0% found this document useful (0 votes)
276 views297 pages

The Future of Sea Power 9780367711160 - 250319 - 141529

The document is a reprint of Eric Grove's book 'The Future of Sea Power', first published in 1990, which explores the complexities of sea power in the modern world, including its military and economic implications. It discusses the evolution of naval warfare, the changing international context, and the strategic importance of navies in both peace and conflict. The book aims to provide a comprehensive analysis of maritime strategy and the future of naval capabilities in light of contemporary geopolitical changes.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
276 views297 pages

The Future of Sea Power 9780367711160 - 250319 - 141529

The document is a reprint of Eric Grove's book 'The Future of Sea Power', first published in 1990, which explores the complexities of sea power in the modern world, including its military and economic implications. It discusses the evolution of naval warfare, the changing international context, and the strategic importance of navies in both peace and conflict. The book aims to provide a comprehensive analysis of maritime strategy and the future of naval capabilities in light of contemporary geopolitical changes.
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 297

ROUTLEDGE LIBRARY EDITIONS:

INTERNATIONAL SECURITY STUDIES

Volume 9

THE FUTURE OF SEA POWER


THE FUTURE OF SEA POWER

ERIC GROVE
First published in 1990 by Routledge
This edition first published in 2021
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
and by Routledge
52 Vanderbilt Avenue, New York, NY 10017
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 1990 Eric Grove
All rights reserved. No part of this book may be reprinted or reproduced or utilised in any form or by any
electronic, mechanical, or other means, now known or hereafter invented, including photocopying and
recording, or in any information storage or retrieval system, without permission in writing from the
publishers.
Trademark notice: Product or corporate names may be trademarks or registered trademarks, and are used
only for identification and explanation without intent to infringe.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
ISBN: 978-0-367-68499-0 (Set)
ISBN: 978-1-00-316169-1 (Set) (ebk)
ISBN: 978-0-367-71116-0 (Volume 9) (hbk)
ISBN: 978-0-367-71118-4 (Volume 9) (pbk)
ISBN: 978-1-00-314938-5 (Volume 9) (ebk)
Publisher’s Note
The publisher has gone to great lengths to ensure the quality of this reprint but points out that some
imperfections in the original copies may be apparent.
Disclaimer
The publisher has made every effort to trace copyright holders and would welcome correspondence from
those they have been unable to trace.
The Future of Sea Power

Eric Grove
First published in 1990
by Routledge
11 New Fetter Lane, London EC4P 4EE
© 1990 Eric Grove
Printed in Great Britain by
Richard Clay Ltd, Bungay, Suffolk

All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by any
electronic, mechanical, or other means, now known or hereafter invented, including photocopying and
recording, or in any information storage or retrieval system, without permission in writing from the
publishers.
British Library Cataloguing in Publication Data
Grove, Eric
The future of sea power.
1. Sea – power
I. Title
359
ISBN 0-415-00482-9
To Elizabeth
Contents

List of tables and figures


Foreword by Admiral Sir Julian Oswald
Acknowledgements

Part I: The state of the art


1 Sea power in the modern world
2 Maritime strategy

Part II: The uses of the sea


3 Economic uses of the sea
4 The military medium

Part III: The changing shape of naval war


5 Weapon systems
6 Platforms

Part IV: The evolving environment


7 The international and legal context
8 Economic stimuli and constraints

Part V: Navies in peace and war


9 Navies in peace
10 Navies in war

Part VI: Conclusion


11 The future of sea power - three theoretical frameworks

Notes and references


Acronyms and abbreviations
Index

Tables and figures

Table
6.1 Most promising future seaborne platform options

Figures
3.1 Main oil movements by sea, 1987
11.1 The ‘use of the sea’ triangle I
11.2 The ‘use of the sea’ triangle II

Foreword
by Admiral Sir Julian Oswald, GCB, ADC, Chief of
Naval Staff and First Sea Lord

It is, I suppose, axiomatic that in relation to Sea Power, its place in the world and
its actual or potential effect on the lives of people, both Maritime Strategy and
the Structure of Navies will be constantly under review. Thus a book which
combines careful analysis of the structure and components of sea power and a
critical, questioning look at the strategy which governs navies’ possible
employments will always be important. It is particularly to be welcomed on two
further scores. First it comes from the pen of Eric Grove, an experienced
commentator, distinguished strategist, and most painstaking author who revels in
this particular subject, and whose background is ideal. Additionally, it comes to
us at a time when the whole structure of inter-state, inter-allied, and inter-bloc
politico-military relationships is undergoing traumatic change. It is therefore
essential that we should look ahead and ask ourselves – ‘What might we expect
of navies in the 21st Century?’ For the United Kingdom, long term exponent of
maritime power and with a unique geographical position and sea heritage, the
question is of the utmost importance. Ships on the drawing board today may well
be in the Fleet in 2030. Will they be adaptable to the requirements of that age?
What might those requirements be? Other countries too must answer difficult
questions if they are to invest sensibly in sea power.
Arms control initiatives now showering down upon us may well start by
altering the balance of forces (and both the likelihood and nature of conflict) on
land – but they will soon introduce seaborne considerations as well. Some
measures may seek to reduce weapon systems at sea, others may encourage the
deployment of weapons on maritime platforms to compensate for those removed
from land. In any event it seems likely that the sea will remain host to all or part
of the strategic deterrents of both the superpowers, France, the United Kingdom
and perhaps others. This of itself will greatly influence maritime operations.
Asymmetric cuts in conventional forces in Europe may reduce the risk of
attack but, paradoxically, increase the importance of reinforcement across the
Atlantic and Channel. Retaining necessary and credible naval forces to effect
this will necessitate not only presenting a convincing and coherent Alliance
Maritime Strategy and rationale to our political leaders, but also insisting, in
discussion with the Soviet bloc, that the entirely different nature of NATO and
the Warsaw Pact – the one on maritime Alliance, the other a great continental
power block – be recognized.
Not to be forgotten is the impact of navies outside the NATO/ Warsaw Pact
area. Can we judge that as a better relationship and understanding grows
between the superpowers and their allies, they will turn their efforts more to the
prevention of war and the resolution of conflict in the rest of the world? If they
do they will find very many navies, some far from small and many well-
equipped and capable, out there. What interactions can be expect, and how will
these perceptions effect naval development around the world?
The questions are endless. Some are unanswerable. To others we can give a
qualified reply. But if we are to get even that right we must be absolutely clear
about which principles of Maritime Structure and Strategy will endure, and
which may not. To guide us is the fascinating and demanding task which Mr
Grove has set himself.

Julian Oswald

Acknowledgements

Three good friends played crucial roles in this book. Philip Towle of Queen’s
College, Cambridge first suggested I should write it. Evan Davies of the
Britannia Royal Naval College, Dartmouth, who was to have been co-author,
was of great assistance in defining its basic structure. Geoffrey Till of the Royal
Naval College, Greenwich provided useful suggestions for improving the final
draft. Michael Ranken and R. S. Woolrych of the British Maritime Charitable
Foundation were more than helpful with information and advice on merchant
shipping matters. British Petroleum kindly allowed reproduction of their chart of
oil movements by sea and D. W. Waters and Ken Booth both very graciously
assented to extended quotations of their work. Norman Friedman, a name to be
found often in the footnotes, has always been a constant source of insight both in
his written works and our private discussions. Peter Sowden of Croom Helm,
now Routledge, tolerated with amazingly good grace the very long, but
unavoidable, delays in final production of the manuscript. My mother and father
typed, re-typed, and printed out the various drafts in their usual efficient manner.
My wife, to whom this book is dedicated, put up with the burden of being an
obsessive author’s partner. To all the above very many thanks. No one but the
author is responsible for the facts or judgements that follow: they do not reflect
the policy of any organization or institution.
Part I

The state of the art


Chapter 1

Sea power in the modern world

‘Sea Power’ means different things to different people. It can be an almost


mystical concept, a magic formula to be mouthed in awestruck tones to scare
away evil spirits such as defence ministers with non-naval priorities or air force
officers with alternative means on offer of providing a state’s military power on
or across the oceans. To others, following in the well-trodden footsteps of Alfred
Thayer Mahan, ‘sea power’ represents a more coherent but equally universal
concept, an interlocking system of forms of sea use, military and civil, which has
a unique contribution to making powers great or greater still. This book will take
a narrower and more measured approach. In series with its two earlier
companions on ‘air power’ and ‘land warfare’ it will treat ‘sea power’ as a
military concept, that form of military power that is deployed at or from the sea.
As Mahan himself put it, ‘the history of sea power, while embracing in its broad
sweep all that tends to make a people great upon the sea or by the sea, is largely
a military history’;1 equally, the future emphasized here will be the military
future of sea power, although the relationship of naval power with the various
forms of sea use must also be considered.
The whole subject has become rather more complex than it once seemed in
the age of Mahan. For a number of economic reasons, which are dealt with more
fully in Chapter 3, the states that possess the world’s most powerful navies do
not possess the world’s largest merchant fleets. Despite many state sponsored
attempts to prove Mahan correct in his famous assertion that, ‘The necessity of a
navy in the restricted sense of the word, springs, therefore, from the existence of
a peaceful shipping and disappears with it’,2 the world’s most powerful naval
power was outranked at the end of 1986 in merchant shipping by Liberia,
Panama, Japan, and Greece.3 Liberia, with 97.8 million deadweight tons of ships
and 15.6 per cent of the world’s total tonnage, deployed a military fleet of six
patrol craft, none bigger than 50 tons, and a land-based helicopter. The largest
weapon available was an 81 mm mortar. Panama, with 69 million tons of
merchantmen, 11 per cent of the world’s tonnage, has eight small patrol boats
and eight transports; one of the latter is a former shrimp boat.4 The two major
mercantile marines of the world seem to prosper with nothing more than a
coastguard back home. Only when we come to the world’s number three, Japan,
with almost 56 million deadweight tons of merchant ships and almost 9 per cent
of world tonnage, do we find the possessor of a significant military navy, one
that ‘has tended to look more and more like an ocean going navy’.5
Nevertheless, its operations are still severely circumscribed and it is still
significantly smaller and less capable than the navies of Britain and France,
countries which no longer make it into the merchant shipping top ten (unless
Britain’s dependencies are included). Something has clearly gone wrong with the
Mahanian paradigm.
One factor has been the growing internationalization of marine activities.
Mahan drew his lessons from the great days of classical mercantilism and the
clashes of autarkic empires competing for slices of a finite ‘cake’ of maritime
trade. He thought this might well recur in the twentieth century. In a way it did,
but the defeat of the German and Japanese attempts to carve out continental and
maritime variations on this theme led to a new world of economic neo-liberalism
that still, despite numerous vicissitudes, survives. In this environment the flag a
ship carries is very often not a reflection of the identity of the owner. If US-
owned ‘flag of convenience’ vessels are taken into account, the tonnage of the
American mercantile marine more than doubles to over 65 million tons; but that
of Japan increases by an almost similar margin, to over 90 million. Even taking
beneficial ownership into account, therefore, the United States remains only
third in merchant ship ownership, behind Japan and Greece.6
Why then is she the number one naval power, closely followed by the USSR,
with a mere 23.4 million tons of merchant ships. The fundamental fact of
twentieth-century sea power is that a country’s naval capability is a direct
reflection of its sheer economic power in all senses and that that power
inevitably reflects its control and exploitation of large land masses. Professor
Paul Kennedy has ably demonstrated that Halford Mackinder not Alfred Mahan
proved the surer prophet of the new century.7 In 1902, in his Britain and the
British Seas, Mackinder argued that Britain’s sea power rested on fragile
foundations in an era of ‘vast Powers, broad based upon the resources of half
continents . . . mere insularity gives no indefeasible title to marine sovereignty.’8
In 1904, before the Royal Geographical Society, he went further by suggesting
that ‘When historians of the remote future come to look back . . ., it may well be
that they will describe the last 400 years as the Columbian epoch, and will say
that it ended soon after the year 1900.’ Mackinder argued that with the world
now virtually fully explored and divided up, ‘we are for the first time in a
position to attempt, with some degree of completeness, a correlation between the
larger geographical and the larger historical generalisations.’9 In other words, as
Kennedy has rather more clearly put it, ‘efficiency and internal development
would replace expansionism as the main aim of modern states and . . . size and
numbers would be more accurately reflected in the sphere of international
developments.’10 The railway already meant that for the first time in human
history, transport by land was as (if not more) effective. And in some ways it was
as economical as transport by water. The beginnings of air (and road motor)
transport promised to confirm this secular change. In one of the most perceptive
comments by a member of Mac-kinder’s audience, Leo Amery, while insisting
that a great power must possess sea power as well as land power, accepted that
‘sea power alone, if it is not based on great industry and has a great population
behind it, is too weak for offence to really maintain itself in the world struggle’.
In future, once the transport revolution had taken place ‘the successful powers
will be those who have the greatest industrial base. It will not matter whether
they are in the centre of a continent or on an island; those people who have the
industrial power and the power of invention and of science will be able to defeat
all others.’11
And so they have. The world’s superpowers, the USA and the USSR, stand
supreme upon the world’s oceans, their navies confronting each other around the
globe as global expressions of the fundamental politico-strategic split that
divides the planet. Both sides’ navies are more expressions of a power that has
its roots elsewhere than, as the Royal Navy was at the height of its power, the
guardians of the foundations of national economic power itself, i.e. seaborne
trade. Nothing better symbolizes the foundations of sea power in the modern
world than the nature of much of both sides’ current naval investment. The
superpowers’ ballistic missile firing submarines prowl their patrol areas ready at
a moment’s notice to destroy the political and economic heartland of the other
side. They use the hiding place provided by the sea to make themselves the least
vulnerable and most stabilizing form of long-range nuclear striking power: they
might even be regarded as a form of air power. For the USSR, perhaps the one
form of sea use that is fundamental to its national security is the ability to use the
seas to deploy this form of strategic reserve as a final sanction against all-out
Western nuclear attack. Certainly the USA seems to think so in its recent
articulations of how to exert potentially decisive maritime pressure on the USSR
in a conventional war.
Modern technology, as explained more fully in Chapter 5, has revolutionized
the ability of sea-based forces to ‘thrust power ashore’, especially in the form of
bombardment. ‘Sea power’ can thus directly affect the land battle in immediate
and short-term ways that it could not do, except on the coastline itself, until the
middle decades of this century. Before that ‘sea power’ tended to be an indirect
factor in the campaign ashore, maintaining sea-dependent states in the war so
that they could exert direct military pressure on the enemy in various ways, be it
subsidies to allies, giving battle on land or, in later years, bombing from the air.
Even at the height of the Columbian era in the eighteenth and early nineteenth
centuries the last and greatest maritime empire, the British, had to maintain
armies in the field against France if she was to ensure success: the one war
where she did not do this was the one she lost – the War of American
Independence.12 Sea power prevented the wars against Napoleon being lost but
Mahan’s famous ‘storm tossed ships upon which the Grand Army never
looked’13 could do little in themselves to win them.
When that greatest of all maritime strategists, Sir Julian Corbett, wrote up in
1911 the lectures of history and strategy with which he had been intriguing (and
annoying) the officers of the Naval War College for the previous five years, he
stated the situation plainly: ‘Since men live upon the land and not upon the sea,
great issues between nations at war have always been decided – except in the
rarest cases – by what your army can do against your territory and national life
or else by the fear of what the fleet makes it possible for your army to do.’14
The events of 1914-18 bore Corbett out. Lord Esher set out the strategic
situation only ten days into the war. ‘No one,’ he wrote, ‘. . . believes that the
naval forces of the allies can force a decision, or do more than render the lives of
the Allied peoples secure and that of the enemy impossible. We shall remain
immune from attack. The French will be free to receive supplies from oversea.
The Germans will become a beleaguered garrison. These achievements are those
of modern sea command.’15 Esher doubted if a major battle would take place
and when it did, off Jutland in 1916, the British fleet commander wisely did not
risk losing the fruits of Britain’s current strategic situation for the somewhat
limited extra gains that would have been obtained from destroying the German
battlefleet. Britain was thus able to continue to draw on the resources of the
world to deploy the Imperial Army that, in its finest hour, defeated the full might
of the German Army in the summer and autumn of 1918. When, following these
land defeats, the Germans gave up most of their battlefleet to the British without
a fight the Admiralty was moved to send a congratulatory signal to the Grand
Fleet ‘on a triumph to which history knows no parallel’ which would ‘remain for
all time the example of the wonderful silence and sureness with which sea power
attains its ends’.16 A soldier from Amiens or the Hindenburg line might have
disputed the silence if not the sureness.17
The Admiralty in its November message spoke of the pressure exerted by the
Grand Fleet itself on Germany, and much has been made of the supposed effects
of the blockade on the German war economy and the morale of her people.18
Sadly the facts do not bear this out. Numerous historians have pointed out the
limitations of blockade against a power like Germany, which for most of the war
controlled most of the continent.19 German inefficiency in food production and
distribution was more responsible for shortages than the blockade itself.20 More
important still, the fact that her resources were tied up in a debilitating land war
was a vital factor: ‘civilian sufferings from the blockade would have been far
fewer had not the great military campaigns swallowed up such astronomical
amounts of foodstuffs, industrial products and especially the men needed to farm
the land.’21 Of course, if the British could not win the war at sea they could
certainly lose it, and the German submarine offensive in 1917 came close to
achieving that success.22
In the second great twentieth-century war of 1939-45 the same pattern was
repeated. Britain was, with some considerable effort both in the air and at sea,
able to survive and act as a forward base for the projection of US power into
Europe, but it was the latter and even more so the great land battles on the
Eastern Front that defeated the might of Hitler’s Reich. What was survival to the
British was an exercise in seaborne power projection to the Americans. In the
Pacific the United States demonstrated even more clearly the new shape of sea
power, as a three dimensional fleet of unparalleled power and reach covered a
succession of amphibious landings that cut the Japanese Empire in two and
gained the bases for an aerial assault that eventually, with nuclear firepower,
proved decisive. In addition, and almost as an afterthought, given Japan’s
extraordinary power of ignoring strategic realities and coming close to
committing national suicide, in the most successful war against seaborne trade in
history, Allied submarines sank over 1150 ships of almost 5 million tons.
Aircraft sank almost 750 more of almost 2.5 million tons.23 By the end of the
War intensive Allied mining had bottled up the remaining ships and a maritime
empire’s arteries had ceased to function. Japan had been effectively strangled.
How long she could have continued fighting without the coups de grace of
Hiroshima and Nagasaki is an interesting point.
The major lessons of the two world wars of the twentieth century seem to be
that sea power, while a necessary part of a great power’s armoury if it wishes to
project military power around the globe, is far from sufficient on its own to
assure victory in war. Indeed, over-dependence on the sea far from being an asset
was a positive vulnerability only limited by extraordinary national determination
and the suicidal propensities of national culture. If extra pressure is required in
these circumstances perhaps only nuclear weapons are capable of providing it.
The year 1945, therefore, in the end turned out to be a victory for air power,
albeit air power that could only operate through seaborne power projection. The
advent of nuclear weapons confirmed the challenge of the salience of sea power.
As the kiloton range nuclear bombs became megaton range thermonuclear
devices in the 1950s, serious doubts arose in the minds of many as to whether a
future war would last long enough for the sea to be relevant at all. Sea power
was slow acting and the next war might be over in hours. For a time navies
argued that ‘broken backed’ war might still be possible and necessary after the
initial nuclear strikes as both sides licked their dreadful wounds and prepared for
a second more intensive phase, but this carried less and less conviction and
proved a weak foundation for the continued survival of naval power. Better to
climb onto the nuclear bandwagon and substitute sea-based systems for the
forward air bases. This was successfully achieved in the 1950s and 1960s by the
navies of both superpowers.24 The technological emphasis varied but the
strategic logic remained the same: sea power as a form of air power against
Mackinder’s heartland. It was not surprising that the relationship of superpower
navies with sister services, be they the USAF or the Soviet Strategic Rocket
Forces, ranged from uneasy to downright bad.
Naval officers, however, were the first to see a fundamental truth of the
nuclear age. Nuclear war is so clearly self-destructive that it can only be credibly
invoked as a threat when the most vital national interests are threatened, if
indeed then. Perhaps nuclear weapons are only fully credible as necessary
deterrents to nuclear attack. If states wish, or need, to use force to defend their
interests and politico-economic health they have to have a range of capabilities
much more discriminating in their effects than the instruments of nuclear
Armageddon. This has a relevance both in superpower confrontation and in less
equal conflicts. Whatever the context, maritime forces are often useful and
sometimes indispensable for exercises of limited force. This might not ease
relations much with the air force, especially if the latter sees limited war
operations as its major role (cf. the controversy over British carriers in the
1960s), but it does provide one solid rationale for the navies of those states that
aspire to intervene militarily beyond their boundaries.25
‘Seaborne power projection’ has thus become the main role for naval forces in
the nuclear age, but the seas themselves remain crucial factors in the world
economy and hence in the overall security interests of states. Although, as we
shall see in Chapter 8, the level of sea dependence of many countries is arguable
there can be little gainsaying the fact that many states have vital interests
connected with an ability to use the sea in various ways. Moreover, interests in
sea use go far wider than a few especially dependent states. The sea provides the
transport medium for the vast bulk of the world’s international trade, about 80
per cent by volume.26 Over 99.5 per cent of the weight of cargoes carried over
the oceans goes by sea, less than 0.5 per cent by air.27 The free flow of these
supplies is of great interest to the whole world, especially the OECD nations,
whose trade tends to be more maritime than that of the communist nations of the
CMEA. The USA, now rather more reliant than before on foreign sources of
both manufactured goods and raw materials, sees it as being in her interest to
provide protection when required to the West’s seaborne trade, to the extent of
lending its flag to ships under threat as in the Gulf in 1987. This ‘return to the
flag’ and restoration of Mahanian proprieties is one of the most interesting
developments of recent years and may presage a growing practice when danger
threatens. Other countries, especially those like Britain whose dependencies can
offer the national flag on virtual open registry (i.e. flag of convenience) terms
and which can also offer naval protection from the metropolitan power, may also
be in sometimes embarrassing demand by ship owners and operators.
The sea has also increased in importance as a source of useful resources.
Food has always been obtained from the sea but now energy and, to a growing
extent, other minerals come from the ocean and its floor. This has given the
ownership of the coasts and islands of the world a new importance as these now
give sole exploitation rights to huge areas of sea. Although the precise legal
regime for the world’s oceans remains in dispute the enclosure of large tracts of
sea and ocean has already taken place. Countries need the means to assert their
rights and carry out their duties in these newly-defined territorial seas and
Exclusive Economic Zones, and this is perhaps the major reason for the
proliferation of what might be called the institutions of sea power – the world’s
navies.
These factors, and the sheer growth in numbers of sovereign states, have
contributed to making navies increasingly popular. Never have there been more
then there are today. Ninety years ago the first edition of Jane’s Fighting Ships
listed the vessels of a mere twenty-two different navies.28 Ten years later, when
what with Edwardian arrogance were dismissed as ‘unimportant’ navies
(Cambodia, Costa Rica, Persia, Zanzibar, etc.) were added, the number doubled
to forty-four.29 The ninetieth issue of Jane’s listed over 150 nations with navies
and/or coastguards.30 Sea power, as a military activity at least, is obviously a
growth industry.
It is even being thought about more. One of the curious features of the growth
of strategic studies over the last quarter century or so has been the relative
inattention paid to sea power. Only with the publication of the US Navy’s
controversial ‘Maritime Strategy’ have we seen the growth of a strategic debate
worthy of the name.31 What is rather extraordinary about this debate is that there
is so little new in the Maritime Strategy, other than the fact of its publication and
the coherence and forcefulness of its drafting. As various authors have shown32
the perceived imperatives of technology and resource constraints forced Western
navies to emphasize the ‘attack at source’ on Soviet naval assets from the earliest
days of NATO. Forward carrier and submarine operations have been a central
and persistent feature of the naval strategy of the Anglo-American maritime
coalition from the start. That their re-emphasis should catch so many by apparent
surprise confirmed the author’s assessment that there was a need to examine the
future of sea power both in terms of a comprehensive assessment of the present
reality and some understanding of the experience of the past.
This book will attempt to answer two main questions. How is sea power
going to develop in the next few decades? Is it possible to develop a new theory
of sea power for the future or are the old ideas still relevant, despite all the
technological changes of the last century? Before going tentatively into the
future however, like Mahan and Corbett, we must look more systematically and
in some detail at the lessons of experience, so we know from where we begin.
Chapter 2

