Visit https://ebookluna.

com to download the full version and

explore more ebooks

An Introduction to Parallel Programming 2. Edition

Pacheco - eBook PDF

_____ Click the link below to download _____

https://ebookluna.com/download/an-introduction-to-
parallel-programming-ebook-pdf/

Explore and download more ebooks at ebookluna.com

Here are some recommended products that might interest you.
You can download now and explore!

(eBook PDF) Introduction to Programming Using Python An 1

https://ebookluna.com/product/ebook-pdf-introduction-to-programming-
using-python-an-1/

ebookluna.com

Parallel programming: concepts and practice 1st Edition -

eBook PDF

https://ebookluna.com/download/parallel-programming-concepts-and-
practice-ebook-pdf/

ebookluna.com

(eBook PDF) Java: An Introduction to Problem Solving and

Programming 7th Edition

https://ebookluna.com/product/ebook-pdf-java-an-introduction-to-
problem-solving-and-programming-7th-edition/

ebookluna.com

(eBook PDF) Java: An Introduction to Problem Solving and

Programming 8th Edition

https://ebookluna.com/product/ebook-pdf-java-an-introduction-to-
problem-solving-and-programming-8th-edition/

ebookluna.com
(eBook PDF) Microsoft Visual C#: An Introduction to
Object-Oriented Programming 7th Edition

https://ebookluna.com/product/ebook-pdf-microsoft-visual-c-an-
introduction-to-object-oriented-programming-7th-edition/

ebookluna.com

Python Programming: An Introduction to Computer Science

3rd Edition by John Zelle (eBook PDF)

https://ebookluna.com/product/python-programming-an-introduction-to-
computer-science-3rd-edition-by-john-zelle-ebook-pdf/

ebookluna.com

(eBook PDF) Introduction to Programming Using Visual Basic

10th Edition

https://ebookluna.com/product/ebook-pdf-introduction-to-programming-
using-visual-basic-10th-edition/

ebookluna.com

Introduction to Java Programming, Comprehensive Version

10th edition- eBook PDF

https://ebookluna.com/download/introduction-to-java-programming-
comprehensive-version-ebook-pdf/

ebookluna.com

(eBook PDF) Introduction to Java Programming, Brief

Version, Global Edition 11th Edition

https://ebookluna.com/product/ebook-pdf-introduction-to-java-
programming-brief-version-global-edition-11th-edition/

ebookluna.com
An Introduction to Parallel
Programming

SECOND EDITION

Peter S. Pacheco
University of San Francisco

Matthew Malensek
University of San Francisco
Table of Contents

Cover image

Title page

Copyright

Dedication

Preface

Chapter 1: Why parallel computing

1.1. Why we need ever-increasing performance

1.2. Why we're building parallel systems

1.3. Why we need to write parallel programs

1.4. How do we write parallel programs?

1.5. What we'll be doing

1.6. Concurrent, parallel, distributed

1.7. The rest of the book

 1.8. A word of warning

1.9. Typographical conventions

1.10. Summary

1.11. Exercises

Bibliography

Chapter 2: Parallel hardware and parallel software

2.1. Some background

2.2. Modifications to the von Neumann model

2.3. Parallel hardware

2.4. Parallel software

2.5. Input and output

2.6. Performance

2.7. Parallel program design

2.8. Writing and running parallel programs

2.9. Assumptions

2.10. Summary

2.11. Exercises

Bibliography

Chapter 3: Distributed memory programming with MPI

 3.1. Getting started

3.2. The trapezoidal rule in MPI

3.3. Dealing with I/O

3.4. Collective communication

3.5. MPI-derived datatypes

3.6. Performance evaluation of MPI programs

3.7. A parallel sorting algorithm

3.8. Summary

3.9. Exercises

3.10. Programming assignments

Bibliography

Chapter 4: Shared-memory programming with Pthreads

4.1. Processes, threads, and Pthreads

4.2. Hello, world

4.3. Matrix-vector multiplication

4.4. Critical sections

4.5. Busy-waiting

4.6. Mutexes

4.7. Producer–consumer synchronization and semaphores

4.8. Barriers and condition variables

 4.9. Read-write locks

4.10. Caches, cache-coherence, and false sharing

4.11. Thread-safety

4.12. Summary

4.13. Exercises

4.14. Programming assignments

Bibliography

Chapter 5: Shared-memory programming with OpenMP

5.1. Getting started

5.2. The trapezoidal rule

5.3. Scope of variables

5.4. The reduction clause

5.5. The parallel for directive

5.6. More about loops in OpenMP: sorting

5.7. Scheduling loops

5.8. Producers and consumers

5.9. Caches, cache coherence, and false sharing

5.10. Tasking

5.11. Thread-safety

5.12. Summary
 5.13. Exercises

5.14. Programming assignments

Bibliography

Chapter 6: GPU programming with CUDA

6.1. GPUs and GPGPU

6.2. GPU architectures

6.3. Heterogeneous computing

6.4. CUDA hello

6.5. A closer look

6.6. Threads, blocks, and grids

6.7. Nvidia compute capabilities and device architectures

6.8. Vector addition

6.9. Returning results from CUDA kernels

6.10. CUDA trapezoidal rule I

6.11. CUDA trapezoidal rule II: improving performance

6.12. Implementation of trapezoidal rule with warpSize thread

blocks

6.13. CUDA trapezoidal rule III: blocks with more than one warp

6.14. Bitonic sort

6.15. Summary
 6.16. Exercises

6.17. Programming assignments

Bibliography

Chapter 7: Parallel program development

7.1. Two n-body solvers

7.2. Sample sort

7.3. A word of caution

7.4. Which API?

7.5. Summary

7.6. Exercises

7.7. Programming assignments

Bibliography

Chapter 8: Where to go from here

Bibliography

Bibliography

Bibliography

Index
Copyright
Morgan Kaufmann is an imprint of Elsevier
50 Hampshire Street, 5th Floor, Cambridge, MA 02139,
United States

Copyright © 2022 Elsevier Inc. All rights reserved.

No part of this publication may be reproduced or

transmitted in any form or by any means, electronic or
mechanical, including photocopying, recording, or any
information storage and retrieval system, without
permission in writing from the publisher. Details on how to
seek permission, further information about the Publisher's
permissions policies and our arrangements with
organizations such as the Copyright Clearance Center and
the Copyright Licensing Agency, can be found at our
website: www.elsevier.com/permissions.

This book and the individual contributions contained in it

are protected under copyright by the Publisher (other than
as may be noted herein).
Cover art: “seven notations,” nickel/silver etched plates,
acrylic on wood structure, copyright © Holly Cohn

Notices
Knowledge and best practice in this field are constantly
changing. As new research and experience broaden our
understanding, changes in research methods,
professional practices, or medical treatment may become
necessary.
Practitioners and researchers must always rely on their
own experience and knowledge in evaluating and using
any information, methods, compounds, or experiments
described herein. In using such information or methods
they should be mindful of their own safety and the safety
of others, including parties for whom they have a
professional responsibility.

To the fullest extent of the law, neither the Publisher nor

the authors, contributors, or editors, assume any liability
for any injury and/or damage to persons or property as a
matter of products liability, negligence or otherwise, or
from any use or operation of any methods, products,
instructions, or ideas contained in the material herein.

Library of Congress Cataloging-in-Publication Data

A catalog record for this book is available from the Library
of Congress

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the
British Library

ISBN: 978-0-12-804605-0

For information on all Morgan Kaufmann publications

visit our website at https://www.elsevier.com/books-and-
journals

Publisher: Katey Birtcher

Acquisitions Editor: Stephen Merken
Content Development Manager: Meghan Andress
Publishing Services Manager: Shereen Jameel
Production Project Manager: Rukmani Krishnan
Designer: Victoria Pearson

Typeset by VTeX
 Visit https://ebookluna.com
now to explore a diverse
collection of ebooks available
in formats like PDF, EPUB, and
MOBI, compatible with all
devices. Don’t miss the chance
to enjoy exciting offers and
quickly download high-quality
materials in just a few simple
steps!
Printed in United States of America

