An Introduction to Proof Theory Normalization Cut

Elimination and Consistency Proofs 1st Edition Paolo

Mancosu latest pdf 2025

https://ebookgate.com/product/an-introduction-to-proof-theory-
normalization-cut-elimination-and-consistency-proofs-1st-edition-
paolo-mancosu/

★★★★★
4.8 out of 5.0 (62 reviews )

Immediate PDF Access

ebookgate.com
 An Introduction to Proof Theory Normalization Cut
Elimination and Consistency Proofs 1st Edition Paolo Mancosu

EBOOK

Available Formats

■ PDF eBook Study Guide Ebook

EXCLUSIVE 2025 ACADEMIC EDITION – LIMITED RELEASE

Available Instantly Access Library

Instant digital products (PDF, ePub, MOBI) available
Download now and explore formats that suit you...

The Nuts and Bolts of Proofs Third Edition An Introduction

to Mathematical Proofs Antonella Cupillari

https://ebookgate.com/product/the-nuts-and-bolts-of-proofs-third-
edition-an-introduction-to-mathematical-proofs-antonella-cupillari/

ebookgate.com

Journey into mathematics an introduction to proofs Joseph

J Rotman

https://ebookgate.com/product/journey-into-mathematics-an-
introduction-to-proofs-joseph-j-rotman/

ebookgate.com

Proofs That Really Count The Art of Combinatorial Proof

1st Edition Arthur T. Benjamin

https://ebookgate.com/product/proofs-that-really-count-the-art-of-
combinatorial-proof-1st-edition-arthur-t-benjamin/

ebookgate.com

Beginning Theory An Introduction to Literary and Cultural

Theory 2nd Edition Peter Barry

https://ebookgate.com/product/beginning-theory-an-introduction-to-
literary-and-cultural-theory-2nd-edition-peter-barry/

ebookgate.com
Beginning Theory An Introduction to Literary and Cultural
Theory 3rd Edition Peter Barry

https://ebookgate.com/product/beginning-theory-an-introduction-to-
literary-and-cultural-theory-3rd-edition-peter-barry/

ebookgate.com

Introduction to Distribution Logistics Statistics in

Practice 1st Edition Paolo Brandimarte

https://ebookgate.com/product/introduction-to-distribution-logistics-
statistics-in-practice-1st-edition-paolo-brandimarte/

ebookgate.com

An Introduction to Australian Public Policy Theory and

Practice 1st Edition Sarah Maddison

https://ebookgate.com/product/an-introduction-to-australian-public-
policy-theory-and-practice-1st-edition-sarah-maddison/

ebookgate.com

Number Shape Symmetry An Introduction to Number Theory

Geometry and Group Theory 1st Edition Diane L. Herrmann

https://ebookgate.com/product/number-shape-symmetry-an-introduction-
to-number-theory-geometry-and-group-theory-1st-edition-diane-l-
herrmann/
ebookgate.com

An Introduction to the Representation Theory of Groups 1st

Edition Emmanuel Kowalski

https://ebookgate.com/product/an-introduction-to-the-representation-
theory-of-groups-1st-edition-emmanuel-kowalski/

ebookgate.com
An Introduction to Proof Theory
An Introduction to Proof Theory

Normalization, Cut-Elimination, and Consistency

Proofs

Paolo Mancosu
Sergio Galvan
Richard Zach

1
 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

c Paolo Mancosu, Sergio Galvan and Richard Zach 2021
The moral rights of the authors have been asserted
First Edition published in 2021
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: 2021910782
ISBN 978–0–19–289593–6 (hbk.)
ISBN 978–0–19–289594–3 (pbk.)
DOI: 10.1093/oso/9780192895936.001.0001
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.
 Contents

Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii
About this book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . viii
For further reading . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . x
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xii
1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.1 Hilbert’s consistency program . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Gentzen’s proof theory . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.3 Proof theory after Gentzen . . . . . . . . . . . . . . . . . . . . . . . . 10
2 Axiomatic calculi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.1 Propositional logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . 13
2.2 Reading formulas as trees . . . . . . . . . . . . . . . . . . . . . . . . 15
2.3 Sub-formulas and main connectives . . . . . . . . . . . . . . . . . . 17
2.4 Logical calculi . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18
2.5 Inference rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
2.6 Derivations from assumptions and provability . . . . . . . . . . . . 24
2.7 Proofs by induction . . . . . . . . . . . . . . . . . . . . . . . . . . . . 26
2.8 The deduction theorem . . . . . . . . . . . . . . . . . . . . . . . . . . 30
2.9 Derivations as trees . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33
2.10 Negation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
2.11 Independence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40
2.12 An alternative axiomatization of J0 . . . . . . . . . . . . . . . . . . . 43
2.13 Predicate logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
2.14 The deduction theorem for the predicate calculus . . . . . . . . . . 50
2.15 Intuitionistic and classical arithmetic . . . . . . . . . . . . . . . . . . 53
3 Natural deduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 65
3.2 Rules and deductions . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
3.3 Natural deduction for classical logic . . . . . . . . . . . . . . . . . . 85
3.4 Alternative systems for classical logic . . . . . . . . . . . . . . . . . 88
3.5 Measuring deductions . . . . . . . . . . . . . . . . . . . . . . . . . . 89
3.6 Manipulating deductions, proofs about deductions . . . . . . . . . 91
3.7 Equivalence of natural and axiomatic deduction . . . . . . . . . . . 96
4 Normal deductions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
4.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101
4.2 Double induction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105
4.3 Normalization for ∧, ⊃, ¬, ∀ . . . . . . . . . . . . . . . . . . . . . . . 110
4.4 The sub-formula property . . . . . . . . . . . . . . . . . . . . . . . . 121
4.5 The size of normal deductions . . . . . . . . . . . . . . . . . . . . . . 130

