You are on page 1of 53

Lectures on Inductive Logic 1st Edition

Jon Williamson
Visit to download the full and correct content document:
https://textbookfull.com/product/lectures-on-inductive-logic-1st-edition-jon-williamson/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Inductive Logic Programming 29th International


Conference ILP 2019 Plovdiv Bulgaria September 3 5 2019
Proceedings Dimitar Kazakov

https://textbookfull.com/product/inductive-logic-
programming-29th-international-conference-ilp-2019-plovdiv-
bulgaria-september-3-5-2019-proceedings-dimitar-kazakov/

Lectures on Astrophysics 1st Edition Steven Weinberg

https://textbookfull.com/product/lectures-on-astrophysics-1st-
edition-steven-weinberg/

Lectures on Imagination 1st Edition Paul Ricœur

https://textbookfull.com/product/lectures-on-imagination-1st-
edition-paul-ricoeur/

Lectures on geometry Woodhouse

https://textbookfull.com/product/lectures-on-geometry-woodhouse/
Lectures on New Testament Theology 1st Edition Baur

https://textbookfull.com/product/lectures-on-new-testament-
theology-1st-edition-baur/

Lectures on Lyapunov Exponents 1st Edition Marcelo


Viana

https://textbookfull.com/product/lectures-on-lyapunov-
exponents-1st-edition-marcelo-viana/

Network Flow Algorithms 1st Edition Williamson

https://textbookfull.com/product/network-flow-algorithms-1st-
edition-williamson/

Lectures on Convex Optimization Yurii Nesterov

https://textbookfull.com/product/lectures-on-convex-optimization-
yurii-nesterov/

Twenty Lectures on Algorithmic Game Theory 1st Edition


Tim Roughgarden

https://textbookfull.com/product/twenty-lectures-on-algorithmic-
game-theory-1st-edition-tim-roughgarden/
LECTURES ON INDUCTIVE LOGIC
Lectures on Inductive Logic

JON WILLIAMSON
Professor of Reasoning, Inference and Scientific Method, University of Kent

3
3
Great Clarendon Street, Oxford, OX2 6DP,
United Kingdom
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
© Jon Williamson 2017
The moral rights of the author have been asserted
First Edition published in 2017
Impression: 1
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 licence or under terms agreed with the appropriate reprographics
rights organization. Enquiries 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
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America
British Library Cataloguing in Publication Data
Data available
Library of Congress Control Number: 2016953026
ISBN 978–0–19–966647–8
Printed and bound by
CPI Group (UK) Ltd, Croydon, CR0 4YY
Links to third party websites are provided by Oxford in good faith and
for information only. Oxford disclaims any responsibility for the materials
contained in any third party website referenced in this work.
‘Tell me, what drove thee from the Table Round,
My brother? was it earthly passion crost?’
‘Nay,’ said the knight;‘for no such passion mine.
But the sweet vision of the Holy Grail
Drove me from all vainglories, rivalries,
And earthly heats that spring and sparkle out
Among us in the jousts’
( from The Holy Grail by Alfred Lord Tennyson)
PREFACE

Inductive logic seeks to determine the extent to which the premisses of an argument entail
its conclusion. This book offers an introduction to the topic and develops a new account of
inductive logic which appeals to developments in Bayesian epistemology (a theory about
how strongly one should believe the various propositions that one can express).
Chapter 1 introduces perhaps the simplest and most natural account of inductive lo-
gic, classical inductive logic, which is attributable to Ludwig Wittgenstein. An inductive logic
needs to capture two kinds of partial entailment: the first kind, logical entailment, quantifies
the extent to which premisses entail a conclusion in virtue of logical connections between
them; the second kind, inductive entailment, quantifies the extent to which a sample of past
observations entails the next outcome to be sampled. Unfortunately, as we shall see, clas-
sical inductive logic fails to adequately capture inductive entailment. Thus there is a need
to develop more sophisticated inductive logics.
Chapter 2 presents enough logic and probability theory for the reader to begin to study
inductive logic, while Chapter 3 introduces the ways in which logic and probability can be
combined in an inductive logic. Then, in Chapter 4, we encounter the most influential ap-
proach to inductive logic, due to W.E. Johnson and Rudolf Carnap. We see that while this
approach can account for inductive entailment, it does so at the expense of logical entail-
ment. Thus we remain in need of an inductive logic that captures both logical and inductive
entailment.
In Chapter 5 we see how an alternative approach to inductive logic follows rather nat-
urally from the philosophical theory of objective Bayesian epistemology. This approach
captures logical entailment because it extends classical inductive logic (Chapter 6). It cap-
tures inductive entailment because of its connections with statistical theory (Chapter 7).
Chapter 8 defends the approach by tackling several key criticisms that are often levelled at
inductive logic. Chapter 9 presents a formal justification of the objective Bayesian approach.
We take stock in Chapter 10.
This book is aimed at researchers and students who have some knowledge of elemen-
tary logic and probability, and who are familiar with mathematical reasoning and notation.
The material presented in this book is well suited to being delivered as a series of lec-
tures. An undergraduate course to philosophy students might omit the starred sections,
which are less central to the main thrust of the argument and can be more technical, and
might skim over some other mathematical details. For graduate students in philosophy,
these lectures could be given in conjunction with a more detailed consideration of Bayesian
epistemology (see, e.g., Williamson, 2010b) and other philosophical notions of probability
(Gillies, 2000). Alternatively, the material in this book could be treated alongside that of
Howson (2000) to provide a thorough introduction to induction from both the subjective
Bayesian perspective (Howson’s view) and the objective Bayesian perspective developed
viii | PREFACE

here. For mathematics students, these lectures could be given in conjunction with material
from Paris and Vencovská (2015), which is a work in the Johnson–Carnap tradition, and
the lectures could be extended by filling in the material sketched in §9.4.2, which can be
found in Landes and Williamson (2013, 2015). For students in computing, these lectures
could be augmented by a thorough discussion of inference in inductive logic using Bayesian
networks (Haenni et al., 2011; Korb and Nicholson, 2003; Neapolitan, 2004; Williamson,
2005).
ACKNOWLEDGEMENTS

I am grateful to the UK Arts and Humanities Research Council for supporting the research
project From objective Bayesian epistemology to inductive logic; that research, conducted in
collaboration with Teddy Groves and Jürgen Landes, informs much of this book. Lectures
based on the material in this book have been given at the fifth Indian School on Logic and
its Applications, the 2015 Spring School on Combining Probability and Logic, and to phil-
osophy students at the University of Kent. I am grateful to the audiences for their feedback.
The book has also benefited immensely from discussions with and comments from David
Corfield, Teddy Groves, Jürgen Landes, Jeff Paris, Soroush Rad, Christian Wallmann and
Michael Wilde.
I would like to thank Dan Taber, Keith Mansfield and all the team at OUP for ensuring
that the process of producing this book ran smoothly.

This book is dedicated to Kika, Charlie and Anouk, with thanks for everything.
CONTENTS

1 Classical Inductive Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1


1.1 From Deductive to Inductive Logic 1
1.2 Patterns of Partial Entailment and Support 3
1.2.1 The Fundamental Inductive Pattern 3
1.2.2 Diminishing Returns 5
1.2.3 Examining a Possible Ground 6
1.2.4 Analogy 7
1.3 Why Inductive Logic? 8
1.3.1 Decision Making 8
1.3.2 Artificial Intelligence 10
1.3.3 The GRAIL Quest 11
1.4 Learning from Experience 11
1.5 Inductive Entailment and Logical Entailment 13
2 Logic and Probability . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 16
2.1 Propositional Logic 16
2.2 Predicate Logic 17
2.3 Probability over Logical Languages 18
2.3.1 Axioms of Probability 18
2.3.2 Properties of Probability 19
2.3.3 Truth Tables and Probability 21
2.3.4 Conditional Probability and Inductive Logic 22
2.4 Entropy, Divergence and Score 25
2.5 Interpretations of Probability 31
2.6 ∗ Probability over Fields of Sets 32
2.6.1 Fields of Sets 32
2.6.2 Axioms of Probability 33
2.6.3 The Valuation Space 34
3 Combining Probability and Logic. . . . . . . . . . . . . . . . . . . . . . . . . . . 40
3.1 Entailment 40
3.2 Support and Consistency 45
3.3 The Languages of Inductive Logic 46
3.4 Inductive Qualities 47
3.5 Probabilistic Logics 49
3.6 * More Examples of Inductive Logics 55


The starred sections contain material which is less central to the main thrust of the argument and which can be
more technical.
xii | CONTENTS

