The Last Century of Sea Power, Volume 1
438 pages
English

Vous pourrez modifier la taille du texte de cet ouvrage

Obtenez un accès à la bibliothèque pour le consulter en ligne
En savoir plus

The Last Century of Sea Power, Volume 1

-

Obtenez un accès à la bibliothèque pour le consulter en ligne
En savoir plus
438 pages
English

Vous pourrez modifier la taille du texte de cet ouvrage

Description

The first volume of a magisterial study of naval power in the 20th century


The transition to modern war at sea began during the period of the Sino-Japanese War (1894–1895) and the Spanish-American War (1898) and was propelled forward rapidly by the advent of the dreadnought and the nearly continuous state of war that culminated in World War I. By 1922, most of the elements that would define sea power in the 20th century were in place. Written by one of our foremost military historians, this volume acknowledges the complex nature of this transformation, focusing on imperialism, the growth of fleets, changes in shipbuilding and armament technology, and doctrines about the deployment and use of force at sea, among other factors. There is careful attention to the many battles fought at sea during this period and their impact on the future of sea power. The narrative is supplemented by a wide range of reference materials, including a detailed census of capital ships built during this period and a remarkable chronology of actions at sea during World War I.


Preface and Acknowledgments
List of Maps
List of Appendices
Part I. Introduction. Definitions and Terms of Reference
Chapter 1 The Sino-Japanese War, 1894-1895
Chapter 2 The Greco-Turkish war of 1897
Chapter 3 The Spanish-American War of 1898
Chapter 4 The Shifting Balance of Power
Part I Appendices
Part I Notes
Part II. Introduction: From Port Arthur to Bucharest, 1898 to 1913
Chapter 5 The Russo-Japanese War: The First Phases
Chapter 6 The Russo-Japanese War: The Battle of Tsushima and its Aftermath
Chapter 7 The Dreadnought Naval Race
Chapter 8 Prelude to the First World War
Part II Appendices
Part II Notes
Part III. Introduction: From Sarajevo to Constantinople, 1914 to 1922
Chronology of the First World War at Sea
Chapter 9 The First World War: The War in Northern Waters
Chapter 10 The First World War: Tsingtao and the Dardanelles
Chapter 11 The First World War: Naval Support of Operations in Africa
Chapter 12 The First World War: Action in the Baltic
Chapter 13 The First World War: The Black Sea, Otranto Strait, and Other Matters
Chapter 14 The Legacy of the First World War
Part III Appendices
Part III Notes
Part IV. Conclusion: Not so much Finis as . . .
Part IV Appendix
Part IV Notes
Index

Sujets

Informations

Publié par
Date de parution 09 juin 2009
Nombre de lectures 3
EAN13 9780253003560
Langue English
Poids de l'ouvrage 1 Mo

Informations légales : prix de location à la page 0,0025€. Cette information est donnée uniquement à titre indicatif conformément à la législation en vigueur.

Exrait

THE LAST CENTURY OF SEA POWER
THE LAST CENTURY OF SEA POWER

Volume One: From Port Arthur to Chanak, 1894-1922
H. P. Willmott
This book is a publication of
Indiana University Press
601 North Morton Street
Bloomington, IN 47404-3797 USA
http://iupress.indiana.edu
Telephone orders 800-842-6796
Fax orders 812-855-7931
Orders by e-mail iuporder@indiana.edu
2009 by H. P. Willmott
All rights reserved
No part of this book may be reproduced or utilized in any form or by any means, electronic or mechanical, including photocopying and recording, or by any information storage and retrieval system, without permission in writing from the publisher. The Association of American University Presses Resolution on Permissions constitutes the only exception to this prohibition.
The paper used in this publication meets the minimum requirements of American National Standard for Information Sciences-Permanence of Paper for Printed
Library Materials, ANSI Z39.48-1984.
Manufactured in the United States of America
Library of Congress Cataloging-in-Publication Data
Willmott, H. P.
The last century of sea power : from Port Arthur to Chanak, 1894-1922 / H.P. Willmott.
p. cm.
Includes bibliographical references and index.
ISBN 978-0-253-35214-9 (cloth : alk. paper) 1. Naval history, Modern-19th century. 2. Naval history, Modern-20th century. I. Title.
D362.W68 2008
359 .0309041-dc22
2008015018
1 2 3 4 5 14 13 12 11 10 09
Dedicated to FY1645 and in Praise of Toleration, Uncertainty, and Dissent
CONTENTS
List of Maps
Preface and Acknowledgments
Part 1. Definitions and Terms of Reference
Introduction
ONE
The Sino-Japanese War, 1894-1895

Appendix 1.1. The Pacific and the East Indies in the Nineteenth Century
TWO
The Greco-Turkish War of 1897
THREE
The Spanish-American War of 1898

Appendix 3.1. The Actions in the Philippines and the American Order of Battle

Appendix 3.2. The Campaign on Puerto Rico
FOUR
The Shifting Balance of Power

Appendix 4.1. The German 1898 Building Program

Appendix 4.2. The Boxer Rebellion and South Africa
Part 2. From Port Arthur to Bucharest, 1898 to 1913
Introduction
FIVE
The Russo-Japanese War: The First Phases

Appendix 5.1. Submarines and the Russo-Japanese War

Appendix 5.2. The Japanese Attack at Port Arthur, 8 February 1904

Appendix 5.3. The Second and Third Blocking Operations

Appendix 5.4. The Order of Battle in the Action of 10 August 1904

Appendix 5.5. The Fate of Russian Ships with the 1st Pacific Squadron at Port Arthur
SIX
The Russo-Japanese War: The Battle of Tsushima and Its Aftermath

Appendix 6.1. The Fate of Russian Ships with the 2nd and 3rd Pacific Squadrons

Appendix 6.2. Japanese Warships Sunk and Damaged at the Battle off Tsushima

Appendix 6.3. The Campaign on Sakhalin

Appendix 6.4. The Opening of the Portsmouth Conference
SEVEN
The Dreadnought Naval Race

Appendix 7.1. British and German Battleship and Battlecruiser Programs and Construction, 1905-1914: Summary by Year

Appendix 7.2. British and German Capital Ship Programs and Construction, 1905-1913

Appendix 7.3. The Battleships and Battlecruisers of the Major Powers, 1905-1913
EIGHT
Prelude to the First World War
Part 3. From Sarajevo to Constantinople, 1914 to 1922
Introduction

Appendix Part 3.Intro.1 British Trade in the First World War

a. The Volume of Imports Entering British ports in 1913, 1917, and 1918 by Commodities

b. Cargoes and British Ports, 1913-1919

Appendix Part 3.Intro.2 Shipping and the Major Allied and Neutral Powers in the First World War: Gains and Losses
NINE
The War in Northern Waters

Appendix 9.1. The Battle off Heligoland, 28 August 1914

Appendix 9.2. The Battles off Coronel, 1 November, and the Falklands, 8 December 1914

Appendix 9.3. The Battle off the Dogger Bank, 24 January 1915

Appendix 9.4. The Battle of Jutland, 31 May-1 June 1916

Appendix 9.5. The Scale of Convoy Escorts Assigned to Various Ports, 1917-1918 and the Expansion of the British Navy between 1914 and 1918

a. The Scale of Convoy Escorts assigned to Various Ports, 1917-1918

b. The Wartime Expansion of the British Navy

Appendix 9.6. Convoys and Losses: May 1917-November 1918

Appendix 9.7. U-boat and Shipping Losses in the First World War

a. Total German U-boat, Aggregate Allied and Neutral, and British Shipping Losses to Enemy Action and Sinkings by German Regional Formations in the First World War

b. German Figures Relating to the Sinkings of Allied and Neutral Merchantmen by Submarines and by Regional Commands

Appendix 9.8 The Campaign against Shipping: High Seas and Coastal Convoy Sailings and Losses

a. Convoys to and from British Waters

b. Coastal and Short-Haul Convoys
TEN
Tsingtao and the Dardanelles
ELEVEN
Naval Support of Operations in Africa
TWELVE
Action in the Baltic
THIRTEEN
The Black Sea, Otranto Strait, and Other Matters

Appendix 13.1 Other German Units outside European Waters

Appendix 13.2 Russian Destroyer and Torpedo-Boat Losses in the Baltic and Black Seas
FOURTEEN
The Legacy of the First World War

Appendix 14.1 Warship and Auxiliary Losses, 1914-1918

a. Tabular Representation of Warship and Auxiliary Losses of the Allied and the Central Powers, 1914-1918

b. Battleships and Battlecruisers Lost in the Course of the First World War

Appendix 14.2 The Arrival of British Imperial and Dominion Formations in Europe
Part 4. Not So Much Finis as . . .
Conclusion
Appendix Conclu.1 The Battleships, Battlecruisers, Aircraft Carriers, and Cruisers with the British Navy, 1913 and 1935
Chronology of the First World War at Sea
Notes
Selected Bibliography
General Index
Index of Warships, Auxiliaries and Merchantmen, and Submarines
MAPS
MAP 1.1. The Japanese Perspective: The main theaters of operations in the Sino-Japanese and Russo-Japanese Wars, 1984-1895 and 1904-1905, respectively.
MAP 2.1. The Greco-Turkish War of 1897: The Thessaly and Epirus sectors.
MAP 3.1. The Battle of Manila Bay, 1 May 1898.
MAP 3.2. The naval battle off Santiago de Cuba, 3 July 1898.
MAP 5.1. The Russo-Japanese War: The initial Japanese operations.
MAP 5.2. The Russo-Japanese War: Second- and third-phase Japanese operations.
MAP 6.1. The journey of the 2nd Pacific Squadron from Europe to the Far East, October 1904-May 1905.
MAP 6.2. The action off Tsushima, 27-28 May 1905.
MAP 8.1. The Italian-Turkish and Balkan wars: The Balkan peninsula.
MAP 8.2. The Italian-Turkish and Balkan wars: The Aegean.
MAP 9.1. The North Sea: The perspective of the British Grand Fleet.
MAP 9.2. The Southwest Approaches, English Channel, and southwest North Sea.
MAP 10.1. The moment of the mine: The Dardanelles operation, 18 March 1915.
MAP 12.1. The Baltic theater of operations: The Russian perspective.
MAP 13.1. The Black Sea theater of operations: The Russian perspective.
MAP 13.2. The Mediterranean theater of operations.
MAP 13.3. The Adriatic theater of operations.
MAP 14.1. The dismemberment of Turkey under the terms of the Treaty of S vres.
PREFACE AND ACKNOWLEDGMENTS
V ERY RESPECTFULLY, AND in light of lengthening shadows of mortality, I would in these few lines set out two matters that together provide the raison d tre of The Last Century of Sea Power . The first matter, relating to one s own rationale as a historian, is something that I had never committed in public, but it is provided here because it forms the basis of the approach to the subject that in its turn gave rise, per se, to The Last Century of Sea Power . This first matter, the shaping of one s own philosophy and career, really has its basis in three episodes, only two of which will be presented here, both of which were only minutes in length.
The first of these episodes was really the first time I thought, and I was some 24 years of age and had just presented my first lecture. A matter of months before I had completed two years post-graduate study at university, and my thesis subject was the Liberal governments and the Navy Estimates/dreadnought building program, 1906-1910, and because I had worked on these subjects I understood navies and therefore I understood the U.S. Navy and therefore I understood the Pacific War, 1941-1945-or that was how logic (of a kind) ordained that I was the member of the department obliged to give this specific lecture. I gave the lecture and at its end my head of department came to me and told me that the lecture had been very good indeed and that he had much enjoyed it: he congratulated me and told me that the lecture had been very well organized and delivered. I went to my office, sat down at my desk, lit a cigarette-how things have changed!-and sat there a moment, and then the thought crossed my mind: I had described the Pacific War but had not explained any aspect of it. I realized, with a start, that my head of department really did not understand the difference between description and explanation, between narrative and analysis, and I resolved at that moment always to explain and never to describe.
I would like to claim that I kept this promise to myself but, of course, for all my best intentions I have confused the two repeatedly and on all too many occasions have failed to provide explanation. With the passing of time I have realized that single explanation really does present intellectual difficulty and indeed dangers, but while I would plead that I have consciously attempted to provide explanation, I would admit that the second episode has probably been more important and a greater single influence than this 1970 intention. The second episode came nearly three years later, in late 1972, as I watched the eleventh program of Jacob Bronowski s television series The Ascent of Man . This remarkable enterprise-the explanation of the relationship between the physical sciences, political philosophy, and politics-had as its eleventh subject of examination the certainty of knowledge and dealt with Einstein, Szilard, and Heisenberg s Principle of Uncertainty (1927). The one piece in this program, the eleventh hour of the eleventh, that I always will remember was the scene in which, crouched over a little stream, Bronowski made the statement that the basis of democracy is tolerance and the basis of tolerance is uncertainty, and that when men behave with the certainty of knowledge that has no test in reality, then one finishes in a place like this. The camera stepped backward and one recognized that Bronowski was in Auschwitz. He then stated that this was where so many members of his family were murdered and where their ashes were washed away, and that this is what happens when men aspire to the certainty of knowledge that has no test in reality, that this is what happens when men aspire to the knowledge of gods. I realized, at that one moment, that, whatever explanation I sought to provide, more important than knowledge and explanation were tolerance and uncertainty. From this one moment, before the screen, there stemmed over a period of time a desire to place before the reader choice that would encourage independent thought, to provoke questions rather than provide answers, and indeed in time came another thought that arose when attending a lecture and hearing myself quoted as an authority on the subject: a determination to write the counter-view lest the original idea and writings commanded acceptance and endorsement.
The second matter deals with the immediate origins of The Last Century of Sea Power , and one would admit that this book does have a rather unusual pedigree. Some four years ago I was considering various options, most obviously with reference to war and warfare at the present time, and was specifically intrigued by one thought-that the United States at the present time has the same capacity that European powers had with respect, or disrespect, to east Asia between about 1840 and 1890 and over most of Africa in the last thirty years of the nineteenth century. Both the European powers more than a hundred years ago and the United States at the present time had acquired the capacity to destroy what may be termed Third World polities, and they were able to wage total war without having to mobilize the economic, social, and military forces as was necessary in their conduct of the two world wars of the twentieth century.
I therefore determined to produce a three-volume enterprise, From Total War to Total War , the first volume from 1815 to 1864 and from the defeat of the Old Guard at Waterloo to the Union defeat at Cold Harbor, the second from the defeat of the French cavalry at Sedan in 1870 to Okinawa and a case of from kamikaze to kamikaze, and the third from Tokyo to Baghdad. But in putting together chapters for these three volumes a totally unforeseen problem emerged. I was working on the wars of national liberation of the mid-nineteenth century, and specifically on the war of 1866, when I was confronted by the realization of the manner in which the detail of this war has been combed from historical accounts over the last hundred years. What was an important series of events after the battle of K niggr tz that had been recorded in pre-1914 history books was largely lost after 1919, being of small account, and by the time I was working my way through university in the sixties what had been but little known had followed Austria into oblivion. My own interests took me in other directions, and four decades were to elapse before I appreciated the manner in which historical bookkeeping had all but ceased to provide proper account of these proceedings. It was only when I acquired books published in the 1880s and 1890s that I was able to acquire some knowledge and understanding of events after K niggr tz, but by the time that I did so another matter had intruded upon my deliberations: if events on land were subject to such treatment, then where stood matters naval?
In seeking answers to this question, I was beset by certain considerations. I have long entertained certain reservations on naval historiography, specifically with reference to three matters. In the two world wars, but specifically and more seriously the second, the defeat of the submarine campaign against shipping has been told primarily in terms of U-boat losses. Any detached consideration of events would seem to suggest that U-boat losses were extremely important in terms of German defeat, but the peak of U-boat numbers at sea in the Second World War was in April 1945, and therefore German submarine losses cannot tell a very full story in terms of overall defeat. Any attempt to explain that defeat must address other matters such as new Allied construction, control of shipping space, and the volume of cargoes safely arriving in port and their composition and relative importance in the maximum effective use of available shipping.
The second matter relates to what has been my main area of study, Japan and the war in the Pacific. My study of this war extends over more than four decades, more than eleven times the length of the war, which, I admit, does leave me a rather sad and pathetic case not least because my adding of knowledge has gone hand-in-hand with a lessening of understanding. There are so many questions about this war that defy understanding, and while this is not the place to parade some of the more obvious ones, just three may be cited for purposes of immediate illustration: the obvious question of when Japan reached the point of defeat, the extent of Japanese losses between the outbreak of war in December 1941 and the start of the American drive across the central Pacific in November 1943, and the scale of American losses between November 1943 and October 1944 in the course of the offensives that took the tide of war from eastern New Guinea and the Gilbert and Ellice Islands to the Philippines. The first of these questions defies easy or simple answer, though I would stand by my view that in conducting the strike against the U.S. Pacific Fleet at its Pearl Harbor base Japan passed the point of defeat. The second and third questions are more straightforward because both are concerned with numbers and the answers are perhaps surprising. Between 7-8 December 1941 and 19 November 1943 the Imperial Japanese Navy lost four fleet and two light carriers, two seaplane carriers, three battleships, four heavy and four light cruisers, forty-seven destroyers, and forty submarines, and such losses prompt an obvious question of whether Japan was assured of defeat on the basis of such losses. Intimation to that effect is patently absurd, and herein the obvious problem of comprehension arises: it is difficult to understand how the sum of the series of defeats incurred by Japanese forces between May 1942 and November 1943 represents national defeat. But what really does confound understanding is the fact that between 24 November 1943, when the escort carrier Liscome Bay was torpedoed off the Gilberts by the submarine I-175, and 24 October 1944, when the Princeton was lost off the northern Philippines as a result of attack by a land-based aircraft, Japanese shells, torpedoes, and bombs failed to account for a single U.S. Navy fleet unit other than the Fletcher -class destroyer Brownson , which was lost on 26 December 1943 off Cape Gloucester, New Britain, to air attack. In other words, the whole of the American effort that resulted in the breaking of the outer perimeter defense in the central Pacific, the carrier rampages into the western Pacific that resulted in the shipping massacres at Truk (17-18 February) and Koror (30-31 March), the landings at Hollandia and Aitape, which took the tide of war from one end of New Guinea to the other in two months, and which finally led to overwhelming victory in the Philippine Sea (19-20 June), cost the United States just one destroyer, plus the destroyer escort Shelton , which was sunk by the submarine Ro. 41 off Morotai on 3 October.
But if these matters represented the raising of real and genuine problems of understanding with reference to naval history, the third matter represented the problem of understanding of the present. If one looks, for example, at Stephen Roskill s definition of sea power, then one can see that certain of what he identified as the key ingredients of sea power-overseas bases, building and repair facilities, a merchant marine and fisheries with their supply of manpower-would seem to be, at best, under sentence. Any casual consideration of the present time would suggest a major, indeed in certain countries a fatal, contraction of shipbuilding capacity, the lack of large numbers of men used to and trained in the ways of the sea, and, perhaps most significant of all, the apparent absence of any real blue-water enemy for the major western powers. My own view is that in 1945 the most important of the services in the United States was the navy because it was by sea that army and army air force formations and units were moved overseas and supplied: the reach of the United States across two oceans was primarily naval and maritime. In a very real sense, in 1945 the U.S. Navy was primus inter pares , and it had its own army and it had its own air force; by 2003 the U.S. Marine Corps had its own navy.
These were perhaps the main ideas that went into the concept of The Last Century of Sea Power , but in seeking to put together a book on this subject everything seemed to go wrong from the start. The various dimensions of sea power-not naval power-seemed to add layer upon layer of material as they were brought to the pages where they would be subjected to analysis, and this left aside one simple fact, namely that in presenting The Last Century of Sea Power one deliberately set aside full and proper consideration of those aspects of wars at sea that have been afforded full, one is tempted to suggest over-full, consideration in most histories. I deliberately sought to bring the little known aspects of war at sea to center stage at the expense of these better-known, and indeed more important, aspects whether these be campaigns, battles, formations and units, and other like matters. But what had been intended to be a single volume became two as the first ran to its prescribed length around the end of the First World War, but in line with a general sense of perversity this volume was ended with Chanak and 1922 in order to allow the second volume to begin in Washington in 1921. . . .
Such were the terms of reference of The Last Century of Sea Power , and in the preparation of this first volume acknowledgement must be made to those who, over many years and whether in the form of general conversation, correspondence, lectures, or conferences, have provided me with the basis of knowledge and critical facility that made this work possible. To attempt to list these people is impossible, but they have the satisfaction of knowing that without them and their guidance this book could never have been written and also that they are not responsible for the various errors, the sins of omission and commission, that litter its pages.
Nonetheless, specific acknowledgement needs be made to certain individuals who have spent many hours helping me try to settle a host of difficulties that presented themselves in the course of this book s writing. I would specifically acknowledge and offer my sincere and unreserved gratitude for all the help and advice I received from Ersan Bas, Anthony Clayton, Kobayashi Go, Sally Paine, Geoffrey Miller, John Norton, Sarandis Papadopoulos, Tohmatsu Haruo, and Stephen Weingartner. And to these I would add those persons who were always at my side that went beyond the call of friendship, namely Michael and Sara Barrett, Bernard Cole, Michael Coles, Gerard Roncolato, William Spencer and Andrea Johnson and family, John Andreas and Tine Olsen, Jack and Gisela Sweetman, and Spencer C. and Beverly Tucker. To all of these people I would simply state my thanks and appreciation for help and camaraderie that are beyond my poor powers to acknowledge properly.
I would also acknowledge the support and encouragement provided by various colleagues and friends in a period of very considerable personal and professional misfortune and without whose quiet companionship what was bad might well have been nigh impossible. Among those I would acknowledge my debt of gratitude to Tim Bean, Patrick Birks, Nigel and Martine de Lee, Christopher Duffy, Paul Harris, Jack and Suzanne Hurley, Cliff Krieger, Jim Mattis, Lars Neilsen, George Raach, Kyle Sinisi, Frederick Snow, Patrick and Jennifer Speelman, David Vance, John Votaw, and David White, and with this acknowledgement I would state my hope that this book is some small token of my appreciation and esteem.
I also wish to acknowledge my debt to those without whose patience, tact, and literary ability this book would probably have gone the way of many of the ships cited in these pages. Specifically I would wish to acknowledge my debts to Robert Sloan and Brian Herrmann of Indiana University Press, to copy editor Sarah Brown, to Keith Chaffer for his professionalism and imaginative work upon the maps, and to the library personnel who professionally and personally have helped me at every stage of proceedings, Gareth Bellis, Edwin Finney, Ken Franklin, Iain Mackenzie, John Montgomery, Andrew Orgill, and John Pearce: I trust they will accept this poor acknowledgement of their support and efforts.
There remains one group that always appears in my acknowledgements section and for one reason: they have been the means of ensuring continuing sanity. I would acknowledge my debt to and love for my beloved woofers. Would that Everton, Sherry, Kondor, Jamie, Suki, and Lancaster be at peace and together, and in terms of my present debt and love for Mishka and Cassie, for Mishka and Cassie, and for Junior and Yanya, I would merely express my hope that much time will pass before they join their predecessors and chase together across the celestial fields.
H. P. W ILLMOTT
Englefield Green
Surrey
United Kingdom
4 October 2008
THE LAST CENTURY OF SEA POWER
PART 1

DEFINITIONS AND TERMS OF REFERENCE
INTRODUCTION
T HE COMING OF THE NEW millennium invited any number of histories, real and alleged, constructed on the basis of noughts. It is one of the curses of history that, depending on the prejudices of the writer, either a decade or a century is an age, its counterpart an era, and that in neighboring periods there are elements of contrast that so determine character. In reality, history concerns itself with elements of constancy and change, and very seldom affords consideration to simplistic, single-cause representation and for very obvious reason: for every complicated human problem there is a simple explanation, which is neat, plausible, and invariably wrong.

Within two decades of the end of the Second World War the British historian Stephen Wentworth Roskill (1903-1982), writing in The Strategy of Sea Power , set out definitions of sea power and its constituent elements. 1 One would provide one s own definition of the historical role of naval power, which is that
The purpose of sea power is to ensure in times of war those rights automatically commanded in times of peace, specifically the security of homeland and overseas possessions against raid and invasion and of sea-borne trade, while denying those same rights to an enemy in terms of the conduct of amphibious operations and attacks on shipping. 2
The crucial point herein is that while in a general war the offensive use of sea power in terms of assault or landing on enemy territory cannot necessarily be undertaken before and until a measure of defensive primacy has been secured, the line of demarcation between the offense and defense at sea is very different from that ashore, and battle itself is very different. The battle at sea does not possess those elements such as rivers, mountains, lines of communication, and settlement that ashore spell out the difference between offense and defense: the battle at sea has terms of reference supplied by latitude and longitude, daylight hours, and factors of time and distance that necessarily ally themselves with coastline and off-shore hazard. The battle at sea has to be fought repeatedly over the same reaches of sea and ocean in a way that the battle on land does not, and lest the point be doubted reference may be made to just one war and campaign. In the course of the Second World War the German offensive against shipping was defeated in May 1945. Various commentators-one hesitates to use the word historians -have tended to focus upon the month of May 1943 as the time when the German campaign againstAllied shipping was defeated, and it cannot be denied that in this month the German U-boat offensive suffered a defeat singular in significance. In this single month the German Navy lost no fewer than forty-one U-boats from all causes, and this total stands in very sharp contrast to the totals of nine, twenty-four, thirty-five, and eighty-seven U-boats lost to all causes in (3 September-31 December) 1939, 1940, 1941, and 1942, respectively. But the point was that the victory that was won by Allied forces in May 1943 had to be repeated until the very end of the European war, and Allied shipping had to be provided with escort and nonetheless took losses virtually to the very last day of the German war. The victory that was won in May 1943 was indeed repeated, most obviously in July-August and again in October-November 1943, and the victories that were recorded in these subsequent months were every bit as important as the victory won in May for the very simple reason that these subsequent losses were sustained by a U-boat service that had been re-organized, re-equipped, and committed afresh to the campaign in the North Atlantic. Losses in July and August 1943 were thirty-seven and twenty-five, respectively, and in October and November twenty-six and nineteen, respectively, 3 and in terms of the war at sea and the proper recounting of history the crucial point is to see these subsequent Allied successes in terms of complementary victories, not episodes complete in their own right. The victories that were won between May and November 1943 undoubtedly served to ensure that the initiative at sea passed finally and irreversibly into Allied hands, but the basic reality-that the defensive commitment remained until the end of the war and that the victories of 1943 had to be fought for and won every week, every month, of what remained of the war-cannot be gainsaid.