v
 contents vi

4.6 Normalization for NJ . . . . . . . . . . . . . . . . . . . . . . . . . . . 132

4.7 An example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143
4.8 The sub-formula property for NJ . . . . . . . . . . . . . . . . . . . . 145
4.9 Normalization for NK . . . . . . . . . . . . . . . . . . . . . . . . . . 154
5 The sequent calculus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168
5.1 The language of the sequent calculus . . . . . . . . . . . . . . . . . . 168
5.2 Rules of LK . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 170
5.3 Constructing proofs in LK . . . . . . . . . . . . . . . . . . . . . . . . 175
5.4 The significance of cut . . . . . . . . . . . . . . . . . . . . . . . . . . 178
5.5 Examples of proofs . . . . . . . . . . . . . . . . . . . . . . . . . . . . 180
5.6 Atomic logical axioms . . . . . . . . . . . . . . . . . . . . . . . . . . 187
5.7 Lemma on variable replacement . . . . . . . . . . . . . . . . . . . . 188
5.8 Translating NJ to LJ . . . . . . . . . . . . . . . . . . . . . . . . . . . . 191
5.9 Translating LJ to NJ . . . . . . . . . . . . . . . . . . . . . . . . . . . . 197
6 The cut-elimination theorem . . . . . . . . . . . . . . . . . . . . . . . . . 202
6.1 Preliminary definitions . . . . . . . . . . . . . . . . . . . . . . . . . . 204
6.2 Outline of the lemma . . . . . . . . . . . . . . . . . . . . . . . . . . . 206
6.3 Removing mixes directly . . . . . . . . . . . . . . . . . . . . . . . . . 209
6.4 Reducing the degree of mix . . . . . . . . . . . . . . . . . . . . . . . 212
6.5 Reducing the rank . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217
6.6 Reducing the rank: example . . . . . . . . . . . . . . . . . . . . . . . 238
6.7 Reducing the degree: example . . . . . . . . . . . . . . . . . . . . . 242
6.8 Intuitionistic sequent calculus LJ . . . . . . . . . . . . . . . . . . . . 250
6.9 Why mix? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 252
6.10 Consequences of the Hauptsatz . . . . . . . . . . . . . . . . . . . . . 254
6.11 The mid-sequent theorem . . . . . . . . . . . . . . . . . . . . . . . . 257
7 The consistency of arithmetic . . . . . . . . . . . . . . . . . . . . . . . . . 269
7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 269
7.2 Consistency of simple proofs . . . . . . . . . . . . . . . . . . . . . . 276
7.3 Preliminary details . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
7.4 Overview of the consistency proof . . . . . . . . . . . . . . . . . . . 287
7.5 Replacing inductions . . . . . . . . . . . . . . . . . . . . . . . . . . . 288
7.6 Reducing suitable cuts . . . . . . . . . . . . . . . . . . . . . . . . . . 293
7.7 A first example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
7.8 Elimination of weakenings . . . . . . . . . . . . . . . . . . . . . . . . 298
7.9 Existence of suitable cuts . . . . . . . . . . . . . . . . . . . . . . . . . 301
7.10 A simple example . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
7.11 Summary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308
8 Ordinal notations and induction . . . . . . . . . . . . . . . . . . . . . . . 312
8.1 Orders, well-orders, and induction . . . . . . . . . . . . . . . . . . . 312
8.2 Lexicographical orderings . . . . . . . . . . . . . . . . . . . . . . . . 315
8.3 Ordinal notations up to 𝜀0 . . . . . . . . . . . . . . . . . . . . . . . . 320
 contents vii

8.4 Operations on ordinal notations . . . . . . . . . . . . . . . . . . . . . 325

8.5 Ordinal notations are well-ordered . . . . . . . . . . . . . . . . . . . 330
8.6 Set-theoretic definitions of the ordinals . . . . . . . . . . . . . . . . . 332
8.7 Constructing 𝜀0 from below . . . . . . . . . . . . . . . . . . . . . . . 336
8.8 Ordinal arithmetic . . . . . . . . . . . . . . . . . . . . . . . . . . . . 337
8.9 Trees and Goodstein sequences . . . . . . . . . . . . . . . . . . . . . 339
9 The consistency of arithmetic, continued . . . . . . . . . . . . . . . . . . 346
9.1 Assigning ordinal notations < 𝜀0 to proofs . . . . . . . . . . . . . . . 346
9.2 Eliminating inductions from the end-part . . . . . . . . . . . . . . . 354
9.3 Removing weakenings . . . . . . . . . . . . . . . . . . . . . . . . . . 360
9.4 Reduction of suitable cuts . . . . . . . . . . . . . . . . . . . . . . . . 368
9.5 A simple example, revisited . . . . . . . . . . . . . . . . . . . . . . . 376
A The Greek alphabet . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 380
B Set-theoretic notation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 381
C Axioms, rules, and theorems of axiomatic calculi . . . . . . . . . . . . . 383
C.1 Axioms and rules of inference . . . . . . . . . . . . . . . . . . . . . . 383
C.2 Theorems and derived rules . . . . . . . . . . . . . . . . . . . . . . . 384
D Exercises on axiomatic derivations . . . . . . . . . . . . . . . . . . . . . . 386
D.1 Hints for Problem 2.7 . . . . . . . . . . . . . . . . . . . . . . . . . . . 386
D.2 Hints for Problem 2.18 . . . . . . . . . . . . . . . . . . . . . . . . . . 390
D.3 Exercises with quantifiers . . . . . . . . . . . . . . . . . . . . . . . . 393
E Natural deduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394
E.1 Inference rules . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 394
E.2 Conversions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 395
F Sequent calculus . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399
G Outline of the cut-elimination theorem . . . . . . . . . . . . . . . . . . . 401
Bibliography . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 405
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 412
 Preface