4 Carnap’s Programme . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 59
4.1 Conditionalizing on a Blank Slate 59
4.2 Pure and Applied Inductive Logic 61
4.3 Conditionalization 63
4.4 The Permutation Postulate 65
4.5 The Principle of Indifference 68
4.6 Which Value in the Continuum? 71
4.7 Which Continuum of Inductive Methods? 72
4.8 Capturing Logical Entailment 72
4.9 Summary 74
5 From Objective Bayesian Epistemology to Inductive Logic . . . . . . . . . 75
5.1 Objective Bayesian Epistemology 75
5.2 * Objective versus Subjective Bayesian Epistemology 77
5.3 Objective Bayesian Inductive Logic 81
5.4 * Language Invariance 85
5.5 * Finitely Generated Evidence Sets 91
5.6 Updating, Expansion and Revision 94
5.6.1 Maxent and Conditionalization 96
5.6.2 Maxent and KL-updating 101
5.7 Summary 103
6 Logical Entailment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
6.1 Truth Tables with Probabilities 105
6.2 Logical Irrelevance Revisited 108
6.3 Context and Chance Constraints 111
6.4 Constraints on Conditional Probabilities 115
6.5 Revision Under Constraints 118
6.6 Lottery and Preface Paradoxes Revisited 120
6.7 The Fundamental Inductive Pattern Revisited 122
6.7.1 A Surprising Consequence 122
6.7.2 An Otherwise Surprising Consequence 123
6.7.3 A Plausible Consequence 124
6.8 * Inferences in Predicate Inductive Logic 126
7 Inductive Entailment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
7.1 Syntactic Relevance 134
7.2 The Calibration Norm 135
7.3 Extended Example 138
7.4 Is this Application of Confidence Intervals Legitimate? 142
7.5 Uniqueness of the Interval 145
7.6 Loss of Information 146
7.7 Generalization 147


The starred sections contain material which is less central to the main thrust of the argument and which can be
more technical.
CONTENTS | xiii

8 Criticisms of Inductive Logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 151


8.1 Language Invariance Revisited 151
8.2 Goodman’s New Problem of Induction 154
8.3 The Principle of Indifference Revisited 159
8.4 Universal Hypotheses 162
8.5 Summary 166
9 Justification . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 167
9.1 Two Problems of Induction 167
9.2 Two Principles of Rationality 168
9.3 Minimal Worst-Case Expected Loss 175
9.4 * Robustness of the Minimax Theorem 180
9.4.1 Key Assumptions 180
9.4.2 Rationality Principles 182
10 Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
10.1 Have we Found the GRAIL? 187
10.2 Open Questions 189
10.2.1 Knowledge Engineering 189
10.2.2 Other Questions 191

References 193
Index 199


The starred sections contain material which is less central to the main thrust of the argument and which can be
more technical.
1

Classical Inductive Logic

In this chapter we shall examine a rather natural and intuitive inductive logic, which we shall
call classical inductive logic, as a means to acquaint ourselves with the goals of inductive logic.
Although it will turn out (in §1.4) that classical inductive logic is not altogether viable, this
logic will provide a useful platform upon which to develop a more successful inductive logic
later on in the book.

1.1 From Deductive to Inductive Logic


One approach to inductive logic is via deductive logic. In deductive logic an argument is
valid just when its premisses force the truth of its conclusion, that is, if and only if the con-
clusion is true under each of the possible assignments of truth values to the basic (‘atomic’)
sentences that make the premisses true.
Consider the following argument in propositional logic: if a then b; b; therefore a. This
may be represented:
a→b
b
a
We can ask whether the argument is deductively valid:

a → b, b | a?

Another way of putting this question is: do the premisses a → b, b deductively entail (i.e.
force the truth of) the conclusion a?
We know the argument is invalid by considering its truth table:
a b a→b b a
T T T T T
T F F F T
F T T T F
F F T F F

Lectures on Inductive Logic. Jon Williamson.


© Jon Williamson 2017. Published 2017 by Oxford University Press.
2 | CLASSICAL INDUCTIVE LOGIC

The two left-most columns go through all the truth assignments to the atomic sentences,
while the next two columns give the corresponding truth values of the premisses and the
right-most column gives the corresponding truth value of the conclusion. The argument
is invalid because the third truth assignment makes the premisses true but the conclusion
false.
Hence this argument, which is often called affirming the consequent, is deemed fallacious,
and we write:

a → b, b  | a.

While the argument seems a poor one from a deductive point of view, we can ask: to what
extent is the conclusion plausible, given the premisses? This can be viewed as a question of
partial entailment: to what extent do the premisses entail the conclusion? If, in the presence
of the premisses, some level of plausibility y attaches to the conclusion, then we can write:

a → b, b |≈ ay .

Here |≈ is used to distinguish an inductive entailment relation from deductive


entailment |.
There is also another sort of question we can ask: do one or more of the premisses make
the conclusion more plausible than it is in their absence? If so, to what extent? Here we are
asking the extent to which the premisses support the conclusion. While partial entailment is
an absolute notion, support is a relative notion. If the premisses entail the conclusion to de-
gree y, that is, a → b, b |≈ ay , and the conclusion has degree of plausibility z in the absence
of the premisses, that is, |≈ az , then the premisses support the conclusion just when y > z.
Moreover, the premisses support the conclusion to the extent to which y goes beyond z.
One view is that degree of partial entailment can be measured by the proportion of all
those truth assignments that make the premisses true which also make the conclusion true.
Ludwig Wittgenstein put forward this view in his Tractatus Logico-Philosophicus:

If Tr is the number of the truth-grounds of a proposition ‘r’, and if Trs the number of
the truth-grounds of a proposition ‘s’ that are at the same time truth-grounds of ‘r’,
then we call the ratio Trs : Tr the degree of probability that the proposition ‘r’ gives to
the proposition ‘s ’.
(Wittgenstein, 1922, §5.15)

Wittgenstein also noted that this proportion can be read off a standard truth table
(Wittgenstein, 1922, §5.151). Consider our example:

a b a→b b a
T T T T T
T F F F T
F T T T F
F F T F F
PATTERNS OF PARTIAL ENTAILMENT AND SUPPORT | 3

Here, one out of the two truth assignments that make the premisses true also makes the
conclusion true, so
a → b, b |≈ a1/2 .

We shall call the inductive logic with this notion of partial entailment classical inductive
logic, or CIL for short. We use this ‘classical’ for three reasons. First, it is perhaps the most
natural and basic inductive logic. Second, it was put forward by Wittgenstein at the same
time that he put forward the truth-table method for deciding propositional validity in what
is now known as classical deductive logic. Third, this inductive logic follows very naturally
from what is generally known as the classical interpretation of probability:

The theory of chance consists in reducing all the events of the same kind to a certain
number of cases equally possible, that is to say, to such as we may be equally undecided
about in regard to their existence, and in determining the number of cases favorable
to the event whose probability is sought. The ratio of this number to that of all the
cases possible is the measure of this probability, which is thus simply a fraction whose
numerator is the number of favorable cases and whose denominator is the number of
all the cases possible.
(Laplace, 1814, pp. 6–7)

The classical interpretation of probability is discussed in more detail by Gillies (2000,


Chapter 2).

1.2 Patterns of Partial Entailment and Support


In order to get a feel for classical inductive logic, we shall take a look at some examples
of partial entailment and support. Several of these were put forward by George Polya in
his book Patterns of Plausible Inference. Now, Polya was somewhat sceptical regarding the
precise numerical measurement of degrees of partial entailment (Polya, 1954, p. 132).
While Jaynes (2003, §5.2.2) suggests that Polya’s scepticism arose from a mistake in his
calculations which led him to believe erroneously that numerical methods give the wrong
results, the fact remains that Polya thought of his patterns of inference as purely qualitative.
Nevertheless, we can ask whether quantitative measures of partial entailment and support
validate these patterns: do the patterns turn out to be warranted in our preferred quantita-
tive inductive logic? As we shall see, classical inductive logic does indeed validate several of
these patterns, which speaks in favour of CIL.

1.2.1 The Fundamental Inductive Pattern


Polya (1954, §12.1) put forward what he calls the fundamental inductive pattern:

a implies b
b true
a more credible
4 | CLASSICAL INDUCTIVE LOGIC

The key point to note about this inference is that the conclusion says that a becomes
more credible; it is thus asserting that there is a change in degree of partial entailment, that
is, a relation of support.
We saw above that the premisses entail the conclusion to degree 12 :

a → b, b |≈ a1/2 .

In the absence of the premisses, we have the following truth table:

a b a
T T T
T F T
F T F
F F F

Two out of the four truth assignments render the conclusion true. Hence,

|≈ a1/2 .

Thus the premisses neither support nor undermine the conclusion—they fail to move its
probability away from 12 . The premisses are neutral with respect to the conclusion, written:

a → b, b ⊥
⊥ a.

It might appear, then, that classical inductive logic fails to validate the fundamental in-
ductive pattern. But Polya actually interprets the fundamental inductive pattern in a slightly
different way. He says, ‘the verification of a consequence renders a conjecture more cred-
ible’ (Polya, 1954, p. 5). This can be interpreted as: given a → b, learning b makes a more
credible. A very similar inference scheme was also advocated by Jaynes (2003, p. 4), who
gives much the same interpretation as Polya. Hempel (1945, p. 103) called this inference
pattern the entailment condition, and gives a similar interpretation: ‘Any sentence which is
entailed by an observation report is confirmed by it.’ Hempel argues that any reasonable
concept of partial entailment should satisfy this condition.
Thus the question is whether b changes the degree to which a is entailed, when a → b is
held fixed. If so, then b does support a relative to a → b and we can write:

[a → b], b  a.

Here the square brackets specify what is held fixed, while  denotes positive support.
Now,
a → b |≈ a1/3 .