Maritime strategy

If sea power is to be used purposively then it must be through the application of


maritime strategy. This can be defined as the art of directing maritime assets (i.e.
those that operate on, over, or under the sea) to achieve the required political
objectives. In 1911 Sir Julian Corbett set down ‘some principles of maritime
strategy’ that were based on an intelligent reading of naval history up to that
date.1 Corbett was criticized at the time on two main grounds. First, it was said
that as a civilian he had no locus standi to comment on such arcane arts as
maritime strategy. The founder of Jane’s Fighting Ships dealt with such
criticisms devastatingly: ‘it would be a bad day ... if the principle ever gets
established that unless a man is an actor he is incapable of criticising the actions
of a drama ... the contention would work out that “you cannot tell whether an
egg is good or bad unless you are a hen”.’2
The second criticism was one that is very familiar today: that technological
progress had so fundamentally altered everything that historical parallels were
inapplicable to contemporary naval affairs. If accepted, this line of argument
would make naval affairs something separate from the rest of human activity, all
of which is based to varying extents on experience. Progress in any field, even a
fundamental breakthrough in science and technology, is usually based on an
intelligent and informed reassessment of existing phenomena. If any constructive
thought is to take place it must have an existing frame of reference. As Corbett
himself put it: ‘Having determined the normal we are in a stronger position. Any
proposal can be compared with it and we can proceed to discuss clearly the
weight of the factors which prompt us to depart from the normal.’3
The tendency to argue that ‘everything has changed’ is a very dangerous one.
Even Corbett fell into the trap in his discussion of the defence of trade in Some
Principles. Only six years after the publication of his book, which fell in with the
current fashion in denying the continued relevance of convoy, Britain was almost
defeated because of the steadfast refusal of her naval leaders to introduce a
convoy system for the defence of trade. Knowledge of convoy’s crucial role in
1917-18 (and 1939-45) does not mean necessarily that convoy is the only answer
in all circumstances to the problem of the defence of merchant ships, but it does
mean that we should be very careful indeed about departing from convoy as the
core of any strategy for sea use.
The use of the sea for movement is itself the core of maritime strategy in its
traditional sense. Corbett put it in slightly more general terms: ‘maritime
communications whether for commercial or military purposes’.4 Control of these
communications in terms of being able to achieve one’s own movement and
deny it to an opponent is the essence of ‘command of the sea’ and, as Corbett
flatly stated, ‘the object of naval warfare must always be directly or indirectly
either to secure the command of the sea or to prevent the enemy from securing
it’.5
There is something rather old-fashioned about the term ‘command of the sea’.
Some modern writers have argued that it implies a degree of dominance that
modern technology makes impossible. One of the most notable of these was
Admiral Stansfield Turner, perhaps the leading figure in American naval thought
in the 1970s. When running the US Naval War College in 1974 he popularized
the notion of ‘sea control’ as ‘a deliberate attempt to acknowledge the limitations
on ocean control brought about by the development of the submarine and the
airplane’.6 He went on to justify his new term as follows:

The new term ‘Sea Control’ is intended to connote more realistic control in
limited area and for limited periods of time. It is conceivable today to
temporarily exert air, submarine and surface control in an area while moving
ships into position to project power ashore or to re-supply overseas forces. It
is no longer conceivable, except in the most limited sense, to totally control
the seas for one’s own use or to totally deny them to an enemy.7

Turner’s thinking reflected both the experience of World War II and the
perceived weakness of the US Navy of the post Vietnam period against the
growth of Soviet naval capability of the Gorshkov era. ‘Sea Control’, however,
is a useful concept in a more general sense in that it demonstrates the
fundamental change in the capacity of states to attack shipping that was brought
about by the advent of submarine and aircraft. Nevertheless, it should not lead to
the complete abolition of the older term. Analysts such as Corbett clearly
recognized that ‘the most common situation in a naval war is that neither side
has the command; ... the normal position is not a commanded sea but an
uncommanded sea’.8 Command of the sea is clearly a relative not an absolute
term. Dr Geoffrey Till admirably encapsulates this dimension in his definition,
which probably remains the best:

Being ‘in command of the sea’ simply means that a navy in that happy
position can exert more control over the use of the sea than can any other.
The degree of command varies greatly and is primarily illustrated by the
extent to which it confers the capacity to use the sea for one’s own purposes
and prevent the enemy using it for his.9

In this sense modern naval operations are still about command of the sea.
In order to secure command of the sea it is necessary to neutralize the
enemy’s main naval forces. The classic way of achieving this is to defeat and
destroy them in battle. Mahan and many other naval historians have been
fascinated by naval battles and have tended to overrate their importance. Perhaps
the classic example of such mis-assessment is the most famous naval battle of
all, Trafalgar. Nelson’s great victory of 21 October 1805, when his numerically
inferior fleet smashed the combined battlefleet of France and Spain, has justly
been remembered as a classic of its type, although one that relied on the gross
qualitative inferiority of the French and Spanish that it was Nelson’s genius to
observe and act upon. What is quite fascinating, however, is the way that, at the
regular celebrations of the event in the modern Royal Navy, the victory is
provided with an almost completely spurious strategic rationale. Senior officers
solemnly lecture their audiences about how Nelson’s victory saved Britain from
invasion, forgetting that at the moment of victory the French Army, its invasion
plans long abandoned, was already on the Danube confronting Britain’s
continental allies, Austria and Russia.10
Even Mahan clearly recognized this and Corbett found the Trafalgar
campaign of such interest precisely because it was the sophisticated combination
of manoeuvre and blockade that had worsted Napoleon’s invasion attempt.11 The
battle achieved something, the blunting of a move on Naples, the demonstration
of British naval superiority, Nelson’s long desired self-immolation as an
immortal national hero, but it did not save Britain from invasion.
Battles are means to an end not ends in themselves. Nevertheless, some
battles have had momentous results. In World War II, Midway in 1942 marked
the end of Japanese expansion and the demise, for a time, of her main instrument
of naval warfare, her carrier striking force. The destruction of the latter was
completed at the Philippine Sea two years later.12 But such decisive
engagements are relatively rare. Most naval battles are but episodes of larger
campaigns and there are serious difficulties in bringing even these about,
especially in the modern world with its developed surveillance techniques. In
1941, before Cape Matapan, Admiral Cunningham, the British C-in-C
Mediterranean Fleet, was forced into a cunning deception plan in order to
prevent the Italians thinking he was about to emerge from harbour. This
unwelcome news would have discouraged the enemy sortie Cunningham needed
to get to grips with his opponent. Nelson had used similar deceptions with
similar purposes.13
The development of the aircraft and the aircraft carrier made such stratagems
less necessary. The carrier gave a new meaning to the traditional maxim of
‘seeking out and destroying’ the enemy’s fleet. At Taranto in 1940 and Pearl
Harbor in 1941 carrier aircraft struck at enemy fleets in harbour with
strategically devastating effect. In 1945 the Japanese Navy ceased to exist in the
repeated US carrier air strikes on its bases. After World War II, as it seemed to
become more and more difficult to counter enemy naval assets actually at sea,
and as nuclear weapons greatly increased the power of the offence, the ‘attack at
source’ began to dominate Western naval thinking.14 Now the new Maritime
Strategy postulates – as the ultimate but not necessarily most likely threat to
Soviet naval power – a conventional battle in which Western naval forces so
threaten Soviet assets that the Soviet Navy is both drawn out to fight and
destroyed in its bases.15
This is heady stuff and traditional US Navy rhetoric about ‘going in harms
way’ is deployed to justify it. Some modern analysts find this rhetorical approach
as disturbing today as Corbett found the equivalent slogans in the early years of
the century. Commenting on the oft-quoted British maxim that ‘the enemy’s
coast is our true frontier’ he said, ‘you might as well try to plan a campaign by
singing “Rule Britannia”.’16 Corbett, in his notes for the naval officers to whom
he lectured in 1906, set down the following important warning against going on
the offensive for its own sake.

In applying the maxim of ‘seeking out the enemy’s fleet’ it should be borne
in mind: (1) that if you seek it out with a superior force you will probably
find it in a place where you cannot destroy it except at heavy cost. (2) That
seeing the defensive is the stronger form of war then it is prima facie better
strategy to make the enemy come to you than to go to him and seek decision
on his own ground.17

When Corbett wrote this the underwater threat was ill developed and the aircraft
merely an unreliable form of observation. The development of these two forms
of naval warfare would seem to make Corbett’s warning even more true today.
The alternative to battle if one side or the other does not wish to seek it is
blockade. Again this is a rather old-fashioned term and NATO prefers
‘containment’ in its current Concept of Maritime Operations.18 If the enemy’s
main forces can be confined to their ports or to restricted areas far from where
one needs to sail one’s own shipping, then one possesses many of the fruits of
victory without the need to take the risks of battle. Blockade, however, consumes
much effort over time and its efforts are often far from complete, especially in
the era of the submarine and aircraft which not only can keep blockaders from
too close a proximity to their own bases but which can evade the blockade more
easily than surface ships. Nevertheless, given the geographical disadvantages
under which the Soviet Union labours with the various ‘choke points’ through
which its naval forces must pass, various forms of blockade (using mines, sonar
arrays, or barrier patrols of various kinds) are useful weapons in the inventory of
the West. Submarines operating forward are also potent blockading devices
sometimes, as in the Falklands War, having the facility to force an opponent not
equipped with ASW to stay in port.19
In the days of sail, battle, and blockade were the two principal weapons in the
inventory of the dominant sea power. The weaker force had two answers. The
obvious one was to attempt a guerre de course, literally a war of the chase
directed against the other side’s merchant shipping. The guerre de course is a
classic ‘sea denial’ strategy, i.e. the attempt to deny the enemy use of the sea
without necessarily being able to use it oneself. For this reason the guerre de
course was always regarded as very much a second best by classical maritime
strategists such as Mahan and Corbett. It had apparently been ineffective when
repeatedly employed by France against Britain. As Paul Kennedy has pointed
out, this might not be a totally accurate perception of historical reality. The
French guerre de course was a very powerful form of pressure on the England of
the 1690s.20 Whatever the power of the guerre de course in the age of sail,
however, the submarine transformed this strategy into a potential war winner
against powers dependent on shipping for vital supplies. Germany came
disturbingly close to victory against Britain in 1917 and again in 1942, and the
United States laid waste the shipping of the Great East Asia C-Prosperity Sphere
in 1944-5.
The only way the guerre de course has been successfully countered in the
past is by the application of the convoy system, i.e. the direct escort of groups of
ships by warships. Before going any further it must be emphasized that close
escort was, and is, only part of the overall convoy ‘system’. In addition to the
escort, ‘warships sent to the convoy assembly ports to escort ships assembled for
passage in company to their various destinations’,21 there were also support
forces of various kinds designed to give additional protection during especially
dangerous parts of the voyage. The forward deployed blockading forces also
played a vital role both to contain and survey enemy movements. Nevertheless,
if warships were not grouped directly with the ships themselves the system did
not work.22
The convoy system was almost taken for granted in the great maritime wars
of the period 1660 to 1815. Mahan was so fixated with his main fleet actions that
he almost overlooked the importance of convoy, but in his study of the war of
1812 he made clear his recognition of its central importance.23 D.W. Waters, a
most articulate and important advocate of convoys, has summed up succinctly
the change that occurred in the nineteenth century when the old mercantilist
system of state control was replaced by the new doctrines of laissez faire. The
revolution in economic thought, he argues,

imposed a revolution in naval thought. The object was no longer to protect


the ships plying the routes, but to protect the routes plied by the ships. The
key stone of the system of trade defence introduced to effect this was
blockade, a measure which had never succeeded in the past and which
modern developments in ships and weapons offered small prospect of
success in the future. The blockade was to contain the enemy fleet until it
offered battle then, as a result of its decisive defeat the world’s sea routes
were confidently expected to be safe to shipping.24

It was also expected that the next war would be short.


There are certain echoes of this thinking in modern naval strategic thought.
Those who argued before that things had fundamentally changed were, however,
proved wrong. The next war was not short and the blockade proved sadly
wanting. Only by introducing convoy was the loss rate drastically reduced to
acceptable proportions.25
One reason convoy was ignored was its ‘defensive’ association. Notionally
‘offensive’ operations seemed much more attractive to self-respecting naval
officers. In this context the following excerpt from the 1925 edition of the
British Naval War Manual is of interest: ‘It is desirable by adopting an offensive
strategy to force the enemy to assume the defensive and thus prevent him from
attacking our trade. ... In other words if the enemy attacks trade, an attack upon
some interest which he will feel bound to defend may result in the withdrawal of
forces from trade attack.’26 Sadly, in World War II such operations were both
impractical and ineffective. Fond hopes that the Germans would not be so stupid
as to reintroduce unrestricted submarine warfare (which had brought in the USA
against them and ensured their defeat) proved groundless. A partial convoy
system was introduced in 1939 but, given the diversion of assets to other
priorities, it was far from complete. This allowed the Germans, with their limited
numbers of submarines, to inflict almost all their losses on Allied merchant
shipping. ‘Up to the fall of France U-boats had sunk some 200 British, Allied
and neutral merchant ships sailing independently. Of the many thousands of
ships sailed in French and British convoys they had sunk fifteen, and had lost
eight of the attacking U-boats at the hands of the escorts – an intolerable
exchange rate for the U-boats.’27
Throughout the war the story was the same, high loss rates among
independent shipping, low loss rates and good exchange ratios in U-boats sunk
for the convoys. U-boat aces made their names against unescorted shipping, not
convoys. At the beginning of 1942, the maturing of the Atlantic convoy system
having made life in the open oceans unprofitable, the German submarines were
unleashed against the coastal shipping off the USA’s eastern seaboard which the
Americans refused to convoy until catastrophic losses had been suffered. Some
505 ships were lost west of 40°W between mid-January and the early part of July
1942.28 Ships safely protected in the open ocean were thrown away in coastal
waters. It is interesting, and not a little ironic, that the standard map used by the
US Navy today to justify its new doctrine of ‘forward operations’ quotes the
losses of this period.29 The reason for these losses, however, was not over-
commitment to the defence of shipping in home waters, rather it was the virtual
complete lack of these defences in a coherent form. Many US and British naval
assets were indeed forward deployed, but their contribution to maintaining the
flow of shipping was indirect and insufficient.
Again convoy came to the rescue. In July 1942 only 0.3 per cent of ships in
US coastal convoys were sunk.30 The U-boats came back to mid-Atlantic to
throw themselves in packs against the convoys. Here the escorts were at their
weakest due to the difficulty of providing air cover. There were apparently
higher priorities for the necessary escort carriers and the long-range maritime
patrol aircraft required. When air escort was finally provided however, and
utilization of escorts made more efficient by increasing the size of convoys in
accordance with the findings of operational research, convoy escorts and support
groups were able to inflict such exchange ratios on the U-boats that Donitz had
to call them off: ‘their sinkings were negligible, their losses crippling’. German
attempts to break back into the Atlantic against a mature convoy system were ‘a
complete failure’.31 In the Pacific the success of the American guerre de course
against the Japanese was in large part due to the latter’s failure to come up with a
proper convoy strategy.
That convoy should be so effective in maintaining shipping should not be
surprising as it combines the advantages of the offensive and defensive forms of
war as analysed by Clausewitz and Corbett. Forcing the enemy to come to you
means that he must often operate at the end of long lines of communication, in
less strength than he might otherwise be able to attain, and perhaps in waters
where you can maximize your own strength in relation to his. The defender can
thus concentrate his assets where they can be doubly effective – both offensively
to mount a crushing counter attack, and defensively to protect from loss the ships
whose movement and survival are the object of the operation. The enemy may
be deterred from attacking at all and, even if a covert enemy strikes, the ‘flaming
datum’ thus obtained gives away his presence opening him up to counter-attack
and destruction if friendly forces are there to do the counterattacking and
destroying.
This brings us to the important paradox that convoy seems often to require
large numbers of defensive forces. The argument that these numbers are
impossibly high has a pedigree almost as long as convoy itself. Astronomical
figures of the total of ships requiring protection were used to delay the
introduction of the convoy system in 1917. In fact, given its combination of the
best aspects of offence and defence it has historically been the best method of
optimizing the effect of relatively few naval assets.
In the Mediterranean in 1917-18 the local commander was unwilling to
introduce convoy: he preferred ‘an increased and unceasing offensive which
should in time enable us to dispense with the convoys and these methods of
defence’.32 Thus the Allied naval forces in the Mediterranean concentrated on
the ‘Otranto barrage’, a combined containment operation and attack at source
composed of offensive patrols by surface warships, aircraft, and submarines,
bombing of U-boat bases, and the laying of minefields. It sank a single U-boat
and conspicuously failed to stop the German and Austrian submarines getting
into the Mediterranean. The limited escort allowed to each convoy, usually a
sloop and two trawlers for thirty ships, sank eight out of twelve lost in the
Mediterranean in 1917 and 1918, and kept the loss rate of Allied ships in convoy
at one per cent.33 To take another example, this time from World War II, between
September/October 1943 and the end of May 1944, an aircraft escorting a
convoy was ten times more likely to sink a U-boat than one on offensive patrol
over the Bay of Biscay.34 Finally, and most tellingly, in the period after the great
withdrawal of the U-boats from mid-Atlantic on 23 May 1943, only a dozen
ships were lost from the many convoys that ran through the mid-Atlantic area
and none from convoys with an air escort.35
It is, of course, true that the technological conditions upon which these
performances were based have been fundamentally changed; we shall return to
this question in Chapter 10. Nevertheless, the overwhelming weight of historical
evidence seems to demonstrate that unless some direct protection is given to
shipping then the exercise of command of the sea may be impossible. In other
words, the battle around the ships themselves may be as vital to gaining
command of the sea as operations elsewhere, perhaps more so. In this sense it is
perhaps misleading to refer, as do the classical strategists, to the direct defence
of trade as merely the exercise of command of the sea. Command is gained
elsewhere only to the extent that the battle/blockading forces neutralize the
enemy’s most capable assets which might otherwise be able to deal with the
relatively light and dispersed convoy escorts. The escorts, however, are just as
important as the heavier and more spectacular forces. Indeed, recent evidence
seems to show that they do a great deal more fighting and their battles are the
most important battles that actually take place. In the world wars it was around
the convoys that Germany’s most dangerous naval assets, i.e. those the country
was able to use, were sunk. It was, therefore, around the convoys that the
decisive Allied naval victories against Germany were won. Defence of shipping
via convoy ought, therefore, to be added to battle and blockade as a crucial
offensive means of gaining command of the sea. Far from being a defensive
measure, convoy allows the naval offensive to be most efficient and effective, a
classic Clause-witzian advantage of the defensive form of war. This deserves
clearer recognition as a principle of maritime strategy, as it explains why a
supposedly ‘defensive’ strategy turns out to be asset intensive. If one does bring
the enemy to battle around a convoy one needs superiority, like any attacker, to
succeed.
Commander Waters has gone so far as to produce twenty-one ‘operational
laws’ of naval warfare based on historical experience. They may be scenario
dependent in respect of the technology available but, using Corbett’s rule, they
demonstrate ‘the normal’ from which technological developments may draw us.
They are thus worthy of note.36

1. The risk to an independent ship of being detected and attacked Rs α Ni


(Rs) varies directly as the number of ships (Ni) at sea.
2. And directly as the number of enemy submarines (U) at sea. Rs α U
3. Therefore combining laws 1 and 2. Rs α NiU
4. The risk to a submarine (Ru) when attacking an independent Ru = 0
ship of being sunk by an escort force (E) is zero because E = 0.
5. It does not vary with the number of ships attacked (Ni) because Ru = Ni × 0
E = 0.
6. The risk of a ship in convoy (Rc) varies inversely as the size (5
= number of ships in the convoy).
7. And directly as the number of convoys (Nc) at sea. Rc α Nc
8. And directly with the number of submarines (U) at sea. Rc α U
9. Therefore combining laws 6, 7 and 8.