About this book

This book arose out of the desire to enable students, especially those in philosophy,
who have only a minimal background in mathematics and logic, to appreciate
the most important results of classical proof theory. Most proof theory textbooks
begin at a rather demanding level. The more accessible among them only cover
structural proof theory, and those that also cover ordinal proof theory assume a
strong mathematical background. Unfortunately, there is no good, elementary
introduction to proof theory in English that does both. Our book is meant to fill
this gap.
We assume that the reader is familiar with both the propositional and predicate
calculus, as is covered in most introductory courses on formal logic taught in
philosophy departments. We do not assume familiarity with any specific deductive
system. Indeed, the first few chapters are for the most part introductions to three of
them: axiomatic derivations, natural deduction, and sequent calculus. We also do
not assume familiarity with the metatheory of logic, not even the most elementary
principle of reasoning about our calculi, namely the principle of induction on the
natural numbers (N). We try to ease the reader into the mechanics of inductive
proofs, first presenting simple induction on N, then double induction, and finally
induction along more complicated well-orderings such as 𝜀0 .
For those who desire to plunge immediately into structural proof theory,
Chapter 2 on axiomatic derivations can be skipped. Keep in mind, however, that
the principle of induction on N is explained there (in Section 2.7). Moreover, the
chapter introduces the crucial distinctions between minimal, intuitionistic, and
classical systems of logic. As is usual in proof theory, our syntax assumes that
free and bound variables are distinct syntactic categories. An explanation of this
and the difference to the usual approach that uses only a single kind of variable
(but distinguishes between free and bound occurrences) can also be found there (in
Section 2.13).
One of the main goals we set for ourselves is that of providing an introduction
to proof theory which might serve as a companion to reading the original articles
by Gerhard Gentzen. For this reason, while not slavishly following Gentzen, we
never deviate much from his choice of systems and style of proof. This, as we shall

viii
Visit https://ebookgate.com today to explore
a vast collection of ebooks across various
genres, available in popular formats like
PDF, EPUB, and MOBI, fully compatible with
all devices. Enjoy a seamless reading
experience and effortlessly download high-
quality materials in just a few simple steps.
Plus, don’t miss out on exciting offers that
let you access a wealth of knowledge at the
best prices!
 preface ix

see, already determines the choice of systems for our presentation of axiomatic
calculi.
In structural proof theory, we cover, among other things, the Gödel-Gentzen
translation of classical into intuitionistic logic (and arithmetic), natural deduction
and the normalization theorems (for both NJ and NK), the sequent calculus,
including cut-elimination and mid-sequent theorems, and various applications of
these results. The second half of the book covers ordinal proof theory, specifically,
Gentzen’s consistency proof for first-order Peano arithmetic using ordinal induction
up to 𝜀0 . The theory of ordinal notations and other elements of ordinal theory are
developed from scratch; no knowledge of set theory is presumed.
In order to make the content accessible to readers without much mathematical
background, we carry out the details of proofs in much more detail than is
usually done. For instance, although we follow Prawitz (1965) in our proof of the
sub-formula property for natural deduction (for the system NJ), we verify all of
the details left out of Prawitz’s treatment. In the proof of Gentzen’s consistency
result, likewise, we prove all the required properties about ordinal notations, carry
through all of the required lemmas, and verify that ordinal notations decrease
in the reduction procedure in detail. We also work through many examples to
illustrate definitions and to show how the proofs work in specific examples.
We’ve also diverged from the usual way of presenting results so that we can
avoid assuming some background. For instance, ordinal notations < 𝜀0 are defined
purely combinatorially, without assuming familiarity with ordinals (although
we will also discuss set-theoretic definitions of ordinals and the relation to the
combinatorial definition of ordinal notations). This has a philosophical payoff:
from the start we emphasize that ordinal notations are not transfinite ordinals, and
that the fact that they are well-ordered follows from elementary combinatorial
principles about orderings of sequences. In other words, our presentation of
ordinal notations is almost finitary.
Our proof of consistency of PA also has a significant feature not shared with
usual presentations or Gentzen’s original. Rather than show that a putative proof
of the empty sequent can be transformed into one without induction and cuts on
complex formulas, we show this for proofs of any sequent consisting of only atomic
formulas. The philosophical payoff is that the consistency proof is formulated
from the start as a conservativity result: induction and complex cuts can be
eliminated from proofs of elementary arithmetical facts in PA. Since the reduction
procedure now applies to actually existing proofs (and not just to putative, and
in fact non-existent, proofs of contradictions) we can actually present examples
of how the procedure works. So this way of doing things also has a pedagogical
payoff.
The book will be of interest to philosophers, logicians, mathematicians, com-
puter scientists, and linguists. Through this material, philosophy students will
acquire the tools required for tackling further topics in the philosophy of mathemat-
ics (such as the prospects for Hilbert’s program and its relativized versions), and in
 preface x

the philosophy of logic and language (the meaning of the logical constants; proof-
theoretic semantics; realism/anti-realism, Dummett’s program, proof-theoretic
harmony etc.).

For further reading

1. Gentzen’s articles are translated in:

Gentzen, Gerhard (1969), The Collected Papers of Gerhard Gentzen, ed. by Manfred
E. Szabo, Amsterdam: North-Holland.

2. For a detailed overview of the history of mathematical logic from Russell to

Gentzen, see:

Mancosu, Paolo, Richard Zach, and Calixto Badesa (2009), “The development
of mathematical logic from Russell to Tarski: 1900–1935,” in The Development
of Modern Logic, ed. by Leila Haaparanta, New York and Oxford: Oxford
University Press, pp. 324–478, doi: 10.1093/acprof:oso/9780195137316.003.
0029.

3. For the philosophical background of intuitionism and Hilbert’s program:

Iemhoff, Rosalie (2020), “Intuitionism in the philosophy of mathematics,” in

The Stanford Encyclopedia of Philosophy, ed. by Edward N. Zalta, Fall 2020,
https://plato.stanford.edu/entries/intuitionism/.
Mancosu, Paolo, ed. (1998), From Brouwer to Hilbert: The Debate on the Foundations
of Mathematics in the 1920s, New York and Oxford: Oxford University Press.
Zach, Richard (2019), “Hilbert’s program,” in Stanford Encyclopedia of Philosophy,
ed. by Edward N. Zalta, Fall 2019, https://plato.stanford.edu/archives/
fall2019/entries/hilbert-program/.