This can be seen by the fact that one out of the three truth assignments that satisfy a → b
also satisfies a:
PATTERNS OF PARTIAL ENTAILMENT AND SUPPORT | 5

a b a→b a
T T T T
T F F T
F T T F
F F T F

On the other hand, we saw that a → b, b |≈ a1/2 . Hence b raises the degree of partial
entailment of a from 1/3 to 1/2, and we do indeed have that:

[a → b], b  a.

Note that the maximum amount that the degree of partial entailment could be raised is
1 – 1/3 = 2/3. So b raises it by 1/2–1/3
2/3
= 25% of the maximum amount. This allows us to
quantify the degree to which b supports a, relative to a → b, and we can write:

[a → b], b 25% a.

In sum, then, classical inductive logic does indeed validate the fundamental inductive
pattern.

1.2.2 Diminishing Returns


Consider a variant of the fundamental inductive pattern, with a ∧ c in place of a:

(a ∧ c) → b
b
a∧c

This has truth table:

a b c (a ∧ c) → b b a∧c
T T T T T T
T T F T T F
T F T F F T
T F F T F F
F T T T T F
F T F T T F
F F T T F F
F F F T F F

The truth table shows several things. First, this is an invalid argument (lines 2,5,6).
Second, lines 1,2,5,6 show that:

(a ∧ c) → b, b |≈ a ∧ c1/4 .
6 | CLASSICAL INDUCTIVE LOGIC

Third, lines 1,2,4,5,6,7,8 show that:

(a ∧ c) → b |≈ a ∧ c1/7 .

So b supports a ∧ c, but not by as much as in the case of the fundamental inductive pattern
considered in §1.2.1:

[(a ∧ c) → b], b 12.5% a ∧ c.

This may be explained by the fact that classical inductive logic considers a ∧ c inherently
less plausible than a: we have that |≈ a ∧ c1/4 while |≈ a1/2 .

1.2.3 Examining a Possible Ground


Polya (1954, §13.2) put forward the following inference pattern:

a implied by b
b false
a less credible

Polya expressed the inference thus: ‘our confidence in a conjecture can only diminish
when a possible ground for the conjecture is exploded’ (Polya, 1954, p. 20). In the inference,
a is the conjecture and b is the ground for the conjecture.
Here is the truth table for this inference:

a b b→a ¬b a
T T T F T
T F T T T
F T F F F
F F T T F

Lines 2 and 4 show that:

b → a, ¬b |≈ a1/2 .

On the other hand, lines 1,2,4 show that:

b → a |≈ a2/3 .

So the premiss ¬b undermines the conclusion a:

[b → a], ¬b 25% a.

Thus classical inductive logic does indeed validate this inductive pattern of Polya’s.
PATTERNS OF PARTIAL ENTAILMENT AND SUPPORT | 7

1.2.4 Analogy
Polya (1954, §13.9) proposes:

a analogous to b
b true
a more credible

Here ‘a is analogous to b’ is understood as there being some common ground g, such that
g → a and g → b. That is,

g→a
g→b
b true
a more credible

Classical inductive logic validates such an inference, as we shall now see. First,

g → a, g → b, b |≈ a2/3 .

This can be seen from the truth table:

g a b g→a g→b b a
T T T T T T T
T T F T F F T
T F T F T T F
T F F F F F F
F T T T T T T
F T F T T F T
F F T T T T F
F F F T T F F

The above truth table also shows that:

g → a, g → b |≈ a3/5 .

Hence,

[g → a, g → b], b 16.6̇% a.

Note that classical inductive logic also validates the inference if the first two premisses
are not held fixed:

|≈ a1/2 ,
8 | CLASSICAL INDUCTIVE LOGIC

so

g → a, g → b, b 33.3̇% a.

Exercise
(i) To what extent does a ∨ b entail a?
(ii) To what extent do the premisses of the following argument entail its conclusion? To
what extent do they support the conclusion?
(a ∨ b) → c
(a ↔ c) → b
c

1.3 Why Inductive Logic?


Let us recap. Deductive logic offers a very coarse grained test of the cogency of an argument:
for premisses to deductively entail a conclusion, they must force the truth of the conclusion.
In contrast, an inductive logic such as CIL offers a much more fine-grained assessment:
we can see the extent to which premisses entail a conclusion, and can determine the extent
to which certain premisses support or undermine the conclusion, relative to other prem-
isses that are held fixed. This more nuanced analysis of an argument offers a key theoretical
motivation for investigating inductive logic, and we have seen that classical inductive logic
appears promising insofar as it captures some important patterns of inductive reasoning.
But there are other, more practical, reasons for developing a viable inductive logic, and
we shall consider some of them in this section.

1.3.1 Decision Making


Suppose that a consultant needs to decide which of two treatments to give a cancer patient
after surgery and a previous round of treatment. They can give chemotherapy or radiother-
apy. Chemotherapy is the more aggressive treatment, with harsher side-effects, and is most
appropriate if the patent’s cancer will recur in the absence of treatment. Here is a table repre-
senting the utility of each treatment according to whether the patient’s cancer will otherwise
recur (r) or not (¬r):

r ¬r
chemotherapy +6 –2
radiotherapy +4 –1

Because chemotherapy is more likely to be successful in killing remaining cancerous


cells, it has higher positive utility should the cancer otherwise recur. On the other hand,
it his more negative utility if the cancer will otherwise not recur, because it has the harsher
side-effects. Of course, the consultant normally does not know for sure whether or not the
WHY INDUCTIVE LOGIC ? | 9

cancer will otherwise recur. Bayesian decision theory, which is arguably our best guide to
decision making in such a scenario, says that the consultant should choose the act which
has maximum expected utility. Thus they should choose chemotherapy if

6P(r) – 2P(¬r) > 4P(r) – 1P(¬r),

where P(r) and P(¬r) are the probabilities of r and ¬r respectively. According to the rules
of probability, P(¬r) = 1 – P(r), so they should choose recurrence if

6P(r) – 2 + 2P(r) > 4P(r) – 1 + P(r),

that is, if

P(r) > 1/3.

The key task that remains, then, is to determine P(r), the probability of recurrence, from
available evidence.
Suppose, for example, that the consultant’s evidence consists of the facts that the cancer
metastasized (m) and that metastasis together with the presence of a particular biological
marker (b) ensure recurrence. (In reality, of course, a consultant’s evidence will be much
more complex, but this simplified example will suffice to illustrate the role, if not the extent,
of the available evidence. A more realistic version will be explored at the end of §6.4.) The
task is to determine what degree of plausibility the evidence gives to r. This is really just a
question of inductive logic:

m, m ∧ b → r |≈ r? .

According to classical inductive logic, we need to build a truth table to answer this
question:

m b r (m ∧ b) → r m r
T T T T T T
T T F F T F
T F T T T T
T F F T T F
F T T T F T
F T F T F F
F F T T F T
F F F T F F

Two of the three truth assignments that satisfy the premisses also satisfy the
conclusion, so

m, m ∧ b → r |≈ r2/3 .
10 | CLASSICAL INDUCTIVE LOGIC

Now 2/3 > 1/3, so giving chemotherapy maximizes expected utility and the consultant
should choose to give the patient chemotherapy.
In sum, inductive logic is essential to decision making. Decision making requires ascer-
taining the probabilities of the various possible outcomes that make a difference to the
utilities of the available acts. These probabilities are not normally a part of one’s evidence,
but need to be derived from the available evidence. It is inductive logic that tells us how to
derive these probabilities: inductive logic tells us the extent to which evidence entails the
various outcomes relevant to the decision problem. Thus we need to investigate whether
there is some inductive logic—such as classical inductive logic—that stands out as being
most appropriate.

1.3.2 Artificial Intelligence


Since the early days of artificial intelligence research, inductive logic has been considered
essential to machine reasoning:

Something like Carnap’s theory [of inductive logic] would be required if an electronic
reasoning machine is ever built.
(Good, 1950, p. 48)

This is because the notion of partial entailment has such wide applicability in artificial
intelligence. Consider a partial entailment relation of the form:

ϕ1 , . . . , ϕk |≈ ψ y .

In a medical decision support system, ϕ1 , . . . , ϕk might be evidence concerning prognos-


tic factors of a patient and their relationship to cancer, and ψ might assert recurrence of
the patient’s cancer as per the above example (§1.3.1). Or, to take another example, a doc-
tor may need to decide whether to prescribe statins, and a medical decision support system
may need to determine the extent to which the hypothesis ψ that the patient will other-
wise develop cardiovascular disease is entailed by information ϕ1 , . . . , ϕk about the drug
and its effects, and about the patient (e.g. cholesterol levels and blood pressure). MYCIN,
an expert system for the diagnosis and treatment of infections, was one of the earliest med-
ical systems to incorporate numerical ‘certainty factors’ (Shortliffe and Buchanan, 1975;
Szolovits and Pauker, 1978; Heckerman, 1986).
In robotics, ψ might be the hypothesis that a robot is in a certain location, with ϕ1 , . . . , ϕk
constituting background information about its environment together with observations
from its sensory apparatus (see, e.g., Thrun et al., 2005).
In financial systems, a bank system may need to decide whether to give a loan. ψ might
be the hypothesis that the money will be paid back on time, and ϕ1 , . . . , ϕk might contain
general bank rules and the individual’s banking history.
In bioinformatics, ψ may be a hypothesis about genetic linkage in the fruit fly Drosophila,
to be assessed on the basis of DNA data and background knowledge about the genetic
mechanisms ϕ1 , . . . , ϕk (Neapolitan, 2009).
LEARNING FROM EXPERIENCE | 11

In natural language processing, ψ might be the hypothesis that a word has a certain
meaning, while ϕ1 , . . . , ϕk might contain information about the language as well as con-
textual information about the words preceding the word in question (Berger et al., 1996).
There are countless other potential applications of inductive logic in artificial intelli-
gence. Consequently, a viable inductive logic would be of enormous interest to researchers
in these fields.

