Fuzzy Logic and Mathematics: A

Historical Perspective 1st Edition

Radim B■lohlávek
Visit to download the full and correct content document:
https://textbookfull.com/product/fuzzy-logic-and-mathematics-a-historical-perspective-
1st-edition-radim-belohlavek/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...

Recent Advances in Intuitionistic Fuzzy Logic Systems

and Mathematics Said Melliani

https://textbookfull.com/product/recent-advances-in-
intuitionistic-fuzzy-logic-systems-and-mathematics-said-melliani/

A First Course in Fuzzy Logic Fuzzy Dynamical Systems

and Biomathematics Theory and Applications 1st Edition
Laécio Carvalho De Barros

https://textbookfull.com/product/a-first-course-in-fuzzy-logic-
fuzzy-dynamical-systems-and-biomathematics-theory-and-
applications-1st-edition-laecio-carvalho-de-barros/

A First Course in Fuzzy Logic 4th Edition Hung T.

Nguyen

https://textbookfull.com/product/a-first-course-in-fuzzy-
logic-4th-edition-hung-t-nguyen/

A Modern Introduction to Fuzzy Mathematics Apostolos

Syropoulos

https://textbookfull.com/product/a-modern-introduction-to-fuzzy-
mathematics-apostolos-syropoulos/
MATLAB Fuzzy Logic Toolbox User s Guide The Mathworks

https://textbookfull.com/product/matlab-fuzzy-logic-toolbox-user-
s-guide-the-mathworks/

Mathematical Modeling using Fuzzy Logic Applications to

Sustainability 1st Edition Abhijit Pandit

https://textbookfull.com/product/mathematical-modeling-using-
fuzzy-logic-applications-to-sustainability-1st-edition-abhijit-
pandit/

Popular Sovereignty in Historical Perspective 1st

Edition Richard Bourke

https://textbookfull.com/product/popular-sovereignty-in-
historical-perspective-1st-edition-richard-bourke/

Discrete Mathematics via Logic and Proof Calvin Jongsma

https://textbookfull.com/product/discrete-mathematics-via-logic-
and-proof-calvin-jongsma/

Elementary Logic with Applications A Procedural

Perspective for Computer Scientists 1st Edition D.M.
Gabbay

https://textbookfull.com/product/elementary-logic-with-
applications-a-procedural-perspective-for-computer-
scientists-1st-edition-d-m-gabbay/
Fuzzy Logic and Mathematics:
A Historical Perspective
 Fuzzy Logic and Mathematics:
A Historical Perspective

Radim Bělohlávek, Joseph W. Dauben, and George J. Klir

1
 1
Oxford University Press is a department of the University of Oxford. It furthers
the University’s objective of excellence in research, scholarship, and education
by publishing worldwide. Oxford is a registered trade mark of Oxford University
Press in the UK and certain other countries.
Published in the United States of America by Oxford University Press
198 Madison Avenue, New York, NY 10016, United States of America.
© Oxford University Press 2017

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

a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by license, or under terms agreed with the appropriate reproduction
rights organization. Inquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above.

You must not circulate this work in any other form

and you must impose this same condition on any acquirer.

Library of Congress Cataloging-in-Publication Data

Names: Bělohlávek, Radim. | Dauben, Joseph Warren, 1944- | Klir, George J., 1932-
Title: Fuzzy logic and mathematics : a historical perspective / Radim
Bělohlávek, Joseph W. Dauben, George J. Klir.
Description: Oxford ; New York, NY : Oxford University Press, [2017] |
Includes bibliographical references.
Identifiers: LCCN 2016024541| ISBN 9780190200015 (hardcover) |
ISBN 9780190200039 (online content)
Subjects: LCSH: Fuzzy logic. | Logic, Symbolic and mathematical.
Classification: LCC QA9.64 .B4525 2017 | DDC 511.3/13–dc23 LC record available
at https://lccn.loc.gov/2016024541
1 3 5 7 9 8 6 4 2
Printed by Sheridan Books, Inc., United States of America
 Contents

Preface ix

Notes for the Reader xi

1 Aims and Scope of This Book 1

2 Prehistory, Emergence, and Evolution of Fuzzy Logic 5

2.1 Prehistory of fuzzy logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5
2.2 Emergence of fuzzy logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17
2.3 Evolving attitudes toward fuzzy logic . . . . . . . . . . . . . . . . . . . . . 27
2.4 Documented debates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 32
2.5 Evolution of supporting infrastructure for fuzzy logic . . . . . . . . . . 38

3 Fuzzy Logic in the Broad Sense 43

3.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43
3.2 Basic concepts of fuzzy sets . . . . . . . . . . . . . . . . . . . . . . . . . . . 44
3.3 Operations on fuzzy sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 46
3.4 Fuzzy intervals, fuzzy numbers, and fuzzy arithmetic . . . . . . . . . . 55
3.5 Fuzzy relations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60
3.6 Approximate reasoning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 69
3.7 Possibility theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 79
3.8 Fuzzy clustering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88
3.9 Methods for constructing fuzzy sets . . . . . . . . . . . . . . . . . . . . . . 94
3.10 Nonstandard fuzzy sets . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99

4 Fuzzy Logic in the Narrow Sense 105

4.1 From classical to fuzzy logic: Principal issues . . . . . . . . . . . . . . . . 106
4.2 Many-valued logics until the 1960s . . . . . . . . . . . . . . . . . . . . . . 112
4.2.1 Łukasiewicz finitely- and infinitely-valued logics . . . . . . . . 112
4.2.2 Gödel finitely- and infinitely-valued logics . . . . . . . . . . . . 129
4.2.3 Other propositional logics . . . . . . . . . . . . . . . . . . . . . . 134

v
vi Contents

4.2.4 Predicate logics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 147

4.2.5 Further developments . . . . . . . . . . . . . . . . . . . . . . . . . 155
4.2.6 The mid-1960s and Zadeh’s idea of fuzzy sets . . . . . . . . . . 159
4.3 Fuzzy logics with graded consequence . . . . . . . . . . . . . . . . . . . . 160
4.3.1 Goguen’s logic of inexact concepts . . . . . . . . . . . . . . . . . 161
4.3.2 Pavelka-style fuzzy logic . . . . . . . . . . . . . . . . . . . . . . . . 165
4.3.3 Further developments . . . . . . . . . . . . . . . . . . . . . . . . . 171
4.4 Fuzzy logics based on t-norms and their residua . . . . . . . . . . . . . . 177
4.4.1 Fuzzy logics based on t-norms until the mid-1990s . . . . . . 178
4.4.2 Product logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 187
4.4.3 Hájek’s logic BL and Metamathematics of Fuzzy Logic . . . 189
4.4.4 Logics related to BL . . . . . . . . . . . . . . . . . . . . . . . . . . 202
4.5 Fuzzy logic and uncertainty . . . . . . . . . . . . . . . . . . . . . . . . . . . 210
4.5.1 Degrees of truth vs. belief and truth functionality . . . . . . 210
4.5.2 Possibilistic logic . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 212
4.5.3 Gerla’s probabilistic fuzzy logics . . . . . . . . . . . . . . . . . . 215
4.5.4 Belief, modality, and quantifiers in fuzzy logic . . . . . . . . . 216
4.6 Miscellaneous issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 222
4.6.1 Relationship to applications . . . . . . . . . . . . . . . . . . . . . 222
4.6.2 Computability and complexity . . . . . . . . . . . . . . . . . . . 224
4.6.3 Further developments . . . . . . . . . . . . . . . . . . . . . . . . . 227

5 Mathematics Based on Fuzzy Logic 231

5.1 Principal issues and outline of development . . . . . . . . . . . . . . . . 231
5.1.1 What is mathematics based on fuzzy logic? . . . . . . . . . . . 231
5.1.2 The problem and role of foundations . . . . . . . . . . . . . . . 233
5.1.3 The problem and role of applications . . . . . . . . . . . . . . . 238
5.1.4 Outline of development . . . . . . . . . . . . . . . . . . . . . . . . 241
5.2 Foundations of mathematics based on fuzzy logic . . . . . . . . . . . . 246
5.2.1 The role of fuzzy logic in the narrow sense . . . . . . . . . . . 246
5.2.2 Higher-order logic approaches . . . . . . . . . . . . . . . . . . . 250
5.2.3 Set-theoretic approaches . . . . . . . . . . . . . . . . . . . . . . . 253
5.2.4 Category-theoretic approaches . . . . . . . . . . . . . . . . . . . 261
5.3 Selected areas of mathematics based on fuzzy logic . . . . . . . . . . . . 267
5.3.1 Sets and relations . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267
5.3.2 Algebra . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 282
5.3.3 Topology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292
5.3.4 Quantities and mathematical analysis . . . . . . . . . . . . . . . 298
5.3.5 Probability and statistics . . . . . . . . . . . . . . . . . . . . . . . 309
5.3.6 Geometry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315
5.3.7 Further developments . . . . . . . . . . . . . . . . . . . . . . . . . 320
 Contents vii

