logic I ii 5, and, for different views, H. W. B.
Joseph, op. cit., p. 150 ff. (2nd ed.).
- 8.Bosanquet, The Essentials of Logic, Lect. V.
28 LOGIC
Description
Description is a partial statement of the connotation of
the term, sufficient to distinguish it from other terms. To
the logician, a description is an imperfect definition, but
to the man of letters description is a fine art and a difficult
art. Individuals can be described, but not, strictly, defined.
‘A gillaroo is a fish ’—that description goes a very short
TERMS 31
way; but it might be better than nothing. ‘A gillaroo
is a trout ’—that is better again. ‘Pleasure is a test of
a formed habit "—that is a good Aristotelian description,
not far from.a definition.
Division
Logical Division is like analysis in the experimental
sciences ; it breaks up the material into its component
parts. Division is the complementary process to defini-
tion, dealing with extension and denotation, as definition
deals with connotation. Division is much like Classifica-
tion; but in Division we start with the whole and work
down to the parts, while in Classification we usually start
with the parts and work up to the whole. For instance,
a treatise on the racial characters of man- would have to
begin by dividing mankind into Aryans, Semites, Magyars,
etc. The main divisions would be sub-divided, and the
sub-divisions perhaps sub-divided. The resulting table
could be called ‘a Division of Man’, or ‘a Classification
of men’. Division is mainly concerned with the orderly
atrangement of class and species; but class and species
can be divided into the individuals composing them.
Individuals themselves are not capable of logical division ;
hence their name.
There are three main Rules of Logical Division :
(1) It must be adequate. ‘The parts must together be
equal to the whole, and no part must be overlooked.
Example:
The contents of Logic are divided into terms, proposi-
tions, and syllogisms. In dividing natural species, etc.,
it is often hard to ensure that no part has been omitted.
(z) It must be distinct. There must be no overlapping
32 LOGIC
ot blurred outlines. Europeans are correctly divided into
British, Irish, French, Germans, Dutch, Swedish, etc., but
not into British, Scottish, Irish, French, Germans, Dutch,
Swedish, Scandinavians, etc. The logic of the shop-sign
‘ Ladies, Gents, and Clerical Tailoring ’ is open to criticism.
(3) There must be only one Principle of Division. The
Principle of Division is the basis on which the division is
made. If we divide mankind by race or by colour, then
race or colour is the Principle of Division. Whatever
Principle we adopt, we must stick to it, as far as possible.
We must not divide mankind into Aryans, Semites, white
men, redskins, intellectuals, workers, ancients, and
moderns. Books may be divided into fiction, history,
travel, poetry, biography, etc., but not into fiction, octavo,
quarto, manuscript, printed, bound and unbound. It is
not always possible or desirable to keep strictly to this
rule; and if the two Principles are very close or con-
gruous, a cross-division will do no great harm.
Dichotomy
Dichotomy is a form of division, suggested by Plato
and criticized by Aristotle. The word means ‘ cutting
into two parts’. Dichotomy divides a genus into two
parts, one of which possesses the differentia and the other
not. ‘Thus animal would dichotomize into rational and
irrational. Rational beings would dichotomize into mortal
and immortal. The division is necessarily exhaustive, and
the species necessarily exclude one another. The method
is of some use, and we still divide into vertebrates and in-
vertebrates; but the members of the class zuvertebrates have,
most of them, little incommon. The historic dichotomy,
known as Arbor Porphyriana, is shown below (p. 134).
TERMS 33
The Copula
The Copula is the hinge of the proposition ; on it turns
the relation between subject and predicate. The copula
is the connecting link. Silence is silence, and golden is
golden; but if you wish to express the golden character
of silence, you must connect the two words by the copula.
34
PROPOSITIONS 35
Equipollent Propositions
Propositions that differ in quality, but have the same
meaning, the same subject, and contradictory predicates,
are called Equipollent. Such are:
Conversation is a pleasant pastime.
Conversation is not an unpleasant pastime.
Singular Propositions
Singular propositions have a singular term for subject.