One is very conscious that in setting out such an argument one comes close to infringing upon a related matter, the impermanence of victory, which comes associated with its complementary point, namely that great powers are not powers that win wars but powers that lose wars but keep going: defeat and failure, and the reaction to failure and defeat, are the measure of a great power, not victory-the weakness of this particular argument being that it would suggest that perhaps the greatest of all powers was Austria-Hungary. That final point aside, the nature of sea power and its related parts demand definition. With reference to the latter Roskill set out naval power in terms of fleets and warships, industrial infrastructure and bases, merchant and fishing fleets, and trained manpower, and with reference to the merchant and fishing fleets and trained manpower there was the obvious link in terms of the former in part providing navies in times of war with manpower trained in the ways of the sea. The element of impermanence to which reference was made presents itself herein because one can seriously question, on the basis of Roskill s definition, what presently provides the foundation of sea power for such countries as the United States and Britain. These two countries were, at the end of the Second World War, possessed of naval strengths that rendered them impervious to challenge at sea, yet at the present time, in the first decade of the new millennium, Britain most certainly does not possess the industrial basis of Roskillian naval power and obvious question marks must be set against the United States in this same dimension: in 1990 the U.S. Navy had to go to suppliers in no fewer than eight countries in order to ready itself for war in the Middle East, a state of affairs that in industrial terms would have been unthinkable even twenty years previously, during the Vietnam War. No less obviously, the lack of shipping lines and trained manpower reserves were barely finessed by the U.S. Navy in 1990 when septuagenarians had to man the engine rooms of supply transports from the reserve that should have been scrapped and replaced at least one, perhaps two, decades earlier. The simple fact is that changed patterns of production, and certainly for European navies the lack of bases beyond Europe, necessarily have involved new definition of the role of naval power, and in ways that would never have been given real consideration even as late as the 1960s and 1970s. And to these matters there is a codicil. The Japanese dimension of the Second World War was unusual in the sense that this was a war decided by sea power, and it was very unusual in the sense that it was decided by sea power directed across an ocean, and it was a war that was won primarily by naval power. Certainly the American effort was necessarily joint in a way seldom properly defined, but the basic pattern was that land-based air power neutralized objectives, naval power isolated the latter, and amphibious assault then secured islands and airfields from which the process began anew. But the basis of this effort was naval: it was the sea that carried supplies for both the air and military endeavors. It was the U.S. Navy that was the main agency of American national power in the war in the Pacific, and at war s end the U.S. Navy had its own air force-the air groups of the fleet and escort carriers-and its own military, the U.S. Marine Corps. By 2003, and say it sotto voce , the U.S. Marine Corps had its own navy. The basic point herein is that in times of peace navies always fare badly: their costs are high and invariably they, historically, took second place to armies. The period immediately after the Second World War conformed to historical precedence, but in the 1950s and 1960s the navies of the great powers were to be strengthened institutionally by the vesting of the main elements of strategic deterrence in submarines, while carriers recovered if not their numbers then their relevance in the aftermath of the Korean War. With the end of the Cold War, however, navies have come under pressure on four counts: the strategic nuclear deterrence role may still be in place but it is of dubious relevance; the demands of peace-enforcement and peace-keeping necessarily are directed primarily to armies, not navies; navies, and particularly the U.S. Navy, have been overtaken by air forces and now are third in the defense pecking order; and the sheer cost of units in terms of money and manpower places obvious question marks against future role and capability. It may very well be that navies, no longer facing blue-water enemies and with very limited capacity to engage targets ashore, are in a decline that will mark the end of their role as defined by fleets in two world wars and summarized by Roskill, and that in the future their role will be with those secondary aspects of sea power, such as suppression of smuggling, piracy, and slavery, that previously have been dismissed as unbecoming. It is perhaps worth noting that by 2007 the British Navy had reached a point of decline, specifically in the number of warships in commission, that really pointed to a decline of status that corresponds to that of the nineteenth-century Dutch Navy.

But in setting out these matters, and these are presented as the basis of discussion and not definitive, one must note the obvious, namely that so many accounts of proceedings set out sea power and naval power as the same, whereas in reality sea power embraces very distinct naval and maritime parts. The story of sea power is not simply concerned with naval formations and units and with the nature and conduct of war, and operations, at sea: matters relating to merchant shipping, the volume and nature of sea-borne trade, and the relationship between (on the one side) power, industry, and commerce and (on the other) the sea need be defined because the various elements herein are complementary. This point is so obvious that it is very seldom addressed. For example, very few accounts of the campaigns against shipping in the two world wars of the twentieth century provide proper analysis of the merchant fleets, cargo capacity, and the ship-building and maintenance facilities of the major combatant states, and most certainly such matters relating to neutrals are even less well documented even though in the First World War the importance of neutral shipping to British survival and Allied victory is not to be under-estimated: in the Second World War such nations as Denmark and Norway, the Netherlands, Belgium and France, and Greece contributed significantly to eventual Allied victory by virtue of their shipping despite defeat and occupation by German forces. Likewise, while accounts of the defeat of the German campaigns against shipping concentrate primarily upon such matters as the gradual extension and increased effectiveness of convoy, the resultant reduction of shipping losses to manageable proportions, and U-boat losses, such matters as the volume of imports arriving in ports, the state provision of objectives and rationing priorities, and details of ship requisitioning are seldom afforded much in the way of attention and consideration in naval histories. The fact is, however, that between August 1914 and October 1918 the British Transport Department provided tonnage sufficient to ensure no fewer than 23,700,000 individual passages-the equivalent of one for every two people in the British Isles-while transports dispatched almost 50,000,000 tons of military stores and supplies, 500,000 guns and motor vehicles, and 2,200,000 animals from British ports to various theaters of operation. The total of military stores and supplies dispatched as part of the national capacity to wage war represented the equivalent of a year s total imports prior to the outbreak of war, yet in most naval histories these are matters that are seldom afforded even as much as historical footnotes. 4
Sea power, in its historical context, has been concerned with naval power and the use of the sea, and in seeking to set out the story of sea power and the twentieth century the two parts necessarily need to be considered together in setting out the definition of the twentieth century. How one defines the twentieth century is a matter of personal persuasion and prejudice, but leaving aside the dictates of chronological exactness and thereby discarding the 1901-2000 option, one would suggest that there are three possible naval matters that might provide a suitable start line: the Russo-Japanese War of 1904-1905, the Spanish-American War of 1898, and the Sino-Japanese War of 1894-1895. One would suggest that in many ways the claims of the second, the Spanish-American War, really do provide a suitable point of departure, and on two counts: it was primarily a naval war and one that involved the defeat of a European imperial power-albeit one long past its peak-by a non-European state, and it was a war in which one battle, fought on 1 May 1898 in Manila Bay, may be regarded as the last battle of the Age of Sail. It was an action fought on the one side by five American cruisers and two gunboats and on the other by four Spanish cruisers, three gunboats, and three other vessels. The Spanish warships, inferior in gun power and of dubious seaworthiness, were anchored under the cover of guns in the fortified base of Cavite. Why this action should be considered the last battle of the Age of Sail is on account of the nature of the action. It was fought without reference to mines, torpedoes, and submarines, and it was fought without reference to central gunnery control systems, radio, airships, and aircraft, all of which were to figure so prominently in war at sea in the twentieth century. It was a battle fought in line ahead with broadsides at ranges that were reduced to 250 yards/230 m, at which distance, and with no means of aiming other than the eye, thirty-nine rounds in every forty still managed to miss their intended target. It was an action that should immediately invoke thoughts of Quiberon Bay (20 November 1759), the Nile (1 August 1798). and, particularly, first and second Copenhagen (2 April 1801 and 2-7 September 1807) in terms of pedigree. Despite the fact that the warships in Manila Bay in May 1898 were steam-powered, the battle that they fought properly belonged to a previous age.
Conversely, the Russo-Japanese War presents itself as a possible start point on account of the fact that this war did see the employment of twentieth-century means absent from Manila Bay, airships and aircraft excepted. 5 But perhaps the most convenient start point for an examination of sea power in the twentieth century is provided by the Sino-Japanese War and on three counts. First, it is a war in which Japan emerged as the greatest indigenous power in eastern Asia and, in effect, as a great power, though recognition of that fact was not forthcoming until 30 January 1902 in the form of the alliance with Britain. 6 Second, it was a war in which naval power provided the basis of Japanese victory. Third, this war came at a time when a number of changes relating to navies, merchant fleets, and trade at sea were in hand, and these were to prove crucial in the unfolding of subsequent events. And, of course, there was the small matter of the events that followed the Treaty of Shimonoseki (17 April 1895) and that set the scene for the Russo-Japanese conflict. 7

Historical attention necessarily has focused upon the Triple Intervention, whereby France, Germany, and Russia combined to relieve Japan of the most important of the gains that it had registered at Chinese expense at Shimonoseki, and thereafter acquired various concessions for themselves. This sequence of events, along with the Sino-Japanese War itself, represents a convenient start line in an examination of sea power and the twentieth century for one immediate reason: such intervention on the part of European powers arguably would have been impossible, or at least intervention on the part of France, Germany, and Russia would have been impossible, before the last decade of the nineteenth century; it most certainly would have been impossible after that time. 8 The first part of this argument is one that needs be exercised with care because it could be asserted that in the period between 1840 and 1870 perhaps Britain alone (but more likely Britain and France together) had an unprecedented capacity to destroy east Asian states and to do so without undue effort: in effect, Britain (or Britain and France) had the ability to wage total war without having to mobilize their populations and industrial and economic resources in order to do so, a Western capability that was lost until the United States entered into such an inheritance in the 1990s. But even if this ability on the part of European powers really did exist, as opposed to being alleged in these lines to have existed, the real point is that the ability reflected developments that came together at this time.
The most important of these changes concerned Britain and the position of industrial, financial, and naval pre-eminence in the world that it had established for itself in the course of the nineteenth century. The last decade of that century saw first the United States and then Germany overtake Britain in gross national product and industrial output. With such differences in population and natural resources among these nations, such a situation, at least in retrospect, would seem to have been inevitable, a question of when, not if. Yet despite being relegated to third place in the industrial league table in the course of the 1890s, Britain retained its position as world leader in finance, shipping, and trade. This decade, however, marked the apogee of British shipping and trade, not in terms of size but in terms of relative position. The decade really marked the end of the sailing ship on major routes; though various kinds of sailing vessels continued to work coastal routes and even major international routes with bulk items such as wheat, barley, jute, nitrates, timber, and (ironically) coal, the 1890s saw the last of the major passenger sailing vessels on the Antipodes routes. The passing of the sailing ship from center stage was primarily the result of the development of high-pressure boilers after 1878 and of compound engines, specifically the triple expansion engine after 1881, which made possible the operating of ships of 15,000 tons and more on a profitable basis. These developments made possible the development of electricity in merchantmen, and this was to mean, with the first refrigeration ships, that produce of the outside world could be brought to European tables: the first refrigerated cargo from the United States to arrive in Britain did so in 1879, from Australia in 1880, from New Zealand in 1882, and from Argentina in 1884. 9
The last decade of the nineteenth century saw the various restrictive measures effected by different states assume significance for a British merchant marine that represented perhaps half the world s tonnage capacity but which had been involved in carrying three-fifths, perhaps as much as two-thirds, of all sea-borne trade. Such measures as state subsidies, mail packaging, and the preventing of foreign vessels working designated coastal routes-which in the case of the United States meant that trade between New York and Honolulu was designated coastal-meant that the merchant navies of a number of major states had secured a viable base on which to expand by the 1890s, and this had to be at the expense of their British counterpart. Over the previous two decades the increase in the size of shipping world-wide had been accompanied by major falls in freight charges, 10 but if in 1890 British steamships totaled 5,414,000 tons and the steamships of the rest of the world mustered 2,293,000 tons, the fact was that the subsequent expansion in carrying trade was to be shared: between 1890 and 1910 world-wide shipping tonnage all but doubled, with Britain, for the first time, falling below the magical 50 percent figure, though in terms of the building of merchantmen Britain still accounted for almost three-fifths of world production even as late as 1914. 11 As it was British building primacy was accompanied by innovation unmatched by any other nation in this same period. In 1894 a British yard produced, with the cargo ship Inchmona , five-cylinder four-stage expansion propulsion, and in that same year another yard constructed the famous Turbinia , which achieved an unprecedented speed of 34.5 knots. After 1902 steam turbines were generally adopted for passenger ships; the hybrid turbine and reciprocating engine were combined for the first time in 1906 and geared turbines made their first appearance in 1911, but it was in 1902 that a British yard built the first ship larger than Brunel s famous Great Eastern , 12 which, at 18,914 tons, had been launched on 31 January 1858 and was scrapped in 1889. The Celtic , however, at 20,904 gross registered tons (GRT) was very quickly surpassed in size: within a decade liners larger than 40,000 GRT were working the oceans; by 1914 two German liners exceeded 50,000 tons. 13 By this time the largest cargo ships were about 7,760 GRT-about the same size as the American-built Liberty ships of the Second World War-and were capable of carrying about 10,400 tons of cargo at a top speed of 13 knots. 14
T ABLE 1. T HE W ORLD S L EADING M ERCHANT M ARINES , 1870-1910

By 1914 three countries, hitherto not really in the lists, had emerged with major mercantile fleets. The most important of these was Germany, which in 1880 had been second tier, alongside France, Norway, and the United States, but which in the 1880s and then again in the first decade of the twentieth century possessed a merchant fleet that doubled in size. If it remained small in comparison to Britain by 1910 it was roughly double the size of the next largest merchant fleet. That fleet was Japanese. Even as late as 1880 there was no Japanese merchant fleet, and the massive growth of the Japanese merchant marine between 1890 and 1910 should be noted alongside two related matters. The ruggedness of the home islands placed a premium on coastal shipping, specifically small wooden ships. Such was the cost of steel in Japan-it was not until the First World War that Japan could produce steel that was cheaper than pre-war British and German imports-that in this period much of the better shipping that Japan acquired was built abroad, chiefly in Britain. In the period 1884-1903, 87.6 percent of all Japanese warships by tonnage were built abroad. The third country was Norway, which, of course, was not independent but was joined with Sweden until 1905. Perhaps somewhat surprisingly, as a union Sweden and Norway more or less matched Germany until about 1900, and their combined merchant fleets were second only to Britain. By 1910 the merchant marine of Norway alone ranked fourth in the world. 15
The contrast between these countries and the United States was most marked by the latter s inability to compete in the price of steel, ship-building, and labor. Only on the Great Lakes, where there was no competition worthy of the name, could American ships make money, and the fact was that whereas in 1830 nine-tenths of American sea-borne trade was handled by American ships, by 1890 this was at one-tenth of its 1830 level. Alone among the leading powers, the United States possessed a merchant marine that declined in size between 1870 and 1890-in part the legacy of the losses incurred during the Civil War (1861-1865)-and it was not until the turn of the century that her merchant fleet began to reach the size that it had possessed some thirty years before.

What was happening to trade and shipping in the last decade of the nineteenth century was the opening of the extra-European world, and this was so in two senses: there was the opening of a world beyond Europe that was either settled or dominated by Europeans-the nineteenth century in South America was known as the British century, when British industry and capital in effect replaced direct Spanish and Portuguese rule-and there was the opening of world markets by direct annexation or the acquisition of spheres of influence. Perhaps the most important single development in this process was the European division of Africa in the wake of the Berlin conference, 1884-1885. 16 In 1880 European presence in Africa was primarily coastal; the only areas where there was a general European presence and control were Algeria and southern Africa. By 1914 the only parts of Africa that were not under direct European control were Ethiopia and Liberia.
The European pre-eminence in Asia had been established primarily in the first half of the nineteenth century, what were perhaps the twin apogees of western power being the Anglo-French razing of the Summer Palace in Peking in October 1860 and the various bombardments, in June-August 1863 and September 1864, of Japanese fortresses by western navies, especially the attack on Shimonoseki, 5-8 September 1864, involving American, British, Dutch, and French warships. The latter bombardment formed part of the prelude to the civil war within Japan that resulted in the overthrow of the Tokugawa shogunate and the Meiji restoration of January 1869. In the event Japan, made acutely conscious of its weakness relative to western powers, was able to escape occidental domination primarily because the main focus of western attention was China. Japan was those extra miles and days beyond the markets of coastal and riverine China, and in that distance and time was the measure of protection that China lacked. In fact, the opening of Japan to western influence is a matter in which the United States has always claimed specific importance and relevance, the basis of this claim being Commodore Matthew C. Perry s mission, with its four black ships, 17 in 1853 and the resultant Treaty of Kanagawa (31 March 1854), under the terms of which the shogunate undertook to open normal trade relations with the United States and to return shipwrecked sailors. 18 This indeed was the first casting aside of the cloak of self-exclusion with which Japan had been clad for more than two centuries.
Seldom afforded any consideration of these events is that American intention resulted in a Russian squadron sailing from the Baltic at the same time as Perry s initial mission, and its ships stayed in Japanese waters after Perry returned to the United States. It was on the basis of this presence that Russia was able to secure the northern Kurile Islands in the Treaty of Shimoda of 7 February 1855. 19 This was but the prelude to gains that were to render Russia the main beneficiary of the occidental penetration of the Far East, albeit it was a process that was primarily military and land-based and not naval. The Treaty of Aigun 20 (28 May 1858) and the Convention of Peking (18 October 1860), respectively, provided for the Russian acquisition of the Amur region and the Maritime Province, the irony being that Russia, which had reached the Pacific coast in 1649 and (as part of the preliminaries to the October treaty) had established a military outpost on what was to become the city and naval base of Vladivostok on 2 July 1860, established Vladivostok before it had secured such old cities as Tashkent (June 1865) and Samarkand (May 1868) in central Asia and much nearer to home. 21 In fact, the Russian gains in the Far East predated the end of the campaigns in the Caucasus while the main Russian effort in central Asia unfolded after 1868. The process of conquest and assimilation within the Russian empire was more or less completed by February 1884, though it was not until the Akbal-Khorasan boundary treaty of 21 December 1881, along with the protocols of 30 January and 6 March 1886, that the border between Persia and Russia was designated, and it was not until 1897 that Russia and Afghanistan concluded a treaty that marked out, finally, the border that joined and separated the two countries.
The Russian conquest of central Asia was provided political, moral, and intellectual justification under the terms of the Prince Gorchakov memorandum, which was circulated throughout the capitals of Europe in December 1864. Annexation was justified in terms of contact with half-savage nomadic populations . . . whose turbulent and unsettled character make them undesirable neighbours and the need to exercise a certain ascendancy over such peoples. In such a situation, Russia was obliged to choose between bringing civilisation to those suffering under barbarian rule and abandoning its frontiers to anarchy and bloodshed, and in such a situation Russia, irresistibly forced, less by ambition than by imperious necessity, into this forward march, necessarily was committed to a forward policy as has been the fate of every country which has found itself in a similar position. 22 Herein was the basis of what all the countries involved in the scramble for empire in the second half of the nineteenth century embraced, mutatis mutandis , and which, by century s end, was over-laid by two basic points of self-justification: Christian duty and the concept of racial superiority. In these matters, as one has noted elsewhere, the nineteenth century was about the cultivation of hatred, the twentieth century about the reaping, 23 but the basic point was that the elements of racial contempt and hatred were in place in the last decade of the nineteenth century in terms of European treatment of east Asians and provided the rationale as the last extra-European territories, in the Pacific, became the object of imperialist aspirations on the part of various powers.
Within the Pacific by the last decade of the nineteenth century the capacity to establish empire had largely passed. The French secured Tahiti and the Marquesas in the 1840s, by which time the British had established undisputed ownership of the whole of Australia and New Zealand and had secured Singapore and Malacca. Hongkong was secured as a result of victory in the First Opium War (1839-1842). Labuan was secured in 1846, but another four decades were to elapse before the British position of domination of northern Borneo manifested itself with the acquisition of Northern Borneo (1881) and Sarawak (1888). With Britain acquiring the Fiji Islands in 1874, the various island groups of the southwest Pacific were then secured: the Ellice Islands in 1886, the Gilbert Islands in 1892, and the middle and lower Solomons in 1893. With the Dutch in general control of the Indies-though Bali was not finally pacified until April 1908, more than three hundred years after the Dutch first arrived in the Indies-the only major territory not properly designated was New Guinea, the northeast part of which became German and Kaiser Wilhelmsland in 1884. Papua was acquired at this time by Britain and passed to Australia in 1906. But with the United States by this time having established itself in the Hawaiian Islands and having taken possession of the Midway Islands in 1867, Pago Pago in 1878, and the Pearl Harbor station in 1887, by the 1890s there were no areas for peaceful expansion, no indigenous territories into which major powers-whether old, established empires or newcomers-could move, at least not move and acquire without serious war.

The last decade of the nineteenth century saw development of naval power that ran in parallel with what had happened to merchant fleets and ships. The most obvious similarity was that this decade marked the apogee of British naval primacy, though a certain care needs be exercised in this matter because the losses incurred by the Japanese and Russians in the war of 1904-1905 and the effect of the Dreadnought allowed Britain a few years of grace. Even more conspicuous was the change of the profile of warship construction, which in this decade resulted in the development of types and silhouettes that are naturally part of the twentieth-century seascape.
The most important of these latter developments concerned battleships built of high-quality steel and equipped with quick-firing guns. The Royal Sovereign class, the first of which were laid down in 1889, were the first British battleships to have all-steel armor protection and they were the first British battleships to be given a secondary armament that consisted of quick-firing guns, initially 4.7-in./1-120-mm but later 6-in./152-mm guns. This combination of steel construction and the development of quick-firing guns was to transform the battleship with a restoration of freeboard as battleships began to assume one characteristic that was to become the standard feature of dreadnoughts. The members of the Royal Sovereign class, the first battleships to have steel armor and displace more than 12,000 tons, carried all their armament on the weather deck and all their secondary armament in casements; with their high freeboard they presented a balanced, symmetrical profile that contrasted very sharply with that of two decades of sullen and misshapen misfits. 24
In a very real sense the Royal Sovereign class represented two markers, two elements of constancy, at a time of very real change. The seven members of the class were laid down between July 1889 and February 1891 and were completed between May 1892 and June 1894 at an average cost of 944,140. The class represented a settled design that successive classes followed. The Barfleur, Centurion , and the Renown , the Majestic class battleships of the Spencer program of December 1893, and the Canopus class battleships of 1896-1897 represented a search for smaller, and less costly, battleships before the Formidable class of 1897, the London class of 1898 and 1900, and the Duncan class of 1898 and 1899 represented a return to the dimensions of the Majestic class at a seven-figure cost. Only the Implacable from the Formidable class and the Bulwark from the London class cost less than 1,000,000, while average cost of the six units of the Duncan class was almost 1,100,000 and the Prince of Wales , from the London class, cost almost 1,200,000. Settled design and profile, therefore, came at a very considerable increase of cost, and with the new century the size and cost of battleships climbed even more quickly: the eight battleships of the King Edward VII class, laid down between March 1902 and February 1904, displaced between 15,610 and 15,885 tons and cost 1,344,804 on average, while the Lord Nelson and Agamemnon , built between 1905 and 1908, displaced some 16,000 tons and cost 1,541,443 exclusive of armament. In a little less than a decade, therefore, the cost of individual ships rose by four-fifths (i.e., the average cost of a Lord Nelson class battleship compared to the Albion of the Canopus class), and in just six years rose by more than half (i.e., the average cost of a Lord Nelson class battleship compared to the Formidable ).
The second element of constancy that the Royal Sovereign class represented was Britain s determination to maintain itself as the world s foremost naval power. The Royal Sovereign class was the first built under the provisions of the Naval Defence Act of March 1889, which authorized the construction of seventy warships between 1889 and 1894 at a cost of 21,500,000. So expensive a program had been foreshadowed by the report in the previous month, February 1889, which stated that Britain would be pressed to conduct a naval war against a single enemy- altogether inadequate to take the offensive in a war with only one Great Power -and would be out-numbered by a combination of two powers. Of course a great deal depended on the identity of these two powers, and the fact was that there were only two powers that could combine to leave the British inferior to themselves-France and Russia. It was the combination of these two powers that resulted in the emergence of the two-power standard on which Britain based her naval construction requirements over the next dozen years. In reality, in 1889 France and Russia were not allies and there seemed no real prospect that they might be, though this was to become fact in August 1892.
From the outset the Naval Defence Act s main provision, that Britain had to measure itself against a two-power standard and that this was in effect a Franco-Russian combination, was wholly unrealistic, and for very simple reason: the Franco-Russian alliance of 1892 was concerned not with Britain but with Germany and her alliance with Austria-Hungary. The alliance between France and Russia came about as a realization of mutual vulnerability, a matter that had been obvious to France ever since the war (and the defeat) of 1870-1871, but was a more recent development on the part of Russia. The latter had come to the realization that Germany sought to balance between Austria-Hungary and Russia but would never support Russia in its dealings with the Hapsburg monarchy. There was on the part of France and Russia in 1892, to invoke a saying that was more than one hundred years old, the recognition of the need to hang together lest they be hanged separately, and the rapid provision of military clauses to the Franco-Russian treaty of alliance secretly provided the evidence of its anti-German, not its anti-British, nature. There might be areas of dispute beyond Europe between Britain and France and between Britain and Russia, and these were serious and major impediments to good relations between Britain and the other two powers, but in the final analysis no European country was ever going to risk a major war with Britain for the sake of naval and non-European matters; no European country was prepared to risk major complications within Europe, a compromising of position relative to a real enemy, on account of matters beyond Europe s shores. And that left, of course, Japan and the United States out of this particular equation.

