Logical Introduction To Probability and Induction Franz Huber Full Chapter PDF

Logical Introduction to Probability and
Induction Franz Huber

Visit to download the full and correct content document:
https://ebookmass.com/product/logical-introduction-to-probability-and-induction-franz-
huber/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Belief and Counterfactuals: A Study in Means-End

Philosophy Franz Huber
https://ebookmass.com/product/belief-and-counterfactuals-a-study-
in-means-end-philosophy-franz-huber/
Introduction to Probability and Statistics Metric

Edition 1925-2009) Mendenhall
https://ebookmass.com/product/introduction-to-probability-and-
statistics-metric-edition-1925-2009-mendenhall/
Introduction to Probability and Statistics (Metric

Version) 15th Edition William Mendenhall
statistics-metric-version-15th-edition-william-mendenhall/
Introduction to Probability Models 12th Edition Sheldon

M. Ross
https://ebookmass.com/product/introduction-to-probability-
models-12th-edition-sheldon-m-ross/
Introduction to Probability and Statistics - Metric
Version, 15e 15th Edition William Mendenhall
statistics-metric-version-15e-15th-edition-william-mendenhall/
Introduction to Linear Regression Analysis (Wiley

Series in Probability and Statistics) 6th Edition
Montgomery
https://ebookmass.com/product/introduction-to-linear-regression-
analysis-wiley-series-in-probability-and-statistics-6th-edition-
montgomery/
Gestión de los cuidados enfermeros y liderazgo (Spanish

Edition) Diane Huber [Huber
https://ebookmass.com/product/gestion-de-los-cuidados-enfermeros-
y-liderazgo-spanish-edition-diane-huber-huber/
Franz Kafka and Chinese Culture Yanbing Zeng
https://ebookmass.com/product/franz-kafka-and-chinese-culture-
yanbing-zeng/
Doing Philosophy: From Common Curiosity to Logical

Reasoning Timothy Williamson
https://ebookmass.com/product/doing-philosophy-from-common-
curiosity-to-logical-reasoning-timothy-williamson/
A LOGICAL
INTRODUCTION TO
PROBABILITY AND
INDUCTION
A LOGICAL
INTRODUCTION
TO PROBABILITY
AND INDUCTION
F R A N Z HU B E R University
of Toronto
1
1
Oxford University Press is a department of the University of Oxford. It furthers
the University’s objective of excellence in research, scholarship, and education
by publishing worldwide. Oxford is a registered trade mark of Oxford University
Press in the UK and in certain other countries.
Published in the United States of America by Oxford University Press

198 Madison Avenue, New York, NY 10016, United States of America.
© Oxford University Press 2019
All rights reserved. No part of this publication may be reproduced, stored in

a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by license or under terms agreed with the appropriate reproduction
rights organization. Inquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above.
You must not circulate this work in any other form

and you must impose this same condition on any acquirer.
CIP data is on file at the Library of Congress

ISBN 978–0–19–084538–4 (Pbk.)
ISBN 978–0–19–084539–1 (Hbk.)
9 8 7 6 5 4 3 2 1
Paperback printed by WebCom, Inc., Canada

Hardback printed by Bridgeport National Bindery, Inc., United States of America
The instructor’s manual can be found online at the following webaddress:

global.oup.com/us/companion.websites/9780190845384/
For password access please contact the author at franz.huber@utoronto.ca
CONTENTS
Preface ix
Acknowledgments xiii
1. Logic 1
1.1. Propositional Logic 1
1.2. Predicate Logic 7
1.3. Exercises 14
Readings 19
2. Set Theory 21
2.1. Elementary Postulates 21
2.2. Exercises 28
Readings 35
3. Induction 36
3.1. Confirmation and induction 36
3.2. The problem of induction 38
3.3. Hume’s argument 41
Readings 46
vi CONTENTS
4. Deductive Approaches to Confirmation 47

4.1. Analysis and explication 47
4.2. The ravens paradox 49
4.3. The prediction criterion 53
4.4. The logic of confirmation 55
4.5. The satisfaction criterion 62
4.6. Falsificationism 65
4.7. Hypothetico-deductive confirmation 69
4.8. Exercises 72
Readings 74
5. Probability 75
5.1. The probability calculus 75
5.2. Examples 81
5.3. Conditional probability 84
5.4. Elementary consequences 87
5.5. Probabilities on languages 92
5.6. Exercises 94
Readings 97
6. The Classical Interpretation of Probability 98

6.1. The principle of indifference 98
6.2. Bertrand’s paradox 100
6.3. The paradox of water and wine 106
Reading 110
7. The Logical Interpretation of Probability 111

7.1. State descriptions and structure descriptions 111
7.2. Absolute confirmation and incremental
confirmation 122
7.3. Carnap on Hempel 123
7.4. The justification of logic 126
CONTENTS vii
7.5. The new riddle of induction 132

7.6. Exercises 137
Readings 138
8. The Subjective Interpretation of Probability 139

8.1. Degrees of Belief 139
8.2. The Dutch Book Argument 140
8.3. The Gradational Accuracy Argument 147
8.4. Bayesian Confirmation Theory 155
8.5. Updating 168
8.6. Bayesian Decision Theory 173
8.7. Exercises 180
Readings 185
9. The Chance Interpretation of Probability 187

9.1. Chances 187
9.2. Probability in physics 190
9.3. The principal principle 195
Readings 203
10. The (Limiting) Relative Frequency

Interpretation of Probability 204
10.1. The justification of induction 204
10.2. The straight(-forward) rule 207
10.3. Random variables 215
10.4. Independent and identically distributed
random variables 217
10.5. The strong law of large numbers 222
10.6. Degrees of belief, chances, and relative
frequencies 227
10.7. Descriptive statistics 230
10.8. The central limit theorem 238
viii CONTENTS
10.9. Inferential statistics 243

10.10. Exercises 255
Readings 257
11. Alternative Approaches to Induction 259

11.1. Formal learning theory 259
11.2. Putnam’s argument 265
Readings 268
References 271
Index 283
P R E FA C E
A Logical Introduction to Probability and Induction is an

introduction to the mathematics of the probability calculus
and its applications in philosophy. On the mathematical side,
we will study those parts of propositional and predicate logic
as well as elementary set theory that we need to formulate
the probability calculus. On the philosophical side, we will
mainly be concerned with the so-called problem of induction
and its reception in the philosophy of science, where it is
often discussed under the heading of ‘confirmation theory.’ In
addition, we will consider various interpretations of probability.
These are philosophical accounts of the nature of probability
that interpret the mathematical structure that is the probability
calculus.
The book is divided into five sections. The first section,
Chapters 1–2, provides us with the relevant background in logic
and set theory. It will occupy the first two weeks.
The second section, Chapters 3–5, covers Hume’s argument
for the thesis that we cannot justify induction; Hempel’s work
on the logic of confirmation and the ravens paradox; Popper’s
falsificationism and hypothetico-deductive confirmation; as
x PREFACE
well as Kolmogorov’s axiomatization of the probability calculus.

It will occupy three to four weeks.
The third section, Chapters 6–8, covers the classical,
logical, and subjective interpretation of probability. Topics
include Carnap’s inductive logic and the distinction between
absolute and incremental confirmation; Goodman’s philosophy
of induction and the new riddle of induction; Haack’s dilemma
for deduction; the Dutch Book and gradational accuracy
arguments for the thesis that subjective degrees of belief ought
to obey the probability calculus; Bayesian confirmation theory;
update rules for subjective probabilities or probabilistic degrees
of belief; as well as Bayesian decision theory. It will occupy four
to five weeks.
The fourth section, Chapters 9–10, is devoted to the chance
and (limiting) relative frequency interpretation of probability.
Topics include probability in physics; Lewis’ principal principle
relating subjective probabilities and chances; Reichenbach’s
“straight-(forward) rule;” the strong law of large numbers
relating chances and (limiting) relative frequencies; descriptive
statistics and the distinction between singular and generic
variables; the central limit theorem relating sample means and
expected values; as well as estimation with confidence intervals
and the testing of statistical hypotheses. It will occupy four
weeks and contains a section on the interplay between the three
major interpretations of probability: subjective probabilities,
chances, and relative frequencies.
Along the way, we will come across probability puzzles
such as Bertrand’s paradox and the paradox of water and
wine, as well as paradoxes from logic and set theory such as
the liar paradox and Russell’s paradox. Sections 10.3.-10.9 are
centered around the strong law of large numbers and the central
limit theorem. They are mathematically advanced and can be
considered optional.
PREFACE xi
The last week of a course is usually best spent reviewing

material from previous weeks, and, perhaps, putting things
in perspective by mentioning alternative approaches. The final
section, Chapter 11, contains a suggestion.
The primary aim of this book is to equip students with the
ability to successfully carry out arguments, which is arguably
(sic!) the most important philosophical skill. Fifty exercises that
may be solved in groups rather than individually will help attain
this end. Another skill that is important in philosophy is the
ability to draw conceptual distinctions. Students are best asked
to explain some of the distinctions introduced in this textbook
jointly in the classroom, and individually, in the form of exam
questions similar to those listed in the instructor’s manual. The
latter also contains the solutions to the fifty exercises.
ACKNOWLEDGMENTS
I am very grateful to Claus Beisbart, Joseph Berkovitz, Michael

Miller, Jonathan Weisberg, and, especially, Alan Hájek, Rory
Harder, and Christopher Hitchcock for their helpful feedback
on an earlier draft of this book. Rory Harder has also created
the figures.
A LOGICAL
INTRODUCTION TO
PROBABILITY AND
INDUCTION
CHAPTER 1
Logic
1.1 PROPOSITIONAL LOGIC
The sentence “Angela Merkel is chancellor of Germany in August

2017” means, or expresses, the proposition that Angela Merkel
is chancellor of Germany in August 2017. Sentences come in
at least two forms: as abstract types and as concrete tokens.
Consider:
Toronto is a city. Toronto is a city. It is not the case that

Toronto is not a city.
Is there one sentence in the above line, or are there two

sentences, or even three? The correct answer depends on
whether we understand sentences as tokens or as types. There
are two sentence types and three sentence tokens in the line.
Both are different from the one proposition that these three
sentence tokens express, or mean, viz. that Toronto is a city.
Logicians are lazy. They use propositional variables, or
sentence letters, as place-holders (“variables”) for sentence
tokens. For instance:
p Angela Merkel is chancellor of Germany in August 2017.

q Toronto is a city.
r The Great Wall of China stretches from east to west.
.. ..
. .
2 INTRODUCTION TO PROBABILITY AND INDUCTION
Sentence tokens can be combined, or connected, to form more

complex sentence tokens by connectives:
Negation: ¬ it is not the case that . . .