Examples :
Socrates taught Plato.
Socrates did not teach Aristotle.
As stated above they rank as universals for most purposes
(A or E). Some logicians place them in a class apart,
neither universal nor particular. On balance the present
arrangement is to be preferred. Four types are easier to
manage than six, and singular propositions are like univer-
sal propositions in that their subjects are wholes, not
aggregates.
Enumerative Propositions
DISTRIBUTION OF TERMS
Non-Categorical Propositions
Hypothetical and Disjunctive Propositions make use of
categorical predications, but are not themselves categori-
cal; they connect two or more categorical predications
in various ways, but do not themselves predicate anything
of a subject categorically. The Hypothetical conjoins the
truth of one predication to the truth of another. The
Disjunctive disjoins the truth of one or more predications
from the truth of one or more predications. Some Hypo-
theticals can be represented by Categoricals, and Dis-
junctives can be represented by Hypotheticals; but in
both cases the representation is inadequate. Hypotheticals
and Disjunctives are distinct and important ways of making
statements. They are in everyday use, and play a con-
siderable part in law and science, and from time im-
memorial they have been treated as part of Logic.
Hypothetical Propositions
The Hypothetical Proposition states by supposition. If
my opponent opens P-K4, says the chess player to him-
PROPOSITIONS age
Disjunctive Propositions
A Disjunctive Proposition affirms alternatives. It, too,
is affirmative as a whole, though its constituent clauses may
be affirmative or negative. It is a useful form for setting
out the constituents of a complex situation without com-
mitting yourself to a premature decision. The farmer
finds no fruit on his tree. He cannot state the cause for
certain ; but he can set out the possibilities, and that may
help him another year. Either the frost nipped the
blossom, or the caterpillar ate the leaf, or the worm
devoured the fruit. He states the possibilities disjunc-
tively, i.e. as alternatives. In theory there is no limit to
the number of alternatives and the clauses that express
them ; but for simplicity’s sake we will confine ourselves
to two.
The terms of the two predications may be disposed in
three different ways, and there are three corresponding
types of Disjunctive Propositions.
(1) The subject of both is the same and the predicates
differ. A is either B or C. He is either a fool or a
knave.
PROPOSITIONS 59
Modal Propositions
In Modal Propositions the predication is qualified by
words such as, may be, must be, is (emphatic), possible,
probable, impossible, necessary. Such words bring the
speaker’s mind into the proposition, and express his
attitude to what he is saying. All propositions are Modal
to some slight extent; for we cannot keep ourselves
entirely out of what we are saying. The farmer almost
succeeded in doing so, who, when congratulated by a
townsman on his fine field of turnips, replied, ‘ Them’s
mangolds.’? He did not mean, they may be, might be,
must be, mangolds. He voiced what is called ‘ the objec-
tive fact’. His statement was all about the crop and not
about his attitude to it; it was factual, not modal.
Modal propositions are of three main types, Assertoric,
Problematic, and Apodeictic. Assertoric Propositions
predicate confidently, usually after doubt; e.g. ‘ Yes,
after all; they are mangolds.’ A Problematic Proposition
62 LOGIC
Subalternation
Contradiction
If I is true, E is false.
If O is true, A is false.
If A is false, O is true.
If E is false, I is true.
If I is false, E is true.
If O is false, A is true.
Contradiction in practice
Contrariety
Of two Contrary Propositions the one denies in its
whole extent what the other affirms. They cannot both
be true, but they may both be false. They must be
universal and of different quality, and therefore, in symbols,
they are A and E. Examples:
CONTRARY PROPOSITIONS
Subcontrariety
Of two Subcontrary Propositions the one affirms par-
ticularly what the other denies particularly. They may
both be true; they cannot both be false. Being both
particular and differing in quality, they are, in symbols,
I and O. Examples:
SUBCONTRARY PROPOSITIONS
SUBALTERNS SUBALTERNS
I SUBCONTRARIES Oo
Obversion
Obvertend Obverse
All S is P. No S is not-P.