1.3.3 The GRAIL Quest


It should be clear by now that the prospect of a viable inductive logic is extremely attract-
ive. The search for a viable inductive logic can be thought of as the quest for a General,
Reasonable, Applicable Inductive Logic, or GRAIL for short. The logic should be general
insofar as it should be able to express a broad range of inferences. It should be reasonable
in that it should be well motivated and its inferences should accord with our inductive in-
tuitions. It should also be applicable: we should be able to use such a logic for assessing the
strength of deductively invalid arguments, for decision making and in artificial intelligence,
for example.
Classical inductive logic appears general, reasonable and applicable, but it is limited in
certain important respects. The fact that the truth-table method can be used to draw in-
ferences suggests that classical inductive logic should be easily applicable. Realistically,
however, truth-table methods can only be applied to small problems, as the size of the truth
table grows exponentially with the number of propositional variables under consideration.
CIL is also quite general insofar as it can cope with any inference in propositional logic;
moreover, as we shall see in Chapter 4, there is an inductive logic c∞ which extends CIL to
monadic predicate logic. Yet CIL is perhaps insufficiently general, in that it can only cope
with categorical premisses. Recall the cancer recurrence example in §1.3.1: in practice a
consultant’s evidence will not be easily expressed as categorical sentences of propositional
logic or monadic predicate logic—it will contain statistical information, for instance. We
have seen in §1.2 that CIL yields several inferences, such as the inferences put forward by
Polya, that are reasonable in that they accord with intuition. Yet classical inductive logic
yields certain other inferences that are by no means reasonable, as we shall see next. The
GRAIL is not so easily obtained.

1.4 Learning from Experience


Carnap (1945, p. 81) argued that classical inductive logic is inadequate because it fails to
capture the phenomenon of learning from experience.
Suppose that we are examining ravens and seeing whether or not they are black. Let Bri
stand for the proposition that the ith observed raven is black. Classical inductive logic will
attach degree of partial entailment 12 to the proposition that the hundred and first raven is
black:

|≈ Br101
1/2
.
12 | CLASSICAL INDUCTIVE LOGIC

This is because half the truth assignments make this proposition true.

Br101
T
F

This degree of partial entailment seems reasonable enough, in the absence of any evi-
dence that indicates otherwise. But classical inductive logic will say that, in the presence of
premisses which assert just that 100 ravens have all been found to be black, the degree of
partial entailment remains 12 :

Br1 , . . . , Br100 |≈ Br101


1/2
.

This is because one out of the two truth assignments that make the premisses true also
makes the conclusion true:

Br1 Br2 ... Br100 Br101


T T ... T T
T T ... T F
T T ... F T
T T ... F F
... ... ... ... ...

Thus, according to classical inductive logic the premisses are neutral with respect to the
conclusion, that is, they neither support nor undermine the conclusion:

Br1 , . . . , Br100 ⊥
⊥ Br101 .

This seems totally unreasonable. Observing 100 ravens and finding them all to be black
should lend some support to the conclusion that the next raven will be found to be black.
This inability to capture learning from experience is a clear deficiency of classical inductive
logic.
Although Carnap highlighted the significance of this problem for inductive logic, a simi-
lar point had previously been made by George Boole, in the context of drawing balls from
an urn with replacement:

It follows, therefore, that if the number of balls be infinite, and all constitutions of the
system be equally probable, the probability of drawing m white balls in succession will
be 21m , and the probability of drawing m + 1 white balls in succession 2m+1 1
; whence
the probability that after m white balls have been drawn, the next drawing will furnish
a white one, will be 12 . In other words, past experience does not in this case affect future
expectation.
(Boole, 1854, pp. 371–2)
INDUCTIVE ENTAILMENT AND LOGICAL ENTAILMENT | 13

Wittgenstein pointed out that this phenomenon is due to the propositions being logically
independent, that is, having no propositional variables in common, when viewed from the
perspective of propositional logic:

Propositions which have no truth-arguments in common with one another we call


independent.
Independent propositions (e.g. any two elementary propositions) give to one
another the probability 12 .
(Wittgenstein, 1922, §5.152)

Classical inductive logic appeals to truth-table methods, which are only suitable for de-
ciding entailment in propositional logic. Thus classical inductive logic treats the Bri as
distinct propositional variables. Since these propositional variables have no logical struc-
ture from the point of view of propositional logic, classical inductive logic cannot capture
any connection between them. Because the premisses are logically independent of the con-
clusion, classical inductive logic cannot but deem them to be neutral with respect to the
conclusion. This deficiency, then, points to an inadequacy of propositional inductive logic,
and motivates a move to predicate logic to try to capture the connection between the Bri .
This was essentially what motivated Carnap’s programme for inductive logic.

1.5 Inductive Entailment and Logical Entailment


It is apparent, then, that an inductive logic needs to capture two kinds of partial entailment:

Logical Entailment. This is the degree to which premisses entail a conclusion by


virtue of the logical dependence between them. For example, a ∨ b supports
proposition a because they have propositional variables in common.
Inductive Entailment. This is the degree to which premisses entail a conclusion by
virtue of their involving samples from the same population. For example, finding
all sampled ravens to be black supports the proposition that the next sampled
raven will be black, if they are sampled in an appropriate way from the same
population of ravens.

That inductive logic needs to capture these two kinds of partial entailment motivates the
following slogan:

Partial entailment in inductive logic = inductive entailment + logical entailment.

Logical entailment is a non-ampliative concept of entailment, attributable to logical


dependence. Classical inductive logic seems to capture this concept very well, as was rec-
ognized by Wittgenstein, as well as Kemeny and Oppenheim (1952), who focused on
developing classical partial entailment to the exclusion of learning from experience. On
the other hand, inductive entailment is an ampliative concept that can link logically inde-
pendent propositions. Classical inductive logic fails to capture this concept. Thus Carnap
14 | CLASSICAL INDUCTIVE LOGIC

(1945) abandoned the classical notion of partial entailment to try to capture learning from
experience.
We see then that the field of inductive logic underwent a schism, with some researchers
focussing on logical entailment and others focussing on inductive entailment. In fact,
Wesley Salmon argued that it is not possible to capture both notions of partial entailment
in a single inductive logic. Here, Salmon uses ‘partial entailment’ to refer to what we call
‘logical entailment’, and ‘degree of confirmation’ for what we call ‘partial entailment’:

If degree of confirmation is to be identified with partial entailment, then c† [i.e. clas-


sical inductive logic] is the proper confirmation function after all, for it yields the result
that p is probabilistically irrelevant to q whenever p and q are completely independ-
ent and there is no partial entailment between them. . . . Unfortunately for induction,
statements strictly about the future (unobserved) are completely independent of
statements strictly about the past (observed). Not only are they deductively inde-
pendent of each other, but also they fail to exhibit any partial entailment. The force
of Hume’s insight that the future is logically independent of the past is very great in-
deed. It rules out both full entailment and partial entailment. If partial entailment were
the fundamental concept of inductive logic, then it would in fact be impossible to learn
from experience.
(Salmon, 1967a, pp. 731–2)

The dilemma seems to be as follows. In order for an inductive logic to capture logical
entailment, the following principle must hold:

Logical Irrelevance. If ϕ and ψ are logically independent then they are neutral with
respect to each other (i.e., neither changes the degree of partial entailment of
the other).

However, in order for an inductive logic to capture inductive entailment, Logical


Irrelevance must fail.
If this dilemma is correct, the prospect of a viable inductive logic—our GRAIL quest—
appears bleak. It appears that no inductive logic would be wholly reasonable: it would
capture one sort of entailment at the expense of capturing the other.
The core argument of this book is that there is something right about this dilemma and
something wrong with it. Some version of this irrelevance principle must plausibly hold if
an inductive logic is to capture logical entailment. Moreover, some version of this principle
must fail if an inductive logic is to capture inductive entailment. We shall see that this di-
lemma is enough to scupper both classical inductive logic, which captures logical entailment
to the exclusion of inductive entailment (§1.4), and Carnap’s programme for inductive lo-
gic, which captures inductive entailment to the exclusion of logical entailment (§4.8). But
the version of the irrelevance principle that must hold is slightly different from the version
that must fail, leaving just enough room to develop an inductive logic that successfully cap-
tures both kinds of partial entailment. A Bayesian inductive logic will be developed that, as
we shall see in Chapters 6 and 7, can claim to capture both inductive entailment and logical
entailment.
Another random document with
no related content on Scribd:
III

Kokonaisen kuukauden Žilin eli näin tovereineen. Isäntä yhä vain


naureskeli: »Jos sinu, Ivan, hyvä, — minu, Abdulin, hyvä». Mutta
ruoka oli kehnoa, — ainoastaan hapattamatonta hirssileipää joskus
hiukan kakkaraksi paistettua, toisinaan taas aivan pelkkää taikinaa.