(¬q) It is not the case that Toronto is a city.
Conjunction: ∧ (&, .) . . . and . . .
(q ∧ r) The Great Wall of China stretches from east to west,
and Toronto is a city.
Disjunction: ∨ (from the Latin word ‘vel’ that has the same
meaning as the combination of English words ‘and/or’) . . .
and/or . . .; . . . or . . . (or both)
(p ∨ q) The Great Wall of China stretches from east to west,
or Toronto is a city (or both).
Material Conditional: → (⊃) if . . ., then . . .; it is not the
case that . . ., or . . . (or both)
(q → r) If Toronto is a city, then the Great Wall of China
stretches from east to west.
Material Biconditional: ↔ . . . if and only if . . . (iff); . . . just
in case . . .
(q ↔ r) The Great Wall of China stretches from east to west
if and only if Toronto is a city.
Officially we define a formal language L for propositional logic

in the following recursive way:
1. Every propositional variable, or sentence letter, ‘p,’ ‘q,’

‘r,’ . . . is a sentence, or well-formed formula (wff), of L.
2. If α and β are sentences of L, then so are (¬α),

¬β , α ∧ β , α ∨ β , α → β ,and α ↔ β .
3. Nothing else is a sentence of L.
Speakers use languages to talk about things. Those things can

be the chancellor of Germany in August 2017, the weather, or
a language. When a language can be used to talk or write about
LOGIC 3
another language, the former has the latter as an object language.

In 1–3 above, the formal language L is our object language. The
language in which we talk or write about an object language is
a metalanguage for this object language. In 1–3 above, ordinary
English is our metalanguage for the object language L.
Here is another example of this distinction. As a matter of
fact, my English has a somewhat funny (German) accent, and
after class, students sometimes talk about it. In this case, the
language the students talk about, the object language, is my
English with the funny accent. In contrast to this, the language
in which the students talk, the metalanguage, is their perfect
English. Of course, as I am writing this, I use my English with
the funny accent, and one of the things I am writing about is the
students’ perfect English. Therefore, two languages can both be
metalanguages for each other.
Another distinction we need is that between use and
mention. The following two sentences are both true.
Toronto is a city.
‘Toronto’ consists of seven letters.
In the first sentence, we are using the word ‘Toronto.’ In the

second sentence, we are mentioning the word ‘Toronto.’ The
convention is to use left and right single quotes—“’ and ‘’,’
respectively—to indicate that one is mentioning rather than
using a symbol. Take a moment to reflect on what you do when
you introduce yourself to someone, as I do at the beginning of
the first class of a course. Determine which of the following
introductions are philosophically correct:
I am Franz. I am ‘Franz.’
My name is Franz. My name is ‘Franz.’
Now consider again the recursive definition above and, in

particular, the Greek letters ‘α’ and ‘β.’ These Greek letters
are symbols of the metalanguage that we use to talk about

the sentences of the object language L such as ‘(p ∧ q)’ and
‘¬r.’ These Greek letters are not part of the object language.
In contrast to this, the symbols ‘¬,’ ‘∧,’ ‘∨,’ ‘→,’ and ‘↔’ are
symbols of the object language L, and so are the left and right
parentheses, ‘( ’ and ‘ ),’ respectively. This means that the string

of symbols ‘ α ∧ β ’ contains both symbols that are, and symbols
that are not, part of the object language L. Strictly speaking, we
would have to write the following:
‘( ’α‘∧’β‘ )’ is a sentence of L.
However, as mentioned, logicians are lazy. They have introduced

the symbols ‘’ and ‘’ to put single quotes around every symbol
between them whenever single quotes are to be placed, and
not when not. These symbols are called (left and right) Quine
quotes—or, to be philosophically correct, these symbols are called
‘Quine quotes’—being named after the philosopher Quine who
introducedtheminQuine(1940).Nowthatwehavediscussedthe
distinction between use and mention we can ignore it again, as it
quickly becomes quite cumbersome to always use these quotes.
Before we move on to the next topic, let me note that
there are languages—such as the English which I am using to
write this book—that are so rich in expressive power that they
can be metalanguages for themselves. That is, we can use such
languages to talk in them about them. In fact, this is exactly what
I am doing in this paragraph! Such languages are given a special
name, viz. ‘self-referential languages.’ And, while they are great,
they also cause a lot of philosophical trouble. This is illustrated
by the following ‘liar sentence’:
L This sentence is false.
If the sentence L in the line above is true, then it is false, and

if it is false, then it is true. So L is true if and only if it is false.
LOGIC 5
Something has gone wrong. However, it is not the mere fact that
the sentence L speaks about itself that leads to this problem. In
the paragraph above, I use English to speak about itself, and no
problem arises. Similarly, the following sentence makes perfect
sense and is true:
This sentence contains five words.
Try to think about which feature, in addition to self-

referentiality, the sentence L possesses that may cause the
problem mentioned above.
What we have dealt with so far is the syntax of propositional
logic, which is concerned with the question which strings of
symbols are well-formed formulas, or sentences, of the formal
language L. In courses on logic, you will learn more about this
syntax, as well as how to derive, or prove, that some sentences
are theorems of propositional logic (this is sometimes called
proof theory). We will now deal with the semantics (or model
theory) of propositional logic, which studies what the sentences
of the formal language L mean, or express—that is, under
which conditions these sentences are true. A third field of
study that you can learn about in courses on the philosophy
of language is pragmatics. It studies how the way speakers
use a language affects the semantic meanings of syntactically
well-formed sentences.
We assume that every sentence is either true or false, but
not both, so that there are exactly two truth values: T, for truth,
and F, for falsehood.
Negation: A negated sentence (¬α) is true just in case the

sentence α is false. A negated sentence (¬α) is false just
in case the sentence α is true.

Conjunction: A conjunctive sentence α ∧ β is true just
in case both conjuncts α and β are true. A conjunctive

sentence α ∧ β is false just in case the conjunct α is false
or the conjunct β is false or both conjuncts α and β are false.

Disjunction: A disjunctive sentence α ∨ β is true just in
case the disjunct α is true or the disjunct β is true or both

disjuncts α and β are true. A disjunctive sentence α ∨ β
is false just in case both disjuncts α and β are false.

Material conditional: A material conditional α → β is
true just in case the antecedent α is false or the consequent

β is true or both. A material conditional α → β is false
just in case the antecedent α is true and the consequent β is
false.

Material biconditional: A material biconditional α ↔ β
is true just in case the sentence α and the sentence β have

the same truth value. A material biconditional α ↔ β
is false just in case the sentence α and the sentence β have
different truth values.
Unfortunately, the meaning of many English conditionals, or

if-thensentences,isnotcapturedbythematerialconditional.For
this reason, philosophers have come up with other conditional
connectives besides the material conditional. Among these
probably the most important one for philosophical purposes is
the counterfactual conditional which captures the meaning of ‘if’
in sentences such as ‘If things had been such and so, things would
have been thus and so.’ The antecedents, or if-clauses, of these
conditionals may involve a contrary-to-fact supposition (hence
the name ‘counterfactuals’). Can you think of a reason why a
contrary-to-fact supposition may cause trouble for the material
conditional?
The above truth conditions can be summarized by what
logicians call a truth table. Also, note that I have stopped using
single quotes when it became too cumbersome. Otherwise
I should have written ‘that logicians call a ‘truth table.”
LOGIC 7

α β (¬ α) (α ∧ β (α ∨ β (α → β (α ↔ β
T T F T T T T T T T T T T T T T
T F F T T F F T T F T F F T F F
F T T F F F T F T T F T T F F T
F F T F F F F F F F F T F F T F
1.2 PREDICATE LOGIC
Sentences talk about objects, the properties these objects have,

and the relations they stand in. Objects are referred to, denoted
by, or named by, names such as ‘Angela Merkel.’ Since logicians
are lazy, they use the shorter individual constants, which are
usually small letters from the beginning of the alphabet.
a Angela Merkel
b Toronto
c Montréal
.. ..
. .
Properties of one object are referred to by predicates, or

predicate symbols, and relations between two or more objects are
referred to by relation symbols. These are usually capital letters
from the middle of the alphabet.
F . . . is chancellor of Germany in August 2017

G . . . has more inhabitants than . . .
.. ..
. .
A predicate and an individual constant can be combined to

form a sentence, similarly for a binary relation symbol and
two individual constants (that may be two tokens of the same
type).
F (a) Angela Merkel is chancellor of Germany in August 2017.

G (b, c) Toronto has more inhabitants than Montréal.
It is customary to identify predicate symbols with unary

relation symbols, and propositional variables, or sentence
letters, with 0-ary relation symbols. This has the consequence
that propositional logic is included in predicate logic as a special
case.
Besides individual constants there are individual variables.
These are usually small letters from the end of the alphabet.
They make predicate logic both powerful and difficult.
x
y
..
.
I have not included a right column for individual variables

because they generally do not occur on their own. Instead
they are generally bound by the existential quantifier ‘∃x’ or the
universal quantifier ‘∀y.’
Existential quantifier: ∃x there exists an x

such that . . . x . . .; some x is such that . . . x . . .; at least one x
is such that . . . x . . .
Universal quantifier: ∀y every y is such that . . .
y . . .; all y are such that . . . y . . .; each y is such that . . . y . . .
When we translate English sentences into well-formed formulas

of predicate logic, it is often helpful to proceed in two steps.
In a first step, we move from the English language to the
regimented English language, which is a clumsy version of
the English language that contains no ambiguities and makes
LOGIC 9
the predicate-logical form of all sentences clear. For instance,

consider the English sentences:
There is a Republican U.S. president in August 2017.