5.4 Miscellaneous issues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 321

5.4.1 Interpretation of truth degrees . . . . . . . . . . . . . . . . . . . 321
5.4.2 Fuzzy logic and paradoxes . . . . . . . . . . . . . . . . . . . . . . 332
5.4.3 Fuzzy logic and vagueness . . . . . . . . . . . . . . . . . . . . . . . 339

6 Applications of Fuzzy Logic 347

6.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 347
6.2 A historical overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 349
6.3 Engineering . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356
6.3.1 Fuzzy control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 356
6.3.2 Other engineering applications . . . . . . . . . . . . . . . . . . . 365
6.4 Decision making . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 371
6.5 Natural sciences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375
6.5.1 Chemistry . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 375
6.5.2 Biology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 376
6.5.3 Physics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 378
6.6 Earth sciences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 382
6.7 Psychology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 385
6.8 Social sciences . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 390
6.8.1 A historical overview . . . . . . . . . . . . . . . . . . . . . . . . . . 390
6.8.2 Economics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 392
6.8.3 Political science . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 396
6.9 Computer science . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 399
6.10 Medicine . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 407
6.11 Management and business . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411
6.12 Other applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 414

7 Significance of Fuzzy Logic 421

7.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 421
7.2 A retrospective overview of fuzzy logic . . . . . . . . . . . . . . . . . . . . 421
7.3 Paradigm shifts in science, mathematics, engineering, and other areas 425
7.4 Assessment of the significance of fuzzy logic . . . . . . . . . . . . . . . . 430
7.5 A prospective view on the 50th anniversary of fuzzy logic . . . . . . . 442

Appendices

A The Enigma of Cox’s Proof 449

B Overview of Classical Logic 453

C Photographs 469
viii Contents

Glossary of Symbols 473

References 475

Name Index 511

Subject Index 521

 Preface

he initial idea for writing a book of this kind emerged from discussions at
T the Eighth World Congress of the International Fuzzy Systems Association
(IFSA) in Taipei in August of 1999. With the approaching new millennium, the fo-
cus of the congress was on reexamining the past and pondering the future of theo-
retical developments as well as applications of fuzzy logic and mathematics based on
fuzzy logic. It was repeatedly suggested in the many discussions at the congress that
the time was ripe for writing a book describing comprehensively and in sufficient de-
tail the relatively short history of fuzzy logic and assessing its significance and likely
prospects in the new millennium. One of us (Klir) participated in these discussions
and the idea of writing such a book was appealing to him. He realized, however,
that the scope of this kind of book was too large for a single person to write, and so
the idea of writing such a book remained idle in his mind for more than a decade. In
2011, while working jointly with a former colleague (Bělohlávek) on another book
involving fuzzy logic (Concepts and Fuzzy Logic, MIT Press, 2011), he mentioned the
idea and they eventually agreed to work on this larger book together. However, due
to the historical nature of the book, they felt it would be useful to enlist someone
with expertise in history, especially the history of mathematics, and so they extended
an invitation to Joseph Dauben to join them in working on this challenging book,
which he accepted.
In 2012, the three of us began by converting the initial, somewhat haphazard
idea into a well-defined writing project. That is, we formulated our vision of the
aims and scope of the book and described how we intended to implement it. We also
defined a general structure for the book and determined roughly the envisioned con-
tents of individual chapters. This overall description of the book ultimately served
as the basis for our introductory chapter 1. We then explored several prospective
publishers and settled eventually on Oxford University Press, which we all consid-
ered the ideal publisher for a book like this.
Although we intended to work on the book as a team, it was reasonable and
practical that each of us would assume a particular role in the actual writing process.
Since two of us—Bělohlávek and Klir—had been active participants in research, ed-
ucation, and other activities regarding theory and applications of fuzzy logic, we

ix
x Preface

took the responsibility for most of the writing and divided it according to our inter-
ests: Bělohlávek assumed responsibility for chapters 4 and 5; Klir for chapters 2, 3,
and 6. Even though several parts—particularly of chapters 2 and 6—represent joint
work, the individual chapters differ slightly in their styles of writing. Because these
distinctions in style also reflect distinct backgrounds and experiences—indeed, Klir
and Bělohlávek represent two distinct generations of fuzzy-logic researchers—we
considered it desirable to preserve them. Although the primary role of Dauben—
a relative newcomer to fuzzy logic—was to provide his expertise in the history of
mathematics, he also critically evaluated the content of chapters 2–6 and partici-
pated actively in writing chapter 7.
Our cooperation with Oxford University Press, primarily via Peter Ohlin—the
editor responsible for this book in the New York office of OUP—has been efficient
and cordial from the beginning, for which we are very grateful. Initially, we under-
estimated the enormous scope of information to be covered in a book like this, but
we were determined to include all of the relevant and valuable historical informa-
tion that we could find, some of which had nowhere been previously documented.
Nevertheless, we have had to be very selective and, in some cases, we have had to
shorten some parts of our original text. However, it is our intent to publish at least
some of the deleted material in some form elsewhere in the near future.
We are aware that the publication of this book coincides with the 50th anniver-
sary of the genesis of fuzzy logic. This is fortunate, even though initially it was not
planned to happen this way. However, we believe that the book can now be suc-
cinctly described as covering in considerable detail the history of fuzzy logic during
the first fifty years of its existence, as well as offering a carefully argued assessment
of the significance of fuzzy logic at the end of its first fifty years. We hope our work
may also serve as a benchmark in so far as our vision of possible future developments
of fuzzy logic may be compared at some future time whenever another historical as-
sessment of fuzzy logic and its significance may be made.

Radim Bělohlávek, Joseph W. Dauben, and George J. Klir

Acknowledgments
We would like to express our gratitude to the following persons: Eduard Bartl, a col-
league of Radim Bělohlávek at Palacký University in Olomouc, Czech Republic,
who played a major role in typesetting the entire book using a LATEX style that he cre-
ated. Similarly, Ellen Tilden, editorial assistant to George Klir, read and copyedited
virtually the entire manuscript from a stylistic point of view, and we are grateful to
her for improving the overall readability of this book.
 Notes for the Reader

hese notes provide information about how the material covered in this book
T is organized. The aims and scope of the book as well as the content of its main
chapters, chapters 2–7, are presented in chapter 1. In addition to these chapters, the
book contains three appendices: appendix A concerns Cox’s theorem, which is dis-
cussed in chapter 2; appendix B provides an overview of classical logic; appendix C
consists of photographs with short biographical sketches of the major contributors
to the early development of fuzzy logic. The book also contains a glossary of mathe-
matical symbols widely used in the text, a list of references, a name index, and a sub-
ject index.
The list of references at the end of the book contains all of the major publi-
cations on which we have drawn in writing this book. Additional references not
directly concerned with its main subject are presented in footnotes, which are num-
bered consecutively within individual chapters, and also contain further informa-
tion related to the main text. For works published in languages other than English,
the original title is given with an English translation in brackets; whenever an English
translation of such works is also available, the respective bibliographic item contains
a complete reference to the translated version as well.
For stylistic matters, the 16th edition of the Chicago Manual of Style is gener-
ally followed. In cross-references, “p.” stands for “page” and “n.” for “footnote.”
In mathematical theorems, we use “iff” for “if and only if” and “w.r.t.” for “with
respect to.” The following abbreviations are used throughout:

ACM Association for Computing Machinery

AMS American Mathematical Society
Ann Annals
Bull Bulletin
IEEE Institute of Electrical and Electronics Engineers
Int International
J Journal
Proc Proceedings
Symp Symposium
Trans Transactions
Univ. University

xi
xii Notes for the Reader

Aeq Math Aequationes Mathematicae

American J Math American Journal of Mathematics
American J Phys American Journal of Physics
Ann Math Annals of Mathematics
Ann Pure Appl Logic Annals of Pure and Applied Logic
Arch Math Logic Archive for Mathematical Logic
Artif Intell Rev Artificial Intelligence Review
Bull AMS Bulletin of the American Mathematical Society
Civil Eng Syst Civil Engineering Systems
Comput Intell Computational Intelligence
Commun ACM Communications of the ACM
Europ J Operational Research European Journal of Operational Research
Fund Inform Fundamenta Informaticae
Fund Math Fundamenta Mathematicae
Fuzzy Sets Syst Fuzzy Sets and Systems
IEEE Trans Automat Contr IEEE Transactions on Automatic Control
IEEE Trans Fuzzy Syst IEEE Transactions on Fuzzy Systems
IEEE Trans Syst Sci Cyb IEEE Transactions on Systems Science and Cybernetics
IEEE Trans Syst Man Cyb IEEE Transactions on Systems, Man and Cybernetics
Inf Control Information and Control
Inf Sci Information Sciences
Int J Approx Reason International Journal of Approximate Reasoning
Int J Gen Syst International Journal of General Systems
Int J Intell Syst International Journal of Intelligent Systems
Int J Man Mach Stud International Journal of Man-Machine Studies
Int J UFKBS International Journal of Uncertainty, Fuzziness and Knowledge-Based Systems
J Applied Non-Classical Logics Journal of Applied Non-Classical Logics
J American Chem Soc Journal of the American Chemical Society
J ACM Journal of the ACM
J Comput Syst Sci Journal of Computer and System Sciences
J Logic Comput Journal of Logic and Computation
J Math Anal Appl Journal of Mathematical Analysis and Applications
J Pure Appl Algebra Journal of Pure and Applied Algebra
J Symb Logic Journal of Symbolic Logic
Lect Notes Comput Sci Lecture Notes in Computer Science
Lect Notes Logic Lecture Notes in Logic
Lect Notes Math Lecture Notes in Mathematics
Logic J IGPL Logic Journal of the Interest Group of Pure and Applied Logic
Math Logic Quart Mathematical Logic Quarterly
Notices AMS Notices of the American Mathematical Society
Philos Sci Philosophy of Science
Proc London Math Soc Proceedings of the London Mathematical Society
Proc Natl Acad Sci USA Proceedings of the National Academy of Sciences of the United States of America
Tatra Mt Math Publ Tatra Mountains Mathematical Publications
Theor Comput Sci Theoretical Computer Science
Trans AMS Transactions of the American Mathematical Society
Z Math Logik Grundlagen Math Zeitschrift für mathematische Logik und Grundlagen der Mathematik
Chapter 1

Aims and Scope of This Book

he subject of this book—fuzzy logic and its role in mathematics—has a rel-

T atively short history of some fifty years. The overall aim of this book is to cover
this short history as comprehensively as possible. This means that we intend to cover
not only theoretical and practical results emanating from fuzzy logic, but also moti-
vations and creative processes that led to these results. Such a retrospective reflection
is in our opinion essential for properly assessing the overall significance and impact
of fuzzy logic, and we feel that the time is ripe for it.
It seems reasonable to expect that the aims and scope of any scholarly book be
expressed in a nutshell by its title. Our choice of “Fuzzy Logic and Mathematics:
A Historical Perspective” was indeed intended to do so. First, although “fuzzy logic”
has multiple connotations, their common ground is the rejection of a fundamental
principle of classical logic—the principle of bivalence. This is basically an assump-
tion, inherent in classical logic, that any declarative sentence has only two possible
truth values, true and false.
By rejecting the principle of bivalence, fuzzy logic does not abandon the classi-
cal truth values—true and false—but allows for additional ones. These truth values,
which are interpreted as degrees of truth, may be construed in various ways. Most
commonly, they are represented by numbers in the unit interval [0, 1]. In this in-
terpretation, 1 and 0 represent the boundary degrees of truth that correspond, re-
spectively, to the classical truth values true and false. The numbers between 0 and 1,
with their natural ordering, represent intermediate degrees of truth. Either all real
numbers from [0, 1] or their various subsets, each containing 0 and 1, may be em-
ployed as truth values. Other sets of truth values are possible as well, provided that
they are at least partially ordered and bounded by the classical truth values.

1
2 Aims and Scope of This Book chapter 1

Embracing additional truth values and interpreting them as degrees of truth has
an appealing motivation, which may briefly be described as follows. Classical logic
is appropriate for reasoning that involves bivalent propositions such as “7 is a prime
number,” that is, propositions that may in principle be only true or false. Most
propositions that people use to communicate information about the physical world
are, however, not bivalent—their truth is a matter of degree.
As an example, consider the proposition “the outside humidity is high.” In clas-
sical logic, one is forced to assign to this proposition the truth value of either false
or true. To do so, one needs to choose a threshold value h such that the proposition
is considered true if and only if the outside humidity is larger than h (or, alterna-
tively, larger than or equal to h ). Then, for an arbitrarily small deviation, ε, in the
humidity from h , the proposition is considered true when the actual humidity is
h + ε and false when the humidity is h − ε. This abrupt change from truth to falsity,
which clearly does not agree with our common-sense understanding of the proposi-
tion “the outside humidity is high,” can be avoided when the principle of bivalence
is rejected.
In fuzzy logic, one may assign to any proposition not only 1 (true) or 0 (false) but
also other, intermediate truth degrees, such as 0.8. For a (slightly) higher humidity,
one naturally assigns the proposition a (slightly) higher truth degree, for example
0.85. Such employment of degrees of truth meets our common-sense understand-
ing of this proposition (and other propositions expressed in natural language) quite
well; it is one of the attractive capabilities of fuzzy logic. How to assign truth degrees
in the context of each particular application of fuzzy logic is an important issue that
we discuss later in this book.
What is the effect of degrees of truth on predicates? As is well known, the ex-
tension of each predicate in classical logic is a unique set defined within some pre-
determined universe of discourse. If the proposition “ x is P ” is true, then object x
is a member of P ’s extension; if it is false, then x is not a member of P ’s extension.
In fuzzy logic, where the requirement that “ x is P ” be either true or false is rejected,
a broader class of predicates is admitted. It includes predicates, such as “high hu-
midity,” “tall man,” or “low inflation rate,” for which the truth of the proposition
“ x is P ” is, in general, a matter of degree. Such predicates are called fuzzy predi-
cates. It is natural to view the degree of truth of each particular proposition “ x is P ”
as a degree of membership of object x in P ’s extension. Such an extension is then
a nonclassical set whose membership is a matter of degree. These nonclassical sets,
which are referred to as fuzzy sets, are thus essential components of predicate fuzzy
logic. However, their role in fuzzy logic extends beyond their connection with fuzzy
predicates, as is explained throughout this book.
So far, we have used the term “fuzzy logic” in its common-sense meaning. It is
in this way that the term is most often used, referring to all aspects of representing
and manipulating knowledge that employ degrees of truth. In addition, two other
chapter 1 Aims and Scope of This Book 3

meanings are recognized, namely fuzzy logic in the narrow sense and fuzzy logic in
the broad sense. Fuzzy logic in the narrow sense, also called mathematical fuzzy
logic, deals with formal logical systems in which truths of propositions are a mat-
ter of degree. These systems provide foundations for fuzzy logic in the broad sense,
which has a much wider agenda. Its primary aim is to emulate common-sense hu-
man reasoning in natural language and other human capabilities and utilize them
in various applications.
These few remarks characterize fuzzy logic in a generic way, as a variety-embrac-
ing concept. In this sense, the term “fuzzy logic” stands for a wide number of special
systems, which are distinguished from one another by various properties such as the
set of truth degrees employed, their algebraic structure, truth functions chosen for
logic connectives, and other properties. Since it is virtually impossible to cover in
detail the entirety of fuzzy logic in a single book, we have been selective and focus
primarily on the principal aims of this book. Whatever we consider essential for
achieving our aims, we cover in detail. The rest we try to cover as satisfactorily as
possible via short notes and accompanying references to the literature.
The literature dealing with fuzzy logic in its various forms and applications has
been rapidly growing for several decades and is now very extensive. It covers, on
the one hand, many important theoretical results concerning foundations of fuzzy
logic and, on the other hand, a broad spectrum of impressive applications of fuzzy
logic in engineering, science, and other areas of human affairs. It seems reasonable
to view these established theoretical results, successful applications, and the ongoing
rapid growth of the literature as indicators of the growing acceptance and viability
of fuzzy logic.
As the title of this book also suggests, our aim is not only to examine fuzzy logic
alone, but also to examine its role in mathematics. The proper way to examine the
role of fuzzy logic in mathematics and to assess its overall impact is to approach this
problem from a historical perspective. Let us elaborate a little on why we consider
this important.
As is well known, most technical publications in virtually any area focus largely
on presenting results and usually pay little or no attention to motivations and cre-
ative processes leading to those results. This means that valuable information re-
garding the emergence, evolution, and propagation of ideas within a given subject
is habitually suppressed. Such details often remain only in the memories of the peo-
ple who created them, and then pertinent historical details are lost forever when
these people pass away.
In fuzzy logic, fortunately, many of the pioneers and key contributors are still
alive, making it to a large degree feasible to reconstruct important but ephemeral in-
formation. This fleeting opportunity motivates us to embark on this project at this
time. Our ultimate goal is to assess the overall significance of fuzzy logic. To this
end, we believe, it is essential to characterize fuzzy logic not only from the standpoint
4 Aims and Scope of This Book chapter 1

