Metric Structures and Fixed Point Theory 1St Edition Dhananjay Gopal Online Ebook Texxtbook Full Chapter PDF

Metric Structures and Fixed Point
Theory 1st Edition Dhananjay Gopal

Visit to download the full and correct content document:
https://ebookmeta.com/product/metric-structures-and-fixed-point-theory-1st-edition-dh
ananjay-gopal/
More products digital (pdf, epub, mobi) instant
download maybe you interests ...
Metric Fixed Point Theory: Applications in Science,

Engineering and Behavioural Sciences
https://ebookmeta.com/product/metric-fixed-point-theory-
applications-in-science-engineering-and-behavioural-sciences/
Fixed Point Theory and Fractional Calculus Recent

Advances and Applications Forum for Interdisciplinary
Mathematics Pradip Debnath (Editor)
https://ebookmeta.com/product/fixed-point-theory-and-fractional-
calculus-recent-advances-and-applications-forum-for-
interdisciplinary-mathematics-pradip-debnath-editor/
Optimization on Solution Sets of Common Fixed Point

Problems 1st Edition Alexander J. Zaslavski
https://ebookmeta.com/product/optimization-on-solution-sets-of-
common-fixed-point-problems-1st-edition-alexander-j-zaslavski/
FRP Composite Structures: Theory, Fundamentals, and

Design 1st Edition Gangarao
https://ebookmeta.com/product/frp-composite-structures-theory-
fundamentals-and-design-1st-edition-gangarao/
Digital Twin Technology 1st Edition Gopal Chaudhary
https://ebookmeta.com/product/digital-twin-technology-1st-
edition-gopal-chaudhary/
Multivariable Calculus Metric Edition Stewart
https://ebookmeta.com/product/multivariable-calculus-metric-
edition-stewart/
Calculus Metric Edition James Stewart
https://ebookmeta.com/product/calculus-metric-edition-james-
stewart/
Photovoltaic Thermal Passive House System 1st Edition

Gopal Nath Tiwari
https://ebookmeta.com/product/photovoltaic-thermal-passive-house-
system-1st-edition-gopal-nath-tiwari/
Breaking Point Book I Of Breaking Point Trilogy 1st

Edition John Rhodes
https://ebookmeta.com/product/breaking-point-book-i-of-breaking-
point-trilogy-1st-edition-john-rhodes/
Metric Structures and
Fixed Point Theory
Metric Structures and
Fixed Point Theory
Edited By
Dhananjay Gopal
Praveen Agarwal
Poom Kumam
First edition published 2021
by CRC Press
6000 Broken Sound Parkway NW, Suite 300
Boca Raton, FL 33487-2742
and by CRC Press
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
© 2021 selection and editorial matter, Dhananjay Gopal, Praveen Agarwal and Poom Kumam; individual
chapters, the contributors
CRC Press is an imprint of Taylor & Francis Group, LLC
The right of Dhananjay Gopal, Praveen Agarwal and Poom Kumam to be identified as the authors of the editorial
material, and of the authors for their individual chapters, has been asserted in accordance with sections 77 and
78 of the Copyright, Designs and Patents Act 1988.
Reasonable efforts have been made to publish reliable data and information, but the author and publisher cannot
assume responsibility for the validity of all materials or the consequences of their use. The authors and publishers
have attempted to trace the copyright holders of all material reproduced in this publication and apologize to
copyright holders if permission to publish in this form has not been obtained. If any copyright material has not
been acknowledged please write and let us know so we may rectify in any future reprint.
Except as permitted under U.S. Copyright Law, no part of this book may be reprinted, reproduced, transmitted, or
utilized in any form by any electronic, mechanical, or other means, now known or hereafter invented, including
photocopying, microfilming, and recording, or in any information storage or retrieval system, without written
permission from the publishers.
For permission to photocopy or use material electronically from this work, access www.copyright.com or contact
the Copyright Clearance Center, Inc. (CCC), 222 Rosewood Drive, Danvers, MA 01923, 978-750-8400. For works
that are not available on CCC please contact mpkbookspermissions@tandf.co.uk
Trademark notice: Product or corporate names may be trademarks or registered trademarks and are used only for
identification and explanation without intent to infringe.
Library of Congress Cataloging- in- Publication Data
A catalog record has been requested for this book
ISBN: 978-0-367-68914-8 (hbk)
ISBN: 978-1-003-13960-7 (ebk)
ISBN: 978-0-367-68917-9 (pbk)
Typeset in Palatino
by Newgen Publishing UK
Contents
Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . vii
Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix
Editors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xi
List of Contributors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii
1 Symmetric Spaces and Fixed Point Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

Pradip Ramesh Patle and Deepesh Kumar Patel
2 Fixed Point Theory in b-Metric Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

Nguyen Van Dung and Wutiphol Sintunavarat
3 Basics of w-Distance and Its Use in Various Types of Results . . . . . . . . . . . . . . . . . . . . . . 67

Dhananjay Gopal and Mohammad Hasan
4 G-Metric Spaces: From the Perspective of F-Contractions and Best

Proximity Points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 103
Vishal Joshi and Shilpi Jain
5 Fixed Point Theory in Probabilistic Metric Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

Juan Martínez-Moreno
6 Fixed Point Theory For Fuzzy Contractive Mappings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199

Dhananjay Gopal and Tatjana Došenović
7 Set-Valued Maps and Inclusion Problems in Modular Metric Spaces. . . . . . . . . . . . . 245

Poom Kumam
8 Graphical Metric Spaces and Fixed Point Theorems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267

Satish Shukla
9 Fixed Point Theory in Partial Metric Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283

Dhananjay Gopal and Shilpi Jain
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
v
Preface
One of the intuitively natural concept in the history of human beings is notion of dis-
tance. Though the notion was known but it was axiomatically formulated by Frechet in
early 1900’s. The realization of the fact that Euclidean distance between two points can
be given by the absolute difference, Frechet come up with a abstract formulation and
generalization of the distance concept (termed as metric). It is an indisputable argument
that the formulation of metric opens a new subject in mathematics called non-linear ana-
lysts after the appearance of Banach fixed point theorem. Because the underlined space
of this theorem is a metric space, the theory that developed following its publications is
known as the metric fixed point theory. It is well known that metric fixed point theory
provides essential tools for solving problems arising in various branches of mathematics
and other sciences such as split feasibility problems, variational inequality problems, non-
linear optimization problems, equilibrium problems, complementarity problems, selection
and matching problems, and problems of proving an existence of solution of integral and
differential equations are closely related with fixed point theory. Due to this reason over
the last seventy years many people have tried to generalize the definition of metric space
and corresponding fixed point theory. As a result, the notions like Hausdorff metric, sym-
metric, quasi-metric, semi-metric, probabilistic metric, fuzzy metric, b-metric, w-distances,
G-metric, D-metric, modular metric and so on have been appeared in the literature and this
trend is still going on. However, most of the materials on these topics are scattered either
in the form of research papers or general articles. Also numerous books are available on
fixed point theory but we felt that no book could be used systematically to follow on the
above topics. Considering these aspect, the aim of this monograph is to provide a system-
atic survey and latest updates on the most popular generalizations of metric spaces and
corresponding fixed point results which fall within the scope of metric fixed point theory.
This monograph is divided into nine chapters. Each chapter contributed by different
authors contains a section “Introduction” which summarizes the material needed to read
the chapter independent of others and contains a necessary background, several examples,
and comprehensive literature to comprehend the concepts presented therein. This could
be helpful for those who want to pursue their research career in metric fixed point theory
and its related areas.
Chapter One presents a brief study of fixed point theorems in symmetric (semimetric)
spaces. Taking into consideration the importance, we also discussed the fixed point of
multivalued mappings and common fixed point theorems of two mappings with various
contractive conditions. In addition, the result due to Suzuki which proves the character-
ization of semicompleteness by means of fixed point property is also presented which
provides quite self containment to this chapter. Finally, some open problems are listed for
future scope and study.
Chapter Two deals with comprehensive survey and motivation of Fixed Point Theory
in b-Metric Spaces. A short history of the b-metric space and its basic properties are pre-
sented in Section 1. The metrization and completion results of b-metric spaces are shown
in Section 2. Section 3 is devoted to show that many fixed point results in b-metric spaces
vii
viii Preface
have been proven similar to that in metric spaces. Multivalued fixed point results in b met-
ric spaces are presented in Section 4. Finally, Section 5 is devoted to present the relaxations
of constant contractions in b-metric spaces.
Chapter Three presents the basics of w-Distances and well known results like Caristi
fixed point theorem, Ekeland variational principle, the nonconvex minimization theorem
according to Takahashi, Ciric fixed point theorem in the setting of w-Distances.
Chapter Four deals with the study of those fixed point results in G-metrics paces that
cannot be obtained from the existing results in the setting of associated metric spaces. The
main motive to introduce such results was to emphasize the applicative approach of fixed
point theory.
Chapter Five presents concise study of fixed point results concerning various classes of
probabilistic contractions including altering distance functions and probabilistic nonlinear
contractions. We hope that the results presented in this chapter illustrate the direction
of research over the last five decades up to the most recent contributions for fixed point
theory of single valued mappings in PM spaces.
Chapter Six deals with a concise study of fixed point theorems for fuzzy contractive
type mappings in fuzzy metric spaces. The results presented in detail were selected to
illustrate the direction research in the field has taken from last four decades up to most
recent contribution in the subject.
Chapter Seven present recent developments of set-valued analysis based mainly on
the fixed point inclusion problems. This chapter is organized as follows: The succeeding
Section 2 is devoted to the mathematical analysis of fixed point inclusion problems and
also the common fixed point of several classes of set-valued maps. In the Section 3, we pro-
vide the applications of fixed point results presented in the previous section. We consider
here two main nonlinear problems- abstract economy and fractional integral inclusion.
Chapter Eight is mainly concern to study of an approach to fixed point theory via graph-
ical metric structure. In view of the fact that the class of contractions in graphical metric
spaces is larger than that in metric spaces, we see that the fixed point results in graph-
ical metric spaces can be applied on a larger class of existence problems which use the
fixed point methods, e.g., existence of solution of an integral equation. As well as, apart
from the topology of usual metric spaces, the topology generated by graphical metric
spaces are non-Hausdorff which can be considered as a useful tool in theories where non-
Hausdorffness occurs, e.g., in algebraic geometry, in representation theory, in the theory
of C?-algebras, in domain theory, in computer applications etc., this topology can be used
as a substitute of existing tools.
Chapter Nine is devoted to provide a systematic survey on the various aspects of partial
metric spaces and concerning fixed point theory. But still there are a lot of topological as
well as fixed point aspects which remain unknown about partial metric spaces.
Acknowlegments
We greatly admire and are deeply indebted to our friends and colleagues working in
the Fixed Point Theory for their encouragement and support, especially Professor Som-
pong Dhompongsa, Chiang Mai University, Thailand, Professor Yeol-Je Cho, Gyeongsang
National University, Jinju, Korea, Professor Ismat Beg, Centre for Mathematics and Statis-
tical Sciences, Lahore School of Economics, Lahore, Pakistan and Professor S. Radenovic,
Faculty of Mathematics and Statistics, Ton Duc Thang University, Vietnam. In particu-
lar, we wish to express our deepest thanks to our colleagues who contributed their recent
research work in the form of chapters for inclusion in this book.
Our deepest gratitude and thanks are also due to our family members who always
encouraged us and refreshed our energies with their sweet words while we were busy
in accomplishment of this project.
The editors are very thankful to Aastha Sharma, Shikha Garg and the staff at CRC Press
for their unfailing support cooperation and patience in publishing this book.
ix
Editors
Dhananjay Gopal has a doctorate in Mathematics from Guru

Ghasidas University, Bilaspur, India and is currently Associate
Professor of Mathematics in Guru Ghasidas Vishwavidyalaya
(A Central University), Bilaspur (C.G.), India. He is author
and co-author of several papers in journals, proceedings and a
monograph on Background and Recent Developments of Met-
ric Fixed Point Theory and a book on Introduction to metric
spaces. He is devoted to general research on the theory of
Nonlinear Analysis and Fuzzy Metric Fixed Point Theory.
Dr. Gopal has active research collaborations with KMUTT,
Bangkok, Thammasat University Bangkok, Jaen University
Spain and in his research pursuits he has visited South Africa,
Thailand, Japan and Iran.
Dr. P. Agarwal earned his Master’s degree in Mathematics

from Rajasthan University in 2000. In 2006, he earned his Ph.D.
(Mathematics) at the Malviya National Institute of Technology
(MNIT) in Jaipur, India, one of the highest ranking universities
in India.
Dr. Agarwal has been actively involved in research as
well as pedagogical activities for the last 20 years. His
major research interests include Special Functions, Fractional
Calculus, Numerical Analysis, Differential and Difference
Equations, Inequalities, and Fixed Point Theorems. He is an
excellent scholar, dedicated teacher, and prolific researcher. He
has published 7 research monographs and edited volumes and
more than 150 publications (with almost 100 mathematicians
all over the world) in prestigious national and international
mathematics journals. Dr. worked previously either as a regu-
lar faculty or as a visiting professor and scientist in universities
in several countries, including India, Germany, Turkey, South Korea, UK, Russia, Malaysia
and Thailand. He has held several positions including Visiting Professor, Visiting Scien-
tist, and Professor at various universities in different parts of the world. Specially, he was
awarded most respected International Centre for Mathematical Sciences (ICMS) Group
Research Fellowship to work with Prof. Dr. Michael Ruzhansky-Imperial College Lon-
don at ICMS Centre, Scotland, UK, and during 2017-18, he was awarded most respected
TUBITAK Visiting Scientist Fellowship to work with Prof. Dr. Onur at Ahi Evran Uni-
versity, Turkey. He has been awarded by Most Outstanding Researcher-2018 (Award for
contribution to Mathematics) by the Union Minister of Human Resource Development of
India, Mr. Prakash Javadekar in 2018. According to Google Scholar, Dr. Agarwal is cited
more than 2,553 times, and on Scopus his work is cited more than 1,191 times. Dr. Agarwal
is the recipient of several notable honors and awards.
Dr. Agarwal provided significant service to Anand International College of Engineer-
ing, Jaipur. Under his leadership during 2010-20, Anand-ICE consistently progressed in
xi
xii Editors
education and preparation of students, and in the new direction of academics, research
and development. His overall impact to the institute is considerable. Many scholars from
different nations, including China, Uzbekistan, Thailand and African Countries came to
work under his guidance. The majority of the visiting post-doctoral scholars were sent to
work under Dr. Agarwal by their employing institutions for at least one month.
Dr. Agarwal regularly disseminates his research at invited talks/colloquiums (over 25
Institutions all over the world).He has been invited to give plenary/keynote lectures
at international conferences in the USA, Russia, India, Turkey, China, Korea, Malaysia,
Thailand, Saudi Arabia, Germany, UK, Turkey, and Japan.
He has served over 40 Journals in the capacity of an Editor/Honorary Editor, or
Associate Editor, and published books as an editor. Dr. Agarwal has also organized
International Conferences/ workshops’/seminars/summer schools.
In summary of these few inadequate paragraphs, Dr. P. Agarwal is a visionary scientist,
educator, and administrator who have contributed to the world through his long service,
dedication, and tireless efforts.
Poom Kumam received his Ph.D. in Mathematics from

Department of Mathematics, Faculty of Science, Naresuan
University. Now, he is the Professor and Associate Dean for
Research and Networking, Faculty of Science, King Mongkut’s
University of Technology Thonburi (KMUTT) and also the
leading of Theoretical and Computational Science (TaCS) Cen-
ter and director of Computational and Applied Science for
Smart Innovation Cluster (CLASSIC). He served on the edi-
torial boards of various international journals and also pub-
lished more than 700 papers in Scopus and Web of Science
(WoS) database. Furthermore, his research interest focuses on
Fixed Point Theory and Optimization with related with opti-
mization problems in both pure science and applied science.
List of Contributors
Tatjana Došenović Juan Martnez-Moreno

Faculty of Technology, University of Novi Department of Mathematics, University of
Sad, Serbia Jaen, Campus las Lagunillas s/n, 23071,
Jaén, Spain Dosenovic
Nguyen Van Dung
Faculty of Mathematics Teacher Education, Deepesh Kumar Patel
Dong Thap University, Cao Lanh City, Department of Mathematics, Visvesvarya
Dong Thap, Province, Vietnam National Insitute of Technology, Nagpur,
India
Dhananjay Gopal
Department of Mathematics Guru Ghasidas Pradip Ramesh Patle
Vishwavidyalaya (A Central University), Department of Mathematics, Visvesvarya
Bilaspur (C.G.) India National Insitute of Technology, Nagpur,
India
Mohammad Hasan
Department of Mathematics, Jazan Satish Shukla
University, Jazan, Kingdom of Saudi Department of Applied Mathematics and
Arabia Humanities, Shri Vaishnav Institute of
Technology and Science Gram Baroli,
Shilpi Jain Indore, India
Department of Mathematics, Poornima
College of Engineering, Jaipur, India Wutiphol Sintunavarat
Department of Mathematics and Statistics,
Vishal Joshi Faculty of Science and Technology,
Department of Applied Mathematics Thammasat University (Rangsit Center),
Jabalpur Engineering College, Jabalpur,
12121 Pathumthani, Thailand
India
Poom Kumam
Fixed Point Theory and Applications
Research Group, Center of Excellence in
Theoretical and Computational Science
(TaCS-CoE), Science Laboratory Building,
Faculty of Science, King Mongkut’s
University of Technology Thonburi
(KMUTT)
xiii
1
Symmetric Spaces and Fixed Point Theory
Pradip Ramesh Patle and Deepesh Kumar Patel
CONTENTS
1.1 Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1
1.2 Basic Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2
1.3 Topology of Symmetric Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
1.4 Completeness Concepts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
1.5 Fixed Points of Single-Valued Mappings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9
1.6 Fixed Points of Multivalued Mappings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
1.7 Conclusion and Future Investigations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30
1.1 Introduction
One of the intuitively natural concepts in the history of human beings is the notion of dis-
tance. The notion was known but it was axiomatically formulated by Fréchet [19] in the
early 1900s. On realization of the fact that the Euclidean distance between two points can
be given by the absolute difference, Fréchet come up with an abstract formulation and
generalization of the distance concept (termed as metric). It is an indisputable argument
that the formulation of metric opens up a new way for analysts. A large number of gener-
alizations, improvements and extensions of the metric concept have appeared in different
directions due to its fundamental role in analytic sciences and applications. As a result,
notions such as Hausdorff metric, fuzzy metric, symmetric, quasi-metric, semi-metric,
metric-like, partial metric, b-metric, G-metric, D-metric, 2-metric, ultra-metric, dislocated
metric, modular metric, partial b-metric and so on have appeared in the literature and this
trend is still going on.
In an attempt to generalize the concept of metric spaces Fréchet [19], Menger [38], Chit-
tenden [13] and Wilson [52] introduced the notion of symmetric spaces and semi-metric
spaces. These spaces have been studied in great detail and have developed into a wide lit-
erature. Cicchese [15] introduced the notion of contraction mappings and proved the first
fixed point theorem in semi-metric spaces. Afterwards various fixed point theorems have
been extended in these spaces.
In this chapter, we restrict ourselves to the merging of two interesting notions; sym-
metric and semi-metric. We focus on the basic properties of symmetric and semi-metric
1
2 Pradip Ramesh Patle and Deepesh Kumar Patel
spaces. Also topology on symmetric spaces and various topological properties are pre-
sented. As our main concern is the fixed point theory in symmetric and semi-metric spaces,
we need to study the concept of completeness in these spaces. There are various complete-
ness concepts in these spaces because of different notions of Cauchy-type sequences. A
huge literature of fixed point theory in these spaces has developed that is impossible to
summarize in a chapter so we present some of the basic results in this chapter.
1.2 Basic Concepts