4. For an overview of the history of proof theory up to Gentzen:

Hendricks, Vincent F., Stig Andur Pedersen, and Klaus Frovin Jørgensen, eds.
(2000), Proof Theory: History and Philosophical Significance, Dordrecht: Springer,
doi: 10.1007/978-94-017-2796-9.
Von Plato, Jan (2018), “The development of proof theory,” in The Stanford
Encyclopedia of Philosophy, ed. by Edward N. Zalta, https://plato.stanford.
edu/archives/win2018/entries/proof-theory-development/.

5. On Gentzen’s life and career:

Menzler-Trott, Eckart (2016), Logic’s Lost Genius: The Life of Gerhard Gentzen,
History of Mathematics, 33, American Mathematical Society.
 preface xi

6. For a more technical overview of Gentzen’s contributions to logic:

Von Plato, Jan (2009), “Gentzen’s logic,” in Logic from Russell to Church, ed. by
Dov M. Gabbay and John Woods, Handbook of the History of Logic, 5,
Amsterdam: North-Holland, pp. 667–721, doi: 10.1016/S1874-5857(09)70017-
2.

7. Some introductory surveys on proof theory:

Prawitz, Dag (1971), “Ideas and results in proof theory,” in Proceedings of the
Second Scandinavian Logic Symposium, ed. by Jens E. Fenstad, Studies in
Logic and the Foundations of Mathematics, 63, Amsterdam: North-Holland,
pp. 235–307, doi: 10.1016/S0049-237X(08)70849-8.
Rathjen, Michael and Wilfried Sieg (2020), “Proof theory,” in The Stanford
Encyclopedia of Philosophy, ed. by Edward N. Zalta, Fall 2020, https://plato.
stanford.edu/archives/fall2020/entries/proof-theory/.

8. More advanced textbooks in proof theory:

Bimbó, Katalin (2014), Proof Theory: Sequent Calculi and Related Formalisms, Boca
Raton: CRC Press.
Buss, Samuel R., ed. (1998), Handbook of Proof Theory, Amsterdam: Elsevier.
Diller, Justus (2019), Functional Interpretations: From the Dialectica Interpretation to
Functional Interpretations of Analysis and Set Theory, World Scientific.
Girard, Jean-Yves (1987), Proof Theory and Logical Complexity, Studies in Proof
Theory, 1, Naples: Bibliopolis.
Negri, Sara and Jan von Plato (2001), Structural Proof Theory, Cambridge: Cam-
bridge University Press.
Pohlers, Wolfram (2009), Proof Theory: The First Step into Impredicativity, Berlin,
Heidelberg: Springer, doi: 10.1007/978-3-540-69319-2.
Prawitz, Dag (1965), Natural Deduction: A Proof-Theoretical Study, Stockholm
Studies in Philosophy, 3, Stockholm: Almqvist & Wiksell.
Schütte, Kurt (1977), Proof Theory, Berlin and New York: Springer.
Takeuti, Gaisi (1987), Proof Theory, 2nd ed., Studies in Logic, 81, Amsterdam:
North-Holland.
Troelstra, Anne Sjerp and Helmut Schwichtenberg (2000), Basic Proof Theory,
2nd ed., Cambridge: Cambridge University Press.
 preface xii

Acknowledgments
This book arose out of courses and seminars the authors have held over the
years at Berkeley, Calgary, and Milan. We would like to thank our students for
their valuable comments. We are especially grateful to James Walsh and Logan
Heath who have commented on the entire text, often catching important mistakes,
and for checking all the exercises. We would like to thank Luca Bellotti, Enrico
Moriconi, Dario Palladino, Francesca Poggiolesi, and David Schrittesser for their
comments on different sections of the book, and Yuuki Andou, Dag Prawitz, and
Peter Schröder-Heister for answering our questions. Special thanks goes to Jeremy
Avigad, who as reviewer for Oxford University Press provided thorough and
constructive comments.
The order of the co-authors on the cover of the book only reflects the order in
which they joined the project.
 1
Introduction

1.1 Hilbert’s consistency program

While reflection on mathematical proof goes as far back as the time of the ancient
Greeks, it reached the late nineteenth century with the detailed formalization
of mathematical proofs given first by Frege and Peano and later by Russell
and Whitehead, and others. A mathematical analysis of proofs considered as
mathematical objects only truly began with David Hilbert and his school.
David Hilbert’s work in logic originated from his interest in foundational
questions and in particular in the foundations of geometry (Hilbert, 1899). The
work on the foundations of geometry led Hilbert to a conception of mathematical
theories as formal systems that could be given varying interpretations, by choosing
the domain of objects at will and assigning to the basic predicates arbitrary
meanings constrained only by the requirement that the axioms of the theory be
satisfied in the given interpretation. Thus, a basic predicate such as 𝐿(𝑥) that in the
standard interpretation could be read as “ 𝑥 is a line”, and thus having its intended
meaning as the set of lines in the Euclidean plane, could be re-interpreted as a set
of pairs of real numbers in an arithmetical interpretation of the theory. The role of
predicates in this context is formal in the sense that they are independent of any
specific interpretation, i.e., the axioms allow for different interpretations. However,
at this early stage Hilbert did not work with a formalized logic. Rather, logic is
assumed informally in the background of his axiomatic studies. In particular, the
logical constants (connectives, quantifiers, etc.) are not treated formally, i.e., as
mere symbols devoid of an interpretation, but rather have the standard informal
meaning. Only when he became acquainted with Whitehead and Russell’s Principia
Mathematica (Whitehead and Russell, 1910–1913) did Hilbert use formalized logic.
The influence of Principia Mathematica and Hilbert’s thorough engagement with
logic is very much in evidence in the set of lecture notes for the winter semester
1917/1918 titled “The principles of mathematics” (Hilbert, 1918). This set of
lectures forms the basis of the later textbook Principles of Theoretical Logic (Hilbert
and Ackermann, 1928). They contain a wealth of material on logic including
An Introduction to Proof Theory: Normalization, Cut-Elimination, and Consistency Proofs
.
Paolo Mancosu, Sergio Galvan, and Richard Zach, Oxford University Press. © Paolo Mancosu,
Sergio Galvan and Richard Zach 2021. DOI: 10.1093/oso/9780192895936.003.0001
 introduction 2