of the current situation—by existing results—but also from the historical perspec-
tive or rather retrospective—by documenting processes leading to these results. To
achieve the above-mentioned aims, the book is organized as follows.
In chapter 2, we first identify instances throughout the long history of classical
logic when the principle of bivalence was challenged. This may loosely be viewed
as a sort of “prehistory” to fuzzy logic. We then describe in some detail circum-
stances associated with the emergence of fuzzy logic. Furthermore, we survey and
present chronologically events that contributed in a major way to the development
of fuzzy logic from its very beginnings to its present state. We also survey evolving
attitudes toward fuzzy logic, especially within the academic community, and sum-
marize various debates between members of the emerging fuzzy logic community
and opponents of fuzzy logic.
In chapter 3, which is devoted to fuzzy logic in the broad sense, we trace the
emergence of a broad spectrum of ideas motivated by the ultimate goal of develop-
ing sufficiently expressive means for emulating common-sense human reasoning in
natural language and some other unique capabilities of human beings.
In chapter 4, we first survey the development of many-valued logic before the
emergence of fuzzy sets. We then cover in a fairly comprehensive way the evolu-
tion of fuzzy logic in the narrow sense since the late 1960s and examine the various
propositional, predicate, and other fuzzy logic calculi that have been advanced.
In chapter 5, we explore foundations of mathematics based on fuzzy logic and
examine some of its areas. We show that the various concepts of mathematics based
on bivalent logic may be viewed as special cases of their extended counterparts in
mathematics based on fuzzy logic, and also that some additional useful concepts
emerge in the latter that have no counterparts in the former. In addition, we exam-
ine various semantic issues such as interpretation of truth degrees, the phenomenon
of vagueness, and resolution of certain paradoxes by fuzzy logic.
In spite of its relatively short history, fuzzy logic has already achieved high vis-
ibility, primarily due to its impressive applications in many areas. These applica-
tions are the subject of chapter 6. There, we cover fairly comprehensively the many
applications of fuzzy logic that have so far been developed in various branches of
engineering, science, medicine, management, and other areas.
An appraisal of the overall significance of fuzzy logic is the subject of chap-
ter 7. We regard fuzzy logic as a new paradigm, provide a retrospective overview of
its foundations and applications, and assess its significance from a variety of view-
points. We conclude by assessing the prospects of fuzzy logic as we see them on its
50th anniversary.
Chapter 2

Prehistory, Emergence,
and Evolution of Fuzzy Logic

n this chapter, various challenges to the principle of bivalence of classical logic

I throughout the whole history of logic are briefly surveyed. Circumstances that
led eventually to the emergence of fuzzy logic are then described in greater detail. We
examine how the initial, strongly negative attitudes toward fuzzy logic by the aca-
demic community have gradually become more positive. In this context, we sum-
marize various documented debates between researchers criticizing fuzzy logic and
researchers supporting it, which usually have also contributed to its development.
Finally, we outline several visible stages in the evolution of supporting infrastruc-
ture for fuzzy logic.

2.1 Prehistory of fuzzy logic

Our aim in this section is to focus solely on one aspect of the long and complex
history of classical, bivalent logic and mathematics based on this logic. Our only
interest here is to identify episodes when the principle of bivalence of classical logic
was challenged. Such episodes were actually extremely rare and of minor influence
prior to the 19th century.
The mainstream view of the history of logic (Bocheński 1961) associates the be-
ginnings of logic with ancient Greece, and Aristotle (384–322 BC) is routinely con-
sidered the founder of classical logic. This is primarily due to his development of
a system of inference schemes (syllogisms) for deductive reasoning, which influenced
the development of classical logic for some two thousand years. This system, usu-

5
6 Prehistory, Emergence, and Evolution of Fuzzy Logic chapter 2

ally referred to as syllogistic, is described in one part of Aristotle’s major work entitled
Organon (McKeon 1941, 62–107).
In developing his syllogistic, Aristotle fully subscribed to the principle of biva-
lence. However, he questioned the applicability of this principle to propositions
concerning future contingencies. This issue is discussed in one part of Organon,
which is entitled De interpretatione (On Interpretation) (McKeon 1941, 40–61). The
following is the essence of his arguments (p. 48):
Everything must either be or not be, whether in the present or in the future, but it is not always
possible to distinguish and state determinately which of these alternatives must necessarily come
about.
Let me illustrate. A sea-fight must either take place to-morrow or not, but it is not nec-
essary that it should take place to-morrow, neither it is necessary that it should not take place,
yet it is necessary that it either should or should not take place to-morrow. Since propositions
correspond with facts, it is evident that when in future events there is a real alternative, and
a potentiality in contrary directions, the corresponding affirmation and denial have the same
character.
This is the case with regard to that which is not always existent or not always nonexistent.
One of the two propositions in such instances must be true and the other false, but we cannot
say determinately that this or that is false, but must leave the alternative undecided. One may
indeed be more likely to be true than the other, but it cannot be either actually true or false. It is
therefore plain that it is not necessary that of an affirmation and a denial one should be true and
the other false. For in the case of that which exists potentially, but not actually, the rule which
applies to that which exists actually does not hold good. The case is rather as we have indicated.

Even though Aristotle certainly questioned the applicability of the principle of bi-
valence for propositions involving future occurrences, he actually did not abandon
the principle. He dealt with the problem regarding propositions involving future
events via the modalities of possibility and necessity.
While the above thoughts of Aristotle are well known, it seems virtually un-
known that he was also interested in a completely different phenomenon which
challenges the principle of bivalence and which is—unlike the problem of future
contingencies—in fact directly related to the idea of fuzzy logic, namely vagueness of
human concepts. As shown by the following excerpt from another part of Organon,
called Categoriae (Categories) (McKeon 1941, 7–37), Aristotle clearly recognized that
certain human categories apply to objects to various degrees and hence do not have
sharp boundaries. He was also aware of borderline cases, which he called interme-
diates, and alluded to the violation of the law of excluded middle by such categories
(pp. 27, 29–30):
Qualities admit of variation of degree. Whiteness is predicated of one thing in a greater or less
degree than of another. . . .
The qualities expressed by the terms ‘triangular’ and ‘quadrangular’ do not appear to admit
of variation of degree. . . . Thus it is not all qualities which admit of variation of degree. . . .
Those contraries which are such that the subjects in which they are naturally present, or of
which they are predicated, must necessarily contain either the one or the other of them, have no
intermediate, but those in the case of which no such necessity obtains, always have an interme-
diate. . . . Odd and even, again, are predicated of number, and it is necessary that the one or the
other should be present in numbers. . . . On the other hand, in those contraries with regard to
which no such necessity obtains, we find an intermediate. Blackness and whiteness are naturally
present in the body, but it is not necessary that either the one or the other should be present in
section 2.1 Prehistory of fuzzy logic 7

the body, inasmuch as it is not true to say that everybody must be white or black. . . . These pairs
of contraries have intermediates: the intermediates between white and black are grey, sallow, and
all the other colours that come between; the intermediate between good and bad is that which
is neither the one nor the other.

In his monumental book entitled Metaphysica (Metaphysics) (McKeon 1941, 689–

926), Aristotle even exercises the related idea that some claims may be more true
than others (p. 743):
Again, however much all things may be ‘so and not so’, still there is a more and a less in the nature
of things; for we should not say that two and three are equally even, nor is he who thinks four
things are five equally wrong with him who thinks they are a thousand. If they are not equally
wrong, obviously one is less wrong and therefore more right. If then that which has more of any
quality is nearer the norm, there must be some truth to which the more true is nearer.