All Canadian cities have at most as many inhabitants as
Toronto.
These two sentences from the English language are transformed

into the following two sentences from the regimented English
language:
There exists at least one object x such that: x is U.S. president

in August 2017 and x is Republican.
All things x are such that: If x is a Canadian city, then x has
at most as many inhabitants as Toronto.
In a second step, we can then transform the sentences from

the regimented English language into the formal language of
predicate logic:
∃x (P (x) ∧ R (x))
∀x (C (x) → M (x, b))
We now subsume the formal language for propositional logic

under the richer formal language for predicate logic, also called
‘L,’ which is defined recursively as follows.
1. If ‘t1 ,’ . . ., ‘tn ’ are n terms, that is, individual constants

or individual variables, and ‘R’ is an n-ary relation
symbol (which includes propositional variables, or
sentence letters, as the special case where n = 0), then
‘R (t1 , . . . , tn )’ is a well-formed formula of L. Specifically,
it is an atomic formula.
2. If α and β are well-formed formulas of L, and if ‘x’ is

an individual variable, then (¬α), ¬β , α ∧ β ,

α ∨ β , α → β , α ↔ β , as well as ∃x (α)
and ∀x (α) are also well-formed formulas of L.
Specifically, they are complex formulas.
3. Nothing else is a well-formed formula, or simply
formula, of L.
In contrast to the previous definition, the first clause now is

more general and includes the previous first clause as a special
case. The same is true for the second clause. Predicate logic thus
covers propositional logic as a special case.
The following is one version of how the truth conditions
for existentially and universally quantified formulas can be
defined. It is not standard, as it assumes that we have a name,
or individual constant, for each object. The standard semantics
does not make this assumption, but it is considerably more
complex (Shapiro 2013: sct. 4, Zach 2016: ch. 5). For our
purposes, the present version will do.
We use the notation ‘α [a/x]’—read: ‘a’ for ‘x’ in α—to
denote that every free occurrence, or token, of the individual
variable ‘x’ in the well-formed formula α has been replaced by
an occurrence, or token, of the individual constant ‘a.’
For instance, consider the well-formed formula ‘∃y (L (x, y))’
in which the individual variable ‘x’ occurs freely, but in which
the individual variable ‘y’ is bound by the quantifier ‘∃y’ and
so does not occur freely. ∃y (L (x, y)) [a/x] is ‘∃y (L (a, y)),’
because ‘x’ occurs freely in ‘∃y (L (x, y))’ and so is replaced

by ‘a.’ ∃y (L (x, y)) b/y is ‘∃y (L (x, y))’ because ‘y’ does not
occur freely in ‘∃y (L (x, y)),’ and so nothing is replaced. Finally,
∃y (L (x, y)) [c/z] is also ‘∃y (L (x, y))’ because ‘z’ does not occur
at all in ‘∃y (L (x, y)).’ Note that the ‘y’ next to the ‘∃’ does not
count as an occurrence of ‘y’ in ‘∃y (L (x, y)).’ Instead it is part of
the quantifier which is ‘∃y’ rather than ‘∃.’
LOGIC 11
Existential quantifier: An existentially quantified formula

∃x (α) is true just in case there is at least one individual
constant ‘a’ such that α [a/x] is true. An existentially
quantified formula ∃x (α) is false just in case all individual
constants ‘a’ are such that α [a/x] is false.
Universal quantifier: A universally quantified formula
∀x (α) is true just in case all individual constants ‘a’ are
such that α [a/x] is true. A universally quantified formula
∀x (α) is false just in case there is at least one individual
constant ‘a’ such that α [a/x] is false.
Now that we have defined the truth conditions, or meaning,

of the connectives and quantifiers, we can define the concepts
that make clear that logic is the study of the validity, or value, of
arguments. An argument consists of one or more premises to the
left of the therefore symbol ‘∴’ and a conclusion to its right. An
argument is logically valid if and only if the premises logically
imply the conclusion. An argument is logically sound if and only
if it is logically valid, and all its premises are true. Thus, the
conclusion of a logically sound argument is also true.
Logical truth: A formula α is logically true, |= α, just in case

α is true in all logically possible cases.
Below I will say more about what these logically possible cases
are. For now, a few examples will do. The sentence ‘Toronto is a
city or Toronto is not a city’ is logically true because it is true in
all logically possible cases: if Toronto is a city, and also if Toronto
is not a city. In symbols: |= (q ∨ (¬q)).
Logical consequence (special version): A formula α logically

implies a formula β, or β is a logical consequence of α, α |= β,
just in case β is true in all logically possible cases in which α
is true.
The sentence ‘Toronto is a city’ logically implies the sentence

‘Toronto is a city, or the Great Wall of China stretches from
east to west’ because the latter sentence is true in all logically
possible cases in which the former sentence is true: if Toronto
is a city and the Great Wall of China stretches from east to west,
and also if Toronto is a city and the Great Wall of China does not
stretch from east to west. In symbols: q |= (q ∨ r).
Since arguments generally contain more than one premise,
this definition needs to be generalized as follows:
Logical consequence (general version): Several formulas

α1 , α2 , . . . logically imply a formula β, α1 , α2 , . . . |= β, just
in case β is true in all logically possible cases in which all
formulas α1 , α2 , . . . are true.
The sentences ‘p,’ ‘q,’ and ‘((p ∧ q) → r)’ logically imply the
sentence ‘r’ because ‘r’ is true in all logically possible cases in
which ‘p’ and ‘q’ (and, hence, ‘(p ∧ q)’) as well as ‘(p ∧ q) → r’ are
true: that is, in the one logically possible case where all of ‘p’ and
‘q’ and ‘r’ are true. In symbols: p, q, ((p ∧ q) → r) |= r.
Logical equivalence: A formula α is logically equivalent to a

formula β just in case: β is a logical consequence of α, and α
is a logical consequence of β.
The sentence ‘Toronto is a city’ is logically equivalent to the

sentence ‘It is not the case that Toronto is not a city’ because
these two sentences are logical consequences of each other. In
symbols: q |= ¬¬q and ¬¬q |= q. Logical equivalence can also be
defined as follows:
Logical equivalence (variant): A formula α is logically

equivalent to a formula β just in case the sentence α ↔ β
is logically true.
LOGIC 13
That is, logically equivalent formulas have the same truth

value in all logically possible cases. This means that logically
equivalent sentences express, or mean, the same proposition.
Of course, these definitions say little if we do not specify
what the logically possible cases (that is, the models of model
theory) are. For now, the logically possible cases are the lines
in a truth table. A formula that is logically true with this
understanding of the logically possible cases is said to be
logically true in propositional logic. A formula that logically
implies, or is logically equivalent to, another formula with this
understanding of the logically possible cases is said to logically
imply, or to be logically equivalent to, the former formula in
propositional logic.
Every logical truth, logical implication, and logical
equivalence in propositional logic is also a logical truth,
logical implication, and logical equivalence in predicate logic,
respectively. The converse is not true, though, because
propositional logic, in contrast to predicate logic, does not
have any rules for quantifiers. It treats quantified formulas
as sentence letters that cannot be analyzed further. This is
illustrated by the following three examples.
The formula ∀x (M (x)) → ∃x (M (x))—read: If everything
is material, then something is material—is logically true in
predicate logic. It is not logically true in propositional logic
because the latter treats ∀x (M (x)) and ∃x (M (x)) as two distinct
sentence letters that cannot be analyzed further. Therefore,
there is a line in the truth table in which ∀x (M (x)) is true
and ∃x (M (x)) is false. This line in the truth table shows that
∀x (M (x))→∃x (M (x)) is not logically true in propositional logic.
The argument F (a) ∴ ∃x (F (x))—read: Angela Merkel is
chancellor of Germany in August 2017; therefore, someone is
chancellor of Germany in August 2017—is logically valid in
predicate logic. It is not logically valid in propositional logic
because the latter treats F (a) and ∃x (F (x)) as two distinct
sentence letters that cannot be analyzed further. Therefore,

there is a line in the truth table in which F (a) is true and
∃x (F (x)) is false. This line in the truth table shows that F (a) ∴
∃x (F (x)) is not logically valid in propositional logic.
The formula ¬∀x (M (x))—read: Not everything is material
—is logically equivalent to the formula ∃x (¬ (M (x)))—read:
Something is not material—in predicate logic. The first formula
is not logically equivalent to the second formula in propositional
logic because the latter treats ∀x (M (x)) and ∃x (¬ (M (x))) as
two distinct sentence letters that cannot be analyzed further.
Therefore, there is a line in the truth table in which ∀x (M (x)) is
false—so that its negation ¬∀x (M (x)) is true—and ∃x (M (x)) is
false. This line in the truth table shows that ¬∀x (M (x)) is not
logically equivalent to ∃x (¬ (M (x))) in propositional logic.
We will come across one principle, the principle of the
substitution of logical equivalents (SLE), where the distinction
between logical equivalence in propositional logic as opposed
to logical equivalence in predicate logic is crucial.
Finally, if you think the first example, much like the
claim that something is or is not material, ∃x (M (x) ∨ ¬M (x)),
should not be a logical truth, and the second example, much
like the argument: Winnie-the-Pooh is a bear and speaks
English; therefore, some bears speak English, B (w) ∧ E (w) ∴
∃x (B (x) ∧ E (x)), should not be a logically valid argument, then
you are a proponent of inclusive logic (Nolt 2014). Inclusive logic
rejects the assumptions of classical logic, which we are using,
that at least one thing exists and that names only refer to
existing things.
1.3 EXERCISES
The truth table for the formula ‘((p ∨ q) ∧ ¬ (q))’ is obtained by

first identifying the different types of propositional variables,
or sentence letters, of the formula. These are ‘p’ and ‘q.’ Next, we
LOGIC 15
list all the possible assignments of truth values to these types of

propositional variables.
p q
T T
T F
F T
F F
Then these truth values of the propositional variables,

or sentence letters, are written underneath all tokens, or
occurrences, of them in the formula:
p q (p ∨ q) ∧ ¬ q
T T T T T
T F T F F
F T F T T
F F F F F
In order to reduce the use of parentheses, we adopt the

convention that ‘¬’ binds stronger than ‘∧’ and ‘∨,’ and that ‘∧’
and ‘∨’ bind stronger than ‘→’ and ‘↔.’ In the above table I have
omitted all parentheses that are not needed to avoid ambiguities
in scope. Next, we work our way from the propositional
variables, or sentence letters, to the first connective, then the
next, and so on. . .
p q (p ∨ q) ∧ ¬ q p q (p ∨ q) ∧ ¬ q
T T T T F T T T T T T F T
T F T F T F T F T T F T F
F T F T F T F T F T T F T
F F F F T F F F F F F T F
. . .until we reach the main connective of the formula, ‘∧’:
p q (p ∨ q) ∧ ¬ q
T T T T T F F T
T F T T F T T F
F T F T T F F T
F F F F F F T F
Exercise 1: Write down the truth table for the following formula
‘((p ∧ q) ∨ (¬ (¬q))),’ or simply ‘(p ∧ q) ∨ ¬¬q.’
In this way we can show, or prove, that the formula ‘¬ (p ∧ ¬p)’ is

logically true because it has a ‘T’ under its main connective ‘¬’
in all lines of the truth table.
p ¬ (p ∧ ¬ p) p ¬ (p ∧ ¬ p)
T T T T T F T
F F F F F T F
p ¬ (p ∧ ¬ p) p ¬ (p ∧ ¬ p)
T T F F T T T T F F T
F F F T F F T F F T F
Exercise 2: Show that the formula ‘(p → (p ∨ q)),’ or simply ‘p →