One would suggest, therefore, that the essential background to the Sino-Japanese War of 1894-1895 was-at least in its occidental setting-the imperialist urge, backed as it was by various political and moral imperatives, a pattern of industrial and trading development, and the search for markets that represented fundamental change from the situation that had prevailed half a century previously, and an ability to project military power on scales and across distances that were unprecedented. A certain care, obviously, needs be exercised on this latter point: the British captured Manila from the Spanish in October 1762, and whatever European operations happened in the Far East in the nineteenth century, none represented projection of force across a greater distance than this singularly seldom-remembered episode.
The events of the late nineteenth century unfolded at a time of great change in ships, whether naval or mercantile, but there was one more matter at work within western societies at this time: the 1890s really marked the emergence of the first literate generation in western society. Such a statement invites the obvious counter that literacy was very uneven and that east and southeast Europe were some way astern of other parts of Europe and North America in this regard. That cannot be denied, but the basic point remains, and one would note that the process of industrialization necessarily involved a literate proletariat and for obvious reason. The nineteenth century as a whole saw the very slow growth of literacy within Europe, but after 1870, and with Germany adopting compulsory primary education, the process was to gather pace until, by the turn of the century, a complete generation in western and central Europe had gone to school. The connection between this development and imperialism lies in two matters, the first being the emergence of a press that was populist and sensationalist, and which is best described in terms of its whereabouts, the gutter.
Perhaps it is more accurate to refer to a generation that was semi-literate and whose opinions were increasingly shaped by a popular, tabloid press. The first newspaper chain began life in 1879, and the Hearst press in the United States was founded in 1895, one year before the first British tabloid, the Daily Mail , appeared. The emergence of a sensationalist press that pandered to the semi-literacy of a population, an increasing portion of which was enfranchised, was vitally important in the emergence of a xenophobic paranoia throughout Europe that was to contribute significantly to the atmosphere of crisis in the ten years before 1914 and which made war, when it came, so welcome in all the major capitals of Europe. A vulgar press and its readership, their diet of imperialism, and a steady stream of reports of war beyond Europe that provided an element of brutalization and added to the value of white-supremacy currency were present in this process, but there were deeper currents at work. The First World War, when it came, was not the product of the labors of a scurrilous press intent on servings of nationalism and militarism or the result of the scribblings of the likes of Erskine Childers and William LeQueux. 25 Certainly popular sentiment-in the form of nationalistic sensationalism and emotionalism-ran ahead of any serious clash of interests, and certainly the fervent nationalism throughout Europe that greeted the outbreak of war in 1914 was in part the product of years of demonology in the gutter press. 26
But, and second, in terms of the last decade of the nineteenth century, imperialism, and navies, the emergence of what might be called the first book-reading generation was significant in one matter: the last decade of the nineteenth century witnessed the publication of Alfred Thayer Mahan s The Influence of Sea Power upon History, 1660-1783 and The Influence of Sea Power upon the French Revolution and Empire, 1793-1812 . Published in 1890 and 1892, respectively, these books provided the title deeds of imperialism and navies. The books owed more to the turn of phrase rather than serious analysis-not that such a matter would ever concern naval officers-but they provided an interpretation of history that stressed the importance of financial power, trade, colonies, and sea power. In a book-reading society the impact of these books was immense, and in a sense raised the horizon, the imperialist and naval horizon, for much of Europe and the United States.
Mahan s works and time came together. Had these books been published in the 1870s, perhaps even the 1880s, their impact would have been minimal; they would have come to possess value as historical curiosities and no more. But for a public that had begun to equate national greatness with navies and empire, their impact was immense. The popular acclaim that greeted Mahan, the indulgence shown him by European royalty, and the use that was made of his writings over the next two decades on a host of naval matters were evidence of the potency of the written word at this time. In one respect, Mahan s books may be thanked for the growth of fleets in the quarter century before the outbreak of war in Europe. Fleets had to be funded and navies were massively expensive: monies had to be found, taxes levied, and the public and potentially volatile electorates had to be persuaded of the value and importance of this effort. Herein, the press, a reading public, imperialism, and navies were linked and in a sense dependent upon one another within a political system that for the most part saw states in the nineteenth century embrace democratic form if not necessarily democratic substance, and in certain cases embrace democratic form in order to avoid democratic substance. 27
Mahan, his writings, and the press provided justification for imperialism and navies just as empires commanded public attention and endorsement, and it was this rising temper within occidental states that was related to increasingly strident nationalism, an increasingly vicious racism that was linked to pseudo-sciences such as craniology, ethnology, phrenology, and eugenics and others of similar ilk that struggled on the brink of acceptability. There were twenty-nine phrenological societies in Britain alone in 1832, and as late as 1883-one year after the publication of Ratzel s Anthropogeographie -their very serious studies were still being duly reported in equally very serious journals. 28 Such activities, coming together with such influential publications as de Gobineau s four-volume Essai sur l Inegalit des Races Humains (1853-1855), 29 were helping to shape a more sharply conscious patriotism that was increasingly identified with race, the notion that what marked different nationalities apart in Europe was not cultural but racial and genetic. And, of course, such assumed qualities were justification for one people-a white people-to take charge of another-a non-white people-under the label of imperialism. Thus, Mahan was building on a foundation that was already in place. These elements that provided the social and intellectual basis of the state, power, imperialism, and concepts of nationalism and racism were there, and Mahan s popularity lay in the fact that just as his histories embraced the Thomas Carlyle concept of the Great Man, so they provided the rationale, the raison d tre, for navies, the justification for states hitherto not naval and imperialist to reach across the oceans.
CHAPTER ONE

THE SINO-JAPANESE WAR, 1894-1895
T HE SINO-JAPANESE WAR , July 1894-April 1895, fits into the context discussed at the end of the introduction with one crucial exception: the racial dimension. But overall the background is provided by the obvious point of contrast: Japan, by a very deliberate process of imitation, had been able to absorb western organization and methods and to provide itself with a military capability that by the last decade of the nineteenth century marked it as perhaps the most powerful single state in eastern Asia, whereas China s process of fragmentation had assumed critical-if largely unsuspected-dimensions by this time. The war was not one that was sought by either side 1 but arose from events in Korea that possessed singular importance to Japan: the Korean peninsula was potentially the point of invasion-the Mongol expeditions of 1274 and 1281 had shown this-and at the same time it was Japan s obvious point of entry on the mainland-as witness the Hideyoshi Toyotomi (1537-1598) expedition of 1592-1598. Both China and Japan saw Korea as lying within their own spheres of influence, China by right of historical precedence and Japan in terms of future intent. With the Tientsin Convention of 18 April 1885, the two states had agreed on a treaty that in effect provided for Korean independence but also for their rights of intervention in Korea and the obligation to consult with one another. But war was to come in 1894 as a result of a complicated power-struggle within Korea that prompted separate Chinese and Japanese intervention and that set in train a series of events that led to confrontation.
The war is one that has commanded little in the way of western attention, a state of affairs that one suspects owed itself to three facts of life. First, the war was immediately overshadowed by the Triple Intervention and was then overtaken by the Russo-Japanese War, 1904-1905. Second, the war proved to be a one-sided affair and the victories that were won by the Japanese were simple and overwhelming; the war was nothing more than the story of successive and easy Japanese victories-Port Arthur, 2 with a 10,000-strong garrison, was taken at the cost of 66 dead and 353 wounded-and hence there was little in the way of real historical interest or lessons to be learned. Third, this was a conflict derisively known in the west as the Pigtail War, 3 and with such a name there was an element of racial disdain that complemented the second point: clearly, for most westerners, there was little to be learned, and very little of interest, in a conflict between two manifestly inferior races. Nonetheless, this was a war that saw the first actions between major forces at sea since Lissa (20 July 1866) and the first action involving modern warships. In a perverse sense, however, this was a war in which naval power was of secondary importance, a state of affairs that sits uneasily alongside the very obvious fact that Japanese forces could not be deployed to Korea and northern China other than by courtesy of sea power.
The secondary status of naval power in this war stems from the fact that the initial deployment of military forces to Korea by sea by both China and Japan took place before the outbreak of war, and after the declaration of war on 1 August 1894 both sides used transports to move additional forces to Korea without let or hindrance by the other side. The immediate defensive commitment, to provide escort to troopships, and the limited range and endurance of warships meant that the first moves by sea were unopposed on both sides. And there was the episode of 25 July that obstinately refuses to accord with this portrayal of events.
Between 21 and 23 July ten Chinese transports left Taku for Korea, 4 and early on the morning of 25 July the Chinese cruiser Tsi-yuen and gunboat Kuang-yi were off Phung-do Island, off Asan Bay, in anticipation of a rendezvous with the gunboat Tsao-kiang and transport Kowshing when they fell into the company of three Japanese cruisers, the Akitsushima, Naniwa , and the Yoshino . 5 Apparently the Japanese salute was met by Chinese fire on the Japanese ships and in the resultant action the Tsi-yuen escaped while the Kuang-yi was run aground in Caroline Bay, where it was engulfed by a fire that set off the magazine. 6 At the same time the Tsao-kiang and Kowshing appeared on the scene, the Akitsushima s presence being sufficient to ensure the surrender of the Tsao-kiang without a fight; along with its auxiliary sail, it was to enter Japanese service as the Soko . The Kowshing , however, proved somewhat more difficult. A British ship flying a British flag, it had on board some 1,100 Chinese troops and, with the main action over, it was ordered to follow the Japanese cruisers to port. The Chinese troops tried to seize the ship in an attempt to return to Taku, and in these circumstances the Naniwa , the captain of which was a certain Togo Heihachiro (1848-1934), sank the Kowshing; there were very few survivors, whether Chinese troops or crew. Thereafter, with the declaration of war following this incident, 7 both sides sent troops by sea to Korea, the Japanese via Fusan (6 August) and Chemulpo (12 September) and the Chinese via the ports on Korea Bay at the mouth of the Yalu. It might also be noted, en passant and because the episode seldom commands as much as a mention, that Japanese forces, in addition to the landings at Fusan in southeast Korea and Chemulpo, on the west coast, also landed at Gensan, on the east coast of northern Korea, on 26 August. 8
In the war that followed there was only one major action at sea, the Battle of the Yalu River/Haiyang, 17 September 1894, though most accounts assert that there were two major actions, the Yalu and Wei-hai-wei, 2-12 February 1895. 9 In reality the latter was not so much an action, or even a series of actions; what happened at sea was the postscript in a siege that began with Japanese landings at Yung-cheng on 20 January 1895 and that resulted in the Japanese capture of the five forts covering the southern approaches to Wei-hai-wei by 30 January. 10 The town, with its harbor, was taken, without opposition, on 2 February, but the naval dimension to this episode concerned the reduction of the forts on the twin islands of I-Tao and Liu-kung and dealing with Chinese warships as they sought either to give battle or to seek safety in flight. 11 Once Japanese forces had secured forts, town, and port, and indeed even before they had done so, the possibility of the Chinese putting together an effective joint operation, involving coordinated efforts by their military and naval forces, was all but nonexistent. What was to happen was the progressive enfeeblement of the Chinese naval force at Wei-haiwei as a result of bombardment by Japanese forces ashore and a series of attacks by Japanese warships between 30 January and 6 February on Liu-kung and units in the outer anchorages. In this time, and as a prelude to February 1904, night attacks by Japanese torpedo-boats accounted for one of the two largest Chinese warships on station, the Ting-yuen , and three other units. 12 With six Chinese gunboats sunk at various times when they tried to escape from the doomed base, the main Japanese effort began on 7 February, and two days later the protected cruiser Ching-yuen was sunk by fire from one of the captured Chinese forts; I-Tao was subjected to landings and cleared, and by 11 February the guns in the fortresses on eastern Liu-kung had been silenced, again primarily as a result of fire from the captured fortresses. On the following day the Chinese admiral, Ting ju-ch ang, asked for terms of surrender, which were afforded on the 13th (and after Ting s suicide) and which were generous: a total of 5,124 Chinese service personnel were granted immediate parole. Less fortunate were the battleship Chen-yuen , the cruisers Ping-yuen and Tsi-yuen , the Kuang-ping , and six gunboats, which were surrendered. 13
The one action that has always been noted, the Battle of the Yalu River or Haiyang Island, depending on individual preference, was fought in the immediate aftermath of the Japanese victory over a Chinese force in front of Pingyang, which forced a general Chinese withdrawal northward to the Yalu. On the morning of 17 September a Japanese formation, operating in support of the army formations then advancing north from Pingyang, intercepted a Chinese formation that the previous day had put men and supplies ashore at Tatungkao. 14 This Chinese formation consisted of two squadrons deployed in line abreast with one line consisting of the of the third-class unprotected cruiser Tsi-yuen , the dispatch vessel Kuang-chia , the third-class protected cruiser Chih-yuen , the armored cruiser King-yuen , the second-class battleships Ting-yuen and Chen-yuen , the armored cruiser Lai-yuen , the third-class protected cruiser Ching-yuen , and the third class unprotected cruisers Chao-yung and Yung-wei , and the other line with the Kuang-ping and armored cruiser Ping-yuen along with the torpedo gunboat Kuang-yi and two extra units, the Fu-lung and Choi-ti , which apparently were torpedo-boats. The Japanese force consisted of three formations, a scouting group that consisted of the Akitsushima, Naniwa, Takachiho and the Yoshino , a battle force that consisted of the cruisers Chiyoda, Fuso, Hashidate, Hiei , and the Itsukushima , and a command group with the formation flag in the cruiser Matsushima , which was in the company of two dispatch vessels, the gunboat Akagi and the armed merchant cruiser Seikyo . Perhaps rather strangely, the last of these ships had a British master and the passengers on board included the chief of staff of the Imperial Japanese Navy (the Kaigun ).
Accounts of this action generally have described events under four headings. The first, simply, is that the Chinese opened fire at a range of about 6,000 yards/5,500 m, which was extremely long-range for that time, but failed to record any hits, and second, the Japanese formation, with marked superiority of speed, was able to close range and to concentrate against part of the Chinese line, with predictable consequences initially for the Chao-yung and then the Yung-wei . Third, no account is complete without the Chinese commander leaving his bridge and failing to exercise command, which evolved upon one of his foreign assistants, a former ensign in the U.S. Navy. It is very western, and most certainly very American, that such a person, of obviously exalted rank, should have been called upon to exercise a command that was beyond the Chinese admiral. And, finally, no comment on the Chinese force is complete without various references to alleged corruption relating to guns and the state of the ships. One is at a loss to assess the credibility of some of these allegations-such as that some of the guns on one ship had been sold on the black market and that there were shells that hit Japanese ships but failed to explode because they were packed with cement and not explosive. But in a sense assessment is unnecessary because these matters describe rather than explain, and the description and explanation of the outcome of this action may be given in a single sentence. The battle was not an action between two fleets or two formations: it was an action between a Japanese fleet or formation and a collection of Chinese warships. In this sense and in terms of tactical handling and result, this action was very similar to Manila Bay in May 1898 and Tsushima in May 1905.
The contrast between the two forces at the Yalu is no more obvious than with reference to the types of ship that were present. The Japanese scouting force consisted of one third-class cruiser, the Akitsushima , and three second-class cruisers that had rated at time of construction as the most powerful (the Naniwa and Takachiho ) and fastest (the Yoshino ) cruisers in the world. The main force consisted of two units (the Fuso and Hiei ) that were almost twenty years old, one of which had been rebuilt, and were a cross between cruisers and coastal defense ships. The remaining units, plus the Matsushima , were cruisers that stood comparison with their contemporary opposite numbers in foreign navies. The Chinese ships defy ready definition. In the main force there were, it appears, three pairs of sister ships-the Chen-yuen and Ting-yuen , the King-yuen and Lai-yuen , and the Chao-yung and Yung-wei -though the difference between these pairs was such that obvious questions of compatibility in handling and role present themselves even before such matters as the breakdown of communications within the Chinese force, on account of the Ting-yuen losing its masts and yards, and the fact that there appears to have been no gunnery practice by Chinese ships for several months, are weighed on the scales.
With the Chinese formation bound by the 6 knots of its slowest members, the Chao-yung and the Yung-wei , and unable to take advantage of the superior number of heavy (12-in./305-mm, 10.2-in./259-mm, and 8.2-in./208-mm) guns, the Japanese were able to close to a range of 3,000 yards/2,750 m. the Chinese problems of coordination being confounded after the Ting-yuen s misfortune by the pre-arranged recourse to the Chinese ships fighting in pairs and thus leaving each pair liable to defeat in detail. The Chao-yung and Yung-wei and the Chih-yuen and the King-yuen were sunk, while the Kuang-chia was run aground in order to avoid sinking but was lost; the Ting-yuen and Lai-yuen were both seriously damaged, but the surviving Chinese warships were able to reach the safety of Port Arthur. They did so primarily because the Japanese did not press what was a clear advantage, presumably because the Japanese commander, Admiral Ito Sukeyuki (1843-1914), was only too aware that what he had under command was not to be risked inasmuch as it represented more or less the sum of national strength and a clear victory had been won. The fact that Japanese ships had been fairly liberally peppered-though only the Hiei was obliged to withdraw from proceedings on account of its damage-was also a factor in Ito s calculations. 15 Moreover, given the fact that the action lasted most of the hours of daylight on this single day, the Japanese warships must have been low in ammunition and probably could not have prevented the escape of Chinese ships, but the latter was inconsequential: the Chinese naval forces in the north had been neutralized and reduced to little more than fugitive status. The Japanese, with the advantages of geography, placed their warships in a position of dominance of these northern waters and won a comprehensive victory.
The measure of Japanese success can be gauged by the fact that in the aftermath of this action the Japanese force advanced into northern Korea and, with the flank now secured, was able to cross the Yalu into Manchuria (24-25 October). At the same time another force, the equivalent of two divisions, was put ashore at the head of the Liaotung peninsula, Port Arthur itself being taken on 19 November after a Chinese defense that varied between the feeble and nonexistent. 16 The Japanese warships were unable to prevent the escape of the Chinese warships from the base, but these chose to make their way not to the north or west but to the south, to Shantung province and Wei-hai-wei where, to mix metaphors, they were no more than condemned men after sentence. 17
At the time the official Chinese line was to place responsibility for defeat at the Yalu upon the responsible commanders, specifically Ting ju-ch ang. In one respect that was accurate: it seems that the Chinese admiral suffered some form of concussion when his flagship fired its first shells, the blast demolishing his bridge and killing a number of officers there assembled. But if Ting s conduct of operations, or perhaps more accurately his failure to conduct operations, was indeed in part the cause of defeat, this line of argument really is misrepresentation. The Chinese could have had the most gifted of commanders and intact communications but the result would have been essentially the same because the difference between forces, the margin between victory and defeat, was not personal but systemic. The basis of Japanese victory lay in the simple fact that in terms of poor organization and quality the Chinese force that was put into Korea Bay was a force that could trace its pedigree to Chinese defeats at French hands off Foochow in 1884-an action all but unknown to history-and at British and French hands between 1839 and 1860. Therein was the point of surprise: the extent to which Japan had implemented change and had absorbed western form in order to avoid subjugation was but little understood at the time. Certainly China, in spring 1894 when its leadership took what was a conscious decision to deal with matters Korean and Japanese by force, had never anticipated Japanese power and effectiveness. But the same was true of the watching occidental powers, and hence the subsequent events.

The Sino-Japanese War was one of those events when the world turned. People at the time, or at least certain people at the time, realized as they watched events unfold that they were witness to something different: they did not know what would follow but in China s defeat at the hands of another Asian state was something that was without precedent. 18 The war was to possess singular significance for the balance of power between China and Japan, but no less importantly it placed Japan in the ranks of those outside powers that had sought a delicate balance in addressing Chinese matters. These powers had sought to deal with a Chinese state too weak to oppose them but strong enough internally to be able to implement the various concessions they sought to extract from her. The war of 1894-1895 revealed an unsuspected level of Chinese weakness, and Japan s defeat of China in effect ushered in the final phase of China s imperial system and the scramble on the part of western powers to secure concessions, to carve out for themselves spheres of influence and privileged positions within China, and in this respect Japanese efforts proved self-defeating. Fear that a major Chinese defeat in the north, and specifically in the area of Peking, would induce foreign intervention prompted the Japanese moves against Wei-hai-wei and Taiwan, but the full measure of Chinese weakness was not lost on the watching international community.

The Sino-Japanese War was ended by the armistice convention 19 and the peace treaty that were concluded at Shimonoseki on 30 March and 17 April 1895, respectively, the latter being ratified by the two sides at Chi-fu on 8 May. 20 The terms provided for Chinese recognition of Korean independence and renunciation of all claims of suzerainty over that Empire, and the surrender to Japan of the southern Fengtien province (i.e., the Liaotung peninsula and Port Arthur), Taiwan, and the Pescadores. In addition, Japan was afforded access to four treaty ports hitherto closed to her-Chungking and Shashih, and Hangchow and Suchow-and the necessary rights of passage on the Yangtze and the Woosung and canals, respectively. 21 In addition, Japan secured an indemnity of 200 million gold taels ( 25,160,256) to be paid in eight installments over seven years and with an interest rate set in the event of default.
At Chi-fu, however, there was an unwelcome presence: anchored off the port was a Russian naval task force deliberately sent to underline the friendly advice offered to Japan on 23 April that the Liaotung peninsula should be restored to China. Russia had designs on Port Arthur and, with the support of France and Germany, sought to ensure that Japan relinquish this specific acquisition. The formal process by which Japan did so proved somewhat difficult and protracted, and certainly more difficult and protracted than the original negotiations, in large measure because Japan s return of southern Fengtien had to be balanced against provision for increased indemnity. 22 But from the time of the Triple Intervention there was no question of Japan s attempting to resist this demand. 23 The war with China had proved very expensive for a country so poor as Japan, and with Britain, the only power with the navy and money that might thwart Russia and its associates, not prepared at this time to countenance either war or the financing of Japan, the latter had no option but to bear the unbearable. 24
The result was that while Japan committed herself to Gashin Sh tan , Perseverance through Hardship, as the full cost of the 1894-1895 war manifested itself, the other powers moved to establish themselves in the vacuum created by Chinese and Japanese weakness. The first move was by Russia in the treaty of 3 June 1896 under the terms of which it secured major concessions in Manchuria with reference to the building of the Trans-Siberian and Chinese Eastern Railways. In truth this was only part of the process whereby Russia, with ambitions in Korea and northern China that ultimately could not be reconciled with those of Japan, sought to establish itself as protector of China, or perhaps more accurately sought to secure China as a client state. The interesting question, of course, was against which power China needed to be protected, the Germans having occupied Tsingtao on 14 November 1897 and secured from China formal recognition, and acceptance, of the fait accompli in the treaty of 6 March 1898. 25 Whatever the answer to this question, in seeking to establish itself as the leading power in the area Russia was not averse to displays of force against China, witness the deployment of naval forces to the Yellow Sea as part of the process whereby it secured China s agreement to its lease of Port Arthur (27 December 1897): the town and base that Japan had been obliged to forego were formally occupied on 27 March 1898.
The French, by dint of the agreement of 27 May 1898, were not to be left out: the concession they gained was Kwangchow-wan. 26 For its part Britain was to secure two major concessions. The treaty of 9 June 1898 provided for the ninety-nine-year lease of the New Territories, opposite Hongkong Island, while on 24 May 1898-on what day but Queen Victoria s birthday-British forces came ashore at Wei-hai-wei and formally took possession of the 285 square miles/730 square kilometers of town, base, and immediately surrounding area on 1 July 1898. 27 It is sometimes asserted that the British possession of Wei-hai-wei took the form of a twenty-five-year lease, but that was the duration of the Russian lease of Port Arthur. The British lease of Wei-hai-wei was given no time limit but by agreement was to last as long as the Russians were in Port Arthur. In any event, Britain took the decision in 1901 not to fortify Wei-hai-wei, and in September 1905 concluded an arrangement with China whereby it would continue to lease Wei-hai-wei as long as the Japanese were in Port Arthur. That, of course, does represent this story getting ahead of events; suffice to note two comments-that this arrangement does seem odd alongside the fact that Britain and Japan were tied by treaty of alliance, and the British move into Wei-hai-wei coincided, to the month, with another event in the Far East, namely the Battle of Manila Bay in what was known throughout Europe at the time as the Yanko-Spanko war.

MAP 1.1. The Japanese Perspective: The main theaters of operations in the Sino-Japanese and Russo-Japanese Wars, 1984-1895 and 1904-1905, respectively.

APPENDIX 1.1.
THE PACIFIC AND THE EAST INDIES IN THE NINETEENTH CENTURY
T AHITI (17 40 SOUTH 149 30 WEST ) was proclaimed a French protectorate in 1842 and was formally annexed the following year: thereafter the island had an undefined and nebulous status until it became a French imperial possession on 29 June 1880. The Marquesas were subjected to an American proclamation of annexation in 1813 that was repudiated by Congress; after the French move into the group in 1842, the islands became part of the French empire, again after several decades of ambiguous status, in 1870.
The British landing on Singapore Island was on 29 January 1819 and the treaty whereby the island was secured from Johore was concluded on 6 February. Initially, Singapore came under the East India Company and was ruled first from Bengal and then Delhi; with the mainland provinces it became the Crown colony, the Straits Settlements, in 1867. The British secured Malacca (present-day Melaka in 02 14 North 102 14 East) from the Dutch by the Treaty of London, 17 March 1824.
Hongkong Island, the opening of five treaty ports, and an indemnity were secured by Britain at the Treaty of Nanking, 29 August 1842. The Kowloon peninsula was secured by the terms of the Convention of Peking of 24 October 1860 in the wake of the Second Opium War. Various adjacent lands, including New Kowloon and Lantau Island, which were and collectively known as the New Territories, were then secured by Britain, by lease for ninety-nine years, after 1 July 1898. The colony was renamed Hong Kong in September 1926 and the New Territories should have reverted to China with Japan s surrender in 1945. Britain re-occupied Hong Kong and controlled the colony until 30 June 1997, when it was ceded to China: Britain was not obliged to surrender the 1842 and 1860 provisions, only the areas bound by the 1898 arrangement.
The first British settlement on Labuan Island was established in 1840 as part of the navy s anti-piracy effort, and formal possession, from Brunei, was forthcoming 18 December 1846; Labuan became a crown colony in 1848. The process whereby North Borneo became British was complicated. The Americans established themselves in North Borneo (Jesselton, present-day Kota Kinabalu, in 05 59 North 116 04 East) in 1865 but relinquished all holdings and ambitions in North Borneo in 1875 when technically it became an Austro-Hungarian possession as a result of local initiative. When Vienna disclaimed interest in and responsibility for North Borneo, the initial move was to offer the territory to Italy as a penal colony, but British money had provided for the Austrian purchase of 1875 and as a result the British North Borneo Company was formed in August 1881 and it took over the territory, as a British possession, on 1 November 1881.
Sarawak (Kuching 01 32 North 110 20 East) was perhaps the strangest of all British acquisitions. It was part of Brunei when, on 24 September 1841, a Britisher was appointed governor; on 18 August 1842 he became rajah, and the territory remained a possession of the Brooke family for all but a century. In that time it was, for all intents and purposes, British, though whether it was ever formally secured is not clear: it seems that there may have been some formal proclamation, perhaps of protectorate status, in 1888, but there was no change in the system of government of Sarawak.
The Gilbert and Ellice Islands were formally proclaimed British protectorates on 27 May 1892. The first British move into the lower Solomons was in 1893, and then there were moves into the central and upper Solomons in 1898 and 1899, respectively. With the latter moves the British came alongside the Germans, who had established themselves in New Britain, subsequently renamed Neu-Pommern (Rabaul 04 13 South 152 11 East) and New Ireland, subsequently renamed Neu-Mecklenburg; in 1900 Germany relinquished all claims on the Solomons Islands other than these in favor of Britain.
There are real problems in definition with reference to the East Indies, in no small measure because for more than two centuries after their arrival in the Indies the Dutch, primarily concerned with informal colonialism, that is, ports, trade and spices, and recruitment of local soldiery, for the most part sought arrangements with local rulers rather than the imposition of their own direct rule, Java in large measure excepted.
The Dutch arrived at Jayakarta (what was to become Batavia and is now Jakarta) in 1596. The campaign that resulted in the Dutch conquest of part of western Java began in 1619 and the Dutch had ensured themselves against expulsion by 1628-1629 with the checking of Mataram and its attempt to secure Jayakarta, but it was not until 1682 that Bantam (present-day Banten in 06 00 South 106 09 East) was taken. Likewise, while the Dutch established a trading post at Banjarmasin (in southeast Borneo in 03 22 South 109 44 East) in 1605-and took Ambon (on Amboina in 03 41 South 128 10 East) in that same year-it was not until 1787, and after a number of altercations with the British, that the town became a Dutch protectorate. The Dutch took Malacca in 1641 and Macassar/Makassar (present-day Ujung Padang in southwest Celebes in 05 09 South 119 28 East) in 1669.
A general Dutch control of Java-albeit with a couple of enclaves that recognized Dutch suzerainty but remained separate and in being until 1942-really had to await the Treaty of Giyanti, 1755, and its aftermath. In the event this proved short-lived, given the British occupation of 1811-1816, and it was in the aftermath of the British return of Java to the Dutch (19 August 1816) that the latter established direct rule throughout most of Java primarily as a result of victory in the Java War (1825-1830) that resulted in an estimated deaths of two hundred thousand Javanese, the population of the island being about three million. This war alternated with conflict in central and northern Sumatra, where the Dutch established a measure of dominance in the course of the Padri War (1821-1837 but in effect two conflicts, 1821-1824 and 1830-1837), but the subjugation of northern Sumatra had to await the Acheh/Atjeh (present-day Banda Aceh in 05 30 North 95 20 East) War, 1873-1908, the town finally being taken in December 1907, though unrest in the area was endemic and certainly lasted for the duration of Dutch rule.
The Dutch conquest of Bali took place between 20 September 1906 and April 1908, the various kings, their wives, and courtiers of the island s states subscribing to the cult of puputan , or mass suicide, in the face of inevitable defeat. The first puputan took place at Denpasar (in 08 40 South 115 41 East) and the last, some 24 miles/40 km away and eighteen months later, in Klungkung (in 08 32 South 115 25 East).
The United States took possession of the Midway Islands (in 28 12 North 177 24 West) on 28 August 1867 when the captain of the sloop Lackawanna proclaimed annexation, and of coaling stations on Pago Pago (in 14 16 South 170 43 West) in the Samoan group in 1878 and at Pearl Harbor after 20 January 1887, when the Senate gave permission to the U.S. Navy to lease a base. The American annexation of the Hawaiian Islands was proclaimed in July-August 1898 though Midway was never part of the group; initially it was under U.S. Navy administration but at the present time as an unincorporated territory and designated a minor outlying island, it comes under the authority of the U.S. Department of the Interior.
The Samoan group was divided between Germany and the United States by local agreement on 10 June 1899 after Britain relinquished its claims on these islands in return for uncontested ownership of the Tongan Islands (which in effect became a protectorate within the British Empire on 12 August 1900). The division of the Samoan Islands was formalized with the Treaty of Berlin, signed 2 December 1899 and ratified on 16 February 1900.
CHAPTER TWO