No S is P. All Sis not-P.
Some S is P. Some § is not not-P.
Some §S is not P. Some S is not-P.
All men are reasonable. No men are unreasonable.
No tigers are merciful. All tigers are merciless.
IMMEDIATE INFERENCE 75
Obvertend Obverse
Some rebukes are deserved. Some rebukes ate not un-
deserved.
Some conifers are not decidu- Some conifers are not-decidu-
ous. ous, or
Some conifers keep their
leaves in winter.
All textbooks on logic are not Some textbooks on logic are
suitable for beginners. unsuitable for beginners.
No women live on Mt. 1 The entire population of
Athos. Mt. Athos is male.
No men are admitted. All men are excluded.
Some tongues wag too freely. Some people do not curb
their tongues.
Some ills are not wholly evil. 1It’s an ill wind that blows
no one any good.
Conversion
Contraposition
Contraposition consists in first obverting, and then con-
verting the obverse. The chief use of the process is to
alter the quality of a negative convertend; it transfers
the negation to the predicate, and thus enables the con-
verse to admit an undistributed predicate. In this way
it does for O propositions what conversion proper can-
not do.
THE SYLLOGISM
1 The two rules for the distribution of the predicate (above, p. 50)
play a large part in the proofs of these Rules, and the student should
refer to them,
THE SYLLOGISM 89
This rule does not require a term to have the same quantity
in premiss and conclusion. There is nothing against a
term, distributed in its premiss, being undistributed in.
the conclusion, except that in such cases one is generally
entitled to conclude about a//; and if one is entitled to.
conclude about a//, it is usually (but not always) pointless.
to conclude about some.
Corollary. From Rules rand 2, taken together, it follows.
that There must be at least one more distributed term in the
premisses than in the conclusion. For a term distributed
in the conclusion must have been distributed in its premiss,
and in addition the premisses contain the middle term, at:
least once distributed.
The next three rules concern Quality.
(3) From two negative premisses nothing follows. If one
man says, ‘ A is not B’, and another adds, ‘B is not C ‘4
no advance is registered. Nothing has been affirmed.
A and C have been excluded from B; but that does not
include A in C, or exclude A from C. The data do not
relate A and C positively or negatively. The universality
of this rule has been challenged from time to time, and
there are notable discussions of it in the Port Royal Logic:
and in Keynes’ Logic.1 ‘The gist of these discussions is
that where cases occur of two negative premisses, appar-
ently yielding a valid conclusion, either one of the pre-
misses is a disguised affirmation, or there are four terms,
or in some other way the inference is non-syllogistic.
There ate no good grounds for doubting the universality
of Rule 3.
(4) From two affirmative premisses a negative conclusion
cannot follow. In other words, if the conclusion is nega-
tive, one of the premisses must be negative. For when.
1jJ. N. Keynes, Formal Logic, 4th ed., p. 295.
90 LOGIC
both premisses are affirmative, the extremes include or
are included in the middle term; but that can give no
ground for asserting that the one extreme excludes the
other.
(5) If either premiss is negative, the conclusion is negative.
In this case the one premiss is affirmative, and the other
negative; that is to say, the one extreme includes, or is
included in, the middle term, and the other extreme
excludes, or is excluded from the middle term. If any-
thing follows, it must be the exclusion of the one extreme
from the other, not the inclusion of the one extreme in
the other; it must be the denial of relation between
them, not the affirmation of relation between them.
The next (and last) two rules concern Quantity.
(6) From two particular premisses nothing follows. Sup-
pose it possible that there could be two particular pre-
misses. These must be II, IO, OI, or OO. II contains
no distributed term and is excluded by Rule 1. IO and
OI contain only one distributed term apiece, and are
excluded because by Rule 5 the conclusions of both must
be negative, and their predicates distributed, and therefore
by the Corollary of Rules 1 and 2 vo distributed terms
apiece in the premisses would be required. OO is ex-
cluded by Rule 3.