10. Therefore the larger the convoy and the fewer the convoys, the
smaller is the risk to a ship in convoy of being attacked.
11. In the absence of air escort the loss of ships (Lc) in convoy Lc α Uc
varies directly as the number of submarines attacking.
12. And inversely with the number of escorts with the convoy.
13. Therefore combining laws 11 and 12.

14. The number of submarines sunk (Kc) by surface escorts when Kc α Uc


attacking a convoy varies directly as the number of attacking
submarines (Uc)
15. And directly as the number of escorts (E) with the convoy. Kc α E
16. Therefore combining laws 14 and 15. Kc α UcE
17. For an enemy attacking shipping the yardstick measuring
victory or defeat, i.e. the exchange rate, is the number of
submarines sunk while attacking convoys (Kc) per ship sunk in
convoy (Lc).
18. The exchange rate also measures the value of escorts (E) with
a convoy. If Lc α Uc/E and Kc α UcE then
therefore:37
19. The operational value of antisubmarine forces, measured in
terms of the exchange rate, is proportional to the square of the
number of escorts only when the latter are used as convoy
escorts, not when they are used in other ways, e.g. on
‘offensive’ hunting patrol. The E2 law also applies to aircraft if
used as convoy escorts. The value of ‘hunting patrols’ is
measured by a different exchange rate, independent ships sunk
by submarine (Li) per submarines sunk by hunting patrols
(Kh). ‘In each war this exchange rate was ten times as costly to
the Allies as the convoy exchange rate (Law 17) . . . The
“convoy” and “independent hunting” exchange rates when
compared give of course a measure of the efficiency of convoy
as opposed to hunting as an operation of war. In each war it
was about ten times as efficient as hunting.’38
20. The number of escorts (E) required to protect a convoy varies
as rather less than the square root of the number of ships in the
convoy (S): Therefore the larger the convoy the fewer the
escorts required.
21. Each doubling of inter-ship spacing in convoy reduces the risk
of loss by 75 per cent but increases the defence perimeter of
the convoy by only about 50 per cent.

One useful way in which the guerre de course can be supported by a weaker
naval power is to maintain a main battlefleet ‘in being’ to draw off assets that
might otherwise be used in convoy protection. The Germans succeeded in doing
this in both world wars. The shortage of destroyers caused by the requirement to
escort the Grand Fleet played a crucial role in delaying the introduction of
convoy in 1917. Only when it seemed that the United States could provide the
requisite escort vessels did Admiralty attitudes to convoy change.39 Pioneered by
the English Admiral of the Fleet the Earl of Torrington as long ago as 1690 as a
way of preventing a French invasion, the ‘Fleet in Being’ strategy has
demonstrated considerable potential at various times. Reduced to its essentials it
can be defined as the use of options provided by the continued existence of one’s
own fleet to constrain the enemy’s options in the use of his. It can be used both
as a sea denial strategy in preventing the superior enemy transporting his
valuable goods or troops across an uncommanded sea, or to help to gain
command of the sea for one’s self by holding down the enemy’s assets to protect
what is valuable to him while one’s own shipping proceeds with little or no
molestation. It is sometimes a difficult strategy to follow as avoiding action is
frustrating to one’s self, one’s crews and one’s political leaders. The German
Fleet mutinied in 1918: Torrington was dismissed after being forced into
premature battle by his Government and had to stand trial for his actions.
Happily, both for maritime strategy and the hapless Admiral, he was acquitted,
although he never served at sea again.40
It is the passage of shipping that is the main fruit of command of the sea. This
shipping can be commercial or military, and it can cover everything from aircraft
carriers to cargo ships. The aim of the passage can be to carry mundane cargoes
like coal or oil, or perhaps troops for landing. Alternatively or additionally, the
aim might be the application of firepower against some target ashore. Such terms
as ‘sea communications’ or ‘sea lines of communication’ are dangerous in that
they tend to mask the fundamental truth that maritime strategy is about ships,
defined as the means of movement on or below the surface. One does not defend
the sea as it is usually a matter of indifference whether explosions take place in
empty water. It is hardly surprising in this sense that some land-based Eastern
European analysts see NATO’s dependence on the sea for supplies as a positive
asset.41 The sea cannot be blown up; but ships can be sunk or disabled.
Western maritime strategy has tended to emphasize the passage of shipping
carrying goods and supplies, as it has largely been based on the experience of
Britain, the last and greatest of the maritime empires of the ‘Age of Vasco da
Gama’ which ran roughly from 1500 to 1900. Continental maritime strategists
have tended to emphasize sea denial like the French ‘Jeune Ecole’ of the late
nineteenth century. The Germans dallied with Mahan in the early twentieth
century but adopted ‘Jeune Ecole’-like ‘tonnage warfare’ concepts in the late
1930s.42 The Soviet Union has taken a slightly different approach by
emphasizing the utility of ‘operations against the shore’, both in terms of
defending against them and carrying them out. Admiral Gorshkov
understandably made much of these in his writings, having organized much
amphibious warfare in the southern sector of the Eastern Front in World War
II.43
Soviet amphibious operations in World War II, although numerous, were
relatively short ranged and coastal or riverine in nature. There was nothing to
compare with the massive cross-Channel invasion of 1944 or the trans-oceanic
amphibious operations in the Pacific carried out by both the Japanese and the
Americans. The techniques of amphibious warfare were developed enormously
by the United States, in particular with the birth of a whole new family of
amphibious warfare vessels. These have made the application of amphibious
military power more practical than it has ever been before, although on the other
side of the equation the increased scale of coast defences has kept up. Judging
the balance of advantage can, therefore, be difficult. The tendency to write off
amphibious landings as impractical has been persistent throughout this century
but the lessons of World War II and since are that, if sufficient surprise is
obtained and/or air power can be brought to bear both to neutralize the inherent
advantages of land communications and help together with naval gunfire support
to overwhelm the coast defence, amphibious warfare even against significant
opposition can still be a practical operation of war.44 We will return to this
matter in Chapter 4.
One of Corbett’s most controversial arguments was that sea power allowed
the use of limited, but significant, pressure against major opponents in order to
force political concessions and achieve the objectives of the war. Corbett quoted
the recent Russo-Japanese War as his main example but, as Dr Andrew Lambert
has pointed out, a closer examination of the Russian War of 1854-6 would have
yielded a more interesting example still.45 Dr Lambert argues convincingly that
the Anglo-French alliance was remarkably successful in exerting maritime
pressure against the Russian Empire. This was less in the Crimea where ‘a
combination of inter allied friction, hesitant leadership and political weakness
allowed the operation to degenerate into a long campaign in which two
entrenched armies fought out a battle of attrition for a ruined city’.46 Given the
absence of railways on the Russian side, the ability of the Allies to capture
Sebastopol, and later Kinburn at the mouth of the Dnieper, was not too
surprising. But neither success in the south directly caused Russia to accept the
Allies’ terms. However, Russia did submit. Dr Lambert may well have identified
the reason in British activities, and threatened activities, in the Baltic.
Once Sebastopol had been occupied, the British next planned to unleash in
1856 a ‘Great Armament’ upon Kronstadt, the fortress guarding the approaches
to St Petersburg. The ‘Great Armament’ was a massive bombardment fleet of
about four hundred steam gunboats and mortar vessels which would attack
Kronstadt with mortars, guns, and rockets. The attacking force also included
eight ironclad floating batteries and no less than twenty-four line of battleships.
With Kronstadt demolished and occupied ‘something the responsible officers
had every confidence in, St. Petersburg would be open to any insult the British
cared to administer.’47 Sweden might also join the Allies to regain Finland. In
these circumstances the Russians gave in and began peace talks during which the
assembly of the ‘Great Armament’ was used by the British ‘to back up their
tough stand . . ., both to beat down Russian resistance and to impress the
neutrals’.48 Russia was not in danger of complete defeat but the balance of
advantage was clearly to submit to British pressure. Dr Lambert concludes that
‘The Russian War demonstrated that Russia was wide open to maritime threat.’49
Both Admiral Gorshkov and his successors, and the authors of the new
American ‘Maritime Strategy’, would hasten to agree that she still is.
Before concluding this survey of the state of the art it is useful to look at what
Corbett called the ‘theory of the means – the constitution of fleets’.50 Corbett
wisely recognized that the differentiation of the ships of his time into three
groups, ‘battleships, cruisers and flotilla’, while it had ‘so long dominated naval
thought that we have come to regard it as normal, and even essential’, was by
‘no means constant.’51 He perceptively described the evolution of the sailing
fleets of the eighteenth century into ‘battleships’, designed primarily to secure
command of the sea, and ‘cruisers’, faster but less powerful ships, designed
primarily to exercise that command by giving direct protection to shipping, and
secondarily to support the battle-fleet by scouting. The flotilla was the mass of
smaller vessels ‘for coastwise and inshore work, despatch service and kindred
duties’.52
Corbett recognized how during his day the classic threefold distinction was
being eroded, both by the construction of large cruisers of similar fighting power
to battleships, the advent of light cruisers designed primarily for fleet work and
the development of the flotilla into the ocean-going torpedo boat destroyer, a
vital component of the battlefleet in both offence and defence. Corbett sensed
that these developments reflected the dominance of the decisive battle or
‘overthrow’ school of naval strategy.53
As usual, he was right. When the teaching of the latter school proved
insufficient to give the Allies complete command of the sea in World War I a
whole new class of non-battleworthy ‘cruisers’ had to be developed, the escort
sloops, to accompany the ocean-going destroyers that were also pressed into
service to escort convoys against submarines. To a considerable extent the
‘mating’ of these two types of vessel – the relatively slow ‘sloop’ optimized for
the escort of merchantmen against submarines and later, aircraft, and the fast
fleet destroyer, which in order to carry out its fleet defence function in World
War II became primarily an anti-air warfare and anti-submarine vessel – has
provided the parentage of the modern major surface combatant in its various
guises and forms. The ships are, however, primarily classical ‘cruisers’ in that
they are not the primary striking forces of the major navies: that role has fallen
to submarine and aircraft. Surface ships are still necessary to screen some of
these striking forces, those based in mobile surface ship platforms, i.e. aircraft
carriers. Moreover, if it is accepted, as argued above, that the battle around the
convoys has become a primary means of gaining fleet command in itself, the
assets engaged may well be ‘battleships’ of a kind. Nevertheless, in that their
main role is the escort of higher-value shipping, modern surface warships,
‘destroyers’ and ‘frigates’, are classical cruisers first and foremost. Indeed, the
larger examples have taken over the name as it was vacated by the last of the
large, gun armed ‘neo-battlecruisers’ of the first half of this century, ships
designed for surface warfare and combat with their own kind. The large
maritime patrol aircraft, designed primarily for anti-submarine work, might also
qualify as latter day ‘cruisers’ of the air.
What has become of the battlefleet? As stated above, it has taken to the air
and gone beneath the waves. It was soon clear that aircraft had the potential to
sink anything that floated, but the air enthusiasts overstated, sometimes grossly,
the capabilities of the aircraft available during the two decades after World War
I. The Naval Staffs of the time were right to stick to the large gun-armed
battleship as the core of their capability to sink the heaviest and toughest ships
available to the other side. However, the richest and least strategically over-
committed of them, the Americans and Japanese, were also able to develop
aircraft, both carrier and sometimes shore based, as complementary striking
arms. About 1940-1 the aircraft did overtake the battleship with the advent of the
high-performance torpedo bomber, supplemented by high-performance dive
bombers with heavy bomb loads. In the Pacific, aircraft soon asserted their
position as the dominant strike arm against even the largest enemy surface
warship. The largest battleships ever built, the 70,000 ton (full load) behemoths
Yamato and Musashi, were overwhelmed by American carrier-based aircraft
despite their massive AA gun armaments and their being able to manoeuvre in
the open sea. The former blew up after being hit by perhaps thirteen torpedoes
and eight bombs, the latter by twenty torpedoes and seventeen bombs.54
Although land-based aircraft could be as effective, as when the Japanese 11th
Air Fleet disposed efficiently of the battleship Prince of Wales and battlecruiser
Repulse in December 1941, carrier-based aircraft tended to be more useful and
flexible under maritime conditions. Hence the large strike carrier become the
new ‘battleship’, one capable of fighting a three-dimensional anti-surface, air
and underwater battle. Sadly, for the rest of the world’s navies only the United
States is now capable of affording such mighty ships. The rest must make do
with smaller carriers (better described as large ‘cruisers’) or land-based maritime
strike aircraft. The Soviet Union was forced to use missile technology to
substitute for non-availability of proper carrier aircraft. From the late 1950s she
produced what were in effect ‘battlecruisers’ to complement her primary
submarine and land-based air striking forces, culminating in the monsters of the
1970s and 1980s, some relying on a mix of aircraft and missiles and others on
missiles alone.
In one important way the modern attack carrier is also better described as a
‘battlecruiser’ than a ‘battleship’ in that she relies overwhelmingly on striking
power to protect herself, not armour protection. In this she is very like Admiral
Fisher’s original conception of a battlecruiser, a ship that could strike at enemy
vessels from ranges where they could not hit her. Sadly, Fisher’s assumption of
British gunnery superiority was let down by his subordinates and successors
refusing till too late the fire control system that might have given their ships this
ability. Nevertheless, in a sense the erratic British naval genius of the twentieth
century was vindicated by the replacement of the traditional battleship by the
carrier as the dominant surface warship.55
In 1919, shortly before the end of his tempestuous life the following year,
Fisher summed up naval development as he saw it: ‘The greatest possible speed
with the biggest practical gun was up to the time of aircraft the acme of sea
fighting. Now there is only one word – submersible.’56 As usual Fisher
overstated his point – he may well indeed have been considering semi-
submersible gun armed battlecruisers rather than submarines proper – but there
was a grain of truth in it nonetheless. By the time he wrote the above the
submarine had passed from being a mere coast defence flotilla vessel to a
powerful commerce-raiding cruiser. Attempts by the British, and later the
Japanese, to make it into a fleet cruiser, however, were less than successful, and
only the Americans proved able in World War II to use submarines as effective
striking forces in fleet actions as a kind of ‘advanced cruiser line’, notably at the
great victories at Philippine Sea and Leyte Gulf in 1944. Since 1945, however,
the submarine has grown into a full-scale strike arm symbolized by the
‘Dreadnought’ names given to Britain’s early nuclear-powered submarines.
Interestingly, the Americans, who in the 1920s and 1930s built heavy cruisers,
unnecessarily large in British eyes, primarily as fleet ‘battlecruisers’, have
chosen cruiser names for their submarine strike force.
The fact that the primary maritime striking force of the world’s major navies
is split between air and submarine platforms makes one over-used term obsolete
– ‘capital ships’. Oddly enough this term came into vogue in the early years of
the twentieth century as an umbrella term for large gun-armed battleships and
battlecruisers just as these vessels were increasing their dependence on lighter
screening forces. Nevertheless, it did have some meaning as ‘the unit of absolute
power . . . unsinkable except in combat with a superior force of similar ships’.57
No vessel is in such a happy position today. Submarines can torpedo carriers,
carriers can take on submarines. Equally submarines cannot shoot down aircraft.
Even old World War II destroyers dropping depth charges can neutralize the
activities of grandly named nuclear submarines. There are no ‘capital ships’ now
and modern naval warfare is a complex three-dimensional kaleidoscope.
However, before going further into exploring its nature we ought to discover
what is being fought about: what uses of the sea are worth all the trouble? And
how important is the sea in the violent resolution of conflicts caused ashore?
Part II

The uses of the sea


Chapter 3

Economic uses of the sea

The main traditional use of the sea is as a medium for transport. Despite the
advent of the transcontinental railway and the transoceanic aeroplane the sea still
dominates international transport movements. Seaborne trade, as we have
already pointed out, accounts for over 80 per cent of international trade by
volume.1 If weight is the criterion then the importance of water transport
increases still further. ‘No less than 95 per cent of all trade that crosses frontiers
is waterborne, including inland. Over 99.5 per cent by weight of all trans-ocean
cargo is carried in ships, under half a per cent in aircraft, and there is no viable
alternative mode for most of it.’2 World seaborne trade is rising again, 3,362
million tonnes being carried in 1986, some eight times the amount carried in
1946. The trend is upwards despite only 3,090 million tonnes being transported
in the trough of the recession in 1983. The previous peak was 3,606 million
tonnes in 1980.3
Crude oil and petroleum products provide by far the largest single category of
seaborne trade, an estimated 940 million tonnes of the former and 310 million
tonnes of the latter being carried in ships in 1986.4 The Middle East dominates
the export of oil by sea with about half the world’s exported oil originating there.
Western Europe and Japan are the main customers, over a third of the former’s
total oil imports and no less than two-thirds of the latter’s being Middle Eastern.5
All of this must move in tankers, the former (mostly) round the Cape and
through the Suez Canal, the latter across the Indian Ocean and then through the
many Straits leading into the South China Sea.6 Large quantities of oil are,
however, carried by sea within Western Europe, from North Africa to Western
Europe, West Africa to Europe and North America, Soviet East Asia to Japan,
Alaska to the continental United States (some through the Panama Canal), and
from the Caribbean to North America and Europe. The complete pattern of the
world seaborne oil trade is shown on Figure 3.1.
Figure 3.1 Main oil movements by sea, 1987
Source: BP Statistical Review of World Energy, June 1988, p. 19.