p ∨ q,’ is logically true by showing that it has a ‘T’ under its main
connective ‘→’ in all lines of the truth table.
In this way we can also show, or prove, that two formulas are
logically equivalent by showing that they have the same truth
value under their main connective in all lines of the truth table.
For instance, we can show in this way that the two formulas
‘(p ∨ q)’ and ‘¬ (¬p ∧ ¬q)’ are logically equivalent because they
have the same truth value under their main connective ‘∨’ and
‘¬,’ respectively, in all lines of the truth table.
LOGIC 17
p q (p ∨ q) ¬ (¬ p ∧ ¬ q)
T T T T T T F T F F T
T F T T F T F T F T F
F T F T T T T F F F T
F F F F F F T F T T F
Exercise 3: Show that the two formulas ‘(p ∧ q)’ and ‘¬ (¬p ∨ ¬q)’
are logically equivalent by showing that they have the same
truth value under their main connective ‘∧’ and ‘¬,’ respectively,
in all lines of the truth table.
Furthermore, we can also show, or prove, in this way that one

formula logically implies another formula by showing that the
second sentence, or formula, has a ‘T’ under its main connective
in all lines of the truth table, if any, where the first formula has a
‘T’ under its main connective. For instance, we can show in this
way that the formula ‘¬p’ logically implies the formula ‘p → q’
because the second formula has a ‘T’ under its main connective
‘→’ in all lines of the truth table in which the first formula has
a ‘T’ under its main connective ‘¬.’
p q ¬ p p → q
T T F T T T T
T F F T T F F
F T T F F T T
F F T F F T F
Exercise 4: We adopt the convention that the main connective of

a propositional variable, or sentence letter, is the propositional
variable, or sentence letter, itself. Show that the formula ‘q’
logically implies the formula ‘p → q’ by showing that the second
formula has a ‘T’ under its main connective ‘→’ in all lines of the
truth table in which the first formula has a ‘T’ under its main
connective ‘q.’
Exercise 5: Show that the formula ‘p ∧ q’ logically implies the

formula ‘¬p ↔ ¬q’ by showing that the second formula has a ‘T’
under its main connective ‘↔’ in all lines of the truth table in
which the first formula has a ‘T’ under its main connective ‘∧.’
The method of truth tables allows us to show, or prove,

logical truths, logical implications, and logical equivalences in
propositional logic. The principle of the substitution of logical
equivalents facilitates this task. SLE says that the formula

α β/γ is logically equivalent to the formula α, if the formula
β is logically equivalent to the formula γ in propositional logic.

Here α β/γ results from α by replacing all occurrences of
γ in α by an occurrence of β. The restriction in SLE that
β and γ are logically equivalent in propositional logic is most
important!
To show, or prove, logical truths, logical implications, and
logical equivalences in predicate logic, we need additional tools.
The first of these is the principle of existential generalization
(EG). EG says that the formula ∃x (α [x/a]) follows logically
in predicate logic from the formula α, provided the individual
variable ‘x’ does not occur in α. Here the formula α [x/a] results
from α by replacing all occurrences of the individual constant
‘a’ in α by an occurrence of the individual variable ‘x.’ We make
use of this principle when we say that, in predicate logic, the
sentence ‘There exists a chancellor of Germany in August 2017’
follows logically from the sentence ‘Angela Merkel is chancellor
of Germany in August 2017,’ F (a) |= ∃x (F (x)).
The second tool is the principle of universal instantiation
(UI). UI says that the formula α [a/x] follows logically in
predicate logic from the formula ∀x (α), where the formula
α [a/x] results from α by replacing all free occurrences of the
individual variable ‘x’ in α by an occurrence of the individual
constant ‘a.’ We make use of this principle when we say that,
in predicate logic, the sentence ‘Muhammad Ali is mortal if
LOGIC 19
Muhammad Ali is human’ follows logically from the sentence

‘All humans are mortal,’ ∀x (H (x) → M (x)) |= H (a) → M (a).
The third tool is the principle of universal generalization
(UG), and it is by far the most difficult one. UG says that the
formula ∀x (α [x/c]) follows logically in predicate logic from the
formula α [c], provided the individual constant ‘c’ is arbitrary
(that is, has not occurred in any formula that was used to
logically infer α [c]), and provided the individual variable ‘x’ is new
(that is, does not occur in α). One way to think of this principle is
as licensing any-all inferences: From the premise that any object
c has a certain property—recall: ‘c’ is arbitrary—one may and
ought to infer the conclusion that all objects have this property.
Before applying these principles in the chapters to follow,
a note on terminology. Logicians often restrict the term
‘sentence’ to these well-formed formulas that do not contain
any free occurrences of individual variables. In this terminology,
‘∀x (F (x))’ is both a sentence and a well-formed formula,
whereas ‘F (x)’ is only a well-formed formula but not also a
sentence. I will try to avoid this terminology, which is why I am
formulating some things in seemingly odd ways.
READINGS
Textbooks that cover similar material as this book are:

Hacking, Ian (2001), An Introduction to Probability and Inductive Logic.
Cambridge: Cambridge University Press.
Skyrms, Brian (1966/2000), Choice and Chance: An Introduction
to Inductive Logic. 4th ed., Belmont, CA: Wadsworth Thomson
Learning.
Recommended further readings for the material in the first
chapter are:
Klement, Kevin C. (2016a), Propositional Logic. In J. Fieser & B.
Dowden (eds.), Internet Encyclopedia of Philosophy.
Papineau, David (2012), Philosophical Devices. Proofs, Probabilities,

Possibilities, and Sets. Oxford: Oxford University Press.
Chapter 10.
and perhaps also
Papineau, David (2012), Philosophical Devices. Proofs, Probabilities,
Possibilities, and Sets. Oxford: Oxford University Press.
Chapter 11.
Shapiro, Stewart (2013), Classical Logic. In E.N. Zalta (ed.), Stanford
Encyclopedia of Philosophy.
CHAPTER 2
Set Theory
2.1 ELEMENTARY POSTULATES
A set is a collection of things, entities, or objects. In the way

philosophers use the term, subjects such as you and I are also
objects. It’s not rude of me to say that you are an object. If
anything, it would be rude of me to say that you are not an
object, as that would be saying something along the lines that
you do not exist.
For instance, the government of Argentina in 2017 is or
can be thought of as the set containing Mauricio Macri and the
members of his cabinet. Similarly, the Library of Alexandria is
or can be thought of as the set containing all its books. We use
the curly brackets ‘{ ’ and ‘ }’ to denote sets, and it is important to
distinguish a set from its members, or elements: Mauricio Macri
and his ministers are human, but the set containing them is not.
The set C containing all and only Canadian cities with more than
1 million inhabitants is

C = Toronto, Montréal, Calgary .
C is also the set of all objects x such that x is a Canadian city with
more than 1 million inhabitants, that is,

C = x : x is a Canadian city with more than 1 million
inhabitants} .
Together with the colon ‘:’ preceded by the individual variable
‘x’ the curly brackets ‘{ ’ and ‘ }’ bind the second occurrence of the
individual variable ‘x’ in the above line. They do so much like the
quantifiers ‘∀x’ and ‘∃y’ bind the individual variables ‘x’ and ‘y’
in ‘∀x∃y (L (x, y)).’
The order in which the members, or elements, of a set are
listed does not matter:

Toronto, Montréal, Calgary = Calgary, Montréal, Toronto
The number of times a member, or an element, is listed does not
matter either:

Toronto, Montréal, Calgary

= Toronto, Toronto, Calgary, Montréal, Toronto
We use ‘∈’ to denote that the object mentioned to the left of ‘∈’
is a member, or an element, of the set mentioned to the right
of ‘∈’, and ‘’ to denote that it is not. For instance, Toronto ∈ C
and Vancouver C.
Sets S and T are identical just in case they contain the
same members, or elements. This is known as the principle of
Extensionality.
Extensionality: For all sets S and T, S = T if and only if for

all objects x, x ∈ S just in case x ∈ T.
In particular, we have for all sets S: S = {x : x ∈ S}, which will turn

out to be a very useful identity. We use ‘⊆’ to denote that all
members of the “subset” mentioned to the left of ‘⊆’ are also
members of the “superset” mentioned to the right of ‘⊆.’ For

instance, C ⊆ x : x is a Canadian city .
We use ‘∅’, or ‘{},’ to denote the empty set that has
no members, or elements, and whose existence set theory
postulates. There are many ways to describe the empty set—for
example, as the set of objects that are not identical to
themselves, or as the set of objects that are both material and
immaterial—but there exists just one empty set.
SET THEORY 23
Furthermore, we have for all sets S and T: S = T if,

and only if, S ⊆ T and T ⊆ S. For instance, since C =

Toronto, Calgary, Montréal , we get that {Toronto, Montréal,

Calgary ⊆ C and C ⊆ Montréal, Toronto, Calgary . The other
direction of this equivalence is useful when we want to prove
that two sets are identical. This follows if we can establish that
they are subsets of each other. In addition, we have for all sets
S: ∅ ⊆ S and S ⊆ S.
Set theory postulates the existence of further sets besides
the empty set. However, unlike the empty set, whose existence
set theory postulates “categorically,” these further sets are only
postulated to exist on the condition that there already are some
sets.
If S and T are sets, then there exists the intersection of S and
T, S ∩ T, which is the set of objects that are elements of both S
and T:
S ∩ T = {x : (x ∈ S) ∧ (x ∈ T)}
For instance, C ∩ {Toronto} = {Toronto}.
S T
If S and T are sets, then there exists the union of S and T, S ∪ T,

which is the set of objects that are elements of S or of T or of
both S and T:
S ∪ T = {x : (x ∈ S) ∨ (x ∈ T)}

For instance, Calgary, Montréal ∪ {Montréal, Toronto} = C.
S T
If S and T are sets, then there exists the complement of S with

respect to T, T \ S, which is the set of objects that are members
of T, but that are not members of S:
T \ S = {x : (x ∈ T) ∧ ¬ (x ∈ S)} = {x : (x ∈ T) ∧ (x S)}

For instance, C \ {Toronto} = Calgary, Montréal .
S T
If S is a set, then there exists the power set of S, ℘ (S), which is the
set of all subsets of S:
℘ (S) = {A : A ⊆ S}
For instance, the power set of C, ℘ {C}, is the set that contains
the following eight sets as elements:

∅, {Toronto} , {Montréal} , Calgary {Toronto, Montréal} ,

Toronto, Calgary , Calgary, Montréal , C
This means that ℘ (C) is the following set of sets:

℘ (C) = ∅, {Toronto} , {Montréal} , Calgary ,

{Toronto, Montréal} , Toronto, Calgary ,

Calgary, Montréal , C
SET THEORY 25
The power set of a set cannot be pictured easily, as it is a set

whose members, or elements, are sets as well.
Note that the above principles have all been formulated
in the language of predicate logic. This means that set theory
can be formulated as a list of sentences of, or as a theory in,
predicate logic. This has the consequence that we can apply all
the logical principles from the previous chapter to prove claims
in set theory. For instance, we can show, or prove, that it is a
logical truth that every object x either is, or is not, an element
of any given set S.
Here is how. First, let ‘y’ and ‘T’ be arbitrary individual
constants. ‘∈’ is the binary relation of set theoretic membership,
or elementhood. In other words, let y and T be arbitrary objects
of which we assume nothing whatsoever. ‘∈ (y, T),’ or more
perspicuously, ‘y ∈ T’ is a well-formed formula, and so (y ∈ T) ∨
¬ (y ∈ T) is logically true. We can show this by the following
truth table, where ‘t’ is the sentence letter for ‘y ∈ T’:
t t ∨ ¬ t
T T T F T
F F T T F
The next step to arrive at our claim consists in noting that
everything is a thing or an object. This includes sets, which are
the objects satisfying the postulates of set theory. If it is true
that, for an arbitrary object y and an arbitrary object T, y does
or does not stand in relation ∈ to T, then it is also true that,
for an arbitrary object y and an arbitrary object T that has the
property of being a set, y does or does not stand in relation
∈ to T. Alternatively, we can use the method of truth tables to
show that (y ∈ T) ∨ ¬ (y ∈ T) logically implies set (T) → (y ∈ T) ∨
¬ (y ∈ T).
Now we apply the principle of universal generalization
(UG), which says that, in predicate logic, set (T) → (y ∈ T) ∨
¬ (y ∈ T) logically implies
∀S (set (S) → (y ∈ S) ∨ ¬ (y ∈ S)) ,
because ‘T’ was arbitrary (that is, it has not occurred before we
introduced it above,) and because ‘S’ is new (that is, it does not
occur in ‘set (T) → (y ∈ T) ∨ ¬ (y ∈ T)).
A second application of the principle of universal
generalization says that, in predicate logic, ∀S (set (S) → (y ∈ S) ∨
¬ (y ∈ S)) logically implies
∀x (∀S (set (S) → (x ∈ S) ∨ ¬ (x ∈ S))) ,
because ‘y’ was arbitrary (that is, it has not occurred before we
introduced it above), and because ‘x’ is new (that is, does not occur
in ‘∀S (set (S) → (y ∈ S) ∨ ¬ (y ∈ S))). This completes our proof.
The above postulates, or axioms, postulate the existence
of various sets: the empty set, the power set of a set, the
union of sets, the intersection of sets, the complement of
a set with respect to a set. Another principle, the so-called
unrestricted comprehension principle, has been postulated by
Frege (1893/1903), who thought that for each property P there
exists the set SP of objects that posses the property P, SP =

x : x possesses property P . For instance, P may be the property
of being a Canadian city with more than 1 million inhabitants.
(Here we count everything as a property that can be described
by a well-formed formula of predicate logic α [x] in which the
individual variable ‘x’ occurs freely.)
As we have just seen, sets can be members of sets, just
as chefs can cook the dinners of chefs. For instance, each set
S is a member of its power set, and so is the empty set, but
not conversely: S ∈ ℘ (S) and ∅ ∈ S, but ℘ (S) S and S ∅.
Russell (1902) used the following property of sets to show that
the unrestricted comprehension principle is logically false, or
contradictory:
Set S possesses the Russell property just in case S S.
Compare: A chef is special if and only if she does not cook

her own dinner. According to the unrestricted comprehension
SET THEORY 27
principle, for each property there exists the set of objects that
have this property. The Russell property is a property, and so the
unrestricted comprehension principle implies that there exists
the set, the so-called “Russell set” SR , containing all and only
the objects that possess the Russell property:
SR = {S : S S}
This cannot be true, though. Consider the question whether the
Russell set has the Russell property. Suppose first it does so that
RS RS . In this case, RS possesses the Russell property and so
is a member of RS , RS ∈ RS . Suppose next it does not so that
RS ∈ RS . In this case, RS is a member of RS and so possesses the
Russell property, RS RS . Hence, RS ∈ RS if and only if RS RS ,
which is logically false.
In the same way we can prove that it is logically false that
there exists a chef who cooks the dinners of all and only these
chefs who do not cook their own dinners, that is, it is logically
false that there exists a chef who cooks the dinners of all and
only the special chefs. Suppose there exists such a chef and
consider the question if she is special and does not cook her own
dinner. Suppose first she does not cook her own dinner and so is
special. Then she is one of these chefs who she is cooking dinner
for, and so she is not special after all. Suppose next she cooks her
own dinner. Then she is not special, and so is not one of these
chefs she is cooking dinner for. Hence, she is special if only if
she is not, which is a contradiction, that is, a sentence that is
logically false.
The set theory we use relies on a weaker version of the
unrestricted comprehension principle that is known as the
restricted comprehension axiom. The latter principle says that
for each set S and each property P there exists the set SP of
objects which are members of S and possess the property P.
The restricted comprehension axiom is not logically false. It
avoids Russell’s paradox because it assumes there to be a set
S—say, the set of Canadian cities—and then merely postulates
the existence of the subset of S whose members have a given

property—for example, the set of Canadian cities with more
than 1 million inhabitants.
2.2 EXERCISES
Let us show that, in set theory, the following is true of every

set P:
P = P∪P
1. P = {x : x ∈ P} from Extensionality.
2. P = {x : (x ∈ P) ∨ (x ∈ P)} from 1. and because x ∈ P is
logically equivalent to (x ∈ P) ∨ (x ∈ P) in propositional
logic, which can be shown by the method of truth
tables.
3. P = {x : x ∈ (P ∪ P)} from 2. and the definition of ∪.
4. P = P ∪ P from 3. and Extensionality.
Here are the relevant truth tables, where ‘p’ is the sentence letter
for ‘x ∈ P’:
p p p ∨ p p p ∨ p p
T T T T T T T T T T
F F F F F F F F F F
Since we are arguing inside the scope of ‘{x : . . . x . . .},’ we need to
be careful and so will restrict ourselves to what is logically true
in propositional logic. In this section, logical equivalence means
logical equivalence in propositional logic. The principle that
allows us to substitute a formula inside the scope of ‘{x : . . . x . . .}’
for another formula that is logically equivalent to the former in
propositional logic is the principle of Extensionality. It implies
that the curly brackets do not create what philosophers call a
“hyperintensional” context in which this is not allowed. This
is different for concepts such as actual belief which creates a
hyperintensional context.
Another random document with
no related content on Scribd:
that environment into which it has its outlet. My conception of my
spiritual environment is not to be compared with your scientific world
of pointer readings; it is an everyday world to be compared with the
material world of familiar experience. I claim it as no more real and
no less real than that. Primarily it is not a world to be analysed, but a
world to be lived in.
Granted that this takes us outside the sphere of exact knowledge,
and that it is difficult to imagine that anything corresponding to exact
science will ever be applicable to this part of our environment, the
mystic is unrepentant. Because we are unable to render exact
account of our environment it does not follow that it would be better
to pretend that we live in a vacuum.
If the defence may be considered to have held good against the
first onslaught, perhaps the next stage of the attack will be an easy
tolerance. “Very well. Have it your own way. It is a harmless sort of
belief—not like a more dogmatic theology. You want a sort of spiritual
playground for those queer tendencies in man’s nature, which
sometimes take possession of him. Run away and play then; but do
not bother the serious people who are making the world go round.”
The challenge now comes not from the scientific materialism which
professes to seek a natural explanation of spiritual power, but from
the deadlier moral materialism which despises it. Few deliberately
hold the philosophy that the forces of progress are related only to the
material side of our environment, but few can claim that they are not
more or less under its sway. We must not interrupt the “practical
men”, these busy moulders of history carrying us at ever-increasing
pace towards our destiny as an ant-heap of humanity infesting the
earth. But is it true in history that material forces have been the most
potent factors? Call it of God, of the Devil, fanaticism, unreason; but
do not underrate the power of the mystic. Mysticism may be fought
as error or believed as inspired, but it is no matter for easy tolerance
—
We are the music-makers

And we are the dreamers of dreams
Wandering by lone sea-breakers
And sitting by desolate streams;
World-losers and world-forsakers,

On whom the pale moon gleams:
Yet we are the movers and shakers
Of the world for ever, it seems.
Reality and Mysticism. But a defence before the scientists may not
be a defence to our own self-questionings. We are haunted by the
word reality. I have already tried to deal with the questions which
arise as to the meaning of reality; but it presses on us so persistently
that, at the risk of repetition, I must consider it once more from the
standpoint of religion. A compromise of illusion and reality may be all
very well in our attitude towards physical surroundings; but to admit
such a compromise into religion would seem to be a trifling with
sacred things. Reality seems to concern religious beliefs much more
than any others. No one bothers as to whether there is a reality
behind humour. The artist who tries to bring out the soul in his
picture does not really care whether and in what sense the soul can
be said to exist. Even the physicist is unconcerned as to whether
atoms or electrons really exist; he usually asserts that they do, but,
as we have seen, existence is there used in a domestic sense and
no inquiry is made as to whether it is more than a conventional term.
In most subjects (perhaps not excluding philosophy) it seems
sufficient to agree on the things that we shall call real, and
afterwards try to discover what we mean by the word. And so it
comes about that religion seems to be the one field of inquiry in
which the question of reality and existence is treated as of serious
and vital importance.
But it is difficult to see how such an inquiry can be profitable.
When Dr. Johnson felt himself getting tied up in argument over
“Bishop Berkeley’s ingenious sophistry to prove the non-existence of
matter, and that everything in the universe is merely ideal”, he
answered, “striking his foot with mighty force against a large stone,
till he rebounded from it,—‘I refute it thus’” Just what that action
assured him of is not very obvious; but apparently he found it
comforting. And to-day the matter-of-fact scientist feels the same
impulse to recoil from these flights of thought back to something
kickable, although he ought to be aware by this time that what
Rutherford has left us of the large stone is scarcely worth kicking.
There is still the tendency to use “reality” as a word of magic
comfort like the blessed word “Mesopotamia”. If I were to assert the
reality of the soul or of God, I should certainly not intend a
comparison with Johnson’s large stone—a patent illusion—or even
with the ′s and ′s of the quantum theory—an abstract symbolism.
Therefore I have no right to use the word in religion for the purpose
of borrowing on its behalf that comfortable feeling which (probably
wrongly) has become associated with stones and quantum co-
ordinates.
Scientific instincts warn me that any attempt to answer the
question “What is real?” in a broader sense than that adopted for
domestic purposes in science, is likely to lead to a floundering
among vain words and high-sounding epithets. We all know that
there are regions of the human spirit untrammelled by the world of
physics. In the mystic sense of the creation around us, in the
expression of art, in a yearning towards God, the soul grows upward
and finds the fulfilment of something implanted in its nature. The
sanction for this development is within us, a striving born with our
consciousness or an Inner Light proceeding from a greater power
than ours. Science can scarcely question this sanction, for the
pursuit of science springs from a striving which the mind is impelled
to follow, a questioning that will not be suppressed. Whether in the
intellectual pursuits of science or in the mystical pursuits of the spirit,
the light beckons ahead and the purpose surging in our nature
responds. Can we not leave it at that? Is it really necessary to drag
in the comfortable word “reality” to be administered like a pat on the
back?
The problem of the scientific world is part of a broader problem—
the problem of all experience. Experience may be regarded as a
combination of self and environment, it being part of the problem to
disentangle these two interacting components. Life, religion,
knowledge, truth are all involved in this problem, some relating to the
finding of ourselves, some to the finding of our environment from the
experience confronting us. All of us in our lives have to make
something of this problem; and it is an important condition that we
who have to solve the problem are ourselves part of the problem.
Looking at the very beginning, the initial fact is the feeling of purpose
in ourselves which urges us to embark on the problem. We are
meant to fulfil something by our lives. There are faculties with which
we are endowed, or which we ought to attain, which must find a
status and an outlet in the solution. It may seem arrogant that we
should in this way insist on moulding truth to our own nature; but it is
rather that the problem of truth can only spring from a desire for truth
which is in our nature.
A rainbow described in the symbolism of physics is a band of
aethereal vibrations arranged in systematic order of wave-length
from about .000040 cm. to .000072 cm. From one point of view we
are paltering with the truth whenever we admire the gorgeous bow of
colour, and should strive to reduce our minds to such a state that we
receive the same impression from the rainbow as from a table of
wave-lengths. But although that is how the rainbow impresses itself
on an impersonal spectroscope, we are not giving the whole truth
and significance of experience—the starting-point of the problem—if
we suppress the factors wherein we ourselves differ from a
spectroscope. We cannot say that the rainbow, as part of the world,
was meant to convey the vivid effects of colour; but we can perhaps
say that the human mind as part of the world was meant to perceive
it that way.
Significance and Values. When we think of the sparkling waves as