In this section, very basic concepts and definitions from literature of symmetric spaces are
presented. Let us begin with the following definition of symmetric.
Definition 1 ([52]). A function s : X × X → [0, +∞) satisfying
(W1) s(x, y) = 0 if and only if x = y and s(x, y) ≥ 0, for x, y ∈ X,

(W2) s(x, y) = s(y, x), for all x, y ∈ X,
is called symmetric on a nonempty set X, whereas the pair (X, s) is called symmetric space
(also called E-space of Fréchet).
If in addition (X, s) has the property s(x, y) ≤ s(x, z) + s(z, y), for x, y, z ∈ X, then (X, s) is
a metric space. This property is known as the triangular inequality. It is clear that every
metric is symmetric but the converse is not true.
Example 1. Let X = [0, 1] and s(x, y) = (x − y)2 . It is clear that (X, s) is a symmetric space but
not a metric space.
Example 2. Let X = { n1 } ∪ {0}. Let us define s : X × X → [0, ∞) as s(0, 1) = 1 = s(1, 0), s(1, n1 ) =
2
3
= s( n1 , 1) for n ≥ 2, s(1, 1) = 0 and s(x, y) = |x−y| for x, y ∈ X \{1}. Then (X, s) is a symmetric
space.
Example 3. Let X = { n1 } ∪ {0} and let

⎧
⎨ 1
if n is odd,
1 n
s ,0 =
n ⎩
1 if n is even.
and ⎧ 1 1

⎪ m−n
⎪ if m + n is even,
⎪⎪
⎨
1 1
s , = 1 − 1 if m + n is odd and |m − n| = 1,
m n ⎪
⎪
m n
⎪
⎪
⎩
1 if m + n is odd and |m − n| ≥ 2.
Then (x, s) is a symmetric space.
Symmetric Spaces and Fixed Point Theory 3
|m − n|
Example 4. Let X = N and s(x, y) = . Then (X, s) is a symmetric space.
2min{m,n}
Example 5. Let X = { n1 } ∪ {0} and let

⎧ 1 1

1 1
⎨ m − n if |m − n| ≥ 2,
s , =
m n ⎩
1 if |m − n| = 1,
s( n1 , 0) = 1 = s(1, n1 ) and s(x, x) = 0 for all x ∈ X. Then (X, s) is a symmetric space.
The following families of subsets of a nonempty set X are considered in this chapter:
N (X) = {P : P ⊆ X and P = 0};

/
C L s (X) = {P : P ∈ N (X) and P̄s = P};
Bs (X) = {P : P ∈ N (X) and P is bounded};
C B s (X) = {P : P ∈ C L s (X) ∩ Bs (X)} and
C (X) = {P : P ∈ N (X) and P is compact}.
We now recall the definition of distance between two sets. For A, B ∈ Bs (X), S and S +
are the distance functions defined as
S (A, B) = max sup s(a, B), sup s(b, A) (1.1)

a∈A b∈B
1
S + (A, B) = sup s(a, B) + sup s(b, A) , (1.2)
2 a∈A b∈B
where s(a, B) = infs(a, b). S is known as the Pompeiu–Hausdorff distance.

b∈B
Proposition 1. [41] (C B s (X), S ) is a symmetric space if (X, s) is a symmetric space.
Proposition 2. [44] (C B s (X), S + ) is a symmetric space if (X, s) is a symmetric space.
Remark 1. [43] S + and S are topologically equivalent, that is,
k1 S (P, Q) ≤ S + (P, Q) ≤ k2 S (P, Q),
where k1 = 12 and k2 = 1. It is worth mentioning here that the equivalence of two symmetric
spaces does not mean that the results proved with one are equivalent to the others. This is
shown by means of examples in [43] in the case of metric spaces.
Many properties and concepts in symmetric spaces and metric spaces are similar (but
not all, due to the absence of the triangular inequality). Let (X, s) be symmetric space. The
open ball with center x ∈ X and radius r > 0 is defined by B(x, r) = {y ∈ X : s(x, y) < r}. Also
if A is a subset of X, then diam(A) = sup{s(x, y) : x, y ∈ X} denotes the diameter of A. In a
symmetric space (X, s), the limit point of a sequence {xn } is defined by lim s(xn , x) = 0 if
n→∞
and only if lim xn = x.
n→∞
1.3 Topology of Symmetric Spaces

In every symmetric space (X, s) we can define the topology τs by means of a family of
closed sets as follows: a set A ⊆ X is closed if and only if for each x ∈ X, d(x, A) = 0 implies
x ∈ A, where d(x, A) = inf{d(x, a) : a ∈ A}. The convergence of a sequence {xn } in the topol-
ogy τs need not imply s(xn , x) → 0, although the converse holds. We can ensure this by
applying the following definition.
Definition 2. A topological space (X, τ ) is semimetrizable if there is a symmetric function

s : X × X → R such that τs = τ and that the mapping X ⊇ A → c(A) = {x ∈ X : d(x, A) = 0}
is a closure operator in τs . In terms of s it can be expressed by saying that the operator c is
idempotent.
In this case, the space (X, s) is called a semi-metric space. It is known that the name semi-
metric is used by many authors for the function that is different from metric in that it need
not satisfy s(x, y) = 0 implies x = y. Therefore it is recommended here not to get confused
with the terminology.
We now give a characterization of topology on symmetric space (X, s) as follows: Let s be
a symmetric on X. For x ∈ X and ε > 0, let B(x, ε ) = {y ∈ X : s(x, y) < ε }. A topology τs on X
is defined by U ∈ τs if and only if for each x ∈ U, there exists an ε > 0 such that B(x, ε ) ⊂ U.
A subset N of X is a neighbourhood of x ∈ X if there exists U ∈ τs such that x ∈ U ⊂ N.
Definition 3. A symmetric is called a semi-metric if for each x ∈ X and each r > 0, B(x, r) is
a neighbourhood of x in the topology τs .
The following two propositions are well known and they play a very important role in
characterizing semi-metric spaces.
Proposition 3. If (x, s) is a symmetric space, then the family {B(x, r) : r > 0} forms a local basis at
x. Also if s(xn , x) → 0 then xn → x (or lim xn = x) in the topology τs .
n→∞
It is worth mentioning here that this basis need not consist of open sets. Heath [23]
proved this, by constructing a semimetrizable space (X, τs ) such that for any s that
generates τs , there exists x ∈ X and r > 0 such that B(x, r) is not open.
Proposition 4. Every symmetric space (x, s) is a semi-metric space if and only if the following
conditions hold:
1. (X, τs ) is first countable;

2. For any sequence {xn } ⊆ X, s(xn , x) → 0 is equivalent to xn → x in the topology τs .
As we have seen, for symmetric spaces the triangular inequality is relaxed. In order
to combat various problems arising while proving results in such spaces some alternate
concepts need to be satisfied, which are listed below:
Definition 4. Let (X, s) be a symmetric space. We have the following properties:
(W3) [52] Given {xn }, x and y in X; lim s(xn , x) = 0 and lim s(xn , y) = 0 implies x = y.
n→∞ n→∞
(W4) [52] Given {xn }, {yn } and x in X; lim s(xn , x) = 0 and lim s(xn , yn ) = 0 implies
n→∞ n→∞
lim s(yn , y) = 0.
n→∞
(CC) [11] Given {xn } and x in X such that lim s(xn , x) = 0 implies lim s(xn , y) = s(x, y) for
n→∞ n→∞
some y ∈ X.
(HE) [3] Given {xn }, {yn } and x in X; lim s(xn , x) = 0 and lim s(yn , x) = 0 implies
n→∞ n→∞
lim s(xn , yn ) = 0.
n→∞
(W) [39] Given {xn }, {yn } and {zn } in X; lim s(xn , yn ) = 0 and lim s(yn , zn ) = 0 implies
n→∞ n→∞
lim s(xn , zn ) = 0.
n→∞
(JMS) [31] Given {xn }, {yn } and {zn } in X; lim s(xn , yn ) = 0 and lim s(yn , zn ) = 0 implies
n→∞ n→∞
lim s(xn , zn ) = ∞.
n→∞
(MT) [33] there exists K ≥ 1 such that for any x, y, z ∈ X, s(x, z) ≤ K[s(x, y) + s(y, z)].
(MC) [15] there exists ε : R+ → R+ and K ≥ 1 such that for any x, y, z ∈ X, s(x, z) ≤
ε (max{s(x, y), s(y, z)}) + K min{s(x, y), s(y, z)} and lim+ ε (t) = 0.
t→0
The following relations hold between the above properties. See [7, 14, 39] for more details.
(i) (W) =⇒ (W4) =⇒ (W3) but (W3) (W4).

(ii) (W) =⇒ (JMS), (W) =⇒ (HE) and (MT) =⇒ (MC).
(iii) (CC) =⇒ (W3) but (W3) (CC).
(iv) (MC) =⇒ (W3), (MC) =⇒ (W4), (MC) =⇒ (HE), (MC) =⇒ (W), (MC) =⇒ (JMS).
(v) The following pairs of properties are independent (in the sense that neither of which
implies the other): (W) and (CC); (W4) and (CC); (W3) and (HE); (W4) and (HE);
(HE) and (CC).
The following characterization of symmetric spaces with property (JMS) is given in [31].
Proposition 5. Let (X, s) be a symmetric space. Then the following conditions are equivalent.
1. (X, s) satisfies property (JMS).

2. There exist δ , ε > 0 such that for any x, y, z ∈ X,
s(x, z) + s(z, y) < δ implies that s(x, y) < ε .
3. There exists r > 0 such that sup{diam(B(x, r)) : x ∈ X} < ∞.
The next theorem gives a characterization of semi-metric space with open balls in terms
of the new semicontinuity property of s (see [7]).
Theorem 1. Let (X, s) be a symmetric space. Then the following are equivalent:
(i) lim s(xn , x) = 0 implies lim sup d(xn , y) ≤ d(x, y) . . . property (SC),
n→∞ n→∞
(ii) (X, s) is a semi-metric space in which all B(x, r) are open sets.
Remark 2. (i) It is obvious that (CC) implies (SC).

(ii) It is shown [7] that there exists a semi-metric space in which all balls B(x, r) (r > 0)
are open, that does not have properties (W4), (JMS), (HE) and (CC).
(iii) A semi-metric space in which all balls B(x, r) (r > 0) are open, is a T1 space, but it
need not be a Hausdorff space.
Proposition 6. Let (X, s) be a compact semi-metric space in which all balls B(x, r) (r > 0) are open
and K ⊆ X a nonempty compact set. Then K is bounded.
The continuity of a function is generalized with the following two notions.
Definition 5. A function f : X → X is s-continuous if lim s(xn , x) = 0 implies lim s(fxn , fx) = 0.

n→∞ n→∞
Definition 6. A function f : X → X is τs -continuous if lim xn = x with respect to τs implies

n→∞
lim f (xn ) = f (x) with respect to τs .
n→∞
Kirk and Shahzad in [35] introduced the following concept of triangular function.
Definition 7. Let (X, s) be a semi-metric space. Then X is said to be regular with a strong
triangle function if there exists a function Ψ : [0, ∞)2 → [0, ∞) satisfying the following:
(i) Ψ (0, 0) = 0.
(ii) Ψ is nondecreasing (i.e., Ψ (s, t) ≤ Ψ (s , t ) holds for all s, t, s , t ∈ [0, ∞) with s ≤ s and
t ≤ t ).
(iii) Ψ is continuous at (0, 0), that is, limΨ (sn , tn ) = 0 provided {sn } and {tn } are sequences
n→∞
in [0, ∞) converging to 0.
(iv) |s(x, y) − s(z, w)| ≤ Ψ s(x, z), s(y, w) .
Suzuki [49] characterized regular semi-metric space with a strong triangular function in
the following lemma.
Lemma 1. Let (X, s) be a semi-metric space. Then the following are equivalent:
(a) X is regular with a strong triangle function Ψ .

(b) There exists a function ψ : [0, ∞) → [0, ∞) satisfying
(i) ψ (0) = 0.
(ii) ψ is nondecreasing.
(iii) ψ is continuous at 0.
(iv) s(x, z) ≤ s(x, y) + ψ o s(y, z).
A lot of analysis of regular semi-metric space has been done. A large number of articles
have been published in fixed point theory of semi-metric space using strong triangular
function. For more detailed study, interested readers can refer to the recent article of Dung
and Hang [17] and references therein.
1.4 Completeness Concepts

We now shift our focus to the study of Cauchy sequences and completeness in symmetric
and semi-metric spaces. As triangular inequality is skipped in such spaces, various ideas
of Cauchy sequences and completeness have been presented. The key references for this
section are [7, 12, 20, 35, 50].
Definition 8. Let (X, s) be a symmetric space. A sequence {xn } in X is called an s-Cauchy

sequence if for given ε > 0 there is n0 ∈ N such that s(xm , xn ) < ε , for all m, n > n0 .
Definition 9. A symmetric space (X, s) is said to be s-Cauchy complete if every s-Cauchy

sequence converges to some x ∈ X in τs .
Definition 10. A symmetric space (X, s) is called S-complete if for every s-Cauchy
sequence {xn }, there exists x ∈ X with lim s(xn , x) = 0.
n→∞
Definition 11. [24] Let (X, τ ) be a topological space and s : X × X → [0, ∞) with s(x, y) = 0
∞
if and only if x = y. Then (X, τ ) is s-complete topological space if ∑ s(xn , xn+1 ) < ∞ implies
n=1
there exists x ∈ X with lim xn = x with respect to τ (s).
n→∞
The idea of completeness in s-complete topological spaces generalizes the completeness

in metric and quasi-metric spaces. Jachymski et al. [31] proved that an s-Cauchy complete
semi-metric space (X, τ (s)) which satisfies (W4) is an s-complete topological space. For
∞
(X, s) a semi-metric space, this means that if ∑ s(xn , xn+1 ) < ∞, there exists x ∈ X such that
n=1
xn → x in τ (s).
∞
Definition 12 ([24]). A symmetric space (X, s) is complete if ∑ s(xn , xn+1 ) < ∞ implies that
n=1
there exists x ∈ X such that lim s(xn , x) = 0.
n→∞
Recall here that, if we let s generate a topology τs , then the topological space (X, τs ) is
called a symmetric space, whereas a semi-metric space means a symmetric space in which
all open balls are neighbourhoods.
Remark 3 ([24]). (1) In a semi-metric space, lim s(xn , x) = 0 if and only if {xn } converges to x
n→∞
in τs .
(2) In a symmetric space, {xn } converges to x in τs , implies lim s(xn , x) = 0; the converse is
n→∞
not true.
The concept of s-weakly completeness of symmetric spaces is defined by Galvin and

Shore in [20].
Definition 13. A symmetric space (X, s) is said to be s-weakly complete if every decreasing
sequence {Fn } of nonempty closed subsets, such that there exists a sequence {xn }, xn ∈ Fn
with Fn ⊆ B(xn , 2−n ) has a nonempty intersection.
Proposition 7. Let (X, s) be a semi-metric space. Then following are equivalent:
(1) (X, τs ) is s-weakly complete,

(2) every s-Cauchy sequence in X has a convergent subsequence,

(3) every decreasing sequence {Fn } of nonempty closed subsets of X such that diam(Fn ) ≤
2−n for each n has a nonempty intersection.
Lemma 2 ([24]). Suppose T : X → Bs (X). Then lim s(xn , Tx) = 0 implies and is implied by there
n→∞
exists yn ∈ Tx satisfying lim s(xn , yn ) = 0.
n→∞
Definition 14 ([50]). Let {xn } be a sequence in a semi-metric space (X, s) and x ∈ X. Let
κ ∈ N and h : X → R be a function.
(A) {xn } is said to converge to x if lim s(xn , x) = 0.

n→∞
(B) {xn } is said to be Cauchy if lim sup{s(xn , xm ) : m > n} = 0.
n→∞
∞
(C) {xn } is said to be ∑-Cauchy if ∑ s(xn , xn+1 ) < ∞.
n=1
∞
(D) {xn } is said to be (∑, =)-Cauchy if xn (n ∈ N) are all different and ∑ s(xn , xn+1 ) < ∞.
n=1
(E) X is said to be Hausdorff if lim s(xn , x) = 0 and lim s(xn , y) = 0 imply x = y.
n→∞ n→∞
(F) X is said to be κ -Hausdorff if
lim S(x, un(1) , . . . , un(κ ) , y) = 0 (1.3)
n→∞
implies x = y, where
S(x, un(1) , . . . , un(κ ) , y) = s(x, un(1) ) + s(un(1) , un(2) ) + . . . + s(un(κ −1) , un(κ ) ) + s(un(κ ) , y). (1.4)
(G) X is said to be complete if every Cauchy sequence converges.
(H) X is said to be ∑-complete if every ∑-Cauchy sequence converges.
(I) X is said to be (∑, =)-complete if every (∑, =)-Cauchy sequence converges.
(J) X is said to be semicomplete if every Cauchy sequence has a convergent subse-
quence.
(K) X is said to be ∑-semicomplete if every ∑-Cauchy sequence has a convergent
subsequence.
(L) X is said to be (∑, =)-semicomplete if every (∑, =)-Cauchy sequence has a
convergent subsequence.
(M) s is said to be sequentially lower semicontinuous if s(x, y) ≤ lim inf s(xn , yn ) provided
n→∞
{xn } converges to x and {yn } converges to y.
(N) h is said to be sequentially lower semicontinuous if h(x) ≤ lim inf h(xn ) provided {xn }
n→∞
converges to x.
(N) h is said to be sequentially lower semicontinuous from above if h(x) ≤ lim inf h(xn )
n→∞
provided {xn } converges to x and {h(xn )} is strictly decreasing.
(O) h is said to be proper if {x ∈ X : h(x) ∈ R} is nonempty.
Proposition 8. Let (X, s) be a semi-metric space. Consider the following statements.
(i) X is ∑-complete.
(ii) X is (∑, =)-complete.
(iii) X is ∑-semicomplete.
(iv) X is (∑, =)-semicomplete.