As noted above, most oil is carried as crude. This allows refineries closer to
the markets to obtain the best blend of crudes for their production. The trend
towards long hauls of crude caused tankers to increase steadily in size from 1950
onwards. By 1959 the first 100,000 deadweight ton ship had been built and over
the next decade the very large crude carrier (VLCC) 200 to 300 thousand
deadweight tons (dwt) was born. With the closure of the Suez Canal after the
1967 Arab-Israeli War forcing long voyages from the Gulf round the Cape to
both Europe and the United States, and the very high profits obtainable as oil
demand steadily increased, large numbers of VLCCs were constructed. By 1973
the ultra large crude carrier had arrived, the monster of half a million deadweight
tons. At this point economic conditions sharply changed. Oil prices increased,
demand diminished and, not least, the large oil companies lost their dominance
in the oil market place. The latter factor ‘reduced the average size of their oil
parcels shipped by sea, increased the number of tanker routes and increased the
number of tanker charterers’.7 Demand for VLCCs slumped by half between
1978 and 1982, and few were constructed. The reopening of the Suez Canal and
its deepening to allow the smaller VLCCs through also has put a new premium
on moderate dimensions for carrying crude oil. Petroleum products like petrol,
diesel oil and paraffin have always been carried in much smaller ships,
nowadays of around 25,000 dwt and smaller ships still are used for coastal and
riverine traffic.8
Another major bulk hydrocarbon traffic is liquified gas. Taking liquid
petroleum gas (LPG – butane and propane) from the oilfields, where it was often
flared off as waste, to consumers in the developed world began in small ships in
the 1950s. For larger ocean-going LPG carriers refrigeration began to be used to
help store the volatile cargoes and this led the way to ocean transportation of
natural gas (LNG – mainly methane). Japan, especially, requires large amounts
of LNG to come by sea and this demand is fulfilled by impressive and expensive
specialized LNG carriers capable of carrying 125,000 or more cubic metres of
volatile cargo at temperatures below its boiling point of —161 °C.
Moving now to solid bulk cargoes, the most important is iron ore, of which
304 million tonnes, 9 per cent of total goods, were carried by sea. About a third
of world production is carried by sea, and the ore producing areas of Sweden,
Australia, Canada, Brazil, Venezuela, India, and West Africa are linked to the
production centres of Japan, West Germany, the United States, the United
Kingdom, Belgium, France, and Italy. All the latter producers use large amounts
of imported ore. Although iron ore can be carried in non-specialized tramps,
most is transported in bulk carriers of around 35,000 to 45,000 dwt tons. For
long voyages, notably those to Japan, the bulk carriers used may be quite large –
in the 100,000 to 150,000 dwt bracket. Some ships in this category are so-called
‘combination carriers’, doubling as transporters for oil and ore. Such ships,
however, tend to be expensive and awkward to operate, factors militating against
their flexibility.9
Almost as much coal is carried by sea as iron, some 268 million tonnes in
1986, 8 per cent of total sea trade. This is required by steel industries with
insufficient supplies locally, and also for electrical power generation. Japan and
Europe are the major importing areas: the former consuming a third of world
imports. Major producers for seaborne distribution are South Africa, North
America, India and Australia. Eastern Europe also exports significant quantities
of coal by sea through the Baltic. Coal is carried in bulk carriers in the 35,000
dwt bracket. Ships of up to twice this size can be used – the largest that can
transit the Panama Canal – a significant consideration with traffic to be carried
from the Eastern Seaboard of North America to Japan.10
Another important dry bulk cargo is grain, 160 million tonnes of which were
shipped by sea in 1986, 5 per cent of total trade by weight. Depending on
demand, North America and Australia export all over the world. Standard bulk
carriers are the normal grain carriers, although even tankers are pressed into
service at times. Grain is also a staple cargo for the small tramping vessel. The
above cargoes total 270 million tonnes, and amount to 8 per cent of total trade.
Moving on to other bulks: bauxite travels from Australia, West Africa, and the
Caribbean to Europe, Japan, and North America; phosphates from the United
States, the Pacific islands, and Morocco to Europe, Japan, and Australasia;
manganese ore from Brazil, Africa, India, and Australia to Western Europe,
North America, and Japan; copper from Canada, the Philippines, and Papua New
Guinea to Japan; nickel from New Caledonia, Indonesia, and the Philippines to
Japan; zinc from Canada, Peru, and Australia to Europe and Japan; chrome from
South Africa to Japan and elsewhere; sulphur from North America and Poland to
all over the world; and sugar from South America, the Caribbean, Africa,
Australia, the Pacific islands, and various parts of Asia to North America,
Europe, and Japan (and from France to Africa and from Asia and Cuba to the
USSR). The medium-sized bulker is the rule for most cargoes, although large
combination carriers and tramps are used for some, e.g. sulphur and phosphorus.
More complex chemicals are carried in small (25,000 ton) specialized tankers.
These ocean-going tankers cover two main routes, from the Gulf of Mexico to
Europe and from Newfoundland to both Europe and Japan. As with petroleum
products, smaller chemical tankers still are used in coastal waters.11
Exactly a third of all seaborne trade by weight in 1986, 1,110 million tonnes,
was classified as ‘other dry cargoes’, e.g. fruit, meat and manufactured goods.
Once the preserve of the developed world, increasing standards of living in the
developing countries are helping maintain demand for such items. Dry cargo
ships can be liners, operating regular routes to a fixed schedule, or tramps sailing
where the cargoes can be found. The traditional type of vessel for both types of
service used to be the ‘break-bulk ship’ with holds loaded with two to five ton
lots of cargo by slow, labour intensive means through hatches into the hold. As,
after 1945, labour increased in cost and ships, cargo liners especially, got faster,
so the relative cost of the masses of dockers required and time spent idly in port
increased. Ships were often tied up for more than half the year. Very large break-
bulk ships were not possible because of the long turn round times required.12
The answer was ‘unitization’, ‘the breaking down or, alternatively the
syntheses up, of goods into standardised units that can be handled by
mechanised and possibly automated equipment’.13 The most important form of
unitization is the use of the standard ISO container, 8 × 8 feet in cross section
and 10, 20 or 40 feet long. The 20 foot container, the usual standard, has a
maximum payload of about 18.5 tonnes.14 The advantages over break bulk are
obvious: ‘As break bulk, drums of paint, wood bales, cars, crates of canned
goods, cement bags, machine tools and so on have to be packaged, crated and
handled in different ways, and stowed differently in the ships themselves, but all
can go in containers which can be handled in one standardised manner no matter
what is inside them.’15 Containers require specialized capital intensive port
facilities, capable of handling ten times the cargo with only a third the labour.16
Between 1965, when the ISO container was born, and 1980, dry cargo sea
transport in the developed world was transformed in ‘even more of a revolution
than that in break-bulk cargo following the introduction of really efficient steam
engines’.17 In addition to the fundamental advantages of lower port costs and
greater time at sea, pilferage was significantly reduced (10 per cent) and crews
were greatly reduced in size (90 per cent).18
Container ships have rapidly evolved through several generations. They were
always relatively large given the increased volume required. A typical ship of the
1970s would be of about 20,000 dwt tons with a capacity of 1,500 twenty-foot
equivalent units (teu) and a speed of 21.5 knots. During the 1970s capacities
increased to 2,000 – 2,500 TEU and speeds to 26-30 knots. The rise in the oil
price soon undermined the economies of the fast container ship, and at the
beginning of the 1980s re-engining with slower and more economical engines,
usually diesels, became the rule. Economical cruising speeds are now 18 knots.
Container ships of the 1980s vary from small ships of under 15,000 dwt tons to
the huge 4,380 teu, 58,000 dwt giants used by United States Lines on their round
the world route. The limiting factor on such vessels is the requirement to transit
the Panama Canal.19
Even container ships suffer from supply outpacing demand and a growing
tendency is to build ships in the 15,000-45,000 dwt bracket that can carry either
containers or bulk cargoes. Such ships can move between trades when squeezed
by problems of over-capacity in one or the other. One example of such a ship is
the British designed 23,000 dwt Ecoflex type designed to carry general cargo,
grain, ore, or containers. The Scandinavians have developed a type of ship
known as the ‘Bo-Ro’ which combines this flexibility with roll-on roll-off
facilities. A Bo-Ro’s voyage pattern demonstrates its enormous versatility:

Loading 5,000 tonnes of cargo in the Far East, including mobile cranes,
motor cars, containers, etc., at rates varying from 300 tonnes per hour for
containers to 1,000 tonnes for the vehicles, the ship was unloaded in North
West Europe. There nearly 10,000 tonnes of oil cargo were lifted for a short
coastal feeder trip. After discharging this, 10,000 tonnes of forest products
were loaded on trailers at nearly 700 tonnes per hour. This was in turn
unloaded in Belgium where 700 motor cars were taken on, together with
9,000 tonnes of fuel oil in Amsterdam and the whole of the cars and fuel
cargo delivered to Scandinavia.20

Roll-on roll-off (Ro-Ro) facilities are space intensive but repay the investment in
terms of ease of loading and unloading. The technology was pioneered by the
landing vessels of the World War II. The first Ro-Ro ships were in fact converted
British tank-landing ships used on short sea operations in British waters from
1946 onwards. The terminals used were also wartime in genesis – pontoons and
bridges used in the Mulberry harbours. Specialized Ro-Ro cargo ferries followed
and by the 1970s ocean-going Ro-Ro ships of 35,000 dwt were in service.21
Some of the most impressive Ro-Ro cargo liners today are the 36,500 dwt
combined Ro-Ro and container vessels used by the Atlantic Container Lines
consortium on the North Atlantic run. With a capacity of over 2,000 teu and over
600 vehicles they are the Atlantic liners of today.22
Modern ships such as these tend to dominate the trades of the developed
world but the break-bulk vessel is far from dead. There are still more ships of
this type than any other in the World’s trading fleet. Where high labour costs can
be afforded for various economic reasons, e.g. in the developing world or in the
Socialist countries, it remains viable, if not essential, for the type of port
facilities available. Container handling facilities are, however, spreading. Ports
with new installations in recent years have included Bombay, Calcutta, Cochin,
Colombo, Curacao, Karachi, Madras, and Rangoon, as well as Chinese,
Egyptian, Libyan, and Thai ports.23 Although old break-bulk vessels are still so
cheap as to be not worth replacing with new construction, one would expect
container vessels with cranes and holds to slowly displace them as the years go
by and containers become even more widely used. The current design of Soviet
general cargo vessel is a 13,500 dwt ship, capable of carrying 320 containers in
its hold, 68 more containers on deck, loads unitized in pallets, or even cargoes of
bulk grain.24
Many old break-bulk vessels are capable of carrying refrigerated cargoes,
largely food. The ‘reefer’ trade has kept relatively healthy, thanks to the rising
standard of living in the developed world and the expanding demand for new
and different kinds of fruit. The Middle East has also become a major food
import market with the rising standards of living there caused by oil price rises
in the last decade. Container ships, normal liners with provision for refrigerated
cargo, have taken some of the reefer traffic but the seasonal and/or more
specialized facilities (e.g. bananas) are becoming the preserve of modern
specialist reefers designed to carry their cargoes in pallets rather than
containers.25
Another kind of specialized ‘unit load’ vessel, and one of the few types of
ship for which demand remains good, is the vehicle carrier. The Asian motor
industry provides a demand for significant fleets of such vessels, sometimes
owned by the motor manufacturers themselves, each one carrying over 2,000
cars at up to about 18 knots from the factories to Europe, North America, or
elsewhere. The vehicles are simply driven on and off the ships.26
A final and spectacular form of ‘unit cargo’ vessel that must be mentioned is
the large carrier (‘lighter or barge aboard ship’ vessel – LASH or BASH). These
are ships that carry lighters or barges to be floated or lifted on and off the ship at
each end of the voyage. The first appeared in 1969 and was soon followed by
ships designed to carry both containers and barges. Such ships have a number of
advantages, notably the ability to land cargoes without complex harbour
facilities and to send barges directly into inland waterway systems. The early
ships were all American, except for one Dutch vessel, and did not prove as
popular as expected, possibly due to the availability of Ro-Ro and container
vessels at all major ports dealing with likely cargoes. Nevertheless, the concept
was taken up once more in the late 1970s in both West Germany, with its
extensive inland waterways, and the Soviet Union. The latter has a less
developed port infrastructure and a tendency to use merchantmen for military
purposes: both considerations made LASH vessels attractive. A number have
been built in Western yards for the USSR, both ocean-going ships and smaller
feeder vessels for operating at the ends of the routes of the larger ships and for
relatively short sea voyages. Soviet yards were used, however, for a new class of
impressive 41,000 dwt ships designed for arctic operations. These Aleksey
Kosygin class ships can carry 82 barges or almost 1,500 TEUs and are used on
the run between Vladivostok and the mouth of the Yenisey river in the Kara Sea.
This Arctic route is the Soviet Union’s only way of providing heavy transport
facilities to its Northern Siberian lands; given the ice conditions, nuclear power
would have advantages. The Kosygins are diesel propelled but it now seems that
the slightly smaller successors of the Sevmorput class (32,000 dwt and 73
barges/1,300 teu) are nuclear powered. This form of propulsion has not found
favour in the West for merchant shipping, despite various experiments. It is only
practicable in Soviet-style economic conditions and even then for special
purposes like Arctic navigation. The USSR has a powerful fleet of icebreakers to
break passages for its more conventional merchantmen: some of these
icebreakers are also nuclear propelled, the first, the Lenin, was in 1957 the
world’s first nuclear-powered surface ship. She suffered a serious nuclear
accident nine years later, but now appears to be back in service.27
The Soviet Arctic sea route is perhaps the most spectacular example of
coastal and short sea trades. All over the world considerable volumes of traffic
are carried on short voyages by ships, many of which can penetrate far up river
or internal canal systems. Modern coasters must be able to carry containers and
the ‘open hatch’ vessel has become popular. Ro-Ro is also a useful asset when
time in port is at a premium. Coasters are usually quite small, varying from
under 500 to over 2,500 dwt, but about 1,100 dwt is the minimum for
profitability, in Western Europe at least where the German Federal Republic is
now perhaps the leading coasting nation.28
Short distance sea traffic also remains competitive in the transport of
passengers, especially if they are accompanied by their cars. Ferry services using
Ro-Ro techniques play a crucial role in the transport systems both of Western
Europe and Japan. An important route, although one with freight and military
cargoes in mind, is the new combined ‘rail and car ferry’ route between Kleipada
in the Soviet Union and Rugen Island in the DDR. This will have a capacity of
over 5 million tons per year by 1990 and will provide more secure transit,
avoiding the Polish railways. Ferries are becoming progressively larger as ferry
operators try to meet increased freight demand, maximize general economies of
scale, and give greater passenger comfort, especially on overnight sailings, all in
a highly competitive environment. In 1987 British Shipbuilders delivered a Ro-
Ro passenger ferry of no less than 31,000 gross registered tons, the largest
passenger ship built in the United Kingdom since the Queen Elizabeth II in
1967.29 Their owners hope these new, large ferries will help stave off such
threats as the Channel Tunnel. Sea transport will, however, remain the only
means of movement in relatively remote areas, e.g. in northern Norway. New
types of small craft, e.g. catamarans and hovercraft, found their main
commercial niche in fast, short distance coastal services where land alternatives,
for various reasons, were unavailable.
The larger passenger liners now remain almost purely as cruise ships,
although sometimes a limited regular passenger service is offered when ships
sail to and from their cruising grounds. One significant by-product of the end of
the ocean passenger liner as a means of transport rather than a means of
recreation has been the drop in speed of such vessels. Once some of the fastest
ships afloat with speeds of 28 knots or more, liners now rarely exceed a more
economical 22 knots. The world’s largest liner, the Norway, had her speed
reduced from 30 knots to 18 knots when she was refitted for cruising, minus two
of her original four turbines. Converting a ship for cruising is not just a matter of
reducing the speed, the whole layout must be altered to optimize its ability as a
floating hotel, swimming pool, and sun lounge.30
In 1983 the United Nations listed the percentage deadweight tonnage of the
world’s mercantile fleet of 37,000 ships as 44.1 per cent tanker, 24.7 per cent
bulk carrier, 16.5 per cent general cargo, 7.1 per cent combined bulk
carrier/tanker, 2.1 per cent container ship, 0.5 per cent vehicle carriers, 0.4 per
cent passenger ferries, 0.1 per cent barge carriers, and 1.2 per cent others. The
total tonnage of 686 million dwt was 47 per cent owned by the developed market
economy states, 29.1 per cent open registry (i.e. ‘flag of convenience’) countries,
15.3 per cent developing countries, and 7.9 per cent Socialist nations. The figure
of about 37,000 was an increase since 1971 of about 4,000 ships.31 Since then
numbers have fallen again. On slightly different criteria, cargo ships and tankers
of over 100 grt, the number of ships in the world trading fleet has declined from
38,307 of 673 million dwt on 31 December 1983 to 36,454 ships of 626 million
dwt on 30 December 1986.32 The breakdown of the world trading fleet is as
follows:

Number million dwt


Container Ship 1,053 21
Ore/Bulk Carrier 4,945 195
Combination Carrier 363 39
Other Dry Cargo 21,833 109
Tanker 7,495 252
Liquified Gas Carrier 765 10

As for ownership, the top ten merchant fleets by flag on 31 December 1986
were as follows:33

Country million dwt world percentage


Liberia 97.8 15.6
Panama 69 11
Japan 55.8 8.9
Greece 46 7.3
USA 25.1 4
USSR 23.4 3.7
Cyprus 23.4 3.7
China 17.9 2.9
Hong Kong 13.7 2.2
Philippines 13.7 2.2

Hong Kong is a UK Dependency using the Red Ensign. This flag is worn by
ships of a number of different registries. If all Red Ensign ships were counted
(notably the United Kingdom itself with 11.1 million dwt tons, Bermuda,
Gibraltar and the Cayman Islands) the fleet would move above the USA to
number five. Except for the United Kingdom these Red Ensign registries are
about as ‘open’ as the classic ‘flag of convenience’ states.
‘Flags of convenience’ proper are those of nations such as Liberia, Panama
and Cyprus, who open their registries to owners who wish to employ lower-cost
crews than would be possible under national regulations, to avoid paying tax to
the national government of the ‘beneficial owners’ and to avoid some, if not all,
regulations for safety, etc. The Red Ensign is the only open registry that can
offer naval protection. Flags of convenience are unpopular with the developing
countries as they feel that those developing nations which are flag of
convenience states do not receive sufficient benefit themselves and that the
system provides unfair competition for the fleets owned and flagged by
developing countries.34 The main beneficial owner of flag of convenience
shipping has been the United States. In 1981, of a total fleet over 100 grt of
almost 2,000 vessels, over 550 were Liberian flagged and over 300 Panamanian.
More deadweight tonnage, by over a third, flew the single star flag of Liberia
than flew the fifty stars of the USA.35 The league table for current beneficial
ownership in dwt is:

1. Japan 89 million tons


2. Greece 71 million tons
3. USA 65 million tons
4. Hong Kong 45 million tons
5. United Kingdom 33 million tons
6. USSR 29 million tons
7. Norway 24 million tons
8. China 18 million tons
9. The Philippines 15 million tons
10. South Korea and Singapore 13 million tons each