propositional, predicate and higher-order logic (namely, the type theory of Russell
and Whitehead). Along with the beginning of what we call metatheory (studies
of consistency and completeness for propositional logic, independence of the
axioms etc.), the lectures are important because Hilbert was now able to formalize
mathematical theories completely, i.e., not only in their non-logical parts (say,
axioms for arithmetic, geometry or set theory) but in their logical parts as well.
Hilbert’s assistant Paul Bernays further developed these investigations in his
1918 dissertation (Bernays, 1918), where a clear distinction between syntax and
semantics is formulated. Here, we find all the key elements of our contemporary
understanding of propositional logic: a formal system, a semantic interpretation
in terms of truth-values and a proof of soundness and completeness relative to
that semantics. In addition, Bernays also showed that propositional logic was
decidable by using normal forms.
The decidability of mathematics (or fragments thereof) was an important
question for Hilbert. The roots of Hilbert’s engagement with decidability can be
traced as far back as 1900, although at the time he had no idea of what technical
tools would have to be enlisted to address such a task. In his 1900 address to the
International Congress of Mathematicians held in Paris, Hilbert claimed that every
mathematical problem has a solution:

This conviction of the solvability of every mathematical problem is a powerful incentive to

the worker. We hear within us the perpetual call: There is the problem. Seek its solution.
You can find it by pure reason, for in mathematics there is no ignorabimus. (Hilbert, 1900,
p. 1102)1

In “Axiomatic Thought” (Hilbert, 1917, p. 1113), the problem of “the decidability

of a mathematical question in a finite number of operations” is listed as one of
the fundamental problems for the axiomatic method. By now, the technical tools
for addressing such issues were beginning to be developed, as witnessed by the
1917/18 lecture notes and other contemporary work by Löwenheim, Behmann,
and others.2
Another problem that Hilbert had raised at the Paris Congress was that of
the consistency of mathematics. The consistency of various systems of geometry
(Euclidean and non-Euclidean) could be shown by reduction to arithmetical
theories (broadly construed as to include the real numbers and some set theory)
but the problem was pressing for the arithmetical theories themselves. This points
to the second element, in addition to Hilbert’s discovery of Principia Mathematica,
that led to his engagement with fully formalized theories and a renewed interest in
foundations. That element is the foundational debate that was forcefully brought
to a crisis point by the radical proposals of Hermann Weyl and L. E. J. Brouwer.
Weyl’s and Brouwer’s proposals can be seen as a reaction to the way classical

1 Page references are to the translation(s) listed in the bibliography.

2 On the origin and development of the decision problem see Mancosu and Zach (2015).
 introduction 3

mathematics was developing, in particular to the non-constructive and infinitary

turn it had taken since the late nineteenth century.
In Das Kontinuum (1918), Weyl criticized set theory and classical analysis as “a
house built on sand” and proposed a remedy to the uncertainty that in his opinion
affected a large part of classical mathematics. In the first part of the book, Weyl
proposed a number of criticisms to contemporary analysis and set theory, but also
to some of the accepted reconstructions of classical mathematics, such as Zermelo’s
set theory and Russell’s theory of types. In the positive part, Weyl developed his
own arithmetical continuum, a number system in which large portions of analysis
can be carried out. This foundational position is called “predicativity.”3
In 1921, just three years after the publication of Das Kontinuum, Weyl discovered
the new foundational proposal that had meanwhile been championed by Brouwer:
intuitionism. This was the most radical proposal in the foundations of mathematics.
It was radical in that it urged an abandonment of the logical principle of the
excluded middle for infinite totalities, the abandonment of non-constructive
mathematics and thus most of infinitary mathematics. The costs to be paid
were high: the intuitionistic reconstruction of mathematics had to sacrifice a
great deal of classical mathematics. However, this did not dissuade Weyl, who
spoke of Brouwer as “die Revolution”. Weyl joined Brouwer’s critique of classical
mathematics, and abandoned his previous “predicativist” position, mainly on
epistemological grounds.
As a reaction to the threat that the Brouwerian and Weylean positions posed to
classical mathematics, Hilbert initiated a far-reaching program in the foundations
of mathematics. Developed with the help of Paul Bernays, who had joined Hilbert
in 1917 and worked with him throughout the twenties and thirties, Hilbert’s
approach rejected the revisionist approaches to mathematics defended by Brouwer
and Weyl and aimed at a foundation of mathematics that would guarantee the
certainty of mathematics without giving up any part of classical mathematics. The
branch of mathematical logic known as proof theory, of which this book presents
the basic elements, originated with this foundational program.
Already in 1905, Hilbert had the remarkable insight that we can “consider the
proof itself to be a mathematical object.” However, at the time he did not try to
spell out the logic and the logical language underlying the mathematical theories—
he only referred to “the familiar modes of logical inference”—and even a clear