However, neither of these considerations led him to abandon bivalence—then

a deeply rooted pillar of logical investigations.
Aristotle was not the only philosopher in ancient Greece who questioned the
universal applicability of the principle of bivalence in logic. The principle was ques-
tioned even more emphatically by one of his contemporaries, Epicurus (341–270
BC), and his followers—Epicureans. These philosophers basically rejected the prin-
ciple of bivalence on the basis of their strong belief in free will and their associated
commitment to the doctrine of indeterminism.
Aristotle’s hesitation regarding the general applicability of the principle of bi-
valence was criticized by Stoics, a group of Greek philosophers with strong interests
in logic, who were fully committed to this principle. Otherwise, it has not received
much attention, contrary to the great and lasting influence of his syllogistic. There
are only a few rare episodes in the history of logic before the end of the 19th century
when it reappeared as a subject of discussion and further development.
The first of these rare episodes occurred in the 14th century and is primarily
attributable to the well-known English philosopher William of Ockham (c. 1287–
1347). Educated in theology at Oxford University, he was a complex figure who
made important contributions to some areas of philosophy as well as theology. In
this book, we are only interested in his careful logical analysis of Aristotle’s discus-
sion about the truth status of propositions regarding future contingencies.1 Our
principal source in this regard is the book edited by Philotheus Boehner (1945),
which contains Latin originals of Ockham’s Treatise on Predestination, God’s Fore-
knowledge, and Future Contingents, and an extensive English commentary on Ock-
ham’s works by Boehner. Details of Ockham’s logical analysis of chapter 9 of Aristo-
tle’s De interpretatione and of parts of Aristotle’s Topics are covered in his Commen-
tary to Perihermenias and in chapter 30 of the third section of part III of his Summa
1
There were apparently other philosophers who were attracted to this issue after Aristotle’s ideas became avail-
able in Latin due to the efforts of Albert the Great (1193–1280) during the second half of the 13th century. However,
Ockham’s analysis seems to be the only one, according to currently available historical facts, which showed that
Aristotle’s arguments lead to the violation of the principle of bivalence and the need for a three-valued logic.
8 Prehistory, Emergence, and Evolution of Fuzzy Logic chapter 2

Logicae, whose Latin originals are included in Boehner 1945 as appendices II and III,
respectively.2
Ockham’s analysis shows that the arguments made by Aristotle about future
contingencies lead inevitably to propositions that are neither true nor false. Boehner
(1945) introduces the name neuter (N) as the third value—neither true nor false—for
such propositions. After examining Ockham’s arguments in greater detail, he shows
(Boehner 1945, 58–66) that in fact they contain a description of a truth function of
three-valued implication, p → q , which is shown in the following table:3
→ F N T
F T T T
(2.1)
N N T T
T F N T
Although Ockham explained by his own analysis that Aristotle’s argument a-
bout the truth status of propositions concerning future contingencies led inevitably
to a violation of the principle of bivalence, he did not accept these arguments, par-
tially for theological reasons and partially due to his strong belief in the principle.
Therefore, he did not take his prospective three-valued logic seriously, as is also con-
firmed by Boehner (1945, 66–69). As a consequence, his idea of three-valued logic
had virtually no influence on the development of logic.
In 1465, Aristotle’s arguments regarding the truth status of propositions of fu-
ture contingencies became a subject of extensive and heated debates at the Univer-
sity of Louvain (now in Belgium). The debates involved members of the Faculty
of Arts, who supported Aristotle’s arguments, and members of one fraction in the
Faculty of Theology, who rejected them on strictly theological grounds. These de-
bates, which lasted for ten years (1465–75) and extended far beyond the university,
are captured in considerable detail in English translations of transcripts of the vari-
ous debates that have been collected by Baudry (1989).4
This book offers fascinating and insightful reading. It contains an extensive
introduction, in which Baudry characterizes in detail the historical context within
which the debates took place, explains how they evolved, and introduces the key
debaters—Peter de Rivo, representing the Faculty of Arts, and Henry de Zomeren,
representing the Faculty of Theology. It is clear that the principal issue in these
debates was the principle of bivalence. In the end, however, neither side in these
ten-year debates was an obvious winner.

2 English translations by Marilyn McCord Adams and Norman Kretzmann of the Treatise (Ockham 1321–24)

and appendices II and III of Boehner 1945 are included in Adams and Kretzmann 1983; the two appendices from
Boehner 1945 are included in Adams and Kretzmann 1983 as appendices II and III. Appendix II contains, in addi-
tion, English translations (from medieval Latin versions) of relevant passages from Aristotle’s De interpretatione.
3 Adapted from p. 63 of Boehner 1945.
4 Translated into English by Rita Guerlac from the French original, L. Baudry, La querelle des futurs contingents:

(Louvain 1465–1475) (Paris: Vrin, 1950).

section 2.1 Prehistory of fuzzy logic 9

Since the end of the 17th century, the considerations on vagueness of human
concepts reappeared, with no apparent influence of Aristotle’s early analyses, first
in the works of John Locke and then continued to appear in textbooks on philoso-
phy and logic, such as in the established textbook of Watts (1724) and that of Bain
(1870). Vagueness of concepts and in particular the existence of their borderline cases
and unsharp boundaries naturally fitted Locke’s empiricism and became clearly rec-
ognized in these works. Interestingly, in order to describe such concepts Bain even
deemed it necessary to use a “numerical scale for stating the amount or degree of each
property.” The respective developments along with relevant quotations are covered
in section 5.4.3 below. Even though the principle of bivalence is clearly challenged
by such considerations, these works did not question it explicitly.
In the late 19th century, the principle was challenged again by logicians and
mathematicians. They argued that the principle is unnecessarily restrictive and that
it is possible to develop various alternative logics by abandoning it. Some alterna-
tive logics, in each of which more than two truth values were recognized, started
to emerge in the late 1910s and early 1920s. However, some relevant initial ideas re-
garding such logics had already appeared in the literature in the late 19th century
and early 20th century. Authors of these earlier ideas, which we cover only briefly,
were Hugh MacColl (1837–1909), a Scottish mathematician, Charles Sanders Peirce
(1839–1914), the well-known American philosopher with strong interests in logic
and mathematics, and Nikolai A. Vasil’ev (1880–1940), a Russian philosopher. It is
nevertheless interesting to note that already in his analysis of classical logic, George
Boole (1815–1864) briefly considered the idea that some naturally derived algebraic
expressions allow for trichotomy in logical values but concluded that such expres-
sions are “not interpretable in the system of logic” (Boole 1854, chapter III).5
Within the period from 1877 through 1908, MacColl published numerous pa-
pers and a book (MacColl 1906), in which he presented many thought-provoking
ideas addressing a broad spectrum of issues pertaining to symbolic logic.6 He is con-
sidered a forerunner of many-valued logics due to his work in modal logic, which
was apparently motivated by his distrust of material implication and his search for
strict implication. He outlined a system of propositional logic in which proposi-
tions are qualified as true, false (both “in a particular case or instance,” in MacColl’s
terms), certain (“always true,” “true in every case,” “its probability is 1”), impossible

5
Several interesting remarks in this respect are contained in a thorough analysis of Boole’s works by
T. Hailperin. In section 3.6 of his Boole’s Logic and Probability, 2nd ed. (Amsterdam: North-Holland, 1986), he
shows that certain expressions, which Boole used in addition to the classical truth values 0 and 1, may be thought of
as truth values of a particular four-valued logic. In “Boole’s algebra isn’t Boolean algebra,” Mathematics Magazine
54 (1981): 172–84, Hailperin argues that what Boole described was actually an algebra more general than a Boolean
algebra, namely a nonidempotent commutative ring with unit and without nilpotents which may be seen as an
algebra of nonclassical logic from today’s perspective.
6
This book and MacColl’s papers are reprinted, together with some of his correspondence and published de-
bates regarding his ideas, in S. Rahman and J. Redmond, eds., Hugh MacColl: An Overview of His Logical Work
with Anthology (London: College Publications, 2007). This book also contains biographical notes and a critical
overview of MacColl’s logical work.
10 Prehistory, Emergence, and Evolution of Fuzzy Logic chapter 2

(“its probability is 0”), and variable (“neither impossible nor uncertain,” “probabil-
ity is . . . some proper fraction” between 0 and 1). MacColl, however, treated these
qualifications as modalities of propositions and seems not to have been consider-
ing explicit assignments of intermediate truth degrees such as probability degrees to
propositions.
In 1902, Peirce wrote an essay entitled “Minute logic” (published only post-
humously in his Collected Papers) in which he envisioned a new mathematics based
on three-valued logic.7 He referred to this mathematics as a trichotomic mathemat-
ics and to the underlying three-valued logic as a triadic logic. Unfortunately, his
work on triadic logic is documented only by occasional remarks in his publications
or letters and, more importantly, in his handwritten working notes. Analyzing this
fragmentary evidence, Fisch and Turquette (1966) came to a reasonable conclusion
that Peirce had succeeded by 1909 in developing basic ideas of some three-valued log-
ics. These logics are among those that were introduced and studied in the 1920s by
the pioneers of many-valued logics, as we describe later in this section. Why Peirce
did not publish these ideas remains, according to Fisch and Turquette, an unsolved
mystery.
In the early 1910s, Nikolai A. Vasil’ev (1880–1940), who was at that time with the
Kazan University in Western Russia, explored the possibility of a new logic, distinct
from classical logic, which he called imaginary logic. He presented his ideas for the
first time in a lecture at the Kazan University on May 18, 1910, and published them
in Vasil’ev 1910. During 1910–13, he developed his logic further and wrote several
papers, including the important Воображаемая (неаристотелва) логика [Imagi-
nary (non-Aristotelian) logic]. Except for a small note written in 1925,8 his works
are written in Russian.9 Vasil’ev chose the term “imaginary logic” because he saw
a parallel between his ideas and those of N. A. Lobachevsky who in the early 19th
century had also been at Kazan University. Lobachevsky developed his imaginary
geometry—a kind of non-Euclidean geometry describing in a sense a different world
than that described by classical, Euclidean geometry. Vasil’ev’s main idea was that
in addition to our world, which he saw as being described by the Aristotelian logic,
there are other, imaginary worlds for which one needs logics different from the clas-
sical one. Each such logic, which he therefore called imaginary, differs from classical
logic in the laws it satisfies. In particular, Vasil’ev considered logics that violate the
law of noncontradiction and the law of excluded middle. His ideas are, however,
somewhat ambiguously described in his writings because he never attempted to for-