Last digit is the print number: 9 8 7 6 5 4 3 2 1

Dedication

To the memory of Robert S. Miller

Preface
Parallel hardware has been ubiquitous for some time
now: it's difficult to find a laptop, desktop, or server that
doesn't use a multicore processor. Cluster computing is
nearly as common today as high-powered workstations
were in the 1990s, and cloud computing is making
distributed-memory systems as accessible as desktops. In
spite of this, most computer science majors graduate with
little or no experience in parallel programming. Many
colleges and universities offer upper-division elective
courses in parallel computing, but since most computer
science majors have to take a large number of required
courses, many graduate without ever writing a
multithreaded or multiprocess program.
It seems clear that this state of affairs needs to change.
Whereas many programs can obtain satisfactory
performance on a single core, computer scientists should
be made aware of the potentially vast performance
improvements that can be obtained with parallelism, and
they should be able to exploit this potential when the need
arises.
Introduction to Parallel Programming was written to
partially address this problem. It provides an introduction
to writing parallel programs using MPI, Pthreads, OpenMP,
and CUDA, four of the most widely used APIs for parallel
programming. The intended audience is students and
professionals who need to write parallel programs. The
prerequisites are minimal: a college-level course in
mathematics and the ability to write serial programs in C.
 The prerequisites are minimal, because we believe that
students should be able to start programming parallel
systems as early as possible. At the University of San
Francisco, computer science students can fulfill a
requirement for the major by taking a course on which this
text is based immediately after taking the “Introduction to
Computer Science I” course that most majors take in the
first semester of their freshman year. It has been our
experience that there really is no reason for students to
defer writing parallel programs until their junior or senior
year. To the contrary, the course is popular, and students
have found that using concurrency in other courses is much
easier after having taken this course.
If second-semester freshmen can learn to write parallel
programs by taking a class, then motivated computing
professionals should be able to learn to write parallel
programs through self-study. We hope this book will prove
to be a useful resource for them.
The Second Edition
It has been nearly ten years since the first edition of
Introduction to Parallel Programming was published.
During that time much has changed in the world of parallel
programming, but, perhaps surprisingly, much also remains
the same. Our intent in writing this second edition has been
to preserve the material from the first edition that
continues to be generally useful, but also to add new
material where we felt it was needed.
The most obvious addition is the inclusion of a new
chapter on CUDA programming. When the first edition was
published, CUDA was still very new. It was already clear
that the use of GPUs in high-performance computing would
become very widespread, but at that time we felt that
GPGPU wasn't readily accessible to programmers with
relatively little experience. In the last ten years, that has
clearly changed. Of course, CUDA is not a standard, and
features are added, modified, and deleted with great
rapidity. As a consequence, authors who use CUDA must
present a subject that changes much faster than a
standard, such as MPI, Pthreads, or OpenMP. In spite of
this, we hope that our presentation of CUDA will continue
to be useful for some time.
Another big change is that Matthew Malensek has come
onboard as a coauthor. Matthew is a relatively new
colleague at the University of San Francisco, but he has
extensive experience with both the teaching and
application of parallel computing. His contributions have
greatly improved the second edition.
About This Book
As we noted earlier, the main purpose of the book is to
teach parallel programming in MPI, Pthreads, OpenMP, and
CUDA to an audience with a limited background in
computer science and no previous experience with
parallelism. We also wanted to make the book as flexible as
possible so that readers who have no interest in learning
one or two of the APIs can still read the remaining material
with little effort. Thus the chapters on the four APIs are
largely independent of each other: they can be read in any
order, and one or two of these chapters can be omitted.
This independence has some cost: it was necessary to
repeat some of the material in these chapters. Of course,
repeated material can be simply scanned or skipped.
On the other hand, readers with no prior experience with
parallel computing should read Chapter 1 first. This
chapter attempts to provide a relatively nontechnical
explanation of why parallel systems have come to dominate
the computer landscape. It also provides a short
introduction to parallel systems and parallel programming.
Chapter 2 provides technical background on computer
hardware and software. Chapters 3 to 6 provide
independent introductions to MPI, Pthreads, OpenMP, and
CUDA, respectively. Chapter 7 illustrates the development
of two different parallel programs using each of the four
APIs. Finally, Chapter 8 provides a few pointers to
additional information on parallel computing.
 We use the C programming language for developing our
programs, because all four API's have C-language
interfaces, and, since C is such a small language, it is a
relatively easy language to learn—especially for C++ and
Java programmers, since they will already be familiar with
C's control structures.
Classroom Use
This text grew out of a lower-division undergraduate
course at the University of San Francisco. The course
fulfills a requirement for the computer science major, and it
also fulfills a prerequisite for the undergraduate operating
systems, architecture, and networking courses. The course
begins with a four-week introduction to C programming.
Since most of the students have already written Java
programs, the bulk of this introduction is devoted to the
use pointers in C.1 The remainder of the course provides
introductions first to programming in MPI, then Pthreads
and/or OpenMP, and it finishes with material covering
CUDA.
We cover most of the material in Chapters 1, 3, 4, 5, and
6, and parts of the material in Chapters 2 and 7. The
background in Chapter 2 is introduced as the need arises.
For example, before discussing cache coherence issues in
OpenMP (Chapter 5), we cover the material on caches in
Chapter 2.
The coursework consists of weekly homework
assignments, five programming assignments, a couple of
midterms and a final exam. The homework assignments
usually involve writing a very short program or making a
small modification to an existing program. Their purpose is
to insure that the students stay current with the
coursework, and to give the students hands-on experience
with ideas introduced in class. It seems likely that their
existence has been one of the principle reasons for the
course's success. Most of the exercises in the text are
suitable for these brief assignments.
 The programming assignments are larger than the
programs written for homework, but we typically give the
students a good deal of guidance: we'll frequently include
pseudocode in the assignment and discuss some of the
more difficult aspects in class. This extra guidance is often
crucial: it's easy to give programming assignments that will
take far too long for the students to complete.
The results of the midterms and finals and the
enthusiastic reports of the professor who teaches operating
systems suggest that the course is actually very successful
in teaching students how to write parallel programs.
For more advanced courses in parallel computing, the
text and its online supporting materials can serve as a
supplement so that much of the material on the syntax and
semantics of the four APIs can be assigned as outside
reading.
The text can also be used as a supplement for project-
based courses and courses outside of computer science
that make use of parallel computation.
Support Materials
An online companion site for the book is located at
www.elsevier.com/books-and-journals/book-
companion/9780128046050.. This site will include errata
and complete source for the longer programs we discuss in
the text. Additional material for instructors, including
downloadable figures and solutions to the exercises in the
book, can be downloaded from
https://educate.elsevier.com/9780128046050.
We would greatly appreciate readers' letting us know of
any errors they find. Please send email to
[email protected] if you do find a mistake.
Acknowledgments
In the course of working on this book we've received
considerable help from many individuals. Among them we'd
like to thank the reviewers of the second edition, Steven
Frankel (Technion) and Il-Hyung Cho (Saginaw Valley State
University), who read and commented on draft versions of
the new CUDA chapter. We'd also like to thank the
reviewers who read and commented on the initial proposal
for the book: Fikret Ercal (Missouri University of Science
and Technology), Dan Harvey (Southern Oregon
University), Joel Hollingsworth (Elon University), Jens
Mache (Lewis and Clark College), Don McLaughlin (West
Virginia University), Manish Parashar (Rutgers University),
Charlie Peck (Earlham College), Stephen C. Renk (North
Central College), Rolfe Josef Sassenfeld (The University of
Texas at El Paso), Joseph Sloan (Wofford College), Michela
Taufer (University of Delaware), Pearl Wang (George Mason
University), Bob Weems (University of Texas at Arlington),
and Cheng-Zhong Xu (Wayne State University). We are also
deeply grateful to the following individuals for their
reviews of various chapters of the book: Duncan Buell
(University of South Carolina), Matthias Gobbert
(University of Maryland, Baltimore County), Krishna Kavi
(University of North Texas), Hong Lin (University of
Houston–Downtown), Kathy Liszka (University of Akron),
Leigh Little (The State University of New York), Xinlian Liu
(Hood College), Henry Tufo (University of Colorado at
Boulder), Andrew Sloss (Consultant Engineer, ARM), and
Gengbin Zheng (University of Illinois). Their comments and
suggestions have made the book immeasurably better. Of
course, we are solely responsible for remaining errors and
omissions.
Slides and the solutions manual for the first edition were
prepared by Kathy Liszka and Jinyoung Choi, respectively.
Thanks to both of them.
The staff at Elsevier has been very helpful throughout
this project. Nate McFadden helped with the development
of the text. Todd Green and Steve Merken were the
acquisitions editors. Meghan Andress was the content
development manager. Rukmani Krishnan was the
production editor. Victoria Pearson was the designer. They
did a great job, and we are very grateful to all of them.
 Our colleagues in the computer science and mathematics
departments at USF have been extremely helpful during
our work on the book. Peter would like to single out Prof.
Gregory Benson for particular thanks: his understanding of
parallel computing—especially Pthreads and semaphores—
has been an invaluable resource. We're both very grateful
to our system administrators, Alexey Fedosov and Elias
Husary. They've patiently and efficiently dealt with all of
the “emergencies” that cropped up while we were working
on programs for the book. They've also done an amazing
job of providing us with the hardware we used to do all
program development and testing.
Peter would never have been able to finish the book
without the encouragement and moral support of his
friends Holly Cohn, John Dean, and Maria Grant. He will
always be very grateful for their help and their friendship.
He is especially grateful to Holly for allowing us to use her
work, seven notations, for the cover.
Matthew would like to thank his colleagues in the USF
Department of Computer Science, as well as Maya
Malensek and Doyel Sadhu, for their love and support.
Most of all, he would like to thank Peter Pacheco for being
a mentor and infallible source of advice and wisdom during
the formative years of his career in academia.
Our biggest debt is to our students. As always, they
showed us what was too easy and what was far too difficult.
They taught us how to teach parallel computing. Our
deepest thanks to all of them.
1 “Interestingly, a number of students have said that they
found the use of C pointers more difficult than MPI
programming.”
Chapter 1: Why parallel
computing
From 1986 to 2003, the performance of microprocessors
increased, on average, more than 50% per year [28]. This
unprecedented increase meant that users and software
developers could often simply wait for the next generation
of microprocessors to obtain increased performance from
their applications. Since 2003, however, single-processor
performance improvement has slowed to the point that in
the period from 2015 to 2017, it increased at less than 4%
per year [28]. This difference is dramatic: at 50% per year,
performance will increase by almost a factor of 60 in 10
years, while at 4%, it will increase by about a factor of 1.5.
Furthermore, this difference in performance increase has
been associated with a dramatic change in processor
design. By 2005, most of the major manufacturers of
microprocessors had decided that the road to rapidly
increasing performance lay in the direction of parallelism.
Rather than trying to continue to develop ever-faster
monolithic processors, manufacturers started putting
multiple complete processors on a single integrated circuit.
This change has a very important consequence for
software developers: simply adding more processors will
not magically improve the performance of the vast majority
of serial programs, that is, programs that were written to
run on a single processor. Such programs are unaware of
the existence of multiple processors, and the performance
of such a program on a system with multiple processors
will be effectively the same as its performance on a single
processor of the multiprocessor system.
 Visit https://ebookluna.com
now to explore a diverse
collection of ebooks available
in formats like PDF, EPUB, and
MOBI, compatible with all
devices. Don’t miss the chance
to enjoy exciting offers and
quickly download high-quality
materials in just a few simple
steps!
 All of this raises a number of questions:

• Why do we care? Aren't single-processor systems

fast enough?
• Why can't microprocessor manufacturers continue
to develop much faster single-processor systems?
Why build parallel systems? Why build systems
with multiple processors?
• Why can't we write programs that will automatically
convert serial programs into parallel programs,
that is, programs that take advantage of the
presence of multiple processors?

Let's take a brief look at each of these questions. Keep in

mind, though, that some of the answers aren't carved in
stone. For example, the performance of many applications
may already be more than adequate.

1.1 Why we need ever-increasing performance

The vast increases in computational power that we've been
enjoying for decades now have been at the heart of many of
the most dramatic advances in fields as diverse as science,
the Internet, and entertainment. For example, decoding the
human genome, ever more accurate medical imaging,
astonishingly fast and accurate Web searches, and ever
more realistic and responsive computer games would all
have been impossible without these increases. Indeed,
more recent increases in computational power would have
been difficult, if not impossible, without earlier increases.
But we can never rest on our laurels. As our computational
power increases, the number of problems that we can
seriously consider solving also increases. Here are a few
examples:

• Climate modeling. To better understand climate

change, we need far more accurate computer
 models, models that include interactions between
the atmosphere, the oceans, solid land, and the ice
caps at the poles. We also need to be able to make
detailed studies of how various interventions might
affect the global climate.
• Protein folding. It's believed that misfolded proteins
may be involved in diseases such as Huntington's,
Parkinson's, and Alzheimer's, but our ability to study
configurations of complex molecules such as
proteins is severely limited by our current
computational power.
• Drug discovery. There are many ways in which
increased computational power can be used in
research into new medical treatments. For example,
there are many drugs that are effective in treating a
relatively small fraction of those suffering from some
disease. It's possible that we can devise alternative
treatments by careful analysis of the genomes of the
individuals for whom the known treatment is
ineffective. This, however, will involve extensive
computational analysis of genomes.
• Energy research. Increased computational power
will make it possible to program much more detailed
models of technologies, such as wind turbines, solar
cells, and batteries. These programs may provide
the information needed to construct far more
efficient clean energy sources.
• Data analysis. We generate tremendous amounts of
data. By some estimates, the quantity of data stored
worldwide doubles every two years [31], but the vast
majority of it is largely useless unless it's analyzed.
As an example, knowing the sequence of nucleotides
in human DNA is, by itself, of little use.
Understanding how this sequence affects
development and how it can cause disease requires
extensive analysis. In addition to genomics, huge
quantities of data are generated by particle
 colliders, such as the Large Hadron Collider at
CERN, medical imaging, astronomical research, and
Web search engines—to name a few.