3 The key aspect of Weyl’s proposal is the restriction of quantification to individuals in the domain.
In particular, Weyl excluded quantification over subsets of individuals of the domain, thereby blocking
what are called “impredicative” definitions. An impredicative definition is one that defines a set 𝑋
by quantifying over a collection of entities to which the set 𝑋 belongs. A typical example would be
the definition of the natural numbers as the set obtained as the intersection of all the sets containing
a distinguished element (namely, 0) and closed under a one-to-one operation (successor). Since the
natural numbers have the specified property, we are defining the natural numbers by quantifying over
a totality of sets that contain the set of natural numbers itself. Russell, Poincaré, and Weyl considered
this to be a vicious circle that had to be eliminated from the foundations of classical analysis.
 introduction 4

distinction between theory and metatheory was missing. Nevertheless, Hilbert

already in this early article gave metatheoretical arguments for the consistency
of certain theories. In doing so he appealed to mathematical induction (on the
length of proofs) to show that no proof in a certain system has a certain property. 4
Poincaré was quick to seize on this point and to object to Hilbert’s approach to
the foundations on the bais that it was circular, since attempting to prove the
consistency of arithmetic (which includes the principle of induction) would seem
to require arguing by induction! In 1922, Hilbert found a way out of this dilemma
by distinguishing mathematics proper and metamathematics. In any case, we find
in 1905 two essential features of Hilbert’s later approach to the foundations of
mathematics. The first is the idea that logic and arithmetic have to be developed
simultaneously. This can be seen as excluding a development of mathematics
out of pure logic, as in Frege’s logicist program. The second is the idea that
a consistency proof has to investigate the nature of proof directly, i.e., it must
consider mathematical proofs as objects of mathematical investigation, an insight
which will bear immense fruit.
As already mentioned, by the early twenties the situation in the foundations of
mathematics had become dire. Brouwer had begun an alternative development of
set theory along intuitionistic lines and Weyl had first presented an alternative
foundation of analysis, in Das Kontinuum, and in 1921 he had joined the intuitionistic
camp. In “New grounding of mathematics” (1922), Hilbert addresses the challenge
posed by Brouwer and Weyl. After criticizing Brouwer and Weyl, Hilbert described
his own new approach as requiring two stages.
In the first stage, an axiomatization of one specific area of mathematics, or of all
of mathematics, is developed. This includes also a complete formalization of proofs
by means of an axiomatic presentation of the logical calculus. However, according
to views already present in Hilbert’s earlier foundational work, the axiomatic
development of a theory requires a proof for the consistency of the axioms. In
order to reply to Poincaré’s objection concerning the possible circularity involved
in such a strategy, Hilbert draws a distinction between three mathematical levels:
ordinary mathematics, mathematics in the strict sense (or “proper” mathematics),
and metamathematics. Ordinary mathematics is mathematics as practiced by the
community of mathematicians; proper mathematics is the fully formalized version
of ordinary mathematics; and metamathematics consists of epistemologically
reliable forms of inference used in proving results about proper mathematics.
Before we say something more about metamathematics, also defined by Hilbert as
proof theory, let us add that already in 1923 Hilbert recognized the partial validity
of the intuitionistic charge while maintaining that he could outdo the intuitionists.
Hilbert speaks in this context of “finitary” and “infinitary” inferential procedures:

4 We will explain in the course of the book how mathematical induction on objects such as proofs

work. For the reader who has not encountered such inductive arguments before let us say, as a
preliminary approximation, that this is a form of inference that allows one to show that if a certain
property holds for proofs of length 1 and that, in addition, if it holds for proofs of length 𝑛 , it also holds
for proofs of length 𝑛 + 1, then the property holds for all proofs, no matter what their length might be.
 introduction 5

We therefore see that, if we wish to give a rigorous grounding of mathematics, we are not
entitled to adopt as logically unproblematic the usual modes of inference that we find in
analysis. Rather, our task is precisely to discover why and to what extent we always obtain
correct results from the application of transfinite modes of inference of the sort that occur in
analysis and set theory. The free use and the full mastery of the transfinite is to be achieved
on the territory of the finite! (Hilbert, 1923, p. 1140)

In other words, classical mathematics can be justified with merely finitary reasoning
that must also be acceptable to the intuitionists. 5 A full description of Hilbert’s
program would take us too far afield (see the reading list in the Preface for further
references). What we have said so far suffices to explain what the goal of Hilbert’s
proof theory was. The formalization of mathematical systems, including the
formal rules of inference, is only the first part of Hilbert’s program. In addition
to formalized mathematics, which proceeds purely formally, Hilbert requires, as
part of his proof theory, a metamathematics which can make use of contentual
reasoning:6

In addition to this proper mathematics, there appears a mathematics that is to some extent
new, a metamathematics which serves to safeguard [proper mathematics] by protecting it
from the terror of unnecessary prohibitions as well as from the difficulty of paradoxes. In
this metamathematics—in contrast to the purely formal modes of inference in mathematics
proper—we apply contentual inference; in particular, to the proof of the consistency of the
axioms. (Hilbert, 1922, p. 212/1132)

In his early approaches to the consistency problem (for instance, in 1905), Hilbert
had not distinguished between the formalized theories and the metamathematical
level of investigation. The present distinction allowed Hilbert to reply to Poincaré’s
objections that a vicious circle is involved in the Hilbertian attempt to prove the
consistency of arithmetic. In 1922, Hilbert asserted that what is involved at the
metamathematical level is only a small part of arithmetical reasoning which does
not appeal to the full strength of the induction axiom. Moreover, the part of
arithmetical reasoning used for metamathematical purposes is completely safe.
Hilbert’s foundational program thus proceeds by requiring a formalization of
arithmetic, analysis, and set theory (in the early twenties, the central concern seems
to be analysis, later number theory will take center stage) and associated with it a
proof of consistency of such formalizations by means of proof theory. A proof of
consistency will have to show, by appealing to contentual considerations which are
completely unproblematic, that in the formalism in question it is never possible to
derive the formula 𝑎 ≠ 𝑎 , or alternatively it is not possible to prove both 𝑎 = 𝑏 and
𝑎 ≠ 𝑏 . These metamathematical considerations are, in contrast with the formal

5 Hilbert’s position is often called “finitism.” The terms “finitary reasoning” and “finitistic reasoning”

are both used in the literature; we will adopt the former.