(v) X is complete.
(vi) X is semicomplete.
Then the following implications hold:
(a) (i) =⇒ (ii) =⇒ (iii) =⇒ (iv) =⇒ (vi).

(b) (i) =⇒ (v) =⇒ (vi).
Proposition 9. A ∑-complete semi-metric space (X, s) is κ -Hausdorff for any κ ∈ N.
Proposition 10. Let (X, s) be a ∑-complete semi-metric space. Then X is Hausdorff.
Proposition 11. Let (X, s) be a (∑, =)-complete, Hausdorff semi-metric space. Then X is complete.
For proofs of the Propositions 8–11 refer to [50]. It is not confirmed whether the
Proposition 11 holds without Hausdorffness.
1.5 Fixed Points of Single-Valued Mappings

This section presents extensions and generalizations of some fundamental metric fixed
point theorems to a symmetric setting. In 1976, Cicchese [15] proved the first fixed point
theorem for contraction mappings in semi-metric spaces. To date, a lot of fixed point the-
orems have been extended and generalized in the setting of symmetric and semi-metric
spaces. We discuss some of the results from the ocean called the theory of fixed points in
symmetric spaces. We recall that x is a fixed point of a mapping T if x = Tx is satisfied.
Proposition 12. Let (X, s) be a Hausdorff semi-metric and s-Cauchy complete space and let T be a
self mapping on X satisfying the following condition
s(Tx, Ty) ≤ h s(x, y), for h ∈ (0, 1) and x, y ∈ X. (1.5)
If (X, s) is bounded, that is M = sup{d(x, y) : x, y ∈ X} < ∞, then T has a unique fixed point p and
for any x ∈ X, the sequence {Tn x} of iterates of T on x converges to p.
Proof. Let us fix an x ∈ X. Since
s(Tn x, Tn+m x) ≤ hn s(x, Tm x) ≤ hn M, for all n, m ∈ N.
As hn M → 0 with n → ∞, {Tn x} is s-Cauchy. The s-Cauchy completeness of X gives rise to

existence of p in X such that {Tn x} τ -converges to p. Since s is semi-metric, (1.5) implies
that T is τ -continuous. Therefore, {Tn+1 x} τ -converges to Tp. Since (X, s) is Hausdorff, we
may infer that p = Tp. Uniqueness of the fixed point is guaranteed using (1.5).

The boundedness of (X, s) is necessary for Proposition 12 to hold and it cannot be

extended to unbounded semi-metric spaces. This is evident from the following example.
|x − y|
Example 6. Let X = N, T : X → X defined by Tx = x + 1, x ∈ X and s(x, y) = , for
2min{x,y}
x, y ∈ X. Then clearly (X, s) is semi-metric space. Let {xn } be an s-Cauchy sequence. Then
{xn } is bounded; otherwise, there is a subsequence {xnj }, such that xnj → ∞, and then for
any j ∈ N,
|xnj − xnk |
lim s(xnj , xnk ) = lim = ∞,
k→∞ k→∞ 2xnk
violating the Cauchy condition. Therefore, we may infer that {xn } is constant for suffi-
ciently large n, since it is s-Cauchy complete, but T has no fixed point though it satisfies
(1.5) with h = 12 .
We now consider a class of functions, sometimes termed the auxiliary functions which
are used to obtain a large number of generalized forms of contractive conditions.
Definition 15. Let Φ denote the set of all monotone nondecreasing functions ϕ : [0, ∞) →
[0, ∞) such that lim ϕ n (t) = 0 for any t > 0, where ϕ n is the nth iterate of ϕ .
n→∞
Lemma 3. If ϕ ∈ Φ then ϕ (t) < t for all t > 0 and ϕ (0) = 0.
The following result is proved by Jachymski et al. [31].
Proposition 13. Let (X, s) be a Hausdorff s-Cauchy complete semi-metric space which satisfies the
property (JMS). Let T be a self mapping on X and ϕ ∈ Φ satisfying
s(Tx, Ty) ≤ ϕ (s(x, y)), for all x, y ∈ X;
then T has a unique fixed point p ∈ X and Tn x → p, for all x ∈ X.
Proposition 13 cannot be applied on metric type spaces because they are not necessarily
semi-metric space. Thus it is quite important to generalize this proposition in symmetric
spaces. Arandelovic and Keckic [7] obtained its symmetric space version. In order to prove
the theorem we need the following lemma.
Lemma 4. Let X be a nonempty set, T be a self mapping on X and n be a fixed positive integer such
that the iterate Tn has a unique fixed point x∗ . Then,
(1) x∗ is a unique fixed point of T.

(2) If X is a topological space and any sequence {Tn x} of Picard iterates converges to x∗ , then the
sequence of Picard iterates defined by T always converges to x∗ .
Lemma 5. Let (X, s) be a complete symmetric space satisfying the properties (W3) and (JMS). Let
T be a self mapping on X and ϕ ∈ Φ and let δ , ε be defined as in (2) of Proposition 5, satisfying
s(Tx, Ty) ≤ ϕ (s(x, y)), for all x, y ∈ X and ϕ (ε ) ≤ δ /2;
then T has a unique fixed point p ∈ X and Tn x → p, for all x ∈ X.
Proof. From Lemma 3, for any a, b ∈ X we have s(Tp, Tq) ≤ ϕ (s(p, q)) ≤ s(p, q), which implies
T is continuous.
Let us fix an x ∈ X. Then s(Tm+n x, Tn x) ≤ ϕ n (s(Tm x, x)) for any m, n ∈ N. Taking m = 1 we

get s(Tn+1 x, Tn x) ≤ ϕ n (s(Tx, x)) which implies that s(Tn+1 x, Tn x) → 0. This means there exists
k ∈ N such that s(Tk+1 x, Tk x) ≤ min{δ /2, ε }. We wish to show that for all n ∈ N,
s(Tk+n x, Tk x) ≤ ε . (1.6)
It is clear from the definition of k that (1.6) is valid for n = 1. Let us assume that (1.6) is
satisfied for some n ∈ N. From
s(Tk+1 x, Tk x) ≤ δ /2
and
s(Tk+1 x, Tk+n+1 x) ≤ ϕ (s(Tk x, Tk+n x)) ≤ ϕ (ε ) ≤ δ /2,
it follows that
s(Tk x, Tk+1 x) + s(Tk+1 x, Tk+n+1 x) ≤ δ .
Thus Proposition 5 yields
s(Tk x, Tk+n+1 x) ≤ ε .
By induction, we get that (1.6) is satisfied for any n ≥ 1. Thus
d(Tk+n x, Tk+n+m x) ≤ ϕ n (ε ), for any m, n ∈ N.
Hence {Tn x} is a Cauchy sequence. Then, there exists p ∈ X such that lim Tn x = p. Since T
n→∞
is continuous, we have lim Tn+1 x = Tp. By virtue of (W3), we get Tp = p.
n→∞
Since s(Tn x, p) → 0, by Proposition 3, we have Tn x → p in the topology τs . To verify
uniqueness, consider p∗ such that Tp∗ = p∗ , then for all n we have
s(p, p∗ ) = s(Tn p, Tn p∗ ) ≤ ϕ n (s(p, p∗ )) → 0, as n → ∞.
Thus p = p∗ .

Theorem 2. Let (X, s) be a complete symmetric space satisfying (W3) and (JMS). Let T be a self
mapping on X and ϕ ∈ Φ satisfying
s(Tx, Ty) ≤ ϕ (s(x, y)), for any x, y ∈ X;
then T has a unique fixed point p ∈ X and the sequence of Picard iterates {Tn x} at x converges in
X, and by Proposition 3, it converges to the same limit in the topology τs .
Proof. Let δ , ε be defined as in (2) of Proposition 5. If ϕ (ε ) ≤ δ /2, then from Lemma 5, it

follows that T has a unique fixed point y ∈ X and for each x ∈ X the sequence of Picard
iterates defined by T at x converges, in the topology τs to y. Now assume that ϕ (ε ) > δ /2.
Then there exists a least positive integer j > 1 such that ϕ j (ε ) ≤ δ /2. Also, we have that
s(Tj x, Tj y) ≤ ϕ j (s(x, y)),
for any x, y ∈ X, ϕ j ∈ Φ . By Lemma 5, Tj has a unique fixed point, say z ∈ X and for each
x ∈ X the sequence of Picard iterates defined by Tj at x converges in the topology τs , to
z. From Lemma 4 it follows that T has a unique fixed point z ∈ X, and for each x ∈ X the
sequence of Picard iterates defined by T at x converges in the topology τs , to z.

The next result due to [6] generalizes the above theorem. This theorem has a quite
different proof technique from the previous one.
Theorem 3. Let (X, s) be an s-Cauchy complete symmetric space satisfying (W3) and (JMS). Let
T : X → X be a τs -continuous map and ϕ ∈ Φ satisfying
s(Tx, Ty) ≤ ϕ (s3 (x, y)),
where s3 (x, y) = max{s(x, y), s(x, Tx), s(y, Ty)} for any x, y ∈ X. Then T has a unique fixed point
p ∈ X and for each x ∈ X, the sequence of Picard iterates of T at x converges to p in the topology τs .
Proof. Define s∗ : X × X → [0, ∞) as follows: s∗ (x, y) = 0 for x = y and s∗ (x, y) = s3 (x, y) oth-
erwise. Then the space (X, s∗ ) is a symmetric space. Also, we have s(x, y) ≤ s∗ (x, y) for
any x, y ∈ X. So, if {xn } ⊆ X is an arbitrary s∗ -Cauchy sequence in (X, s∗ ), then {xn } is an
s-Cauchy sequence in (X, s). Let x, y ∈ X. From
s(T2 x, Tx) ≤ ϕ (s(Tx, x)),

s(T2 y, Ty) ≤ ϕ (s(Ty, y)), (1.7)
s(Tx, Ty) ≤ ϕ (s3 (x, y)),
it follows that
s∗ (Tx, Ty) ≤ ϕ (s∗ (x, y)). (1.8)
Let δ , ε be defined as in (2) of Proposition 5. Then there exists the least positive integer j ≥ 1
such that ϕ j (ε ) ≤ δ /2. Let g = Tj . We have that T is continuous (in τs ). Then
s∗ (gx, gy) = s∗ (T(Tj−1 x), T(Tj−1 y))

≤ ϕ (s∗ (Tj−1 x, Tj−1 y)) (1.9)
∗
≤ ϕ (s (x, y)).
j
Let x ∈ X and ψ = ϕ j . Then ψ ∈ Φ and
s∗ (gm+n x, gn x) ≤ ψ n (s∗ (gm x, x)) for any m, n ∈ N. (1.10)
Therefore
s∗ (gn+1 x, gn x) ≤ ψ n (s∗ (gx, x)) (1.11)
∗ ∗
which implies that s (g x, g x) → 0. That means there exists k ∈ N such that s (g
n+1 n k+1
x, gk x) ≤
min{δ /2, ε }.
We wish to show that for all n ∈ N,
s∗ (gk+n x, gk x) ≤ ε . (1.12)
It is clear from the definition of k that (1.12) is valid for n = 1. Let us assume that (1.12) is
satisfied for some n ∈ N. From
s∗ (gk+1 x, gk x) ≤ δ /2
and
s∗ (gk+1 x, gk+n+1 x) ≤ ψ (s∗ (gk x, gk+n x)) ≤ ϕ (ε ) ≤ δ /2,
it follows that
s∗ (gk x, gk+1 x) + s∗ (gk+1 x, gk+n+1 x) ≤ δ .
Thus Proposition 5 yields
s∗ (gk x, gk+n+1 x) ≤ ε .
By induction, we get that (1.12) is satisfied for any n ≥ 1. Thus
s∗ (gk+n x, gk+n+m x) ≤ ψ n (ε ), for any m, n ∈ N.
Hence {gn x} is an s∗ -Cauchy sequence in (X, s∗ ), which implies that {gn x} is an s-Cauchy
sequence in (X, s). Then there exists z ∈ X such that lim gn x = z. Since g is τs -continuous, we
n→∞
have lim gn+1 x = gz. By virtue of (W3), we get gz = z.
n→∞
If y is another fixed point of g, then for all n we have
s∗ (y, z) = s∗ (gn y, gn z) ≤ ϕ n (s(y, z)) → 0, as n → ∞.
Therefore, z is a unique fixed point of g. By Lemma 4, z is a unique fixed point of T.

From
s∗ (z, gn+1 x) ≤ ϕ (max{s(z, gz), s(z, gn x), s(gn x, gn+1 x)}),
it follows that for each x ∈ X, the sequence of Picard iterates defined by g = Tj at x con-
verges, in the topology τs∗ , to z, which implies their convergence in the topology τs .
Therefore, by Lemma 4, we obtain that for each x ∈ X the sequence of Picard iterates
defined by T at x converges, in the topology τs , to z.

Remark 4. The continuity of T in Theorem 3 is necessary and cannot be omitted; this is

shown by means of the following example.
Example 7. Let X = {0} ∪ { n1 : n ∈ N} and let s be defined as follows: s(0, 1) = s(1, 0) = 1;

s(1, n1 ) = s( n1 , 1) = 23 for n ≥ 2; s(1, 1) = 0; otherwise s(x, y) = |x − y|. Then clearly (X, s) is a
bounded s-Cauchy complete symmetric space satisfying (W3) and (JMS). Let T : X → X be
defined by ⎧ x
⎨ 4 if x = 0,
T(x) =
⎩
1 if x = 0.
Then T satisfies s(Tx, Ty) ≤ ϕ (s3 (x, y)) for all x, y ∈ X and ϕ (t) = 23 t. But T does not have any
fixed point in X. Note that T is not continuous.
We now present some fixed point results in semi-metric spaces. Let O(x) denote the orbit
of T at a point x which is nothing but a sequence {xn } defined by xn = Tn x. Let O(x, y) =
O(x) ∪ O(y).
Theorem 4. Let (X, s) be a bounded s-Cauchy complete semi-metric space satisfying (W4). Suppose
T : X → X satisfies that for x ∈ X, there exists ν (x) ∈ N such that, for any ν ≥ ν (x) and y ∈ X,
s(Tν x, Tν y) ≤ ϕ (diam(O(x, y))) (1.13)
with ϕ ∈ Φ . Then there exists z ∈ X such that lim Tn x = z in the topology τs .

n→∞
We skip the proof.
Corollary 1. If in addition to the hypothesis of Theorem 4, we assume that T is τs -continuous then

T has a fixed point.
Proof. Since lim Tn x = z in the topology τs , by τs -continuity of T, lim Tn+1 x = Tz in the topol-
n→∞ n→∞
ogy τs . Therefore, using the fact (W4) implies (W3) and (X, s) satisfies (W4), we have Tz = z.
Hence z ∈ X is a fixed point.

Theorem 5. Let (X, s) be a bounded s-Cauchy complete semi-metric space with (W4), (CC) and
(JMS). Suppose T is a self mapping on X satisfying
s(Tx, Ty) ≤ ϕ (diam(O(x, y))) for x, y ∈ X. (1.14)
Then T has a unique fixed point and lim Tn x = z in the topology τs , for every x ∈ X.
n→∞
Proof. Following the conclusion of Theorem 4 we have that there exists a z ∈ X such that
lim s(Tn x, z) = 0 for all x ∈ X. Next assume that z = Tz; that means diam(O(z)) = β > 0. Thus
n→∞
it is possible to choose two sequences {i(n)} and {j(n)} such that
lim s(Ti(n) z, Tj(n) z) = β .

n→∞
Therefore one can choose δ > 0 with a corresponding ε > 0, such that ε ≤ β /2. Since there
exists n0 ∈ N such that
s(Tn z, z) ≤ δ /2, s(Tm z, z) ≤ δ /2, for every n, m ≥ n0 ,
and therefore s(Tn z, z) + s(Tm z, z) ≤ δ , by (2) of Proposition 5 we get
s(Tn z, Tm z) ≤ ε ≤ β /2, for every n, m ≥ n0 .
Hence i(n) = i for all infinitely many n with 0 ≤ i ≤ n0 . Thus there exists a sequence {r(n)} ⊆
{j(n)} such that lim s(Ti z, Tr(n) z) = β . Therefore either r(n) = j for infinitely many n or there
n→∞
exists a sequence {t(n)} ⊆ {r(n)} with r(n) → ∞ as n → ∞ which implies s(Ti z, z) = β . In
both cases, we can conclude that there exist i, j ≥ 0 such that s(Ti z, Tj z) = β .
If s(Tj z, z) = β , since lim s(Tn x, z) = 0 and by using (CC), we get
n→∞
β = d(Tj z, z) = lim s(Tn z, Tj z)

n→∞

≤ lim ϕ diam O(Tn−1 z, Tj−1 z)
n→∞

≤ lim ϕ diam O(z) = ϕ (β ),
n→∞
which is a contradiction with ϕ (β ) < β , for β > 0. On the other hand, if i, j ≥ 1, by (1.14),
we have
β = d(Ti z, Tj z) ≤ ϕ diam O(Ti−1 z, Tj−1 z) ≤ ϕ (β ),
which is again a contradiction. Hence β = 0; that means z is a fixed point of T.