THE GRECO-TURKISH WAR OF 1897
T HE GRECO-TURKISH WAR of 1897, which lasted a little more than a month, is one that has all but disappeared from history books: Dupuy and Dupuy s The Encyclopedia of Military History affords just fifty-six words and numbers to this conflict, and this would seem to be par for the course. There is indeed no disputing the simple fact that this was a war that was of little importance and consequence, and it is a war that has been pushed to the side by the greater conflicts that came over the next twenty-five years. But it was a war that was the one exception in the process of Turkish contraction on the Haemus: it was the only war in the nineteenth-century Balkans in which Turkey was not obliged to cede territory.
What was to become known as the Thirty Days War began on 17 April 1897 as the by-product of the situation that had arisen on Crete, where the collapse of Turkish authority and the activities of Greek militias produced a situation that tethered on the brink of civil war and foreign intervention, the great powers not wishing to see any conflict that might lead to a wider war on the mainland. 1 The great powers were able to prevent Turkey from sending to Crete army formations that might have restored order if not law but could prevent neither a landing on Crete by a Greek force of some two thousand troops on 15 February 1897 nor a Greek declaration of annexation (16 February). What the great powers were able to do, however, was to demand-under threat of naval blockade of all Greek ports for non-compliance-the Greek withdrawal of its military and naval forces from Crete (2 March), but with one unforeseen consequence: frustrated on Crete, the Greeks sought compensation in Macedonia via the encouragement of rebellion among the Greek population in the area and a dual offensive, to the west in the Epirus from the Arta area and in the east from the Larissa area. 2 Confounding this intention, however, were two simple facts of life: the unfolding of events on Crete had given the Turks two months in which to ready themselves for a campaign, and the Turkish Army possessed clear numerical advantage on both sectors but more specifically in the east. In the west the Greeks were able to drive Turkish forces beyond artillery range from Arta and, with Turkish forces withdrawing to positions in front of Philippiada, Greek formations were able to advance some 20 miles/32 km northward, roughly half the distance to Janina, by 25 April. 3 By the time they did so, however, the situation in the east had unravelled. The intended Greek offensive in the direction of Elassona had come to nothing with the Greek forces gathered around Mati, 4 having been outflanked without ever having managed to get over the border, being obliged to conduct a general withdrawal that ultimately resulted in the whole of the area north of Pharsala, including Larissa, Trikkala, and Karditza, 5 being abandoned. This, however, proved only the first part of what was to be comprehensive defeat. With the Greek formations in front of Janina simultaneously forced into a disastrous retreat, the Greek defeats of 15-17 May in front of Domoko and then in the Phurka Pass laid bare the whole of the area to the north of Lamia and the river Sperchcheios. 6 With the Turks also securing first Volo and then virtually the whole of the coastal area between Volo and Lamia, 7 there was little to prevent a Turkish advance to Athens. It was at this point that great power intervention, and a ceasefire in place from 20 May, 8 ensured Greece against the consequences of her own impetuosity and bad judgment. Under Russian brokerage, and after international consultations involving both Greece and Turkey and then those two countries being obliged to negotiate directly with one another, a peace was agreed and a treaty signed on 4 December at Constantinople. The latter provided for a very minor border adjustment in Turkey s favor, and receipt of a very small-indeed derisory-indemnity. Turkey was not able to register gains that its position of military advantage suggests would have been within its grasp but for the great powers.
What is especially interesting about this brief affair is the fact that the Greeks do not appear to have tried to use what was at least on paper an appreciable margin of superiority at sea to turn Turkish positions in Macedonia, most obviously in the area of Salonika, which was all but defenseless; moreover, there was no attempt to take the tide of war to the many islands of the Aegean with their largely Greek populations. One can only assume that events unfolded too rapidly, and to an unforeseen end, for the Greeks to be able to use their margin of superiority at sea to any real effect, though perhaps two matters provide the basis of real explanation. Greece possessed three modern coastal defense ships, the 4,808-ton Hydra, Psara , and Spetsai , built between 1889 and 1892, with a main armament of three 10.6-in./269-mm guns and 17 knots/31.3 kph, and it seems that in winter 1897 the Psara was undergoing major modification and that the other two were undergoing refit and overhaul in a French yard and only with difficulty returned to home waters and then too late to be effective: 9 in the public inquest that followed the war the Greek government was forced to admit that there had been no war plan and that the navy was wholly unprepared for the war that came in April 1897. 10 Be that as it may, the Greek naval performance in this war has drawn the acidic comment that what was achieved was the futile bombardment of Pr veza, the capture of a cargo of vegetables at Sante Quarante and that of a Turkophile British member of parliament. 11
The Greek effort at sea initially involved the bombardment of Pr veza from within the Gulf of Atva by gunboats and from seaward by cruisers and then, between 21 and 23 April, the bombardment on successive days of Targa, Murto, and Haghii Saranda; the only other operation on the west coast was the movement of troops for an intended assault on Pr veza from the north, but this particular effort developed too late to turn back the tide of defeat and indeed the force that was put ashore was hammered on the Luros River and all that remained for the Greek Navy to attempt by mid-May was the support and evacuation of these defeated troops. 12 To the east Greek warships were involved in an attempt to sever the rail line from the east to Salonika by the landing of demolition parties on 20 April 13 and then to cut the railway from Salonika to the south by the bombardment of Platamona, Lephto Karya, and Katerina on the following day. 14 On 24 April Greek warships did conduct a bombardment of one of the forts at the entrance to the Gulf of Salonika, 15 and apparently troops were put ashore at Stylida on the Gulf of Lamia (alternatively called the Gulf of Zituni at this time) on 18 May, but this represented the sum of naval activity. 16 Perhaps wholly unnecessary in light of previous events, the naval armistice included a clause that prohibited Turkish warships leaving the Dardanelles, though Brassey s Naval Annual sets out the claim that Turkish warships did indeed clear the Dardanelles for one demonstration off the straits; 17 it would seem that no one noticed.
But perhaps more interesting than these nickel-and-dime naval events are three other matters, the first being that the great powers, after having ensured a Greek withdrawal from Crete, put what amounted to an international peace-keeping force on the island. After proclaiming a blockade of the island, which was declared a protectorate under international control on 21 March, on the following day various troops and marines came ashore in the first stage of a process that was to result in Austro-Hungarian, British, French, German, Italian, and Russian units being put ashore on Crete. The forces thus committed were modest, no state putting more than the equivalent of a couple of battalions ashore and in the case of Germany just fifteen naval personnel, but the very fact that the six powers combined to attempt to sort out the problem of Crete possibly represents a first-of-its-kind. 18 Perhaps predictably, however, the problem proved one that could not be resolved by such a measure. Crete s somewhat notorious reputation for unruliness, and sheer nastiness of the Greek treatment of Turks on the island, persisted. Crete remained nominally Turkish but under a Greek governor, internationally supervised and with Italian control of the gendarmerie, but after 1897 there was a steady departure of Turks from Crete, particularly after 1905, and Crete became part of Greece after the Balkan Wars, 1912-1913. On the mainland the Turkish Army evacuated Thessaly in 1898 after the Greek payment of indemnity; when it did so virtually the entire Moslem population of the area, which had lived under Greek rule since 1881 but was only too aware of the fate that awaited it at the hands of a returning and vengeful mob, left with it. 19
The second matter of possible interest is that at the time when the great powers demanded the Greek withdrawal from Crete there were three Greek warships on station, the cruiser Wykali being in the process of returning home. The Greeks were obliged to withdraw the gunboats Alpheios and Peneois on 18 March, on which date, allegedly but with no evidence to support the assertion, the Austro-Hungarian cruiser Sebenico sank a Greek warship, unnamed, that had fired upon it. Interestingly, it seems that Austro-Hungarian gunboats were prominent in mounting patrols that enforced the blockade of Crete, presumably the Dual Monarchy s disinterest being at least part of the reason for such a state of affairs, though a certain caution perhaps need be exercised on this point. It seems that after August 1896 Austria-Hungary took the lead in trying to bring about intervention by the great powers in order to ensure some form of peace and stability on Crete, and among her proposals, which were opposed by Britain, was for a blockade of the island in order to prevent Greek or Turkish intervention. Austria-Hungary also proposed, in March 1897, the blockade of Greek ports in order to forestall the possibility of Greek mobilization, specifically to prevent the deployment of forces into Thessaly via Volo and the railway; this proposal was again opposed by Britain and was lost. 20 One assumes that at least in part, Austria-Hungary was motivated by concern lest matters Cretan transposed themselves to Bosnia-Herzegovina, suffice to note that the Greek landing of 15 February provoked a Greek uprising throughout the island and the massacre of thousands of Moslems, most notably in Sitia, and that it was this that finally ensured great power intervention. 21
The third and final point is to note that in the aftermath of this war there appears to have been very considerable efforts on the part of both Greece and Turkey to modernize their naval forces, though in the case of Turkey lack of hard currency precluded much being attempted before 1905. 22 For both sides the units that were in service in 1897 were for the most part old and of little value. For Greece, in addition to the Hydra, Psara , and the Spetsai , the other units considered first-line included four gunboats and twenty torpedo-boats, but the latter, built between 1878 and 1885 and none displacing more than 85 tons, had clearly long lost whatever little relevance they had once possessed. The various units considered second-line, and basically allocated coastal and defensive duties, included one gunboat, the Basileus Georgios , which was built in 1862 or thereabouts, the gunboats Amorakia and Akteion , launched in 1881 and 1883, respectively, and three minelayers-the Aegialia, Monemvasia , and the Naupaktia -built around 1881 and obviously of limited value: five of the remaining twelve gunboats in service dated back to 1856 and three were of only 86 tons displacement. On the Turkish side there were a number of new gunboats and destroyers, built in Germany, but still in service were three casement ships, and the very name is a comment on age and usefulness: for example, the Messoudieh had been launched in 1874 and was completed in 1876: at 9,120 tons it was equipped with twelve muzzle-loading and three 7-in./1-78-mm guns, while the 5,600-ton Assari-Tevfik had been laid down in 1868 and completed in 1870. Four other ironclad turret ships may have been useful as transports but as warships were useless, while seven iron-clad corvettes carried assorted muzzle-loaders and modern guns. The list can be continued but to no real purpose, suffice it to note three points. The first Jane s Fighting Ships , published in 1898, has drawings of Turkish warships, and the largest and most important of these warships are duly represented, fully rigged; that same year, the Brassey s annual listed twelve Turkish armored warships, of which nine were launched in the 1860s, two in the 1870s, and one in the 1880s. Second, it seems that prior to 1897 the last occasion when Turkish warships had operated or even exercised in formation was in 1877. Third, and last, one ship, the frigate Hamidieh , has drawn a comment (written in 1897) worthy of inclusion for obvious reason: laid down in 1870 and completed in 1885, when launched . . . she proved unmanageable: accordingly she was towed back into the arsenal, where she has since spent her life in philosophic contemplation. 23

MAP 2.1. The Greco-Turkish War of 1897: The Thessaly and Epirus sectors.
CHAPTER THREE

THE SPANISH-AMERICAN WAR OF 1898
P ERHAPS, AT THIS DISTANCE in time, the most interesting aspect of the war of 1898 is American attitudes, and specifically the support afforded revolutionary cause against legally constituted and proper authority by the United States; one wonders how congressional motions of this period would be received in Washington today. 1 Moreover, there is the small matter of the commission of inquiry that established, on whatever factual basis has never been determined, that the battleship Maine was sunk in Havana harbor on 15 February 1898 by an external explosion, that is, as a result of Spanish malevolence. 2 Iraq, the Hussein regime, and arms procurement programs would seem to have an ancestral pedigree in terms of reports that situated the appreciation and which presented as conclusive evidence what authority in the United States deemed essential in the pursuit of national interest.

The war of 1898, at least with respect to the war at sea, is a difficult war to summarize because it does not really accord with previous experience or what was to unfold in the first half of the twentieth century. This was not a war that involved genuine naval powers. Spain had long since ceased to be a great power-arguably Trafalgar (21 October 1805) was the last time a Spanish fleet saw battle-and the United States was not yet of such exalted naval status. In light of such facts perhaps the most surprising aspect of the war was that neither side had genuine global reach and capability. Yet even if the two states never took the tide of conflict to the other s metropolitan homeland, this was a war that reached around the globe. It was a war that did not witness prolonged blockade-there was blockade and it did not accord with the various American-proclaimed rights reference sea-borne trade-there was little blue-water action in terms of a guerre de course , and there were no assault landings. The only military campaign was one that owed more to the American public need for heroes and sensation than to real historical substance.
In no small measure, such a state of affairs was the product of the war being one between mismatched opponents, and indeed states as ill-matched as Spain and the United States seldom resolve their differences by war. The outcome of this war was never in any doubt, and indeed it is possible to argue that for all the uncritical acclaim afforded Theodore Roosevelt (1858-1919), the Rough Riders, and San Juan and El Caney, the Spanish military most certainly had not been defeated, whether on Cuba or in the Philippines, by the time that national defeat was conceded by Madrid. But the fact was that despite Spain having more troops on Cuba than the U.S. Army could muster, the imbalance of power, when combined with the two overwhelming victories that were won by the U.S. Navy in Manila Bay and Santiago Bay (3 July), ensured comprehensive American victory overall, and one that for the victor left a lingering and increasingly difficult commitment over the next four decades.

Concerning the naval dimension of this war there are perhaps five matters worthy of note. These involved the twin battles, and dual American victories, in Manila Bay and off Santiago. These are complemented by the Spanish dispatch of a force from C diz under orders to effect the relief of the Philippines. There was in addition the campaign on Cuba of which the battle of Santiago was but one episode, and it is worth noting that this campaign really possessed no joint dimension, for the battles ashore and the naval effort most definitely were not complementary. The last element lay in the post-war disposal of the greater part of the Spanish empire, and with it a redefinition of the balance of power in the western Pacific.
Of these five matters, the third, the Spanish dispatch of a force under Rear Admiral Manuel de la C mara y Libermoore (1836-1920) in an attempt to undo the effect of the defeat in Manila Bay, can be summarized briefly: the attempt was abandoned after the Santiago defeat, and on two counts. With the victory in front of Santiago the American naval forces in the North Atlantic and the Caribbean were freed for offensive operations, and by various means the U.S. naval high command let it be known that a move against Spanish ports was an option that presented itself. C mara s force therefore was needed to safeguard against such an eventuality. More importantly, the U.S. victory in front of Santiago was the clear indication of defeat that Madrid heeded. There was no point in seeking to reverse what was certain to be repeated and that could only delay, not change, the course of events.
In any event, the Spanish force, which consisted of the aging battleship Pelayo , the brand-new armored cruiser Carlos V , two Hamburg-Amerika liners purchased in order to serve as armed merchant cruisers, three destroyers, and five transports, sailed from C diz on 16 June. 3 But while two of the destroyers (the Audaz and Prosperina ) were allowed to pass through the Suez Canal into the Red Sea on 4 July, the other ships were held at Port Said between 26 June and 11 July, and by the latter date Madrid had settled on recall. It would appear that the British, through their occupation of Egypt and control of the canal company, stalled the progress of the Spanish formation on the issue of rights of passage, denying the Spanish warships access to the coaling facilities at Port Said. Even so, the Spanish ships would have been beset by the logistical nightmare in attempting to reach the Philippines with no bases and supplies en route and then by the problem of having to fight an action, at the very limits of their supplies of coal, against a numerically superior enemy. Certainly the Carlos V -with an endurance of 12,000 nautical miles at 10 knots-had the ability to reach the Philippines without resupply, but it was but one of seven warships and the other six most definitely were not so well provided. Just how an action might have developed, with the defense of the transports the first priority of these seven warships, is difficult to discern. The Pelayo and Carlos V were definitely a qualitative match for any American warship in the Philippines at this time, but overall, in terms of formation effectiveness, one suspects that C mara s command would have fared little better than had the command of Rear Admiral Don Patrico Montojo y Pasar n (1839-1917) in what had been the first battle of this war.

The battle of Manila Bay has entered into American national lore, and for one very obvious reason: it was the first battle fought by an American naval formation against a recognizable foreign enemy at sea. Previous to this time American warships had fought individual actions: three actions against the British on inland lakes-Valcour Island (11-13 October 1776), Lake Erie (1813), and Lake Champlain (11 September 1814)-and in the course of the Civil War (1861-1865) had then put together riverine and coastal operations that reached the length of the Confederacy coastline and the Tennessee and middle and lower Mississippi rivers, the most obvious actions being those that resulted in the capture of New Orleans (April 1862) and Vicksburg (July 1863) and the action in Mobile Bay (August 1864). Manila Bay, therefore, was different from the past and it heralded something new for the future: the battle marked the coming of age of the U.S. Navy and of the United States as a great power.
The American declaration of war on Spain on 22 April 1898 found the U.S. Asiatic Squadron at Hongkong. This formation, commanded by Commodore George Dewey, was obliged to leave British waters and to loiter close at hand while final preparations were made before sailing for Manila and the Philippines on 27 April. The formation had under command the protected cruisers Baltimore, Boston, Olympia , and the Raleigh , the gunboats Concord and Petrel , and the revenue cutter McCulloch , along with two auxiliaries, the Nanshan and Zafiro , which were British colliers purchased prior to the outbreak of war. The deployment of the formation to Philippine waters invites two comments about the point made earlier that the battle of Manila Bay was the last battle of the Age of Sail. The first is that the battle was the last action prior to the gunnery revolution that was to produce the dreadnought battleship: this battle, along with Tsushima, was to be invoked by both advocates and opponents of the new battleship concept. If with the battle fought in Manila Bay in May 1898 one can see the origins of the dreadnought battleship, then it seems singularly appropriate that these same waters, the Surigao Strait on the night of 24-25 October 1944, should have been witness to the last action between ships of this type.
Even more notable is that the area of American development was represented by the track of a formation of two columns along the 628 miles/1,005 km that separate Hongkong and Manila, plus the 32 miles/51 km between the previous American anchorage in Mirs Bay and Hongkong. 4 The area of operations was perhaps 10 and certainly no more than 15 square miles/38 square kilometers. The Chief of Naval Operations between 1942 and 1945 was Fleet Admiral Ernest J. King (1878-1956). He entered the Naval Academy at Annapolis in September 1897, and in the following year served in the cruiser San Francisco during anti-shipping patrols off the coasts of Florida and Cuba. 5 In one service lifetime the terms of reference of the naval battle were to undergo such profound change that the battle area in the actions that go under the name of Leyte Gulf covered 115,000 square miles/295,000 square kilometers, an area equivalent to the British Isles or Arizona; the area of interest 6 was three times as large-an area of about 450,000 square miles/1,150,000 square kilometers, an area greater than France, Belgium, Luxembourg, the Netherlands, Germany, Switzerland, and Austria combined or in American terms an area greater than Utah, Colorado, Arizona, and New Mexico together; the area of deployment extended over thousands of miles from Singapore to the Japanese home islands to Ulithi to Humboldt Bay, in Dutch New Guinea, via Seeadler harbor, Manus. 7

The action fought in Manila Bay in 1898 was a much more modest affair, and in a sense its very name is misleading: the Spanish squadron was moored beneath the guns of Cavite and away from Manila, the calculation of Montojo being to spare the city the unintended bombardment of overs and to fight his ships in shoal waters, close to the shore, in order to increase the chances of his men s survival after their ships had been sunk. In fact it had been his original intention to place his force in Subic Bay under the cover of guns of the fortress on Isla Grande at the mouth of the bay. Montojo had no illusions about the outcome of any action on the part of a Spanish collection of ships, individually and collectively, that was hopelessly outclassed by any American force that was brought against it. Indeed such was the un-seaworthiness of the Spanish ships that only one unit, Montojo s flagship Reina Cristina , managed to get under way during the subsequent battle, and it was a comment on the state of the Spanish Navy that one of its units in the subsequent battle, the Castilla , was the last wooden ship to stand in line of battle. But the positions at the head of Subic Bay had not been prepared and guns were not in place, and Montojo had little option but to deploy his ships off Sangley Point with two of his units, the Castilla and the small (1887) cruiser Don Antonio de Ulloa , immobilized as a result of defective machinery. The leaks in the hull of the Castilla were such that flooding could only be stemmed by cement, which meant the loss of engines and shaft, and it had to be supported, very literally, by barges loaded with sand that hopefully were to protect its waterline.
The disparity between Spanish and U.S. ships and formations was marked in virtually every dimension, perhaps most obviously in terms of firepower. The largest Spanish ships were about the same size as the smallest of the American units, the result being that the seven Spanish units had twenty-four 4.7-in./120-mm guns compared to the twenty 5-in./127-mm guns of the six U.S. warships; 8 the American ships had ten 8-in./203-mm and twenty-three 6-in./152-mm guns compared to the seven 6.2-in./158-mm and four 5.9-in./150-mm guns of the Spanish ships; and of the latter six of the 6.2-in. guns were in the Reina Cristina and the 5.9-in. guns were in the Castilla . The disparity of firepower manifested itself very quickly when battle was joined shortly after 0500 on 1 May. The American formation, having discounted the possibility that the Spanish would have mined the entrance to Manila Bay, had entered the Boca Grande shortly before midnight and had drawn desultory fire from the El Fraile position to the south. It was not until after dawn, however, that the American force approached Manila and found the Spanish force to starboard.
The American ships moved in line-ahead formation with the flagship Olympia leading; the Spanish ships were roughly in two lines with the sister ships Don Antonio de Ulloa and Don Juan de Austria moored inside the line provided by the other five ships. The American ships drew long-range but ineffective fire before closing and then turning to the west in order to open broadsides, the Americans commencing general fire on what proved to be the first of five passes, three to the west and two to the east, in the course of which range fell to 2,000 yards/1,835 m. At one stage, because of the apparent ineffectiveness of fire, Dewey intended to have the Petrel move to the east of the Spanish ships while the other American ships closed the range, but this was frustrated by two matters: fire had been more effective than was immediately apparent, and the American flagship had begun to close the range when the Reina Cristina was seen to be under way in what was a vain attempt to ram. Inevitably the Reina Cristina drew concentrate fire that resulted in its being disabled and, lacking power and steering, the ship began to drift helplessly toward Cavite. Montojo ordered the Reina Cristina to be abandoned and ordered two of his units to pick up survivors, and while he was able to transfer his flag to the Castilla , its fires very quickly forced abandonment. 9
With the Spanish line mauled Dewey broke off the action around 0755 at the end of the fifth pass, having received a false report indicating that the Olympia had all but exhausted her ammunition. The American ships withdrew to a position some 5 miles/8 km north of Cavite, but indecision regarding future action was set at nought by the evident distress of Spanish ships; just one of their number, the Don Antonio de Ulloa , remained in its original position while other ships sought whatever safety Bacoor Bay had to offer. Accordingly, the American ships came south to engage the battery on Sangley Point and the solitary Spanish cruiser, the American ships stopping in order to ensure accuracy of fire. The crew of the Don Antonio de Ulloa fought until every gun had been disabled, and the ship was then abandoned as it sank. The Concord moved in an attempt to seize the Isla de Mindanao , aground near Las Pi as, but it was scuttled before the American boarding parties could reach it. The Petrel , perhaps belatedly, was able to move to the east and thence take the smaller units in the inner harbor under fire, but shortly after noon the Spanish colors were struck. Thereafter the abandoned Spanish ships in the inner harbor were set on fire by a boarding party from the Petrel . In the course of the afternoon the American warships anchored off Manila with both sides observing a de facto ceasefire.
The following day an American force occupied the Cavite town and base and immediately afterward two of the U.S. cruisers together effected the surrender of Corregidor. 10 By this action the Americans in effect secured base and bay and imposed blockade on Manila, but they lacked the one requirement to take Manila and bring proceedings to a halt: there was no expeditionary force available for operations on Luzon. The first troop convoy bound for the Philippines-the cruiser Charleston with three transports and 2,500 troops-sailed from San Francisco on 25 May, but across the width of the Pacific, and taking on fresh coal, more than a month was to elapse before American troops came ashore at Cavite. 11 By the time that they did so, on 30 June, two complications had arisen. On 12 June Emilio Aguinaldo (1869-1964), leader of Filipino resistance to Spanish rule, had proclaimed the archipelago s independence and, potentially more serious, British, German, and Japanese warships arrived in Manila Bay. The German presence-and German forays ashore-could have led to serious complications for the Americans, not least because by July the German force was larger and more powerful than their own. In the event, however, the Germans were not prepared to challenge the American presence in the Philippines, and after the peace treaty were to content themselves with the purchase of the Caroline and Mariana groups from Spain, which divested herself of all her Pacific possessions after nearly four centuries of imperial presence in that ocean. With some 11,000 troops and 22 artillery pieces, the American expeditionary force, with Filipino insurgents, were able to mask Manila and, after protracted negotiations, on the morning of 13 August staged what amounted to a sham battle-except to those who were killed-that would enable the Spanish to capitulate honorably, their main concern at this stage seemingly being to avoid any surrender to Filipino rebels. 12 The surrender took place in the early afternoon of the 13th and at 1743 the Spanish flag was lowered for the last time: for the Americans the campaign was over, and, unknowingly, their commitment to a war was about to begin.