Several countries crowd into the 8-12 million bracket: Italy, West Germany,
France, Spain, Taiwan, India, and Brazil.36
The merchant fleets of the less developed countries are being helped by the
‘UN Convention on a Code of Conduct for Liner Conferences’, which intended
to put trade into ships wearing the flags of the states from whose ports the
cargoes originate. This is one part of a general pattern of political intervention in
the merchant shipping business, especially the liner trades. According to one
British study, up to 20 per cent of the world’s liner trade is affected by cargo
reservation of one kind or another, about 10 per cent or slightly over of world
trade in general. Other ways of helping national shipping are by preferential
berth assignment and demanding non-national ships to possess especially
elaborate documentation. The United States does not live up to its free market
ideology in its merchant shipping policy with cargo-sharing arrangements with
Latin American countries and the forcing of all trade between American ports
into American ships. The USA also pays considerable operating subsidies to its
shipping lines. The Socialist countries, notably the USSR, do not, of course,
work to Western notions of profit and loss: this allows rate cutting below the
agreed ‘western’ Liner Conference figures although service can also be
correspondingly poor.37
The Socialist countries play an important role in the second, even older, form
of sea use – fishing. This is probably more important today than ever, although
the world fishing industry has changed greatly in structure over the last decade
with the shutting out of the Western European fleets from their traditional fishing
areas in the Atlantic that now form part of the other nations’ Exclusive Economic
Zones (EEZs); 95 per cent of the world’s catch is taken within 200 miles of the
shore.38 Although these Western nations still own substantial fishing grounds of
their own they also import substantial, perhaps greater, amounts of fish from
other countries, often straight from foreign fishing boats. The United Kingdom
caught 832,500 tonnes of fish in 1985 of which it exported 304,100 tonnes; UK
fish imports, however, totalled 891,200 tonnes.39
Fish provides 25 per cent of the world’s supply of animal protein and is
especially important to the less developed countries. In these nations it provides
the only supplement to rice or maize. ‘If countries are ranked by reliance on
animal protein derived from fish, 39 of the first 40 places are occupied by
developing countries.’40 Among the developed states both Japan and the USSR
rely heavily on fish for food. Both countries have for various reasons a deficient
agricultural base and need protein from the sea. The Japanese diet obtains half its
protein from fish and the Soviet Union is trying to stimulate domestic fish
consumption to 401b per head per annum.41 Fishing is also important, especially
in the Soviet Union, as animal feed and fertilizer.
Fishing is carried out in relatively restricted areas of the world’s oceans.
Almost all the world’s catch comes from about a third of the world’s sea area.
Most fish (about 75 per cent) comes from the Sub Arctic and Continental Shelf
areas of the North Atlantic and North Pacific, and the waters to the west of
Africa and the Americas, although other significant areas are the South Atlantic,
the northern Indian Ocean, the seas around South East Asia, and the area
between the Philippines through Indonesia to Northern Australia.42 Given their
state of economic development and their demand for fish, it is not surprising that
the greatest fishing nations of the world are Japan and the USSR, which caught
11.4 million and 10.6 million tonnes respectively in 1984. Both use fish factory
ships to give their fleets the capacity to range over the world’s oceans, the
Soviets having to make the greater effort due to their more restricted sea access.
Agreements with distant states provide both access and facilities, although in
certain areas fish has to be purchased from local catchers. Both the USSR and
Japan have explored new types of catch, for example the Arctic krill, used as a
form of animal feed, although finding markets for this product has not proved as
attractive as originally hoped.43
Smaller countries have invested in distant deep-sea fishing industries too.
South Korea and Taiwan follow Japan’s model. South Korea has a substantial
catch of 2.65 million tonnes gained from the South West and Eastern Pacific and
Indian Ocean, as well as waters closer to home. Communist states with domestic
difficulties of agricultural productivity find distant water fishing attractive;
Cubans, Poles, Bulgarians, and East Germans are to be found in the South
Atlantic and round Cape Horn in the South Eastern Pacific. In the opposite
direction, China, already the world’s largest producer of fresh water fish (almost
a million tonnes) has also begun to spread over the Pacific and round Cape Horn.
Her sea catch totalled 6.78 million tonnes in 1985, against 4.77 million tonnes
for the United States. Other countries with substantial sea fisheries of over 2
million tonnes per annum in 1985 were: Chile (4.8 million tonnes – and the
world’s greatest fish exporter, 1.3 million tonnes), Peru (4.2 million), India (2.8
million), Norway (2.1 million), Thailand (2.1 million), and Indonesia (2.07
million). Between one and two million tonnes were the Philippines, Denmark,
North Korea, Iceland, Spain, Canada, and Mexico. Brazil and Ecuador were both
getting close to the million mark.44
Although the post-1945 era saw the growth of ‘industrial fishing’ using large
self-contained fish factory ships, the legal restructuring of the industry to allow
local states greater control of their coastal waters has restored the balance in
favour of smaller craft. Now, however, the large vessel is acquiring new uses.
Previously industrial fishing was largely to obtain fish meal for agriculture. In
the 1980s, however, fish processing techniques are making possible human
consumption of otherwise unacceptable species. Pioneered by the Japanese, this
technique of producing what is effectively synthesized shellfish has led to a
resurgence in large trawlers capable of taking in 400 tonnes of various fish at a
time and freezing it. Fish processing will, it is hoped, take the pressure off some
more traditionally acceptable species and allow the exploitation of abundant
species of ocean fish that have not previously been considered good for anything
but animal feed and fertilizer. By such means the world’s annual fish catch may
be more able to keep up with population increases without over stressing the
oceans’ fish stocks.45
One form of ‘fishing’ which may become extinct because of environmentalist
pressures is whaling. Since October 1985 there has been an official ban on
commercial whaling, although Japan, the USSR, South Korea, Norway and
Iceland all continued to catch whales. Japan stopped commercial whaling in
1987 and Norway announced that she would stop commercial whaling by 1988.
But both countries intend to continue with whale catching for ‘scientific research
purposes’, as does South Korea. Sea-dependent Iceland is the major exploiter of
this loophole. Having officially stopped ‘commercial’ whaling in 1985, about
120 whales were caught in 1986 in a ‘research programme’ that looked very like
‘a front for continued commercial exploitation’; 1987 saw economic pressure
from the USA that led to a deal allowing a limited catch on condition Iceland
clears further whaling with the International Whaling Commission.46 It may take
some time for the final remnants of this traditional form of maritime hunting to
disappear, although it is now of very marginal economic importance.47
As well as animal products the sea now provides substantial quantities of
minerals, notably oil and gas. Offshore hydrocarbons have transformed the
economies of even well-developed industrial states. By 1980, when offshore oil
production accounted for 20 per cent of total world output, the United Kingdom,
totally dependent on foreign oil supplies ten years before, was producing 1.65
million barrels a day, second only to Saudi Arabia in offshore oil production.48
In 1985 the United Kingdom produced almost 50 per cent more crude oil than
she consumed in 1986; petroleum and petroleum products accounted for over 11
per cent of the UK’s exports, although her refineries still required significant
amounts of overseas oil for an optimum range of products. Indeed, a surprising 5
per cent of total British imports remained petroleum and its products.49
Nevertheless, North Sea Oil has been a very significant cushion for the decline
of the UK’s traditional industrial base.
As well as oil, gas for fuel has become a major product of the sea. In Western
Europe the United Kingdom, the world’s largest offshore gas producer outside
the USA, has replaced traditional coal gas completely with the products of the
North Sea. Other major gas producers are Norway, Abu Dhabi (UAE), the
USSR, and The Netherlands.50
The level of demand and price will dictate the extent of oceanic hydrocarbon
extraction in the future. The persistence of the boom has been more fragile than
once thought, and even in the longer term alternative energy sources (e.g. coal)
may well prove cheaper and more convenient. Similar factors will also govern
the viability of the maritime extraction of non-fuel minerals. Dredging of useful
minerals in coastal areas has a long history, as does mining out from the coast.
Oil bearing silts and brines have been found in the Red Sea but their exploitation
is expensive. The geographically limited supplies of manganese may, however,
put a premium on the collection of manganese nodules from the deep ocean
floor. The technology is becoming available but the profitability of the operation
remains questionable, and debate over the ownership of seabed resources
militates against confident investment. The United States currently argues that
the interests of its technologically advanced deep-sea extraction industries are
best served by refusing to accede to the current law of the sea treaty. It fears the
assignment to an international authority of valuable resources to which it would
be convenient for the United States to have access, and which it is perhaps the
only country in a position to exploit. An international authority would reflect the
interests of existing land producers of minerals in the Third World. One suspects,
however, that the legal uncertainties involved in this particular American
exercise in unilateralism will hinder the expansion of marine mineral extraction
more than it will help it. It may also find that unexpected competition from
international consortia, possibly joining developed with less developed states,
makes a totally free market less attractive.51
It can be seen from the above, therefore, that the sea remains a crucial
element in the world economy. It remains its main form of international transport
and it is a major source of food. It provides significant quantities of energy and
useful reserves of minerals. The extent to which these economic forms of sea use
may cause conflict will be addressed in later chapters. Now we must move to
examine the ways the sea can be used as a military medium in conflicts caused
both at sea or, more usually, on land.
Chapter 4

The military medium

The various economic uses of the sea provide mankind with assets worth
defending and fighting about. Nevertheless, the oceans also provide an important
medium for the application of military power, if not for its own sake, then to
achieve political objectives that might have little connection with the sea itself.
The seas provide the most easily crossed two-thirds of the earth’s surface and
man has always used them as a way of gaining access to his opponent. With the
development of new kinds of longer-ranged weapons systems the use of the seas
to ‘project power’ has become even more important than it has ever been.
Indeed, as we have seen, this rather than the protection of economic sea use is
probably now the most important function of naval forces in the late twentieth
century.
As Mahan memorably put it, the sea remains ‘a wide common, over which
men may pass in all directions’.1 The growing moves to enclose the sea for
economic purposes have so far not limited very significantly the capability of
nations to use the high seas to pass their military forces across the globe. They
must still use the sea if they wish to apply military power against distant states.
Despite the advent of large transport aircraft of the American C-141 and Soviet
Candid categories with payloads in the 40 tonne class, and the even larger
aircraft in the 80-125 tonne class, the US C-5 and the Soviet Cock and Condor,
the capacity of aircraft to move large amounts of heavy equipment remains quite
limited. One of the United States’ Military Sealift Command’s Algol class, 30
knot container ships can lift over 23,000 tonnes.2 Only if equipment is
prepositioned can the superior speed of aircraft be exploited to move personnel
to be reunited with it. Otherwise aircraft have to make so many journeys that air
movement is not as rapid as it might at first seem. As Joel Sokolsky has pointed
out in the context of US airlifted reinforcements in a European war, ‘. . . without
prepositioning, it would take nearly 70 days to move five mechanised divisions
to Europe after initiation of mobilisation. Sealifted reinforcements, with
equipment, could begin arriving in European ports after fourteen days.’
Moreover, ‘without prepositioning, additional airlift capacity hardly improves
delivery times because of the difficulty of organising an airlift of equipment as
compared to sealift.’3
Sea transport’s efficiency in moving heavy and bulky materials extends also
to supplies and the long logistical tail required by modern armed forces. No
military operation can be sustained for long without large amounts of supplies,
and modern armies are more voracious than ever in their consumption. In these
circumstances there is often no alternative to seaborne transport over long
distances. As the Chairman of the Joint Chiefs of Staff has put it: ‘In any major
overseas deployment sealift will deliver about 95 per cent of all dry cargo and 99
per cent of all petroleum products.’4 In 1966 the Vietnam War required 20
million ‘measurement tons’ of sealift per annum to sustain operations ashore.5
For countries like the United States the key word is ‘overseas’. Geography is,
of course, crucial to the military importance of the sea. If a state is surrounded
by land frontiers and has no claims to great power privileges or responsibilities,
then the sea is of little or no importance to its military posture. The opposite is
the case when a state is an island, of whatever size. Not for nothing did Mahan
put geographical position at the head of his list of ‘conditions affecting the sea
power of nations’.6 The United States, as a semi-island continent, must go to sea
if she is to exert her military power against states beyond Canada and Mexico.
For the Soviet Union the situation is more complex. Surrounded by what she
perceives to be threats beyond her land frontiers her fundamental strategic
requirements are land and air forces. Indeed, if she wishes to project power, air
forces and airborne formations are of greater relative importance. It is no
coincidence that Soviet airborne forces outnumber similar American forces by
about 1.5:1 in men and 3.5:1 in divisions, and American Marines outnumber
Soviet naval infantry by 11:1 in personnel and 3:1 in divisions.7 Soviet airborne
forces also tend to have more specialized heavy equipment than their American
counterparts. On the other hand, US amphibious capabilities, both as regards
specialized shipping and forces for landing, are unmatched anywhere else.
Nevertheless, as Admiral Gorshkov repeatedly pointed out, if the Soviet Union
is to exert world-wide military power there is no alternative than to use the seas
to do it.8
Many other states must go to sea to project military power. The United
Kingdom, Mahan’s archetypal imperial sea power, may have lost many of her
world-wide military commitments but she has to cross the sea even to maintain
her land and air commitment to the continent of Europe, let alone defend herself
from the maritime threat to the North. Other complete ‘island’ states, like Japan
and Australia, must see the sea as being of especial importance as a two-way
channel for the projection of military power. Both, like Britain, were threatened
with seaborne invasion in World War II. Such nations as these (and their smaller
counterparts like the West Indian and South Pacific islands) lie at one extreme of
geographical strategic sea dependence. Other states are less open to maritime
attack but most have some coastline that gives them both naval opportunities and
vulnerabilities.
A coastline is always ambivalent strategically. It can be a source of access
both for the owner and for a stronger state’s maritime offensive forces. Not for
nothing were small islands the first and last colonial possessions. Given the
traditional dominance of maritime trade, a remarkable amount of the world’s
population and economic activity is carried out not far from the sea: more than
two-thirds of the world’s inhabitants live within 300 km of the coast.9 These
people are potentially vulnerable to maritime military power, i.e. sea power, of
some kind.
Not only does the sea provide simple access, it also does it in ways that are
militarily advantageous. The very extent of the sea provides opportunities for
concealment and surprise. Although the revolution in surveillance that has
resulted from the development of space-based platforms is undermining this
advantage, it is still remarkably easy for maritime forces to lose themselves to
view in the world’s oceans. The main requirement is that they do not give away
their position by emitting electronically or acoustically.
It is this combination of mobility and surprise that gives amphibious
operations much of their continuing validity. As discussed in Chapter 2,
amphibious warfare has been consistently written off by military commentators,
not least after the Gallipoli fiasco of 1915-16. As with all particular historical
contexts, the factors that militated against this operation were to a considerable
extent scenario dependent. At the outset,

there was little choice in method of deployment, owing to the mountainous


terrain, the restricted beaches and the generally improvised nature of the
opposition. Limited movement of some formations between landing sites was
possible and was done, but there could be no shifting internally of the main
weight, for most of the rifles ashore were sorely needed to hold the existing
ring, the overall tactical picture could only be affected by the entry at a
selected point of major reinforcements from outside the theatre. And these
were by no means readily available.10
Execrable operational leadership in the later landings, and an overall commander
insufficiently ruthless and too tolerant to deal properly with such incompetence,
made failure certain.11
In the inter-war period the US Marine Corps, faced with the need to prepare
plans for operations against Japan, examined the lessons of Gallipoli and
identified where the failures had occurred in command and control, equipment,
communications, fire support, and logistics.12 Measures were taken to solve the
various problems. Doctrine was developed and the Fleet Marine Force
established. Thus was the background laid for the Pacific counter-offensive of
World War II.
Japan, with her amphibious experiences of the War of 1904-5 (which had
stimulated Corbett to argue that there was still a place for amphibious warfare in
the strategic repertoire), also developed amphibious warfare capabilities and
doctrines in the inter-war period which allowed her to take her Great East Asia
Co-Prosperity Sphere in the space of a few weeks in 1941-2.13 Even Britain,
understandably the most affected by the Gallipoli fiasco, did not neglect
‘combined operations’ completely. As one study has concluded: ‘the fact is that
there existed a doctrine, known only to a handful but nevertheless something on
paper, a training and development centre with insufficient personnel and funds
but in-being, and a sub-committee of government to put existing doctrine into
effect when needed.’14
Thus were the foundations laid for amphibious operations that began with the
commando raids of 1941 and culminated with the reoccupation of Western
Europe. In the latter, air power was a vital component of the amphibious
offensive. As Admiral Schofield has put it with reference to the Normandy
landings of June 1944,

the success or failure of the operation depended largely on the rate of build
up of the Allied force in the bridgehead exceeding that of the enemy. The
Germans possessed interior lines of communication in the form of a network
of roads and railways which, unless interdicted, would undoubtedly give
them a great advantage in this respect over the sea-borne lines of
communication on which the Allies had to depend.15

The victory of the Allied air forces, the destruction of the Luftwaffe and the
dropping of 76,200 tons of bombs on 80 road and rail targets successfully
neutralized much of the Wehrmacht’s land power and allowed the Allies to use
the sea to put ashore a huge army and its supplies. Between June 6 and June 15,
1944, half a million men and 77,000 vehicles were landed.16
If such operations were practical, small-scale operations supported by integral
air forces were even more so. World War II was for the United States, and even
for the United Kingdom, essentially an exercise in the projection of military
power by sea, it might even be argued essentially a ‘maritime strategy’. Captain
Stephen Roskill, the official British naval historian of the war, tended to argue
thus: ‘Our total expulsion from the European continent in 1940 forced us to
change from a continental to a predominantly maritime strategy: and the reason
for the change was simply that no other means of achieving victory then
remained to us.’17
Roskill concluded that ‘the experiences of the last war seem to suggest that,
provided all the basic requirements for success in amphibious enterprises are
met, they can bring as great benefits as were derived from them in earlier
conflicts – both in the field of minor strategy (through small scale raids) and in
that of major strategy (by large overseas expeditions)’. ‘However’, he went on,
‘the hazards of such undertakings remain as high as ever.’18 These have been
well summarized by two American analysts. ‘Amphibious assault is perhaps the
most intricate and difficult of all tactical operations. Preparation for it requires
both the creation of peculiar force structures and tactical doctrines, and the
development of specialised and often costly weapons and equipment.’19
The US Marine Corps has played the leading role in developing these
structures and doctrines, its institutional momentum and bureaucratic self-
interest helping overcome initial US Army and Air Force, and indeed Navy,
doubts about the continued relevance of the amphibious mission post-1945. The
Inchon landings of 1950 helped vindicate the amphibious role once more, albeit
that this was a relatively small-scale operation carried out by less than one
division. Nevertheless, it set the scene for the post-war era, the tendency for
amphibious operations to be rather limited and often, unlike Inchon,
unopposed.20
The British record was similar, but anti-amphibious post-war reaction was
more extensive, given the less favourable financial and bureaucratic
environment. Despite this, operationally the Suez landings went remarkably well
and their lessons were constructively digested afterwards. Using American
precedents, and what equipment that could be afforded, a significant British
amphibious capability was built up after 1957 for post-colonial policing
operations. This was still largely intact in 1982 to retake the Falklands by sea.21
The US Marine Corps has used its post-war experience to continue to
legitimize a continued investment in amphibious forces that has been massive by
the standards of the rest of the world. The officers who man this force have also
taken the lead in refining concepts of amphibious operations. The potential of
new technologies such as hovercraft landing vessels and tilt rotor aircraft has
fostered some radical and interesting thinking on how in the future the sea can be
more fully exploited as a military medium. As one young officer put it recently:

The operational significance of coastal waters has never been fully


appreciated. Unlike inland terrain, with its hills, streams, forests, and various
other obstacles, the ocean is relatively flat, even in weather conditions that
often slow or stop land campaigns, offering amphibious forces a plain on
which to conduct initial operations. The advantages offered by this plain can
be exploited using new landing tactics based on multiple landing points and
rapid shifting of forces. Instead of relatively static and predictable broad
landing beaches currently used, much narrower landing points of no more
than tens of yards width offer an opportunity to seek out enemy weaknesses.
By landing his forces across multiple landing points perhaps in waves of
companies, a commander retains the ability to develop situations while
committing minimal forces. If successful, initial landing forces can be
immediately reinforced by uncommitted units, if not, they can be quickly
withdrawn and shifted to more successful landings.22