7
C. Hartshorne and P. Weiss, eds., Collected Papers of Charles Sanders Peirce, vol. 4 (Cambridge, MA: Harvard
Univ. Press, 1933), paragraphs 4.307–4.323, 248–62.
8 “Imaginary (non-Aristotelian) logic,” in Atti del Quinto Congresso Internazionale di Filosofia, ed. G. della

Valle (Naples, 1925), 107–9.

9
They are available in N. A. Vasil’ev, Избранные труды [Selected works] (Moscow: Nauka, 1989). Vasil’ev life
and work are covered in V. A. Bazhanov, Николай Александрович Васильев (1880–1940) [Nikolai Alexadrovich
Vasil’ev (1880–1940)] (Moscow: Nauka, 1988); see also the short account in English, V. A. Bazhanov, “The fate of
one forgotten idea: N. A. Vasiliev and his imaginary logic,” Studies in Soviet Thought 39 (1990): 333–41.
section 2.1 Prehistory of fuzzy logic 11

malize them. As a consequence, the numerous existing works assessing his contribu-
tions contain mutually incompatible views—while some regard Vasil’ev as a pioneer
of many-valued logic, paraconsistent logic, and sometimes even intuitionistic logic,
some regard such conclusions as unwarranted and claim that he never propounded
the idea of multiple truth values. His idea of alternative logics nevertheless clearly
makes him a forerunner of nonclassical logics.
The first formal systems of many-valued logics were developed by Jan Łukasie-
wicz (1878–1956), Paul Bernays (1888–1977), and Emil Post (1897–1954). Although
they invented their logics independently and at about the same time, Łukasiewicz
not only has a slight priority in being first but his logics were also most influential
in subsequent developments. We therefore cover his ideas in some detail.
Łukasiewicz first developed a three-valued logic and presented it in a lecture to
the Polish Philosophical Society that he delivered in June of 1920 in Lvov. The
essence of this lecture was summarized in a short, two-page paper (Łukasiewicz
1920). In this paper, Łukasiewicz allows each proposition to have three values. Two
of them are the classical values 1 and 0, which represent truth and falsity, respectively.
The third value represents neither truth nor falsity and is interpreted as “possibil-
ity.” This value is denoted in this early paper by 2, but later on, the convenient sym-
bol 1/2 was adopted. Łukasiewicz defines the truth function, →, of his three-valued
implication, as shown by the following table:
→ 0 1/2 1
0 1 1 1
1/2 1/2
(2.2)
1 1
1 0 1/2 1

Furthermore, he defines the truth function, ¬, of negation for any truth value a as
¬a = a → 0,
which yields
a 0 1/2 1
,
¬a 1 1/2 0
and defines the truth functions of disjunction and conjunction in terms of → and
¬ via the formulas
a ∨ b = (a → b ) → b ,
a ∧ b = ¬(¬a ∨ ¬b ),
respectively. When defining his implication, Łukasiewicz remarks that this can be
done in different ways in three-valued logics, but he does not explain in this paper his
rationale for choosing his particular definition. He addressed this issue in another
paper ten years later (Łukasiewicz 1930). We describe his motivations in detail in
chapter 4. It is significant that the three-valued implication defined by Łukasiewicz
(2.2) is the same except for notation as the one defined by Ockham (2.1).
12 Prehistory, Emergence, and Evolution of Fuzzy Logic chapter 2

Although Łukasiewicz’s first written presentation of his three-valued logic ap-

peared in 1920, it is clear from the Farewell Lecture he delivered at Warsaw University
on March 7, 1918 (Borkowski 1970, 84–86) that he had already developed it in the
summer of 1917. The following are a few excerpts from this interesting lecture:
In this farewell lecture, I wish to offer a synthesis of my research, based on autobiographical
confession. I wish to describe the emotional background against which my views have developed.
I have declared a spiritual war upon all coercion that restricts man’s free creative activity.
There are two kinds of coercion. One of them is physical, which occurs either as an exter-
nal force that fetters the freedom of movement, or as internal impotence that incapacitates all
action. . . .
The other kind of coercion is logical. We must accept self-evident principles and the theo-
rems resulting therefrom. . . .
That coercion originated with the rise of Aristotelian logic and Euclidean geometry. The
concept was born of science as a system of principles and theorems connected by logical relation-
ship. The concept came from Greece and has reined supreme. The universe was conceived after
the pattern of a scientific system: all events and phenomena are connected by causal links and
follow from one another as theorems in scientific theory. All that exists is subject to necessary
laws.
In the universe conceived in this way there is no place for a creative act resulting not from
a law but from a spontaneous impulse. . . .
The creative mind revolts against this concept of science, the universe, and life. A brave
individual, conscious of his value, does not want to be just a link in the chain of cause and effect,
but wants himself to affect the course of events. . . .
And what does [sic] a scientist to do? . . .
He has two paths to choose from: either to submerge himself in skepticism and abandon
research, or to come to grips with the concept of science based on Aristotelian logic. . . .
Logical coercion is most strongly manifested in a priori sciences. In 1910 I published a book
on the principle of contradiction in Aristotle’s work, in which I strove to demonstrate that the
principle is not so self-evident as it is believed to be.10 I strove to construct non-Aristotelian logic,
but in vain.
Now I believe to have succeeded in this. My path was indicated to me by antinomies, which
prove that there is a gap in Aristotle’s logic. Filling the gap led me to a transformation of the
traditional principles of logic.
This gave rise to system of three-valued logic, which I worked out in detail last summer.
That system is coherent and self-consistent as Aristotle’s logic, and is much richer in laws and
formulae.
The possibility of constructing different logical systems shows that logic is not restricted to
reproduction of facts but is a free product of man, like a work of art. Logical coercion vanishes
at its very source.

Later in the 1920s, Łukasiewicz also worked on generalizing his three-valued logic to
n -valued logics and decided to label, for each particular n , the truth values by a set
Łn of equidistant rational numbers in [0, 1] defined in the following way:
0 1 n−1
Łn = {0 = ,
n−1 n−1
, . . . , n−1 = 1}.
Under this more natural notation the logic operations of his three-valued logic can
be conveniently expressed for all a, b ∈ {0, 12 , 1} by the following formulas:
a → b = min(1, 1 − a + b ),
¬a = 1 − a,
a ∨ b = min(1, a + b ),
10
See Łukasiewicz 1910.
section 2.1 Prehistory of fuzzy logic 13

a ∧ b = max(0, a + b − 1).
These formulas also describe every n -valued generalization of Łukasiewicz’s three-
valued logic, as presented together with other results in an important paper by Łu-
kasiewicz and Tarski (1930). They also describe the case when the set of truth values
are all real numbers in the unit interval [0, 1], as shown earlier by Łukasiewicz (1923).
Łukasiewicz’s finitely- and infinitely-valued logics play an important role in fuz-
zy logic and we therefore discuss them in detail in chapter 4. In that chapter, we also
thoroughly examine the above-mentioned logics of Bernays and Post as well as ad-
ditional many-valued logics which started to emerge since the early 1930s. They in-
clude the logic developed in the context of intuitionistic logic by Kurt Gödel (1906–
1978), which plays an important role in fuzzy logic in the narrow sense, the three-
valued logic of Dmitri A. Bochvar (1903–1990) for analyzing some paradoxes of clas-
sical logic and set theory, in which the third value is interpreted as “meaningless,”
the three-valued logic of Stephen C. Kleene (1909–1994), in which the third value
is interpreted as “not defined,” as well as some probabilistic logics and logics devel-
oped in the context of quantum mechanics. This information and much more on
many-valued logics is covered in chapter 4.
The literature on many-valued logics has been rapidly growing since the early
1920s. Early major sources in this area include the excellent and comprehensive sur-
vey of the literature on many-valued logics from the very beginnings until about the
mid-1960s prepared by Rescher (1969). His book covers almost 600 relevant refer-
ences. More references, covering the period from 1966 to 1974, are given in Wolf
1977. We should also mention the early classic monographs written by Rosser and
Turquette (1952), Zinov’ev (1963), and Ackermann (1967), as well as the edited books
by Dunn and Epstein (1977) and Rine (1984).
The late 19th century witnessed a revival of interest in vagueness. Peirce, whom
we mentioned in connection with his three-valued logic, elaborated on distinctions
between various facets of linguistic imprecision. Moreover, he recognized that it
is primarily the vagueness of general terms, ubiquitous in natural language, which
makes classical logic incapable of formalizing common-sense reasoning with state-
ments described in natural language. His definition of vagueness is clear and con-
vincing:11