By this time, however, the second major naval action of this conflict had been fought and decided. It was, in many ways, curiously similar but at the same time very different from the 1 May action. The point of similarity was battle and outcome, but the chief point of difference between two theaters separated by half the world was that the closeness of the United States to Cuba and Puerto Rico provided the basis of close and effective blockade, landing operations, and the conduct of battle on the part of the United States to which Spain had no counter. One simple fact points to the outcome: the Americans were able to deploy the battleships Indiana, Iowa, Massachusetts , and the Oregon and the second-class battleship Texas for the purposes of blockade and battle, and this number, modest though it might have been to first-line European navies, made for overwhelming advantage against any Spanish force dispatched to the Caribbean to seek battle. On the Spanish side only the battleship Pelayo and armored cruiser Carlos V stood real comparison with their American counterparts, and disparity of numbers was comment enough on the imbalance of advantages in the North Atlantic and off the Antilles.
For Spain the outbreak of war brought the immediate problem presented by American proximity to the Caribbean islands: the American imposition of blockade pointed to the prevention of the regular and proper reinforcement and re-supply of Spanish forces on Cuba and Puerto Rico. The response of Madrid was to order a squadron, under the command of Rear Admiral Pascual Cervera y Topete (1839-1909), to the Caribbean in an attempt to counter the American blockade, but this move was opposed by Cervera for obvious reasons: his formation was clearly massively inferior to any that it was likely to meet, while individually the ships in his command left a very great deal to be desired. His flagship, the armored cruiser Crist bal Col n , was missing its main armament, while the state of repair of the armored cruisers Almirante Oquendo, Infanta Maria Teresa , and the Vizcaya and the sea-keeping qualities of the destroyers Furor, Plut n , and the Terror were somewhat limited, 13 and Cervera was to leave the Terror at Martinique. Not so immediately obvious a weakness was the lack of colliers, which was to mean that once Cervera s formation reached the Caribbean it lacked the range to return to Spanish waters because the French at Martinique denied the Spanish warships coal (12 May) while the Dutch at Cura ao permitted the Spanish warships just 600 tons of coal and two days respite (14-15 May). 14 With such limited range, Cervera decided to take his formation to Santiago de Cuba rather than to Havana or Cienfuegos, or even to San Juan, and in truth he had little real option because American superiority meant that the Spanish formation had little or no chance of breaching the American blockade. 15
The American first-line naval forces in the North Atlantic and Caribbean were divided between two formations that, in the form of command, were to represent the division that caused so much dissension within the U.S. Navy in the two decades following this Splendid Little War. The main force, the North Atlantic Squadron, was commanded by Rear Admiral William Thomas Sampson (1840-1902) and was initially deployed at Key West, Florida. The second force, the Flying Squadron, was commanded by Commodore Winfield Scott Schley (1839-1909) and was initially deployed at Hampton Roads, Virginia, in Chesapeake Bay. Sampson s command was initially charged with the imposition of blockade on Havana while Schley s formation was held against the possibility of Spanish tip-and-run raids on the potential profusion of attractive East Coast targets. The Flying Squadron initially included the Massachusetts and Texas , two cruisers, and a number of lesser units, but in mid-May, with the immediate danger of Spanish raids seemingly passed, it was ordered south to Charleston, South Carolina, in order either to reinforce Sampson s command or to provide direct protection to the Key West base. Arriving at Charleston, Schley s command was ordered forward in order to cover the transports then being gathered in Tampa Bay, but this task was set aside and the formation was dispatched first to Cienfuegos and then to Santiago de Cuba in the search for Cervera s formation.
Schley s formation sailed for Cienfuegos on 19 May, the same day as Cervera s formation arrived at Santiago de Cuba. The latter fact was reported to the naval authorities in Washington by an agent employed as a telegrapher within the Spanish headquarters in Havana. After some hesitation Schley was ordered on 23 May to proceed to Santiago, to ensure its blockade, and to ascertain whether or not the Spanish force was there, but for reasons that have never been satisfactorily explained it was not until 28 May that the Flying Squadron took station off Santiago, and it was only on the following day that this formation finally identified certain of Cervera s ships in harbor. On 29 May Sampson s formation sailed from Key West and, after joining with the Oregon and two minor warships that had been the blockading force off Havana, it moved via the Nicholas Channel and Windward Passage to join Schley s formation off Santiago on 1 June.
By this stage the Americans had secured a position of overwhelming advantage but nonetheless faced a double problem: they could not force Cervera s force to give battle but at the same time they could not be certain of ensuring against a sortie. On the night of 3-4 June there was an attempt to sink a collier in the channel, but the ship was crippled by Spanish fire and it ran aground but not in a position to prevent the Spanish ships leaving harbor. In this situation the Americans moved to the alternative of landing operations in eastern Cuba that were designed to secure Santiago by an overland advance, the first landings taking place in Guant namo Bay on 10 June, the second landings at Daiquir and Siboney on 22 June. 16 The concentration of Spanish forces in western Cuba, and specifically in the area of Havana, and the smallness of the scattered but numerous garrisons in eastern Cuba in effect precluded any effective Spanish counter. There was a certain inaction and passivity on the part of the Spanish military, in Cuba as a whole but specifically to eastern Cuba, that invites the comment that if this was indeed a case of sangfroid then it was indistinguishable from rigor mortis.
The American conduct of operations once ashore was less than impressive-the only less impressive aspect was the Spanish conduct of operations-but these operations nonetheless entered American national mythology and Theodore Roosevelt and the Rough Riders afforded pride of place. The significance of these events, other than the political, lay in the simple fact that once the Americans established themselves ashore the close investment of Santiago was but a matter of days away, and the position of Cervera s formation thereby rendered untenable. Accordingly, on 3 July the Spanish formation sailed, and it chose to do so when Sampson in the New York and the Massachusetts were absent, Sampson being involved in consultation with the army commander, Major-General William Rufus Shafter (1835-1906), 17 while the Massachusetts was taking on coal.
In Sampson s absence command of the American force (which consisted of the battleships Indiana, Iowa, Oregon , and the Texas , the armored cruiser Brooklyn , the torpedo-boat Ericsson , and three auxiliaries) was exercised by Schley, and herein was to be the basis of the subsequent Sampson-Schley controversy over who should take credit for the subsequent victory, Schley and the popular press being in accord on this particular matter. In fact the initial exchanges were to place question marks against Schley s exercise of command because, with the Spanish having secured an element of surprise by making their sortie in daylight, in calm weather, and on a Sunday, their basic plan was to have their flagship, the Infanta Maria Teresa , ram the Brooklyn , thus taking Schley s ship out the action and hopefully increasing the chances of the other ships being able to escape. The ploy certainly induced a measure of confusion in the American line as the Brooklyn , in her attempt to elude the Teresa , almost rammed the Texas , but the fact was that Cervera s flagship drew concentrated American fire, and its admiral, who took direct command of his flagship when its captain was wounded, ran the Teresa , heavily damaged and on fire, aground near Punta Cabrera about one hour after it had entered the main channel. By that time, however, American fire had shifted and two more Spanish ships were run aground within a matter of minutes, first the heavily damaged Almirante Oquendo (fourth in line) and then the destroyer Plut n (sixth in line), its engine-room wrecked by a shell from the battleship Indiana . At the same time the Furor (fifth in line) was sunk, somewhat improbably by the barely armed converted yacht Gloucester . 18 Only two of the Spanish ships managed to get clear of the channel and the immediate concentration of American warships, but both were subjected to a stern chase and an overwhelming concentration of firepower as they sought to make their way to Cienfuegos. The battered Vizcaya , originally second in line, was caught on a reef as it turned for the shore in an attempt to run itself aground, and the Crist bal Col n , originally third in line and its coal all but exhausted, was run aground and then scuttled at the mouth of the River Turquino, some 50 miles/80 km west of Santiago. 19 With just two casualties, the American victory was overwhelming, comprehensive, and final, and marred only by several hundred Spanish sailors being murdered by Cuban insurgents as they tried to come ashore. But for Spain, of course, it was something much more than that: it was a defeat that closed a chapter more than four centuries in the writing, and that reached back to Grenada, the Alhambra, and Columbus. 20
The battle, and in a sense the war, was over. Santiago de Cuba was not surrendered until 17 July, and probably prematurely: malaria, yellow fever, and dysentery combined to wrack the American forces outside the city certainly to the extent that precluded deliberate assault, and, so easily overlooked, there was a campaign, or perhaps more accurately a f te des fleurs , in Puerto Rico with San Juan bombarded on 12 May as part of the operations in search of Cervera s force. American formations were put ashore at Gu nica on 25 July and at Ponce two days later. With the destruction of Cervera s formation the Spanish had to recall C mara s formation as, for the first time, the possibility of American moves into the eastern Atlantic manifested itself. For the Spanish, no options remained and the defeats that had been incurred could not be reversed. At this point Spain, which in June had extended pourparlers and had sought to learn what terms the United States sought, accepted defeat, and with the surrender of Manila in the Philippines the armistice that should have come into effect the previous day did so. The first meeting in the peace negotiations opened in Paris on 1 October. The Treaty of Paris, concluded on 10 December 1898, provided for Cuban independence (albeit under American tutelage), the annexation of Guam and Puerto Rico by the United States, and the American purchase, at a cost of 20 millions, of the Philippines. The by-product of the war was the congressional resolution, passed by the House on 15 June and the Senate on 6 July and signed into law by President William McKinley (1843-1901) on 7 July, that provided for American annexation of the Hawaiian Islands.

MAP 3.1. The Battle of Manila Bay, 1 May 1898.

MAP 3.2. The naval battle off Santiago de Cuba, 3 July 1898.

APPENDIX 3.1.
THE ACTIONS IN THE PHILIPPINES AND THE AMERICAN ORDER OF BATTLE
P RESENT IN THE AMERICAN order of battle on 13 August and the bombardment of Fort San Antonio de Abad were two ships other than those that had been involved in previous operations, Albert A. Nofi, The Spanish-American War, 1898 , p. 286, citing the captured gunboat Callao and the monitor Monterey , which had arrived as scheduled on 4 August.
With reference to the Callao , this ship was captured on 12 May when, unaware of the outbreak of war and the events of 1 May, it returned to Manila Bay after an extended cruise in the Philippines; see Donald H. Dyal, Historical Dictionary of the Spanish-American War , p. 54. It was commissioned into U.S. service on 2 July; see the ship s entry in Dictionary of American Fighting Ships , Vol. 2, p. 16.
With reference to the Monterey , there is some confusion about its arrival in the Philippines not least because the official record, in the form of the Movements of Vessels section in the Annual Report of the Navy Department. Report of the Secretary of the Navy. Miscellaneous Reports for 1898 and 1899 are different. The 1898 report (p. 346) states that the ship sailed on 11 June from San Diego (in the company of a collier, according to its entry in Dictionary of American Fighting Ships , Vol. 4, pp. 426-427), that it was at Pearl Harbor between 24 June and 1 July, and that it arrived in Manila Bay on 9 August. The 1899 report (p. 432) states that after leaving the Hawaiian Islands, the Monterey was at Guam between 23 and 25 July and that it was in Manila Bay after 4 August, which is the date cited by Dyal, p. 226, though other details this source gives are different. It does not appear, however, that the Monterey was actively involved in the bombardment of Fort San Antonio. Given the detail of the Guam landfall, it would seem that this second account, with the 4 August schedule, is likely to be correct.
Dyal, p. 225, also states that the monitor Monadnock did not arrive in Manila Bay until 16 August; the latter date is confirmed by the ship s entry in Dictionary of American Fighting Ships , Vol. 4, p. 411-412, which states that it sailed from San Francisco on 23 June. This entry also states that this ship was laid down in 1874, was launched 19 September 1883, and was commissioned into service 20 February 1896. Given that it was decommissioned on 24 March 1919, it would seem that the Monadnock was almost as long being built as it was in service. 1
NOTE
1 . The checking of details, and specifically the 1898 and 1899 reports, was recorded by Sarandis Papadopoulos, Naval Historical Center, and communicated to a very grateful author by e-mail on 8 August 2006.

APPENDIX 3.2.
THE CAMPAIGN ON PUERTO RICO
T HE AMERICAN LANDINGS involved ten transports (with the Massachusetts and four auxiliaries) that sailed from Guant namo Bay on 21 July. At both landings, at Gu nica (in 17 59 North 66 51 West) and at Ponce (in 18 01 North 66 36 West), the Americans encountered minimal resistance and were able to secure intact port facilities; reinforcements arrived at both ports on 31 July and another landing was conducted at Arroyo (in 17 59 North 66 03 West) on 2 August. 1 Despite various supply problems, in large measure the result of faulty loading of transports and inadequate inventories, American forces from Arroyo secured Guayamo (in 17 58 North 66 55 West) while one division that had been landed at Gu nica, having secured Yauco (in 18 02 North 66 51 West) on 26 July, first moved eastward to Ponce and thence north to secure Adjuntas (in 18 10 North 66 42 West), while the division that had been landed at Ponce moved eastward to secure Juana Diaz (in 18 03 North 66 31 West) on 7 August and Coama (in 18 05 North 66 22 West) on the 9th. At the same time a brigade from Yauco advanced westward to secure Hormigueros (in 18 08 North 67 08 West) on 10 August and Mayaguez (in 18 13 North 67 09 West) the following day.
The armistice found American forces just short of Los Marias (in 18 18 North 66 59 West), Utuado (in 18 17 North 66 41 West), and Aibonito (in 18 10 North 66 13 West) and on or about the main ridge-line across the island. Interestingly, the American plan for the Puerto Rican campaign envisaged the use of an expeditionary force of about 16,000 officers and men, a total about the same as the force in Cuba and twice the force dispatched to the Philippines: overall about 14,500 officers and men (including medical personnel) were landed on Puerto Rico prior to the armistice.
NOTE
1 . Sources: Dyal, Historical Dictionary , pp. 147 and 268-269, and Nofi, The Spanish-American War , pp. 227-260 and 334-336.
CHAPTER FOUR

THE SHIFTING BALANCE OF POWER
T HE SINO-JAPANESE and Spanish-American wars in effect marked the closing of an era. The mark that the world wears is primarily a European mark. The state and the capitalist system were primarily European creations, and a global economy and global war were likewise of European pedigree. The calendar and time are similarly European, and Europeans drew virtually every border in the world, usually with little or no reference to indigenous populations. The Sino-Japanese and Spanish-American wars really marked the apogee of empire: from this time, around 1895-1898, virtually every part of the world other than Europe and North America was under either direct European control or a dominating European influence: as noted earlier, the nineteenth century in South America was known as the British century because the states of that continent found that British money, investments, and trade slotted into place with the end of Spanish empire. By the dawn of the new century there were no areas in which Europeans, and Americans, might establish themselves without war, or the very real prospect of war, with another European power. The conclusive evidence of this truth was provided in the Fashoda crisis of September-November 1898 when France quite deliberately chose not to confront Britain over claims over the Sudan. In a very real sense this was the acid test of extra-European issues: if France was prepared to acquiesce in British primacy on this occasion then there was never going to be an issue that could produce real crisis, at least not between Britain and France. Relations between the two countries were not good, but the simple fact was that France could never afford to challenge Britain outside Europe given its military, industrial, and demographic inferiority relative to Germany. Fashoda, however, was one in a series of events that came together in very rapid succession, and which in very large measure re-wrote naval terms of reference.

The crucial development was the emergence of Germany as a major naval power, but inevitably events were more complicated than any single cause would suggest. The last quarter of the nineteenth century witnessed the industrialization and urbanization of Europe and the United States. The American Civil War and the Franco-Prussian War of 1870-1871 were not fought between industrialized societies that lived by manufacture and trade: as late as 1860 9 in 10 Americans lived in settlements of less than 2,000 people. Prior to this time perhaps only two states, Britain and Belgium, were properly industrialized in the sense that they paid their way primarily through industrial production: the majority of their peoples lived by manufacture and trade. In the last quarter of the nineteenth century, Europe became industrialized, with the pattern of industrialization thinning to the east. Railways, even properly metalled roads, reached out across countries, and certainly major cities were being marked by electricity. Patterns of production were in the process of change, and at this time there was technological revolution in the making. The turn of the century witnessed the first submarine, 1 the first trans-Atlantic wireless message being sent (11 December 1901), and the first controlled flight by a heavier-than-air machine (17 December 1903), and significantly all possessed American dimensions. The fact that all had an American ancestry pointed to the industrial and technological changes that were in the making. There were, nonetheless, the necessary elements of constancy, those matters that did not change, and two were of singular, naval, importance.
The first was that Britain remained pre-eminent in terms of finance and trade. Less than one-fiftieth of all American manufacture was for export, and German industrial primacy in Europe was not general: it was marked in certain fields, most notably in the petrochemical and engineering industries, and in primary, non-finished, products. Britain retained her previous primacy in overseas investments, insurance, world-wide shipping, and volume of trade. Second, in one vital respect Britain continued to lead the world, the United States and Germany included. The American steel industry was in no small measure reared on the demands of a navy deliberately raised in the last two decades of the nineteenth century as the means of strengthening and expanding national capability, and certainly by 1914 German iron and steel production was roughly three-fifths greater than that of Britain, but in terms of ship-building, both capacity and speed of construction, and in naval armament, Britain led the world. Germany, not surprisingly, led the world in terms of armament manufacture, but Britain retained until the First World War primacy in naval construction precisely because its military requirements were so slender.
A certain caution needs to be exercised in any examination of ship-building and naval armaments industries at the turn of the century, for the simple reason that the British lead in these industries was in the process of being eroded. Britain was able to sustain itself throughout the nineteenth century s last decade, after the 1889 Naval Defence Act, at more than a two-power standard because its lead in these fields still remained, but by the turn of the century the British position was increasingly strained. There was a short-term aspect of British difficulties that asserted itself around this time, and this was a series of strikes, lock-outs, and the generally appalling state of labor relations in the shipyards, which meant the building and completion of the Duncan class of battleships was little short of disastrous: the shortest building time for any ship of this class was longer than the average building time of all but one class of battleships built between 1889 and 1907. But the crucial point was that Britain s advantages in ship-building and naval armaments were being eroded by the process of industrialization that imposed itself on Europe and the United States in the last quarter of the nineteenth century.
The process of industrialization of Europe and the United States meant more shipyards, and shipyards that could build more quickly than hitherto. This was general, and applied to Britain as well as those states in the process of industrialization and that previously did not have to be considered seriously by Britain in terms of speed of and numbers under construction. Crucial in this process was the example provided by the Vengeance , a Canopus -class battleship laid down on 23 August 1898 and completed in April 1902. It was the first major warship to be built in its entirety by one firm. The Vengeance was provided with armor, guns, and engines by Vickers at Barrow-in-Furness, and though it did not break any records for speed of construction, how it was built was to point to the future direction of warship construction with very few specialist yards but each independent in terms of available resources. That development in itself was to ensure the building of warships faster than at any time over the previous three decades, though it needs to be pointed out that the general trend after 1889, at least in British yards, was for shorter periods of construction. The average time of construction of the seven battleships of the Royal Sovereign class 2 built between July 1889 and June 1894 was 42.94 months, and for the eight units of the King Edward VII -class battleships built between March 1902 and January 1907 it was 33.69 months. It needs to be noted, however, that the building of the Dreadnought considerably delayed the completion of the last three of this class; at a conservative estimate the average building time could well have been three or four months less than was the case. 3
By 1900 the world was capable of building more ships more quickly than at any time over the previous century, and the implication for Britain was self-evident: the day was bound to come when her pre-eminence would end. Her assured pre-eminence in numbers of ships could not be maintained indefinitely, and there was no reason to assume that her superiority in speed of construction could be maintained permanently. On both counts by 1900 the signs were becoming very clear, though, paradoxically, Britain remained in a position to out-build any potential rival within Europe in no small part because of the shortcomings of Italian and Russian yards and the military distractions of Austria-Hungary and France. Given these four countries financial and other commitments and their problems in terms of volume and speed of warship construction, Britain possessed a margin of superiority within Europe, but outside Europe the position of Britain was more difficult. Japan presented no immediate problems because, at the turn of the century, it was still dependent upon foreign yards for first-line heavy ships; it was not until May 1905, less than two weeks before the battle of Tsushima was fought, that Japan laid down a battleship. 4 But during Theodore Roosevelt s presidency (14 September 1901-3 March 1909), the United States launched no fewer than fourteen battleships, and, perhaps more relevantly, in just thirty-seven months between 2 April 1902 and 1 May 1905, the United States laid down no fewer than twelve battleships. 5 Moreover, even between the two classes that together provided ten of this total there was a very significant acceleration of average building time, a cut of more than a year per ship. 6 The four ships of the Virginia class 7 averaged more than fifty months between keel being laid and completion; the six-strong Connecticut class 8 had an average time with the shipyards of 38.59 months; and the Vermont and New Hampshire were the first American battleships to be less than three years in their construction. What was no less significant was that no fewer than seven different yards handled the construction of the last three classes of battleship to be built before the appearance of the Dreadnought ; the days when battleship orders meant assured contracts divided between William Cramp and Sons, Ltd, of Philadelphia, Pennsylvania, and Newport News navy yard had passed.

The changing patterns of production of shipping in general but warships in particular was one of a number of matters that came together at the turn of the century to produce the dreadnought revolution. Within Europe a pattern of naval construction had been established in the last two decades of the nineteenth century that stressed programs over a period of time, not individual construction orders based on yearly parliamentary timetables. The first such program had been signed into existence on 20 May 1882, when Russia set out a construction program for twenty battleships, the time allowed for this program being twenty years to the completion of the last unit. The more relevant example for the rest of Europe to follow, however, was provided by Britain with the 1889 Naval Defence Act that in effect set out a two-power standard. 9 As significant was that even in the few years that had elapsed since the start of the Russian program design had largely stabilized and with two results: states could set about building classes in the sure knowledge that these, unlike their predecessors of the previous two decades, would not be more or less bordering on obsolescence when first laid down, and states could set about seeking increased capacity and increased efficiency in shipyards in order to ensure real addition to strength. Thus in November 1890 Russia devised a program for the construction of six second-class (7,500-ton) battleships, four armored coastal defense ships, and three large armored cruisers for the Baltic, and in 1891 France introduced a ten-year program that was to see the construction of ten battleships, one coastal defense ship, and no fewer than forty-five cruisers. When France and Russia concluded their alliance the British reaction was a revision of existing programs and a specific program, in December 1893, for seven battleships and thirty cruisers. The Russian response came in 1895 in the form of a program that would result in the building of five battleships, four coastal defense ships, and six armored cruisers by 1901, and three years later Russia again implemented another building program, this time for the construction of five battleships and sixteen cruisers within seven years.
Such measures could have pitted these nations against one another in a full-scale naval construction race but for three matters: the final (1898) Russian program was crafted with a view to the provision of major naval units in the Far East; the Russian 1895 program was not geared to the British program but to the Baltic situation; and the French program proved a case of aspiration and performance not being at one with each other. Certainly these various endeavors produced one ship of more than en passant interest: the second-class battleship Rostislav , on 8,800 tons, was the first major warship in the world to be fitted with oil-fired boilers and it was also the first warship to carry a secondary armament of 8 6-in./152-mm guns in four twin turrets. Laid down in 1894, and launched on 20 August 1896, the Rostislav was twenty years ahead of its time on both counts. But such matters counted for little when set against one reality that slowly took shape in the 1980s: Germany began to emerge as a major naval power, and indeed by 1905, despite numerical inferiority to other countries, it was second only to Britain in available, modern, first-line units.
The emergence of Germany as a major naval power was one of the most profoundly important developments in the period between 1892 and 1914. Before this period the Imperial German Navy, formed in 1848 and perhaps the only survivor of that disastrous year that saw the defeat and failure of liberalism in Germany, was of no account. German concern was military and reference France and Russia; naval ambition promised only to be an unnecessary and unacceptable burden. The temper of the times, however, changed such terms of reference: a new kaiser, the pursuit of naval doctrine as laid down by Mahan, and the great powers search for advantage relative to one another in securing for themselves concessions in the Far East were all at work in the 1890s in preparing Germany to accept naval obligation.
Under a new naval secretary, Admiral Alfred von Tirpitz (1849-1930), who was appointed in January 1897, the German Navy found itself the recipient of a Reichstag vote of 10 April 1898 that set national requirements as nineteen battleships (the Oldenburg , the four ships of the Sachsen class, the four ships of the Brandenburg class, the five members of the Kaiser class then under construction, and the five members of the Wittelsbach class then being planned), eight coastal defense ships (the members of the Siegfried class), and six large and sixteen small cruisers, with another six large and fourteen small cruisers assigned to foreign stations. This force was to be built in six years, with individual ships allocated a life span of twenty-five years. With the major units already in service counted on establishment, the 1898 act provided for the replacement of units to be carried out at the rate of two ships a year between 1906 and 1909, one ship a year over the next seven years, and two ships in 1917. 10
This involved a major change for an Imperial Navy that in the 1890s underwent fundamental reconstruction in terms of its small number of coastal defense ships and cruisers in commission. Many of these ships were aging and of dubious worth, and indeed some of the older units had been originally sailing ships. The rebuilding program provided the Imperial Navy with real additions to strength, and the 1898 Navy Law clearly set out the service objective of numerical respectability. 11 Within two years, however, this had been set aside in favor of the idea that quantity had a quality all of its own. The naval bill published on 20 June 1900 set out a program that provided for the construction of two flagships and four squadrons each with eight battleships, eight large and 24 small cruisers, a reserve of four battleships, three large, and four small cruisers, and a total of three large and ten small cruisers for service on foreign stations; the destroyer allocation was fixed at 96 units. Replacements were calculated on the basis of twenty-five years for a battleship and twenty years for a cruiser, but perhaps the real sting in these provisions was in the fact that it was proposed that two of the battleship squadrons were to be maintained on a war footing with the other two squadrons forming the reserve; half of the ships thus earmarked, however, were to be maintained in permanent commission. In other words, Germany aimed to build a fleet with thirty-eight battleships, of which a minimum of twenty-five were to be fully operational at all times. 12
The German naval law of 1900 has attained such notoriety on account of its stated raison d tre that the provisions of the subsequent naval laws of 1906, 1908, and 1912 are seldom afforded the consideration they are due. The combined effect of these laws was to add extra numbers to those already allocated to the extent that ultimately the Imperial Navy found itself committed to programs that totalled 41 battleships, 20 battlecruisers, and 40 light cruisers; how such numbers were to be secured does not seem to have been properly addressed. The immediate point, however, was the rationale for such a force, which, as laid down in the naval law of 1900, stated:
To protect Germany s sea-trade and colonies, in the existing circumstances, there is only one means: Germany must have a battle fleet so strong that, even for the adversary with the greatest sea power, a war against it would involve such dangers as to imperil his position in the world.
For this purpose it is not absolutely necessary that the German battle fleet should be as strong as that of the greatest naval power, because a great naval power will not, as a rule, be in a position to concentrate all its striking force against us. But even if it should succeed in meeting us with considerable superiority of strength, the defeat of a strong German fleet would so substantially weaken the enemy that, in spite of a victory he might have obtained, his own position in the world no longer be secured by an adequate fleet. 13
Much, some might argue too much, has been written about the so-called Risk Theory, the Risikoprinzip embodied in the preamble of the 1900 Law, and these pages could not hope to add anything new, suffice to note three matters. First, the greatest military power in the world by this act sought to secure for itself a position at sea second only to Britain, and this could not be anything other than a direct threat to the latter s security and one that Britain could never return with interest. Britain did not have an army that could pose a direct or even indirect threat to Germany: at the end of the day the famous response of Imperial Chancellor Otto von Bismarck (1815-1898) to the question of what he would do if a British Army landed in Schleswig-Holstein-that he would send a policeman to arrest it-still held good. Second, the very wording of the Risk Theory was provocative, almost deliberately insulting to a Britain that at this time was beset with the revelation of a general European hostility as a result of the opening exchanges in the second South African war (1899-1902). This war had seen the British seizure of a number of German ships in African waters and had provoked considerable anti-British feeling within Germany. To this affront had been added other incidents that favored the anti-British and navalist cause within Germany, most notably the acquisitions of Spanish holdings in the western Pacific in the wake of Spain s defeat at American hands in the war of 1898 and the British intervention between the Americans and Germans in the Samoan dispute, the British in effect having prevented the two from coming to blows and settling the dispute in American favor.
The third point, however, was the Risk Theory s unlikely combination of offering the greatest possible offence with the least possible chance of success, to which must be added an additional matter, which was the sanctioning of a fleet primarily for political reason and without reference to its geographical position and its numerical inferiority. In these matters the whole of the Risk Theory was utter nonsense and for a reason that manifested itself very quickly once war came in 1914: the construction of a battle fleet in the southern North Sea, where it was wholly incapable of protecting Germany s sea trade and colonies, left it with a desire for battle that was but the least tactical response to a lost strategic cause. On political, geographical, and strategic criteria, the 1900 Naval Law failed its own terms of reference, but, of course, in no small measure the Risk Theory sought to evade these because the primary objective was to force concessions from Britain and to obtain from it an undertaking either of neutrality or an alliance with Germany in the event of a general war in Europe between, on the one hand, Germany and Austria-Hungary, and, on the other, France and Russia.
There were within the conservative government in Britain individuals, such as Colonial Secretary Joseph Chamberlain (1836-1914), quite willing to seek some form of arrangement with Germany that very largely would have met these requirements, but various negotiations failed. Britain refused to involve itself in a commitment that, in effect, would have stripped it of real choice. The Risikoprinzip statement of 20 June 1900 began a search for security that was to last until 15 May 1940, the black day in British twentieth-century history. In these four decades British national security came to rest upon a guarantee to maintain France as a great power relative to Germany. It was not a case that Britain moved immediately to this conclusion after 1900. The closing of ranks with the French came in the aftermath of the 1909 Bosnian and 1911 Agadir crises, and, of course, was consummated in 1912 with the cabinet and Committee of Imperial Defence decisions that in the event of war the British Army would be deployed to France.
After 1900 Britain made a series of negative decisions, the most obvious being confirmation that it would seek any political and peaceful solution to any problem that arose between itself and the United States rather than go to war with that country. The basis of this decision was to be found in the War of 1812 and the decision by both countries thereafter to solve disputes by arbitration. The American Civil War had created more than a few problems, most notably the United States claiming the rights that it had denied Britain in 1812. But the basis of the British decision was very simple: Britain could not defeat the United States militarily and had various colonies throughout the Caribbean that would be very vulnerable to American attack. The United States was Britain s most important single market and was critical in terms of overseas earnings with reference to property, investments, and carrying trade. The cultural, historical, and shared democratic practice of the two countries rendered recourse to war all but unthinkable. In many parts of the world the United States established missions and embassies if not under British patronage then in certain places with a closeness between British and American representatives that was very notable.
In 1902, moreover, Britain made a second decision-the alliance with Japan. Wars in which naval power has proved the decisive element in terms of power of decision have been very few, and alliances between naval powers even fewer. It is possible to argue that neither Britain nor Japan was primarily a naval power, and in Japan there was one organization, the Nippon Teikoku Rikugun or Imperial Japanese Army, that held certain reservations on that particular score. Britain in the period between 1900 and 1902 made two basic calculations about the Far East. The first was that Russo-Japanese differences were fundamental and could only be resolved by war, and that there existed a basis for an understanding with Japan. Britain had a certain standing with Japan on account of its naval mastery, and it had not been involved in the Triple Intervention. Britain sought to check Russian ambitions in the Far East, and at the turn of the century these were concentrated on Manchuria and Korea. Britain reasoned that these areas were Japan s natural areas of interest on the mainland, hence the basis of an understanding between the two countries.
The Russo-Japanese War (8 February 1904-5 September 1905), or at least two of its battles, demands consideration in its own right. Here it is sufficient to note that 1904 was critical to Britain not so much because of Russian defeats in Korea and Manchuria, and not even on account of the British military occupation of Lhasa in August, but because of the decision whereby most extra-European disputes with France were laid to rest. The Anglo-French entente of 8 April 1904, followed by the rapprochement with Russia with the treaty of 31 August 1907, marked a fundamental change in British foreign and defense policy, involving as it did an attempt to settle outstanding disputes, some many decades in the making, 14 and to ensure that they were not attended by risk of war. Even more critically, the entente with Russia was seen as a means of strengthening that country at a time of serious weakness, the defeat in the Japanese war having been compounded by revolution within Russia that had been barely contained. Less than eight years separated the last of these British decisions from the realization that Splendid Isolation might not be so splendid after all. It is a small matter, of no importance in itself, but the extent to which Britain involved itself in European affairs in the first decade of the twentieth century can be understood by reference to the simple fact that in her sixty-four years as queen, Victoria undertook just one state visit abroad, in 1855 to Paris. With the new century the world was indeed on the point of major change.