These and a host of other problems won't be solved without

tremendous increases in computational power.

1.2 Why we're building parallel systems

Much of the tremendous increase in single-processor
performance was driven by the ever-increasing density of
transistors—the electronic switches—on integrated circuits.
As the size of transistors decreases, their speed can be
increased, and the overall speed of the integrated circuit
can be increased. However, as the speed of transistors
increases, their power consumption also increases. Most of
this power is dissipated as heat, and when an integrated
circuit gets too hot, it becomes unreliable. In the first
decade of the twenty-first century, air-cooled integrated
circuits reached the limits of their ability to dissipate heat
[28].
Therefore it is becoming impossible to continue to
increase the speed of integrated circuits. Indeed, in the last
few years, the increase in transistor density has slowed
dramatically [36].
But given the potential of computing to improve our
existence, there is a moral imperative to continue to
increase computational power.
How then, can we continue to build ever more powerful
computers? The answer is parallelism. Rather than building
ever-faster, more complex, monolithic processors, the
industry has decided to put multiple, relatively simple,
complete processors on a single chip. Such integrated
circuits are called multicore processors, and core has
become synonymous with central processing unit, or CPU.
In this setting a conventional processor with one CPU is
often called a single-core system.
1.3 Why we need to write parallel programs
Most programs that have been written for conventional,
single-core systems cannot exploit the presence of multiple
cores. We can run multiple instances of a program on a
multicore system, but this is often of little help. For
example, being able to run multiple instances of our
favorite game isn't really what we want—we want the
program to run faster with more realistic graphics. To do
this, we need to either rewrite our serial programs so that
they're parallel, so that they can make use of multiple
cores, or write translation programs, that is, programs that
will automatically convert serial programs into parallel
programs. The bad news is that researchers have had very
limited success writing programs that convert serial
programs in languages such as C, C++, and Java into
parallel programs.
This isn't terribly surprising. While we can write
programs that recognize common constructs in serial
programs, and automatically translate these constructs into
efficient parallel constructs, the sequence of parallel
constructs may be terribly inefficient. For example, we can
view the multiplication of two matrices as a sequence
of dot products, but parallelizing a matrix multiplication as
a sequence of parallel dot products is likely to be fairly slow
on many systems.
An efficient parallel implementation of a serial program
may not be obtained by finding efficient parallelizations of
each of its steps. Rather, the best parallelization may be
obtained by devising an entirely new algorithm.
As an example, suppose that we need to compute n
values and add them together. We know that this can be
done with the following serial code:
 Now suppose we also have p cores and . Then each
core can form a partial sum of approximately values:

Here the prefix indicates that each core is using its own,
private variables, and each core can execute this block of
code independently of the other cores.
After each core completes execution of this code, its
variable will store the sum of the values computed by
its calls to . For example, if there are eight
cores, , and the 24 calls to return the
values

1, 4, 3, 9, 2, 8, 5, 1, 1, 6, 2, 7, 2, 5, 0, 4, 1, 8, 6, 5,
1, 2, 3, 9,
then the values stored in might be

Here we're assuming the cores are identified by

nonnegative integers in the range , where p is the
number of cores.
 When the cores are done computing their values of ,
they can form a global sum by sending their results to a
designated “master” core, which can add their results:

In our example, if the master core is core 0, it would add

the values .
But you can probably see a better way to do this—
especially if the number of cores is large. Instead of making
the master core do all the work of computing the final sum,
we can pair the cores so that while core 0 adds in the result
of core 1, core 2 can add in the result of core 3, core 4 can
add in the result of core 5, and so on. Then we can repeat
the process with only the even-ranked cores: 0 adds in the
result of 2, 4 adds in the result of 6, and so on. Now cores
divisible by 4 repeat the process, and so on. See Fig. 1.1.
The circles contain the current value of each core's sum,
and the lines with arrows indicate that one core is sending
its sum to another core. The plus signs indicate that a core
is receiving a sum from another core and adding the
received sum into its own sum.
 FIGURE 1.1 Multiple cores forming a global sum.

For both “global” sums, the master core (core 0) does

more work than any other core, and the length of time it
takes the program to complete the final sum should be the
length of time it takes for the master to complete. However,
with eight cores, the master will carry out seven receives
and adds using the first method, while with the second
method, it will only carry out three. So the second method
results in an improvement of more than a factor of two. The
difference becomes much more dramatic with large
numbers of cores. With 1000 cores, the first method will
require 999 receives and adds, while the second will only
require 10—an improvement of almost a factor of 100!
The first global sum is a fairly obvious generalization of
the serial global sum: divide the work of adding among the
cores, and after each core has computed its part of the
sum, the master core simply repeats the basic serial
addition—if there are p cores, then it needs to add p values.
The second global sum, on the other hand, bears little
relation to the original serial addition.
The point here is that it's unlikely that a translation
program would “discover” the second global sum. Rather,
there would more likely be a predefined efficient global
sum that the translation program would have access to. It
could “recognize” the original serial loop and replace it
with a precoded, efficient, parallel global sum.
We might expect that software could be written so that a
large number of common serial constructs could be
recognized and efficiently parallelized, that is, modified so
that they can use multiple cores. However, as we apply this
principle to ever more complex serial programs, it becomes
more and more difficult to recognize the construct, and it
becomes less and less likely that we'll have a precoded,
efficient parallelization.
Thus we cannot simply continue to write serial programs;
we must write parallel programs, programs that exploit the
power of multiple processors.

1.4 How do we write parallel programs?

There are a number of possible answers to this question,
but most of them depend on the basic idea of partitioning
the work to be done among the cores. There are two widely
used approaches: task-parallelism and data-parallelism.
In task-parallelism, we partition the various tasks carried
out in solving the problem among the cores. In data-
parallelism, we partition the data used in solving the
problem among the cores, and each core carries out more
or less similar operations on its part of the data.
As an example, suppose that Prof P has to teach a section
of “Survey of English Literature.” Also suppose that Prof P
has one hundred students in her section, so she's been
assigned four teaching assistants (TAs): Mr. A, Ms. B, Mr. C,
and Ms. D. At last the semester is over, and Prof P makes
up a final exam that consists of five questions. To grade the
exam, she and her TAs might consider the following two
options: each of them can grade all one hundred responses
to one of the questions; say, P grades question 1, A grades
question 2, and so on. Alternatively, they can divide the one
hundred exams into five piles of twenty exams each, and
each of them can grade all the papers in one of the piles; P
grades the papers in the first pile, A grades the papers in
the second pile, and so on.
In both approaches the “cores” are the professor and her
TAs. The first approach might be considered an example of
task-parallelism. There are five tasks to be carried out:
grading the first question, grading the second question,
and so on. Presumably, the graders will be looking for
different information in question 1, which is about
Shakespeare, from the information in question 2, which is
about Milton, and so on. So the professor and her TAs will
be “executing different instructions.”
On the other hand, the second approach might be
considered an example of data-parallelism. The “data” are
the students' papers, which are divided among the cores,
and each core applies more or less the same grading
instructions to each paper.
The first part of the global sum example in Section 1.3
would probably be considered an example of data-
parallelism. The data are the values computed by
, and each core carries out roughly the same
operations on its assigned elements: it computes the
required values by calling and adds them
together. The second part of the first global sum example
might be considered an example of task-parallelism. There
are two tasks: receiving and adding the cores' partial sums,
which is carried out by the master core; and giving the
partial sum to the master core, which is carried out by the
other cores.
When the cores can work independently, writing a
parallel program is much the same as writing a serial
program. Things get a great deal more complex when the
cores need to coordinate their work. In the second global
sum example, although the tree structure in the diagram is
very easy to understand, writing the actual code is
relatively complex. See Exercises 1.3 and 1.4.
Unfortunately, it's much more common for the cores to
need coordination.
In both global sum examples, the coordination involves
communication: one or more cores send their current
partial sums to another core. The global sum examples
should also involve coordination through load balancing.
In the first part of the global sum, it's clear that we want
the amount of time taken by each core to be roughly the
same as the time taken by the other cores. If the cores are
identical, and each call to requires the same
amount of work, then we want each core to be assigned
roughly the same number of values as the other cores. If,
for example, one core has to compute most of the values,
then the other cores will finish much sooner than the
heavily loaded core, and their computational power will be
wasted.
A third type of coordination is synchronization. As an
example, suppose that instead of computing the values to
be added, the values are read from . Say, is an array
that is read in by the master core:

In most systems the cores are not automatically

synchronized. Rather, each core works at its own pace. In
this case, the problem is that we don't want the other cores
to race ahead and start computing their partial sums before
the master is done initializing and making it available to
the other cores. That is, the cores need to wait before
starting execution of the code:
 Visit https://ebookluna.com
now to explore a diverse
collection of ebooks available
in formats like PDF, EPUB, and
MOBI, compatible with all
devices. Don’t miss the chance
to enjoy exciting offers and
quickly download high-quality
materials in just a few simple
steps!
Other documents randomly have
different content
“East of the river Liao,” in a vague way to signify the territory which
lies to the left of that river; and foreign geographers, in ignorance of
the meaning of the term, had applied it to the bay into which the
river flows, which appears in atlases as the “Liaotung Gulf.” When
the intervention took place, it was probably found convenient to
make use in the Notes of protest of a term already given in foreign
atlases to the bay that forms the western boundary of the territory in
question. Hence the adoption of the term “the Liaotung peninsula,”
which was an error in geographical nomenclature. Once adopted, or,
as may be said, invented, the convenience of the term led to its
employment again when the Russo-Chinese Agreement for the lease
of Port Arthur was made in 1898, though the territory then leased
was limited to what is now known as the peninsula of Kwantung. It
reappears in the additional Russo-Chinese Agreement of the same
year. From that time the term seems to have passed into general use,
for we find it in the Portsmouth Treaty of 1905.
The intervention of the three Powers had far-reaching
consequences, none of which, in all probability, were foreseen at the
time by any of the Governments concerned, though each may have
felt that it had established a claim to the goodwill of China. Four
months after Japan had agreed to the retrocession of the territory
ceded to her by the Shimonoséki Treaty Russia, who had been the
prime mover in the matter, proceeded to lay China under further
obligations by rendering her financial assistance which facilitated the
liberation of her territory. This took the form of a Chinese loan of
£15,000,000, floated in Paris under Russia’s guarantee.
In January, 1896, one of the consequences above mentioned was
seen in the settlement of various questions which the French
Government had been pressing on the attention of the Government
of China for some time. These questions related to the rectification of
the Tonkin frontier, and to railway and mining concessions in the
provinces of Yunnan, Kwangsi and Kwantung. This was only an
instalment of the recompense for her services which France was to
obtain. The arrangement with France regarding the Tonkin frontier
constituted a breach of the Burma Convention of 1886, and of a later
Convention of 1894, regulating the boundaries separating British and
Chinese territories, which provided, inter alia, that no portion of two
small States assigned to China should be alienated to any other
Power without previous agreement with Great Britain. The dispute
which arose over this question was eventually settled—as between
Great Britain and France—by the joint Declaration of January 15th,
1896, fixing the boundary between the possessions, or spheres of
influence, of the two Powers as far as the Chinese frontier, and
arranging for all privileges conceded by China in the provinces of
Yunnan and Szechwan to the two Powers, respectively, under their
Agreements with China of 1894 and 1895 to be made common to
both Powers and their nationals; and—as between Great Britain and
China—by an Agreement signed on February 4th, 1897, modifying
the previous boundary in favour of Great Britain, and opening the
West river, which flows into the sea at Canton, to foreign trade.
Russia was the next to profit. She had already decided in 1892 to
construct what is now the Trans-Siberian Railway with the object of
linking up the eastern and western extremities of the Empire, and
thus aiding the development of Siberia, as well as strengthening her
position on the Pacific coast. The line, as then projected, was to run
from Chiliabinsk in the Ural Mountains to the south-western shore
of Lake Baikal, and from the south-eastern shore of the lake to
Vladivostok, following for some distance the course of the Amur
river; communication across the lake to be maintained by vessels
specially constructed for the purpose. Work was commenced at both
ends of the railway, and when the Shimonoséki Treaty was signed the
line had been finished as far east as Chita, a town south-east of Lake
Baikal, and within two hundred miles of the Chinese frontier.
The war between China and Japan had served a useful purpose for
Russia in revealing both the weakness of China and the strength and
ambitions of Japan. To check these ambitions in the direction of
Manchuria, and forestall Japan by establishing herself in the coveted
territory, was the task to which she now directed her energies. In the
preliminary step by which the retrocession of the Liaotung peninsula
was effected she was, as we have seen, aided by both France and
Germany. Between the latter and herself some sort of roughly
formulated understanding seems to have been arrived at, described
by Reventlow in his Deutschland’s Auswärtige Politik as a secret
agreement between the Kaiser and the Tsar, the results of which
were to be seen later. With France she worked throughout in the
closest accord in the development of the new line of policy she had
marked out for herself in the Far East, to which Belgian financiers
also lent their co-operation. In return for Russia’s support in
European affairs, as arranged by the Entente concluded between the
two countries, France, for her part, was only too willing to encourage
Russian aims in the Far East; and she was the more ready to do so,
since this course assured her of reciprocal help in the prosecution of
her own interests in China. Russia had been the connecting link
between the three Powers whose intervention had restored the
Liaotung peninsula to China. It was the relations she continued to
maintain with her two associates after that incident—in the one case
an informal understanding, in the other definite concerted action—
which shaped the course of subsequent events in the Far East.
In Ma Mission en Chine, M. Gérard, who was French Minister in
Peking during the period 1893–7, gives an account of the secret
negotiations with China by means of which Russia succeeded in
forestalling Japan in Manchuria. His book supplies the key to a
correct understanding of the course of events, and throws much light
on the political situation at the time of which he speaks. We learn
how close was the accord then maintained between France and
Russia; how skilfully Russia made use of the complaisant attitude of
her two associates; and with what unscrupulous determination to
compass her ends she traded on the weakness of China, on the claims
she had established on the latter’s goodwill, and on the vanity and
corruption of Chinese officials.
In May, 1896, according to M. Gérard, a secret Treaty was signed
at St. Petersburg by Prince Lobanoff, then Minister for Foreign
Affairs, and Li Hung Chang, Viceroy of Chihli, who had been sent to
Russia as China’s representative at the Coronation of the late Tsar
Nicholas II. The full text of this Treaty has never been published, but
it promised to China Russian protection against Japan; China, in
return for this guarantee of assistance, granting to Russia the
privilege of using, in time of war, the harbours of Ta-lien Wan, in the
Kwantung peninsula, and Kiaochow, in the province of Shantung, as
bases for her fleet. Three months later (August 27th) a secret Railway
Agreement was signed at St. Petersburg by Li Hung Chang and the
representatives of the Russo-Chinese Bank. This institution, half the
capital of which was French, had been created at the end of the
previous year. M. Gérard explains that, in consequence of so large a
portion of the bank’s capital being furnished by a French syndicate,
the French Government insisted on receiving definite information
regarding the negotiations in question. His statements regarding the
French financial interest in the Russo-Chinese Bank are confirmed
by other writers: by Chéradame, in his interesting book, Le Monde et
La Guerre Russo-Japonaise, and by Débidour in Histoire
Diplomatique de l’Europe. We learn also from M. Gérard that the
Chinese Government had contributed, under the title of a deposit,
5,000,000 taels to the capital of the bank, explaining at the time, in
answer to enquiries, that this sum represented China’s share of the
cost of construction of the Chinese Eastern Railway; that for the
building of this line a company called the Chinese Eastern Railway
Company was formed, which, although Russo-Chinese in name, was
a purely Russian concern; and that it was agreed that on the
completion of the line in question the sum “deposited” by China
should be returned to her. He adds that the President of the bank
was Prince Ouchtomsky, who afterwards visited Peking at the head
of a Russian Mission.
Both the Treaty and the Railway Agreement were ratified by the
Chinese Government on the 18th September, and came into force on
that date. The popular rumour which credited the Russian Minister
in Peking with the negotiation of these two instruments was, it
appears, due to the presence of M. Cassini at the Chinese Capital,
where it was considered necessary for him to remain in order to
secure their ratification by China. As a glance at a map of North-
Eastern Asia will show, the Railway Agreement constituted a
concession of the greatest importance to Russia. The Chinese
Eastern Railway, the name of the new line which Russia obtained
leave to construct, became the eastern section of the Trans-Siberian
Railway, connecting Lake Baikal with Vladivostok, Russia’s outlet to
the Pacific. The new line, which would traverse Northern Manchuria
via Kharbin, Tsitsihar and Hailar, would shorten the distance by
more than 300 miles. Moreover, the more level country through
which the line was to pass presented few engineering difficulties, as
compared with the Amur route, a fact which would greatly diminish
the period and the cost of construction. The Agreement was
subsequently rendered complete in every detail by the elaboration of
what were termed the Statutes of the Chinese Eastern Railway. These
were confirmed by the Tsar on the 4th December in the same year.
Although these Statutes (given in Rockhill’s Treaties and
Conventions) provided that the President of this railway company
should be Chinese, the stipulation was purely nominal. The Chinese
Eastern Railway, like the Russo-Chinese Bank, was an exclusively
Russian undertaking, the raising of the capital required, as well as
the construction of the line, being entirely in Russian hands.
Meanwhile the Kaiser, who personally directed the foreign policy
of Germany, was forming plans for claiming his share of reward for
the triple intervention, and he had, it appears, already approached
the Peking Government on this subject, though without any success.
What, assuming its existence, was the nature of the understanding
arrived at between the Courts of St. Petersburg and Berlin in regard
to Far Eastern affairs will probably remain for ever a State secret. In
any case, however, it is clear, from his own repeated declarations as
to Germany’s need for “a place in the sun,” and from the proceedings
of the German Minister at Peking, that he was bent on obtaining a
foothold of some sort in China, whence Germany’s future expansion
in the Far East might be conveniently developed. His opportunity
came in 1897. In the autumn of that year two German missionaries
were murdered in the province of Shantung. A few weeks later a
German force landed in that province at Kiaochow, one of the two
harbours the use of which in time of war Russia had acquired
eighteen months before under her secret Treaty with China. M.
Gérard in his book above mentioned states that the German Emperor
had before the departure of the German ships on this errand
informed the Tsar by telegraph of his intentions, and, receiving no
reply objecting to the proposed step, took the Tsar’s silence for
consent. Germany’s occupation of this strategic position, which had
the further advantage of being in a region of the Chinese mainland
sufficiently distant from points where other foreign interests were
centred to obviate objections on the part of other Powers, and, at the
same time, ensure an ample and undisturbed field for German
enterprise, was confirmed by a Treaty concluded with China on
March 6th, 1898. By this Treaty China granted to Germany a lease
for ninety-nine years of the port of Kiaochow and a considerable
stretch of “hinterland.” Germany also acquired under it certain rights
of railway construction in the neighbourhood of the port.
 The author of Japan: The Rise of a Modern Power, tells us, on the
authority of a statement said to have been made by Prince Henry of
Hohenzollern, that the Kaiser’s next step was to invite the Tsar to
take Port Arthur and Ta-lien Wan. Whatever truth there may be in
the statement attributed to Prince Henry—M. Gérard thinks the
suggestion may have been made in the telegram announcing his own
intentions—the fact remains that Germany’s abrupt action resulted
in an immediate scramble on the part of several European Powers for
various portions of Chinese territory. Russia led the way in these
undignified proceedings, for which a harsher word might with justice
be substituted. Two months after the occupation of Kiaochow by
Germany, Russian men-of-war anchored in Port Arthur. Thither they
were followed by British cruisers, and for a moment it looked as if
history would repeat itself, and that Russia might have to reckon
with British interference in her designs. Other counsels, however,
prevailed. The British ships were withdrawn, and on March 27th,
three weeks after the conclusion of the Kiaochow Agreement, a
similar Treaty was signed at Peking by Li Hung Chang and the
Russian Chargé d’Affaires. This Treaty, the text of which was not
published by the Russian Government, provided for the lease to
Russia of Port Arthur, Ta-lien Wan and adjacent waters for a period
of twenty-five years, renewable by arrangement at the expiration of
the term. It was further agreed that the right to construct the Chinese
Eastern Railway across Northern Manchuria, secured by Russia
under the secret Railway Agreement of August 27th, 1896, should be
extended so as to include the construction of branch lines from a
point on that railway to Ta-lien Wan and other places in the Liaotung
peninsula. The Treaty also provided for a subsequent definition of
the boundaries of the leased area and—a point of some importance in
the light of after events—of a neutral strip of territory separating the
Chinese and Russian spheres. Port Arthur, moreover, was declared
to be a naval port, and as such closed to all vessels save those of the
two contracting parties. Subsequently, on May 7th, a supplementary
Agreement, signed at St. Petersburg, defined the boundaries of the
leased area, and arranged for their delimitation.
It was not long before France, whose services to China at the time
of the triple intervention had, as we have seen, already met with
recognition in the shape of the prompt settlement of various
outstanding questions, obtained, in her turn, a territorial concession
of the same nature—though, perhaps, not so important—as those
granted to Germany and Russia. By a Convention signed at Peking
on May 27th, 1898, China granted to her a ninety-nine years’ lease
for the purpose of a naval station and coaling depôt of the Bay of
Kwang-chow and adjacent territory in the peninsula of Leichow,
together with the right to construct a railway connecting the bay with
the peninsula. The area of this concession was in the province of
Kwangsi, which adjoins the French territory of Tonkin.
Unlike the three Powers associated in the triple intervention,
whose subsequent action justifies the supposition that they regarded
themselves as brokers entitled to a commission for services
rendered, Great Britain had no special claim on the goodwill of
China. Nevertheless, she joined in the scramble for Chinese territory.
A Convention, signed at Peking on June 9th, 1898, gave her an
extension of territory at Hongkong under lease for a period of ninety-
nine years, the reason assigned for the concession being that this
extension was necessary for the proper protection and defence of
that colony. Three weeks later (July 1st), by another Convention,
signed also at Peking, it was agreed that the Chinese Government, “in
order to provide Great Britain with a suitable naval harbour and for
the protection of British commerce in the neighbouring seas,” should
lease to her Weihaiwei and the adjacent waters “for so long a period
as Port Arthur shall remain in the occupation of Russia.” The area
thus leased comprised the island of Liu-kung, and all other islands in
the bay of Weihaiwei.
In defence of Great Britain’s action it may fairly be pleaded that
her interests in China, and in the Far East generally, which were
more extensive than those of any other Power, with the possible
exception of Japan, made it necessary for her Government to take
prompt measures to counteract the effect of any proceedings on the
part of other Powers which might be prejudicial to those interests.
The political situation created in the Far East by the actions of the
three Powers associated in the triple intervention was the reverse of
reassuring. Russia’s occupation of Port Arthur was in direct
contradiction to the grounds of the joint protest against the
annexation of the Liaotung peninsula by Japan. Neither with France
nor with Russia at that time were our relations what they afterwards
became. Between British and Russian policy there was a scarcely
veiled antagonism, while the French and ourselves had long been
rivals in China, as elsewhere. The concerted action of these two
Powers, not to speak of their support by a third, whose exact relation
to her associates was dubious, was thus calculated to give rise to
apprehensions which would doubtless have been increased had
British Ministers then known all that has since come to light.
Additional gravity was given to Germany’s sudden appearance on the
scene in a new rôle by, to use M. Gérard’s words, her “occupation by
force and at a moment of complete peace of a port belonging to the
Empire the integrity of whose territory she claimed to have
safeguarded against Japan.” Under these circumstances the British
Government may well have felt that it was justified in regarding
these proceedings as fraught with possibilities of injury to British
interests and prestige, and in adopting what in the light of these
occurrences might reasonably be held to assume the character of
precautionary measures. Such, beyond a doubt, was the general
interpretation given by impartial observers to Great Britain’s action
in arranging for her occupation of Weihaiwei. It was, as the terms of
the Agreement clearly indicated, a direct counter-move to Russia’s
occupation of Port Arthur. As such it was welcomed by Japan, who,
when the time for the evacuation of Weihaiwei arrived, willingly
handed it over to the Power who was shortly to become her ally.
 CHAPTER XXIV
American Protest against Foreign Aggression
in China—Principle of “Open Door and Equal
Opportunity”—Financial Reform—Operation
of Revised Treaties—The Boxer Outbreak—
Russia and Manchuria.