Kostylin kirjoitti uudelleen kotiinsa, odotteli rahakirjettä ja ikävöi.


Päiväkaudet hän istui vajassa ja laski päiviä, milloin kirjeen pitäisi
saapua, tahi nukkui. Mutta Žilin tiesi, ettei hänen kirjeensä tullut
perille, eikä toista enää kirjoittanut.

»Mistä äitimuori semmoiset rahat ottaisi minun lunnaikseni», hän


mietti. »Tähän asti on enimmäkseen elänyt minun lähettämistäni
rahoista. Ja jos saisikin miten kokoon viisisataa, niin sitten olisi puilla
paljailla. Ehkäpä hyvässä lykyssä suoriudun tästä omin neuvoin.»

Kaikki paikat hän tutki ja tarkasteli, miettien miten pääsisi


karkaamaan.

Hän käveli vihellellen kartanolla, välistä istuutui jotakin


näpertelemään; teki savinukkeja ja punoi koreja vitsaksista. Žilin oli
ihmeen kätevä, oikea mestari kaikkeen tällaiseen.
Muovailipa hän kerran savesta nuken, jolla oli oikein nenä, kädet
ja jalat, vieläpä tatarilainen mekkokin yllä, ja pisti sen seisomaan
katolle.

Tatarilaistyttöjä kulki ohi, he menivät kaivolle. Talon tytär Dina


huomasi nuken ja kutsui toverinsakin sitä katsomaan. Kaikki laskivat
astiansa maahan, katsoivat ja nauroivat. Žilin otti nuken ja tarjosi sitä
heille, mutta he vain nauroivat, eivätkä uskaltaneet siihen koskea.
Hän pani sen takaisin katolle ja meni vajaan nähdäkseen, mitä
seuraisi.

Dina juoksi nuken luo, vilkaisi ympärilleen, sieppasi lelun ja livisti


tiehensä.

Aamulla päivän sarastaessa Žilin näki Dinan tulevan portaille


nukke kädessään. Nukke oli nyt koristettu punaisilla tilkuilla ja hän
tuuditteli sitä kuin lasta ainakin, hyräillen sille kehtolaulua.

Kohta tuli äitikin ulos, alkoi torua häntä, tempasi nuken hänen
kädestään, löi sen kappaleiksi ja lähetti Dinan jotakin toimittamaan.

Žilin teki vielä toisenkin nuken, entistä paremman, ja lahjoitti sen


Dinalle. Tyttö toi kerran ruukun Žilinille, laski astian maahan ja
istuutui katselemaan häntä. Hän osoitti nauraen astiaa.

»Mikähän sitä niin huvittaa?» ajattelee Žilin. Hän otti astian ja


rupesi juomaan luullen astiassa olevan vettä. Mutta siinä olikin
maitoa. Hän joi maidon ja virkkoi: Hyvää oli!» Tästäpä tyttö vasta
riemastui!

— Hyvä, Ivan, hyvä! — sanoi hän hypähtäen seisoalleen, taputti


käsiään ja juoksi pois.
Siitä lähtien hän toi joka päivä salaa Žilinille maitoa. Ja kun tatarit
tekivät vuohenmaidosta juustoja, joita kuivatettiin katolla, kiidätti hän
niitäkin varkain Žilinille. Kerran isäntä tappoi pässin, ja tyttö toi
hihassaan viipaleen lampaan paistia vangille heittäen sen ovesta ja
juosten heti tiehensä.

Sattui kerran kauhea ukonilma; vettä satoi tunnin ajan kuin


saavista kaataen. Purot pursuivat sameina. Mistä ennen pääsi
kahlaamalla, siinä oli nyt vettä kolmen arssinan syvyydeltä ja virta
käänteli kiviä. Kaikkialla virtasi puroja, ja vuorilla kävi veden pauhu.
Sadekuuron mentyä ohi virtasi ojia kylän joka haaralla. Žilin pyysi
isännältä veitsen, teki telan sivulaudat ja niiden väliin veisti rattaan.
Rattaan kummallekin puolelle hän kiinnitti nuken.

Tyttöset toivat hänelle tilkkuja, ja hän vaatetti niillä nuket; toisesta


teki tytön, toisen puki pojaksi. Sitten hän kiinnitti vehkeet telaan ja
asetti puroon. Ratas alkoi pyöriä ja nuket hypähdellä.

Koko kylä oli kohta siinä koolla: poikia, tyttöjä, vanhoja eukkoja ja
ikämiehiäkin. Kaikki solkkasivat vierasta kieltään.

— Ai urus, ai Ivan!

Abdulilla sattui olemaan rikkinäinen kello. Hän kutsui Žilinin


luokseen, näytteli hänelle kelloa ja melskasi jotakin.

Žilin sanoi heti:

— Annahan tänne, niin korjataan.

Hän otti kellon, väänteli veitsellä rattaat auki, pani ne jälleen


paikoilleen ja antoi kellon takaisin. Kello käydä naksutti.
Isäntä ihastui tästä; toi vanhan, rikkinäisen alusmekkonsa ja
lahjoitti sen Žilinille. Mitäs muuta kuin ota — kelpasihan se yöllä
peitteeksi.

Tästä lähtien Žilinin kätevyydestä kulki maine, että hän oli oikea
mestari. Alkoi tulla väkeä hänen luokseen kaukaisista kylistä; ken toi
pyssyn tai pistoolinlukon, ken kellorähjän korjattavaksi. Toipa isäntä
hänelle työkalujakin, hohtimet, poran ja viilan.

Kerran eräs tatari sairastui, ja heti tultiin Žiliniä pyytämään:


»Tulehan parantamaan!» Žilinillä ei ollut aavistustakaan miten oli
lääkittävä, mutta hän meni kuitenkin katsomaan potilasta tuumien:
»Tottapahan siitä omia aikojaan paranee.» Hän meni vajaan, otti
vettä, sekoitti siihen hiekkaa ja lausui puoliääneen jotakin tatarien
kuullen, antaen sitten juoman sairaalle. — Onneksi hänen
maineelleen potilas parani.

Žilin alkoi jo vähän ymmärtää heidän kieltäänkin. Toiset, jotka jo


olivat tutummiksi käyneet, kutsuivat häntä tuttavallisesti nimeltä:
»Ivan». Muut taas pälyilivät häneen kuin villipetoon.

Punapartainen tatari ei sietänyt Žiliniä. Heti kun hän vain sattui


tämän näkemään, synkistyivät miehen kasvot ja hän kääntyi
poispäin tai alkoi ärhennellä. Olipa täällä vielä eräs ukkokin, hän ei
tosin asunut itse kylässä, vaan tuli aina jostakin vuoren juurelta. Žilin
näki hänet vain silloin, kun hän kulki talon ohi moskeijaan Allahia
kumartamaan. Tämä oli lyhyt mies, lakin ympäri oli kierretty valkea
liina, suipoksi leikattu parta ja viikset olivat valkeat kuin untuva.
Kasvot olivat ryppyjä täynnä ja tiilenpunaiset. Nenä oli käyrä kuin
haukan nokka, harmaat silmät kiiluivat ilkeästi ja hampaattomasta
suusta törrötti vain kaksi torahammasta. Turbaani päässään hän
mennä köpitteli sauvaansa nojaten ja tähyili ympärilleen kuin susi.
Huomatessaan Žilinin hän murahti ja kääntyi pois.

Kerran Žilin lähti vuoren juurelle katselemaan ukon asuntoa. Hän


kulki polkua pitkin jonkin matkaa ja näki kiviaidan ympäröimän
puutarhapahasen; aidan takana kasvoi kirsikka- ja persikkapuita ja
vähän etempänä oli pieni, tasakattoinen hökkeli. Hän astui
lähemmäksi, näki oljista punottuja mehiläispönttöjä vierekkäin;
mehiläiset lentelivät suristen niiden ympärillä. Ukko oli polvillaan
maassa, hääräillen mehiläispesän ääressä. Žilin nousi siitä hiukan
ylemmäs katsomaan ja sattui kolistamaan jalkapuitaan. Ukko vilkaisi
taakseen, älähti, vetäisi vyöltä pistoolin ja laukaisi Žiliniä kohden.
Tämä ehti töintuskin painautua kiven taakse suojaan.

Ukko tuli kantelemaan isännälle. Tämä kutsui Žilinin luokseen ja


naurussa suin kysyi:

— Mitä sinulla oli tekemistä ukon luona?

— Kävinpähän vain katsomassa, miten ukko siellä elelee. En minä


hänelle mitään pahaa tehnyt.

Isäntä kertoi ukolle Žilinin vastauksen.

Mutta äijä oli vain kiukuissaan, murisi jotakin torahampaat törröllä


ja pui nyrkkiä Žilinille.