(7) If either premiss is particular, the conclusion is par-
ticular. First case. Both premisses are affirmative. These
must be A and I, which between them have only one
distributed term, and ¢wo would be required by the above
Corollary, if the conclusion were universal. The con-
clusion therefore is particular. Second case. One of the
premisses is negative. The pairs must be either AO or EI.
Each of these pairs contains only two distributed terms,
and since the conclusions are negative, ‘ree would be
THE SYLLOGISM 91
S P Ss P Se Ss P
Dk
SP
(1) If the major premiss is affirmative, the minor
premiss is universal.
(2) If the minor premiss is affirmative, the conclusion
is particular.
(3) If a premiss is negative, the major premiss is
universal.
Proofs. (1) If the major is affirmative, its predicate,
the middle term, is undistributed. The middle term in
THE SYLLOGISM 95
The first step is to find its figure and mood and mood
name. It is in the second figure; its mood is EAE;
its mood name is Cesare. The name Cesare tells us that
REDUCTION AND PRINCIPLES OF REASONING 103
All P is M.
All S is P.
.. All S is M.
This conclusion contradicts the O premiss, given true,
and is therefore false. All P is M is given true. ‘There-
fore the other premiss, All S is P, is false. Therefore its
contradictory, Some S is not P, is true. O.E.D. Bocardo
is reduced in a similar way.
I will now take the moods seriatim, and will indicate in
symbols how each is to be reduced.
Second Figure
Cesare No P is M s No M is P
to All S is M AllS is M
Celatrent No S is P No S is P
s No S is P
Festino No P is M s NoM is P
to Some S is M Some S is M
Ferio Some S is not P Some S is not P
Third Figure
Some S is P
Felapton No M is P No M is P
to All Mis S p Some S is M
Ferio Some Sis not P Some S is not P
Ferison No M is P No M is P
to Some MisS_ 5s Some S is M
Ferio Some S is not P Some S is not P
Fourth Figure
Bramantip All P is M te All M is S
by All M is S All P is M
Barbara Some S is P All PisS p Some Sis P
Camenes All P is M Se No M is S
to No M is S All P is M
Celatent No S is P No Pis S 5 No Sis P
Fesapo No P is M s No M is P
to AllMisS p SomeSisM
Ferio Some Sis not P Some S is not P
Datisi Darit
All men are mortal. All men ate mortal.
Some men are philosophers. | Some philosophers are men.
Some philosophers are .. Some philosophers are
mortal. mortal.
Aristotle’s Dictum
The Predicables
Living
Sensible Insensible
ae
Animal
Rational Irrational
Pane
Rational Animal
Pe
Mortal Immortal
ae
Man
THE CATEGORIES AND PREDICABLES 135
IV Vv
Enthymeme?
An Enthymeme is a syllogism with one premiss or the
conclusion suppressed. It is in fact an abbreviated syllo-
gism, such as those that occur in actual discourse of in
writing. To express all three propositions would be
pedantry in most contexts, and bad policy. The art of
convincing often lies in hiding the fact that you are trying
to convince. Hearers or readers, confronted with the para-
phernalia of syllogistic argument, are on their guard, and
the combative instinct is aroused. Partly for policy, and
partly from laziness, if we want to ‘ get a syllogism across’,
as a rule we utter no more of it than is necessary, and we
leave the rest to the imagination. Any two of the three con-
stituent propositions are usually enough to suggest the third.
A premiss, major or minor, is left unsaid because, as a
tule, it says itself. The suppression of the conclusion is
not so common. When done, it is done for effect, and
it can be a most effective device. A suppressed conclusion
is like a sword suspended by a hair, not falling, but ever
ready to fall. Innuendos are often conveyed in this way,
and pointed thrusts which words would blunt. Examples:
Major premiss suppressed. Afforestation schemes remove
land from cultivation, and are therefore of little benefit
to the neighbourhood. The suppressed major is: All
schemes that remove land from cultivation are of little
benefit to the neighbourhood.
1 Aristotle used the term in quité a different sense. He says “an
enthymeme is a syllogism starting from probabilities or signs”.