A proposition is vague when there are possible states of things concerning which it is intrinsically
uncertain whether, had they been contemplated by the speaker, he would have regarded them
as excluded or allowed by the proposition. By intrinsically uncertain we mean not uncertain in
consequence of any ignorance of the interpreter, but because the speaker’s habits of language
were indeterminate.

Peirce repeatedly returned to vagueness in his many writings. He considered it un-

avoidable and thus important to study, and he captured his strong views about
11
J. M. Baldwin, ed., Dictionary of Philosophy and Psychology, vol. 2 (New York: Macmillan, 1902), 748.
14 Prehistory, Emergence, and Evolution of Fuzzy Logic chapter 2

vagueness in his short and remarkably clear statement: “Vagueness is no more to

be done away with in the world of logic than friction in mechanics.”12
Shortly after the death of Peirce, the famous Engllish philosopher, logician, and
mathematician Bertrand Russell (1872–1970) wrote a paper fully devoted to vague-
ness (Russell 1923). In this paper, Russell straightforwardly exposed the limitations
of classical two-valued logic, to which he himself had immensely contributed. The
following excerpts from the paper capture quite well the essence of Russell’s criti-
cism (italics added):
Vagueness and precision alike are characteristics which can only belong to a representation, of
which language is an example. They have to do with the relation between a representation and
that which it represents. Apart from representation, whether cognitive or mechanical, there can
be no such thing as vagueness or precision; things are what they are, and there is an end of it . . .
The law of excluded middle is true when precise symbols are employed, but it is not true
when symbols are vague, as, in fact, all symbols are. . . .
The notion of “true” and “false” can only have a precise meaning when the symbols em-
ployed—words, perceptions, images or what not—are themselves precise. Since propositions
containing nonlogical words are the substructure on which logical propositions are built, it fol-
lows that logical propositions also, so far as we know them, become vague through the vagueness
of “truth” and “falsehood.” We can see an ideal of precision, to which we can approximate indef-
initely; but we cannot obtain this ideal. . . . All traditional logic habitually assumes that precise
symbols are being employed. It is therefore not applicable to this terrestrial life, but only to imag-
ined celestial existence.

It is interesting to see, at the beginning of this quotation, that Russell viewed

vagueness and precision as characteristics of a representation rather than as charac-
teristics of the things being represented. This is compatible with Peirce’s view, as
well as with our own view. Notice also Russell’s casual and rather radical remark
in one of these excerpts that all symbols are vague. He elaborates on this point as
follows:
Let us consider the various ways in which common words are vague, and let us begin with such
a word as “red.” It is perfectly obvious, since colours form a continuum, that there are shades
of colour concerning which we shall be in doubt whether to call them red or not, not because
we are ignorant of the meaning of the word “red,” but because it is a word the extent of whose
application is essentially doubtful. . . . All words describing sensible qualities have the same kind
of vagueness which belongs to the word “red.” This vagueness exists also, though in a lesser de-
gree, in the quantitative words which science has tried hardest to make precise, such as a metre
or a second. . . . The metre, for example, is defined as the distance between two marks in a cer-
tain rod in Paris, when the rod is at a certain temperature. Now the marks are not points, but
patches of a finite size, so that the distance between them is not a precise conception. Moreover,
temperature cannot be measured with more than a certain degree of accuracy, and the tempera-
ture of the rod is never quite uniform. For all these reasons the conception of a metre is lacking
in precision. The same applies to a second. The second is defined by a relation to the rotation
of the earth, but the earth is not a rigid body, and two parts of the earth’s surface do not take
exactly the same time to rotate; moreover all observations have a margin of error. There are some
occurrences of which we can say that they take less than a second to happen, and others of which
we can say that they take more, but between the two there will be a number of occurrences of
which we believe that they do not last equally long, but none of which we can say whether they
last more or less than a second. Therefore, when we say an occurrence lasts a second, all that it is
worth while to mean is that no possible accuracy of observation will show whether it lasts more
or less than a second.
12 C. Hartshorne and P. Weiss, eds., Collected Papers of Charles Sanders Peirce, vol. 5 (Belkamp Press of Harvard

University, 1934), paragraph 5.512, 361.

section 2.1 Prehistory of fuzzy logic 15

This thought-provoking paper by Russell is now considered a classic in the literature

on vagueness, but it was Russell’s rather singular paper on the subject. After its
publication, strangely, Russell showed no interest in developing further his ideas
about vagueness.
Vagueness was discussed again by Cohen (1927) who spoke of concepts’ “twi-
light zones,” and, more thoroughly than ever before, in a paper by the American
philosopher Max Black (1909–1988). After publication of his paper (Black 1937),
which we examine in some detail in the next section, interest in vagueness began
to grow. We return to the various views and aspects of vagueness throughout this
book, in particular in section 5.4.3.
In addition to arguments based on vagueness of natural language, other argu-
ments against the principle of bivalence can be found in philosophical literature in
the late 19th century. One type of argument emerged from metaphysical views of
a major English philosopher, Francis H. Bradley (1846–1924); another type of argu-
ment, based on epistemological grounds, was put forward by the influential French
physicist, mathematician, and philosopher, Pierre Duhem (1861–1916).
Bradley developed his ideas primarily in his book Appearance and Reality (Brad-
ley 1893). This book, which is based on Bradley’s metaphysical view that true reality
is immune to self-contradiction, is divided into two parts. In the first part, which is
rather short, Bradley basically argues that any human experiential domain, regard-
less of how it is organized, is subject to unavoidable inconsistencies. This means
that, according to Bradley’s metaphysical view, it is only an appearance of reality,
but not true reality. In the much longer second part of the book, Bradley’s princi-
pal aim is to reconcile the inconsistent and fragmentary appearance of reality with
the consistent, harmonious, and all-inclusive true reality. Through a long sequence
of thoroughly discussed arguments, each including a response to potential criticism,
Bradley concludes that the only sensible way to reconcile reality with appearance of
reality is by employing degrees of truth, which means in turn rejecting the principle
of bivalence. His argumentation leading to this conclusion is so extensive that even
any meaningful summary of it would be long; instead, we include here only a few
representative short quotations from his book (Bradley 1893, 364–65, 396):

[To] be more or less true, and to be more or less real, is to be separated by an interval, smaller or
greater, from all-inclusiveness or self-consistency. Of two given appearances the one more wide,
or more harmonious, is more real. It approaches nearer to a single, all-containing, individuality.
To remedy its imperfection, in other words, we should have to make a smaller alteration. The
truth and the fact, which, to be converted into the Absolute, would require less re-arrangement
and addition, is more real and truer. And this is what we mean by degrees of reality and truth.
To possess more the character of reality, and to contain within oneself a greater amount of the
real, are two expressions for the same thing. . . . There cannot for metaphysics be, in short, any
hard and absolute distinction between truths and falsehoods. With each assertion the question
is, how much will be left of that assertion, if we suppose it to have been converted into ultimate
truth? . . .
A predicate, we may say, in no case is, as such, really true. All will be subject to addition, to
qualification and re-arrangement. And truth will be the degree up to which any predicate, when
made real, preserves its own character.
16 Prehistory, Emergence, and Evolution of Fuzzy Logic chapter 2

It is worth mentioning that Bradley’s metaphysical arguments for rejecting the

principle of bivalence were further advanced almost half a century later by the Amer-
ican philosopher, Brand Blanshard (1892–1987), in his large, two-volume book Blan-
shard 1939. In fact, almost half of the second volume is devoted to this issue. Blan-
shard is basically reformulating Bradley’s metaphysical ideas in the framework of
a coherence theory of truth. It suffices for our purposes to present here his summary
in paragraph 12 of chapter XXII on “Coherence and Degrees of Truth” (306–7):
We have been setting out the logical ground for the doctrine of degrees of truth. Is its connection
with the doctrine now clear? To make sure, let us briefly repeat: To think of any object whatever
is to think of it in the relation to what is beyond it. There are always some of these relations that
are so vital to the thing’s nature, and therefore to our concepts of its nature, that neither could
be what it is if cut off from them. Thus our concept can never be adequate till we have embraced
these in our thought. And since we never do grasp them all, our thought remains inadequate.
Not that as it stands it is worthless; to say that would be self-contradictory. It is plainly not
wholly adequate; it is equally plainly not wholly worthless. Its adequacy is a matter of degree.