APPENDIX 4.1.
THE GERMAN 1898 BUILDING PROGRAM
T HE LARGE CRUISERS WERE defined as the K nig Wilhelm, Kaiser, Deutschland , and the Kaiserin Augusta , the members of the five-strong Viktoria-Luise class, which were the nameship, the Freya, Hansa, Hertha , and the Vineta , and the F rst Bismarck , which was completed in 1900, and the Prinz Heinrich and Prinz Adalbert , then being built.
The K nig Wilhelm, Kaiser , and Deutschland were built as armored frigates on the Thames and were the last major German warships built abroad. The K nig Wilhelm was launched in 1868 and was commissioned into service in 1869. It displaced 9,760 tons and carried thirty-three 72-lb. cannon. In 1878 it sank the armored ship Grosser Kurf rst in a collision off Folkestone. It was reclassified as an armored cruiser in 1897 but was laid up in 1904 and used thereafter as an accommodation ship. The Kaiser was launched in 1872 and was commissioned in 1875. The Deutschland was launched in 1872 and commissioned in 1875. In their original state both had two funnels and three masts for a full rig. Originally they displaced 7,600 tons and carried eight 10.2-in./2-60-mm cannon. The Kaiser was rebuilt between 1891 and 1895 as a heavy cruiser. The Deutschland s reconstruction as a heavy cruiser was completed on 25 January 1897.
The Kaiserin Augusta and the members of the Viktoria-Luise class were second-class cruisers. The Kaiserin Augusta displaced 6,056 tons and was the first triple-screw ship in the German Navy; it was launched in 1892 and was completed in 1896 and carried four 5.9-in./150-mm and eight 3.46-in./88-mm guns and five torpedo tubes. The Freya, Hertha , and the Viktoria-Luise all displaced 5,660 tons, the Hansa and Vineta 5,885 tons. The Freya, Hertha , and the Viktoria-Luise were launched in 1897 and were completed in 1898; the Vineta was launched in 1897 and the Hansa in 1898, and both were completed in 1899. All carried two 8.3-in./210-mm, eight 5.9-in./150-mm, and ten 3.46-in./88-mm guns and three torpedo tubes.
The last three were armored cruisers. The F rst Bismarck was launched in 1897 and was completed in 1900. It displaced 10,570 tons and carried four 9.45-in./240-mm, twelve 5.9-in./150-mm, and ten 3.46-in./88-mm guns, plus three torpedo tubes. The Prinz Heinrich was launched in 1900 and was completed in 1902. It displaced 8,759 tons and carried two 9.45-in./240-mm, ten 5.9-in./150-mm, and ten 3.46-in./88-mm guns, plus four torpedo tubes. The Prinz Adalbert was slightly larger at 8,858 tons; it was launched in 1901 and was completed in 1903. It carried four 9.45-in./240-mm, ten 5.9-in./150-mm, and twelve 3.46-in./88-mm guns, plus four torpedo tubes.

APPENDIX 4.2.
THE BOXER REBELLION AND SOUTH AFRICA
T HE SUPPRESSION OF the Chinese xenophobic movement known as the Boxer Rebellion was not primarily a naval matter, but it was an episode that did involve the bombardment and capture of the Taku fortresses on 17 June 1900 and certain other matters that intrude on matters naval. For example, the first attempt to effect a relief of the Legations in Peking was conducted under the command of the senior officer present, who just so happened to be a British naval officer, and if the result was a perhaps predictable failure, this effort, between 10 and 26 June, does stand in rather odd contrast to what subsequently happened. The alliance of the eight nations -Austria-Hungary, Britain, France, Germany, Italy, Japan, Russia, and the United States-witnessed the deployment of a combined force that numbered some 18,700 troops, marines and sailors but no overall command; command, in effect, was by committee and operations conducted on the basis of improvised cooperation. Tientsin was taken on 23 July and Peking on 14 August, with the Imperial City subjected to attack on the following day, but while operations continued into May 1901 and were only closed by the Boxer Protocol of 12 September, perhaps four matters naval are worthy of note. First, among the British wounded in June 1900 during the first relief effort was a certain John Jellicoe, while his b te noire , David Beatty, was a member of the naval brigade put ashore at Tientsin, was twice wounded, and at the age of 29 was promoted to the rank of captain. Second, the warships involved in the bombardment of the Taku forts, drawn from no fewer than six navies, were the American paddle-steamer gunboat Monocacy (which had been on station since 1867), the British torpedo-boat destroyers Fame and Whiting and sloop Algerine , the French gunboat Lion , the German gunboat Iltis , the Japanese gunboat Atago , and the Russian gunboats Bobr, Gilyak , and the Koveetz . Third, paradoxically, while there was no action between Chinese and foreign warships, the Chinese managed to lose four torpedo-boats: these were being built in a German yard and were seized, Britain, France, Germany, and Russia each taking one and all of them being given the name of Taku . 1 Fourth, and perhaps the least known of these, was that among the British contingent that was involved in the advance to Peking were Indian and locally recruited Chinese troops but also naval gunners from the first-class protected cruiser Terrible who, in February, had been involved in the operations that had resulted in the relief of the British military garrison at Ladysmith in Natal in what was the second South African War (1899-1902).
This latter point-just six months between Ladysmith and Peking-is perhaps the point of real importance in terms of unprecedented reach and scale of operations. The suppression of the Boxer Rebellion did not necessitate the deployment of forces from Europe, the various powers having naval and military units and formations on station in China or generally in theater, and the fact that these formations were involved in operations did not constitute the most distant of operations, but most certainly by 1900 the scale of operations, and specifically the scale of operations in terms of the movement by sea of military formations to the theater of operations, was unprecedented. Within six months of the outbreak of war, the British had deployed some two hundred thousand military personnel to South Africa, and these included British, Indian, and dominion troops, the latter being drawn from Australia, Canada, and New Zealand. Certainly such numbers were overshadowed by those of 1812 and 1904-1905, but the fact was that the impact of the Industrial Revolution and the process of imperialist expansion in the nineteenth century really did come together with reference to the deployment of formations in South Africa in terms of time, scale, and distance that were unparalleled.
NOTE
1 . The 1901 edition of Brassey s Naval Annual , p. 66.
PART 2

FROM PORT ARTHUR TO BUCHAREST, 1898-1913
INTRODUCTION
T HE YEARS BETWEEN 1904 and 1922 are all but synonymous with the Anglo-German naval race and the First World War, yet this was a period that saw five major wars involving great powers and two, not one, major naval races. It was a period that opened with the Russo-Japanese War (8 February 1904-6 September 1905) and then moved to the Italian-Turkish (29 September 1911-15 October 1912) and the Balkan (17 October 1912-10 August 1913) wars, the latter in many ways serving as the overture to the First World War (28 July 1914-11 November 1918), which in turn gave rise to a series of wars, the most notable being the Russian Civil War and Allied intervention (December 1917-October 1922). The latter, of course, was accompanied by the Russo-Polish War (April 1920-18 March 1921), and there was also the small matter of the Greco-Turkish War (May 1919-October 1922). Between 1906 and 1914, the unfolding Anglo-German naval race was one of the most important single items on a political and diplomatic agenda that made for an increasing militant and strident assertiveness that went hand in hand with an increasing sense of insecurity on the part of all the powers. Yet the First World War was witness to a second naval race in the Pacific between Japan and the United States, which after 1919-1920 was to be curbed in the attempt by the great powers to craft a new international order that included the first arms limitation arrangements.
Inevitably this is a period dominated by the First World War. But the provision of explanation, as opposed to the mere recounting of events, presents immediate and very considerable difficulty, not least in terms of relating sea power to the outcome of events. The main difficulty is the distinction between causation and occasion, between the causes of events and the point of time when these manifested themselves, and, of course, the latter provide cause and momentum. If one were permitted, one would cite two matters as examples of the difficulty of distinguishing between first explanation and narration and second causation and manifestation with reference to the First World War.

With reference to explanation and narration, one would cite the subject of trench-lock and its treatment by generations of historians. Here the problem of interpretation is obvious: several generations of historians and military commentators have provided the answer and, unfortunately, it is the wrong answer. There is no single answer because what is termed trench-lock was not the product of any one cause but the result of the coming together of a number of factors. Two, whether singly or in combination, always form the first line of alleged explanation: the superiority of defensive firepower over offensive firepower and the superiority of strategic mobility over tactical movement. One more matter is often cited as complementary to these two: the lack of the systems that in the Second World War were to unlock fronts. These individual systems are usually identified as the tank and aircraft, the point being that during the First World War their very limited capabilities precluded their use as the means of breakthrough. This raises the wider issue of context because the assumption that tanks and aircraft were the means of breakthrough is contentious and the fact is that these, in terms of their absence or very limited offensive capabilities, do not explain why there was deadlock in the first place. The critical development was not tanks, aircraft, or motor transport but the miniaturization of the radio, which made possible effective command and control at the point of contact, though in terms of unlocking of fronts it was the combination of tanks, aircraft, motor transport, and radio that was important, not one single development.
The secondary factors usually paraded as explanation of trench-lock are that terrain worked against the attack, that surprise was difficult to achieve, and that the historical means of ensuring mobility, the use of an open flank, was not available. And to these can be added another: given the rapid degradation of formations committed to offensive operations, any attack invariably reached its culminating point very quickly-witness the returns registered in the British offensive at Amiens in August 1918. All these facts of life, and others, were at work and contributed to the tactical impasse of the First World War. It was very difficult to register surprise, and no enemy position could be outflanked. The Germans had the pick of the ground after November 1914, and therefore Allied armies, committed to the offensive because the war was being fought on Belgian and French soil, were faced with major difficulty, and one that worsened over time because of the defense added to its power in terms of depth and firepower with every year. A rudimentary trench system in 1914 evolved by 1917 into a defensive system with three main lines of resistance, sited on reverse slopes wherever possible and with the forward positions held lightly. The German defensive systems on the Western Front between 1914 and 1917 successively involved a tier a year and evolved a step ahead of the attack, at least until November 1917 at Cambrai, and in this evolution the defense acquired a depth that ensured that it could not be broken in a single offensive operation. To put the matter in reverse, successive Allied offensives were conducted a year behind requirement. With each successive year there were tactical innovations but, in effect, what was attempted in 1915 was what would have been needed in 1914 to have overcome a defensive position, and this phenomenon repeated itself with every passing year, at least until March 1918, and therein was irony. The German offensive that was unleashed in 1918 was again a year behind reality, not so much because of tactical considerations but because of the strategic reality Germany had created by itself by bringing the United States into the ranks of its enemies in April 1917.
Herein one begins to get to the real reasons for trench-lock on the Western Front: the identification of a mental rather than a material problem at the heart of indecision. The offensive, burdened as it was by problems that did not encumber the defense, could not match the latter s rate of learning, at least not until March 1918. But leaving aside the detail of the German spring offensive, even the basic point-the faster rate of learning of the German defense compared to the Allied offence-begs the question of how to overcome a defense that was too big to be defeated. At the heart of the indecisiveness of the Western Front in the First World War are two realities: armies had become so large and possessed such powers of recuperation that they could not be defeated in the course of a single battle or campaign. The fact was that militarily deadlock in the First World War was the result of a decisive victory being beyond any power because all armies were too strong to be overwhelmed in a single attack. And herein lies part of the reason for the deadlock of the Western Front: for the most part high commands were committed to the idea of the Vernichtungsschlacht , the integral and decisive battle of annihilation, which was incapable of realization.
The inability to break the deadlock of the Western Front between November 1914 and March 1918 stemmed from a basic failure of understanding on the part of the high commands of the nature of war and the nature of a campaign, or more precisely to the confusion of a campaign with a single battle. An investment of belief in the decisive battle served to obscure the reality that only a campaign that embraced simultaneous and separate efforts offered any chance of victory. The failure of commands to understand the distinction between a campaign and a battle and to realize that the only possible way in which a front might be opened was by a series of offensives, set pieces, and related battles, and not the attempted single battle of annihilation, was crucial to offensive failure. But the majority of military historians and commentators very seldom define the capacity of warring states to wage war by generating resources of unprecedented scale in such terms. Refuge is taken in the notion that wars between great, industrialized powers necessarily are protracted and attritional and hence cannot produce decisive campaigns or battles. But the problem with any and all of these explanations of deadlock on the Western Front is that they are not explanations: singly and together they describe the battlefield rather than explain the indecisiveness of battle.
Deadlock in the First World War was not the result of imbalances of firepower and movement, size of armies, conditions of ground, or technical factors affecting the conduct of operations. Deadlock had nothing to do with either the capacity of the powers to wage total war or even possession of the means to do so. It was about their willingness to wage total war, their willingness to continue to prosecute war despite the indecisiveness of battle, their hardening determination to fight to a finish in justification of the losses that had been incurred already that explains the phenomenon of trench-lock. It was the willingness of societies to fight on, despite and because of the elusiveness of success on the battlefield-a social cohesion and a failure, and perhaps inability, of societies to collapse under the strain of total war that by rights should have destroyed them. Trench deadlock was a military phenomenon, but primarily it was a military reflection of a political and mental phenomenon. 1

The second matter, causation and occasion, is perhaps even more contentious, not least because matters beg ready definition. If one looks at the collapse of the Central Powers in autumn 1918, then the various ingredients of defeat can be identified with relative ease. There were elements of political, economic, and military defeat that had come together by this time in what was a siege war in which the powers of central Europe were defeated by enemies that were able to wage global war by mobilizing manpower and economic resources across the world. By the time that Germany sued for an armistice in an attempt to avoid final defeat, she and her allies were on the brink of political, economic, and social collapse. Industry was in or at the point of entering end-run production, while shortages of food and domestic heating pointed to a winter of something beyond discontent, though conditions in Germany were very much better than anywhere else in central and eastern Europe. But in the attribution of defeat the general account suggests that the failure of the German spring offensive, the Kaiserschlacht , and then the accumulation of defeats beginning with 8 August and The Black Day of the German Army were crucial, but again, as with trench-lock, this describes rather than explains.
There was no single cause but rather a series of matters coming together that made for victory and defeat, but in this process one would suggest a single episode, very seldom afforded much in the way of historical attention, and certainly not much in the way of American or British attention, that may possess an importance that has been overlooked: the collapse of Bulgaria in September 1918 as a result of the Allied offensive from Salonika, which ultimately saw French formations reach the Danube. The reason this one episode may be argued to possess singular importance lies in two matters. First, it was the first offensive of the war that, at least from the Allied perspective, saw a military victory translate into the defeat of an enemy state. It was not just a military victory over the Bulgarian Army but a defeat that embraced the nation, and that simple fact pointed to a hardening of Allied determination on the Western Front as, for the first time, the prospect of real victory dawned. Second, it was a defeat that brought the Hapsburg monarchy to the realization that its end was nigh. 2 Bulgaria asked for an armistice on 25 September 1918, and it cannot be mere coincidence that in the next two days Austria-Hungary informed Germany of its need to leave the war and the German chancellor, Count Georg von Hertling (1843-1919), was obliged to resign: on the 29th, the day after the infamous seizure and collapse of Quartermaster-General Erich Ludendorff (1865-1937), the German military demanded that a request for an armistice be made immediately. Bulgaria s collapse was not simply the result of the defeat in front of Salonika. There were a number of matters in the making, not least the realization that the German spring offensive had failed, that Germany s certainty of victory was no more than an empty boast. There was by September 1918 the virtual collapse of Bulgarian industry and trade, and by this time Bulgaria faced the prospect of a disastrous winter after an abysmal harvest. These matters, along with such items of importance as the Russian example of social revolution, were there in what was a two-way process between cause and occasion. The Bulgarian collapse was the result of a defeat some three years in the making, but the defeat of September 1918 drew together these various elements of national defeat and brought the process to its end. Its importance lay in the fact that it initiated the final phase of a process in which what bound the Central Powers together unravelled, and did so with remarkable rapidity and totality: just eleven days separated the Bulgarian defection and Germany s request for an armistice.
And, of course, that leaves just the small matter of where navies and sea power enter the equation.
CHAPTER FIVE

THE RUSSO-JAPANESE WAR: THE FIRST PHASES
I F ONE ACCEPTS the idea, expounded previously, that the Battle of Manila Bay was the last battle of the Age of Sail, then it would follow that the Russo-Japanese War represented the first naval war of the modern era. This conflict did not witness the employment of aircraft, 1 but it obviously embraced a series of actions that marked out the road to Jutland.
In terms of war at sea, there are a number of matters that should command attention and careful consideration, not least the fact that this war was the first in which electronic counter-measures made their appearance: Russian wireless operators first jammed the radio signals of Japanese destroyers operating off Port Arthur in February 1904. It was, moreover, the first war in which mines were of very real significance for the conduct of operations. Both defensively and offensively, mines were to be important in two world wars, as Japan found to its cost in 1945 when American mining and Operation Starvation formed part of the process that completed Japan s defeat. But, and with exception, probably neither war saw mines afforded the strategic significance registered in the Russo-Japanese War.
The mine had claimed its first victim in the course of the American Civil War, and if it had not played any real role for the remainder of the nineteenth century, it emerged as a formidable weapon that wrought immense strategic consequence in terms of denial of sea areas and the infliction of losses. Arguably the moment of the mine was 18 March 1915, in the Dardanelles, and in a sense the strategic significance of the mine in the First World War as a check upon British sea power represented the peak of the strategic achievement of the mine. But the fact was in the Russo-Japanese War the only Japanese battleship losses were to mines, and the impact of the events of 15 May 1904 when the Hatsuse and Yashima were lost can hardly be understated. Overall the mine was the main cause of Japanese naval losses, and this latter situation, mutatis mutandis , never repeated itself in any subsequent conflict.
More significant for the history of war at sea, the actions between major units in this war were fought, or at least were opened, at ranges that were unprecedented, 2 and most certainly the experience of battle seemed to point to the crucial advantage of superior speed in the conduct of fleet operations. These matters, plus the fact that the Japanese ships generally had more medium and secondary guns than Russian ships, provided the basis for much of the subsequent argument about the dreadnought concept, but in real terms this war was important in what was not present. There were no aircraft or airships, there was no real contribution on the part of submarines, and in real terms there was no guerre de course . In these aspects of naval warfare this conflict did not point in the direction of the First World War.
The war on land most definitely did so in two respects, namely the protracted battle conducted over extended fronts and the first, very tentative moves toward the concept of the operational-as distinct from the strategic and tactical-level in the conduct of war. Moreover, there were certain curious similarities between the origins of this conflict and those of the First World War. What was at stake was security and status with so much tied to the single issue of Port Arthur. In truth, however, Port Arthur was more symptom than cause, and in many ways there was a curious similarity in Russian policy with reference to the real issue-China and the relative position of the powers-and Russian policy through much of the first half of the nineteenth century with reference to the Ottoman Empire. In both cases Russia hesitated between seeking to exploit the weakness of its neighbor in order to exact major concessions for itself and playing the role of protector-albeit at a price to the protected-against the demands of third parties. The Sino-Russian treaty of 3 June 1896 was very blunt in this matter, and it was primarily directed against one country, Japan. 3 At stake was the position of Japan and Russia relative to China and Korea, the immediate conflict of interest being Korea-and Japan s concerns for its security were obvious given the immediate threat that Korea in Russian hands would present-and Manchuria in terms of economic interest and spheres of influence. In these matters the Japanese position was more basic than the Russian: Russian interest was more concerned with status, but Japan was concerned with access to, and indeed control of, food and raw materials. Japan wanted Manchuria for its resources that were essential to great power status, and as an area of colonization, Manchuria s sparseness of population contrasted with what seemed at the time-and which by present-day standard would seem most modest-the overcrowded home islands.