Kaikkea Žilin ei ymmärtänyt, sen hän kuitenkin tajusi, että ukko


käski isäntää tappamaan venäläiset eikä pitämään heitä talossaan.
Lopulta ukko lähti kotiinsa.

Žilin alkoi kysellä isännältä lähemmin ukosta, mikä hän oikein oli.
Isäntä selitti:
— Siinä vasta mies oli! Ei olekaan hänen veroistaan džigittiä näillä
mailla, monta venäläistä on ottanut hengiltä ja rikas oli aikoinaan.
Kolme eukkoa on ukolla ollut ja kahdeksan poikaa. Kylässä asuivat
kaikki yhdessä. Saapuivat sitten venäläiset, hävittivät kylän ja
tappoivat seitsemän poikaa. Yksi poika vain jäi henkiin ja antautui
venäläisille. Ukko matkusti perästä ja heittäytyi itsekin venäläisten
armoille. Eleli heidän parissaan kolmisen kuukautta, löysi sieltä
poikansa, surmasi hänet omin käsin ja sitten pakeni. Sen päivän
jälkeen ei ole enää sotaa käynyt, vaan Mekkaan matkusti Allahia
rukoilemaan. Siksi nyt turbaania kantaa. Mekassa käynyttä sanotaan
Hadziksi, ja sellaisilla on turbaani päässä. Ukko ei suosi teikäläisiä.
Käskee tappamaan sinut, mutta minun ei käy sitä tekeminen, kun
olen maksanut sinusta rahat. Pidän sinusta, Ivan, en hennoisi
tappaa, en edes luotani laskea, ellen olisi antanut sanaani. — Isäntä
nauraa virnisteli virkkaessaan venäjäksi: »Sinu, Ivan, hyvä, minu,
Abdul, hyvä.»
IV

Žilin eleli näin jälleen kuukauden päivät. Hän käveli päivisin ympäri
aulia (kylää) tai askarteli jotakin. Mutta kun yö saapui ja aulissa
kaikki äänet vaikenivat, hän vajassaan tonki maata kaivellen. Kivet
tekivät kaivaessa vastusta, mutta hän kuopi kivet irti viilan tyngällä ja
sai lopulta kaivetuksi seinän alle aukon, josta sopivana hetkenä voi
ryömiä ulos. »Kunpa vain tuntisin, seudut ja tietäisin, mihin päin
lähteä», hän mietiskeli. »Tataritpa eivät virka mitään.»

Žilin päätti käyttää tilaisuutta hyväkseen, kun isäntä lähti matkalle.


Hän meni iltapäivällä aulin taa vuorelle seutua tarkastelemaan. Mutta
isäntä oli, lähtiessään käskenyt poikansa vartioida Žiliniä eikä
päästää häntä näkyvistä. Poika juoksi nyt ihan Žilinin kintereillä
kirkuen:

— Älä mene! Isä kielsi laskemasta. Taikka minä huudan ihmisiä!

Žilin rupesi poikaa taivuttelemaan.

— Enhän minä kauas mene, virkkoi hän, — nousen vain tuonne


vuorelle; haen lääkeruohoja, niitä tarvitaan. Tule mukaan. Enhän
minä jalkapuissa pakoon pääse. Huomenna sitten teen sinulle
jousen ja nuolia.
Poika taipui ja he lähtivät. Vuori näytti olevan lähellä, mutta
jalkapuissa oli työlästä kompuroida. Kiipeäminen oli kova urakka.
Viimein oltiin vuorella, Žilin istahti ja alkoi tarkastella seutua.
Eteläpuolella, vajan takana notkossa, oli parvi hevosia, ja alempana
syvänteessä näkyi toinen tatarilaiskylä. Sen takana kohosi toinen
vuori, vielä jyrkempi, ja sen takana taas vuori. Vuorten välissä siinsi
metsä ja sitten oli taas vuoria, yhä korkeampia, ja kaikkein
korkeimmat olivat lumipeitteisiä, valkeita kuin sokeri. Yksi niistä
kohosi muita korkeammalle; sillä näytti olevan aivan kuin hattu
päässä. Katsoipa itään tai länteen, aina vain näkyi samanlaisia
vuorenhuippuja; siellä täällä pilkotti rotkoista tatarilaiskyliä, joista
nousi savu. »Tämä kaikki on nyt sitä heidän puoltaan», mietti Žilin.
Sitten hän alkoi silmäillä venäläiselle puolelle: alhaalla virtasi joki —
ja sen rannalla näkyi tatarilaiskylä puutarhoineen. Joella oli naisia
pyykkiä huuhtomassa, he näyttivät pieniltä kuin nuket. Kylän takana
alempana oli vuori ja sen takana vielä toisia metsän peittämiä vuoria.
Mutta kahden vuoren välitse siinsi etäällä tasanko, ja siellä hyvin
kaukana näkyi aivan kuin savu nousevan. Žilin koetti muistella ja
palauttaa mieleensä mistäpäin aurinko nousi ja mihin se laski silloin
kun hän oli linnoituksessa. Hän katseli yhä ja hänestä tuntui
varmalta, että heidän linnoituksensa oli juuri tässä laaksossa.
Tuonne, noiden kahden vuoren väliin oli siis koetettava paeta.

Aurinko alkoi jo mennä mailleen. Punertava hohde levisi vuorten


lumihuipuille, ja lumettomat vuoret tummenivat. Notkoista kohosi
sumua, mutta tasanko, jossa linnoituksen täytyi olla, leimusi
pelkkänä tulimerenä laskevan auringon valossa. Žilin terästi
katseensa, jotakin häämötti etäällä alangolta, sieltä nousi todellakin
savua rakennuksesta. Hän uskoi nyt varmasti, että sen täytyi tulla
juuri heidän linnoituksestaan.
Jo alkoi olla myöhä. Tatarilainen mulla huusi iltahuutonsa. Karjaa
ajettiin kotiin, — lehmät ammuivat. Poika kirkui yhtenään:
»Lähdetään jo», mutta Žilin ei olisi vielä tahtonut lähteä.

He palasivat kotiin. »Hyvä on», ajattelee Žilin, »nyt tunnen tien, ei


muuta kuin karkuun!» Jo samana yönä hän tahtoi karata. Yöt olivat
pimeät — oli uudenkuun aika. Pahaksi onneksi saapuivat tatarit
yöksi kotiin. Näin retkiltään palatessaan he tavallisesti ajavat laumaa
edellään ja pitävät iloa. Mutta nyt he eivät olleet saaneet mitään
käsiinsä, ainoastaan erään tatarin ruumis roikkui satulassa; —
vainaja oli punaparran veli. He olivat kokoontuneet kuljettamaan
vainajaa hautaan ja olivat äkeissään. Žilinkin tuli katsomaan. Kuollut
käärittiin palttinaan, kannettiin kylän taakse plataanipuiden alle ja
laskettiin nurmelle. Mullakin saapui siihen, ja kaikki sitoivat liinan
lakkiinsa ja kävivät istumaan kyykkyyn vierekkäin kuolleen ääreen.

Etumaisena oli mulla, hänen takanaan kolme turbaanipäistä ukkoa


ja näiden takana vielä muita tatareja. Kaikki istuivat äänettöminä,
päät painuneina rinnalle. He olivat kauan ääneti.

Vihdoin mulla kohotti katseensa ja lausui:

— Allah! — Hän sanoi vain tämän ainoan sanan, ja kaikki


painoivat jälleen päänsä ja olivat kauan vaiti istuen liikkumatta.

Mulla kohotti taas päänsä:

— Allah! — ja kaikki toistivat:— Allah! — vaieten jälleen. Kuollut


makasi nurmella liikkumattomana, ja toverit kyyköttivät ympärillä.
Kukaan ei hievahdakaan. Tuulenhenki ainoastaan hiukan leyhytteli
plataanin lehtiä. Sitten mulla luki rukouksen; kaikki nousivat ylös,
nostivat kuolleen ja kantoivat käsillään haudan reunalle. Hauta ei
ollut tavallinen, se oli kaivettu kellarin tapaisesti törmään. Miehet
tarttuivat kuollutta sääriin ja kainaloihin, laskivat hiljalleen hautaan
istuvaan asentoon ja asettivat kädet vatsalle.

Nogaijilainen toi sylyksen vihreitä kaisloja. Näillä täytettiin hauta,


se luotiin kiireesti umpeen, kumpu tasoitettiin ja pään puolelle
pystytettiin kivi. Sitten maa tallattiin, ja kaikki istuutuivat taas
vierekkäin haudan ääreen ja olivat kauan ääneti.

— Allah! Allah! Allah! — Saattajat huokasivat ja nousivat.

Punapartainen tatari jakeli ukoille rahaa, hypähti ylös ja


sivallettuaan kolme kertaa piiskalla otsaansa lähti kotiinsa.

Seuraavana aamuna Žilin näki punaparran taluttavan tammaa


kylän taakse ja kolmen muun tatarin kulkevan hänen perässään.
Tultuaan kylän taakse punaparta heitti takin yltään, kääri
paidanhihansa ylös — käsivarret olivat vankat — veti tikarinsa
vyöltään ja hioi sitä kovasimeen. — Tatarit tempoivat tamman päätä
taaksepäin, punaparta astui luokse ja viilsi eläimeltä kurkun poikki,
kaatoi tamman pitkälleen ja alkoi nylkeä elukkaa tunkemalla
nyrkkinsä nahan alle. Tuli akkoja ja tyttöjä suolia ja sisälmyksiä
puhdistamaan. Miehet paloittelivat sitten hevosen kappaleiksi ja
laahasivat tupaan. Ja koko kylän väki saapui punaparran taloon
vainajan muistoa viettämään.