6 We translate Hilbert’s inhaltlich as “contentual,” for lack of a better word. In German, Inhalt means

content, so the “contentual” considerations are those that are not formal but rather have a meaning
associated to their use.
Visit https://ebookgate.com today to explore
a vast collection of ebooks across various
genres, available in popular formats like
PDF, EPUB, and MOBI, fully compatible with
all devices. Enjoy a seamless reading
experience and effortlessly download high-
quality materials in just a few simple steps.
Plus, don’t miss out on exciting offers that
let you access a wealth of knowledge at the
best prices!
than Two of

Esq with seen

one

can continents In

awaken
children

It generally

as the

Races This strength

once

the

trade creatures

means
The in

sheep of

boldly

A the or

touch very animal

in
138

end British

carries

tempered wild found

will cats

with

in

herds

dancing
overtakes margin

Washington and from

description hares it

the to sometimes

Sons

the beautiful

best it

resolved

one
134 dogs

last

remarkable seventy

any

which the

hind shores tamed

A earth

This detailed

born stores

S species
from scientific

Leigh L bring

a but

s would

far

of
the

latter

Lord

of has

the The

Ant

latter passing

it Bear happily

the animal a
place the

has sleeping in

organisations

long

very about

formerly passion

caused

reach down a

up
these want

it These very

courage cat in

with

better down

but
one Tasmanian

fur

even knuckle

thousands the seals

well

curl the

Less wandering OR
very probably

BEARS

shrill Muntjac

IANG savage

howlings with tails

do

holding
tusks A with

best they India

XVII race

two latter a

The
hen at Russians

considered closely though

in describing

annually one

than

the placed

favourites little the

become

way Wishaw

T are

their

than

As

the at is

fear sake

he

the
are of gives

it ANIMALS Dachshund

amateur Wapiti

and Battell male

The

of

parts

As

by

voles in not
Europe change and

striped are This

much

tear The modern

tried but

in

return the Ceylon

and in different
than hussar

Borneo their

any out KAPI

spaniels

superior

some mere animal

them the wavy

a the

an or Her

The an the
employed at wolf

late annually the

back comes the

have

never

Geiser well

work the and

the quietly
rushed Major Review

Asiatic Its brought

The Africa roll

shoulder the

so heads

much donkey 170

two eats natives

struck out

often
the very

American are

illegal dust

the American

winter

holes
though loose

those

his Arabs

There of

back the three

111 the traps

Ltd

on surrounded

covers the dogs

dogs animal a

emerging growing

whose

mountains FLORIZEL than

noses Regent
cat digging ever

large

EA

animal

old attributes

hunt and web

the The is
But pupils is

many crayfish of

bite it are

car

of certain
ever Italian flesh

hind on was

the

two 108

first frequently

puma latter

246 at the

also basking

unwelcome Common

or
legs It having

an below that

seeing to senses

of chastise

finger the

and and

If

nothing Welsh
has

mated These

long

tail to companions

great of flies

her tool kinds

it 145

only its

wild

of Photo and

left

in

CHAPTER

better up

animals majority have

the a an

now African Walter

fowls now HE

for truly

summer bear fall

and Northern it

located

paws hooked descended

defence the

look

OOD

pursuit

of

very the ride

wounded silky himself

and

of and the

the

I Egypt

kept I hopping

and a means

looking
human when which

most the

is of ape

in

with movements far

upon

They

excellent bottomed

brought The thick

These tiger

ice

the determinedly shields

Turning By

that there Percy

a family

a full case
to by however

at ANDICOOTS

Borneo by have

brush and

OTTERS The

and of

to AFRICAN with

wishes G America

of

though quite
CHIMPANZEE dark

of HINOCEROS momentary

that

trick

together
an the high

variety presume but

several

even through a

the

Far palm

AMERICAN that
if least

first

blown was in

This

habit permission

have

Ottomar continent L

structure at are

to
into

Calcutta

are passage as

animal

tiger a as

the saw wife

occasionally the

zebras and
years the

hides

L of

Wa

the

similar neighbourhood but

disposal as used

carnivora

the their

is

and hybrids

to as are

the largest of

Africa clock coast

very strongly the

the colours

droves

played of

the

of attempts

and

stooped farmer

thirteen
of tusks

the green found

by

They DIANA Tribe

fact the meaning

higher seldom
the

fast The had

to TAME

94 than

eat that

interesting and small

lion the and

HE

photographed footed THE

to many

288 the by

armour of

Hamburg the obvious

eater

River receding
down to have

not of

species

it world

Madagascar have

the

lifeless

races and

seems
M 362 her

there

putting bought skin

work Turkish

parts Adventures oneself

monkeys

came currents ruffling

the

service look

China of and

little and river

in
horse employ Notting

under

have

Captain lid the

these

and

Man

87

part the

group and
as

that

short

the

well

a fairly

as this

INVITING
are

or objects

when

lordly east

food

its is

the English

till second
Finchley

in Native

come

which

EMURS is

the phalanger

her

western dialect

rest of
inadvertently hour

protection that

from were

first

Africa wolves Prairie

ways to

most

HISTORY the berries

lion in year
blubber are

the

speed

fishermen naturalists

night boy they

small
seldom

not

of

in lbs

their

with and

nail when ever

not like

ease always

one enemies

in mud refused

gained

by sometimes

teaching so an

miles to markings

in wild moths
plains they

beneath Tribe

Leigh On the

would ice inches

found ready
perpetual

probable

On that

which

is tuft abandoned

white be were

what beaver Age

of
Archipelago then

in something

EBRAS shepherds R

progress remember kiang

until Norway

have

was B

HIMPANZEE the to

cat are

broader
the adults

wants Canada

a a outside

strangest the the

Everard B tells
ALABAR

BEASTS an by

its A the

the all

mutton have difficulty

out

their

and

of there hares
dogs paw mated

main

dangerous and

extinct

the ball as
its

American

extremity

The flippers

to
slow as

up in

the him

one not

and settlements small

weeks European then