Theorem 6. Let (X, s) be a bounded s-Cauchy complete semi-metric space with (W) and (CC).
Suppose T is a self mapping on X satisfying (1.14); then T has a unique fixed point and lim Tn x = z
n→∞
in the topology τs for every x ∈ X.
The following result is the extension of Kirk–Saliga [36] type fixed point theorem in ∑-
semicomplete semi-metric spaces. This result is obtained by Suzuki in [49] (see also [50]).
For an arbitrary set A, car(A) denotes its cardinal number. Let α be an ordinal number. We
denote by α + and α − the successor and the predecessor of α , respectively. α is said to be
isolated if α − exists. On the other hand, α is said to be a limit if α = 0 holds and α − does
not exist. For κ ∈ N we define α + κ by
α + κ = α +...+(κ times) .
Theorem 7. Let (X, s) be a ∑-semicomplete semimetric space and let T be a mapping on X. Let h :
X → (−∞, +∞] be a function which is proper and bounded from below. Assume that h is sequentially
lower semicontinuous from above in the sense of Definition 14. Assume also that there exists κ ∈ N
satisfying the following:
(i) h(Tx) ≤ h(x) for all x ∈ X.

(ii) h(Tκ x) + s(x, Tx) ≤ h(x) for all x ∈ X.
Then T has a fixed point.
Proof. Let us define a function H from X into (−∞, +∞] by

κ −1
H(x) = ∑ h(Tj x),
j=0
where T0 is the identity mapping on X. Then from (ii) we have

H(Tx) + s(x, Tx) ≤ H(x), for x ∈ X. (1.15)
Arguing by contradiction, we assume Tx = x for any x ∈ X. Let Ω be the first uncountable
ordinal number. Using transfinite induction, we define a net {uα : α ∈ Ω } satisfying the
following:
(P1 : α ) h(uα ) ≤ h(uβ ) and H(uα ) < H(uβ ) for any β ∈ Ω with β < α .
(P2 : α ) h(uα ) < h(uβ ) for any β ∈ Ω with β + κ ≤ α .
(P3 : α ) For any ε > 0 and for any β ∈ Ω with β < α , there exists a finite sequence
(γ0 , . . . , γn ) ∈ Ω n+1 satisfying
β = γ 0 < γ 1 < . . . < γn = α ,
n−1
H(uα ) + ∑ s(uγj , uγj+1 ) < H(uβ ) + ε . (1.16)
j=0
Fix u ∈ X with h(u) < ∞. It follows from (i) that H(u) < ∞ holds. Put u0 = u. Then (P1 : 0)–
(P3 : 0) hold obviously. Fix α ∈ Ω with 0 < α and assume that (P1 : β )–(P3 : β ) hold for β < α .
We consider the following two cases:
(a) α is isolated.
(b) α is a limit.
Put uα = Tuα − in the first case. For any β < α , since β ≤ α − and uα − = uα hold, we have by
(P1 : α − ), (i) and (1.15)
h(uα ) ≤ h(uα − ) ≤ h(uβ ),
H(uα ) ≤ H(uα ) + s(uα − , uα ) ≤ H(uα − ) ≤ H(uβ ). (1.17)
This means we have shown (P1 : α ). For β ∈ Ω with β + κ ≤ α , we have by (P1 : α ) and (ii)
h(uα ) ≤ h(uβ +κ ) = h(Tκ uβ ),
≤ h(Tκ uβ ) + s(uβ , Tuβ ) ≤ h(uβ ). (1.18)
This means we have shown (P2 : α ). Fix ε > 0 and β ∈ Ω with β < α . In the case where
β = α − , putting γ0 = β and γ1 = α , we have by (1.15)
n−1
H(uα ) + ∑ s(uγj , uγj+1 ) = H(Tuβ ) + s(uβ , Tuβ )
j=0
< H(uβ ) < H(uβ ) + ε . (1.19)

In the other case, where β < α − , from (P3 : α − ), there exists a finite sequence (γ0 , . . . , γn ) ∈
Ω n+1 satisfying
β = γ 0 < γ 1 < · · · < γn = α − ,
n−1
H(uα − ) + ∑ s(uγj , uγj+1 ) < H(uβ ) + ε . (1.20)
j=0
Putting γn+1 = α , we have by (1.15)

n−1 n−1
H(uα ) + ∑ s(uγj , uγj+1 ) ≤ H(uα − ) + ∑ s(uγj , uγj+1 ) < H(uβ ) + ε . (1.21)
j=0 j=0
Thus (P3 : α ) is proved. Therefore uα satisfying (P1 : α )–(P3 : α ) is defined for the first case.
In the second case of α being a limit, let us take {βn } an increasing sequence in Ω con-
verging to α ; that is, the following hold: βn < α for n ∈ N and for any β < α , there exists
n ∈ N satisfying β < βn . For any n ∈ N, by (P3 : βn+1 ), we can choose a finite sequence
(n) (n)
(γ0 , . . . , γνn ) ∈ Ω νn +1 satisfying
βn = γ0(n) < γ1(n) < · · · < γν(n)

n
= βn+1 ,
νn −1
∑ s(uγ (n)
j
, uγ (n) ) < H(uβn ) − H(uβn+1 ) + 2−n .
j+1
(1.22)
j=0
Since h is bounded from below, H is also bounded from below. Therefore we have
∞ νn −1
∑ ∑ s(uγ (n)
j
, uγ (n) ) < H(uβ1 ) − lim H(uβn ) + 1 < ∞.
j+1 n→∞
(1.23)
n=1 j=0
Since X is ∑-semicomplete, the sequence

(1) (2) (2) (3) (3)
(β1 =)γ0 , . . . , γν(1)
1
(= β2 = γ0 ), γ1 , . . . , γν(2)
2
(= β3 = γ0 ), γ1 , . . .
has a subsequence {δn } such that {uδn } converges to some uα ∈ X. Note that {δn } is strictly
increasing and converges to α . Without loss of generality, we may take a subsequence with
the assumption that δn + κ ≤ δn+1 for n ∈ N. By (P2 : δn+1 ) we have
h(uδn+1 ) < h(uδn ) for n ∈ N.
Thus, {h(uδn )} is strictly decreasing. Fix ε > 0 and β ∈ Ω with β < α . Choose ν ∈ N satisfying
β < δν , s(uδν , uα ) < ε .
Since h is sequentially lower semicontinuous from above, from (i) we have
h(uα ) ≤ lim h(uδn ) < h(uδν ) ≤ h(uβ ),

n→∞
H(uα ) ≤ κ h(uα ) ≤ κ lim h(uδn )

n→∞
= κ inf{h(uγ ) : γ < α } = lim H(uδn ) (1.24)

n→∞
< H(uδν ) < H(uβ ).
We have shown (P1 : α ) and (P2 : α ). We can choose a finite sequence (γ0 , . . . , γn ) ∈ Ω n+1
satisfying
β = γ 0 < γ 1 < · · · < γn = δ ν ,

n−1
H(uδν ) + ∑ s(uγj , uγj+1 ) < H(uβ ) + ε . (1.25)
j=0
Putting γn+1 = α , we have by (P1 : α )

n−1 n−1
H(uα ) + ∑ s(uγj , uγj+1 ) < H(uδν ) + ∑ s(uγj , uγj+1 ) + ε < H(uβ ) + 2ε . (1.26)
j=0 j=0
Thus uα satisfying (P1 : α )–(P3 : α ) is defined for the second case.

Therefore by transfinite induction, we have defined the net {uα : α ∈ Ω } satisfying (P1 :
α )–(P3 : α ) for any α ∈ Ω . Since the net {H(uα ) : α ∈ Ω } is strictly decreasing, we obtain
car(Q) = car(N) < car(Ω ) < car(Q),
which is a contradiction. Therefore there exists a fixed point of T.

The following result is the extension of the very famous and celebrated Caristi fixed
point theorem in semi-metric spaces. Its proof follows from the proof of Theorem 7.
Theorem 8. Let (X, s) be a ∑-semicomplete semimetric space and let T be a mapping on X. Let h :
X → (−∞, +∞] be a function which is proper and bounded from below. Assume that h is sequentially
lower semicontinuous from above in the sense of Definition 14. Assume also
h(Tx) + s(x, Tx) ≤ h(x) for all x ∈ X. (1.27)
Suzuki in [50] characterized ∑-semicompleteness of semi-metric space (X, s) via Theorem

8, which is our next result.
Theorem 9. Let (X, s) be a semimetric space. Then X is ∑-semicomplete if and only if every self
mapping T on X has a fixed point provided there exists a proper sequentially lower semicontinuous
function h : X → [0, +∞] satisfying (1.27).
Proof. The ‘if’ part follows from Theorem 8. We need to prove the ‘only if’ part. For this we
assume X is not ∑-semicomplete. Then by Proposition 8, X is not (∑, = )-semicomplete.
Hence there exists a sequence {xn } in X such that all xn are different for every n ∈ N,
∞
∑n=1 s(xn , xn+1 ) < ∞ holds and it has no convergent subsequence. We define two mappings
T : X → X and h : X → [0, ∞] as follows:
⎧
⎨ xn+1 if x = xn , for some n ∈ N
Tx =
⎩
x1 if x = xn .
⎧ ∞
⎪
⎨ ∑s(xj , xj+1 ) if x = xn , for some n ∈ N
⎪
hx = j=n
⎪
⎪
⎩
∞ if x = xn .
Note that T and h are well defined because xn are all different and ∑∞n=1 s(xn , xn+1 ) < ∞
holds. Then h is proper. We will prove that h is sequentially lower semicontinuous. We will
make use of a contradiction argument to prove it. Let {yn } be a sequence in X converging
to y ∈ X. We assume that
h(y) > lim inf h(yn ).
n→∞
Then from the definition of h, there exists a subsequence {f (n)} of {n} in N such that
yf (n) ∈ {xm : m > ν } for any n ∈ N, where we put ν ∈ N ∪ {0} by
⎧
⎨ n if y = xn , for some n ∈ N
ν=
⎩
0 if y = xn .
Define a function g from N to {m ∈ N : m > ν } by yf (n) = xg(n) . We consider the following
two cases:
(a) lim sup g(n) = ∞.

n→∞
(b) μ = lim sup g(n) < ∞.
n→∞
In case (a), {xn } has a subsequence converging to y. This is a contradiction. In case (b), we
have
∞ = car({n ∈ N : g(n) = μ }) = car({n ∈ N : yf (n) = xμ }) ≤ car({n ∈ N : yn = xμ }),
which implies that y = xμ . This is also a contradiction. Therefore we obtain that h(y) ≤
lim inf h(yn ). This means h is sequentially lower semicontinuous.
n→∞
Also T satisfies (1.27) for any x ∈ X but T does not have a fixed point, which is a
contradiction with the hypothesis. Thus X is ∑-semicomplete.

The following results can be obtained as corollaries to Theorems 7 and 8.
Corollary 2. Let (X, s) be a (∑, =)-complete semimetric space and let T be a self mapping on X.
Let h : X → (−∞, +∞] be a function which is proper and bounded from below. Assume that h is
sequentially lower semicontinuous from above in the sense of Definition 14. Assume also that there
exists κ ∈ N satisfying the following:
(i) h(Tx) ≤ h(x) for all x ∈ X.

(ii) h(Tκ x) + s(x, Tx) ≤ h(x) for all x ∈ X.

Corollary 3. Let (X, s) be a (∑, =)-complete semimetric space and let T be a self mapping on X.
Let h : X → (−∞, +∞] be a function which is proper and bounded from below. Assume that h is
sequentially lower semicontinuous from above in the sense of Definition 14. Assume also
h(Tx) + s(x, Tx) ≤ h(x) for all x ∈ X. (1.28)
The next result is an extension of the Banach Contraction Principle for semi-metric spaces
with ∑-completeness and ∑-semicompleteness which is a corollary to Theorem 8.
Corollary 4. Let (x, s) be a semi-metric space. Assume that either of the following holds:
(i) X is ∑-complete.
(ii) X is ∑-semicomplete and s is sequentially lower semicontinuous.
Let T be a mapping on X satisfying
s(Tx, Ty) ≤ k s(x, y),
for k ∈ (0, 1) and for all x, y ∈ X. Then T has a fixed point z ∈ X. Moreover, {Tn x} converges to z
for all x ∈ X.
We now discuss coincidence and common fixed points in symmetric spaces. For us it is
impossible to discuss all the results here, so we just give one of the basic result proved by
Hicks and Rhodes [25]. Recall that, x ∈ X is said to be a coincidence point of mappings f
and g on a symmetric space (X, s) if fx = gx holds and y ∈ X is called a common fixed point
of f and g if y = fy = gy holds.
Definition 16. Let (X, s) be a symmetric space and f , g : X → X be two mappings. Then f
and g are said to be
(i) commuting if fgx = gfx for all x ∈ X;

(ii) compatible if lim s(fgxn , gfxn ) = 0 for each sequence {xn } in X such that lim fxn =
n→∞ n→∞
lim gxn ;
n→∞
(iii) noncompatible if there exists a sequence {xn } in X such that lim fxn = lim gxn but
n→∞ n→∞
lim s(fgxn , gfxn ) is either nonzero or nonexistent;
n→∞
(iii) weakly compatible if they commute at their coincidence points.
Theorem 10. Let (X, s) be a S-complete (s-Cauchy complete) symmetric (semi-metric) space with
bounded s and (W3). Suppose f , g : X → X are two s-continuous (τs -continuous) mappings
satisfying
(i) f and g are commuting and g(X) ⊂ f (X);

(ii) s(gx, gy) ≤ ks(fx, fy) for all x, y ∈ X.
Then f and g have a unique common fixed point.

Proof. First we prove the Theorem for s a symmetric. Let M = sup{s(x, y) : x, y ∈ X}. Let us
fix x0 ∈ X. We choose x1 ∈ X such that gx0 = fx1 . Again we choose x2 ∈ X such that gx1 = fx2 .
In general, we choose xn such that f (xn ) = g(xn−1 ). Now,
s(fxn , fxn+m ) = s(gxn−1 , gxn+m−1 )

≤ k s(fxn−1 , fxn+m−1 )
≤ ...
≤ kn s(fx0 , fxm ) ≤ kn M.
Therefore {fxn } is an s-Cauchy sequence and the S-completeness of (X, s) gives us x ∈ X

with lim s(fxn , x) = 0. Since g is s-continuous, we have lim s(gfxn , gx) = 0. Now fxn = gxn−1
n→∞ n→∞
so that lim s(gxn , x) = 0. As f is s-continuous lim s(fgxn , fx) = 0. Since fg = gf , lim s(fgxn , fx) =
n→∞ n→∞ n→∞
lim s(gfxn , gx) = 0. By (W3), we get fx = gx. Also fgx = gfx. Thus f (fx) = f (gx) = g(fx) = g(gx)
n→∞
and s(gx, g(gx)) ≤ k s(fx, f (gx)) = k s(gx, g(gx)) implies g(x) = g(gx). Hence g(x) = g(gx) =
f (gx) so gx is common fixed point of f and g.
To prove uniqueness: If x = fx = gx and y = fy = gy, then (ii) gives s(x, y) = s(gx, gy) ≤
k s(fx, fy) = k s(x, y). Thus x = y.
For s a semi-metric, s(xn , x) → 0 if and only if xn → x in the topology τs . Thus for a τs -
continuous f , xn → x implies fxn → fx and the above proof holds without change.