Analytica Priora, Il, 27, 70a.
142
SYLLOGISTIC ARGUMENTS 143
Polysyllogisms. Sorites?
Several syllogisms, joined together like links in a chain,
the conclusion of one forming a premiss for another,
make a Polysyllogism. St. Paul’s argument in Romans viii
29-30 is in the nature of a polysyllogism, “ For whom he
did foreknow, he also did predestinate’. . Moreover
whom he did predestinate, them he also called ; and whom
he called, them he also justified; and whom he justified,
them he also glorified.”
The most interesting type of polysyllogism is the
Sorites ; the word comes from the Greek owpd¢, a heap.
The Sorites is a heaped argument; it is defined as A
polysyllogism in the first figure, with the intermediate con-
clusions suppressed, and supplied mentally as minor pre-
misses with the following major premiss. It has the form:
Sorites Suppressed Conclusions
Yas
Bais, G SESE.
C is D alrwarkiog
Dis E sa Ny tel S)
E is F
sree Ip 7]
1 Three syllables; pronounced Saw-ri-tees.
144 LOGIC
The minor premiss is written first, then its major
ptemiss; each suppressed conclusion becomes in turn
minor premiss to the subsequent major premiss. The
predicate of each premiss is subject of the next, and the
subject of the first premiss, and the predicate of the last
premiss are, respectively, subject and predicate of the final
conclusion. ‘The first and second premisses are minor and
major, respectively, of the first syllogism ; its conclusion
(suppressed) and the third premiss are minor and major,
respectively, of the second syllogism; and so on, till the
conclusion of the Sorites is reached.
The Sorites has two rules:
The first premiss alone can be particular.
The last premiss alone can be negative.
To prove these rules break the Sorites up into its com-
ponent syllogisms. ‘The first premiss is minor of the first
syllogism, and each suppressed conclusion acts in turn as
minor of the next following syllogism. If the first premiss
is particular, the intermediate and final conclusions will be
particular; but if any premiss, other than the first, were
particular, there would of necessity be an undistributed
middle in the syllogism that contained it; for each
expressed premiss after the first acts as major, having as
subject the middle term, which is required to be distributed
there, since it is undistributed in the corresponding minor,
which is affirmative and has the middle term as predicate.
Therefore the first premiss alone can be particular.
The last premiss can be negative, and if it is negative,
the conclusion of the whole Sorites is negative. No other
premiss can be negative; for if it were, there would be
an Illicit Process of the major term in the syllogism, next
following. All the intermediate conclusions, as we have
SYLLOGISTIC ARGUMENTS 145
Hypothetical Syllogisms?®
MODUS TOLLENS
If A is B, it is C. If A is B, X is Y.
A is not C. X is not Y.
.. It is not B. ~. A is not B.
The Dilemma
If either A is B or C is D, E is F.
But either A is B or C is D.
ul Dis Gea
COMPLEX CONSTRUCTIVE
If A is B, E is F, and if C is D, G is H.
But either A is B or C is D.
.. Either E is F or G is H.
If A is B, C is D and E is F.
But either C is not D or E is not F.
~, A is not B.
If A is B, either C is D or E is F.
But neither is C D, nor is E F.
.. A is not B.
156 LOGIC
If bodies can move, they must move either where they are or
where they are not.
Bodies cannot move either where they are or where they are not.
”. Bodies cannot move:?
COMPLEX DESTRUCTIVE
If A is B, E is F, and if C is D, G is H.
But either E is not F, or G is not H.
~, Either A is not B, or C is not D.
Analogy
A brief discussion of Analogy may serve to link deduc-
tive Logic to inductive; for analogy is an inference from
a resemblance between particular things, events, or occa-
sions to a further resemblance. Peisistratus, a political
leader at Athens, asked for a bodyguard, and by means
of it made himself tyrant. On this analogy, said Aristotle,
it might have been inferred that when Dionysius of
Syracuse asked for a bodyguard, he was aiming at supreme
power.