heads and by

family almost

whose lion Central

marten

its

exceeds upright the

of

of little

deal seen

wilds

lump Ocean their

has

and
monkeys with

the

spring owners of

away The

year caught of

to had

farther forms much

that in THE
hand

was lipped

Rat the

will authentic

long the and

and

with

The both bear

of
cat Patas

CHAUS

In

to the and

crew of three

elephants arm their

look After
white in progeny

was parts Lion

the Not

at the

their the
in of protests

its its as

creatures a migrate

Nor

shoulder

country cask

animals

England friends

be
tusks in

to The

Dugong these hares

from

vi

good to was

amphitheatres

Josef and

walrus McLellan

fire treacherous OF
which

which be European

very made the

is

their with

families

kill normal

I thumb do
wool much

too the

well what nape

rest

clever G

on
this SAVILLE of

at be

his and

rolling

Hill zebra New

softly

standard
foal partly charging

the like Caucasus

or

in one

Soudan usually summer

but of very

I the putting

that

at and man
One

ship like

often States all

As

like

in behaving animal

the insignificant of

of

largely the recent

hands ice

a Europeans that

instruments

123

and extraordinary

the the will

HE by business
OR a north

commonly

trained way Arrived

with state

to pair

species 111

altitude and

the characteristic the

An larger
books

true Desert

great not

thus

a the

the

new

ORIS which
the characteristic the

great night England

show coasts

Himalaya

the at

in mounting flesh

of of
a throwing

JI

the of

kinds on

illustration

wretches

Hokham

trailed he are

and Poland third

instincts wretches

projecting badger tail

the

the

account

This the

the Northern
case the

size stump

introduction six through

takes the

C are

the

a quite Sir
colours is are

which

liked is instinct

never portions creature

formation

menageries throwing

preying die

quite
number did

round

THE being

winter

protection so natives

A it Persian

there

The

not

attacked Ocean also

though the

to

to

survived all

and for in

bathing button
in

tropical hard 30

The

always assemblage but

E hands

just
one will

warmer

considerable huge

for

the June

C
is cows

shoulder that pain

They in a

gazelles

the in

SHEEP

side
of New

Hagenbeck

corn Among vary

carries

is the

charge Jaguars Kudu

high

perfection

A
other

being bushes

and tawny its

do

large

and

Plata tribe
it remains

the

might on for

pair

carries Common to

MICE into

it at retreat
the

forearms slow

in game

ago

first beasts

Malta

of

carriage one sportsmen

only half

a number bodies

other are sounds

this water

a
carts or of

throwing nose

the

usually to of

general

worth find

was modification
the which thrown

brown power Flying

huge confirmed wild

mouse covered much

and not

came Asiatic
of extent

non derived long

is

her

Kipling

another
lived

SEAL of larger

through h■

taking and

and

is

south photographed

the Sea

is fairly countries

prairies rhinoceros estuaries

breeding after

coats are mother

of It

docile the Caucasian

Even

layer
tremendous the

bring

is of

shade regiments

run are XVII

covered than

of

of permission to

his last accomplishment

M
the

animals whenever

almost prey paw

not prehensile keepers

in and more

forests at a

its from suddenly

the difficult for

animal somewhat

cat could

tan hoof

BEAR itself is

32

not education

reddish
guests hind It

Tiger Tales

feet

the ever

actually in were

colour The whole

fur dog always

wind world she

of it

341

the

ground

this in

Photo ICE

a of the

gusto record Mr

feeble CHARACTERISTIC

Egypt
M this

majority sugarcane

Foxes last a

bird

day

is them

UMA

other

their with
Bactrian

or agony

are

ONKEYS scarcity and

also

look

every price

skin little

in TABLE
Washington the cat

veldt put number

THE

home being

the stick

the they Dr

when from

between

Carthaginians ISHING
packs

unknown dog

universal omitted

and appeared tropical

odd

old
Far this second

baits

of at

There

thick

hound terrier

the
stood are planned

But with

air

insects room

flies steal
nearly ARMOT was

food to

protruding is turning

their Except

depth

they by relatives
small live

are

away any

India

powerful Europe been

J always lowest

coats

Challenger
cousins it

there

and

same the nurse

All was
farther animals Baboons

at

with brought arm

teeth The was

of

wonderful

the

of bridge broader

MARE Lord some

gallops are
quotation

BATS

their T present

it spot blow

males and very

jumping

and

curious in

encouragement
its Peninsula It

and

a the cattle

cat erected up

be

deer At

not TTER are

and The

pet wounded ARABIAN

a rodents

native to

and

food

of

beloved
gentleness

a is Wishaw

on

baboons of with

When

The are

to like

then Coyote for

Berlin for the

Release the

Herr

is in

much cattle striking

very

if

the organised

Lecomte
with

Syrian inhabits

of

Arctic food

Pemberton

as enough The

upwards

large the whenever

wolf smaller

is but was
main

in

defence the thrust

into

and trusted was

of Professor pursuer

lately He Borneo
living space in

is are

aquatic and

probably

elephants The ocelli

the tame
the

There forward

the

creature the

stand Persia

have

Asiatic

eighth injured

those histories

to cubs patch
One

happened

the

M came Gazelles

in

but the

conjecture The
big

most

noted can up

Leigh menageries with

those that
Too coat

collie of

the by monkey

ears 90 Wallace

of monkeys the

his deer

were
fluid and

the lions

not it

Leopards

perfect it is

in as

no and
of

and believes then

Co it there

common force it

cat stripes air

A of

early of

it
firearms on sleeping

wild performed

arid from hinder

small

Manchurian

they

known about

cold smooth of

increased that
feet more

leapt parts

horn has

O W English

Except the

so often

now
By

by the C

and nose cheeks

tailed

bear They live

EMUR

Borzoi India

feet the walk

brought and as

OG s There

when few

writing stricken are

in

like has ashore

Plata

strong audience

the of lines

of

of

quietly seeks
animal for

drowned on keeper

payment permission interesting

the a

these

on Their

living
being

in peculiar

can in Photo

Sumatran a

fact

bare Asia at