Kolme päivää syötiin sitten tamman lihaa, juotiin tatarilaisolutta ja


vietettiin vainajan muistoa. Kaikki tatarit olivat kotosalla. Neljäntenä
päivänä Žilin näki päivällisen tienoissa miesten hommaavan lähtöä
jonnekin. Talutettiin hevosia pihalle, varustauduttiin. Kymmenkunta
miestä lähti matkalle, punaparta heidän mukanaan. Ainoastaan
Abdul jäi kotiin. Yöt olivat vielä pimeät, uusi kuu oli vastikään
syntynyt.

— Nyt, tuumi Žilin, on tullut karkaamisen hetki. — Hän puhui


asiasta
Kostylinillekin. Mutta tätä yritys peloitti.

— Mihin me tästä karkaamme? Kun emme tunne edes tietäkään.

— Minä kyllä tunnen tien.

— Mutta yössä ei ehditä perillekään.

— Ollaan metsässä yötä, jollei ehditä. On minulla jokunen


kakkarakin säästettynä. Sitäkö tässä rupeaa istumaan? Kuka tietää
lähettävätkö rahojakaan. Ja tatarit ovat nyt kiukuissaan, kun
venäläiset ovat tappaneet heidän miehensä.

Kostylin tuumaili kotvan aikaa.

— No, olkoon menneeksi.

Žilin ryömi aukon suulle ja kaivoi sitä suuremmaksi, jotta


Kostylinkin mahtuisi. He istuivat sitten odottamaan, kunnes kylässä
äänet vaikenivat.

Kun ei koko aulista enää kuulunut ihmisääntä, Žilin ryömi vajasta


seinän alitse. Hän kuiskasi Kostylinille: »Ryömi!» Kostylinkin pääsi
reijästä, mutta jalallaan tuli potkaisseeksi kiveä ja kolistelleeksi.
Isännällä oli talon vahtina kirjava koira, hyvin äkäinen, Uljašin
niminen. Tätä Žilin oli jo etukäteen vartavasten syötellyt. Koira kuuli
nyt kolinan, ärähti haukkumaan ja ryntäsi vajan luo perässään vielä
muitakin koiria. Žilin vihelsi hiljalleen ja heitti piirakan palan koiran
eteen. Uljašin tunsikin jo hänet, alkoi heiluttaa häntäänsä ja lakkasi
haukkumasta.

Isäntä kuuli tupaan haukunnan ja huusi sieltä koiralle: — Uskii!


Uskii!
Uljašin!

Mutta Žilin syyhytteli koiran niskaa. Koira oli ääneti, hankasi


itseään hänen polviansa vasten ja heilutteli häntäänsä.

Nurkan takana he sitten istuivat odotellen. Kaikki oli taas hiljaista;


kuului ainoastaan lampaan yninää karsinasta ja alhaalla kivien
lomitse virtaavan veden kuohu. Oli pimeä, tähdet kiiluivat korkealla
taivaalla; vuoren takaa alkoi uusi kuu punertavana kohota, sakarat
ylöspäin. Notkoissa leijaili maidonvalkeana sumu.

Žilin nousi ja virkkoi toverilleen: »No, veliseni, nyt!»

He lähtivät astumaan, mutta juuri kun he olivat päässeet liikkeelle,


kuulivat he katolta papin huutavan: »Allah! Besmillah! Il rahman!»
Väki siis meni moskeijaan. Miehet istuutuivat taas, painautuen
seinän viereen. He istuivat kauan ja odottivat, kunnes kaikki
moskeijaan menijät olivat kulkeneet ohi. Taas oli hiljaista.

— No, Herra meitä auttakoon! — He tekivät ristinmerkin ja


läksivät. He kulkivat pihan poikki jyrkkää rinnettä alas joelle,
kahlasivat sen poikki ja kulkivat notkoa pitkin eteenpäin. Alhaalla
laaksoissa leijaili sakea sumu, ylhäällä pään päällä tuikkivat tähdet.
Niiden mukaan Žilin ohjasi suunnan. Sumu oli viileää, ja oli kevyt
kävellä, saappaat vain olivat epämukavat, länttään astutut. Žilin heitti
ne jalastaan menemään ja kulki paljain jaloin. Hän hyppelehti kiveltä
kivelle ja katsahti tähtiin. Kostylinin oli vaikea seurata.
— Älä mene niin kovasti, pyysi hän. Kirotut ruojut kalvavat jalkani
verille.

— Heitä menemään, niin on kevyempi kulkea.

Kostylin käveli avojaloin, — mutta entistä huonommin kävi kulku.


Hän löi jalkansa kiviin, ja jäi yhä jälemmäs. Žilin virkkoi:

— Vähät naarmuista, ne paranevat; mutta jos saavat kiinni, niin


leikki on lopussa, tappavat.

Kostylin ei virkkanut mitään, kulki vain ähkien. He kulkivat kauas


notkoa pitkin. Yhtäkkiä kuuluu oikealta koirien haukuntaa. Žilin
pysähtyi, katseli ympärilleen, kiipesi vuorelle, kopeloiden käsillään.

— Ohhoh, harhaan olemme menneet, sanoi hän.

— Me olemme kulkeneet liiaksi oikealle. Tuolla on vieras kylä, näin


sen vuorelta; on mentävä takaisin vasemmalle, vuoren rinnettä.
Tuolla täytyy olla metsä.

Mutta Kostylin sanoi:

— Odotahan toki hiukan, levähdetään; jalkani ovat ihan verillä.

— Voi veikkoseni, kyllä ne paranevat; koeta hyppiä kevyemmin.


Tällä lailla.

Ja Žilin juoksi takaisin ja kääntyi vasemmalle vuoren kylkeä pitkin


metsään. Kostylin jäi jälkeen ja kulki ähkien. Žilin hoputti häntä
vähänväliä eikä pysähtynyt.

He nousivat jälleen vuorelle. Aivan oikein — siinä olikin metsän


reuna. He painautuivat metsään ja repivät okaihin siekaleiksi
viimeiset vaaterääsynsä. Lopulta he osuivat polulle ja lähtivät
kulkemaan sitä pitkin.

— Seis! — Tieltä kuului kavion kapsetta. He seisahtuivat


kuuntelemaan. Kuului kuin hevosen askelten töminää. Sitten oli
hiljaista. He lähtivät liikkeelle, ja töminä kuului uudelleen. He
pysähtyivät, ja sekin taas lakkasi kuulumasta. Žilin heittäytyi
pitkälleen maahan ja koetti katsella valoa vasten, — tiellä seisoi
jokin. Se näytti ihan hevoselta, ja oli kuin sen selässä olisi joku
istunut, mutta se ei näyttänyt oikein ihmiseltä. Kuului hevosen
korskumista. — Mitä ihmettä! — Žilin vihelsi hiljalleen, — se ryntäsi
metsään ja ryske kävi, kun se tuulispäänä katkoi mennessään oksia.

Kostylin aivan lyyhistyi maahan. Mutta Žilin virkkoi nauraen:

— Se oli hirvi. Kuuletko miten se sarvillaan katkoo oksia. Me


pelkäsimme sitä ja se meitä.

He kulkivat yhä eteenpäin. Linnut alkoivat puissa liikehtiä, aamuun


ei ollut enää pitkälti. Miehet eivät oikein tienneet, olivatko he oikealla
tiellä vai ei. Žilinistä tuntui kuin hänet olisi tuotu tätä samaa tietä
kylään ja että toverien luo oli noin kymmenkunta virstaa. Varma ei
hän tiestä kuitenkaan ollut, kun yö oli vielä pimeä. He tulivat vihdoin
aholle. Kostylin istuutui ja puhui:

— Tee niinkuin tahdot, mutta minä en jaksa enää astua, jalkani


eivät nouse.

Žilin alkoi häntä innostaa. — Ei tule mitään, vastasi Kostylin. — En


jaksa.

Žilin kiukustui, sylkäisi ja alkoi sättiä:


— Lähdenpähän sitten yksin. Hyvästi!

Kostylin hypähti pystyyn, alkaen kiiruhtaa jäljestä. He olivat


kulkeneet noin neljä virstaa. Sumu metsässä tiheni tihenemistään, ei
mitään voinut erottaa ympärillä, tuskin tähtiäkään enää näkyi.

Yht'äkkiä kuului taas hevosen kavioiden töminää. Kengät


kilahtelivat kiviä vasten. Žilin heittäytyi vatsalleen ja kuunteli korva
maassa.

— Tosiaankin, joku ratsastaa tänne meihin päin.

He juoksivat tieltä pensaikkoon odottamaan. Žilin ryömi tielle päin


katsoakseen, — tatari ajoi ratsain lehmää edellään, mutisten jotakin
nenäänsä itsekseen. Nyt hän oli ohi. Žilin palasi Kostylinin luo.