The terms of reference of the Sino-Japanese and Russo-Japanese wars were to be very different, and the immediate point of difference can be defined very easily. First, though the Sino-Japanese War definitely had a naval dimension and was very properly a war that assumed joint service dimensions, in the final analysis this war was primarily military in a way that the Russo-Japanese War was not. In the 1894-1895 war defeat at sea was of minor consequence to Imperial China and most certainly was of minor consequence when set against the defeat that was taking shape on land as 1894 gave way to 1895. This was not true of the war of 1904-1905: Russian defeats on land and at sea ran in tandem and supplemented one another, and, of course, it was the final defeat at sea that persuaded the tsarist regime to cut its losses rather than seek to increase its commitment in the Far East in the belief and hope that Japan s defeat in a protracted conflict would be achieved. Second, timing and geography in these two wars was very different in the sense that Japan could not undertake landing operations on the Korean east coast lest Russian forces at Vladivostok, closer to Gensan than Japanese forces at Sasebo, 4 be able to intervene with obvious advantages of distance, time, and choice, and Japan could not undertake any initial move into northern Korea, in the form of landings at Chinampo 5 or further to the north around the mouth of the Yalu, because it was not until March that the ice melted sufficiently to allow major landing operations. 6 Such problems had not been present in the 1894-1895 conflict, not least because of the third point of difference. In the 1894-1895 war the main theater of operations was Korea, and Japan s initial and primary concern had been to secure control of Korea and then to carry the war outside its borders. In the 1904-1905 war Korea was but one of three theaters-the Liaotung peninsula and Port Arthur being the second and southern Manchuria the third-and in a very obvious sense the Korean theater, with its very limited rail facilities and with tracks rather than roads that rapidly became impassable in heavy rain or thaw, presented major problems of time and distance for the movement of major Japanese formations. Politically for Japan it was very important, perhaps the most important single matter, that it secure Korea, but more than 400 miles/640 km separated Fusan and southeast Korea from the lower Yalu. With a best rate of advance of perhaps 12 miles/19 km a day, the Japanese needed to avoid an advance up the length of the Korean peninsula. But a powerful Russian naval force was concentrated at Port Arthur, making landings on the west coast of Korea problematical unless the Japanese made recourse to a pre-emptive attack that took the form of the Iai and that, in a later context, assumed notorious dimensions.
The Iai is a stroke in the repertoire of the Japanese swordsman. It is a surprise blow struck at the outset of combat without the preliminary ritual or the customary exchange of courtesies. Delivered by a right-handed swordsman, the blow is struck in a single sweep as the sword leaves the scabbard, cutting up and through the opponent from the right hip to left shoulder. In resorting to the Iai , the swordsman s intention is to strike a surprise blow from which there can be no recovery. In February 1904, and again in December 1941, by resorting to such expediency the Japanese military sought to destroy or incapacitate the enemy and thus secure decisive advantage.
The Japanese attack on the Russian naval force outside Port Arthur on the night of 8-9 February 1904, two days after Japan had severed diplomatic relations with Russia and two days before the declaration of war, provides real problems of comprehension, not least because accounts of the war invariably begin with the attack on the Russian warships gathered outside Port Arthur, whereas the first act of war, the Japanese occupation of Fusan, Masampo, and Chinhae Bay, 7 took place on 6 February. 8 What is difficult to understand about the Port Arthur operation is the Japanese decision to divide the destroyer force that was to mount the attack between Port Arthur and Dalny. If it was so important to neutralize Russian naval power, and it was known that the greater part of Russian naval numbers was gathered at Port Arthur, then the sending of two of the five destroyer flotillas to Dalny 9 in order to attack whatever Russian ships might be there makes little sense. In the event these eight destroyers encountered no Russian warships, 10 while the destroyers from the three flotillas committed to the Port Arthur attack conducted uncoordinated attacks over an hour. The number of destroyers that conducted these attacks varies according to source, 11 but the Japanese records indicate that ten destroyers from three flotillas carried out the initial torpedo attack. 12 The Japanese destroyers fired two torpedoes apiece, and it does beg the imagination what the Japanese hoped to achieve on the basis of an attack with just twenty torpedoes, but three Russian ships-the three ships at the rear of the two columns that were nearest to attacking Japanese destroyers-sustained single hits to port. The first-class protected cruiser Pallada was hit first, amidships near a bunker. Thereafter two first-class battleships were hit: the Retvizan was struck forward and the Tsarevich was hit aft and suffered the flooding of her steering compartment. All three Russian ships were able to get under way with the intention of reaching the shallows. The Pallada ran herself aground near the lighthouse on the western side of the harbor entrance, while the two battleships ran themselves aground at the entrance to the harbor, fortunately but only by a narrow margin not blocking the channel in so doing. 13 Both the Pallada and the Tsarevich were moved into harbor that same day, but Port Arthur lacked a dry dock that could accommodate the Russian battleship. 14
As planned this attack was followed in the morning by a reconnaissance by the Japanese cruiser division that was intended to ascertain the results of the previous night s endeavors. The Japanese were able to discern three damaged Russian warships, but with the greater part of the Russian battleships and cruisers outside the harbor, and apparently in an unprepared and disorganised state, 15 the decision was taken to seek a gunnery action in an attempt to compound the assumed success. In an action that lasted about forty minutes minimal results were registered. Four Russian armored cruisers were lightly damaged, the most extensively damaged needing just ten days repairs, while no Japanese warships sustained even modest damage. The most notable feature of this exchange was the fact that the Japanese line engaged at ranges between 7,000 and 9,000 yards/6,420 and 8,260 m, which was unprecedented. Yet there was no way the Japanese warships could prevail given the Russian refusal to seek sea-room and a fight, the Russian formations staying under the cover of their shore batteries. It was the latter that provided the more effective fire and that led to the Japanese decision to break off the action. 16 The negligible results registered by the Japanese with this attack suggest one conclusion: if there was to be an action involving the Japanese battleships-and it is possible to argue that with the main force landing then taking place this action placed the secondary objective ahead of the first-then the battleships should have been committed either immediately after the destroyers attack or with the dawn s first light. 17
But there are other questions concerning this attack that relate equally to the two sides. On the Japanese side there was an assumption of victory that over time became part of national mythology. It seems incredible that this should be the case given that half the attack force was needlessly diverted to a secondary objective and achieved nothing while the main attack failed to sink a single Russian ship. Over time, and specifically after the twenty-fifth anniversary celebrations and the funerals of leading commanders in this war, there emerged a popular tradition about the fighting that allowed for no questioning. But perhaps the most appropriate comment on the example of the Russo-Japanese War was provided, quite inadvertently, in the first contact of another war.
When the first wave of Japanese aircraft appeared over Pearl Harbor, the strike force leader, Lieutenant Commander Fuchida Matsuo, could scarcely believe what he saw. The Pacific Fleet, resting placidly at anchor, with the battleships neatly aligned in two rows beside Ford Island, put him in mind of an earlier war, one that also began (when he was three years old!) with a surprise attack by Japanese naval forces on an enemy fleet at anchor. Have these Americans never heard of Port Arthur? he thought, just before sending an encoded radio transmission ( Tora ! Tora ! Tora !) informing his superiors that Japan had achieved surprise at the beginning of this war as well.
Military historians analyzing the Pearl Harbor operation in the larger context of a Japanese way of war owe Fuchida an eternal debt of gratitude for his observation. In referencing the incident that kicked off the Russo-Japanese War of 1904-1905, Fuchida, who spoke with the voice of authority and soon-to-be-acquired experience, greatly simplified the historians job of establishing the historical links needed to legitimize and explain the idea of a Japanese way of war. The Port Arthur-Pearl Harbor analogy is obvious, so much so that Fuchida s acumen would be open to question had it escaped his notice. But here the obvious tends to obscure what is truly important. Port Arthur specifically and the Russo-Japanese War generally presage Pearl Harbor and the Pacific War in ways that Fuchida (and most historians) have failed to detect. Fuchida is to be pardoned for this failure: he was, after all, quite busy at the time. Otherwise it might have occurred to him to ask the same question from the Japanese perspective: Have we, the Japanese, never heard of Port Arthur? The answer, fully considered, should have given him and his countrymen pause.
Like the Pearl Harbor operation, the Japanese attack at Port Arthur was executed in two stages. First came the initial surprise attack by torpedo-firing destroyers. This occurred ten minutes before midnight (Port Arthur time) on 8 February 1904 against Russian warships of the 1st Pacific Squadron, then anchored in the roadstead outside the harbor. Some twelve hours later the main body of the Japanese fleet, commanded by Vice Admiral Togo Heihachiro, steamed into range and commenced shelling the Russian ships and the port.
It is important to grasp that the dual assault on Port Arthur achieved little of substance. Serious damage was inflicted on Russian cruiser Pallada and battleships Retvizan and Tsarevich . All were grounded; all were raised and repaired and would participate in subsequent engagements with Togo s ships. In fact Retvizan would be sunk yet again and raised again, by the Japanese, who renamed it Hizen and incorporated it into their fleet. The Japanese coup de main failed to deliver the knockout blow they had sought. The Port Arthur squadron survived the Japanese surprise attack and in so doing helped prolong Russian resistance in the siege that followed, thus prolonging the war as well. The doctrine of surprise, a cornerstone of Japanese military strategy, had proved a nonstarter. Yet it would remain a key element of the Japanese way of war, as the Pearl Harbor operation would demonstrate. 18
The point is well made, and would seem to be beyond serious dispute, not least because of the very nature and terms of reference of the Russo-Japanese War that western treatment of this conflict very seldom properly considers but which most certainly was identified in this note. It was that just as the Japanese paid various Manchurian bandits and outlaws to carry out attacks on the Russian rail lines and generally to engage in what amounted to a guerrilla campaign in Russian rear areas in order to ensure a defensive dispersal of Russian forces, so Japan sought to ensure sympathy for her position and objectives within the United States by a deliberate orchestration of the political process and press, and with one basic objective that really makes a neat counterpoint to what was to happen in 1941. Japan calculated that it could defeat Russia in a short, limited war, but that it would lose a protracted war. Japan therefore sought American mediation to ensure it of the gains it intended to record by force of arms, and this was to be achieved. The Japanese script was almost word perfect: anticipated early victories would be consolidated and underwritten by diplomacy, American diplomacy. In effect, American diplomacy would secure the victory that Japan could not secure by its own efforts. And in 1941 there was the initiation of hostilities by Japan, its basic aim being to secure the defeat of the United States in a limited war in which anticipated early victories would be consolidated and underwritten by diplomacy in the wake of American acceptance of defeat. The irony of such a situation is profound, though in truth it could be argued that the basic characteristic is not irony but nonsense. This point aside, however, the thesis may well be right, and perversely so: it was the Japanese, not their enemies, who had never heard-or had forgotten the real lessons-of Port Arthur.
But, and there is always a but, the Japanese attack achieved its basic aim. Despite what really were minimal losses, there was no coherent Russian response, and this lack of response is perhaps one of the strangest features of this episode. The damage that was incurred did not really amount to much in terms of the naval balance of power in the Far East. The Russian intent had been to concentrate units in the Far East with the intention of matching Japanese numbers by 1905, and by February 1904 the Russians could match the Japanese more or less in battleship and destroyer numbers, but they remained decidedly inferior in cruisers and torpedo-boats and were less well provided than their enemy in terms of base facilities, auxiliary shipping, and geographical position. Regarding the latter there was nowhere for Russian warships in Port Arthur and Vladivostok to go unless it was to a Sakhalin port or Petropavlovsk-Kamchatski. 19 Port Arthur might have had attractions as a base-and its docking facilities were very limited 20 -but any Russian force was isolated and Japan s geographical position in effect provided the basis of distant blockade.
Nonetheless Russian passivity in the immediate aftermath of the attacks of 8-9 February does defy immediate understanding. There was no measured Russian response to the Japanese attacks until there was a change of command with the arrival at Port Arthur on 8 March of Vice-Admiral Stepan Osipovich Makarov (1849-1904), 21 and quite clearly the personnel matter was important: while the commander of the Pacific Squadron, Vice-Admiral Oskar Viktorovich Stark (1846-1928), cannot be blamed for the February 1904 setback any more than Admiral Husband E. Kimmel (1882-1968) be reasonably held to account for the 1941 debacle, the lack of any coherent and sustained response after 9 February clearly did reflect the limitations of command arrangements at Port Arthur at and immediately after the onset of hostilities, though a codicil has to be added: in the immediate aftermath of the attack the preservation of the fleet-in-being was clearly very important to the Russians. But that point aside, one is left with only one conclusion: that the Japanese attack was not unlike the British infantry because it never seeks to surprise but to amaze the enemy. The joke aside, and the British infantry doing anything other than telling all and sundry of its merits is not a joke but true, the attacks of 8-9 February clearly had a wholly disproportionate importance relative to results achieved not so much in terms of surprise as shock.

The twin attacks on the Russian Pacific Squadron at Port Arthur went hand in hand with the first landings in Korea by Japanese troops, at Chemulpo. The Japanese plan of campaign embraced alternative courses depending on the degree of success registered in these attacks. If success was limited, then the Japanese military contemplated landings in Fusan and then an overland advance on Seoul; but if success was substantial and the Russian capacity to interfere with Japanese operations was compromised, then the main Japanese landings would be undertaken first at Chemulpo, with the main overland offensive effort directed first to Seoul and thence to Pingyang, and second at Chinampo, also directed toward Pingyang, with Japanese forces then advancing across a narrow front to the lower Chechen, between Anju and Yongpyon, 22 and thence to the lower Yalu, between Wiju and Chyangsong. 23 In the event of the main effort being made at Chemulpo there would supplementary landings at Fusan and Masampo, while Gensan was to be secured by troops advancing overland from Seoul. If necessary the Chemulpo landing would be abandoned in favor of landings in Asan Bay followed by an advance northward to Seoul.
The Japanese plan of campaign seems to have provided for every contingency except, perhaps, for the one that really counted. It is possible to argue that the initial Japanese moves in 1904 were exactly like the attack on the U.S. Pacific Fleet at Pearl Harbor in December 1941 in that both represented the wrong effort. In 1941 what was needed, if Japan was to take war to the United States, was not a carrier attack but a carrier attack in conjunction with assault landings whereby Oahu could be secured. In 1904 what was really needed was not a small-scale attack by destroyers and landings in Korea but landings directly on the Liaotung peninsula and immediate moves against Port Arthur. Such an action might well have forced the Russian hand and provoked a fleet action and thereby spared the Japanese the onerous and protracted responsibility of blockade, if, of course, that battle had been won.
The formations committed to the Port Arthur (and Dalny) operations sailed from Sasebo on 6 February and did so at the same time as three transports, with some twenty-five hundred men from the 12th Infantry Division, sailed in the company of four cruisers; late on the afternoon of the following day these effected a rendezvous off Single Island 24 with one cruiser and eight torpedo-boats from two flotillas that had proceeded separately. 25 The force then sailed to Baker Island 26 and a meeting at 0800 with the third-class cruiser Chiyoda , which over the previous ten months had been the duty ship at Chemulpo, and which had sailed from that port shortly before midnight. The Chiyoda reported the presence in port of the first-class cruiser Varyag and the aging gunboat-dispatch vessel Koveetz , along with single American, British, French, and Italian warships, and a Korean gunboat. 27 Given the clear disparity of strength, the decision to proceed with the Chemulpo landing was a foregone conclusion and it was decided that the Naniwa, Niitaka , and the Takachiho should remain outside the harbor, in the general area of Philip Island, while the other units covered landings that began that same evening, that is, before the Port Arthur attack was conducted. By 0300 next morning all Japanese troops were ashore and more than half their number had been dispatched by rail to Seoul. 28
As the Japanese ships approached the harbor they encountered the Koveetz leaving for Port Arthur. The Russian ship, divining Japanese intent, turned back and the following morning the Russian ships were given an ultimatum: at the same time the neutral warships were advised to clear the harbor. The latter refused such invitation and delivered a protest at what was and would be a clear violation of Korean sovereignty and neutrality, but as the Japanese prepared to enter the harbor the two Russian warships were seen to be leaving the harbor in order to fight an action that all knew could only result in their destruction. The Varyag was simply overwhelmed in a 55-minute action that ended with the Russian ships making their way back to their moorings, the Russian cruiser being steered by engines, on fire, and with all but two of her twelve 6-in./152-mm guns destroyed. Once secured, the Russians scuttled their ships, the Varyag by opening the seacocks and the Koveetz by charges. 29 Russian personnel were transferred to the British, French, and Italian warships, which along with local facilities took Russian wounded, and all personnel eventually returned to Russia under parole; the American ship that was present, the Vicksburg , refused to provide any facility for Russian personnel. 30 After the action the Japanese warships remained off Chemulpo until the following morning (9 February), when they left for Asan Bay. There they effected a rendezvous with the main force, after its operations off Port Arthur on the 9th, shortly after 1400 on 10 February. 31

Between 9 and 11 February Russian priorities were immediate: the bringing of all units into the Port Arthur harbor, the recovery of the Pallada and Tsarevich , the fitting of the Retvizan as a guard ship pending its final recovery, and the strengthening of the shore batteries around Port Arthur by adding to their number and ensuring that those that were in place could be properly used. 32 It was not until 11 February that all the Russian ships were able to negotiate the entrance and thereafter enjoy the security afforded by the Tiger peninsula, but by the time they did so misfortune had imposed itself on two Russian warships. First, one of the two minelayers stationed at Port Arthur, the 3,000-ton Yenisei , was sunk by one of its own mines in Talien Bay, and, second, the 3,200-ton cruiser Boyarin was mined when, along with four destroyers, it was ordered to sortie, following reports that the Yenisei had been sunk in an action with Japanese destroyers. 33 Such actions on the Russian part were necessarily defensive, and in effect ceded the initiative and use of the sea to the Japanese, but the latter s attempt to conduct another torpedo attack on whatever ships might still be in the roads on the night of 13-14 February foundered upon gale, blinding snow, and bitter cold. Two Japanese destroyers, unaware that the operation had been abandoned, separately made their way to Port Arthur and conducted attacks at less than 1,000 yards/920 m range against the Retvizan , but to no effect. The Russian shore batteries failed to register a single hit on either of the Japanese ships, but the attack did achieve one result: it brought to the Japanese high command the knowledge that the Russian warships were secure inside Port Arthur. On 16 February, and despite the fact that they had no watch on Port Arthur, the Japanese took the decision to proceed with the movement of the main formations to Chemulpo. At the same time five auxiliaries were taken in hand in preparation for an attempt to block the entrance to Port Arthur and thus prevent the Russian squadron from venturing forth to contest Japanese moves, rendering the Russian warships the prisoners of the security that they sought.

This was to be the first of three operations mounted-on 24-25 February, 34 27 March, and 2-3 May-with the aim of blocking the channel, and it was one of five quite separate naval efforts that unfolded at this time and that together form the second phase of the war at sea, between mid-February and the beginning of May 1904. The war itself can be divided into five phases-the initial phase; this second phase of mining and associated operations; a third phase of Japanese landings on the Liaotung peninsula and the two actions that separately spelled the end of the Russian forces at Port Arthur and Vladivostok, the battles fought in the Yellow Sea on 10 August and off Ulsan four days later; 35 the fourth phase between mid-August 1904 and January 1905 that saw the Japanese complete first the close investment of Port Arthur and then the capture of 203-Metre Hill, which facilitated the bringing of directed fire against Russian warships in the East Port and roads astride the Tiger s Tail; and the last phase, between January and July 1905, that witnessed the final moves of Russian formations that had been sent from Europe to the Far East and the battle of Tsushima, 27-28 May, and the Japanese landings in and occupation of Sakhalin in June-July 1905. This second phase of the naval war involved the blocking efforts, both Japanese and Russian mining operations, the sorties of the cruiser squadron based on Vladivostok, the Japanese landings at Chinampo, and the fleet operations initiated by Makarov on 8 March that ended on 13 April with the loss of the flagship Petropavlovsk to a mine outside Port Arthur, the battleship Pobyeda being mined but not lost at the same time.
The three blocking operations involved totals of five, four, and twelve Japanese auxiliaries, and the most suitable comment on these operations and their failure is provided in one simple fact. At the start of hostilities the Kaigun had a total of sixty-one merchantmen of 161,530 tons that had been requisitioned for auxiliary service, and the three blocking operations involved 21 ships of 48,026 tons; the fact that the Japanese were willing to write off what was the equivalent of one-third of their requisitioned auxiliaries at the start of the war was comment on the importance that they attached to these operations. 36 The major increase in size of the third effort compared to the previous two reflected the increased importance of containing the Russian naval force within Port Arthur given the Japanese plan of campaign that provided for two simultaneous efforts, namely first the crossing of the Yalu followed by a landing at Takushan 37 with provision for an advance into southern Manchuria and, second, a landing at Pitzuwo at the neck of Liaotung peninsula. The very odd point about the third operation was that the Yalu was crossed on 29 April before the third blocking operation, and the landing on the Liaotung peninsula was conducted despite the clear failure of this third effort: no less strange was the Kaigun claim that the 2-3 May effort had been a success. It was one of the features of Japanese reporting in the Second World War that claims of sinkings were all but inversely related to reality, but at every level of command there was a ready acceptance of claims, even those that treaded on the borderline between the fantastical and the grotesque. Why Togo, the Japanese commander, should have seen fit to claim that this third blocking operation had been successful is not clear unless it was prompted by inter-service calculations, and most certainly the Kaigun could not admit to have failed at the very time when the Rikugun was pressing into southern Manchuria.

The mining effort by both sides was to be crucial in terms of the major losses incurred by both sides, though the major Japanese losses-of the battleships Hatsuse and Yashima -came in the next phase of operations. The main Russian mining efforts were across the entrances of Kerr and Deep Bays, outside Port Arthur, off the Laotieh peninsula, and in Eight Ships Bay, some 8 miles/13 km north of Port Arthur on the reverse side of the Kwantung peninsula, either off the headlands or in the bays where Japanese warships might seek to shelter as the need arose. Japanese mining, naturally enough, was directed to the sealing of Port Arthur, and these mining efforts, within sight of the shore, could not be countered by a Russian command that lacked anything other than contact mines.
The mining efforts inevitably had to lead to a clash at some time, and with Makarov taking up command the clash was immediate. The first encounter was between four destroyers on both sides on the morning of 9 March. On the following day there took place an action between a Japanese destroyer flotilla and two Russian destroyers that had become separated from their parent formation. One Russian destroyer, badly damaged, managed to escape, but the second, the Steregushchi , was disabled just as it came within range of shore-based batteries. It was abandoned and then sank after having been boarded by the Japanese. Makarov took the third-class cruiser Novik (with the armored cruiser Bayan ) in vain support of the Steregushchi only to encounter Japanese battle and cruiser formations arriving on the scene. The Russian ships were able to return to Port Arthur, but the personal behavior of the admiral in this episode did much in the way of restoring morale and belief in the leadership. 38
Nevertheless, the basic Russian naval position remained unchanged. The real problems confronting the Russian command at this stage were two-fold. The Japanese operations were staged in part as distant cover for the landings by troops of the 12th Infantry Division at Chinampo, where an advance party was put ashore on 10 March with the main force landing on the 13th. The inescapable fact was that whatever the Russians might try to claim as a result of these exchanges, the Japanese position, with troops advancing on and securing Pingyang, had been strengthened by what amounted to all but complete control of the Korean peninsula. 39 Of more immediate concern, and second, as Makarov was to find in the coming days, was that the exchanges demonstrated a very low state of training of Russian formations and individual ships. There was very little he was able to effect in this matter in the few days of life left to him. 40
The Japanese force coming on the scene did so because it had been tasked to conduct a very deliberate bombardment of Port Arthur from a position south of the Laotieh peninsula, beyond the range of most of the shore batteries and at an unprecedented range of 15,250 yards/13,990 m achieved only by the Japanese ships listing in order to provide sufficient elevation for their guns. 41 Despite four hours of shelling damage was minimal. The only damage of any note, ironically, was to the Retvizan : her cofferdam was punctured and she was beached on a shoal. Comment on the effectiveness of the Japanese attack was provided the following day when Makarov took his fleet to sea for an exercise, the Russians very deliberately picking the time when the Japanese main force could not be present. The Japanese returned and mounted another attack on Port Arthur on the night of 21-22 March, this time by torpedoes directed against any Russian warships that might be off the harbor entrance, and followed this attack with another bombardment. This time a Russian force left harbor, but with the Russians unwilling to leave the cover provided by shore batteries and the Japanese wary of closing, no serious action resulted. The Russian force, which had taken more than four hours to leave harbor, returned to base in mid-afternoon. 42 This was the first occasion in the war to date when a Russian task force, with battleships, cruisers, and destroyers, sailed from and returned to base on a single tide and exchanged fire with the enemy.
These exchanges prompted moves by the two sides that were perhaps predictable, at least in retrospect. The Russian response to the Japanese attack was to mine the waters off the Laotieh peninsula and to establish a number of gun positions covering this minefield. The Japanese response, in addition to continuing patrols and mining, was a second blocking operation on the night of 26-27 March. This operation proved one ship too few since Russian battleships could still negotiate the entrance, albeit with special care. The next day the Russians put to sea and stood off the Laotieh peninsula, but the Japanese battle force declined action. In the following days the Russians defensive measures in the form of boom defenses, blockships of their own, new gunnery positions ashore, and constant patrolling off the port by destroyers that had been some time in the preparation 43 began to slot into place, with obvious implications for future Japanese intent. For their part the Japanese adopted what was in effect a precursor of 1914 and distant blockade with the establishment of a cruiser line of observation between the Shantung peninsula and the Sir James Hall group of islands. 44 On 11 April the Japanese fleet went to sea with the intention of conducting a mining operation as the prelude to a series of operations that would see a third blocking operation and landings on the Liaotung peninsula in the sector between the Lilan and Tasha rivers, 45 to be followed by an advance on Chin-chou. The occupation of this town would place Japanese formations astride the rail line linking Port Arthur with the outside world and would then be followed by advances in two very different directions, toward Port Arthur and with Japanese formations crossing the Yalu into southern Manchuria back northward via Pulantien, Kaiping, and Haicheng toward Liaoyang. 46
The Japanese mining effort on the night of 12-13 April was seen by the Russians. On the following morning there was a series of confused actions initially involving Russian destroyers that had become detached from a force sent the previous day to reconnoiter the Elliot Islands and to attack any Japanese warships, the Russian suspicion being that the group was being used by the Japanese as an advance base. In the initial exchange one destroyer, the Strashni , which in the dark had mistakenly joined the company of a Japanese destroyer flotilla, was sunk. A Japanese attempt to intercept other Russian destroyers was frustrated by the appearance of a Russian cruiser, but as a Japanese force of ten cruisers (operating in support of the destroyers) closed the Bayan it was joined by first three other cruisers and then two battleships. Despite their numerical advantage the Japanese formations withdrew with the intention of drawing the Russian ships forward and onto the guns of the Japanese battleships that were in support. The latter s presence, shrouded in the morning mists, was unknown to the Russians, but when the Japanese battle force was sighted the Russian force turned away in order to get under the cover provided by shore-based artillery. In so doing Makarov s flagship, the Petropavlovsk , ran into the minefield that had been laid the previous night. It hit a mine that set off explosions in both a magazine and boiler room, and it sank in some two minutes with only seventy survivors. At this stage the Russian force did not seek the safety of the Port Arthur base but remained on station until a second battleship, the Pobyeda , was mined. 47 It was able to get back to Port Arthur and was followed by the other Russian ships; the Japanese formations, after a wholly ineffectual indirect bombardment of Port Arthur the next day, then proceeded to an advance base, in Thornton Haven off Haiyang Island.
The action of 13 April in effect closed this second phase of the naval war, the activities of the Vladivostok force notwithstanding. The loss of two battleships-one sunk, one damaged-left the remaining Russian formations at Port Arthur massively inferior in numbers to the enemy, while for many Makarov s death spelled the end of hope. But there was also another factor at work, and one that was wholly unintended. In the aftermath of the 13 April episode the Russian government took the decision to send formations from the Baltic as reinforcements for the Pacific Squadron, leaving the latter with nothing more than a defensive, and largely passive, commitment. The formation had to be held in anticipation of the arrival of reinforcements and could not be used in any attempt to redeem present misfortunes. The passing of this second phase can be said to have been marked by the Japanese laying of three minefields off Skryplev Island, Shotka Island, and the Currie Channel in the approaches to Vladivostok on 28 April, after which time Japanese forces were re-concentrated in readiness for the next (and third) phase of operations, which would see the crossing of the Yalu and the landings near Pitzuwo.
The Yalu was initially crossed on 29 April in the area just above Wiju by formations from the 12th Infantry Division, but the main effort, with the Guards Division coming into the line, was made on 1 May. A third blocking operation was conducted on the night of 2-3 May and was pronounced a success by Togo, apparently on the grounds that Russian ships had not put to sea in order to fight the Japanese ships off the harbor as they had on the previous occasion. In reality, this operation came less close to success than the previous blocking attempt, in large measure because of the booms, blockships, and increase of guns that had been set in place by the Russians since the end of February. But the Japanese landings went ahead. Delayed by weather for a day, the first landings on the Liaotung peninsula were conducted on the morning of 5 May, and by 13 May the 2nd Army had two divisions ashore with a third in the process of being landed. The landings at Takushan, by a 10th Infantry Division that was to serve as the nucleus of the 4th Army, were conducted on 19 May, by which time other developments were afoot.