In addition to the various Agreements for the occupation of Chinese

territory mentioned in the preceding chapter, negotiations were
conducted with the Chinese Government about the same time by the
European Powers concerned, and also by Japan, for the purpose of
obtaining Declarations regarding the non-alienation by China of
certain territories which were regarded by them as coming,
respectively, within their special spheres of interest. As a result of
these negotiations the French Minister at Peking received in March,
1897, a verbal assurance, confirmed later in writing, that the Chinese
Government would “in no case, nor under any form, alienate to
another Power the island of Hainan off the coast of the province of
Kwantung.” In February, 1898, a similar Declaration concerning the
riverain provinces of the Yangtse was made to Great Britain. In the
following April the assurance previously given to France was
extended so as to include the three southern provinces of Yunnan,
Kwangsi and Kwantung bordering on Tonkin; while Japan in the
same month received an assurance of a corresponding nature
regarding the province of Fukien, the Chinese Government signifying
its intention “never to cede or lease it to any Power whatsoever.” In
thus obtaining from China a Declaration of non-alienation respecting
the province of Fukien, similar to those given to Great Britain and
France regarding other portions of Chinese territory, Japan
established her claim to rank as one of the leading Powers in the Far
East, a position which, as will be seen, received further recognition in
the following year. Her success in this respect—due to her victory in
the war with China, and to the alteration in her status as a nation
which resulted from the conclusion of revised treaties with several
foreign Powers—was rendered the more noticeable by the failure of
Italy, after prolonged negotiations, to gain China’s consent to a
territorial concession similar to those granted to other European
Powers.
The years 1898 and 1899 witnessed the negotiation by European
Powers with each other of two other arrangements relating to China
of a somewhat different character. One of these was the Declaration
made by Great Britain to Germany on April 19th, 1898, binding
herself not to construct any railway connecting Weihaiwei, and the
adjoining leased territory, with the interior of the province of
Shantung. The other was the Agreement, effected through an
exchange of Notes at St. Petersburg on April 28th, 1899, by which the
British and Russian Governments recorded their intention to regard,
for the purpose of railway concessions, the basin of the Yangtse and
the region north of the Great Wall as the special spheres of interest of
the two Powers, respectively, confirming, at the same time, the
understanding arrived at between them in regard to the railway
between Shanhaikwan and Newchwang.
The outbreak of war between the United States and Spain in the
spring of 1898 led to the introduction of a new factor into the
situation created in the Far East by the events above described. One
of the results of the war was the cession of the Philippine Islands to
America, who had already, by annexing Hawaii, secured a stepping-
stone across the Pacific. By the acquisition of these former Spanish
possessions, which provided her with a naval base in the Eastern
Pacific for the protection of her commerce in Far Eastern waters,
America’s attitude towards Far Eastern questions was at once
affected. Hitherto in her relations with the Far East—with China,
Japan and Korea—she had maintained a detached attitude in
keeping with her traditional policy of non-interference in foreign
questions. In China, where she came late into the field, she had been
content to follow, at a distance, in the wake of other Powers; sharing
in whatever commercial or extra-territorial privileges might be
obtained, but never breaking the ice for herself, nor—to her credit, be
it said—betraying any aggressive tendencies. As the pioneer of
Western nations in putting an end to the seclusion of Japan and
Korea, she had opportunities for exercising a powerful influence, of
which her traditional policy forbade her to make full use. Regarding
both countries somewhat in the light of protégés, her policy in
respect to each soon settled down into one of benevolent inaction,
varied only by occasional half-hearted opposition to the less
complaisant policy of other Governments, whenever the duty of a
patron, so to speak, seemed to call for her interposition. We have
seen how she was thus led on two occasions in the matter of Treaty
revision into a premature encouragement of Japanese ambitions,
which was the cause of embarrassment both to herself, and to the
nation whose wishes she was willing to further. The course thus
pursued by America, which precluded concerted action with other
Powers, was in some respects simply an extension to the Far East of
the policy she had previously adopted in regard to European
questions. Well as the traditional principle of holding aloof from
affairs outside of the American continent, through fear of political
entanglements, may have suited the conditions of her earlier
existence as a nation, a too rigid adherence to this principle, when
those conditions were fast disappearing, might lead to consequences
more unpleasant than those she sought to avoid. An attitude of
detachment carried too far might result in her exclusion from a voice
in the regulation of matters of international interest. Towards some
such position America appeared to be drifting, when, to borrow the
phrase used by Mr. Hornbeck in Contemporary Politics of the Far
East, she suddenly “stumbled into World Politics” through her
occupation of the Philippines. From that moment her political
isolation was ended. She began to take a more active and intelligent
interest in Far Eastern questions, though the reluctance to abandon
her traditional policy, which was still noticeable in her action when
she did move, was liable to be mistaken for timidity.
The territorial concessions obtained, one after the other, by
Germany, Russia, France and Great Britain, and the ear-marking of
other Chinese territory by arrangements made either by the Powers
concerned, as well as by Japan, with China, or by certain of those
European Powers between themselves, caused uneasiness in
Washington. There was a fear lest the new activity displayed by
various Governments might result in the closure, or restriction, of
Chinese markets hitherto open to all countries, in which case serious
injury might accrue to American commerce and enterprise. The
apprehension was not unfounded, even so far as the Declarations
regarding the non-alienation of Chinese territory were concerned.
Although the actual wording of these Declarations did not of itself
justify an inference of this nature, from the fact that they were made
at all it was generally held that their effect was to establish, in each
instance, a sort of priority of right—a position of exceptional
advantage in favour of the Power to whom the Declaration was made.
The inference derived support from the vagueness of the term
“spheres of interest” applied to the regions affected by the
Declarations in question, and was also strengthened by the common
impression formed at the time that this ear-marking of Chinese
territory portended an eventual partition of China. This seems to
have been the view taken by the United States Government.
In September, 1899, the American Secretary of State addressed
Circular Notes to the British, French, German and Russian
Governments, expressing the hope that they would “make a formal
declaration of an ‘open door’ policy in the territories held by them in
China.” An assurance was sought from each Power: that it would “in
no way interfere with any treaty port or any vested interest within
any so-called sphere of interest, or leased territory, it might have in
China”; “that the Chinese Treaty tariff of the time being should apply
to all merchandise landed or shipped to all such ports as are within
the said ‘sphere of interest’” ... and “that duties so leviable should be
collected by the Chinese Government”; and that it “would levy no
higher harbour dues on vessels of another nationality frequenting
any port in such ‘sphere’ than should be levied on vessels of its own
nationality, and no higher railroad charges over lines built,
controlled or operated within its ‘sphere’ on merchandise belonging
to citizens or subjects of other nationalities transported through such
‘sphere’ than should be levied on similar merchandise belonging to
its own nationals transported over like distances.” In the following
November similar, though not identical, Notes were addressed to the
Governments of France, Italy and Japan, asking them to join in these
formal declarations of policy.
The reason for the distinction thus made both in the dates and
tenor of the two series of communications may, perhaps, be found in
the fact that the territories leased by the three first-named Powers,
besides their greater strategic importance, were situated in a part of
China where American interests were more closely concerned than in
the region further south affected by French action, and that Japan,
though interested in the Declaration regarding Fukien, had neither
sought nor obtained any cession of territory; while Italy had failed in
her endeavour to emulate the example of her nearest continental
neighbours.
The assurance received from China by Japan regarding the non-
alienation of the province of Fukien was, as we have seen, in effect,
an admission of the position of power and influence she had by this
time acquired. Her inclusion in the list of States consulted by
America on this occasion was indirectly an endorsement of this
admission, and is the first public recognition of her new status as a
leading Power in the Far East.
Favourable replies were received from all the Powers consulted;
each, however, with the exception of Italy, making the reservation
that assent to the proposals was subject to the condition that all the
Powers interested should participate in the Declarations. Thereupon,
in March, 1900, the American Secretary of State sent instructions to
the American representative at each of the capitals of the Powers
consulted to inform the Government to which he was accredited that,
in his opinion, the six Powers in question and the United States were
mutually pledged to the policy of maintaining the commercial status
quo in China, and of refraining, each within what might be
considered its sphere of influence, “from measures calculated to
destroy equality of opportunity.”
The Notes thus exchanged between the United States and the six
other Powers above mentioned explain the origin, as they also
constitute “the formal basis” (to use Mr. Hornbeck’s words) of what
has ever since been known as the policy of the “Open door and equal
opportunity” in China. The latter part of the phrase was afterwards
used in the Anglo-Japanese Treaty of Alliance to designate the policy
of Great Britain and Japan in Korea as well as in China. To the
former country, now annexed to Japan, it no longer applies; but the
policy has theoretically, if not always practically, been in force as
regards China, for the last twenty-one years, and there is reason to
think that more may yet be heard both of the phrase, and of the
policy it represents, in connection with affairs in China, and possibly
in other parts of Eastern Asia.