The landing of forces from the sea remains an operation of war with potential in
many circumstances. Forces found at sea may be sunk before landing by
relatively few weapons. Modern mines make the landings themselves more
hazardous. On some occasions, superior land and air communications may be
able to bring up a rapid military response that can contain and destroy
amphibious incursions. Nevertheless, a proper exploitation of the assailant’s
advantages makes the result often depend on marginal factors that vary from one
case to another. The sea gives much potential for surprise and if the enemy lacks
the benefits of a shore transport infrastructure, especially likely on islands or in
remote areas, the assailant may actually be better able to support and sustain
operations than the defender. Amphibious capabilities, therefore, remain as
useful and as challenging as ever.
Amphibious landings, of course, affect primarily littoral areas, as do shore
bombardments. The range of bombardment has, of course, greatly increased in
the last half century until no significant part of the world’s land mass is now
beyond the range of naval forces. This is fundamentally a commentary on the
capability of sea transport to move bulky objects, like long-range weapons
systems, more easily than other means. The combination of first the long-range
heavy attack aircraft and the aircraft carrier and then, more importantly, the
intermediate and later the intercontinental ballistic missile with the submarine,
especially the nuclear-powered submarine, has had a fundamental effect on the
world’s strategic situation. The condition of ‘mutual assured destruction’ that
provides the strategic underpinning of the modern world is a direct result of the
utility of the sea as a medium for the deployment of invulnerable launch
platforms.
Once one goes beneath the surface of the sea one enters a murky ultra-
complex world where electromagnetic radiations, the main media of both optical
and electronic sensors, cannot penetrate easily. Water is a relatively good
transmitter of sound waves but even here the uncertainty and unreliability of the
sea come more to the aid of the hider than the seeker. The molecules of the water
itself absorb sound waves by converting them to heat: the higher the frequency
the higher the loss. The sea is also teeming with animal life of all shapes and
sizes which both scatter and absorb acoustic radiations and produce a cacophony
of sounds of their own. Man-made noises, both ships and drilling operations, add
to the noise level. Inorganic matter and bubbles contribute further to the
confusion, while the nature and shape of the ocean floor varies enormously, all
with serious effect on the acoustic environment.
Although the sea’s salinity is usually constant and its pressure increases with
depth, it is far from uniform in temperature. There are at least three main
temperature ‘layers’. The lowest in both depth and temperature is ‘isothermic’,
roughly constant in temperature at between about 3° and 5°C. Coming up at well
over 1,000 metres below the surface one then enters the ‘thermocline’ layer,
where the water increases in temperature over 1,000 metres or so from about 5°C
to an average of about 20°C. Then there is a surface layer up to a few hundred
metres’ depth which is again more or less isothermic. Of course, actual surface
temperatures vary with time of day, latitude and season from below zero to
above 30°C, depending on the area of the world. The boundaries between these
layers create an unpredictable acoustic environment, sometimes helping sound
propagation along ducts between layers, sometimes blocking off one layer from
another. Sound can penetrate from one layer to the next if it is travelling steeply
enough but, as might be expected, it suffers both attenuation and refraction.
Once in the lowest layer, or deep sound channel, however, sound may go easily
for thousands of miles.23
As well as these basic horizontal boundaries, which, it must again be
emphasized, vary from ocean to ocean, one has vertical discontinuities caused by
currents, eddies, fronts, and underwater mountains and valleys. The effect is to
create huge shadows in which to disappear from acoustic ‘view’. The interaction
of the temperature layers and the increase in pressure with depth creates an
environment where sound moves up and down in a series of skipping motions.
The deflection of sound in the thermocline can both shield submarine platforms
suitably placed and allow their detection at longer ranges as the sound channel
rises to the surface duct once more, having been deflected upwards by the cool,
higher pressure water of the deep isothermic layer. Sound tends to move down to
the deep sound channel and then upwards again, close to the surface, at distances
of about 50 km from the source, so creating ‘convergence zones’ about 5 km
wide at distances of 50, 100 and 150 km.
With sound at the best of times travelling in these jagged paths and the
manifold barriers to detection provided by the discontinuities and background
noises of the sea, it is hardly surprising that once a submersible platform loses
itself in the oceans it is often hard to find. Despite all these difficulties, acoustics
will remain the basis of finding objects beneath the ocean, although future
detection systems will also probably be developed to exploit submarine-
generated electromagnetic effects. A submarine both creates a ‘magnetohydro-
dynamic’ effect by displacing water molecules across the earth’s magnetic field
and a galvanic current resulting from the electrical potential between the
submarine’s brass propeller and ferrous hull. These extremely low frequency
(ELF) effects can be detected using magnetometers. Current resulting from
‘underwater electrical potential’ (UEP) is detectable over the longest ranges
(measured in miles); magnetohydrodynamic extremely low frequency effects are
probably the more persistent as they interact with physical hydrodynamic
displacement effects, notably ‘internal waves’. The latter can also cause ripple
effects on the surface detectable by synthetic aperture radar. Submarines can also
cause detectable ‘humps’ on the surface by physical displacement, temperature
variations, and even ‘bioluminescence’ among marine life. Leaving aside the
question of active decoying, all of a submarine’s effects are also produced by
natural means, winds, natural currents, whales, etc. Long-range, high certainty
area submarine surveillance is thus still a long way off, even if supplementary
non-acoustic forms of tactical and operational detection are becoming available.
UEP mines are reportedly already deployed by the Soviet Union. Magnetic
anomaly detection (MAD) of the distortions caused by a submarine in the earth’s
magnetic field has been useful for short-range ASW since the World War II
years, and could be developed further.24
The key to all this is that if the sea was a uniform static medium it would be
relatively easy to detect a number of indicators that objects were present.
However, it is neither uniform nor static. Only the most sophisticated computer
analysis allows the natural phenomenon to be distinguished from the potential
target and most, if not all, non-acoustic indicators produce insufficient
information Upon which computers can even base their analysis. Sonar applied
to modern signal processing techniques does provide some transparency to the
seas, but it is a variable and uncertain transparency even now. There is also
always the possibility of deliberate deception. The sea will for long remain the
best place in which to hide and will thus remain the world’s most important
medium for the deployment of those forces upon whose invulnerability strategic
stability rests.
It is not, however, just the physical characteristics of the oceans that make
them very attractive as a military medium, both in normal times and in crisis.
The interaction of the sea’s ease of movement, especially of stealthy movement,
with its international status as the ‘great common’ gives it unique advantages as
a medium for the deployment of military power, especially in nuanced situations
short of open conflict. Naval power projection forces can be kept out of sight or
deliberately displayed. They can be manoeuvred in threatening or non-
threatening ways. They can be sent to cruise in international waters in or close to
a potential crisis area without raising all the problems of acceptability and
commitment that a deployment of land forces or land-based air forces would
entail. Naval forces can bring air and missile power to bear without the need to
negotiate overflying rights and/or engage in difficult and elaborate airborne
refuelling exercises. Usually, proximity to the target also facilitates greater
accuracy and discrimination in the delivery of firepower. Forward presence can
be maintained for long periods if required. Even amphibious forces can be kept
poised at sea for weeks on end, giving options for intervention that can be
unveiled or utilized as and when required. If circumstances do not develop in the
war feared, perhaps because of the maritime deployment or perhaps despite of it,
perhaps for other reasons, those forces can be withdrawn without too much
diplomatic penalty. Maritime forces also lend themselves to quick ‘in–out’
operations, such as those required to evacuate nationals from dangerous
situations.
It is in the area of providing contingency forces for decision makers that sea
power really lives up to its reputation for silence of operation. In 1967 the
British commando carrier (LPH) HMS Albion disappeared into the Atlantic for
many weeks with her embarked commando, artillery battery and helicopter
squadron. This gave Britain the option to take limited action in the opening days
of the Nigerian Civil War, perhaps to create a neutral zone if requested, perhaps
to act more unilaterally to evacuate civilians. The ship’s absence from scheduled
exercises was covered by the Arab–Israeli crisis in the Middle East that broke
about the same time.25
Even more covert by its very nature was the despatch of the nuclear-powered
submarine HMS Dreadnought to the vicinity of the Falkland Islands at the end
of 1977, to be unveiled should a threatening Argentina be thought to be about to
invade. The aim was to do this without ‘provocation’ and the two frigates sent to
provide vital communications links were kept as far away as possible. There is
some debate as to whether Argentina was informed of these deployments at
some stage, but the main aim seems to have been not to force Buenos Aires to
back down but to provide a means of setting up a defensive ‘exclusion zone’
immediately should the situation warrant it. It is interesting to speculate what
would have happened in 1982 if the Thatcher Government had shown the same
foresight as that of its Labour predecessor.26
Submarines by their very nature are usually less useful than surface ships if
overt presence is required. Even in the above case the only way HMS
Dreadnought could have proved her presence would have been by sinking
something. They are, however, useful in other covert roles such as intelligence
gathering. Both superpowers and others exploit submarines to enter each other’s
sovereign waters, to gather information on each other’s sensitive installations
and assets on, above, and below the waves, and to plug into communications
cables to monitor sensitive traffic. Incidents often occur, such as collisions and
even submarines running aground, as happened in the Soviet ‘Whiskey on the
rocks’ incident in Sweden. Except in such blatant cases, however, things can
easily be kept quiet to both sides’ benefit.27
Given the extent of the high seas, however, surface vessels have a key role to
play in intelligence gathering and analysis. The platforms used include warships,
merchant ships and specialized auxiliaries (AGIs). Jane’s Fighting Ships lists no
less than sixty-three specialized intelligence collection vessels in Soviet service.
Most are based on trawler hulls but there are also vessels in the 3,000-4,000 ton
class.28 The United States has no ships actually designated ‘AGI’ but its use of
specialized surveillance vessels, like the ill-fated Liberty and Pueblo, is well
known. Perhaps America’s most notable current AGI is the ‘missile range
instrumentation ship’ Observation Island that monitors Soviet missile tests and
electronic emissions in the Northern Pacific.29 Surveillance, especially by
surface ships, can be an important tool of crisis management, having the double
effect of demonstrating an interest in the area in question and the forces under
surveillance, and providing an accurate monitor of the other side’s activities. The
actions of surveillance platforms can also be used to show resolve and exert
pressure.
Those states likely to be the victims of these uses of maritime power may
resent the intrusion, especially if they are unable to take appropriate and
effective countermeasures. The 1986 UN Study, ‘The Naval Arms Race’,
articulated concerns that are widely held in the Third World.

When employed on normal deployments as part of national peacetime tasks,


activities by world-wide and blue-water navies outside their own territorial
and regional areas can become a significant political factor in regional and
local situations. . . . Many regional states may become concerned at the
implications of such deployments for regional security and strongly dislike
the implied threat, real or perceived, of external intervention in the regional
or internal affairs of other states.30

It called for ‘the exercise of moderation and restraint on the part of all, . . . if
further threats to security are to be contained’.31
Happily, the nature of maritime forces, working in a neutral medium under
tight national control (a control that is becoming tighter all the time), lends itself
to the exercise of moderation and restraint. Although the very presence of
warships may act as a provocation to an opponent or a too readily available
resort to force, more usually the actors in a particular crisis exploit the
opportunities for limitation and discrimination in response. Tight ‘rules of
engagement’ (ROEs) are usually laid down to regulate the way weapons can be
used and platforms manoeuvred, and the circumstances in which actions can be
carried out or threatened. They are especially important at sea, where there are
not the clear demarcations provided ashore by national boundaries. ROEs can be
manipulated and sometimes ignored by commanders on the spot, as when in the
Falklands War the British commander tried to unleash his submarine in contact
with the cruiser General Belgrano. The signal was intercepted and
countermanded, but the unauthorized action became an exercise in effective
manipulation as it demonstrated all too effectively to London the gravity of the
situation. The ROEs were, therefore, changed to allow attacks on all Argentine
warships outside territorial waters.32 The natural dynamics of naval
confrontation can sometimes force escalation despite attempts by the actors in
the crisis to keep the lid on the situation; when ships are sunk the symbolic sense
of loss can be great and the casualties can be heavy.
These escalatory pressures must be borne in mind in analysing ideas for
limited conflicts at sea between major powers. There are some precedents, such
as the ‘Phoney’ period of World War II over the winter of 1939-40. This period
of restraint, advised by the conjoint fear of the air raid and renewed trench
warfare, seemed of particular relevance when the nuclear age made the price of
total war unacceptable. A conflict limited to the oceans might, in the words of
Professor Martin, be ‘more likely to remain within the bounds of a regulated
policy contest than one that wrought destruction on homelands and civilians
even only as the collateral effect of a would-be intra-military contest’.33 During
the Berlin crisis of 1959-61, the United States studied various graduated
responses against Soviet, DDR or other Warsaw Treaty Organization (WTO)
shipping ‘which would be preferable to directly challenging the Soviets along
the NATO frontiers because it carried less risk of escalation to nuclear war’.34
By 1967 there was an Atlantic ‘War at Sea Plan’ which was ‘specifically
designed to apply a measure of force in support of a national policy of
determined persuasion, or alternatively may be implemented as a controlled
response to Soviet military aggression short of strategic nuclear attack’. Using
well-defined ROEs the US Navy would have a number of options ‘to employ
military action against the Soviet Union under circumstances in which
implementation of the general war plan is neither a rational nor credible
threat’.35
This kind of planning exploited the very ‘secondary strategic value of the sea
to the Soviets. If Russia did not depend heavily on the seas to support its
economy or its military power in Central Europe then striking at Soviet vessels
would not elicit a massive retaliation by the Soviets.’36 Since the late 1960s
Soviet dependence on the sea has increased further, but it remains less than the
West’s. This might, however, be held to advise against the West, which has more
at stake, deliberately exposing a vulnerable flank to Soviet pressure. Turning the
above argument on its head it is sometimes argued that the Soviets themselves
might begin a maritime offensive against Western shipping. The graduated
nature of the pressure might allow time for negotiation. However, a number of
problems immediately present themselves to such a plan. The international
nature of modern shipping would force the attacker to take on many countries,
not just NATO. Given the use made by the West of land-based sensor arrays and
aircraft, it would be exceptionally difficult for land operations to be dispensed
with entirely. Moreover, the balance of maritime forces is such that Western
maritime power might be projected in several ways that the USSR would find
unwelcome, especially against its more far-flung bases at home and overseas.
Even if the maritime inderdiction campaign was a success, an increasingly
desperate West might well consider such ‘horizontal escalation’, perhaps even
increasing the intensity of conflict; the results of a successful blockade can be
anything but ‘limited’ in principle or practice.
Despite all its apparent advantages as a neutral place d’armes, one must be
careful not to overstate the utility of the sea as a self-sufficient military medium.
It is the very interaction of the sea and the land, where, as Corbett pointed out,
people actually live, that gives sea power its fundamental strategic salience both
now and in the future.37 Maritime coercive power is exerted in a remarkable
number of different ways and from an equally large variety of mobile platforms:
a very wide range that helps give sea power its inherent flexibility and
controllability, both now and in the future. As we shall see in Part III, every
contingency is covered from mild constabulary pressure to all-out nuclear
holocaust.
Part III

The changing shape of naval war


Chapter 5

Weapon systems

This part of the book addresses itself entirely to the current state of naval
technology and the paths of its likely future development. First, we look at the
weapons, together with the sensor, information, and control systems that are the
keys to their effectiveness. As weapons systems are means to ends not ends in
themselves it is probably best to examine them in the context of their functions
in the various dimensions of naval operations: anti-surface ship warfare
(ASUW), anti-submarine warfare (ASW), anti-air warfare (AAW), and
operations against the shore. The special question of nuclear weapons will be
dealt with last.