The most important of these developments was that with landing area secured the 2nd Army conducted an offensive that left it astride the Port Arthur-Liaoyang rail line just to the north of Chin-chou on 16 May; two divisions turned to the south while one turned to the north, though it was not until 13 June that the latter s advance to the north began. 48 The resultant isolation of Port Arthur, acknowledged from the time of the Japanese landings in the form of Petersburg s recall of the resident viceroy, has been largely overshadowed by the loss of the Japanese battleships Hatsuse and Yashima on the previous day. These two warships, like the Russian units on 13 April, blundered into a minefield laid some 24 hours earlier, and in both cases mines had been laid very deliberately in areas where the enemy had been seen to frequent. Also on this same day (15 May), and in its first two hours in thick fog and in a position some 60 miles/96 km southeast of Port Arthur, the Japanese lost the second-class cruiser Yoshino as a result of a collision with the armored cruiser Kasuga . These losses were bracketed by the loss of Torpedo Boat 48 and the dispatch vessel Miyako to mines off Cape Robinson on 12 and 14 May, respectively, and the loss of the destroyer Akatsuki on 17 May to a mine some 15 miles/24 km to the northeast of where the Hatsuse and Yashima had been lost. 49
The seriousness of the loss of two battleships notwithstanding, 50 these operations presented what amounted to a series of impossible strategic and tactical dilemmas for the Russian naval command at Port Arthur, not the least of which was the Japanese advance that resulted in the taking of Chin-chou and Nanshan on 26 May. With the capture of Dalny and the sweeping of Talien Bay, the Japanese were able to establish a secure base within easy reach of Port Arthur. The timing of the initial landings on the Liaotung peninsula meant that there could be no effective Russian response to the landings in the Pitzuwo-Cape Terminal area: by the time short-ranged Russian torpedo-boats and destroyers could reach the area, the transports would have left, while those that were en route, with follow-up formations, would be defended in such numbers as to preclude any chance of success. In addition, with the Russian government then in the process of deciding whether or not to send naval reinforcements from Europe to the Far East, the local command in Port Arthur necessarily was wedded to the fleet-in-being policy: the formations at Port Arthur had to remain intact in order to ensure that reinforcements lived up to that label. Only by a combination of two sets of forces, those at Port Arthur and those that might be sent from Europe, would the Russians have any chance of meeting the Japanese on the basis of equality and with any chance, however remote, of equalizing accounts at sea.
The Russian dream of winning superiority was probably wholly unrealistic at this stage of proceedings. This did not mean, however, that in this phase of the war the formations at Port Arthur were passive. At various times Russian cruisers, gunboats, and destroyers put to sea in order to provide fire support for forces ashore. There were many mining efforts, some in the absence of proper minelayers by harbor launches, 51 and there was, rather perversely, one fleet encounter that is seldom afforded much in the way of historical attention. Admittedly the 23 June action was one that was not deliberately sought by either side. The Russian command was intent on an exercise with all units and did not seek an action with Japanese formations, but a delayed departure from Port Arthur meant that the Russian force could not get back into base the same day because high tide was missed, and there was on the Russian side no anticipation that the enemy would be able to close as rapidly as was the case. But the Japanese did not press an action because an examination of the Russian line revealed the presence of all six enemy battleships and hence a marked enemy superiority of numbers. Japanese destroyers attacked at dusk, but ineffectually, allowing the Russians to get behind the booms at the entrance of the harbor that evening. The only damage was sustained by the battleship Sevastopol , which struck a mine. It and the other ships nonetheless were able to enter base the following morning, the Sevastopol being afforded the care and attention of the cofferdam that had previously seen employment with the Retvizan . 52
This third phase of the naval war did see two battles on the high seas, and they were occasioned by the Russian attempt to concentrate their entire naval force in the Far East at Vladivostok. This intent was prompted by the sudden vulnerability of Port Arthur: on 26 July Russian forces stood on the Shuangtaikou-Antzuling-Mount Laotso line, 53 well clear of Port Arthur, but by 30 July the Japanese had not simply overrun this line but had captured the last fortress on the Wolf Hills north of Port Arthur from which, for the first time, their artillery could shell the harbor and its warships. 54 In a manner of days, two Russian battleships had been hit, the unfortunate Retvizan some seven times, but none had been seriously damaged before the Russian squadron sailed on the morning of 10 August. At this time the main Japanese force was off Round Island, 55 and some three hours after the Russian force had cleared Port Arthur battle was joined with the Japanese having clear advantages in numbers, firepower, and speed.
The general treatment afforded the record of the war in general and this battle in particular has been very generous to the Japanese, but certainly in the first exchanges there was little distinction on either side. Despite what should have been considerable Japanese advantage in speed and maneuver, the Russian force secured a fleeting advantage of position that might have enabled it to clear the Yellow Sea. As it was the Russian fire was accurate but unlucky, with many straddles and some hits but none that caused serious damage to any single Japanese ship. With the ships at the rear of the Russian line taking increasing punishment, the difference between the two forces was to be summed up in one exchange. The Japanese flagship Mikasa was hit and the commander, Togo, was struck by a splinter in the nose and was soaked in the blood of officers who had been killed alongside him. The Russian flagship Tsarevich was hit by two shells that destroyed her foremast and conning tower, killing the force commander, Rear Admiral Wilgelm Karlovich Vitgeft (1847-1904). Everyone on the bridge was either killed or wounded. Bodies jammed the steering wheel with the result that the ship turned in full circle-narrowly missing the Sevastopol and Peresvyet in the process-and brought chaos to the Russian line. The Retvizan initially followed the Tsarevich , while the Pobyeda stayed on its original course as the other three battleships turned to starboard. Order was restored after minutes of confusion, and with the Japanese line trying to close to a range of 4,000 yards/3,670 m, the Russian formation turned back to Port Arthur. The Peresvyet, Poltava , and the Tsarevich had all sustained hits on or about the waterline, and maneuvering in anything other than a flat calm would have lead to serious difficulty. Most certainly these ships could not have attempted to make the passage to Vladivostok, though the serious condition of these three warships could not have been known by Rear Admiral Prince Pavel Ukhtomski in the Peresvyet when he gave the order to turn back to Port Arthur. 56
In any event, in so doing the Russian formation was to shed various warships. The Tsarevich , which was hit a total of nineteen times and at one stage was seemingly at the mercy of the Japanese battle line, managed to escape and reached Tsingtao, 57 where it was interned. The Diana also reached Tsingtao, where it took on coal and sailed to Saigon, where it was disarmed. The Askold and the destroyer Grozovoi reached Shanghai, where they too were disarmed. The Novik alone of the Russian ships managed to escape from the Yellow Sea and, after hurriedly taking on coal at Tsingtao, on 13 August passed through the Osumi Strait in an attempt to work its way to Vladivostok via the Japanese east coast and then via the Kunashiri Channel. 58 But it was sighted and Japanese cruisers were able to secure the advantage of position by moving across the Sea of Japan at speed. The Novik finished at Korsakov, in southern Sakhalin, 59 where it was scuttled on 20 August after an action with the Tsushima that saw both ships extensively damaged. In this action the Russian ship s steering was damaged beyond repair, and the arrival of a second Japanese cruiser, the Chitose , outside Korsakov 60 ensured that the Novik was doomed. It was the only warship of either side present at the battle of 10 August to be sunk 61 -an extraordinary situation given the clear superiority of numbers and position of Togo s formations in the aftermath of the damage sustained by the Tsarevich . Of the Russian warships that returned to Port Arthur only the Pallada sustained any serious damage after the main force actions. It was torpedoed when about to enter harbor, but it was able to reach a berth.

The fact that no warship was sunk in this main action, and perhaps more importantly the twin facts that no Russian warship was sunk despite the level of disorganization that engulfed the Russian squadron after the Tsarevich was hit and the separation of the Russian force into groups that nonetheless managed to reach various ports without loss, does seem rather remarkable, as indeed would seem to be history s treatment of this battle. Togo has long since entered the pantheon of national heroes that permitted no serious questioning or reproach, and quite clearly the pro-Japanese sentiment in Britain and the United States at that time, when combined with the prevalent Great Man concept, meant that there was an interpretation of this action that was somewhat questionable but that has basically remained in place since that time. But any objective assessment of this battle is one that would hardly afford Togo much in the way of status. His conduct of operations in the course of this battle seems somewhat hesitant and most certainly not authoritative, and the fact that at one stage the Russian formation secured positional advantage does take some explaining. Whether that advantage could have been anything other than temporary and could have been turned to real purpose is questionable, but the fact that the Russian formation was nonetheless able to leave the Japanese force astern is surprising. But three matters-perhaps pleas of mitigation might be a better term-do bear examination. First, the Japanese were outnumbered six to four in battleships, and there was no escaping the fact that these four battleships did represent Japan s bottom line. A certain caution, therefore, was perhaps in order, though this would hardly seem to extend to one argument-the second point-paraded as justification for Togo s decision, namely concern regarding the Russian formations coming from Europe. There may very well have been a certain level of concern, but the task immediately at hand, and the desirability of winning a crushing victory that might eliminate the threat presented by the Pacific Squadron, would seem to be more weighty considerations than concern about a force that at this time was still some two months from sailing from the Baltic. Third, notwithstanding the Japanese failure to sink a single Russian warship in this action, the battle nonetheless was a very real Japanese victory, albeit with one codicil.
The problem herein is definition and acknowledgment that issues of victory and defeat are not necessarily settled in terms of sinkings and captures, though in most cases these do go hand in hand. In this case, however, the elements of victory and defeat lay in different directions: without realizing it, Togo s refusal to close the Russian ships in an attempt to ensure destruction of units and formation for fear of sustaining damage to his own ships that perhaps could not be made good nonetheless resulted in victory. The Russian formation that returned to Port Arthur did so in some disarray and quite clearly its morale, previously at best brittle, after the obvious failure to reach Vladivostok and an enforced return to a besieged base, must have been about as high as a rattlesnake s tool. But there is one matter relating to the return of the force to Port Arthur that simply is never afforded any real historical consideration.
In the aftermath of the return to Port Arthur the Russian ships were stripped of their guns in order to strengthen the shore batteries, and that fact alone was evidence of defeat and the lack of any real future role. In effect, after this battle the Japanese were left with undisputed control of Far Eastern waters and the Russians at Port Arthur no longer had even a fleet-in-being: any Russian hope of a reversal of naval fortunes had to be vested in formations sent from Europe. 62 But the fact was that this sortie was staged when the Russian defenses at Port Arthur had exhausted their supply of 10-in./254-mm and 6-in./1-52-mm shells, and the return of the warships to Port Arthur provided remedy of a kind. It could be argued that the assessment that Port Arthur may well have been lost in either August or September had it not been for the shells that the warships brought back to the base is somewhat fanciful, a case of a naval service s argument that sought some form of ex post facto vindication of its actions, but nonetheless this is an assessment, or at least a claim, that needs be noted, as indeed must one related matter. With the transfer of ammunition and with some twenty-five hundred officers and men put ashore after the 10 August sortie, the remaining Russians warships had neither the ammunition nor the crews needed to conduct any future sortie. 63 For the Russian squadron, the action of 10 August and return to Port Arthur most definitely represented the closing of the circle, indeed a rather vicious circle.

This third phase of the war saw another action related to the 10 August engagement. Again its result was modest in terms of losses but important in terms of strategic result. This was the action fought some 50 miles/80 km east of Ulsan on 14 August that marked the end of the Vladivostok formation as an effective naval force. From the start of the war this formation, which consisted of the armored cruisers Gromoboi, Rossiya , and the Rurik , and the first-class cruiser Bogatuir and seventeen torpedo-boats, proved an irritant to the Japanese, and on one occasion more than just an irritant. The Japanese attempt of 23-26 April to confine the Russian squadron to port by the laying of extensive minefields in the approaches involved no fewer than nine cruisers, 64 four destroyers, six torpedo-boats, and some fleet auxiliaries, a not-inconsiderable effort on the part of the Kaigun -and a very irritating one given the fact that dense fog prevented the realization of intent.
The cruiser squadron at Vladivostok was there primarily because there was no room for it at Port Arthur, and with the start of the war it conducted two sweeps with the intention of drawing Japanese warships away from the main theater of operations to the south. Its first sortie began on 9 February, courtesy of its ice-breakers, its intention being to stage an initial demonstration in the Tsugaru Strait 65 and then to conduct a reconnaissance of Gensan in anticipation of the Japanese use of that port. In the approach to the home islands the four Russian cruisers encountered two small Japanese coasters, and somewhat improbably one of these escaped to Fukushima on southern Hokkaido. 66 Thereafter the Russian force set course for Gensan but encountered heavy weather with the result that this second part of the mission was abandoned, the Russian ships re-entering Vladivostok on the afternoon of 14 February. Their second mission (24-28 February) saw the Russian cruisers off Gensan, 24-26 February, but there was no encounter with any Japanese ships, whether naval or mercantile, but the mere presence of a Russian force outside a port that it intended to use caused the Kaigun to detach a cruiser formation from the main force outside Port Arthur for use in the Sea of Japan. This formation, with seven units, conducted a bombardment of Vladivostok on 6 March and staged a demonstration outside the port on the following day: the Russian ships, which escaped damage because they were in one area that was not shelled, declined action. In bitterly cold conditions, the Japanese ships then withdrew and on 16 March rejoined the main force. 67
It was not until 8 April that Russian ships were able to leave Vladivostok without the aid of ice-breakers, and it would appear that the combination of bitter winter conditions, a change of command at Vladivostok, and the Port Arthur priority imposed a halt to proceedings over the next month. But in the aftermath of the loss of the Petropavlovsk the two sides staged simultaneous operations, somewhat improbably the Japanese attempting to mine Vladivostok from Gensan while the Russians (with three cruisers and two torpedo-boats) mounted a reconnaissance of Gensan from Vladivostok that was to be followed by a bombardment of Hakodate. 68 Dense fog in the Sea of Japan ensured that the two formations passed with a matter of a few miles of one another on 23-24 April but without contact, the Russians from radio surveillance knowing that the Japanese were close. On 24 and 25 April the fog frustrated Japanese mining efforts with the result that the mission was abandoned, the Japanese ships reaching Gensan shortly after noon on 26 April. There they found that in their absence, on the previous day, the Russian formation had encountered a small merchantman, the 600-ton Goyo Maru , and had sunk it with a single torpedo; later that same afternoon the Russian force had also encountered the 219-ton Haginoura Maru , complete with fish and vegetables, and this ship had been scuttled after her crew had been removed. But more seriously, the Russian formation had moved back northward along the coast, and on 25 April, in response to what proved to be a false report that Russian troops were advancing along the coast road in the direction of Pukcheng, 69 the local command at Gensan had dispatched about three hundred troops to Iwon, on Pallada Bay, 70 in a single 3,853-ton transport Kinshu Maru that was afforded escort by torpedo-boats. Once ashore in early afternoon the Japanese found that there was no enemy force in the area but, with troops re-embarked, weather conditions precluded the transport being escorted to Gensan.
Shortly before midnight the Kinshu Maru encountered the Russian formation, which she mistakenly believed was Japanese. Those on board the Kinshu Maru were given an hour to leave the ship, but it seems that most of the military personnel in the transport refused, and when the Russian ships sought to administer the coup de gr ce the crew returned fire with their personal weapons before the ship was sunk by a torpedo from the Rossiya . Some Japanese managed to reach the coast in lifeboats. Very few allowed themselves to be rescued and taken prisoner, but from these the Russians learned that a Japanese cruiser formation had indeed been at Gensan and had sailed. Armed with such information the proposed bombardment of Hakodate was abandoned in favor of a return to base, the Russian formations making first for Cape Kruilov 71 in order to ascertain if the Japanese force was still off Vladivostok. In fact the latter had returned to Gensan and, obviously in some state of irritation that was exacerbated when the fate of the Kinshu Maru was discerned, the force sailed on 27 April and on the following day laid minefields off the islands and in the channel in the approaches to Vladivostok. 72
With the Japanese offensives into northern Korea, southern Manchuria, and the Liaotung peninsula thereafter taking center stage, the Vladivostok force and the Sea of Japan were relegated to secondary, indeed tertiary, status over the next weeks, the first Russian move after the return to Vladivostok being delayed until 15 May when the Bogatuir sailed for Possiet Bay. In thick fog, the cruiser ran aground off Bruce Point, where it remained until 16 June when it was pulled clear and returned to base; it played no further part in the war. A German collier was mined off Skryplev on 10 June but was nonetheless able to reach Vladivostok, but two days later the Russian squadron, under a new commander, sailed on what was intended as its most ambitious foray to date: it was to patrol in the area of Quelpart and Ross Islands, off southwest Korea, the intention being to return to Vladivostok but with the option of proceeding to Port Arthur. Dawn of 15 June found the Russian formation between the Shimonoseki Strait, Iki Island, 73 and Tsushima, and very quickly it encountered four Japanese ships. One was able to escape and another, the 6,222-ton requisitioned transport Sado Maru , somehow managed to survive two torpedoes launched by the Rurik ; it was towed to Moji by another auxiliary on 17 June. 74
The other two Japanese ships were both sunk. The first was the 3,229-ton auxiliary transport Izumi Maru , which was being used as an unmarked hospital ship: she refused to stop when ordered and even when shots were fired across her bows. Only when hit by Russian fire did the ship stop and the personnel go over the side; after saving about a hundred Japanese, the Gromoboi sank the transport. The second was the 6,172-ton transport Hitachi Maru , which, with many hundreds of troops and supplies, likewise refused to heave-to and was sunk, with heavy loss of life, again by the Gromoboi . 75
It is a rather curious fact that in these exchanges with four Japanese ships one Russian cruiser managed to sink two Japanese transports and two Russian cruisers failed to sink a single ship between them, but equally curious is the fact that a Japanese cruiser, the Tsushima , was in company but certainly did not attempt to place itself between the Russian ships and their intended prey. It nonetheless did report the Russian presence, and within a matter of a couple of hours a cruiser squadron sailed from its base on Tsushima, but in an alternating combination of mist and heavy rain it failed to make contact with the Russian formation, which, anticipating that any Japanese formation coming on the scene would set a direct course for Vladivostok, chose to hug the coast of Honshu. On 16 and 17 June a prize, the British steamer Allanton , was captured and a number of small Japanese craft were boarded and three, all sailing ships, were seized and sunk; at the same time the prisoners in the Russian ships were transferred to a Japanese sailing ship. After reaching Cape Povorotnii during the afternoon of 19 June, the Russian ships reached Vladivostok the following morning.
This third Russian sortie had its ripples of sorrow impact within the Japanese high command, and there were various public demonstrations demanding proper protection of transports, and indeed if the episode proved anything then it was that valuable transports had to be afforded proper escort. Clearly the invulnerability of Japanese shipping to date had precluded the provision of escorts, and the naval ministry defended itself, in a manner that foreshadowed its British counterpart in 1917, on the basis that there were too many sailings for the institution of convoy. But the point was that the Russians were encouraged by their modest success and undertook another operation, again primarily directed to the Strait of Korea and shipping but with the addition of a search of Gensan, on 28 June. At Gensan two tiny Japanese ships were sunk but one Russian torpedo-boat (No. 204) ran aground and, despite being re-floated, had to be abandoned. While the transport Lena and other torpedo-boats returned to Vladivostok, the three Russian cruisers proceeded to the strait, but with timing gone astray their appearing off Okinoshima around noon on 1 July was noted and they were intercepted by a much superior cruiser force in late afternoon off Iki. 76 Very fortunately for the Russians, initial sighting left the Japanese formations trailing by some 12 miles/19 km, and with night, and after the failure of an attack by a separate torpedo-boat formation, there was safety: another British merchantman was taken as a prize the next morning. The Russian ships reached Vladivostok shortly before dusk on 3 July. They were to sail again two weeks later but this time their area of operations was to be very different.
The fifth and what was to be the squadron s penultimate sortie was to last fifteen days and was to take the formation through the Tsugaru Strait to the eastern coast of Honshu, the original Russian intention being to get among the shipping between Yokohama and the Inland Sea. The sortie began on the afternoon of 17 July and the Russian cruisers negotiated the Tsugaru narrows in the early hours of the 20th, sinking one small steamship and two schooners that day. On the 22nd and in a position about 100 miles/160 km north of Yokohama, the Russian cruisers took a German merchantman as prize and then captured and sank a British merchantman, ostensibly because it had insufficient coal to reach a Russian port. Two days later the Russian cruisers sank two small schooners and, in a position 75 miles/120 km southwest of Yokohama, captured a German coaster that was sunk the next day; also on 25 July the Russian ships captured a British merchantman (that was sent to Korsakov) outside Tokyo Bay. The following day the Russian ships turned back to the north, passing through fog and rain and then (on 30 July) the Tsugaru Strait before arriving at Vladivostok on the afternoon of 1 August.
The total of three steamers of 6,238 tons and four schooners sunk and two steamers of 11,186 tons taken as prizes represented by far the most successful return of the sorties conducted by a formation that was to come to grief on its next endeavor. In fact the degree of Russian success in this one sortie was far greater than mere numbers and tonnage sunk or taken: Japanese overseas trade was temporarily paralyzed with insurance rates reaching unprecedented levels: foreign shipping companies suspended trade with Japan and the American West Coast ports were all but closed to Japanese trade. 77 But given Japanese superiority of numbers and the relatively restricted area of operations, that the Vladivostok cruisers would be caught at some stage has all the elements of inevitability, and certainly elements of good fortune had attended their April and May sorties. But the defeat at the action off Ulsan on 14 August was in one respect self-induced by the Russian failure to coordinate operations between the Port Arthur and Vladivostok formations that left both exposed to successive defeat. The naval command at Vladivostok was not aware that the Port Arthur force was to attempt to break out until the consulate at Chi-fu, courtesy of the Ryeshitelni , 78 signalled this information on 11 August, that is, after the defeat in the Yellow Sea and the return of the main force to Port Arthur. Obviously this latter information was not at hand when the Russian authorities at Chi-fu passed to Vladivostok notice of intent. The three remaining cruisers of the Vladivostok formation then sailed on the morning of 12 August with a view to effecting a rendezvous with the Port Arthur force.
Unfortunately a torpedo-boat sent to recall the squadron once the information regarding the fate of Vitgeft s force was received in Vladivostok failed to find the cruisers. Instead, soon after dawn on 14 August, the three Russian cruisers encountered four Japanese armored cruisers, the Azuma, Iwate, Izumo , and the Tokiwa from the 2nd Division off Ulsan, under circumstances very different from the previous encounter. The Japanese ships were faster and had virtually all the hours of daylight available to them, which initially gave them a very marked advantage. Fire was opened at approximately 9,000 yards/8,260 m, and very quickly the 2:1 Japanese advantage in firepower became evident. All three Russian ships were hit and set on fire, the Rurik being reduced to perilous state. The other two Russian cruisers attempted to stand by the Rurik , the Rossiya having most of its guns silenced in the process, until the decision was taken to abandon the stricken Rurik and to attempt to return to port. The Japanese cruisers, leaving the Rurik to be sunk by the Naniwa and Takachiho , which now came upon the scene, then set up a stern chase of the fleeing Gromoboi and Rossiya , but at the very time they appeared to be on the point of overhauling the Russian cruisers, the Japanese cruisers broke off the action and turned back, thus allowing the two Russian warships to escape. 79 If the action had been continued for even a very brief period it seems likely that the Rossiya , and perhaps both Russian cruisers, would have suffered the same fate as the Rurik . As it was the two Russian cruisers were obliged to stop in order to patch a total of seventeen holes near their waterlines and it was not until 16 August that they reached Vladivostok with their total of 459 killed and wounded. It was to be two months before the two surviving Russian cruisers were ready to put to sea, and on the first occasion when the Gromoboi did so it ran aground and was severely damaged. Like the Bogatuir , it was rendered hors de combat , and with just one cruiser on station the Vladivostok formation, in effect, was no more. The Rurik lost 170 of her crew, but Japanese ships, with three more cruisers and torpedo-boats coming on the scene, were able to rescue 625 of its officers and men. 80
The two actions, in the Yellow Sea on 10 August and off Ulsan four days later, marked the end of the third phase of the naval war and indeed they marked the end of the naval war in terms of Russian formations and units within the theater of operations. Russian naval strength, by this time, was spent, and the next phase of the war was to witness the destruction of the Pacific Squadron at Port Arthur as a direct result of the Japanese capture of 203-Metre Hill, a little more than a mile northwest of the New Town, from which observation of the harbor, inner basin, and dockyards and directed fire against warships were possible. The two months after the capture of the Wolf Hills witnessed three major Japanese offensive efforts defeated with major losses, 81 after which the Japanese concentrated upon bringing into position heavy siege artillery and then preparing for a renewed assault after protracted bombardment of Russian defensive positions. Nonetheless the fourth assault, 30 October-1 November, was defeated, but the fifth assault, which opened on 26 November in a blizzard, was to prove successful: on 5 December 203-Metre Hill was taken. 82 In these weeks the war at sea saw the Japanese loss of just three units, two of which were ex-Chinese warships that had been captured in 1895. All three were lost to mines, the destroyer Hayatori off Port Arthur while on blockade duty on 3 September, the gunboat Heien off Iron Island on 18 September, and the cruiser Sai-en off Pigeon Bay on 30 November. For their part the Russians lost the gunboat Zabiyaka at Port Arthur on 25 October and the destroyer Stroini to a mine outside the base on 13 November. 83
The destruction of the major units of the Pacific Squadron then followed, that of the battleship Poltava the same day that 203-Metre Hill was captured; on the next day the battleship Retvizan was sunk. On the 7th the battleship Pobyeda and first-class cruiser Pallada were sunk and the battleship Peresvyet , after sustaining damage from Japanese fire that could not be righted, was scuttled. On 8 December the gunboat Gilyak was sunk; the next day the armored cruiser Bayan was sunk. In these five days directed Rikugun fire achieved what the Kaigun had manifestly failed to do over ten months, and thereafter the continuing Japanese artillery bombardment was directed against dockyard installations and buildings with Russian warships continuing to be sunk, the torpedo-gunboat Vsadnik on the 15th, the Amur on the 18th, and the Bobr on the 26th. By mid-December, therefore, just the battleship Sevastopol remained of major units, and it was to escape destruction by virtue of the fact that it, and the gunboat Otvazhni , moved into the roadstead and therefore were not in line of sight of Japanese observation posts on 203-Metre Hill; during hours of daylight these two ships were joined by various surviving torpedo-boats and auxiliaries that moved back into the harbor, and away from Japanese torpedoes, during the night. Outside the harbor the Sevastopol and Otvazhni were subjected to repeated attack by torpedo-boats after 13-14 December, the initial Japanese attacks faring less than well. In the course of a reconnaissance mission on the afternoon of 13 December in which it was employed in providing cover and support for the torpedo-boats, the cruiser Takasago was mined and sunk; Torpedo Boat 53 suffered a similar fate the next day and Torpedo Boat 42 was sunk by the destroyer Serditi on 15 December. 84 But even with boom defenses in place and the Russian warships offering resistance to successive attacks, the Sevastopol could not escape torpedo hits, and it was grounded in order to prevent it sinking before, in the early hours of 2 January and in anticipation of the surrender of the Port Arthur garrison that same morning, it was taken into deep water and scuttled, as was the Otvazhni ; the gunboat Razboinik and the destroyers Bditelni and Razyashchi were scuttled inside the Port Arthur base on this same day: the torpedo-gunboat Gaidamak and destroyer Silni had been scuttled inside the base on the previous day. The auxiliary yacht Arinitoi , the destroyers Serditi, Skori , and the Statni , and the torpedo-boat Parastsni sailed from Port Arthur in these last hours of its Russian life in an attempt to reach Tsingtao, but pursued by Japanese units they put into Chi-fu, where they were interned. It was a comment on Russian losses over an extended period, and also on the escape of various units in August 1904 and now in January 1905, that at the end of the siege there were just four destroyers remaining in the base-the battleships and cruisers obviously escorted destroyers and not the other way around-and that all were scuttled in order to preclude capture, though in the event one, the Silni , was returned to service, Japanese service, as the Fumisuki . But in terms of Port Arthur and Russian naval formations that had been in the Far East at the outbreak of war, the Russo-Japanese War was over.

MAP 5.1. The Russo-Japanese War: The initial Japanese operations.

MAP 5.2. The Russo-Japanese War: Second- and third-phase Japanese operations.

APPENDIX 5.1.
SUBMARINES AND THE RUSSO-JAPANESE WAR
T HE RUSSO-JAPANESE WAR saw the employment of submarines but to no effect in terms of the sinking or damaging of warships, auxiliaries, and merchantmen or, indeed, in terms of circumscribing in any way the employment of naval formations.
Apparently the first Russian boats-pedal-driven submersibles with torpedo tubes rather than submarines-were delivered by the transport Dagmar in late 1900; one of these is supposed to have been transported to Port Arthur. In 1903 the Tsarevich brought a French-built submarine with it from Europe to Port Arthur. These craft apparently did not see action.
According to Liudmila Spiridonova, et al., The Navy of the Russian Empire , p. 261, there were thirteen submarines at Vladivostok at the end of summer 1905 and one, the Som , the ex- Fulton , purchased from the Holland firm in the United States, notched up more than 1,000 miles/1,600 km of patrolling in six months in a series of missions that reached to distances of 120 miles/190 km from Vladivostok.
According to various entries on the website, one of the other submarines, the Delfin , entered service in February 1905 but was destroyed in May 1905 when its fuel exploded and another submarine, name not given, supposedly attempted to attack a Japanese destroyer in Amur Bay at some time, not known. The problem here, clearly, is lack of reliable written records.
The Japanese ordered five Holland boats on 14 June 1904 and these, built at Fore River yard at Quincy, Massachusetts, were completed 5 October and delivered, in sections, at Yokohama on 12 December 1904; after assembly, the first was ready for employment on 1 August 1905. The Japanese also built two more Hollands, one assumes under license, but these were not readied until spring 1906. 1
NOTE
1 . Sources: Evans and Peattie, Kaigun , p. 177, while Jentschura, Warships of the Imperial Japanese Navy , p. 160, states that the American order was placed on 14 June 1904 and all five were delivered, in sections, at Yokohama on 12 December 1904; after assembly, the first was ready for employment on 1 August 1905. The details differ slightly from Warner, The Tide at Sunrise , pp. 242-243.

APPENDIX 5.