In touching on the subject of financial reform in a previous chapter

attention was called to the monetary confusion which existed after
the abolition of the feudal system, when the new Government which
had come into power found itself saddled with clan debts and with
clan paper money, mostly depreciated and of many different kinds. It
was pointed out how, as a natural consequence of this monetary
confusion and of financial embarrassments due to other causes, the
monetary transactions of the country were for many years conducted
on the basis of an inconvertible paper currency; and how by
successive steps, taken as opportunity offered, to remedy this state of
things, specie resumption on a silver basis was at length effected in
1886.
It was not till eleven years later, in 1897, that Japan adopted her
present gold standard. The reasons for this step are given in the
chapter on Finance contributed to Marquis Ōkuma’s book, already
mentioned, by Marquis Matsugata, who also explains the means by
which it was accomplished.
“When,” says this authority on Japanese financial matters, “the
Government opened places for the redemption of paper money in
1886, silver coins only were offered in exchange. Such being the case,
the currency of Japan at that time was based practically on a silver
standard, although legally the system was bimetallic. The price of
silver, however, owing to various reasons, gradually fell, and artificial
checks to its fall were effective only for a short time. Fluctuation after
fluctuation in foreign exchange seemed to follow each other in
endless succession. In the meantime Western countries commenced
to adopt gold monometallism. Our authorities knew very well that, to
insure a healthy growth of finance, Japan must adopt, sooner or
later, a monometallic gold standard, and this was impressed on the
minds of financiers so keenly that the Government determined to
effect the reform as soon as possible. The desired opportunity came
with the Peace Treaty of 1895, when China began to pay to our
country an indemnity of 200,000,000 taels” [sic the amount was
really 230,000,000 taels]. “Further negotiations between our
Government and the Chinese authorities resulted in the payment of
the indemnity, not in Chinese money, but in pounds sterling. This
was important, since a large gold reserve was indispensable for the
establishment of gold monometallism.”
The experience of 1886, referred to by Marquis Matsugata, proved
that confidence in the Government’s ability to meet its obligations in
paper money was all that was needed. This confidence once
established, no further difficulty presented itself in the passage from
an inconvertible to a convertible paper currency. Prepared for heavy
calls on the specie resources of the Treasury, the Government had on
that occasion accumulated a reserve of £5,000,000. When, at the
end of a few days after the date fixed for the resumption of specie
payments, the demand for specie ceased, it was found that the total
value of notes presented for conversion did not exceed £30,000. The
change from a silver to a gold standard in 1897 was conducted with
equal facility, a large portion of the Chinese indemnity being
transferred abroad. There it served a useful purpose in maintaining
Japan’s financial credit, and, as a natural consequence, the market
price of the Bonds of her numerous foreign loans, which for several
years, to the surprise of private investors, were quoted at higher rates
abroad than at home.

The year 1899, when the revised treaties came into operation,
marked a fresh stage in the progress of Japan towards attaining a
footing of equality with Western Powers—the aim which her
statesmen had set before themselves ever since the Restoration, and
which had in so many ways been the guiding principle of both
domestic and foreign policy. With the object of allowing time for the
negotiation of similar treaties with other foreign Powers, the revised
British Treaty, signed in London in 1894, had, as already mentioned,
provided that it should not come into force until five years after the
date of signature. Before the expiration of the period named similar
treaties had been concluded with all the other Powers concerned,
those with France and Germany containing a few modifications of
minor importance. In the meantime, moreover, the conditions
specified in the Treaty regarding the new Japanese Codes and
Japan’s adhesion to the International Conventions concerning
Copyright and Industrial Property had been fulfilled. The way was
thus cleared for the operation of the new revised treaties, which,
accordingly, came into force on the 17th July, 1899, the earliest date
possible. Though in these new treaties, recognizing the territorial
jurisdiction of Japan, the stipulation of previous conventions which
chiefly offended Japanese susceptibilities found no place, she still
remained bound for a further period of twelve years—the term of the
revised treaties—by a tariff of a unilateral character. Only when that
period expired would she recover full tariff autonomy and be free to
negotiate reciprocal treaties with the various Powers concerned on a
footing of complete equality. This opportunity came to her in 1911,
and she at once availed herself of it.

In the spring of the following year (1900) what is known as the

Boxer Rising took place. In its inception it was a protest against
missionary enterprise. As it developed, it became the expression of a
feeling of exasperation among the official and lettered classes of
Northern China engendered by the action of European Powers in
occupying under the guise of leases various portions of Chinese
territory in that region. During the previous autumn a society called
I-Ho-C’uan (Patriot Harmony Fists) had been formed in the province
of Shantung. Its formation was encouraged by the reactionary
tendencies which made their appearance about this time at Peking,
where the Empress Dowager, after the successful coup d’état by
which she had crushed the ill-conducted reform movement in 1898,
was again in power. The magical powers claimed by its members
produced on the ignorant masses an impression that was heightened
by the incantations they performed. As the movement grew, it
attracted the attention of the Governor of the province, who
supported it with, apparently, the twofold idea of utilizing it against
foreign aggression, and gaining favour at Court. As a result of his
outspoken sympathy the Boxer movement assumed formidable
dimensions. Though eventually, through the energy of Yuan Shih-
k’ai, who was at one time, as we have seen, Chinese Resident in
Korea, order was restored in Shantung, the movement spread
northwards towards Peking. There, as Mr. Campbell explains in the
China Handbook prepared under the direction of the Foreign Office,
it gained the powerful support of the ignorant and reactionary
statesman Prince Tuan, the selection of whose son as Heir-Apparent
to the Throne gave him a commanding influence in the councils of
the Empire. In April, 1900, bands of Boxers were drilling in the
outskirts of the Capital, their appearance in every district they
invaded being accompanied by murders of missionaries and
massacres of native converts. Some weeks later the situation became
so threatening that arrangements were made for bringing up to
Peking small contingents of foreign troops for the protection of the
Legations and such portion of the foreign community as still
remained. These guards arrived opportunely at the end of May, by
which time swarms of Boxers infested the Capital, and the Legations
were practically isolated. Prince Tuan chose this moment for openly
espousing the Boxer cause. This step on his part was followed by the
murders of the Chancellor of the Japanese Legation and the German
Minister, the two outrages occurring within a few days of each other.
The subsequent course of events is well known: the storming of the
Taku forts (June 16th); the siege of the Legations by Chinese troops
and Boxers; the failure of Admiral Seymour’s attempt to re-establish
communications with the Capital; the equipment of foreign
expeditionary forces to operate against Peking; the issue of an
Imperial Decree ordering a general massacre of foreigners in the
Chinese dominions; the attack on the foreign settlements at Tientsin;
the arrival of Russian and British reinforcements, and the taking of
Tientsin city (July 14th); the relief of the Legations, and occupation
of the Chinese capital on the 13th and 14th August by the allied
forces; and the flight of the Chinese Court to Sian-fu, the ancient
capital in the province of Shensi. With the flight of the Court from
the capital Chinese resistance collapsed, and when Count Waldersee
arrived in September with several thousand German troops to take
supreme command of the allied expeditionary forces, there was no
enemy to fight. Hostilities gave place to negotiations between the
foreign Governments concerned and China for the settlement of the
various issues raised by the Boxer outbreak. The negotiations
resulted in two preliminary exchanges of Notes, dated, respectively,
December 22nd, 1900, and January 16th, 1901, embodying the
conditions for the re-establishment of normal relations with China,
and in the signature of a final Protocol on September 7th, 1901.
Three days before its signature Prince Ch’un, who had proceeded on
a mission to Berlin to apologize for the murder of the German
Minister, was received in audience by the Kaiser.
 The chief conditions imposed on China by these arrangements
were the payment of an indemnity of 450,000,000 Haikwan taels
(equivalent at the rate fixed—3s. per tael—to £67,500,000); the
permanent occupation of certain places, including Tientsin and
Shanhaikwan, for the purpose of preserving free communications
between Peking and the sea; the razing of the Taku and other forts
which threatened those communications; and the construction of a
separate fortified quarter in the Capital for the foreign Legations, for
the further protection of which permanent foreign guards were to be
retained. Other terms included special reparation for the murders of
the German Minister and the Chancellor of the Japanese Legation
and the desecration of cemeteries; the punishment of Prince Tuan, as
well as other personages and officials responsible for the attacks on
foreigners; and the prohibition of the import of arms.
Thanks, as we learn from the Handbook already quoted, to the
good sense of the leading provincial authorities, such as the Viceroys
of Nanking and Wuchang and the new Governor of Shantung, who
had the courage to disobey the Imperial Decree, the Boxer movement
was stifled in the central and southern regions of China. There, in
spite of considerable unrest, order was preserved. But further north
in Manchuria the Governors were not so judicious. In obedience to
instructions from the Court they declared war on the Russians. The
sudden attacks made by Chinese forces created a panic on the Amur,
and brought about the savage reprisals which occurred at
Blagovestchensk on that river, and the occupation of the whole of
Manchuria by Russian troops. The folly of the Empress Dowager and
of the ignorant clique by whose counsels she was guided gave Russia
the opportunity she desired for pursuing her designs of aggression in
the Far East. Her subsequent conduct throughout the negotiations,
and after their conclusion, destroyed the good effect produced by her
valuable co-operation in the fighting at Tientsin, where the Russian
reinforcements were, undoubtedly, the chief factor in saving the
foreign settlements from destruction.
In the military operations against Peking, and in the protracted
negotiations which succeeded them, Japan played a conspicuous
part. She had suffered injury similar to that sustained by other
foreign Powers in connection with the Boxer Rising, and she had a
common interest with them in adopting whatever measures might be
necessary in the international emergency which had arisen. Her
proximity to China and her military resources enabled her to strike
quickly, and with effect. To the invitation to take part in the
expeditionary force in process of organization, which was addressed
to her by the other interested Powers, with the exception of Russia,
she responded with alacrity; and in a short space of time a well-
equipped Japanese force took its place with the troops of other
Powers, and joined in the march on Peking for the relief of the
besieged Legations. The discipline and efficiency of the Japanese
contingent won well-deserved praise from those best qualified to
judge. In the subsequent negotiations the readiness shown by Japan
to act in harmony with other Powers, whose attitude was influenced
by consideration for the general interests of all concerned, facilitated
the solution of many difficulties; and, when the question of claims
for indemnity came to be discussed, the moderation of her demands
was equalled only by that of Great Britain and the United States.
 CHAPTER XXV
Agreement between Great Britain and
Germany—The Anglo-Japanese Alliance.