Anti-surface warfare
Anti-surface warfare is in a sense the easiest of naval tasks, given the
vulnerability of the target to damage above and under the water. A range of
technologies are available for this role: guns, missiles, and torpedoes. From the
middle of the sixteenth century to the beginning of the twentieth, guns were the
most important instruments of bringing force to bear against an enemy’s surface
ships – although for a surprisingly long period they were more a means of a
‘softening up’ for a human assault or obtaining a surrender than a very efficient
means of exerting decisive damage on the enemy’s fabric or watertight integrity.
Only with the advent of the long-range shell firing gun in the middle of the
nineteenth century did artillery become capable of decisive results on its own.
But its reign was soon challenged by underwater weapons such as effective
mines and torpedoes, which were born at about the same time. Nevertheless,
guns retained a considerable range advantage and, directed with increasing
accuracy by some of the world’s pioneer analogue computers, remained the
dominant naval weapon until well into the twentieth century. Only in the early
1940s did they have to yield supremacy to air-delivered bombs and torpedoes as
the most effective way of neutralizing an enemy’s major naval assets. They had,
however, been replaced by the torpedo as the main means of sinking
merchantmen a quarter of a century before.
One does not, however, necessarily always want to sink an opponent,
especially in conditions short of all-out conflict. At the time of the Cuban missile
crisis in 1962 it dawned on the US Navy that their latest all missile Leahy class
frigates (now classified as cruisers) could not fire a shot across the bows of
Soviet merchantmen: 76 mm (3 inch) guns were soon added, and the design of
the follow-on Belknap class was altered to include a 127 mm (5 inch) weapon.
At about the same time the Royal Navy was finding that 114 mm (4.5 inch) guns
were a somewhat inappropriate answer to small Indonesian sampans; old 20 mm
guns of World War II vintage soon reappeared. These lessons retain their
relevance. A weapon with a relatively rudimentary fire control that can
deliberately miss the target as well as inflict precise and limited damage upon it
will remain a warship’s minimum necessary armament. Tube artillery in the 20
mm to 76 mm range would seem the obvious answer.
Guns, however, will remain useful for more demanding ASUW roles also.
The combination of rocket assistance and simple infra-red/laser homing
technology is giving artillery the characteristics of range and accuracy of the
shorter ranged anti-ship missiles. Given the capacity of a ship to carry a much
larger number of rounds of gun ammunition than relatively large and bulky
missiles, this would seem to give guns advantages in ship-to-ship engagements
of medium range, e.g. out to 30-40 km. Not only can more targets be engaged
but the capacity to fire a number of projectiles at each individual target increases
the chances of a kill. It will be more difficult to give ‘cannon-launched guided
projectiles’ (CLGP) the autonomy of missiles given the limited payload available
for guidance systems and the robustness required to withstand the shocks of tube
launch. Near-term CLGP will also require some external source of designation,
either from a helicopter or RPV, or possibly from the laser range finders now
being increasingly fitted to surface ships themselves.1
To give CLGP an adequate payload, a relatively large calibre is probably
essential. This will probably cause future gun calibres to edge up slightly from
114-127 mm (4.5-5 inch) to 155-203 mm (6-8 inch). Anything bigger, however,
will cause the gun to dominate the design of a ship to an undesirable and
unnecessary degree, given the need for a future warship to be capable of
operating against underwater and air targets as well as against other ships. Very
heavy calibre artillery, as mounted in the superannuated American battleships,
could notionally deliver large and powerful CLGP to 80 kilometres or so. New
and relatively expensive ships built primarily around super-heavy artillery would
probably be over-specialized and not cost-effective, given available ASUW
alternatives. It is possible, however, that, as proposed in the United States, new
gun-armed ships with up to 203 mm (8 inch) main armament will be built, but
the primary purpose of such vessels will be anti-shore warfare rather than anti-
ship.
Until the Falklands War gave a dramatic demonstration of their capabilities,
torpedoes were falling out of fashion as weapons of ASUW. This was partly
because ships had rarely been targets in post-1945 naval actions. Also, torpedoes
are highly lethal: they cannot be used except actually to sink the enemy. In the
three and a half decades after 1945 only in the Indo-Pakistan War of 1971 were
torpedoes operationally used in conflict by a Pakistani submarine to sink an
Indian frigate. Then in 1982 came the dramatic sinking of the Argentine cruiser
General Belgrano by the British submarine Conqueror, and the less successful
attacks by the Argentine submarine San Luis on British surface combatants. The
weapons used against the Belgrano were classic indeed: Mark 8s, half a century
old in basic design, although perhaps the best torpedoes of their kind. Powered
in this case by hot gas, or sometimes by electricity, such weapons have their
course set before launch and are best fired in salvoes. (Conqueror fired three of
which two hit.) The effect of two 340 kg warheads exploding below the
waterline of even a well-built surface ship like a World War II cruiser was
devastating, but these classic torpedoes have to be released close to the target.
Conqueror was only 1,400 yards away. Achieving this is difficult against an
enemy who is capable in ASW techniques, or indeed one who is manoeuvring
evasively. This is not to say that such torpedoes will not continue to find a place
in naval operations, but their usefulness will be limited to unescorted and
unarmed shipping and/or vessels of lesser states.2
Long-range capability of an indeterminate kind can be obtained by pattern
running, i.e. causing the torpedo to steer a zig-zag course after being fired into a
group of ships. This might still have some utility. It is better, however, to fit the
torpedo with some acoustic homing device to guide it on to its prey. Wire
guidance from the launching platform can be used to guide the torpedo to the
vicinity of the target, as revealed by the launching platform’s sensors, e.g. a
submarine’s sonar. When the torpedo is safely locked on to the target’s noises it
can be left to run its course. This can have its problems: as the Argentinians
found when the guidance wires of their German torpedoes snapped.3 Wire
guidance will still be used in many future ASUW torpedoes and in common
torpedoes for ASW and ASUW use, but the latter may not be the most economic
solution. Surface targets are relatively noisy, and acoustic homing torpedoes
often only need to be fired into the target’s area seeking passively to find them.
Whether wire guidance is used or not, ASUW torpedoes can be much less
sophisticated than their ASW brethren, and the US Navy plans to procure a low-
cost acoustic torpedo (LCAT) for ASUW duty at a cost of only 20 per cent that
of its fully capable ASW cousin.4
Torpedoes are, par excellence, the weapon of the submarine, but small surface
craft have traditionally been used as delivery means also. This continues with,
for example, tubes for wire-guided torpedoes being fitted to supplement the
missiles of West German fast-attack craft. Sweden deploys large wire guided
ASUW torpedoes in shore batteries. The latter are some of the few torpedoes to
use hydrogen peroxide (HTP) propulsion. Batteries are the more common
modern propulsion systems for ASUW weapons, although modern
monopropellent heat engines are making an appearance and will become more
common in the future.5
Anti-surface torpedoes are traditionally 21 inch (533 mm) weapons, although
smaller torpedoes have been deployed, primarily as anti-escort self-defence
systems in submarines. In World War II, the Japanese successfully used long-
range 24 inch weapons to give ranges comparable to gunfire, and such large
torpedoes may be making something of a comeback in an interesting novel form.
It is reported that the Soviet Union is now deploying large Type 65 torpedoes of
650 mm calibre (26 inch) with advanced closed-cycle (probably HTP powered)
propulsion systems. Their reported range is 100 km at 30 knots or 50 km at 50
knots, using ‘wake-homing’ techniques by which sensors look for the turbulence
in the wake of a target and then home in upon its source. This is probably done
acoustically with a small torpedo-mounted sonar sensing the turbulent water in
the ship’s wake. Other more advanced types of wake-homing devices are
conceivable, e.g. ‘a passive optical or pressure/temperature sensitive instrument
or one which measures variations in sea-water electrical conductivity associated
with turbulence’.6 Long-range wake-homers can be fired into the wake of an
enemy’s surface force at some considerable range. They then ‘look for’
indications of wakes and by means of a series of diminishing zig-zag paths direct
themselves towards the source of the effect. Such weapons concentrate on the
vulnerable stern areas of a ship, where the propulsion and steering gear of the
vessel is concentrated. They might not sink a very large enemy ship such as a
carrier but they would be almost certain to put it out of action. A torpedo of this
size might carry a rather large warhead (400 kg) with a large kill radius. It is
interesting that the introduction of such weapons has coincided in the Soviet
Navy with the end of the production of submarines with shorter-range anti-
surface ship missiles; the role has apparently been taken over by SSNs that are
equipped with the new torpedo tubes. The effect of a hit of a single long-range
‘wake-homing’ torpedo on a US carrier would be potentially catastrophic: much
more so than even a couple of missile hits.
More effort, therefore, may have to be given to torpedo countermeasures.
Major navies already have acoustic decoys, like the US ‘Nixie’ or the British
Type 182, that can be streamed behind a ship to divert homing torpedoes. These
systems are already inadequate against later torpedoes. Under development are
more advanced reactive decoy systems, such as the British ‘Expendable
Intelligent Decoy’, that can be dropped into the water either to use noise
jamming to distract passive homers or to seduce active homing torpedoes by
retransmitting the torpedoes’ transmissions in such a form that the enemy
weapon homes in on the wrong spot. The French are exploring a reactive
expendable decoy system to be applied first to their new carrier. ‘Pyro-acoustic’
countermeasures are being used, pyrotechnics to create noise or gas generators to
create bubbles to ‘mask’ the ship. Such systems may help deal with modern
acoustic homers, although the advent of formidable ‘wake-homers’ may make
anti-torpedo torpedoes, as suggested recently by one officer, the only answer.
This could be an expensive and complex option.7 Another potential defence
against this underwater threat is the use of short-range ASW rocket launchers as
anti-torpedo defences. All large Soviet combatants either carry special RBU-
1000 sextuple launchers on each quarter for this duty or, in the case of the
‘Udaloy’ class ASW cruiser, mount their longer range RBU-6000 weapons, so
they can apparently double in this role. The Soviets deny that this is so but
similar systems might have some anti-torpedo potential in the future.
The most currently fashionable method of attacking surface vessels is to use
anti-ship missiles. This form of naval warfare was pioneered by the Germans
and Japanese in World War II, the latter using manned guidance. It can be
effective, as demonstrated by the Argentines in 1982 when Exocet missiles were
used to sink an important supply ship and a destroyer. The Soviet Union has
developed a range of anti-ship missiles for deployment on a spectrum of
platforms: airborne, submarine, and surface. The smallest of these have scored
successes in the Middle East and Indo-Pakistan conflicts. In general, however,
Soviet missiles, and the Chinese copies of them, have tended to be relatively
large and fly relatively high.8 Many current longer ranged Soviet missiles are
also relatively slow, giving a requirement for mid-course corrections and the
launching platform, if a submarine, having to stay in a vulnerable surfaced
position as it guides its weapons to the target before they lock on. This drawback
is being overcome by newer, more autonomous missiles like the SS-N-19, which
is an advanced weapon capable, if necessary, of submerged launch and with a
range of some 450 km at supersonic speed. Once fired at a target provided by an
external source, e.g. satellite or aircraft, the missile can get quickly to the point
where its radar head can acquire and home in on the target.9
Some Soviet missiles already at sea have the skimming characteristics of
modern Western missiles such as Exocet. This rocket-propelled weapon is an air-
launched missile with a range of about 60 km and a speed of just under Mach 1,
giving a flight time of about two and a half minutes. The launching aircraft,
usually a small fighter-type machine, exposes itself very briefly to radar
detection to programme the weapon’s autopilot. The missile then skims over the
sea at a height of a few metres (dependent on sea state) turning its radar homing
head about 30 seconds from impact to give the target the minimum chance of
taking countermeasures.
Exocet, which has surface and sub-surface launched variants, is in the middle
range of anti-ship missiles. Jet propulsion gives longer range, as with the
American Harpoon and British Sea Eagle, although problems of target
acquisition after a long flight act as a potent constraint on effective range as
opposed to theoretical maximum. This is an especially serious constraint for very
long range anti-ship missiles like the American Tomahawk. The latter’s ability to
find a target covering its advertised range of 400 km at Mach 0.7 has been
doubted by leading analysts. Data links to update the guidance of missiles for an
external surveillance source may be necessary for such weapons to exploit this
kind of range bracket.10
The new ANS (Anti Navires Supersonique) being developed cooperatively by
France and Germany will use ramjet propulsion to enable it to cover its range of
approximately 175 km at no less than Mach 2 at sea level (these speeds have
previously only been achieved by Soviet AS4 and AS6 air-launched missiles,
both of which have a high altitude mission profile). ANS will be an ‘intelligent’
missile, i.e. fitted with sufficient on-board computer capability to enable it more
fully to cope with enemy movements and countermeasures, and even if the latter
are deployed, weapons of its generation may well be able to think their way
through many of the more obvious techniques. The weapon will be easily
reprogrammable as the measure/countermeasure dialectic evolves.11
Aircraft are probably going to remain the most effective anti-ship missile
launchers as they are best able to find and identify targets as well as only having
to remain for the most limited period in the latter’s vicinity. Modern integrated
surveillance techniques using satellites and reconnaissance aircraft, however,
might well provide missiles from surface or even sub-surface platforms with
sufficiently good initial targeting information for long-range engagements. The
introduction of maritime surveillance aircraft fitted with inverse synthetic
aperture radar (ISAR), with its ability to give a high-definition picture to any
receiving platform, allows long-range missiles, such as Tomahawk, to be fully
exploited by surface ships over ranges of hundreds of kilometres. Modern towed
array sonars provide surface ships or submarines with autonomous very long
range initial detection and surveillance capabilities. These do not always work
(e.g. HMS Splendid’s inability to find the Argentines’ carrier battle group in the
Falklands War), but they do greatly extend the detection horizon of the platform
on or in the sea and, if the weapon carried has a sufficiently wide search pattern,
allow an extended attack horizon. Helicopters or drones can be used to give
supplementary surveillance and targeting information to surface ships. The key
to all this, however, remains discriminating real targets from what will probably
be a jumble of contacts.12
Increased missile speed will also make long-range engagements much easier.
An alternative approach to high-speed cruise missiles might be the use of
terminally homing ballistic missiles. For a time in the 1970s the Soviets seemed
to be developing such a weapon specifically for use against ships, the SS-NX-13.
A two-stage rocket launched a manoeuvring post boost vehicle to an altitude of
250 km where its radar sensors searched for the target. Having picked it up and
locked on, the PBV reoriented itself and using two rapid rocket burns propelled
itself in the direction of the target at very high speed. The actual re-entry vehicle
was then released just above the atmosphere to fall on the target. The system
seems to have depended on the use of a nuclear warhead but modern
manoeuvring re-entry vehicle technology could allow it to home in all the way.
The inevitable difficulties of development, coupled with SALT limits on ballistic
missile launchers, caused the abandonment of this system, but it possibly could
be resurrected to formidable effect. The range was between 150 and 1,000 km.13
Over short ranges modern, semi-active, radar-homing surface-to-air missiles
now have quite good secondary surface-to-surface capabilities over ranges of
about 35 km or so, out to the horizon. The main problem is how to decide on the
relative priorities for these valuable weapons, given limited ammunition loads.
At the lowest end of the scale, specialized anti-ship missiles of around 20 km
range will still be useful for use from fast and manoeuvrable platforms, such as
fast-attack vessels and helicopters. The Israelis demonstrated the utility of such
missiles when their fast-attack craft equipped with Gabriel missiles worsted the
SS-N-2 equipped opposite numbers of Egypt and Syria in 1973. The latter’s
missiles were more suitable for engaging larger warships than other fast-attack
craft. Other advanced Western short-range missiles, e.g. the Norwegian Penguin,
will continue to be effective in relatively enclosed waters. Helicopters also can
usefully be fitted with such devices to defend their operating platforms from
attack by smaller craft, and to expand the ASUW coverage of a surface ship. A
missile designed specifically for this duty is the semi-active homing Sea Skua
used in the Falklands to damage an Argentine patrol vessel.
The limited lethality of this weapon illustrated a general problem with all
missiles. A hit above the water line is much less catastrophic than one below it,
unless the target has special vulnerabilities. The Exocets that scored successes in
the Falklands seem to have been incendiary devices as much as anything else,
their rockets setting fire to vulnerable wiring systems or cargo. The same seems
to be true of hits by the Exocet missiles in the Gulf War. Indeed, there is a
persistent doubt about the reliability and effectiveness of Exocet’s warhead. Only
one of the pair that hit USS Stark exploded. Even if it does explode, glancing
blows, e.g. as suffered by HMS Glamorgan in the Falklands, do not cause
catastrophic damage. Measures are being taken to provide ships with lightweight
armour to protect against warhead damage, and to harden their electrical systems
against the incendiary effect of missile hits. These will be countered by
improvements in both warhead lethality and reliability, but medium-sized
combatants may gradually become significantly less vulnerable to anti-ship
missiles. As the Gulf War demonstrated, unless they are hit in flammable cargo
or storage areas, large merchantmen posses considerable resistance to such
weapons; warships are, paradoxically perhaps, much more vulnerable although
again the bigger the ship the better in damage resistance terms. Smaller vessels,
like the Libyan fast-attack craft, and the unknown mystery ship sunk by the US
Navy’s Harpoon missiles in the incident of March 1986, will remain the most
lucrative missile targets.
As steps are taken to diminish the vulnerability of targets so the benefits of
combining missiles with final underwater attack might seem attractive. Such
systems were indeed considered late in World War II when armoured targets
were still a significant problem. The British developed a large unguided rocket
called Uncle Tom to strike underwater, and various more exotic air-launched
torpedo devices were worked upon. The relative lack of toughness of most post-
war naval targets, and the priority for ASW and the advent of nuclear weapons,
sounded the death knell of these advanced systems, but the Americans did
deploy briefly in the mid-1950s an interesting system called Petrel, an air-
launched, acoustic homing torpedo that was launched by a maritime patrol
aircraft five miles from its surface target. A jet engine and radio command
guidance system placed the torpedo in the correct spot for it to complete its run
under water. Technical progress over the last thirty years might make a more
modern variant of such a device a truly formidable weapon.14
Moving to underwater attack would also help deal with the classic forms of
countering missile attack, i.e. covering the target area with a chaff cloud. The
chaff prevents the incoming missile locking on to the target ship. This technique
was used with some success by the Israelis in 1973, and also by the British in the
Falklands War, although as the sinking of the Atlantic Conveyor demonstrated,
missiles can be potentially diverted from escorts on to higher-value shipping in
their care. Chaff equipment is being made much more effective. The current
Barricade rocket equipment used by NATO navies is controlled by a computer
system. It distracts by putting up decoys before the missile acquires the real
target, or confuses by placing additional ship-like echoes on the missile’s
acquisition radar, thus delaying or preventing target selection. If a persistent
missile does manage to lock on, the chaff cloud can be used together with
jamming to ‘steal’ the target from the missile, or the chaff can be put in the right
place to seduce the missile away from its target. The system computer generates
an optimum course for the ship to steer for the missile to break lock. Even more
advanced active countermeasures are becoming available, e.g. Marconi’s Siren
system, which sends back to an incoming missile a precisely tuned radar ‘song’
to draw it away from target. Only one rocket is required per threat which helps
cope with multiple attacks. Floating decoys have also been developed. For
example, Marconi produce simple ‘towed offboard active decoys’ for use with
any kind of ship. Developed for the Gulf, such TOADs can be ‘tuned’ to any
likely threat. The little boat-like decoy searches for missiles automatically. If it
detects a threat it sends jamming signals to seduce the missile from the target
vessel.15
This is not to say that missiles will be denied effectiveness in the future. They
will remain most potent threats, especially when deployed in large numbers to
swamp defences. Not all targets will be equipped with modern countermeasures
and a large number of especially vulnerable targets will be at risk. Nevertheless,
it seems unlikely that the anti-ship missile will drive shipping from the surface
of the sea. Like the torpedo, which made a somewhat similar entry onto the
naval stage a century or so ago, the missile’s capabilities have been overrated.
No single ‘wonder weapon’ can affect something as complex as maritime
operations. In certain circumstances it might prove decisive, in others
ineffective, especially as countermeasures develop.
A good example of the complexity of the factors governing the effectiveness
of naval weapons systems is the effectiveness of ‘iron’ bombs dropped by small
fighter aircraft. Given the strength of modern ship defences, it is not easy for an
aircraft to use an ordinary bomb against a well-protected naval target. As the
Argentines found in 1982, even if hits are scored the delivery altitude required is
usually so low that the weapon often fails to explode. If the defences are weak,
however, the attacker’s delivery altitude can be higher and these weapons can be
all too effective. Retarded bombs help solve the problem of ultra-low-level
delivery and these may well be a significant threat to ships, especially in coastal
waters, for some time to come. The fate of HMS Ardent in 1982 was a potent
demonstration of this.16 Cluster bombs, as used against Libyan patrol craft in
1986, also have advantages, especially against relatively small ships, while laser-
guided bombs homing on the reflected energy of a laser designator proved
formidable anti-ship weapons in the Gulf in 1988.
A final form of anti-surface warfare to be considered is an old and, until
recent incidents in the Gulf, often a neglected one: mining. Still perhaps the most
effective means of closing a harbour in war or violent peace is the underwater
explosive device set to be exploded by a passing ship. Contact mines are still
available, but ‘influence’ mines have greater potential using the three main
signatures of a ship: magnetic, water pressure, and acoustic. More than one
sensor can be combined in the same weapon and ship counters can be fitted to
delay action to a later target than the first to pass over the weapon. Modern
electronics allow mines to be given a remarkable degree of ‘intelligence’ to deal
with the information provided from their various sensors. Mines can be
programmed to respond to specific signatures, e.g. large, lucrative ship targets
rather than small, mine countermeasures (MCM) vessels.
The classic mine consists of a large explosive charge placed on the bottom or
moored to it. These are only useful if exploded relatively close to the target and
are only effective, therefore, in shallow water. The last quarter of a century or so
has seen the development of ‘mobile’ mines which can be laid in deeper water.
More primitive kinds ‘oscillate’, i.e. move up and down on a pattern searching
for targets. Others ‘home’ on to a target either themselves or by releasing a
‘homing’ torpedo. Such mines tend to be used primarily against submarines, but
they have a wider utility and will probably be more widely used against surface
vessels too. They also create a set of new hazards for mine countermeasure
forces, which will have to worry increasingly about being ‘outranged’.
Mines are the least appreciated of naval weapons. Their use is often seen as
an unwelcome and rather passive and ‘defensive’ alternative to more ‘normal’
types of naval warfare. There is perhaps some truth in this, and the Russian
propensity for mine warfare has much to do with their traditional lack of
expertise in more conventional naval operations. This, however, should not lead
one to discount the effectiveness of mine warfare. It has been very effective in
the past (e.g. in the opening weeks of World War II in British coastal waters, in
the final US blockade of Japan, and in closing Haiphong in 1972). It could be
even more effective in the future. Mines can be laid by a wide variety of
platforms: aircraft and submarines as well as surface ships of all shapes and
sizes. Covert use by unofficial groups as well as by governments has taken place
and may well occur again.
Mine countermeasures thus have a growing importance and there has indeed
been a notable increase in interest in the subject among the world’s navies. The
Royal Navy, for example, which began a large fleet of MCM vessels in the
1950s but then rather lost interest as strategies changed and alternative, more
spectacular priorities were preferred, is currently investing over a billion pounds
in efforts to improve its MCM potential. The enhanced security of NATO
reinforcement shipping is a major aim of this programme (as well as the free
passage of RN and USN ballistic missile submarines).
A real problem, however, is that countermeasures against modern mines are
far from easy. The easiest and most traditional method of dealing with mines is
sweeping, using wire sweeps to cut the moorings of fixed mines causing them to
come to the surface for disposal. Influence mines laid on the bottom in shallow
waters can be exploded by magnetic or acoustic sweeps simulating passing
ships. The effect to which pressure mines react cannot, however, be easily
simulated and simple ship counters or more intelligent discriminating techniques
can relatively easily defeat sweeping efforts. Unless the sweeping is done by
helicopter, the MCM vessel itself is always at risk. One way of overcoming this
problem is to use robot minesweepers controlled from a mother ship, as
pioneered by the West Germans with their Troika system. The Swedes have also
developed remotely-operated acoustic-magnetic sweep vessels called SAMs for
use with their Landsort class MCM vessels.
The manifold drawbacks of sweeping in the modern era mean that mine
hunting often has to be resorted to, which is always a time-consuming business.
The normal technique is to use human divers inspecting and disposing of, or
removing, objects discovered by a precise sonar equipment in the minehunter.
The latter may be neutralized by adverse water conditions, but variable depth
mine-hunting sonars are being developed to deal with this problem. As an
alternative to human divers, TV equipped mine-neutralization vehicles can be
sent down into clear water of limited depth to inspect objects and, if necessary,
lay charges next to the mines. Advanced Remote Minehunting System (ARMS)
is a British produced, remotely operated vehicle that does not require a
specialized MCM platform. Mine hunting has its own hazards: mines may be
booby trapped or programmed, if mobile, to home on to mine-hunting sonar.
Much work has still to be done to improve MCM capabilities (e.g. pressure
sweeps) to counter the guile of the offensive user of mines. These unspectacular
weapons look like retaining a considerable degree of effectiveness for sea denial
for many years to come.17

Anti-submarine warfare
ASW has come a long way since the first submersible boats led to such
desperate expedients as training seals or seagulls or patrolling harbours with
bags to place over submarines’ periscopes. Antisubmarine torpedoes, deployed
from a variety of sonar-equipped platforms, now hunt down their quarry using
their own effective acoustic sensors. These are the major ASW weapons systems
and are likely to remain so, but other methods, mines and even traditional depth
charges, will still have a significant part to play in certain circumstances.
An anti-submarine torpedo must conform to a set of most demanding
characteristics. One expert had identified six criteria for any such weapon:

It has to be fast enough to attack and if necessary overtake any evading


target.
It requires sufficient range to reach, attack and re-attack if needed.
Its detection capability must be more than adequate to overcome
inaccuracies in placement.
It must be able to distinguish between real and false targets and be able to
handle or disregard countermeasures.
It must be launchable into and effective in the shallowest depths in which
submarines can operate and yet be able to attack them at their maximum
depth.
It must be lethal.18

These requirements sometimes conflict. A torpedo must catch its target


without creating so much noise that it alerts it or masks the weapon’s own sonar
or that of its launching platform. A physically small sensor must be combined
with torpedo electronics that can distinguish between the sound of the quarry and
other sounds, including the torpedo’s own. As the expert quoted above went on
to write, torpedo development is a problem-filled exercise:

A wide variety of techniques are brought into play – mechanical or electrical


propulsion, hydrodynamics, warhead design, electro-transducer development,
amplifier design, the latest signal processing including correlation
techniques, beam steering, digital processing, algorithm definition, computer
architecture guidance, radiated noise suppression, advanced power controls,
flow noise controls, simulation techniques, analysis and tracking – a
formidable list which makes it difficult to succeed without prodigious effort
and allocation of large financial and manpower resources.19