— No, ei hätää, — nouse, lähdetään!

Kostylin yritti nousta, mutta kaatui jälleen.

— En voi, — jumal'auta, en jaksa; voimani, ovat lopussa.

Pullea, tukeva mies oli helposti hiestynyt. Kylmässä sumussa jalat


kolhiutuneina verille hän nääntyi tyyten. Žilin koetti auttaa häntä
pystyyn, mutta Kostylinilta pääsi tuskanhuuto:

— Älä, koskee!

Žilinkin oli jähmettyä.

— Hittoako sinä huudat? Tatari tuossa lähellä vielä kuulee. —


Samalla hän ajatteli: »Onpa mies tosiaan uupunut; mitä oikein olisi
tehtävä? Ei käy ystävää jättäminenkään.» Hän virkkoi Kostylinille:
— No, nousehan, käy selkääni, — minä kannan, koska sinä et voi
kävellä.

Hän nosti Kostylinin selkäänsä, piteli reisistä kiinni ja lähti tietä


pitkin laahustamaan.

— Älä kurista kurkkuani. Herran tähden; pitele olkapäistä!

Žilinin oli raskasta kulkea, hänenkin jalkansa olivat verillä ja hän oli
ihan uupumaisillaan. Hän kumartui, kohotteli Kostylinin paremmin
selkäänsä ja kantaa retuutti häntä pitkin tietä.

Tatari näkyi kuulleen Kostylinin huudon. Žilin kuuli jonkun


ratsastavan perässään ja huutavan jotakin tatarin kielellä. Hän
heittäytyi pensaikkoon tien viereen. Tatari tempasi pyssynsä ja
laukaisi. Ei toki onneksi sattunut. Tatari vihelsi omalla tavallaan ja
ratsasti pois tietä pitkin.

— Nyt kävi hullusti, veikkoseni, virkkoi Žilin. — Se koira kerää heti


tatarit ajamaan meitä takaa. Ellemme ehdi vielä kolmisen virstaa,
niin olemme hukassa. Hän ajatteli itsekseen: »Piruko minua riivasi
ottamaan tuon pölkyn niskaani? Yksin olisin jo kauan sitten ollut
turvassa.»

Kostylin kehoitti häntä:

— Lähde yksin, mitä sinä jäät minun takiani kuolemaan?

— En, en lähde yksin, en jätä toveriani!

Žilin otti taas Kostylinin selkäänsä. Virstan verran he kulkivat näin


edelleen. Metsää jatkui yhä, ei laitaa näkynyt. Sumu alkoi jo hälvetä
leijaillen kevyinä hattaroina ilmassa, eikä tähtiäkään enää näkynyt.
Žilin alkoi tuskastua.

He joutuivat tien vieressä olevalle lähteelle, joka oli kivillä


reunustettu. He pysähtyivät sen ääreen ja Žilin laski Kostylinin
maahan.

— Annahan, kun hiukan juon ja huoahdan. Maistetaan kakkaraa.


Varmaankin ollaan jo lähellä.

Kun hän kumartui juomaan, kuulivat he jälleen takanansa


kavioiden kapsetta. He painautuivat tieltä oikealle pensaikkoon ja
heittäytyivät jyrkänteen alle pitkälleen.

Kuului tatarien puhetta. Nämä olivat pysähtyneet samaan kohtaan,


mistä Žilin ja Kostylin olivat poikenneet tieltä. He keskustelivat
hetkisen ja sitten usuttivat koiria. Toverukset kuuntelivat, pensaissa
rapisi ja samassa juoksi vieras koira suoraapäätä heitä kohti. Se
pysähtyi äkkiä ja alkoi haukkua.

Tataritkin tunkeutuivat pensaikkoon — vieraita näkyivät olevan.


Pakolaiset otettiin kiinni, pantiin köysiin, kiskottiin satulaan ja
toimitettiin matkaan.

Kun he olivat ratsastaneet kolmisen virstaa, tuli Abdul isäntä


vastaan kahden muun tatarin seuraamana. Tulijat puhuivat jonkin
sanan, siirsivät karkurit omien hevosten selkään ja lähtivät viemään
takaisin kylään.

Abdulin suu ei vetäytynyt nauruun eikä hän virkkanut heille


sanaakaan koko matkalla.
Päivän koittaessa tultiin perille auliin, ja karkurit jätettiin pihalle
istumaan. Heidän ympärilleen kokoontui heti lapsilauma, joka
kirkuen heitti heitä kivillä ja huitoi raipoilla.

Tatarit kokoontuivat piiriin pihalle; heidän joukossaan oli


vuorenjuurella asuva ukkokin. He alkoivat puhella. Žilin kuuli, että
puhe koski heitä ja että pohdittiin mitä vangeille oli tehtävä. Toiset
neuvoivat lähettämään heidät kauemmaksi vuorille, ukko taas
kehoitti tappamaan. Mutta Abdul ei suostunut:— Minä olen maksanut
heistä rahat ja tahdon saada lunnaat! Ukko vain vakuutteli: — Eivät
maksa kopeekkaakaan, tuhoa vain tekevät. Synti on ruokkia
venäläistä. Parasta on tappaa — sillä pääset!

He hajaantuivat. Isäntä astui Žilinin luo ja alkoi puhella hänelle:

— Ellei teidän lunnaitanne parin viikon päästä lähetetä, pehmitän


nahkanne piiskalla. Mutta jos vielä kerran karkaamaan rupeat, tapan
sinut kuin koiran. Kirjoita kirje ja tee se kunnollisesti.

Vangeille tuotiin paperia, ja he kirjoittivat. Heidät kytkettiin


jalkapuuhun ja laahattiin moskeijan taakse, missä heidät heitettiin
viiden arssinan syvyiseen kuoppaan.
VI

Heidän elämänsä oli nyt niin tukalaa kuin suinkin. Jalkapuista ei


päästetty eikä heitä laskettu ollenkaan raittiiseen ilmaan
jaloittelemaan. Heille heitettiin kuoppaan kuin koirille melkein
paistamatonta taikinaa ja tuotiin ruukussa vettä. Kuopassa oli
raskas, kostea, löyhkäävä ilma; Kostylin tuli ihan sairaaksi, rupesi
pöhöttymään koko mies, ja ruumista alkoi kovasti särkeä; hän milloin
voihki, milloin taas yritti nukkua. Žilininkin mieli oli ihan lamassa;
asiat olivat nyt huonolla kannalla. Eikä hän keksinyt keinoa, millä
olisi täältä selvin nyt. Hän ryhtyi kuopimaan reikää, mutta ei tiennyt
minne mullan heittäisi. Isäntä huomasi sen ja uhkasi tappaa.

Kyyröttäessään kerran kuopassaan hän muisteli taas entistä


vapaanaoloaan ja ikävöi. Samassa putosi piirakas suoraan hänen
polvilleen, pian senjälkeen tipahti toinen ja kirsikoita vielä lisäksi.
Hän katsahti ylös ja huomasi siellä Dinan. Tämä katseli häntä
hetkisen, nauroi ja juoksi sitten tiehensä. Žilinin päähän juolahti heti:

— Jospa tuo Dina auttaisi!

Hän puhdisti kuopan nurkan, kaivoi esiin savea ja alkoi muovailla


nukkeja.
Hän valmisti ihmisiä, hevosia, koiria ja ajatteli itsekseen: — Kun
Dina taas tulee, heitän nämä hänelle.

Dinaa ei kuulunut koko seuraavana päivänä. Sen sijaan Žilin kuuli


kavioiden kapsetta; ratsastajat ajoivat ohitse, ja tatarit kokoontuivat
moskeijan luokse. Siinä he kiistelivät, huusivat ja puhuivat jotakin
venäläisistä. Žilin erotti ukon äänen. Hän ei saanut sanoista tarkoin
selvää, mutta sen verran hän arvasi, että venäläisiä oli tullut
lähitienoille ja että tatarit pelkäsivät heidän saapuvan kylään ja olivat
nyt ymmällä mihin panisivat vangit ja mitä näille oli tehtävä.

He keskustelivat Žilin kuuli kopinaa ja näki Dinan hetkisen ja


poistuivat. Äkkiä ylhäältä. Hän katsahti ylös kyykistyvän kuopan
reunalle; tytön polvet olivat korvia korkeammalla. Tyttö kurottautui,
kaulahelmet riippuivat, heiluivat kuopan yläpuolella, ja silmät loistivat
kuin kaksi tähteä. Hän vetäisi hihastaan kaksi juustopiirakkaa ja heitti
Žilinille. Žilin otti ne ja virkkoi:

— Miksi et ole pitkään aikaan käynytkään? Minä olen tehnyt


sinulle leikkikalujakin. Ota! — Hän alkoi viskellä niitä yksitellen
tytölle.

Mutta tyttö pudisti vain päätään eikä katsonutkaan niitä.

— En minä tahdo! — virkkoi hän. Hän oli ääneti ja istui.

Sittenhän jatkoi:— Ivan, he tahtovat sinut tappaa! — Hän osoitti


kädellään kaulaansa.

— Kuka tahtoo tappaa?

— Isä. Ukot yllyttävät. Mutta minun on sinua sääli.

You might also like