Soon after the opening of negotiations for the re-establishment of

friendly relations with China the Governments of Great Britain and
Germany concluded an Agreement of a self-denying character which
confirmed, though in different words and with special application to
the situation then existing in China, the principle of the “open door
and equal opportunity,” as set forth by the United States, and
accepted by the Powers consulted, in the autumn of 1899 and the
spring of the year following. By this Agreement, signed in London on
October 16th, 1900, the two Powers bound themselves to support the
principle above mentioned; to abstain from making use of the
existing troubles in China to “obtain for themselves any territorial
advantages”; and to co-operate for the protection of their interests in
the event of any attempt on the part of another Power to obtain such
advantages under existing conditions. The Agreement was, as
prearranged, communicated to other interested Powers, who were
invited “to accept the principles recorded in it.” Replies more or less
favourable were received from the Powers addressed. The French
Government referred to its prompt adhesion to the proposals of the
United States in the previous year as a proof of its long-entertained
wishes in the direction indicated; while the Russian reply, which, like
the French, took the form of a Memorandum, went so far as to say
that Russia had been “the first to lay down the maintenance of the
integrity of the Chinese Empire as a fundamental principle of her
policy in China.” The Japanese Government, in its answer, stated
that, in view of the assurance received that in adhering to the
Agreement Japan would be placed in the same position as she would
have occupied had she been a signatory instead of an adhering State,
it had no hesitation in adhering to the Agreement, and accepting the
principles embodied therein.
Subsequently, when it became apparent that Russia had no idea of
evacuating the territory she occupied in Manchuria, the German
Government explained that the Agreement was never intended to
apply to that territory.
The course pursued by Russia from the outset of the negotiations
in Peking was in marked contrast to the attitude adopted by the
other Powers concerned, and in direct contradiction to the principles
embodied in the Anglo-German Agreement in which she professed to
acquiesce. From some of the demands made by the other Powers
conjointly she dissociated herself, while her conduct in keeping her
troops stationed in the furthest positions to which they had
penetrated during the Boxer outbreak indicated an intention to give
a permanent character to her occupation of Manchuria. Her attitude
in this latter respect was doubtless encouraged by the fact that,
whereas the Final Protocol provided for the withdrawal of foreign
troops, under certain conditions, from Peking, and the province of
Chihli, it contained no reference to the evacuation of Manchuria.
Further proof of her designs was furnished by the conclusion in
January, 1901 (subject to confirmation by the Peking Government),
of an Agreement between Admiral Alexeieff and the Tartar General
at Moukden, placing the province of Fêng-t’ien (Shenking) under
Russian control, and by the subsequent opening of negotiations at St.
Petersburg for a formal Convention, which would have established a
Russian Protectorate over the whole of Manchuria, besides giving her
exclusive, or preferential, rights in Mongolia and Chinese Turkestan.
These attempts to obtain China’s consent to her occupation of
Manchuria, and to secure for herself a position of exceptional
advantage elsewhere, were frustrated by the vigilance of Great
Britain, the United States and Japan, and by the general indignation
they aroused in China. The Government at Peking, yielding to the
pressure thus brought to bear upon it, withheld its confirmation of
the Moukden Agreement; the Chinese Minister at the Russian capital
was forbidden to sign the Convention under negotiation; and
eventually, in August, 1901, the Russian Government issued an
official communiqué announcing the shelving of the proposed
Convention owing, as it was explained, to the misrepresentation of
Russia’s intentions. Russian troops, nevertheless, remained in
Manchuria, and it was not until after the conclusion of the Anglo-
Japanese Alliance that Russia at length made an Agreement with
China for the evacuation of the territory she had occupied, an
Agreement which, as M. Witte afterwards explained to the British
Ambassador in St. Petersburg, she never intended to observe.
On the 30th January, 1902, the Anglo-Japanese Treaty of Alliance
was signed in London by the Marquess of Lansdowne and the
Japanese Minister there, the late Count (then Baron) Hayashi, who
was afterwards Japanese Minister for Foreign Affairs. The Treaty
related to affairs in “the Extreme East,” and came into effect
immediately after signature. It was terminable after five years’
duration, at one year’s notice on either side, subject to the condition
that should either of the contracting parties be at war when the
period of the Treaty came to an end it should remain in force until
peace was concluded. By this Agreement the contracting parties
recognized the independence of China and Korea, and the special
interests therein of Great Britain and Japan respectively. They bound
themselves to maintain strict neutrality in the event of either of them
being involved in war, and to come to one another’s assistance in the
event of either being confronted by the opposition of more than one
hostile Power. The Treaty also, as we have seen, affirmed the
principle of “equal opportunity.”
In his despatch to the British Minister in Tōkiō notifying the
signature of the Agreement the Marquess of Lansdowne observed
that it might be regarded as the outcome of the events which had
taken place during the last two years in the Far East, and of the part
taken by Great Britain and Japan in dealing with them. Count
Hayashi, in his Secret Memoirs, published in London in 1915 after
his death, confirms this statement, but puts the date at which
tendencies began to take shape in this direction somewhat further
back. The idea of an alliance between the two countries first came, he
says, into the minds of Japanese statesmen soon after the triple
intervention of 1895, and was favoured by Count Mutsu, who was at
the time Minister for Foreign Affairs. The effect of that intervention,
he explains, was to cause a regrouping of Powers in the Far East:
France, Russia and Germany forming one group, while Great Britain,
Japan and the United States represented another. Having this
regrouping in view, he himself, in the summer of that year, suggested
the desirability of such an alliance, should the unfriendly attitude of
certain Powers towards Japan be continued. The suggestion was
made in articles contributed to a leading Tōkiō journal after he had
ceased to be Vice-Minister for Foreign Affairs, and on the eve of his
appointment as Japanese Minister to China.
The following extracts from a summary of these articles, which is
given in the Memoirs, show how, undismayed by the retrocession of
the Liaotung peninsula, Japanese statesmen still held firmly to their
settled policy of attaining for the nation a footing of equality with
Western Powers, realizing perhaps more clearly than before that the
increase of Japan’s naval and military strength was the only means of
attaining their object.
“We must,” the writer of the articles says, “continue to study
according to Western methods, for the application of science is the
most important item of warlike preparations that civilized nations
regard. If new ships of war are considered necessary, we must build
them at any cost. If the organization of the army is found to be
wrong ... the whole military system must be entirely changed. We
must build docks to be able to repair our ships. We must establish a
steel factory to supply guns and ammunition. Our railways must be
extended so that we can mobilize our troops rapidly. Our oversea
shipping must be developed so that we can provide transports to
carry our armies abroad. This is the programme that we have to keep
always in view.... What Japan has now to do is to keep perfectly
quiet, to lull the suspicions that have arisen against her, and to wait,
meanwhile strengthening the foundations of her national power,
watching and waiting for the opportunity which must one day surely
come in the Orient. When that day comes, she will be able to follow
her own course.”
How sedulously all the steps indicated were subsequently carried
out is now common knowledge. Preparations on a scale so extended
could mean only one thing—provision against the possible
eventuality of war with the Power that might stand in the way of
Japan’s “following her own course.”
 Marquis Saionji. General Prince Katsura.

Descended from an ancient family of Rendered distinguished services in the

Court Nobles. A prominent figure in war with China and Russia; he was
diplomacy and parliamentary life. He conspicuous both as soldier and
was chief delegate for Japan at the statesman.
Versailles Conference.