moved with laughter we are evidently attributing a significance to the
scene which was not there. The physical elements of the water—the
scurrying electric charges—were guiltless of any intention to convey
the impression that they were happy. But so also were they guiltless
of any intention to convey the impression of substance, of colour, or
of geometrical form of the waves. If they can be held to have had
any intention at all it was to satisfy certain differential equations—
and that was because they are the creatures of the mathematician
who has a partiality for differential equations. The physical no less
than the mystical significance of the scene is not there; it is here—in
the mind.
What we make of the world must be largely dependent on the
sense-organs that we happen to possess. How the world must have
changed since man came to rely on his eyes rather than his nose!
You are alone on the mountains wrapt in a great silence; but equip
yourself with an extra artificial sense-organ and, lo! the aether is
hideous with the blare of the Savoy bands. Or—
The isle is full of noises,

Sounds, and sweet airs, that give delight, and hurt not.
Sometimes a thousand twangling instruments
Will hum about mine ears; and sometimes voices.
So far as broader characteristics are concerned we see in Nature

what we look for or are equipped to look for. Of course, I do not
mean that we can arrange the details of the scene; but by the light
and shade of our values we can bring out things that shall have the
broad characteristics we esteem. In this sense the value placed on
permanence creates the world of apparent substance; in this sense,
perhaps, the God within creates the God in Nature. But no complete
view can be obtained so long as we separate our consciousness
from the world of which it is a part. We can only speak speculatively
of that which I have called the “background of the pointer readings”;
but it would at least seem plausible that if the values which give the
light and shade of the world are absolute they must belong to the
background, unrecognised in physics because they are not in the
pointer readings but recognised by consciousness which has its
roots in the background. I have no wish to put that forward as a
theory; it is only to emphasise that, limited as we are to a knowledge
of the physical world and its points of contact with the background in
isolated consciousness, we do not quite attain that thought of the
unity of the whole which is essential to a complete theory.
Presumably human nature has been specialised to a considerable
extent by the operation of natural selection; and it might well be
debated whether its valuation of permanence and other traits now
apparently fundamental are essential properties of consciousness or
have been evolved through interplay with the external world. In that
case the values given by mind to the external world have originally
come to it from the external world-stuff. Such a tossing to and fro of
values is, I think, not foreign to our view that the world-stuff behind
the pointer readings is of nature continuous with the mind.
In viewing the world in a practical way values for normal human
consciousness may be taken as standard. But the evident possibility
of arbitrariness in this valuation sets us hankering after a standard
that could be considered final and absolute. We have two
alternatives. Either there are no absolute values, so that the
sanctions of the inward monitor in our consciousness are the final
court of appeal beyond which it is idle to inquire. Or there are
absolute values; then we can only trust optimistically that our values
are some pale reflection of those of the Absolute Valuer, or that we
have insight into the mind of the Absolute from whence come those
strivings and sanctions whose authority we usually forbear to
question.
I have naturally tried to make the outlook reached in these
lectures as coherent as possible, but I should not be greatly
concerned if under the shafts of criticism it becomes very ragged.
Coherency goes with finality; and the anxious question is whether
our arguments have begun right rather than whether they have had
the good fortune to end right. The leading points which have seemed
to me to deserve philosophic consideration may be summarised as
follows:
(1) The symbolic nature of the entities of physics is generally
recognised; and the scheme of physics is now formulated in such a
way as to make it almost self-evident that it is a partial aspect of
something wider.
(2) Strict causality is abandoned in the material world. Our ideas
of the controlling laws are in process of reconstruction and it is not
possible to predict what kind of form they will ultimately take; but all
the indications are that strict causality has dropped out permanently.
This relieves the former necessity of supposing that mind is subject
to deterministic law or alternatively that it can suspend deterministic
law in the material world.
(3) Recognising that the physical world is entirely abstract and
without “actuality” apart from its linkage to consciousness, we restore
consciousness to the fundamental position instead of representing it
as an inessential complication occasionally found in the midst of
inorganic nature at a late stage of evolutionary history.
(4) The sanction for correlating a “real” physical world to certain
feelings of which we are conscious does not seem to differ in any
essential respect from the sanction for correlating a spiritual domain
to another side of our personality.
It is not suggested that there is anything new in this philosophy.
In particular the essence of the first point has been urged by many
writers, and has no doubt won individual assent from many scientists
before the recent revolutions of physical theory. But it places a
somewhat different complexion on the matter when this is not merely
a philosophic doctrine to which intellectual assent might be given,
but has become part of the scientific attitude of the day, illustrated in
detail in the current scheme of physics.
Conviction. Through fourteen chapters you have followed with me

the scientific approach to knowledge. I have given the philosophical
reflections as they have naturally arisen from the current scientific
conclusions, I hope without distorting them for theological ends. In
the present chapter the standpoint has no longer been
predominantly scientific; I started from that part of our experience
which is not within the scope of a scientific survey, or at least is such
that the methods of physical science would miss the significance that
we consider it essential to attribute to it. The starting-point of belief in
mystical religion is a conviction of significance or, as I have called it
earlier, the sanction of a striving in the consciousness. This must be
emphasised because appeal to intuitive conviction of this kind has
been the foundation of religion through all ages and I do not wish to
give the impression that we have now found something new and
more scientific to substitute. I repudiate the idea of proving the
distinctive beliefs of religion either from the data of physical science
or by the methods of physical science. Presupposing a mystical
religion based not on science but (rightly or wrongly) on a self-known
experience accepted as fundamental, we can proceed to discuss the
various criticisms which science might bring against it or the possible
conflict with scientific views of the nature of experience equally
originating from self-known data.
It is necessary to examine further the nature of the conviction
from which religion arises; otherwise we may seem to be
countenancing a blind rejection of reason as a guide to truth. There
is a hiatus in reasoning, we must admit; but it is scarcely to be
described as a rejection of reasoning. There is just the same hiatus
in reasoning about the physical world if we go back far enough. We
can only reason from data and the ultimate data must be given to us
by a non-reasoning process—a self-knowledge of that which is in our
consciousness. To make a start we must be aware of something. But
that is not sufficient; we must be convinced of the significance of that
awareness. We are bound to claim for human nature that, either of
itself or as inspired by a power beyond, it is capable of making
legitimate judgments of significance. Otherwise we cannot even
reach a physical world.[49]
Accordingly the conviction which we postulate is that certain
states of awareness in consciousness have at least equal
significance with those which are called sensations. It is perhaps not
irrelevant to note that time by its dual entry into our minds (p. 51) to
some extent bridges the gap between sense-impressions and these
other states of awareness. Amid the latter must be found the basis of
experience from which a spiritual religion arises. The conviction is
scarcely a matter to be argued about, it is dependent on the
forcefulness of the feeling of awareness.
But, it may be said, although we may have such a department of
consciousness, may we not have misunderstood altogether the
nature of that which we believe we are experiencing? That seems to
me to be rather beside the point. In regard to our experience of the
physical world we have very much misunderstood the meaning of
our sensations. It has been the task of science to discover that
things are very different from what they seem. But we do not pluck
out our eyes because they persist in deluding us with fanciful
colourings instead of giving us the plain truth about wave-length. It is
in the midst of such misrepresentations of environment (if you must
call them so) that we have to live. It is, however, a very one-sided
view of truth which can find in the glorious colouring of our
surroundings nothing but misrepresentation—which takes the
environment to be all important and the conscious spirit to be
inessential. In our scientific chapters we have seen how the mind
must be regarded as dictating the course of world-building; without it
there is but formless chaos. It is the aim of physical science, so far
as its scope extends, to lay bare the fundamental structure
underlying the world; but science has also to explain if it can, or else
humbly to accept, the fact that from this world have arisen minds
capable of transmuting the bare structure into the richness of our
experience. It is not misrepresentation but rather achievement—the
result perhaps of long ages of biological evolution—that we should
have fashioned a familiar world out of the crude basis. It is a
fulfilment of the purpose of man’s nature. If likewise the spiritual
world has been transmuted by a religious colour beyond anything
implied in its bare external qualities, it may be allowable to assert
with equal conviction that this is not misrepresentation but the
achievement of a divine element in man’s nature.
May I revert again to the analogy of theology with the supposed
science of humour which (after consultation with a classical
authority) I venture to christen “geloeology”. Analogy is not
convincing argument, but it must serve here. Consider the proverbial
Scotchman with strong leanings towards philosophy and incapable
of seeing a joke. There is no reason why he should not take high
honours in geloeology, and for example write an acute analysis of
the differences between British and American humour. His
comparison of our respective jokes would be particularly unbiased
and judicial, seeing that he is quite incapable of seeing the point of
either. But it would be useless to consider his views as to which was
following the right development; for that he would need a
sympathetic understanding—he would (in the phrase appropriate to
the other side of my analogy) need to be converted. The kind of help
and criticism given by the geloeologist and the philosophical
theologian is to secure that there is method in our madness. The
former may show that our hilarious reception of a speech is the
result of a satisfactory dinner and a good cigar rather than a subtle
perception of wit; the latter may show that the ecstatic mysticism of
the anchorite is the vagary of a fevered body and not a transcendent
revelation. But I do not think we should appeal to either of them to
discuss the reality of the sense with which we claim to be endowed,
nor the direction of its right development. That is a matter for our
inner sense of values which we all believe in to some extent, though
it may be a matter of dispute just how far it goes. If we have no such
sense then it would seem that not only religion, but the physical
world and all faith in reasoning totter in insecurity.
I have sometimes been asked whether science cannot now
furnish an argument which ought to convince any reasonable atheist.
I could no more ram religious conviction into an atheist than I could
ram a joke into the Scotchman. The only hope of “converting” the
latter is that through contact with merry-minded companions he may
begin to realise that he is missing something in life which is worth
attaining. Probably in the recesses of his solemn mind there exists
inhibited the seed of humour, awaiting an awakening by such an
impulse. The same advice would seem to apply to the propagation of
religion; it has, I believe, the merit of being entirely orthodox advice.
We cannot pretend to offer proofs. Proof is an idol before whom
the pure mathematician tortures himself. In physics we are generally
content to sacrifice before the lesser shrine of Plausibility. And even
the pure mathematician—that stern logician—reluctantly allows
himself some prejudgments; he is never quite convinced that the
scheme of mathematics is flawless, and mathematical logic has
undergone revolutions as profound as the revolutions of physical
theory. We are all alike stumblingly pursuing an ideal beyond our
reach. In science we sometimes have convictions as to the right
solution of a problem which we cherish but cannot justify; we are
influenced by some innate sense of the fitness of things. So too
there may come to us convictions in the spiritual sphere which our
nature bids us hold to. I have given an example of one such
conviction which is rarely if ever disputed—that surrender to the
mystic influence of a scene of natural beauty is right and proper for a
human spirit, although it would have been deemed an unpardonable
eccentricity in the “observer” contemplated in earlier chapters.
Religious conviction is often described in somewhat analogous terms
as a surrender; it is not to be enforced by argument on those who do
not feel its claim in their own nature.
I think it is inevitable that these convictions should emphasise a
personal aspect of what we are trying to grasp. We have to build the
spiritual world out of symbols taken from our own personality, as we
build the scientific world out of the metrical symbols of the
mathematician. If not, it can only be left ungraspable—an
environment dimly felt in moments of exaltation but lost to us in the
sordid routine of life. To turn it into more continuous channels we
must be able to approach the World-Spirit in the midst of our cares
and duties in that simpler relation of spirit to spirit in which all true
religion finds expression.
Mystical Religion. We have seen that the cyclic scheme of physics