The proof of the following corollary can be obtained easily, so we skip the proof.
Corollary 5. Let (X, s) be an S-complete (s-Cauchy complete) symmetric (semi-metric) space with
bounded s and (W3). Suppose f , g : X → X to be two commuting, s-continuous (τs -continuous)
mappings satisfying g(X) ⊂ f (X). If there exists a positive integer m and k ∈ (0, 1) such that
s(gm x, gm y) ≤ ks(fx, fy), for all x, y ∈ X, then f and g have a unique common fixed point.
1.6 Fixed Points of Multivalued Mappings

We know that multivalued mappings play a major role in many areas such as studying
disjunctive logic programs. We present some fixed point results for multivalued mappings
in symmetric spaces. The key contribution in the fixed point theory of multivalued map-
pings was made by Hicks [24], Moutawakil [41] etc. A mapping T which associates with
each element x of a set, a subset T(x) of set X is called a multivalued (or set-valued) map-
ping. A multivalued mapping T can be treated as a single-valued mapping of X into power
set 2X , that is, the set of all subsets of X. Let us recall that for a multivalued mapping T, x
is a fixed point of T if x ∈ Tx. Let us begin with the following theorem of Hicks.
Theorem 11. Let (X, s) be a complete symmetric space with bounded s and (W4). Let T : X →
(Bs (X), S ) satisfies lim s(xn , x) = 0 implies lim S(Txn , Tx) = 0. Then, there exists x ∈ X with
n→∞ n→∞
x ∈ Tx if and only if there exists a sequence {xn } in X with xn+1 ∈ Txn and ∑∞n=1 s(xn , xn+1 ) < ∞. In
this case, lim s(xn , x) = 0.
n→∞
Proof. If x ∈ Tx, let xn = x for every n. Suppose the condition holds. Since (X, s) is complete,
there exists x ∈ X with lim s(xn , x) = 0 . Since xn+1 ∈ Txn ,
n→∞
s(xn+1 , Tx) ≤ sup{s(y, Tx) : y ∈ Txn } ≤ S(Txn , Tx) → 0 as n → ∞.
Now, lim s(xn , Tx) = 0, so Lemma 2 gives yn ∈ Tx such that lim s(xn , yn ) = 0. Since we have
n→∞ n→∞
lim s(xn , x) = 0, (W4) gives lim s(yn , x) = 0 which in turn gives s(x, Tx) = 0 or x ∈ Tx = Tx.

n→∞ n→∞
The following result is the extension of Nadler’s [42] fixed point theorem in the setting
of symmetric spaces.
Theorem 12. Let (X, s) be a complete symmetric space with bounded s and (W4). Let T : X → B(X)
satisfy
S (Tx, Ty) ≤ k s(x, y), (1.29)
for all x, y ∈ X and k ∈ (0, 1); then T admits at least one fixed point.
Proof. Choose a fix x0 ∈ X and x1 ∈ Tx0 such that x1 ∈

/ Tx1 . We choose x2 ∈ Tx1 such that
s(x1 , x2 ) ≤ k + s(x1 , Tx1 ) ≤ k + S (Tx0 , Tx1 ).
Following similar analogy, there exists x3 ∈ Tx2 with
s(x2 , x3 ) ≤ k2 + S (Tx1 , Tx2 ).
Continuing in this manner, we obtain a sequence {xn } such that xn+1 ∈ Txn , xn ∈
/ Txn and
s(xn , xn+1 ) < kn + S (Txn−1 , Txn )

≤ kn + k s(xn−1 , xn )
≤ kn + k[kn−1 + S (Txn−2 , Txn−1 )]
≤ 2kn + k2 s(xn−2 , xn−1 )
...
≤ nkn + kn s(x0 , x1 ).
Since ∑∞n=1 nkn and ∑∞n=1 kn both converge, ∑∞n=1 s(xn , xn+1 ) < ∞. Clearly, lim s(un , x) = 0 implies
n→∞
lim S(Tun , Tx) = 0. Therefore Theorem 11 gives x ∈ X with x ∈ Tx.

n→∞
The following two results are fixed point theorems for the more general contractive con-
dition proved in [24]. We skip the proof of these theorems as they can be easily proved
with the help of the proof of Theorem 12.
(Bs (X), S ) be a continuous mapping and Tx be compact for each x ∈ X. For fixed x assume y →
s(x, y) is continuous. Suppose S (Tx, Ty) ≤ k(s(x, y)) holds for all x, y ∈ X, and k : [0, ∞) → [0, ∞)
is a nondecreasing mapping with k(0) = 0. Then T has fixed point in X if and only if there exists x0
∞
in X with ∑ kn (s(x0 , Tx0 )) = ∞. In this case, we can choose xn+1 ∈ Txn with n→∞
lim s(xn , x) = 0.
n=1
(B(X), S ) be a continuous mapping and Tx be compact for each x ∈ X. For fixed x, assume
y → s(x, y) is continuous. Suppose that

S (Tx, Ty) ≤ k max{s(x, y), s(x, Tx), s(y, Ty)}
holds for all x, y ∈ X, and k : [0, ∞) → [0, ∞) is a nondecreasing mapping with k(0) = 0. Then T has
∞
a fixed point in X if and only if there exists x0 in X with ∑ kn (s(x0 , Tx0 )) = ∞. In this case, we can
n=1
choose xn+1 ∈ Txn with lim s(xn , x) = 0.
n→∞
Before going further we recall two definitions from [41].
Definition 17. Let (X, s) be a symmetric space and A a nonempty subset of X. Then
s s
(i) A is s-closed if and only if A = A, where A = {x ∈ X : s(x, A) = 0} and s(x, A) =
inf{d(x, y) : y ∈ A}.
(ii) A is s-bounded if and only if diam(A) < ∞, where diam(A) = sup{s(x, y) : x, y ∈ A}.
The following result is required to prove Theorem 15.
Lemma 6. Let (X, s) be an s-bounded symmetric space. Let A, B ∈ C(X) and α > 1. For each x ∈ A,
there exists y ∈ B such that s(x, y) ≤ α S (A, B).
Theorem 15. Let (X, s) be an s-bounded and S-complete symmetric space satisfying (W4) and
T : X → C(X) be a multivalued mapping such that
S (Tx, Ty) ≤ k s(x, y), k ∈ [0, 1), for every x, y ∈ X. (1.30)
Then there exists z ∈ X such that z ∈ Tz.
Proof. Let x1 ∈ X and α ∈ (k, 1). Since Tx1 is nonempty, there exists x2 ∈ Tx1 such that
s(x1 , x2 ) > 0 (otherwise x1 is a fixed point of T). By (1.30), we have
s(x2 , Tx2 ) ≤ S (Tx1 , Tx2 ) ≤ k s(x1 , x2 ) < α s(x1 , x2 ) < α s(x1 , x2 ).
It follows that there exists x3 (say) in Tx2 such that
s(x2 , x3 ) < α s(x1 , x2 ).
Similarly, there exists x4 ∈ Tx3 such that
s(x3 , x4 ) < α s(x2 , x3 ).
Continuing in this manner, we get a sequence {xn } in X such that xn+1 ∈ Txn , s(xn , xn+1 ) > 0
and
s(xn , xn+1 ) < α s(xn−1 , xn ).
We claim that {xn } is an s-Cauchy sequence. Indeed, we have
s(xn , xn+m ) < α s(xn−1 , xn+m−1 )

< α 2 s(xn−2 , xn+m−2 )
...
< α n−1 s(x1 , xm+1 )
< α n−1 diam(X).
This implies {xn } is an s-Cauchy sequence. Hence due to S-completeness of X, there exists
z in X such that lim s(xn , z) = 0. Now, from Lemma 6, we have that for each n ∈ N there
n→∞
exists yn ∈ Tz such that for ε > 1,
s(xn+1 , yn ) ≤ ε S (Txn , Tz) ≤ ε ks(xn , z), n = 1, 2, . . . ,
which implies that lim s(xn+1 , yn ) = 0. In view of (W4), we have lim s(yn , z) = 0 and therefore
n→∞ n→∞
s
z ∈ Tz = Tz.

Recently, the notion of S + -type contractions was introduced in [44] and proved fixed
point theorems for multivalued mapping. The authors used the notion of α -admissibility
defined by Samet et al. [46] in order to relax the requirement of satisfying the contractive
condition at every pair of points in a space without altering the outcome. The idea of
α -admissible mappings is interesting as it includes the case of discontinuous mappings,
unlike the contraction mapping. In the rest of this chapter, the used mapping α (unless
mentioned) is considered as α : X × X → [0, ∞), where X is nonempty.
Definition 18. A self-mapping T : X → X is called α -admissible if for x, y ∈ X, the condition

α (x, y) ≥ 1 implies that α (Tx, Ty) ≥ 1.
Subsequently, multivalued α -admissibility was proposed by Mohammadi et al. [40] as

follows:
Definition 19. A set-valued mapping T : X → N (X) is called an α -admissible mapping if

for all x ∈ X and y ∈ Tx, α (x, y) ≥ 1 implies α (y, z) ≥ 1 for each z ∈ Ty.
The α -admissible pair of multivalued mapping is defined as follows.
Definition 20. Let T, S : X → N (X) be two mappings. The ordered pair (T, S) is said to be
α -admissible if for all x, y ∈ X, α (x, y) ≥ 1 implies α (p, q) ≥ 1, for all p ∈ Tx and q ∈ Sy.
First, we extend the idea of α -completeness to the symmetric space (X, s) along the lines
of [26].
Definition 21. A symmetric space (X, s) is said to be α -complete if for every sequence {xn }
∞
in X satisfying ∑ s(xn , xn+1 ) < ∞ with α (xn , xn+1 ) ≥ 1 for all n ∈ N, there exists x ∈ X such
n=1
that lim s(xn , x) = 0.
n→∞
Remark 5. If (X, s) is complete, then it is also α -complete. The converse need not be true as
shown in the following example.
Example 8. Let X = { n1 : n ∈ N} ∪ {1 + 1j : j ∈ N}. Define s : X × X → [0, ∞) by s(x, y) = |y − x|

for all x, y ∈ Y. Then (X, s) is a complete symmetric space. Consider α : X × X → [0, ∞) as
⎧
⎨ 1 if x, y ≥ 1,
α (x, y) =
⎩
0 otherwise.
Here, (X, s) is also α -complete. In fact, for every sequence {xn } in Y satisfying
∞
∑ s(xn , xn+1 ) < ∞ with α (xn , xn+1 ) ≥ 1 for all n ∈ N, we have xn ∈ {1 + 1
n
: n ∈ N}. There
j=1
exists a = 1 ∈ Y such that lim s(1 + n1 , 1) = 0.
n→∞
Let Φ denote the set of all monotone nondecreasing functions ϕ : [0, ∞) → [0, ∞) such that
∞
∑n=1 ϕ n (t)
< ∞ for each t > 0, where ϕ n is the nth iterate of ϕ .
Lemma 7. Given the following statements:
(i) ϕ ∈ Φ ;
(ii) lim ϕ n (t) = 0;
n→∞
(iii) ϕ (t) < t for all t > 0.
Then (i) implies (ii) implies (iii).
The notion of α -S + -continuity of a mapping weakens the continuity.
Definition 22. A set-valued mapping T : X → C L (X) is called α -S + -continuous on

C L (X) if lim s(xn , x) = 0 and α (xn , xn+1 ) ≥ 1 for all n ∈ N implies lim S + (Txn , Tx) = 0.
n→∞ n→∞
Lemma 8. [24] Let (X, s) be a symmetric space and T : X → Bs (X). Then lim s(xn , Tx) = 0 if
n→∞
and only if there exists yn ∈ Tx satisfying lim s(xn , yn ) = 0.
n→∞
We now introduce the notion of α -ϕ -S + -contractive multivalued mapping.
Definition 23. Let (X, s) be a symmetric space. A set-valued mapping T : X → N (X) is

called α -ϕ -S + -contractive
(1) if there exist two functions ϕ ∈ Φ and α such that

α (x, y)S + (Tx, Ty) ≤ ϕ s(x, y) for all x, y ∈ X, (1.31)
(2) for every x ∈ X, y ∈ Tx, q ≥ 1, there exists z ∈ Ty such that
s(y, z) ≤ q S + (Tx, Ty).
Definition 24. In the above definition, if we put m(x, y) = max s(x, y), s(x, Tx), s(y, Ty)
instead of s(x, y) in (1.31), then the mapping T is called generalized α -ϕ -S + -contractive.
Remark 6. The condition (2) in the above definition holds automatically if we replace S +
by H .
Theorem 16. Let (X, s) be an α -complete symmetric space with (W4). Let T : X → C B s (X) be a
generalized α -ϕ -S + -contraction. Assume that
(i) T is α -admissible;
(ii) there exist x0 in X and x1 in Tx0 such that α (x0 , x1 ) ≥ 1;
(iii) T is α -S + -continuous.
Then T admits a fixed point.
Proof. From (ii), we have x0 ∈ X and x1 ∈ Ta0 such that α (x0 , x1 ) ≥ 1. Assume x0 = x1 , oth-
erwise x0 is a fixed point of T. Assume also x1 ∈ / Tx1 , otherwise x1 will be a fixed point of
T.
Define a sequence {xn } in X by x1 ∈ Tx0 , x2 ∈ Tx1 , . . ., xn+1 ∈ Txn , for all n ∈ N such that
/ Txn . Further, using (i), we obtain α (xn , xn+1 ) ≥ 1. Because of condition (2) in Definition
xn ∈
23, we now have
s(xn , xn+1 ) ≤ α (xn−1 , xn )S + (Txn−1 , Txn )

≤ ϕ max s(xn−1 , xn ), s(xn−1 , Txn−1 ), s(xn , Txn )

≤ ϕ max s(xn−1 , xn ), s(xn , xn+1 ) . (1.32)
If max s(xn−1 , xn ), s(xn , xn+1 ) = s(xn , xn+1 ), then from (1.32) we have s(xn , xn+1 ) ≤
ϕ (s(xn , xn+1 )) and then by Lemma 7, we get s(xn , xn+1 ) < s(xn , xn+1 ), a contradiction.
Therefore,
s(xn , xn+1 ) ≤ ϕ (s(xn−1 , xn )).
Repeating the above process, we get s(xn , xn+1 ) ≤ ϕ n (s(x0 , x1 )). Since, ∑∞n=1 ϕ n (t) < ∞ for all
t > 0, so we have ∑∞n=1 s(xn , xn+1 ) < ∞.
As X is an α -complete symmetric space, so there exists x ∈ X such that lim s(xn , x) = 0. Also
n→∞
the α -H + -continuity of T gives us
lim S + (Txn , Tx) = 0.

n→∞
Since xn+1 ∈ Txn , by using condition (2) in Definition 23 for q ≥ 1, we get
s(xn+1 , Tx) ≤ qH + (Txn , Tx) → 0 as j → ∞.
Thus, lim s(xn+1 , Tx) = 0. This is equivalent to lim s(xn , Tx) = 0. Therefore by Lemma 8, there
n→∞ n→∞
exists yn ∈ Tx such that lim s(xn , yn ) = 0. Since lim s(xn , x) = 0, (W4) implies lim s(yn , x) = 0
n→∞ n→∞ n→∞
which in turn implies s(x, Tx) = 0 and since Tx is closed, x ∈ Tx.

The following result can be proved with similar lines of proof to Theorem 16.
Theorem 17. Let (X, s) be an α -complete symmetric space with (W4) and T : X → C B s (X) be an
α -ϕ -S + -contractive mapping. Assume that (i)–(iii) of Theorem 16 are true. Then T admits a fixed
point.
In the next result the α -H + -continuity of the mapping is relaxed.
Theorem 18. Let (X, s) be an α -complete symmetric space with (CC) and T : X → C B s (X) be a
generalized α -ϕ -S + -contractive mapping. Assume that
(ii) there exist x0 in X and x1 in Tx0 such that α (x0 , x1 ) ≥ 1;
(iii)’ if {xn } is a sequence in X with lim s(xn , x) = 0 and α (xn , xn+1 ) ≥ 1 for all j ∈ N then
n→∞
α (xn , x) ≥ 1.
∞
Proof. Following the proof of Theorem 16, we have that ∑ s(xn , xn+1 ) < ∞ and α (xn , xn+1 ) ≥ 1
n=1
for all n ∈ N. Then by α -completeness of (X, s), there exists x ∈ X such that lim s(xn , x) = 0.
n→∞
Using (iii)’, we get α (xn , x) ≥ 1 for all n ∈ N. We now claim that x ∈ Tx. Assume that x ∈
/ Tx,
then s(x, Tx) > 0. By using (1.31), we have
s(xn+1 , Tx) ≤ α (xn , x)H + (Txn , Tx)

≤ ϕ max{s(xn , x), s(xn , Txn ), s(x, Tx)} . (1.33)
s(x, Tx) s(x, Tx)

Let ε = . Since lim s(xn , x) = 0, we can find n1 ∈ N such that s(xn , x) < for all
2 n→∞ 2
n > n1 .
s(x, Tx)
Also, as lim s(xn , xn+1 ) = 0, we can find n2 ∈ N such that s(xn , Txn ) ≤ s(xn , xn+1 ) <
n→∞ 2
for all n > n2 . Thus we get
max{s(xn , x), s(xn , Txn ), s(x, Tx)} = s(x, Tx)
for all n ≥ n0 = max{n1 , n2 }. Therefore, (1.33) yields us
s(xn+1 , Tx) ≤ ϕ (s(x, Tx)) (1.34)
for n ≥ n0 . Taking the limit as n → ∞ in (1.34) and in view of condition (CC), we get s(x, Tx) ≤
ϕ (s(x, Tx)), which is a contradiction to the consequence of Lemma 7. Thus our assumption
is wrong. Hence x ∈ Tx.