The idea of an alliance, or some sort of understanding, between the

two countries thus put forward in 1895 seems to have gradually
made way both in Japan and in Great Britain. We learn from the
same Memoirs that in 1898 Mr. Joseph Chamberlain, then Colonial
Minister, expressed to Viscount (then Mr.) Kato, who was at that
time Japanese Minister in London, the readiness of Great Britain to
enter into an agreement with Japan for the settlement of affairs in
the Far East, and that the latter, in reporting the conversation to the
Foreign Minister in Tōkiō, strongly supported the suggestion. The
subject, it appears, was again discussed in the course of a
conversation which Count Hayashi had with the late Marquis Itō and
with Marquis (then Count) Inouyé in Tōkiō in 1899, prior to his
(Count Hayashi’s) appointment as Minister in London. His account
of what passed on this occasion shows that the Japanese
Government was at that time hesitating between two opposite
courses—an agreement, or alliance, with Great Britain, and an
understanding with Russia; and it seems to have been thought that
the latter Power was in a position to offer better terms. Soon after his
arrival, early in January, 1900, to take up his post in London the new
Minister met the late Dr. Morrison, then Times correspondent in
Peking, with whom he discussed the question of an alliance between
the two countries. He seems then to have formed the impression that
most British journalists were in favour of an Anglo-Japanese
alliance.
It was not, however, until the following year that the question
began to assume a practical aspect. The first move came from an
unexpected quarter, the German Embassy in London. In March,
1901, Freiherr von Eckhardstein, who was then, owing to the illness
of the German Ambassador, in the position of Chargé d’Affaires,
called on Count Hayashi and expressed the opinion that a triple
alliance between Germany, Great Britain and Japan was the best
means of maintaining peace in the Far East. He suggested that he
(Count Hayashi) should take the initiative in proposing this alliance.
The latter, who had, as we know, been one of the first to advocate an
Anglo-Japanese alliance, reported the suggestion to his Government,
and was instructed to sound the British Government unofficially on
the subject. Much light is thrown on the subsequent course of
negotiations by the Memoirs already mentioned, and Freiherr von
Eckhardstein’s “Reminiscences” (Lebens Erinnerungen und
Politische Denkwürdigkeiten), published in Leipzig in 1920. The ball
thus set rolling, the question was, we learn, discussed informally
from time to time, on the one hand between the Japanese Minister
and Lord Lansdowne, and, on the other, between the latter and the
German Chargé d’Affaires; but it was never reopened by the German
Embassy with the Japanese Minister.
There seems to have been little enthusiasm for the project of a
triple alliance on the part of any of the foreign Ministries concerned.
Great Britain appears to have shown more inclination in this
direction than the other two Powers, for until a late stage in the
negotiations with Japan the point would seem to have been kept in
view by the British Cabinet. If the German Government ever
seriously entertained the idea—which is very doubtful—it was merely
for the reasons mentioned by the Foreign Office in Berlin, that the
inclusion of Japan might be acceptable to her on general grounds,
since she would “find herself in good company,” and might make
negotiations with Great Britain easier, “as Japan was popular in
Germany.” The alliance with Great Britain was regarded as the main
consideration; and even in this matter there is no reason to think
that the German overtures were sincere, for Berlin’s insistence on
Austria’s being brought into the business, though not as a
contracting party, added to the difficulties already in existence. Nor,
on the side of Japan, where the part played by Germany in the
Liaotung incident was not forgotten, does there seem to have been
any marked desire for the inclusion of that Power in any
understanding between herself and Great Britain. This explains the
separate character of the negotiations carried on in London. As
between Great Britain and Germany, they lasted no longer than a few
weeks, during which time they appear to have been kept alive only by
the efforts of the German Chargé d’Affaires, to whose initiative the
project was due. After the resumption of his duties by the German
Ambassador the negotiations were transferred to Berlin, where they
soon came to an end. Their failure is described by the author of the
Reminiscences as “the starting-point of the encirclement
[Einkreisung] of Germany, and of the world-war which was the
mathematical consequence.”
The parallel negotiations between Great Britain and Japan were
not interrupted by the inability of the British and German
Governments to arrive at an understanding. No obstacles of the kind
that stood in the way of an agreement between the two other Powers
existed. The cordial relations which had been established as a result
of the settlement of the long-pending question of Treaty revision had
been improved by the close co-operation of the two countries in the
international measures in which both had joined at the time of the
Boxer outbreak, and by the harmony of views that was developed
during the Peking negotiations. The only difficulty which presented
itself lay in the fact, already referred to, that the Japanese
Government was hesitating between two opposite courses—an
understanding with Russia and an agreement with Great Britain. The
decision rested with the leading statesmen, who on this point were
divided into two parties, one led by the late Prince Itō and the late
Marquis Inouyé, the other by Prince (then Marquis) Yamagata and
the late Prince Katsura. Itō, whose pro-German tendencies were well
known, was in favour of coming to an understanding, if possible,
with Russia, and his opinion was shared by Inouyé. Yamagata and
Katsura, on the other hand, were inclined towards an alliance with
Great Britain. Fortunately for the London negotiations, the cleavage
of opinion did not follow clan lines. The Chōshiū party, to which the
four statesmen in question all belonged, was itself divided.
Fortunately, also, Katsura was then Premier. His and Yamagata’s
policy was adopted by the Cabinet, and finally prevailed. In his
opposition to the Cabinet’s policy Itō went so far as to arrange that a
visit he was about to make to America in connection with
celebrations at the University of Yale should be extended to Russia,
where he seems to have exchanged views with Russian statesmen.
His action threatened at one moment to imperil the success of the
London negotiations, and it became necessary for the Japanese
Government to explain that his visit to Russia had no official
character. In the face of this disavowal he could do little. Whatever
plans he and those who supported him may have formed came to
nothing, and in the end he was forced to content himself with
criticizing unfavourably the draft of the Anglo-Japanese Treaty
which embodied the final amendments proposed by Japan. The
strength of his position in the country at the time, as well as his
influence with the late Japanese Emperor, may be gathered from the
fact that these last amendments were transmitted by the
Government to him in Russia by special messenger, with a request
for his opinion.
It is unnecessary to emphasize the importance of the Anglo-
Japanese alliance. Count Hayashi, in speaking of it as “an epoch-
making event,” does not overstate the case. For both countries it was
a new and grave departure in policy, ending an isolation which was a
source of weakness to each in the quarter of the world to which it
applied. For Japan it had a treble value. It practically assured her
against a repetition of the Liaotung incident, while the mere fact of
her becoming the ally of one of the leading Powers of the world
added greatly to her prestige, and it facilitated the floating of loans
on the London market. If the benefit accruing to Great Britain may
seem to have been less, the alliance was nevertheless opportune in
view of the close understanding between Russia and France in the
Far East, the open menace to her interests offered by Russian
designs in Manchuria and the danger to be apprehended from their
further extension. The fact that the alliance was renewed in an
extended form three years later, was again renewed in 1911, and is
still in force, shows that both Governments have reason to be
satisfied with its results.
The conclusion of the Anglo-Japanese alliance drew from the
Russian and French Governments a Declaration, signed in St.
Petersburg on March 3rd, 1902, which left no doubt as to the
interpretation placed on it in St. Petersburg and Paris. In this
Declaration the two Governments, while approving of the
fundamental principles affirmed in the Anglo-Japanese Agreement,
reserved to themselves the right to consult each other, if necessary,
regarding the protection of their interests. The comment of the
author of Le Monde et la Guerre Russo-Japonaise on this counter-
move was that “it had almost no value as an answer to the Anglo-
Japanese Treaty.”
The action of Russia in prolonging indefinitely her occupation of
Manchuria, in spite of the protests of other Powers, and her attempts
to strengthen her position there by secret arrangements with China,
in defiance of the principle of “the open door and equal opportunity”
which she had united with other Powers in accepting, caused fresh
uneasiness in Washington. On February 1st, 1901, almost
simultaneously with the signature of the Anglo-Japanese Treaty, the
American Secretary of State, to whose initiative in 1899 the
acceptance of this principle had been due, addressed Circular Notes
to the Governments of China, Russia and nine other Powers on the
subject of the situation created in Manchuria by the Russian
occupation. Any agreement, he pointed out, by which China ceded to
corporations, or companies, exclusive industrial rights and privileges
in connection with the development of Manchuria constituted a
monopoly, and, being a distinct breach of the stipulations of treaties
between China and foreign Powers, seriously affected the rights of
American citizens. Such concessions would be followed by demands
from other Powers for similar exclusive advantages in other parts of
the Chinese Empire, and would result in “the complete wreck of the
policy of absolute equality of treatment of all nations in regard to
trade, navigation and commerce within the confines of the Empire.”
 Influenced, perhaps, by the Anglo-Japanese alliance and the
written protest of the United States, Russia at length, on the 8th
April, 1902, concluded at Peking an Agreement for the evacuation of
Manchuria. The Agreement was to come into force from the date of
signature, and was to be ratified within a period of three months, but
this latter stipulation was never observed. It provided for the
evacuation to be conducted in three stages, and to be completed in
eighteen months—that is to say, by October, 1903. The evacuation
was, however, made dependent on two conditions: the absence,
meanwhile, of disturbances in the province, and the abstention of
other Powers from any action prejudicial to Russian interests
therein. The first stage fixed by the Agreement, the withdrawal of
Russian troops from the south-western portion of the province of
Moukden (Fêng-t’ien), was duly carried out by the date agreed upon,
the 8th October, 1902. Before, however, the date fixed for the
completion of the next stage of evacuation (March, 1903), the
withdrawal of Russian troops from the remainder of the province of
Moukden and from the province of Kirin, other and quite new
conditions were formulated by the Russian Government, one being
that no “treaty ports” should be opened in the evacuated territory. In
the face of the well-known fact that the fresh commercial treaties
which America and Japan were negotiating with China contemplated
the opening of additional places for foreign trade in Manchuria,
these sudden demands indicated no intention on Russia’s part to
abide by the Agreement. If any doubt in this respect existed, it was
removed by her action in reoccupying early in 1903 districts she had
already evacuated, this step being followed by the issue in July of the
same year of an Imperial Ukase appointing Admiral Alexeieff Viceroy
of the Amur and Kwantung territories—the latter being, as already
mentioned, the name of the small peninsula in which Port Arthur is
situated.
 CHAPTER XXVI
War with Russia—Success of Japan—
President Roosevelt’s Mediation—Treaty of
Portsmouth—Peace Terms.

The threatening attitude of Russia, who no longer made any pretence

of masking her designs in China, was regarded with increasing
anxiety in Japan, where the necessity of preparing to meet force with
force had already been foreseen. But the high-handed proceedings of
the Russians in Manchuria were not the only cause of the tension
that from this moment began to appear in the relations between the
two countries. Mischief of a kind which had already led to war
between China and Japan was also brewing in Korea. By the Treaty
of Shimonoséki, which ended the war, the independence of that
country was recognized. China in relinquishing her claim to
suzerainty no longer maintained Chinese guards for her Legation in
Seoul, and ceased from all political activity in the peninsula, where
the influence of Japan for a time became predominant. But history
was about to repeat itself. Into the place vacated by China, Russia at
once stepped, and Japan found herself confronted by another and far
more dangerous competitor. The positions of the two new rivals in
Korea were very different. The alliance forced by Japan on the
Korean Government at the outset of the war with China had enabled
her to strengthen her political influence, while the energy she threw
into the development of business projects of various kinds had
increased her material interests in the peninsula. The lion’s share of
Korea’s foreign trade and maritime transport was in the hands of
Japan. She had also constructed and was in charge of the working of
telegraphic communications in that country; she had secured a
concession for the construction of railways; and she had her own
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebookluna.com