presupposes a background outside the scope of its investigations. In
this background we must find, first, our own personality, and then
perhaps a greater personality. The idea of a universal Mind or Logos
would be, I think, a fairly plausible inference from the present state of
scientific theory; at least it is in harmony with it. But if so, all that our
inquiry justifies us in asserting is a purely colourless pantheism.
Science cannot tell whether the world-spirit is good or evil, and its
halting argument for the existence of a God might equally well be
turned into an argument for the existence of a Devil.
I think that that is an example of the limitation of physical
schemes that has troubled us before—namely, that in all such
schemes opposites are represented by + and -. Past and future,
cause and effect, are represented in this inadequate way. One of the
greatest puzzles of science is to discover why protons and electrons
are not simply the opposites of one another, although our whole
conception of electric charge requires that positive and negative
electricity should be related like + and -. The direction of time’s arrow
could only be determined by that incongruous mixture of theology
and statistics known as the second law of thermodynamics; or, to be
more explicit, the direction of the arrow could be determined by
statistical rules, but its significance as a governing fact “making
sense of the world” could only be deduced on teleological
assumptions. If physics cannot determine which way up its own
world ought to be regarded, there is not much hope of guidance from
it as to ethical orientation. We trust to some inward sense of fitness
when we orient the physical world with the future on top, and
likewise we must trust to some inner monitor when we orient the
spiritual world with the good on top.
Granted that physical science has limited its scope so as to leave
a background which we are at liberty to, or even invited to, fill with a
reality of spiritual import, we have yet to face the most difficult
criticism from science. “Here”, says science, “I have left a domain in
which I shall not interfere. I grant that you have some kind of avenue
to it through the self-knowledge of consciousness, so that it is not
necessarily a domain of pure agnosticism. But how are you going to
deal with this domain? Have you any system of inference from
mystic experience comparable to the system by which science
develops a knowledge of the outside world? I do not insist on your
employing my method, which I acknowledge is inapplicable; but you
ought to have some defensible method. The alleged basis of
experience may possibly be valid; but have I any reason to regard
the religious interpretation currently given to it as anything more than
muddle-headed romancing?”
The question is almost beyond my scope. I can only
acknowledge its pertinency. Although I have chosen the lightest task
by considering only mystical religion—and I have no impulse to
defend any other—I am not competent to give an answer which shall
be anything like complete. It is obvious that the insight of
consciousness, although the only avenue to what I have called
intimate knowledge of the reality behind the symbols of science, is
not to be trusted implicitly without control. In history religious
mysticism has often been associated with extravagances that cannot
be approved. I suppose too that oversensitiveness to aesthetic
influences may be a sign of a neurotic temperament unhealthy to the
individual. We must allow something for the pathological condition of
the brain in what appear to be moments of exalted insight. One
begins to fear that after all our faults have been detected and
removed there will not be any “us” left. But in the study of the
physical world we have ultimately to rely on our sense-organs,
although they are capable of betraying us by gross illusions; similarly
the avenue of consciousness into the spiritual world may be beset
with pitfalls, but that does not necessarily imply that no advance is
possible.
A point that must be insisted on is that religion or contact with
spiritual power if it has any general importance at all must be a
commonplace matter of ordinary life, and it should be treated as
such in any discussion. I hope that you have not interpreted my
references to mysticism as referring to abnormal experiences and
revelations. I am not qualified to discuss what evidential value (if
any) may be attached to the stranger forms of experience and
insight. But in any case to suppose that mystical religion is mainly
concerned with these is like supposing that Einstein’s theory is
mainly concerned with the perihelion of Mercury and a few other
exceptional observations. For a matter belonging to daily affairs the
tone of current discussions often seems quite inappropriately
pedantic.
As scientists we realise that colour is merely a question of the
wave-lengths of aethereal vibrations; but that does not seem to have
dispelled the feeling that eyes which reflect light near wave-length
4800 are a subject for rhapsody whilst those which reflect wave-
length 5300 are left unsung. We have not yet reached the practice of
the Laputans, who, “if they would, for example, praise the beauty of
a woman, or any other animal, they describe it by rhombs, circles,
parallelograms, ellipses, and other geometrical terms”. The
materialist who is convinced that all phenomena arise from electrons
and quanta and the like controlled by mathematical formulae, must
presumably hold the belief that his wife is a rather elaborate
differential equation; but he is probably tactful enough not to obtrude
this opinion in domestic life. If this kind of scientific dissection is felt
to be inadequate and irrelevant in ordinary personal relationships, it
is surely out of place in the most personal relationship of all—that of
the human soul to a divine spirit.
We are anxious for perfect truth, but it is hard to say in what form
perfect truth is to be found. I cannot quite believe that it has the form
typified by an inventory. Somehow as part of its perfection there
should be incorporated in it that which we esteem as a “sense of
proportion”. The physicist is not conscious of any disloyalty to truth
on occasions when his sense of proportion tells him to regard a
plank as continuous material, well knowing that it is “really” empty
space containing sparsely scattered electric charges. And the
deepest philosophical researches as to the nature of the Deity may
give a conception equally out of proportion for daily life; so that we
should rather employ a conception that was unfolded nearly two
thousand years ago.
I am standing on the threshold about to enter a room. It is a
complicated business. In the first place I must shove against an
atmosphere pressing with a force of fourteen pounds on every
square inch of my body. I must make sure of landing on a plank
travelling at twenty miles a second round the sun—a fraction of a
second too early or too late, the plank would be miles away. I must
do this whilst hanging from a round planet head outward into space,
and with a wind of aether blowing at no one knows how many miles
a second through every interstice of my body. The plank has no
solidity of substance. To step on it is like stepping on a swarm of
flies. Shall I not slip through? No, if I make the venture one of the
flies hits me and gives a boost up again; I fall again and am knocked
upwards by another fly; and so on. I may hope that the net result will
be that I remain about steady; but if unfortunately I should slip
through the floor or be boosted too violently up to the ceiling, the
occurrence would be, not a violation of the laws of Nature, but a rare
coincidence. These are some of the minor difficulties. I ought really
to look at the problem four-dimensionally as concerning the
intersection of my world-line with that of the plank. Then again it is
necessary to determine in which direction the entropy of the world is
increasing in order to make sure that my passage over the threshold
is an entrance, not an exit.
Verily, it is easier for a camel to pass through the eye of a needle
than for a scientific man to pass through a door. And whether the
door be barn door or church door it might be wiser that he should
consent to be an ordinary man and walk in rather than wait till all the
difficulties involved in a really scientific ingress are resolved.
[49] We can of course solve the problem arising from certain data
without being convinced of the significance of the data—the
“official” scientific attitude as I have previously called it But a
physical world which has only the status of the solution of a
problem, arbitrarily chosen to pass an idle hour, is not what is
intended here.
CONCLUSION
A tide of indignation has been surging in the breast of the matter-of-
fact scientist and is about to be unloosed upon us. Let us broadly
survey the defence we can set up.
I suppose the most sweeping charge will be that I have been
talking what at the back of my mind I must know is only a well-
meaning kind of nonsense. I can assure you that there is a scientific
part of me that has often brought that criticism during some of the
later chapters. I will not say that I have been half-convinced, but at
least I have felt a homesickness for the paths of physical science
where there are more or less discernible handrails to keep us from
the worst morasses of foolishness. But however much I may have
felt inclined to tear up this part of the discussion and confine myself
to my proper profession of juggling with pointer readings, I find
myself holding to the main principles. Starting from aether, electrons
and other physical machinery we cannot reach conscious man and
render count of what is apprehended in his consciousness.
Conceivably we might reach a human machine interacting by
reflexes with its environment; but we cannot reach rational man
morally responsible to pursue the truth as to aether and electrons or
to religion. Perhaps it may seem unnecessarily portentous to invoke
the latest developments of the relativity and quantum theories merely
to tell you this; but that is scarcely the point. We have followed these
theories because they contain the conceptions of modern science;
and it is not a question of asserting a faith that science must
ultimately be reconcilable with an idealistic view, but of examining
how at the moment it actually stands in regard to it. I might sacrifice
the detailed arguments of the last four chapters (perhaps marred by
dialectic entanglement) if I could otherwise convey the significance
of the recent change which has overtaken scientific ideals. The
physicist now regards his own external world in a way which I can
only describe as more mystical, though not less exact and practical,
than that which prevailed some years ago, when it was taken for
granted that nothing could be true unless an engineer could make a
model of it. There was a time when the whole combination of self
and environment which makes up experience seemed likely to pass
under the dominion of a physics much more iron-bound than it is
now. That overweening phase, when it was almost necessary to ask
the permission of physics to call one’s soul one’s own, is past. The
change gives rise to thoughts which ought to be developed. Even if
we cannot attain to much clarity of constructive thought we can
discern that certain assumptions, expectations or fears are no longer
applicable.
Is it merely a well-meaning kind of nonsense for a physicist to
affirm this necessity for an outlook beyond physics? It is worse
nonsense to deny it. Or as that ardent relativist the Red Queen puts
it, “You call that nonsense, but I’ve heard nonsense compared with
which that would be as sensible as a dictionary”.
For if those who hold that there must be a physical basis for
everything hold that these mystical views are nonsense, we may ask
—What then is the physical basis of nonsense? The “problem of
nonsense” touches the scientist more nearly than any other moral
problem. He may regard the distinction of good and evil as too
remote to bother about; but the distinction of sense and nonsense, of
valid and invalid reasoning, must be accepted at the beginning of
every scientific inquiry. Therefore it may well be chosen for
examination as a test case.
If the brain contains a physical basis for the nonsense which it
thinks, this must be some kind of configuration of the entities of
physics—not precisely a chemical secretion, but not essentially
different from that kind of product. It is as though when my brain
says 7 times 8 are 56 its machinery is manufacturing sugar, but
when it says 7 times 8 are 65 the machinery has gone wrong and
produced chalk. But who says the machinery has gone wrong? As a
physical machine the brain has acted according to the unbreakable
laws of physics; so why stigmatise its action? This discrimination of
chemical products as good or evil has no parallel in chemistry. We
cannot assimilate laws of thought to natural laws; they are laws
which ought to be obeyed, not laws which must be obeyed; and the
physicist must accept laws of thought before he accepts natural law.
“Ought” takes us outside chemistry and physics. It concerns
something which wants or esteems sugar, not chalk, sense, not
nonsense. A physical machine cannot esteem or want anything;
whatever is fed into it it will chaw up according to the laws of its
physical machinery. That which in the physical world shadows the
nonsense in the mind affords no ground for its condemnation. In a
world of aether and electrons we might perhaps encounter
nonsense; we could not encounter damned nonsense.
The most plausible physical theory of correct reasoning would
probably run somewhat as follows. By reasoning we are sometimes
able to predict events afterwards confirmed by observation; the
mental processes follow a sequence ending in a conception which
anticipates a subsequent perception. We may call such a chain of
mental states “successful reasoning”—intended as a technical
classification without any moral implications involving the awkward
word “ought”. We can examine what are the common characteristics
of various pieces of successful reasoning. If we apply this analysis to
the mental aspects of the reasoning we obtain laws of logic; but
presumably the analysis could also be applied to the physical
constituents of the brain. It is not unlikely that a distinctive
characteristic would be found in the physical processes in the brain-
cells which accompany successful reasoning, and this would
constitute “the physical basis of success.”
But we do not use reasoning power solely to predict
observational events, and the question of success (as above
defined) does not always arise. Nevertheless if such reasoning were
accompanied by the product which I have called “the physical basis
of success” we should naturally assimilate it to successful reasoning.
And so if I persuade my materialist opponent to withdraw the
epithet “damned nonsense” as inconsistent with his own principles
he is still entitled to allege that my brain in evolving these ideas did
not contain the physical basis of success. As there is some danger
of our respective points of view becoming mixed up, I must make
clear my contention:
(a) If I thought like my opponent I should not worry about the
alleged absence of a physical basis of success in my reasoning,
since it is not obvious why this should be demanded when we are
not dealing with observational predictions.
(b) As I do not think like him, I am deeply perturbed by the
allegation; because I should consider it to be the outward sign that
the stronger epithet (which is not inconsistent with my principles) is
applicable.
I think that the “success” theory of reasoning will not be much
appreciated by the pure mathematician. For him reasoning is a
heaven-sent faculty to be enjoyed remote from the fuss of external
Nature. It is heresy to suggest that the status of his demonstrations
depends on the fact that a physicist now and then succeeds in
predicting results which accord with observation. Let the external
world behave as irrationally as it will, there will remain undisturbed a
corner of knowledge where he may happily hunt for the roots of the
Riemann-Zeta function. The “success” theory naturally justifies itself
to the physicist. He employs this type of activity of the brain because
it leads him to what he wants—a verifiable prediction as to the
external world—and for that reason he esteems it. Why should not
the theologian employ and esteem one of the mental processes of
unreason which leads to what he wants—an assurance of future
bliss, or a Hell to frighten us into better behaviour? Understand that I
do not encourage theologians to despise reason; my point is that
they might well do so if it had no better justification than the
“success” theory.
And so my own concern lest I should have been talking nonsense
ends in persuading me that I have to reckon with something that
could not possibly be found in the physical world.
Another charge launched against these lectures may be that of
admitting some degree of supernaturalism, which in the eyes of
many is the same thing as superstition. In so far as supernaturalism
is associated with the denial of strict causality (p. 309) I can only
answer that that is what the modern scientific development of the
quantum theory brings us to. But probably the more provocative part
of our scheme is the rôle allowed to mind and consciousness. Yet I
suppose that our adversary admits consciousness as a fact and he
is aware that but for knowledge by consciousness scientific
investigation could not begin. Does he regard consciousness as
supernatural? Then it is he who is admitting the supernatural. Or
does he regard it as part of Nature? So do we. We treat it in what
seems to be its obvious position as the avenue of approach to the
reality and significance of the world, as it is the avenue of approach
to all scientific knowledge of the world. Or does he regard
consciousness as something which unfortunately has to be admitted
but which it is scarcely polite to mention? Even so we humour him.
We have associated consciousness with a background untouched in
the physical survey of the world and have given the physicist a
domain where he can go round in cycles without ever encountering
anything to bring a blush to his cheek. Here a realm of natural law is
secured to him covering all that he has ever effectively occupied.
And indeed it has been quite as much the purpose of our discussion
to secure such a realm where scientific method may work
unhindered, as to deal with the nature of that part of our experience
which lies beyond it. This defence of scientific method may not be
superfluous. The accusation is often made that, by its neglect of
aspects of human experience evident to a wider culture, physical
science has been overtaken by a kind of madness leading it sadly
astray. It is part of our contention that there exists a wide field of
research for which the methods of physics suffice, into which the
introduction of these other aspects would be entirely mischievous.
A besetting temptation of the scientific apologist for religion is to
take some of its current expressions and after clearing away
crudities of thought (which must necessarily be associated with
anything adapted to the everyday needs of humanity) to water down
the meaning until little is left that could possibly be in opposition to
science or to anything else. If the revised interpretation had first
been presented no one would have raised vigorous criticism; on the
other hand no one would have been stirred to any great spiritual
enthusiasm. It is the less easy to steer clear of this temptation
because it is necessarily a question of degree. Clearly if we are to
extract from the tenets of a hundred different sects any coherent