Following the proof of above theorem, the next result can be proved easily.
Theorem 19. Let (X, s) be an α -complete symmetric space with (CC). Let T : X → C B s (X) be
an α -ϕ -S + -contraction. Then if conditions (i), (ii) and (iii)’ of Theorem 18 hold, T admits a fixed
point.
Now we will discuss the common fixed point theorems for multivalued mappings using
S + distance in symmetric spaces. We first introduce the concept of an α -ϕ -S + -contractive
pair of mappings.
Definition 25. Let (X, s) be a symmetric space. Given T, S : X → N (X). (T, S) is called an
α -ϕ -S + -contractive pair if
(1) there exist ϕ ∈ Φ and a symmetric function α : X × X → [0, ∞) such that

α (x, y)S + (Tx, Sy) ≤ ϕ mT,S (x, y) (1.35)
for all x, y ∈ X, where mT,S (x, y) = max s(x, y), s(x, Tx), s(y, Sy) ,
(2) for every x ∈ X,
(a) y ∈ Tx, q ≥ 1, there exists z ∈ Sy such that
s(y, z) ≤ qS + (Tx, Sy),
(b) y ∈ Sx, q ≥ 1, there exists z ∈ Ty such that
s(y, z) ≤ qS + (Sx, Ty).
We are now ready to prove a common fixed point result for the class of α -ϕ -S + -
contractive pair of multivalued mappings.
Theorem 20. Let (X, s) be an α -complete symmetric space with (CC) and (T, S) of mappings T, S :
X → C B s (X) be an α -ϕ -S + -contractive pair. Assume that
(i) (T, S) is α -admissible;

(ii) there exist x0 in X and x1 ∈ Tx0 such that α (x0 , x1 ) ≥ 1;
(iii) if {xn } is any sequence in X with lim s(xn , x) = 0 and α (xn , xn+1 ) ≥ 1 for all n ∈ N, then we
n→∞
have α (xn , x) ≥ 1.
Proof. Let x0 ∈ X be arbitrary and x1 ∈ Tx0 . We assume x0 = x1 ; otherwise there is nothing

to prove. This means s(x0 , x1 ) > 0. From (ii), we have α (x0 , x1 ) ≥ 1. Thus by virtue of 2(a) of
Definition 25, we choose x2 ∈ Sx1 such that
s(x1 , Sx1 ) ≤ s(x1 , x2 )
≤ α (x0 , x1 )S + (Tx0 , Sx1 )

≤ ϕ mS,T (x0 , x1 )

≤ ϕ max{s(x0 , x1 ), s(x0 , Tx0 ), s(x1 , Sx1 )}

= ϕ max{s(x0 , x1 ), s(x1 , Sx1 )} . (1.36)
Clearly, from the above inequality we can conclude that max{s(x0 , x1 ), s(x1 , Sx1 )} = s(x0 , x1 ).
Otherwise, the second case would lead us to a contradiction. Thus (1.36), yields us
s(x1 , x2 ) ≤ ϕ (s(x0 , x1 )). (1.37)
As x1 ∈ Tx0 and x2 ∈ Sx1 and due to α -admissibility of (T, S), we have α (x1 , x2 ) ≥ 1. Thus
by virtue of 2(b) of Definition 25, we choose x3 ∈ Tx2 such that
s(x2 , Tx2 ) ≤ s(x2 , x3 )
≤ α (x1 , x2 )S + (Sx1 , Tx2 )

≤ ϕ mS,T (x1 , x2 )

≤ ϕ max{s(x1 , x2 ), s(x1 , Sx1 ), s(x2 , Tx2 )}

= ϕ max{s(x1 , x2 ), s(x2 , Tx2 )} . (1.38)
Again, we have max{s(x1 , x2 ), s(x2 , Tx2 )} = s(x1 , x2 ). Otherwise, the second case would lead
to a contradiction. Thus from (1.38), we get
s(x2 , x3 ) ≤ ϕ (s(x1 , x2 )) = ϕ 2 (s(x0 , x1 )). (1.39)
Persisting this way, a sequence {xn } in X is generated such that x2n+1 ∈ Tx2n , x2n+2 ∈ Sx2n+1
satisfying α (xn , xn+1 ) ≥ 1 and
s(xn , xn+1 ) = ϕ n (s(x0 , x1 )) for all n ∈ N. (1.40)
Since ∑∞n=1 ϕ n (t) < ∞, we have ∑∞n=1 s(xn , xn+1 ) < ∞. As the symmetric space (X, s) is
α -complete, there exists x ∈ X such that
lim s(xn , x) = 0.
n→∞
From (iii), we have α (xn+1 , x) ≥ 1 for all n ∈ N. We now claim that x ∈ Tx ∩ Sx. First, let us
assume x ∈ / Tx; then s(x, Tx) > 0. By 2(a), we have
s(x2n+2 , Tx) ≤ α (x2n+1 , x)S + (Sx2n+1 , Tx)

≤ ϕ max{s(x2n+1 , x), s(x, Tx), s(x2n+1 , Sx2n+1 )} . (1.41)
Since lim s(xn , x) = 0, we can find an integer N1 ∈ N such that s(x2n+1 , x) < ε = s(x,Tx)
2
for all
n→∞
n > N1 . Also as {xn } is a sequence such that lim s(xn , xn+1 ) = 0, we can find an integer N2 ∈ N
n→∞
such that d(x2n+1 , Sx2n+1 ) ≤ s(x2n+1 , x2n+2 ) < ε = s(x,Tx)
2
for all n > N2 . Thus, we get
max{s(x, x2n+1 ), s(x, Tx), s(x2n+1 , Sx2n+1 )} = s(x, Tx),
for all n ≥ N0 = max{N1 , N2 }. Therefore we have
s(Tx, x2n+2 ) ≤ ϕ (s(x, Tx)) for all n ≥ N0 .
Taking n → ∞ and in view of (CC), we get s(Tx, x) < s(x, Tx), which gives us s(x, Tx) = 0.
As Tx is closed, we have x ∈ Tx. Arguing in a similar way, we can get x ∈ Sx and hence
x ∈ Tx ∩ Sx.

The following result is a fixed point theorem without using H or S + distance functions
in symmetric spaces.
Definition 26. Let (X, s) be a symmetric space. A multivalued mapping T : X → C B s (X) is

called generalized pointwise α -ϕ -contractive if there exist functions ϕ ∈ Φ and α : X × X →
[0, ∞) such that for x1 , x2 ∈ X, y1 ∈ Tx1 , y2 ∈ Tx2 ,

α (x1 , x2 )s(y1 , y2 ) ≤ ϕ M3 (x1 , x2 ) , (1.42)
where M3 (x1 , x2 ) = max s(x1 , x2 ), s(x1 , y1 ), s(x2 , y2 ) .
Definition 27. A mapping T is called pointwise α -ϕ -contractive if we replace M3 (x1 , x2 ) by

s(x1 , x2 ) in Definition 26.
Theorem 21. Let (X, s) be an α -complete symmetric space with (W4) and mapping T : X →
C B s (X) be generalized point-wise α -ϕ -contractive. Then T admits a fixed point if the following
hold:
(ii) there exist x0 in X and x1 ∈ Tx0 such that α (x0 , x1 ) ≥ 1;
(iii) for every sequence {xn } in X such that lim s(xn , x) = 0 with α (xn , xn+1 ) ≥ 1, there exists a
n→∞
sequence {yn } in Txn such that lim s(yn , y) = 0 for some y ∈ Tx.
n→∞
Proof. Initiating with arbitrary x0 ∈ X and x1 ∈ Tx0 such that α (x0 , x1 ) ≥ 1, then following
the proof of Theorem 16 we get a sequence {xn } defined by x1 ∈ Tx0 , x2 ∈ Tx1 , . . . , xn+1 ∈ Txn
for all n ∈ N such that xn ∈/ Txn . Since T is α -admissible, we have α (xn , xn+1 ) ≥ 1 for all
n ∈ N ∪ {0}. By (1.42), we have
s(xn , xn+1 ) ≤ α (xn−1 , xn )s(xn , xn+1 )
≤ ϕ (M3 (xn−1 , xn ))

≤ ϕ max s(xn−1 , xn ), s(xn−1 , xn ), s(xn , xn+1 )

≤ ϕ max s(xn−1 , xn ), s(xn , xn+1 ) . (1.43)
If max s(xn−1 , xn ), s(xn , xn+1 ) = s(xn , xn+1 ), then from (1.43) we have s(xn , xn+1 ) ≤
ϕ (s(xn , xn+1 )). Using Lemma 7, we get s(xn , xn+1 ) < s(xn , xn+1 ); that is a contradiction. Thus
(1.43) gives
s(xn , xn+1 ) ≤ ϕ (s(xn−1 , xn )).
Repeating this process, we get
s(xn , xn+1 ) ≤ ϕ n (s(x0 , x1 )).
As ∑∞n=1 ϕ n (t) < ∞ for all t > 0, so we obtain
∞
∑ s(xn , xn+1 ) < ∞.
n=1
Due to α -completeness of the symmetric space X, there exists x ∈ X such that lim s(xn , x) = 0
n→∞
and by (iii), we get a sequence {yn } ∈ Txn such that lim s(yn , y) = 0 for some y ∈ Tx.
n→∞
Since xn+1 ∈ Txn , we get
s(xn+1 , Tx) = inf{s(xn , Tx) : an ∈ Tan }
≤ s(yn , y) → 0 as n → ∞.
Thus, we find that lim s(xn+1 , Tx) = 0. This is equivalent to lim s(xn , Tx) = 0. Therefore by
n→∞ n→∞
Lemma 8, there exists zn ∈ Tx such that lim s(xn , zn ) = 0. Since lim s(xn , x) = 0, (W4) implies
n→∞ n→∞
lim s(zn , x) = 0 which in turn implies s(x, Tx) = 0 and since Tx is closed, x ∈ Tx.

n→∞
The following results follow in a similar way to the above proof.
Theorem 22. Let (X, s) be an α -complete symmetric space with (W4) and mapping T : X →
C B s (X) be pointwise α -ϕ -contractive. If conditions (i)–(iii) in Theorem 21 hold, then T admits a
fixed point.
Remark 7. Theorem 21 and 22 also hold if condition (iii) is replaced by the α -continuity
of T.
Fixed point theorems in symmetric spaces have found applications in probabilistic met-
ric spaces as the very general probabilistic structures admit a compatible symmetric or
semi-metric such that the distribution function Fx,y is related to the symmetric s by s(x, y) < t
if and only if Fx,y (t) > 1 − t where t > 0. Hicks and Rhodes [25] discussed the application of
Theorems 12, 13 and 14 whereas Patle et al.[44] applied Theorem 16 in probabilistic metric
spaces.
1.7 Conclusion and Future Investigations

Some of the earliest results in the theory of abstract spaces and abstract distances were
the contributions of Fréchet and Menger. One of the major problems in point set topol-
ogy in the 19th century was the metrization problem. The study of these problems led
to the development of study of semi-metrization. Early investigations in this area were
made by Wilson. Starting from the work of Wilson on symmetric spaces, in this chapter
we first discussed the topology of symmetric spaces. The absence of a triangular inequality
in defining symmetric spaces leads to the various concepts of Cauchy-type sequences and
completeness of symmetric spaces. All these concepts have been studied in detail from the
point of view of their usage.
The major part of this chapter dealt with the study of fixed point theorems in symmetric
spaces. Taking into consideration their importance, we also discussed the fixed point of
multivalued mappings and common fixed point theorems of two mappings with various
contractive conditions. In addition the result due to Suzuki, which proves the characteri-
zation of semi-completeness by means of the fixed point property, is also presented, which
provides some self-containment to this chapter.
Although the area of fixed point theory for symmetric spaces has developed into a wide
literature there is still a lot to be explored. Some of the open questions we are discussing
here:
• Suzuki [50] (as discussed in Theorem 9) characterized semicompleteness of semi-