158 LOGIC
Analogy always contains a pinch of uncertainty, and is
therefore distinct from logical inference. _Good democrats
have been known to ask for a bodyguard. The analogy
may hold, or it may not. Yet in spite of the uncertainty
analogy and logical inference have much in common ;
they both build on the genus, the kind, the repeated
instance. ‘‘ Probability is the very guide of life ”,* wrote
Bishop Butler, and analogy is probability, systematically
used, consciously directed, and woven into experience.
I am driving a car in traffic; an urchin darts across my
driving-field, nimbly threads his risky way, and reaches
the safety of the pavement. Though he is safe, instinc-
tively I slow down; for where there is one urchin, there
is probably another. Arguing on analogy I expect him,
too, to dart across ; for he, too, argues on analogy uncon-
sciously, and seeing his chum make the crossing safely
expects a similar success.?
Likeness and repetition—these are the foundations of
analogy. Where they are not, there is no analogy ; where
they are found, there is a basis, however slight, for analogi-
cal inference. We expect the sun to rise tomorrow,
because we think of tomorrow on the analogy of yesterday
andtoday. We notice similarities in the changing altitudes
of the sun, in the waxing and waning of the light, and in
our own sensations of freshness and fatigue; when we lie
down to rest, on analogy we expect to wake up next
morning refreshed, and to find the sun arising or arisen.
Analogy attends and uses belief in the uniformity of nature
and in natural law. Because objects or the relations
1 J. Butler (1692-1752), The Analogy of Religion, Introduction, 1736.
2 The probability which comes close to Analogy and in practice
guides a large part of human life must be distinguished from the
Probability which is confined to the numerical and calculable aspects
of events and is treated in the mathematical Theory of Probability.
SYLLOGISTIC ARGUMENTS 159
Fallacies
Men err, and most men err in similar ways. The stock
errors in logic to which the human mind is prone have
long been named and listed. They are known as Fallacies,
ice. Deceptive Arguments. To be made aware of them,
to know their names and just why they are fallacies, is
part of the discipline of logic. They are not the only
fallacies; but they are the chief ones. Every rule in
logic can be broken, and from that angle there are just
as many fallacies as there are rules.
In syllogistic reasoning the three main Fallacies are:
(1) Undistributed Middle. Example:
All followers of Izaak Walton love virtue and angling.
Jones loves virtue and angling.
“, He is a follower of Izaak Walton.
The middle term ‘love virtue and angling’ is undis-
tributed in both premisses ; therefore the conclusion does
not follow from the premisses, and the syllogism is
fallacious.
People who judge by general impression and not by
rule fall readily into this fallacy. In the above instance
Jones clearly is a follower of Izaak Walton in the ordinary
meaning of that term. None the less that proposition
cannot be inferred from the given premisses, and for all
Lye S. Mill; op er#., Ble. IU, esscx}75°! cf. Bk. TP iis;
FALLACIES 161
we (as logicians) know to the contrary, there may be
a ‘splinter’ group of virtuous anglers who owe no
allegiance to the Compleat Angler and its author.
(2) Idicit Process. ‘This fallacy consists in taking uni-
versally in the conclusion a term given particularly in its
ptemiss; that is, in arguing from a term given undis-
tributed to the same term distributed. Illicit Process.
may be of the major, or of the minor. Examples:
Aristotle’s Induction
that of Mr. Smith, and the others, and found that ‘ All
these defective chimneys were boiler chimneys, and liable
to internal damp.’
Clearly then the mind of man does generalize along the
lines indicated by the Inductive Syllogism, even though the
completeness of the enumeration is an ideal, rarely realized.
But was Aristotle right in saying that we arrive at our first
principles by a process of Induction?! He was, provided
we do not take these words of his to mean that the enumer-
ation of instances is enough by itself. He was thinking
of the whole complex process of Induction, not of part.