Logical Introduction To Probability and Induction Franz Huber Full Chapter PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Logical Introduction To Probability and Induction Franz Huber Full Chapter PDF

Uploaded by

Copyright:

Available Formats

Logical Introduction to Probability and

Induction Franz Huber

Belief and Counterfactuals: A Study in Means-End

Introduction to Probability and Statistics Metric

Introduction to Probability and Statistics (Metric

Introduction to Probability Models 12th Edition Sheldon

Introduction to Linear Regression Analysis (Wiley

Gestión de los cuidados enfermeros y liderazgo (Spanish

Franz Kafka and Chinese Culture Yanbing Zeng

Doing Philosophy: From Common Curiosity to Logical

Published in the United States of America by Oxford University Press

© Oxford University Press 2019

All rights reserved. No part of this publication may be reproduced, stored in

You must not circulate this work in any other form

CIP data is on file at the Library of Congress

Paperback printed by WebCom, Inc., Canada

The instructor’s manual can be found online at the following webaddress:

4. Deductive Approaches to Confirmation 47

6. The Classical Interpretation of Probability 98

7. The Logical Interpretation of Probability 111

7.5. The new riddle of induction 132

8. The Subjective Interpretation of Probability 139

9. The Chance Interpretation of Probability 187

10. The (Limiting) Relative Frequency

10.9. Inferential statistics 243

11. Alternative Approaches to Induction 259

A Logical Introduction to Probability and Induction is an

well as Kolmogorov’s axiomatization of the probability calculus.

The last week of a course is usually best spent reviewing

I am very grateful to Claus Beisbart, Joseph Berkovitz, Michael

1.1 PROPOSITIONAL LOGIC

The sentence “Angela Merkel is chancellor of Germany in August

Toronto is a city. Toronto is a city. It is not the case that

Is there one sentence in the above line, or are there two

p Angela Merkel is chancellor of Germany in August 2017.

Sentence tokens can be combined, or connected, to form more

Negation: ¬ it is not the case that . . .

Officially we define a formal language L for propositional logic

1. Every propositional variable, or sentence letter, ‘p,’ ‘q,’

Speakers use languages to talk about things. Those things can

another language, the former has the latter as an object language.

In the first sentence, we are using the word ‘Toronto.’ In the

Now consider again the recursive definition above and, in

are symbols of the metalanguage that we use to talk about

However, as mentioned, logicians are lazy. They have introduced

L This sentence is false.

If the sentence L in the line above is true, then it is false, and

This sentence contains five words.

Try to think about which feature, in addition to self-

Negation: A negated sentence (¬α) is true just in case the

Unfortunately, the meaning of many English conditionals, or

1.2 PREDICATE LOGIC

Sentences talk about objects, the properties these objects have,

Properties of one object are referred to by predicates, or

F . . . is chancellor of Germany in August 2017

A predicate and an individual constant can be combined to

F (a) Angela Merkel is chancellor of Germany in August 2017.

It is customary to identify predicate symbols with unary

I have not included a right column for individual variables

Existential quantifier: ∃x there exists an x

When we translate English sentences into well-formed formulas

the predicate-logical form of all sentences clear. For instance,

There is a Republican U.S. president in August 2017.