metric spaces through the fixed point property. Can such a result be proved for
completeness instead of semicompleteness?
• A basic best proximity point result is discussed in a paper by Felhi [18]. Is there any
possibility of establishing the new best proximity point results in symmetric spaces?
• Can fixed point results for Z-contractions and F-contractions and other contractive
conditions be extended to the notion of symmetric spaces?
• Is it possible to prove Proposition 11 without Hausdorffness?
• Does every symmetric/semimetric space have a completion? How it can be proved?
References
1. M. Aamri, A. Bassou, D. El Moutawakil, Common fixed points for weakly compatible maps
in symmetric spaces with applications to probabilistic spaces, Appl. Math. E-Notes 5 (2005),
171–175.
Another random document with
no related content on Scribd:
insulted her and is expected to pay damages. If a man meets his
mother-in-law coming along the road and does not recognise her,
she will fall down on the ground as a sign, when he will run away. In
the same way a father-in-law will signal to his daughter-in-law; the
whole idea being that they are unworthy to be noticed till they have
proved that they can beget children.”79.2 However, if a wife should
prove barren for three years, the rules of avoidance between the
young couple and their parents-in-law cease to be observed.79.3
Hence the custom of avoidance among these people is associated in
some way with the wife’s fertility. So among the Awemba, a Bantu
tribe of Northern Rhodesia, “if a young man sees his mother-in-law
coming along the path, he must retreat into the bush and make way
for her, or if she suddenly comes upon him he must keep his eyes
fixed on the ground, and only after a child is born may they converse
together.”79.4 Among the Angoni, another Bantu tribe of British
Central Africa, it would be a gross breach of etiquette if a man were
to enter his son-in-law’s house; he may come within ten paces of the
door, but no nearer. A woman may not even approach her son-in-
law’s house, and she is never allowed to speak to him. Should they
meet accidentally on a path, the son-in-law gives way and makes a
circuit to avoid encountering his mother-in-law face to face.79.5 Here
then we see that a man avoids his son-in-law as well as his mother-
in-law, though not so strictly.
Among the Thonga, a Bantu tribe about
The custom of Delagoa Bay, when a man meets his mother-in-
avoiding mother-in-
law and wife of law or her sister on the road, he steps out of the
wife’s brother road into the forest on the right hand side and sits
among the Thonga down. She does the same. Then they salute each
of Delagoa Bay.
other in the usual way by clapping their hands.
After that they may talk to each other. When a man is in a hut, his
mother-in-law dare not enter it, but must sit down outside without
seeing him. So seated she may salute him, “Good morning, son of
So-and-so.” But she would not dare to pronounce his name.
However, when a man has been married many years, his mother-in-
law has less fear of him, and will even enter the hut where he is and
speak to him. But among the Thonga the woman whom a man is
bound by custom to avoid most rigidly is not his wife’s mother, but
the wife of his wife’s brother. If the two meet on a path, they carefully
avoid each other; he will step out of the way and she will hurry on,
while her companions, if she has any, will stop and chat with him.
She will not enter the same boat with him, if she can help it, to cross
a river. She will not eat out of the same dish. If he speaks to her, it is
with constraint and embarrassment. He will not enter her hut, but will
crouch at the door and address her in a voice trembling with
emotion. Should there be no one else to bring him food, she will do it
reluctantly, watching his hut and putting the food inside the door
when he is absent. It is not that they dislike each other, but that they
feel a mutual, a mysterious fear.80.1 However, among the Thonga,
the rules of avoidance between connexions by marriage decrease in
severity as time passes. The strained relations between a man and
his wife’s mother in particular become easier. He begins to call her
“Mother” and she calls him “Son.” This change even goes so far that
in some cases the man may go and dwell in the village of his wife’s
parents, especially if he has children and the children are grown
up.80.2 Again, among the Ovambo, a Bantu people of German South-
West Africa, a man may not look at his future mother-in-law while he
talks with her, but is bound to keep his eyes steadily fixed on the
ground. In some cases the avoidance is even more stringent; if the
two meet unexpectedly, they separate at once. But after the
marriage has been celebrated, the social intercourse between
mother-in-law and son-in-law becomes easier on both sides.81.1
Thus far our examples of ceremonial avoidance
The custom of between mother-in-law and son-in-law have been
avoiding the
mother-in-law drawn from Bantu tribes. But in Africa the custom,
among other than though apparently most prevalent and most
the Bantu tribes of strongly marked among peoples of the great Bantu
Africa.
stock, is not confined to them. Among the Masai of
British East Africa, “mothers-in-law and their sons-in-law must avoid
one another as much as possible; and if a son-in-law enters his
mother-in-law’s hut she must retire into the inner compartment and
sit on the bed, whilst he remains in the outer compartment; they may
then talk. Own brothers-in-law and sisters-in-law must also avoid one
another, though this rule does not apply to half-brothers-in-law and
sisters-in-law.”81.2 So, too, among the Bogos, a tribe on the outskirts
of Abyssinia, a man never sees the face of his mother-in-law and
never pronounces her name; the two take care not to meet.81.3
Among the Donaglas a husband after marriage “lives in his wife’s
house for a year, without being allowed to see his mother-in-law, with
whom he enters into relations only on the birth of his first son.”81.4 In
Darfur, when a youth has been betrothed to a girl, however intimate
he may have been with her parents before, he ceases to see them
until the ceremony has taken place, and even avoids them in the
street. They, on their part, hide their faces, if they happen to meet
him unexpectedly.81.5
To pass now from Africa to other parts of the
The custom of world, among the Looboos, a primitive tribe in the
avoiding relations
by marriage in tropical forests of Sumatra, custom forbids a
Sumatra and New woman to be in her father-in-law’s company and a
Guinea. man to be in his mother-in-law’s society. For
example, if a man meets his daughter-in-law, he
should cross over to the other side of the road to let her pass as far
as possible from him; but if the way is too narrow, he takes care in
time to get out of it. But no such reserve is prescribed between a
father-in-law and his son-in-law, or between a mother-in-law and her
daughter-in-law.82.1 Among the Bukaua, a Melanesian tribe of
German New Guinea, the rules of avoidance between persons
connected by marriage are very stringent; they may not touch each
other or mention each other’s names. But contrary to the usual
practice the avoidance seems to be quite as strict between persons
of the same sex as between males and females. At least the writer
who reports the custom illustrates it chiefly by the etiquette which is
observed between a man and his daughter’s husband. When a man
eats in presence of his son-in-law, he veils his face; but if
nevertheless his son-in-law should see his open mouth, the father-in-
law is so ashamed that he runs away into the wood. If he gives his
son-in-law anything, such as betel or tobacco, he will never put it in
his hand, but pours it on a leaf, and the son-in-law fetches it away. If
father-in-law and son-in-law both take part in a wild boar hunt, the
son-in-law will abstain from seizing or binding the boar, lest he
should chance to touch his father-in-law. If, however, through any
accident their hands or backs should come into contact, the father-
in-law is extremely horrified, and a dog must be at once killed, which
he gives to his son-in-law for the purpose of wiping out the stain on
his honour. If the two should ever fall out about anything, the son-in-
law will leave the village and his wife, and will stay away in some
other place till his father-in-law, for his daughter’s sake, calls him
back. A man in like manner will never touch his sister-in-law.82.2
Among the low savages of the Californian
The custom of peninsula a man was not allowed for some time to
avoiding relations
by marriage among look into the face of his mother-in-law or of his
the Indian tribes of wife’s other near relations; when these women
America. were present he had to step aside or hide
himself.83.1 Among the Indians of the Isla del
Malhado in Florida a father-in-law and mother-in-law might not enter
the house of their son-in-law, and he on his side might not appear
before his father-in-law and his relations. If they met by accident they
had to go apart to the distance of a bowshot, holding their heads
down and their eyes turned to the earth. But a woman was free to
converse with the father and mother of her husband.83.2 Among the
Indians of Yucatan, if a betrothed man saw his future father-in-law or
mother-in-law at a distance, he turned away as quickly as possible,
believing that a meeting with them would prevent him from begetting
children.83.3 Among the Arawaks of British Guiana a man may never
see the face of his wife’s mother. If she is in the house with him, they
must be separated by a screen or partition-wall; if she travels with
him in a canoe, she steps in first, in order that she may turn her back
to him.83.4 Among the Caribs “the women never quit their father’s
house, and in that they have an advantage over their husbands in as
much as they may talk to all sorts of people, whereas the husband
dare not converse with his wife’s relations, unless he is dispensed
from this observance either by their tender age or by their
intoxication. They shun meeting them and make great circuits for
that purpose. If they are surprised in a place where they cannot help
meeting, the person addressed turns his face another way so as not
to be obliged to see the person, whose voice he is compelled to
hear.”83.5 Among the Araucanian Indians of Chili a man’s mother-in-
law refuses to speak to or even to look at him during the marriage
festivity, and “the point of honour is, in some instances, carried so
far, that for years after the marriage the mother never addresses her
son-in-law face to face; though with her back turned, or with the
interposition of a fence or a partition, she will converse with him
freely.”84.1
It would be easy to multiply examples of similar
The custom of customs of avoidance between persons closely
avoiding relations
by marriage cannot connected by marriage, but the foregoing may
be separated from serve as specimens. Now in order to determine
the similar custom the meaning of such customs it is very important
of avoiding relations
by blood; both are to observe that similar customs of avoidance are
probably practised in some tribes not merely between
precautions to
prevent improper
persons connected with each other by marriage,
relations between but also between the nearest blood relations of
the sexes. different sexes, namely, between parents and
children and between brothers and sisters;84.2 and
the customs are so alike that it seems difficult or impossible to
separate them and to offer one explanation of the avoidance of
connexions by marriage and another different explanation of the
avoidance of blood relations. Yet this is what is done by some who
attempt to explain the customs of avoidance; or rather they confine
their attention wholly to connexions by marriage, or even to mothers-
in-law alone, while they completely ignore blood relations, although
in point of fact it is the avoidance of blood relations which seems to
furnish the key to the problem of such avoidances in general. The
true explanation of all such customs of avoidance appears to be, as I
have already indicated, that they are precautions designed to
remove the temptation to sexual intercourse between persons whose
marriage union is for any reason repugnant to the moral sense of the
community. This explanation, while it has been rejected by theorists
at home, has been adopted by some of the best observers of savage
life, whose opinion is entitled to carry the greatest weight.85.1
That a fear of improper intimacy even between
Mutual avoidance of the nearest blood relations is not baseless among
mother and son, of
father and daughter, races of a lower culture seems proved by the
and of brother and testimony of a Dutch missionary in regard to the
Battas or Bataks of Sumatra, a people who have
sister among the attained to a fairly high degree of barbaric
Battas.
civilization. The Battas “observe certain rules of
avoidance in regard to near relations by blood or marriage; and we
are informed that such avoidance springs not from the strictness but
from the looseness of their moral practice. A Batta, it is said,
assumes that a solitary meeting of a man with a woman leads to an
improper intimacy between them. But at the same time he believes
that incest or the sexual intercourse of near relations excites the
anger of the gods and entails calamities of all sorts. Hence near
relations are obliged to avoid each other lest they should succumb to
temptation. A Batta, for example, would think it shocking were a
brother to escort his sister to an evening party. Even in the presence
of others a Batta brother and sister feel embarrassed. If one of them
comes into the house, the other will go away. Further, a man may
never be alone in the house with his daughter, nor a mother with her
son. A man may never speak to his mother-in-law nor a woman to
her father-in-law. The Dutch missionary who reports these customs
adds that he is sorry to say that from what he knows of the Battas he
believes the maintenance of most of these rules to be very
necessary. For the same reason, he tells us, as soon as Batta lads
have reached the age of puberty they are no longer allowed to sleep
in the family house but are sent away to pass the night in a separate
building (djambon); and similarly as soon as a man loses his wife by
death he is excluded from the house.”85.2
In like manner among the Melanesians of the
Mutual avoidance of Banks’ Islands and the New Hebrides a man must
mother and son and
of brother and sister not only avoid his mother-in-law; from the time
among the when he reaches or approaches puberty and has
Melanesians. begun to wear clothes instead of running about
naked, he must avoid his mother and sisters, and
he may no longer live in the same house with them; he takes up his
quarters in the clubhouse of the unmarried males, where he now
regularly eats and sleeps. He may go to his father’s house to ask for
food, but if his sister is within he must go away before he eats; if she
is not there, he may sit down near the door and eat. If by chance
brother and sister meet in the path, she runs away or hides. If a boy,
walking on the sands, perceives footprints which he knows to be
those of his sister, he will not follow them, nor will she follow his. This
mutual avoidance lasts through life. Not only must he avoid the
persons of his sisters, but he may not pronounce their names or
even use a common word which happens to form part of any one of
their names. In like manner his sisters eschew the use of his name
and of all words which form part of it. Strict, too, is a boy’s reserve
towards his mother from the time when he begins to wear clothes,
and the reserve increases as he grows to manhood. It is greater on
her side than on his. He may go to the house and ask for food and
his mother may bring it out for him, but she will not give it to him; she
puts it down for him to take. If she calls to him to come, she speaks
to him in the plural, in a more distant manner; “Come ye,” she says,
not “Come thou.” If they talk together she sits at a little distance and
turns away, for she is shy of her grown-up son. “The meaning of all
this,” as Dr. Codrington observes, “is obvious.”86.1
Mutual avoidance of
a man and his
When a Melanesian man of the Banks’ Islands
mother-in-law marries, he is bound in like manner to avoid his
among the mother-in-law. “The rules of avoidance are very
Melanesians.
strict and minute. As regards the avoidance of the
person, a man will not come near his wife’s mother; the avoidance is
mutual; if the two chance to meet in a path, the woman will step out
of it and stand with her back turned till he has gone by, or perhaps if
it be more convenient he will move out of the way. At Vanua Lava, in
Port Patteson, a man would not follow his mother-in-law along the
beach, nor she him, until the tide had washed out the footsteps of
the first traveller from the sand. At the same time a man and his
mother-in-law will talk at a distance.”87.1
It seems obvious that these Melanesian
It is significant that customs of avoidance are the same, and must be
mutual avoidance
between blood explained in the same way whether the woman
relations of opposite whom a man shuns is his wife’s mother or his own
sexes begins at or mother or his sister. Now it is highly significant that
near puberty.
just as among the Akamba of East Africa the
mutual avoidance of father and daughter only begins when the girl
has reached puberty, so among the Melanesians the mutual
avoidance of a boy on the one side and of his mother and sisters on
the other only begins when the boy has reached or approached
puberty. Thus in both peoples the avoidance between the nearest
blood relations only commences at the dangerous age when sexual
connexion on both sides begins to be possible. It seems difficult,
therefore, to evade the conclusion that the mutual avoidance is
adopted for no other reason than to diminish as far as possible the
chances of sexual unions which public opinion condemns as
incestuous. But if that is the reason why a young Melanesian boy, on
the verge of puberty, avoids his own mother and sisters, it is natural
and almost necessary to infer that it is the same reason which leads
him, as a full-grown and married man, to eschew the company of his
wife’s mother.
Similar customs of avoidance between mothers
Mutual avoidance of and sons, between fathers and daughters, and
mother and son, of
father and daughter, between brothers and sisters are observed by the
and of brother and natives of the Caroline Islands, and the writer who
sister in the records them assigns the fear of incest as the
Caroline Islands.
motive for their observance. “The prohibition of
marriage,” he says, “and of sexual intercourse between kinsfolk of
the same tribe is regarded by the Central Caroline natives as a
divine ordinance; its breach is therefore, in their opinion, punished by
the higher powers with sickness or death. The law influences in a
characteristic way the whole social life of the islanders, for efforts are
made to keep members of families of different sexes apart from each
other even in their youth. Unmarried men and boys, from the time
when they begin to speak, may therefore not remain by night in the
huts, but must sleep in the fel, the assembly-house. In the evening
their meal (âkot) is brought thither to them by their mothers or
sisters. Only when a son is sick may his mother receive him in the
hut and tend him there. On the other hand entrance to the assembly-
house (fel) is forbidden to women and girls except on the occasion of
the pwarik festival; whereas female members of other tribes are free
to visit it, although, so far as I could observe, they seldom make use
of the permission. Unmarried girls sleep in the huts with their
parents.
“These restrictions, which custom and tradition have instituted
within the family, find expression also in the behaviour of the
members of families toward each other. The following persons,
namely, have to be treated with respect—the daughters by their
father, the sons by their mother, the brothers by their sisters. In
presence of such relations, as in the presence of a chief, you may
not stand, but must sit down; if you are obliged on narrow paths to
pass by one of them you must first obtain permission and then do it
in a stooping or creeping posture. You allow them everywhere to go
in front; you also avoid to drink out of the vessel which they have just
used; you do not touch them, but keep always at a certain distance
from them; the head especially is deemed sacred.”88.1
In all these cases the custom of mutual
Mutual avoidance of avoidance is observed by persons of opposite sex
male and female
cousins in some who, though physically capable of sexual union,
tribes. are forbidden by tradition and public opinion to
have any such commerce with each other. Thus
far the blood relations whom a man is forbidden to marry and
compelled to avoid, are his own mother, his own daughter, and his
own sisters. But to this list some people add a man’s female cousins
or at least certain of them; for many races draw a sharp line of
distinction between cousins according as they are children of two
brothers or of two sisters or of a brother and a sister, and while they
permit or even prefer marriage with certain cousins, they absolutely
forbid marriage with certain others. Now, it is highly significant that
some tribes which forbid a man to marry certain of his cousins also
compel him to adopt towards them the same attitude of social
reserve which in the same or other tribes a man is obliged to
observe towards his wife’s mother, his own mother, and his own
sisters, all of whom in like manner he is forbidden to marry. Thus
among the tribes in the central part of New Ireland
Mutual avoidance of (New Mecklenburg) a male and a female cousin,
male and female
cousins in New the children of a brother and a sister respectively,
Ireland. are most strictly forbidden by custom to marry
each other; indeed this prohibition is described as
the most stringent of all; the usual saying in regard to such relations
is, “The cousin is holy” (i tábu ra kókup). Now, in these tribes a man
is not merely forbidden to marry his female cousin, the daughter of
his father’s sister or of his mother’s brother; he must also avoid her
socially, just as in other tribes a man must avoid his wife’s mother,
his own mother, his own daughter, and his own sisters. The cousins
may not approach each other, they may not shake hands or even
touch each other, they may not give each other presents, they may
not mention each other’s names; but they are allowed to speak to
each other at a distance of some paces. These rules of avoidance,
these social barriers erected between cousins, the children of a
brother and a sister respectively, are interpreted most naturally and
simply as precautions intended to obviate the danger of a criminal
intercourse between persons whose sexual union would be regarded
by public opinion with deep displeasure. Indeed the Catholic
missionary, to whom we are indebted for the information, assumes
this interpretation of the rules as if it were too obvious to call for
serious discussion. He says that all the customs of avoidance “are
observed as outward symbols of this prohibition of marriage”; and he
adds that “were the outward sign of the prohibition of marriage, to
which the natives cleave with genuine obstinacy, abolished or even
weakened, there would be an immediate danger of the natives
contracting such marriages.”90.1 It seems difficult for a rational man
to draw any other inference. If any confirmation were needed, it
would be furnished by the fact that among these tribes of New
Ireland brothers and sisters are obliged to observe precisely the
same rules of mutual avoidance, and that incest between brother
and sister is a crime which is punished with hanging; they may not
come near each other, they may not shake hands, they may not
touch each other, they may not give each other presents; but they
are allowed to speak to each other at a distance of some paces. And
the penalty for incest with a daughter is also death by hanging.90.2
Amongst the Baganda of Central Africa in like
Mutual avoidance of manner a man was forbidden under pain of death
certain male and
female cousins to marry or have sexual intercourse with his
among the cousin, the daughter either of his father’s sister or
Baganda; marriage of his mother’s brother; and such cousins might
or sexual
intercourse not approach each other, nor hand each other
forbidden between anything, nor enter the same house, nor eat out of
these cousins under
pain of death.
the same dish. Were cousins to break these rules
of social avoidance, in other words, if they were to
approach each other or hand each other anything, it was believed
that they would fall ill, that their hands would tremble, and that they
would be unfit for any work.90.3 Here, again, the prohibition of social
intercourse was in all probability merely a precaution against sexual
intercourse, for which the penalty was death. And the same may be
said of the similar custom of avoidance which among these same
Baganda a man had to observe towards his wife’s mother. “No man
might see his mother-in-law, or speak face to face with her; she
covered her face, if she passed her son-in-law, and he gave her the
path and made a detour, if he saw her coming. If she was in the
house, he might not enter, but he was allowed to speak to her from a
distance. This was said to be because he had seen her daughter’s
nakedness. If a son-in-law accidentally saw his mother-in-law’s
breasts, he sent her a barkcloth in compensation, to cover herself,
lest some illness, such as tremor, should come upon him. The
punishment for incest was death; no member of a clan would shield
a person guilty thereof; the offender was disowned by the clan, tried
by the chief of the district, and put to death.”91.1
The prohibition of marriage with certain cousins
Marriage between appears to be widespread among African peoples
certain cousins
forbidden among of the Bantu stock. Thus in regard to the Bantus of
some South African South Africa we read that “every man of a coast
tribes but allowed tribe regarded himself as the protector of those
among others.
females whom we would call his cousins, second
cousins, third cousins, and so forth, on the father’s side, while some
had a similar feeling towards the same relatives on the mother’s side
as well, and classified them all as sisters. Immorality with one of
them would have been considered incestuous, something horrible,
something unutterably disgraceful. Of old it was punished by the
death of the male, and even now a heavy fine is inflicted upon him,
while the guilt of the female must be atoned by a sacrifice performed
with due ceremony by the tribal priest, or it is believed a curse will
rest upon her and her issue.… In contrast to this prohibition the
native of the interior almost as a rule married the daughter of his
father’s brother, in order, as he said, to keep property from being lost
to his family. This custom more than anything else created a disgust
and contempt for them by the people of the coast, who term such
intermarriages the union of dogs, and attribute to them the insanity
and idiocy which in recent times has become prevalent among the
inland tribes.”91.2
Among the Thonga, a Bantu tribe about
Marriage between Delagoa Bay, marriages between cousins are as a
cousins allowed in
some African tribes rule prohibited, and it is believed that such unions
on condition that an are unfruitful. However, custom permits cousins to
expiatory sacrifice marry each other on condition that they perform an
is offered.
expiatory ceremony which is supposed to avert the
curse of barrenness from the wife. A goat is sacrificed, and the
couple are anointed with the green liquid extracted from the half-
digested grass in the animal’s stomach. Then a hole is cut in the
goat’s skin and through this hole the heads of the cousins are
inserted. The goat’s liver is then handed to them, quite raw, through
the hole in the skin, and they must tear it out with their teeth without
using a knife. Having torn it out, they eat it. The word for liver
(shibindji) also means “patience,” “determination.” So they say to the
couple, “You have acted with strong determination. Eat the liver now!
Eat it in the full light of the day, not in the dark! It will be an offering to
the gods.” Then the family priest prays, saying: “You, our gods, So-
and-so, look! We have done it in the daylight. It has not been done
by stealth. Bless them, give them children!” When he has done
praying, the assistants take all the half-digested grass from the
goat’s stomach and place it on the wife’s head, saying, “Go and bear
children!”92.1 Among the Wagogo of German East Africa marriage is
forbidden between cousins who are the children of two brothers or of
two sisters, but is permitted between cousins who are the children of
a brother and sister respectively. However, in this case it is usual for
the wife’s father to kill a sheep and put on a leather armlet, made
presumably from the sheep’s skin; otherwise it is supposed that the
marriage would be unfruitful.92.2 Thus the Wagogo, like the Thonga,
imagine that the marriage of cousins is doomed to infertility unless
an expiatory sacrifice is offered and a peculiar use made of the
victim’s skin. Again, the Akikuyu of British East Africa forbid the
marriage of cousins and second cousins, the children and
grandchildren of brothers and sisters. If such persons married, they
would commit a grave sin, and all their children would surely die; for
the curse or ceremonial pollution (thahu) incurred by such a crime
cannot be purged away. Nevertheless it sometimes happens that a
man unwittingly marries a first or second cousin; for instance, if a
part of the family moves away to another district, it may come about
that a man makes the acquaintance of a girl and marries her before
he discovers the relationship. In such a case, where the sin has
been committed unknowingly, the curse can be averted by the
performance of an expiatory rite. The elders take a sheep and place
it on the woman’s shoulders; there it is killed and the intestines taken
out. Then the elders solemnly sever the intestines with a sharp
splinter of wood taken from a bush of a certain sort (mukeo), “and
they announce that they are cutting the clan kutinyarurira, by which
they mean that they are severing the bond of relationship which
exists between the pair. A medicine man then comes and purifies the
couple.”93.1 In all these cases we may assume with a fair degree of
probability that the old prohibition of marriage between cousins is
breaking down, and that the expiatory sacrifice offered when such a
marriage does take place is merely a salve to the uneasy conscience
of those who commit or connive at a breach of the ancient taboo.
Thus the prohibition of marriage between
The mutual cousins, and the rules of ceremonial avoidance
avoidance of male
and female cousins observed in some tribes between persons who
is probably a stand in that relationship to each other, appear
precaution against both to spring from a belief, right or wrong, in the
a criminal intimacy
between them. injurious effects of such unions and from a desire
to avoid them. The mutual avoidance of the
cousins is merely a precaution to prevent a closer and more criminal
intimacy between them. If that is so, it furnishes a confirmation of the
view that all the customs of ceremonial avoidance between blood
relations or connexions by marriage of opposite sexes are based
simply on a fear of incest.
The theory is perhaps confirmed by the
The mutual observation that in some tribes the avoidance
avoidance between
a man and his between a man and his wife’s mother lasts only
wife’s relations until he has had a child by his wife;94.1 while in
seems to be partly
grounded on a fear others, though avoidance continues longer, it
of rendering the gradually wears away with time as the man and
wife infertile.
woman advance in years,94.2 and in others, again,
it is observed only between a man and his future mother-in-law, and
comes to an end with his marriage.94.3 These customs suggest that
in the minds of the people who practise them there is a close
connexion between the avoidance of the wife’s relations and the
dread of an infertile marriage. The Indians of Yucatan, as we saw,
believe that if a betrothed man were to meet his future mother-in-law
or father-in-law, he would thereby lose the power of begetting
children. Such a fear seems to be only an extension by false analogy
of that belief in the disastrous consequences of illicit sexual relations
which we dealt with in an earlier part of this chapter,94.4 and of which
we shall have more to say presently.94.5 From thinking, rightly or
wrongly, that sexual intercourse between certain persons is fraught
with serious dangers, the savage jumped to the conclusion that
social intercourse between them may be also perilous by virtue of a
sort of physical infection acting through simple contact or even at a
distance; or if, in many cases, he did not go so far as to suppose that
for a man merely to see or touch his mother-in-law sufficed to blast
the fertility of his wife’s womb, yet he may have thought, with much
better reason, that intimate social converse between him and her
might easily lead to something worse, and that to guard against such
a possibility it was best to raise a strong barrier of etiquette between
them. It is not, of course, to be supposed that these rules of
avoidance were the result of deliberate legislation; rather they were
the spontaneous and gradual growth of feelings and thoughts of
which the savages themselves perhaps had no clear consciousness.
In what precedes I have merely attempted to sum up in language
intelligible to civilized man the outcome of a long course of moral and
social evolution.
These considerations perhaps obviate to some extent the only
serious difficulty which lies in the way of the theory here advocated.
If the custom of avoidance was adopted in order to
The mutual guard against the danger of incest, how comes it
avoidance between
persons of the that the custom is often observed towards persons
same sex was of the same sex, for example, by a man towards
probably an his father-in-law as well as towards his mother-in-
extension by false
analogy of the law? The difficulty is undoubtedly serious: the only
mutual avoidance way of meeting it that I can suggest is the one I
between persons of
different sexes.
have already indicated. We may suppose that the
deeply rooted beliefs of the savage in the fatal
effects of marriage between certain classes of persons, whether
relations by blood or connexions by marriage, gradually spread in his
mind so as to embrace the relations between men and men as well
as between men and women; till he had worked himself into the
conviction that to see or touch his father-in-law, for example, was
nearly or quite as dangerous as to touch or have improper relations
with his mother-in-law. It is no doubt easy for us to detect the flaw in
this process of reasoning; but we should beware of casting stones at
the illogical savage, for it is possible or even probable that many of
our own cherished convictions are no better founded.
Viewed from this standpoint the customs of
The custom of ceremonial avoidance among savages assume a
mutual avoidance
between near serious aspect very different from the appearance
relations has of arbitrariness and absurdity which they are apt to
probably had the present to the civilized observer who does not look
effect of checking
the practice of below the surface of savage society. So far as
inbreeding. these customs have helped, as they probably
have done, to suppress the tendency to
inbreeding, that is, to the marriage of near relations, we must
conclude that their effect has been salutary, if, as many eminent
biologists hold, long-continued inbreeding is injurious to the stock,
whether animal or vegetable, by rendering it in the end infertile.95.1
However, men of science are as yet by no means agreed as to the
results of consanguineous marriages, and a living authority on the
subject has recently closed a review of the evidence as follows:
“When we take into account such evidence as there is from animals
and plants, and such studies as those of Huth,95.2 and the instances
and counter-instances of communities with a high degree of
consanguinity, we are led to the conclusion that the prejudices and
laws of many peoples against the marriage of near kin rest on a
basis not so much biological as social.”96.1 Whatever may be the
ultimate verdict of science on this disputed question, it will not affect
the result of the present enquiry, which merely affirms the deep and
far-reaching influence which in the long course of human history
superstition has exercised on morality. Whether the influence has on
the whole been for good or evil does not concern us. It suffices for
our purpose to shew that superstition has been a crutch to morality,
whether to support it in the fair way of virtue or to precipitate it into
the miry pit of vice. To return to the point from which we wandered
into this digression, we must leave in suspense the question whether
the Australian savages were wise or foolish who forbade a man
under pain of death to speak to his mother-in-law.
I will conclude this part of my subject with a few
Other examples of more instances of the extreme severity with which
the severe
punishment of certain races have visited what they deemed
sexual crime. improper connexions between the sexes.
Among the Indians who inhabited the coast of
The Indians of Brazil near Rio de Janeiro about the middle of the
Brazil. sixteenth century, a married woman who gave
birth to an illegitimate child was either killed or
abandoned to the caprice of the young men who could not afford to
keep a wife. Her child was buried alive; for they said that were he to
grow up he would only serve to perpetuate his mother’s disgrace; he
would not be allowed to go to war with the rest for fear of the
misfortunes and disasters he might draw down upon them, and no
one would eat any food, whether venison, fish, or what not, which
the miserable outcast had touched.96.2 In Ruanda,
The natives of
Ruanda.
a district of Central Africa, down to recent years
any unmarried woman who was got with child
used to be put to death with her baby, whether born or unborn. A
spot at the mouth of the Akanyaru river was the place of execution,
where the guilty women and their innocent offspring were hurled into
the water. As usual, this Puritanical strictness of morality has been
relaxed under European influence; illegitimate children are still killed,
but their mothers escape with the fine of a cow.97.1
The Saxons.
Among the Saxons down to the days of St.
Boniface the adulteress or the maiden who had dishonoured her
father’s house was compelled to hang herself, was burned, and her
paramour hung over the blazing pile; or she was scourged or cut to
pieces with knives by all the women of the village till she was
dead.97.2 Among the Slav peoples of the Balkan
The Southern
Slavs.
peninsula women convicted of immoral conduct
used to be stoned to death. About the year 1770 a
young betrothed couple were thus executed near Cattaro in
Dalmatia, because the girl was found to be with child. The youth
offered to marry her, and the priest begged that the sentence of
death might be commuted to perpetual banishment; but the people
declared that they would not have a bastard born among them; and
the two fathers of the luckless couple threw the first stones at them.
When Miss M. Edith Durham related this case to some Montenegrin
peasantry, they all said that in the old days stoning was the proper
punishment for unchaste women; the male paramours were shot by
the relations of the girls whom they had seduced. When “that
modern Messalina,” Queen Draga of Servia, was murdered, a
decent peasant woman remarked that “she ought to be under the
cursed stone heap” (pod prokletu gomilu). The country-folk of
Montenegro, who heard the news of the murder from Miss Durham,
“looked on it as a cleansing—a casting out of abominations—and
genuinely believed that Europe would commend the deed, and that
the removal of this sinful woman would bring prosperity to the
land.”97.3 Even down to the second half of the nineteenth century in
cases of seduction among the Southern Slavs the people proposed
to stone both the culprits to death.98.1 This happened, for example, in
Herzegovina in the year 1859, when a young man named Milutin
seduced or (to be more exact) was seduced by three unmarried girls
and got them all with child. The people sat in judgment upon the
sinners, and, though an elder proposed to stone them all, the court
passed a milder sentence. The young man was to marry one of the
girls, to rear the infants of the other two as his legitimate children,
and next time there was a fight with the Turks he was to prove his
manhood by rushing unarmed upon the enemy and wresting their
weapons from them, alive or dead. The sentence was fulfilled to the
letter, though many years passed before the culprit could carry out
the last part of it. However, his time came in 1875, when
Herzegovina revolted against the Turks. Then Milutin ran unarmed
upon a regiment of the enemy and found among the Turkish
bayonets a hero’s death.98.2 Even now the Old Catholics among the
South Slavs believe that a village in which a seducer is not
compelled to marry his victim will be punished with hail and
excessive rain. For this article of faith, however, they are ridiculed by
their enlightened Catholic neighbours, who hold the far more
probable view that thunder and lightning are caused by the village
priest to revenge himself for unreasonable delays in the payment of
his salary. A heavy hail-storm has been known to prove almost fatal
to the local incumbent, who was beaten within an inch of his life by
his enraged parishioners.98.3
It is difficult to believe that in these and similar
Inference from the cases the community would inflict such severe
severe punishments
inflicted for sexual punishment for sexual offences if it did not believe
offences. that its own safety, and not merely the interest of a
few individuals, was imperilled thereby.
If now we ask why illicit relations between the
Why should illicit sexes should be supposed to disturb the balance
relations between
the sexes be of nature and particularly to blast the fruits of the
thought to disturb earth, a partial answer may be conjecturally
the balance of suggested. It is not enough to say that such
nature?
relations are displeasing to the gods, who punish
indiscriminately the whole community for the sins of a few. For we
must always bear in mind that the gods are creations of man’s fancy;
he fashions them in human likeness, and endows them with tastes
and opinions which are merely vast cloudy projections of his own. To
affirm, therefore, that something is a sin because the gods will it so,
is only to push the enquiry one stage farther back and to raise the
further question, Why are the gods supposed to dislike and punish
these particular acts? In the case with which we
The reason why the are here concerned, the reason why so many
gods of savages
are supposed to savage gods prohibit adultery, fornication, and
punish sexual incest under pain of their severe displeasure may
crimes so severely
may perhaps be perhaps be found in the analogy which many
found in a mistaken savage men trace between the reproduction of the
belief that human species and the reproduction of animals
irregularities of the
human sexes and plants. The analogy is not purely fanciful, on
prevent the the contrary it is real and vital; but primitive
reproduction of peoples have given it a false extension in a vain
edible animals and
plants and thereby attempt to apply it practically to increasing the food
supply. They have imagined, in fact, that by
strike a fatal blow at
the food supply. performing or abstaining from certain sexual acts
they thereby directly promoted the reproduction of
animals and the multiplication of plants.99.1 All such acts and
abstinences, it is obvious, are purely superstitious and wholly fail to
effect the desired result. They are not religious but magical; that is,
they compass their end, not by an appeal to the gods, but by
manipulating natural forces in accordance with certain false ideas of
physical causation. In the present case the principle on which
savages seek to propagate animals and plants is that of magical
sympathy or imitation: they fancy that they assist the reproductive
process in nature by mimicking or performing it among themselves.
Now in the evolution of society such efforts to control the course of
nature directly by means of magical rites appear to have preceded
the efforts to control it indirectly by appealing to the vanity and
cupidity, the good-nature and pity of the gods; in short, magic seems
to be older than religion.100.1 In most races, it is true, the epoch of
unadulterated magic, of magic untinged by religion, belongs to such
a remote past that its existence, like that of our ape-like ancestors,
can be a matter of inference only; almost everywhere in history and
the world we find magic and religion side by side, at one time allies,
at another enemies, now playing into each other’s hands, now
cursing, objurgating, and vainly attempting to exterminate one
another. On the whole the lower intelligences cling closely, though
secretly, to magic, while the higher intelligences have discerned the
vanity of its pretensions and turned to religion instead. The result has
been that beliefs and rites which were purely magical in origin often
contract in course of time a religious character; they are modified in
accordance with the advance of thought, they are translated into
terms of gods and spirits, whether good and beneficent, or evil and
malignant. We may surmise, though we cannot prove, that a change
of this sort has come over the minds of many races with regard to
sexual morality. At some former time, perhaps, straining a real
analogy too far, they believed that those relations of the human
sexes which for any reason they regarded as right and natural had a
tendency to promote sympathetically the propagation of animals and
plants and thereby to ensure a supply of food for the community;
while on the contrary they may have imagined that those relations of
the human sexes which for any reason they deemed wrong and
unnatural had a tendency to thwart and impede the propagation of
animals and plants and thereby to diminish the common supply of
food.
Such a belief, it is obvious, would furnish a
Such a belief would sufficient motive for the strict prohibition of what
account both for the
horror with which were deemed improper relations between men
many savages and women; and it would explain the deep horror
regard such crimes, and detestation with which sexual irregularities are
and for the severity
with which they viewed by many, though certainly not by all,
punish them. savage tribes. For if improper relations between
the human sexes prevent animals and plants from
multiplying, they strike a fatal blow at the existence of the tribe by
cutting off its supply of food at the roots. No wonder, therefore, that
wherever such superstitions have prevailed the whole community,
believing its very existence to be put in jeopardy by sexual
immorality, should turn savagely on the culprits, and beat, burn,
drown or otherwise exterminate them in order to rid itself of so
dangerous a pollution. And when with the advance of knowledge
men began to perceive the mistake they had made in imagining that
the commerce of the human sexes could affect the propagation of
animals and plants, they would still through long habit be so inured
to the idea of the wickedness of certain sexual relations that they
could not dismiss it from their minds, even when they discerned the
fallacious nature of the reasoning by which they had arrived at it. The
old practice would therefore stand, though the old theory had fallen:
the old rules of sexual morality would continue to be observed, but if
they were to retain the respect of the community, it was necessary to