It is not wise to trace our principles and generalizations,
be they homely or scientific, to any one source; some of
them are born of experience, some are dictates of reason;
some are bred in the bone; some are inferences; some
have a basis in vital instinct and some in intellectual intui-
tion. Aristotle was just the man to appreciate the mixed
character of our principles, and the varied sources of our
generalizations and premisses. He believed in reason, and
he believed in experience ; he was a realist about sensible
experience, and yet his mind moved easily in geometry
and the other deductive sciences. Now geometry sup-
plies the clue that explains his attitude to Induction; for
in geometry we never actually see the triangles and other
figures about which we reason so confidently. We see
very thin lines, possessing some breadth and a/most straight,
and triangles approximately isosceles; those approxima-
tions set us thinking about the ideal figures and properties
of space. Similarly Induction, as an observation and
enumeration of the individual facts of nature, was for
Aristotle a process that awoke thought, and ‘led on’ to
the general proposition. He wished Induction to be
1 See Analytica Posteriora, Tl, 19, 100b.
176 LOGIC
‘Perfect Induction’ resting on an ideally Complete
Enumeration; but he knew that in practice we have to
be content with less. Broadly, he viewed Inductive Logic
as a supplementary discipline, required by the limitations of
Syllogistic Logic, and he saw Induction as a systematic,
if imperfect, collection of scientific facts, as complete as
each case permits, to be followed by a leap of intuition
(Greek, votc), by which we see with the mind’s eye the
laws of nature, exemplified in particular instances, as the
geometer sees with his mind’s eye the laws of space
exemplified in the imperfect figures of space, which he
draws on paper or traces in the sand.
Bacon’s Criticism
1Jt is of some interest to note that a woollen fleece was used in the
earliest recorded experiment with dew (Judges vi ad fin.). To test
his ‘ call’ Gideon exposed a fleece of wool; next morning he wrung
a bowl of dew out of it, though the ground was dry. He tried the
same experiment next night, and the following morning the situation
was teversed ; the fleece was dry, and all the ground around it was
wet with dew. The ancient Semites, as dwellers in tents, were
naturally interested in the problem of dew, and the story of Gideon
looks as if they knew something about the Method of Difference.
2 Cf. the discovery of penicillin by the late Sir Alexander Fleming.
INDUCTION AND SCIENTIFIC METHOD 189
SYMBOLIC LOGIC
Historical
The Symbols
Symbolic Logic takes its name from its free and inde-
pendent use of symbols. Symbols have great educational
value; a good system of symbols constitutes a mental
shorthand which enables much to be said in a small space.
A symbol is something, e.g. a tally, that makes you think
of something else. Letters of the alphabet are symbols,
and so are words and figures ; but we are so accustomed
to them, and the mind passes so quickly from them to their
meaning that we scarcely think of them as symbols. The
word symbol today generally means an unusual sign, such
as an ideogram. Ordinary symbols, as we have seen, are
1 Tbid., Preface.
SYMBOLIC LOGIC 197
Punctuation
THEOLOGY LIBRARY
CLAREMONT, CALIF.
4432.
4 tog 4
ents homase oa
Je
", Ep!
th i isPt
.? v a it
i ‘ ;
vee: F ot i ~
é
het
Fe
=
~ 1
by
a . ; 2 a
7. ¥ ,
"Ty ae
it ‘
i ?
t
4 >
‘8 *
By
me
P =" ?
i
=
’
v
F a
vi,
_ -
> :
.
‘
.
~
4 —.
. «
aoe
he ‘
¢
=4 ‘ : i i]
YAR) : ¥ POIOINT ny i, ea 7 1S
ea Bu * we
AAD TROMIAALD
ORR
vost
i
~~ 7
i ‘
a
2 :
6
a a]
andl oe
a
iy
>
ay.
14982
eomneneeresese
ee eRe
sSetrresstreafis
oo
etsae ere.rs
ea
sae
aa 7 oe
— >
ese
Ter
Seeanos
S
se
ees
>: 5 pss
ae
atta
ee
vie tate ere
—_ ~
gent pester
tea
een
tar out
ae ne = nae a Oe OS Pree borers
cebpeetcettce
TSE
Selgeyrteeriet
cates
aee
ST
.o8 eevee iris ins]