Switching now to Duhem, we feel that he should be mentioned in this sec-

tion due to his original, and in many ways unorthodox, views about the role of
mathematics in science. These views, expressed primarily in Duhem 1906, envi-
sioned in some sense mathematics based on fuzzy logic—the subject of our chapter 5.
Duhem’s ideas are characterized in a nutshell through the following excerpts:13
Between the concrete facts, as the physicist observes them, and the numerical symbols by which
these facts are represented in the calculations of the theorist, there is an extremely great difference.
It is impossible to describe a practical fact without attenuating by the use of the word “ap-
proximately” or “nearly”; on the other hand, all the elements constituting the theoretical fact are
defined with rigorous exactness.
A practical fact is not translated therefore by a single theoretical fact but by a kind of bundle
including an infinity of different theoretical facts. Each of the mathematical elements brought
together in order to constitute one of these facts may vary from one fact to another; but the vari-
ation to which it is susceptible cannot exceed a certain limit, namely, the limit of error within
which the measurement of this element is blotted. The more perfect the methods of measure-
ment are, the closer is the approximation and the narrower the limits but they never become so
narrow that they vanish. . . .
Thus every physical law is an approximate law. Consequently, it cannot be, for the strict
logician, either true or false; any other law representing the same experiments with the same
approximation may lay as just a claim as the first to the title of a true law or, to speak more
precisely, of an acceptable law. . . .
A mathematical deduction is of no use to the physicist so long as it is limited to asserting
that a given rigorously true proposition has for its consequence the rigorous accuracy of some
such other proposition. To be useful to the physicists, it must still be proved that the second
proposition remains approximately exact when the first one is approximately true. And even
that does not suffice. The range of these two approximations must be delimited; it is necessary
to fix the limits of error which can be made in the result when the degree of precision of the
methods of measuring the data is known; it is necessary to define the probable error that can be
granted the data when we wish to know the result within a definite degree of approximation.
Such are the rigorous conditions that we are bound to impose on mathematical deduction
if we wish this absolutely precise language to be able to translate without betraying the physi-
cist’s idiom, for the terms of this latter idiom are and always will be vague and inexact like the
perceptions which they are to express. On these conditions, but only on these conditions, shall
we have a mathematical representation of the approximate.

13
From Duhem 1954 (pp. 133–35, 171–72, 143)—an English translation of the second edition of Duhem 1906.
Another random document with
no related content on Scribd:
The Project Gutenberg eBook of Chinese
pictures
This ebook is for the use of anyone anywhere in the United States
and most other parts of the world at no cost and with almost no
restrictions whatsoever. You may copy it, give it away or re-use it
under the terms of the Project Gutenberg License included with this
ebook or online at www.gutenberg.org. If you are not located in the
United States, you will have to check the laws of the country where
you are located before using this eBook.

Title: Chinese pictures

notes on photographs made in China

Author: Isabella L. Bird

Release date: October 27, 2023 [eBook #71968]

Language: English

Original publication: London: Cassell and Company, 1900

Credits: The Online Distributed Proofreading Team at

https://www.pgdp.net (This file was produced from images
generously made available by The Internet Archive)

*** START OF THE PROJECT GUTENBERG EBOOK CHINESE

PICTURES ***
the tb
 CHINESE PICTURES

Notes on Photographs
Made in China

BY
MRS. J. F. BISHOP
F.R.G.S., etc.

CASSELL AND COMPANY, Limited

London, Paris, New York & Melbourne
1900. All Rights Reserved
INTRODUCTION.
This little book is the outcome of talks with Mrs. Bishop over some of
the photographs which were taken by her in one or other of her
journeys into and across China. Some of the photographs have
already appeared in her published works, “The Yangtze Valley and
Beyond” and “Korea and Her Neighbourhood” (2 vols., Murray). The
notes were, in substance, dictated by Mrs. Bishop. It is hoped they
contain some real information on the people, their surroundings, and
habits which, though slight in form, may be helpful to a better
understanding of a very difficult problem.
According to our newspaper press to-day, the Chinese are simply
cruel barbarians. According to Mrs. Bishop, when you know them
they are a likeable people—and she has formed this opinion in spite
of the fact that, in their deeply-rooted hatred of the foreigners, they
twice attacked her with violence. A real understanding of the people
is for us, with our different modes of thought, most difficult to arrive
at; but we shall not advance towards it by accepting all the evil
reports and shutting our ears to the good ones. That the problem of
China is, and will for some time continue to be, the most interesting
question to the rest of the world is certain. The future of its people is
all unknown, but there are in it possibilities which make it a terror to
all other nations.
ERRATA.
The illustrations on p. 79 and p. 81 have been transposed. The
former represents “The Tablet of Confucius,” the latter, “The Altar of
Heaven.”

Chinese Pictures.

[Transcriber’s Note: this error has been corrected.]

CONTENTS.
PAGE
The Private Entrance to the Imperial Palace, 8
Peking

The Entrance to the British Legation 10

Entrance to the College of the Student 12

Interpreters

The State Carriage of the British Legation 14

The Great Imperial Stone Road from Peking to 16

Chengtu, the Capital of Sze Chuan

A Mule Cart 18

A Manchurian Family Travelling 20

Carriage by Bearers 22

A Traveller Arriving at an Inn in Manchuria 24

Carriage of Merchandise 26

The Mode of Carrying Oil and Wine 28

Wheelbarrow Traffic on the Chengtu Plain 30

The Wheelbarrow of North China 32

A Small Houseboat on the Yangtze Kiang 34

A Foot Boat Found in Central China 36

Hsin Tan Rapid on the Yangtze River 38

A Boat on the Min River, Used for Running the 40

Rapids

Part of a Fringe of Junks or River Boats at 42

Wan Hsien

The Bridge of Ten Thousand Ages, Foochow 44

A Bridge at Wan Hsien of the Single Arch Type 46

The Bridge of Mien Chuh Sze Chuan 48

A Simple Country Bridge 50

A Dragon Bridge 52

The Zig-zag Bridge of Shanghai 54

The Garden of the Guild of Benevolence, 56
Chung King

A Burial Charity 58

A Baby Tower, Foochow 60

Bottle Seller and Hospital Patient 62

The Dying Coolie 64

The Mode of Sepulchre throughout Southern 66

China

Coffins Kept Above Ground 68

The Temple of the God of Literature at Mukden 70

The Temple of the Fox, Mukden 72

Wayside Shrines 74

The Ficus Religiosa 76

The Altar of Heaven 78

The Tablet of Confucius 80

A Porcelain-fronted Temple on the Yangtze 82

Child Eating Rice with Chopsticks 84

Fort on the Peking Wall 86

Another Fort on the Wall of Peking 88

Colossal Astronomical Instruments on the 90

Peking Wall

Chien Mun Gate 92

The Gate of Victory, Mukden 94

The West Gate of Kialing Fu 96

The West Gate of Hangchow 98

The Gate of a Forbidden City 100

Silk Reeling 102

A Typical Entrance to a House 104

The Guest Hall in a Chinese House, Wan Hsien, 106

Sze Chuan

A Chinese Village 108

A Farmhouse in the Hakka Country, Southern 110

China

A Market Place or Market Street in Sze Chuan 112

The Cobbler 114

Carrying Liquid Manure to the Fields 116

The Marriage Chair 118

Mode of Carrying Cash and Babies 120

A Pai-fang, or Widow’s Arch 122

Two Soldiers of Sze Chuan 124

Opium Culture Encroaching on the Rice Lands, 126

Sze Chuan
THE PRIVATE ENTRANCE
TO THE
IMPERIAL PALACE, PEKING.

A subject of considerable interest, owing to the mystery

surrounding the members of the Imperial Family. The photograph
was taken from the wall of the Purple or Forbidden City, in which
only the Imperial Family and their entourage have the right to dwell.
The building in the centre, which is roofed with yellow tiles, is
supposed to be the residence of the Emperor, but where he does
actually reside remains a mystery. The entrance to the Palace is
through the arches in the building on the left.