Metric Structures and Fixed Point Theory 1St Edition Dhananjay Gopal Online Ebook Texxtbook Full Chapter PDF

Uploaded by

Copyright:

Available Formats

You might also like

Metric Structures and Fixed Point Theory 1St Edition Dhananjay Gopal Online Ebook Texxtbook Full Chapter PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Metric Structures and Fixed Point Theory 1St Edition Dhananjay Gopal Online Ebook Texxtbook Full Chapter PDF

Uploaded by

Copyright:

Available Formats

Metric Structures and Fixed Point

Theory 1st Edition Dhananjay Gopal

Metric Fixed Point Theory: Applications in Science,

Fixed Point Theory and Fractional Calculus Recent

Optimization on Solution Sets of Common Fixed Point

FRP Composite Structures: Theory, Fundamentals, and

Multivariable Calculus Metric Edition Stewart

Calculus Metric Edition James Stewart

Photovoltaic Thermal Passive House System 1st Edition

Breaking Point Book I Of Breaking Point Trilogy 1st

1 Symmetric Spaces and Fixed Point Theory . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1

2 Fixed Point Theory in b-Metric Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 33

3 Basics of w-Distance and Its Use in Various Types of Results . . . . . . . . . . . . . . . . . . . . . . 67

4 G-Metric Spaces: From the Perspective of F-Contractions and Best

5 Fixed Point Theory in Probabilistic Metric Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 149

6 Fixed Point Theory For Fuzzy Contractive Mappings . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 199

7 Set-Valued Maps and Inclusion Problems in Modular Metric Spaces. . . . . . . . . . . . . 245

8 Graphical Metric Spaces and Fixed Point Theorems . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 267

9 Fixed Point Theory in Partial Metric Spaces . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 283

Dhananjay Gopal has a doctorate in Mathematics from Guru

Dr. P. Agarwal earned his Master’s degree in Mathematics

Poom Kumam received his Ph.D. in Mathematics from

Tatjana Došenović Juan Martnez-Moreno

Pradip Ramesh Patle and Deepesh Kumar Patel

1.2 Basic Concepts

Definition 1 ([52]). A function s : X × X → [0, +∞) satisfying

(W1) s(x, y) = 0 if and only if x = y and s(x, y) ≥ 0, for x, y ∈ X,

Example 3. Let X = { n1 } ∪ {0} and let

Example 5. Let X = { n1 } ∪ {0} and let

s( n1 , 0) = 1 = s(1, n1 ) and s(x, x) = 0 for all x ∈ X. Then (X, s) is a symmetric space.

N (X) = {P : P ⊆ X and P = 0};

S (A, B) = max sup s(a, B), sup s(b, A) (1.1)

where s(a, B) = infs(a, b). S is known as the Pompeiu–Hausdorff distance.

Proposition 1. [41] (C B s (X), S ) is a symmetric space if (X, s) is a symmetric space.

Proposition 2. [44] (C B s (X), S + ) is a symmetric space if (X, s) is a symmetric space.

Remark 1. [43] S + and S are topologically equivalent, that is,

k1 S (P, Q) ≤ S + (P, Q) ≤ k2 S (P, Q),

1.3 Topology of Symmetric Spaces

Definition 2. A topological space (X, τ ) is semimetrizable if there is a symmetric function

1. (X, τs ) is first countable;

Definition 4. Let (X, s) be a symmetric space. We have the following properties:

(i) (W) =⇒ (W4) =⇒ (W3) but (W3) (W4).

1. (X, s) satisfies property (JMS).

s(x, z) + s(z, y) < δ implies that s(x, y) < ε .

3. There exists r > 0 such that sup{diam(B(x, r)) : x ∈ X} < ∞.

Remark 2. (i) It is obvious that (CC) implies (SC).

The continuity of a function is generalized with the following two notions.

Definition 5. A function f : X → X is s-continuous if lim s(xn , x) = 0 implies lim s(fxn , fx) = 0.

Definition 6. A function f : X → X is τs -continuous if lim xn = x with respect to τs implies

(a) X is regular with a strong triangle function Ψ .

1.4 Completeness Concepts

Definition 8. Let (X, s) be a symmetric space. A sequence {xn } in X is called an s-Cauchy

Definition 9. A symmetric space (X, s) is said to be s-Cauchy complete if every s-Cauchy

The idea of completeness in s-complete topological spaces generalizes the completeness

The concept of s-weakly completeness of symmetric spaces is defined by Galvin and

Proposition 7. Let (X, s) be a semi-metric space. Then following are equivalent:

(1) (X, τs ) is s-weakly complete,

(2) every s-Cauchy sequence in X has a convergent subsequence,

N (X) = {P : P ⊆ X and P = 0};

(iv) X is (∑, =)-semicomplete.

which is a contradiction. Therefore there exists a fixed point of T.