And not even necessarily then as the British experience, at least until recently,
shows.20
At a price in both time and effort, formidable weapons can be created. The
United States is now deploying, and the United Kingdom developing, submarine
launched 533 mm (21 inch) torpedoes which combine high energy ‘monofuel’
propellents (stable forms of nitroglycerine), external combustion, reciprocating
gas engines or gas turbines and water jet propulsion to propel them quietly out to
30-40 km at speeds of about 50-60 knots. These torpedoes, the American Mark
48 and British Spearfish, are highly intelligent devices with on-board computers
which can adapt to the complex changing environment in which the torpedo
finds itself. Guidance wires allow the torpedo to interact with the parent
submarine and its more powerful sonar and computer capability. The weapons
are fully capable of attacking autonomously, however, if the wire was
accidentally broken or when the launching submarine is sure of ‘lock on’. At a
million dollars per unit, however, these weapons are expensive and most navies
will have to rely on similar weapons of reduced capability like the current
propeller driven, battery electric powered British Mark 24, with only half the
range and half to two thirds the speed of the Mark 48.21
If further increases in performance are required then new propulsion
technology will be necessary. Open exhaust engines like those used in the Mark
48 or Spearfish suffer from back pressure due to the surrounding water,
especially at great depth. This lowers power, increases fuel consumption, and
puts extra stress on the engine. The reciprocating engine suffers less than the
turbine, but both have limited development potential. If greater performance is
required, and submarine development will probably enforce it, a closed cycle
system is desirable. This might take the form of a stored chemical energy
propulsion system (SCEPS), using lithium and sulphur hexa-fluoride as fuel. The
heat generated by the combustion of these substances can be used to create
steam to drive a turbine on a closed Rankine steam cycle. Alternatively,
advanced battery and electric motor technologies promise similar potential
power outputs. A lithium/thionyl chloride battery might yield almost twice the
amount of energy per unit weight than a SCEPS system. Batteries, however,
have less development potential and there are also problems to be solved with
high currents interfering with electronics, and explosion dangers due to high
electrical density. Future heavyweight torpedoes using advanced propulsion
systems, and with improved hydrodynamic shape, might be able quietly to
engage targets at virtually any operational submarine depth out to 50 km and at
speeds of 80 knots. Wake-homing techniques have ASW as well as ASUW
potential.22
Heavyweight torpedoes have the great advantage of a large warhead which
can cope with the newer types of tougher, double-hulled submarines. It is a
weakness of aircraft and surface ships that they usually have to rely on smaller
‘lightweight’ torpedoes for ASW purposes. These are usually autonomous
weapons, being dropped over the target or fired in the correct direction from an
on-board torpedo tube. Based on information from dipping sonar or sono-buoys,
of from a ship’s sonar, the torpedo then begins a programmed search pattern
using sonar, first to detect and then to ‘home on’ to the enemy. Such a weapon
needs to be able automatically to adopt the optimum sonar beam for the
environment, searching out to the maximum range in minimum time. The
weapon ought to be able to have sufficiently powerful electronics to cope with
enemy tactics, evasion techniques, and countermeasures as programmed into its
‘memory’. All this must be crammed into a size less than half that of a
‘heavyweight’ or ‘long’ torpedo.
The most widely used lightweight torpedo in the West is the 324 mm
American Mark 46, later versions of which use the Otto propellent of their larger
counterparts to drive them at up to 40 knots and up to 11,000 metres. The
active/passive acoustic head has an acquisition range of about 450 metres.23
Lightweight torpedoes need even more efficient power plants than their larger
cousins, and they have pioneered the use of advanced closed cycle systems. The
new American Mark 50 has a SCEPS power unit, giving speeds of up to 55
knots and ranges of 15,000 metres. The British are developing a SCEPS/Rankine
cycle system known as CCTS (Closed-Cycle Thermal System) for possible use
in a future lightweight torpedo. Advanced battery technologies, with their high
specific outputs, also have obvious application in lightweight torpedoes. The
current British Stingray also already uses a magnesium silver chloride battery
which gives similar performance to the Mark 46’s Otto system, but more quietly
and down to greater depths.24
Britain’s Stingray and the American Mark 50 are good examples of the way a
torpedo’s guidance system can be improved by the application of more advanced
computer software. These increasingly ‘smart’ weapons can better distinguish
targets from clutter or countermeasures, select the optimum homing mode for the
conditions, and be adjusted to cope with changes in the enemy’s tactics or
countermeasures.
France and Italy, who have produced rather crude lightweight homing
torpedoes, are both now developing weapons comparable to Stingray in
guidance and propulsion. Sweden has its own lightweight torpedo, the Type 42, a
wire-guided passive homer specially designed for ASW and ASUW use in the
Baltic from a wide range of platforms. The Swedes face sonar problems in these
shallow waters, which have discouraged the use of the active mode in homing
torpedoes. But the advent of quieter submarines is forcing a move in this
direction, one aided by modern electronic and computer control.25
The Soviet Union utilizes electrically-powered, acoustically-homing,
heavyweight torpedoes from its submarines and surface ships for ASW. Long-
range wake-homers may also have a role here. It also deploys a somewhat larger
lightweight torpedo than any other producer: a 406 mm weapon carried in
aircraft, smaller surface ships, and submarines. Earlier models were passive only,
but the current type is an active/passive device with wire guidance.26
To protect itself from torpedoes the submarine can either use decoys or make
itself physically hard to destroy. Bubble generators, intended to deceive an
enemy into thinking that a submarine has been destroyed, can act as underwater
chaff, while torpedo-type underwater ‘training targets’ simulate submarines’
acoustic characteristics. Specialized torpedo acoustic decoys known in America
as ADCs (acoustic device countermeasures) are available to be ejected from the
submarine when it is attacked, and hang in the water specifically programmed to
jam the threat. Future anti-torpedo torpedoes also have relevance to submarines
as well as surface ships. If a hit cannot be avoided a double hull, as used in
recent Soviet submarines, may offer protection. Specialized penetrating
warheads go some way to solving this problem for the attacker, although they
provide an extra challenge to the guidance mechanism to deliver the weapon at
the correct angle. The problem of lethality, coupled with increased speed and
performance of the potential targets, casts a real shadow over the long-term
viability of the smaller torpedo which, as Norman Friedman has pointed out,
would have troublesome operational effects given the importance of these
weapons in the armouries of the world navies. Certainly the challenge of
developing all kinds of torpedo to defeat ever more formidable countermeasures
is going to limit production to an even narrower range of states than
previously.27
Homing ASW torpedoes can also be delivered by ‘stand off’ weapons of
various kinds. The United States and its allies deploy the anti-submarine rocket
(ASROC) missile system on surface ships, which uses a ballistic rocket to send a
Mark 46 torpedo over several kilometres out to the computed position of the
submarine. A new vertical-launch version will soon be deployed. Australia has
developed a guided missile, Ikara, to deliver a similar payload over a rather
longer distance (20 km). Both will be used for some time to come and similar
systems are under development. The Soviets deploy a broadly similar, but larger,
system as their primary ASW surface ship armament. The SS-N-14 has an
estimated range in the region of 50 km. The Soviet Union also deploys a
torpedo-carrying, underwater-launched ASW missile in submarines. The United
States, which has so far only used this technology with nuclear warheads, plans
to fit a Mark 50 homing torpedo on its 185 km (100 mile) range ‘Sea Lance’
ASW-SOW (Anti Submarine Warfare Stand-Off Weapon), which will be
deployed aboard submarines. Given adequate communications this could be
targeted on the basis of sensors and computerized data bases external to the
launching platform, but doubts have been expressed over its ability to engage
targets at maximum range without a nuclear warhead (which was delayed
indefinitely in 1986).28
An alternative to taking the torpedo to the submarine is to get the submarine
to come to the torpedo. The ubiquitous Mark 46 acts as the kill mechanism of the
CAPTOR ASW mine, laid in deep water by aircraft, submarines, or surface
ships. The mine listens for the distinctive sound signature of an enemy
submarine and when it detects a target (at up to 1,000 metres) it ranges on it,
using more accurate active sonar. When the target is within range the torpedo is
released. The Soviets have rapid rising ASW mines of their own reportedly
utilizing some form of underwater jet/rocket propulsion and either passive/active
acoustic or UEP detection and homing. Honeywell and Plessey are developing
the Crusader Advanced Sea Mine that is a deep-water rising weapon that can be
laid beyond the continental shelf.29 As an alternative to acoustic homing, ASW
mines can be activated by the passing submarine’s electrical field. All these
mines are designed for use off known enemy submarine bases and in known
transit ‘corridors’. Submarine operators, such as the British, are having to
emphasize special means of sweeping such weapons from relatively deep-water
areas.
Mines have a long pedigree of ASW devices, as do depth charges, the first
specifically ASW weapons to be developed. They were first of all dropped over
the stern or fired by mortars over the side, but World War II saw the
development of various ‘ahead throwing’ systems which allowed for a better
interface of sensors and weapons, and a more controlled engagement. Both
rockets and mortars were utilized in this manner, and still are in many of the
world’s navies.
Depth charges have several advantages still. They can be dropped to exert
pressure on a submerged target rather than with the intention of sinking it. They
thus have a graduated response capability that a more deadly homing torpedo
lacks. Although they have little capability against, for example, a nuclear
submarine, the threat of underwater explosion can be enough to cause a
commander to move away, e.g. HMS Conqueror in 1982. Similarly they have
the flexibility to damage a submarine that is too close to the surface to be
vulnerable to lightweight homing torpedoes, as in the case of the Argentine
submarine Santa Fe when attacked by British helicopters in the same conflict.30
For some time rockets or projectors for hydrostatically exploded depth
charges have been replaced by lightweight torpedoes on most modern surface
combatants built in the West, although most navies are retaining the capability to
drop depth charges from helicopters or other aircraft. Older surface vessels,
however, often continue to carry them using either racks or more sophisticated
projectors, like the excellent British Mark X mortar, capable of delivering three
accurately fused depth bombs in any direction from the ship out to 1,000 metres.
The French still deploy an interesting 305 mm mortar in their older frigates, with
a range of 3,000 metres, which can be used in a secondary shore bombardment
role. It will take some time for such systems to disappear completely and naval
commanders may miss them sufficiently to order their return. The weight of the
charges is certainly a potential counter to even the new generation of tough
submarines.
Still in production are various ASW rocket launcher systems, descendants of
the British Hedgehog of World War II. These project a number of rockets, fused
for contact, time, proximity (hydro-acoustic), or depth. Those navies with
shallow water ASW problems continue to be specially interested in these
systems. The Norwegians carry the Terne II sextuple ASW rocket launcher in
their destroyers and corvettes. This fires six time- or proximity-fused rounds out
to a range of 900 metres against targets whose range and depth have been
computed. The 48 kg warheads have a lethal distance of about 20 feet against a
normal submarine. The Swedish firm of Bofors produces a widely exported
tubed rocket launcher for 375 mm proximity/time/direct action rockets. A French
sextuple variant was fitted to new, small coastal escorts of the French and
Belgian navies built in the late 1970s and early 1980s. Japan continues to fit
quadruple launchers of this type to its latest frigates, in addition to lightweight
torpedo launchers.31
An especially interesting variant on this theme is the ELMA system
developed by Sweden to inflict limited damage upon peacetime underwater
intruders into her territorial waters. This consists of a number of multiple
launchers for small, shaped charge grenades which are scattered in the chosen
pattern and quantity over a range of about 250-350 metres. The shaped charge
projectiles are designed to drive a narrow 30-50 mm hole into the hull of the
delinquent submarine thus forcing it to the surface. This ‘imaginative’ limited
response to the problem of the covert underwater intruder is in service on
Sweden’s fast-attack craft.32
The Soviets, with their tradition of military rocketry dating back to World
War II, remain the widest users of ASW rocket launchers. The projectiles are
fired on to a contact’s position, using either depth or influence fusing. The
standard 6,000 metre range version is carried in all Soviet surface ships from
Petya class corvettes to Kirov class ‘battlecruisers’, and smaller vessels carry
smaller variants. Such devices will give the Soviet Navy considerable combat
flexibility in shallow water ASW for some time to come.33 ASW is an activity
fraught with so many uncertainties that all kinds of weapons will have a part to
play in countering the invisible and increasingly quiet menace of the modern
submarine. Indeed, it seems likely that the increasing quietness of the new
targets, both nuclear and conventional, will reduce the effectiveness of the
passive sonars upon which most ASW techniques have relied in recent years.34
Large, low-frequency, long-range, passive sonar arrays have long been mounted
on the ocean floor in the American SOSUS (Sound Surveillance System). This is
being supplemented both by denser, shorter-range static arrays covering certain
areas (the so called Fixed Distributed System) and mobile strategic surveillance
arrays towed behind unarmed surface ships known as T-AGOS. Combined with
modern electronic communications analysis and dissemination techniques these
systems have done much to deny noisy submarines their cover. By the 1970s this
had already reached the extent that it was being publicized that Western naval
commanders have constant access to the rough location of those potentially
hostile submarines that were at sea. This has provided the technical rationale to
the operational doctrines adopted and developed over the last thirty years by
NATO. ASW has tended to be viewed as a kind of air warfare with targets
detected at long range and platforms directed accordingly to deal with them.
These latter assets also rely on long-range passive detection for much of their
capability, in particular the tactical towed arrays streamed by both submarines
and surface ships with ranges of many tens of kilometres.35
As the number of submarines with quiet machinery increase there will be less
low-frequency machinery sound to penetrate long distances in the ocean carrying
its wealth of information to the passive signal processing equipment of the whole
ASW system. Informed analysts now argue that ASW in the future may well
have to re-emphasize much shorter-range passive detection opportunities.
Ranges of 50 km or so (the first convergence zone) may become the maximum
realistically attainable in all except the most favourable circumstances.36 Often
ranges will be much less. Nevertheless, the tactical towed array for the time
being remains a crucial long-range detection asset and it has allowed the surface
ship to reassert a new importance in ASW. Surface ships can carry longer, more
capable arrays and, unlike submarines, are freer to share their information with
other forces. Aircraft can be quickly ‘cued’ in to prosecute possible contacts
using shorter-ranged sonobuoys or dipping sonars. There is no doubt that much
work will continue with signal processing, e.g. in hitherto little used parts of the
noise spectrum, to retain the long-range capability of the passive towed array.
The performance of more traditional active/passive hull-mounted and variable
depth transducers can be further improved by means of still greater improvement
in digital electronic signal and data processing. Convergence zone active
detection and engagement should, therefore, still be possible for these systems
against even the quietest submarine in good water conditions, at least. The data
processing challenge, however, is enormous. A modern active sonar gives
around 4,000 pieces of data per second, a rate that generates false alarms every
2-5 seconds or so. In a passive sonar the rate is over 60,000 items per second,
giving a false alarm every 15 milliseconds. This makes automatic ‘target
extraction’ of particular importance. One possibly significant development is the
use of combination active transducer/passive array systems. These allow long-
range sonars to be fitted to small vessels unable to carry a large transducer array
themselves.37
The combination of active ‘transmitters’ and separate passive array ‘receiver’
technology will be the key to new, low-frequency active sonar systems that will
be developed as the wholly passive sonar window slowly closes. These will take
novel ‘bistatic’ or ‘multi-static’ forms with static, towed, or hull-mounted sound
sources sending out omnidirectional signals to be reflected by the target and
received on long-range passive arrays either static or towed behind other units.
Such systems have been proposed, in both Europe and the United States. Very
long-range, strategic, very low-frequency systems as well as shorter-range
tactical systems are conceivable, like today’s wholly passive arrays. As C.C.F.
Adcock put it to the 1988 Undersea Defence Technology Symposium:

In general, remote receivers can yield additional areas of possible detection


of submarines outside the area that would be covered by a monostatic
receiver located close to the sound source. An additional advantage is that,
although alerted to the sonar transmission, submarine commanding officers
have no means of discovering the location of the acoustic receivers. They,
therefore, lack the necessary information to be able to manoeuvre to keep
away from receivers, and present low target strength or minimal doppler.
‘Beaconing’ (i.e. revealing the transmitter’s position), might also be less
dangerous if the signals were made ‘broad band’ and thus hard to distinguish
from normal background noise. Beaconing, however, is not the only problem;
care would also have to be taken not to give away the position of one’s own
units which would also reflect the radiated sound. The tactics of long range,
low frequency active sonar will pose a fascinating set of conundra for the
tactician of the future.38

While surface ships and submarines equipped with active sonars are obviously
vulnerable to beaconing and reflection, an aircraft is less so. Longer-range, low-
frequency, active dipping sonars carried in helicopters are, therefore, likely to be
one key to maintaining effectiveness in ASW; in suitable water conditions
convergence zone ranges may be possible, ten times the 5,000 metre range of
current medium frequency systems. One such system under development is the
Anglo-American HELRAS (Helicopter Long-Range Sonar).39
Closing the passive sonar window is bound to limit the utility of the air-
dropped passive sonobuoy. Since 1960 low-frequency analysis and recording
(LOFAR) buoys, using the Jezebel analysis capability developed in the 1950s for
SOSUS, have allowed detection at ranges of up to 100 nautical miles. Over the
next fifteen years DIFAR (directional LOFAR) buoys were developed and the
United States has vertical line array DIFAR (or VLAD) buoys for deep
surveillance. Extending the capabilities of these systems will be necessary, with
improved signal processing capable of monitoring larger numbers of buoys of
more varieties. Active buoys, always important for target acquisition, will
however inevitably acquire a new relevance with new designs both to utilize
better batteries, and to make better use of those power sources. Rotary winged
aircraft will probably also be fitted with control systems designed to make
integrated use of sonobuoys and dipping transducers. Larger passive sonar arrays
that can be dropped into the water and then recovered may be practical as
‘Osprey’ type aircraft gain more general acceptance (see Chapter 6.)40
Submarine launched weapons with fire control solutions based on what will
still often be the best sonar information, co-ordinated by increasingly
sophisticated and integrated command and control systems and carrying the most
effective warheads, are probably going to remain individually the most effective
ASW weapons. Nevertheless, the developments just described will emphasize
the multidimensional character and complexity of anti-submarine warfare.41

Anti-air warfare (AAW)


Ever since the invention of the aircraft, the threat it poses has been a central
concern for naval officers anxious about the survival of both their forces and the
objects of their protection. The more recent advent of the missile has made the
problem even more serious as this has diversified the threat (submarines can now
pose ‘air’ problems), and created the need to destroy the threat physically rather
than just put if off its aim. Much of the earlier effectiveness of AA gunfire was
rather that of the modern chaff dispenser, putting the enemy’s ‘guidance system’
off its aim. There was greater scope for effectiveness as a man’s determination is
less predictable than that of an automatic system. Now and in the future,
however, AA systems must inflict catastrophic material damage if they are to be
effective.
This is not to say that traditional systems like guns, even those without very
sophisticated fire control, cannot be effective against a range of air threats. A
tendency to define the air threat too narrowly in technical terms, to emphasize
long-range aircraft and the missile rather than small aircraft armed with bombs,
led the Royal Navy for one to remove light AA armaments from ships in the
1950s and 1960s and rely on dual-purpose medium guns and missiles. That this
was insufficient to cope with the air threat disposed by a relatively weak power
like Argentina in 1982 demonstrated the shortcomings of this policy. Light guns
in the 20-40 mm bracket have made a dramatic come-back and will probably
remain. They can at least make a close aerial attacker’s task much more difficult,
and more advanced fire control and sensors ought to enhance effectiveness still
further.
At the upper end of the light gun spectrum come the sophisticated anti-missile
gun systems combining advanced autonomous automatic radar tracking with
ultra-high rates of fire. These are also going to become standard fitments to the
ships of the more sophisticated navies. The Vulcan Phalanx CIWS (Close-In
Weapons System) is widely deployed in the US Navy, and alternative European
systems are available in the form of Goalkeeper and Seaguard. Such weapons
fire ultra-high-velocity kinetic energy rounds designed physically to destroy the
incoming weapon. These guns cover the gap in the minimum ranges of many AA
missile systems, and are effective against current missile threats. They are,
however, relatively expensive and the extent of their deployment is thus limited.
Their effectiveness against future faster missiles is also in some doubt. Larger
artillery in the 114-127 mm bracket is also likely to have a continued rule, albeit
a relatively subordinate one for the AAW task. The US Navy has experimented
with infra-red guided shells for antiaircraft work and stayed with the more rapid
firing 5 inch (127 mm) gun, largely due to its AA potential. The 4.5 inch (114
mm) weapons in British ships did score some isolated successes in the Falklands
and added to the generally hostile environment faced by Argentine pilots.42
Despite the continued relevance of guns, rocket propelled missiles seem to be
the key instruments of short-range air defence. This revolution has been longer
in coming in reality than it might have seemed in theory, given the extent of
missile deployment in ships. It must be stated, however, that most of the first
generation systems were so limited in capability that their real combat
effectiveness was highly dubious. At best the longer range systems could destroy
individual high-flying aircraft, e.g. long-range bombers. The shorter-range
systems stood some chance of dealing with a closer-range attack, although one
of the most widely deployed, Seacat, has stood persistently low in the esteem of
its operators, at least in its earlier versions. A simple radio command weapon, it
depends very much on the skill of its human operator. Seacat itself has been
improved, both in sensor and control technology, and the missiles are now much
more reliable and effective, and are likely to become even more so.
Nevertheless, it is significant that the most widely-deployed first generation
point defence missile systems tend to rely on more automatic semi-active
homing radar techniques, usually using air-to-air missile technology.
Anti-air warfare operates in an easier medium than ASW, but the speed of
reaction required makes it complex enough. We may be approaching a situation
where very few navies will be capable of deploying systems able to cope with air
threats of the highest level in both quality and quantity. The only such system
available currently is the American Aegis, which combines phased array radars
with the most advanced electronic data handling techniques. The system can
cope with many targets and, in its initial form, launch about thirty-four missiles
per minute. These are command and auto-pilot guided into the vicinity of the
enemy, on whom an illuminator is switched to provide the necessary reflected
waves for final homing. More effective homing trajectories allow a greatly
extended range (by 60 per cent) for any given missile. Aegis is currently
deployed with the Standard SM-2MR missile, giving it an impressive maximum
range of about 150 km (90 miles). It can track targets out to 370 km (200 miles).
The system is highly resistant to electronic jamming and offers formidable
defensive potential – at equally formidable cost. It is currently only deployed by
the US Navy in its Yorktown class cruisers, of which twenty-seven are planned
in total. Aegis is also to be fitted in modified form to the new US class of
Arleigh Burke destroyers now under construction, and Japan plans to put it into
its four new 6,000 ton destroyers.43
The European navies seem to have been put off not so much by sheer cost as
by the industrial implications of not being able to interface European hardware
with the expensive software that is the core of the system. However, the
multinational anti-air warfare system currently being proposed for the NATO
frigate of the 1990s may be based around Aegis, which might allow European
navies to get their hands on the vital software cheaply. France, the major
opponent of an Aegis software buy, has reportedly been assured of the
compatibility of Aegis with its new Aster 90 vertically launched surface-to-air
missile. In its initial Aster 15 form this is a relatively short-range system (10 km
against missiles, 17 km against aircraft). France hopes to develop a longer-
ranged Aster 30 as an area defence weapon, but a general trend in the European
navies seems to be towards pushing out the range of point defence missiles
rather than emphasizing expensive new area defence systems. The new vertically
launched Sea Wolf is another example of this kind of development. The use of
inertial navigation systems in these new missiles results in more efficient
trajectories, longer range, and diminished demand on the ship’s fire control
system. Vertical launch allows very high rates of fire, one missile per second.
Such projectiles could be combined with cheaper radar hardware: one British
firm, Plessey, is claiming a breakthrough in gallium arsenide micro-circuitry to
provide significantly smaller and less expensive ‘multi-function electronically
scanned array radars’ (MESAR). The whole combination might well amount to
an affordable combination of hardware and software that will allow European
ships to operate in high threat environments. Co-operative procurement is,
however, the key. It is unlikely that any single national European navy could
afford its own system of this level of capability.44
As one might expect, the Soviets seem to have some difficulty in coping with
the computer technology associated with this level of capability. The only
system in any sense comparable is the SA-N-6 of the Slava and Kirov classes. In
these two classes the relatively large missile is controlled by a large Top Dome
phased array radar that is steered in azimuth, unlike the fixed multiple radar
faces of the Americans. According to one American critic this ‘suggests a
limited ability to handle threats from different relative positions
simultaneously’.45 While the Kirov class carries two such radars, the Slava class
carries only one, which must limit further its all-round coverage. The only Soviet
ship with electronically scanned phased arrays similar to those used in Aegis is
the fourth aircraft-carrying cruiser Baku. Oddly, however, this vessel does not
seem to carry a compatible area defence missile system. The phased array
radar’s role seems to be fighter direction and air warning.46
Moving down the scale, the world’s navies are well equipped with missiles
which use semi-active homing to attack an incoming aircraft: the missile homes
on the reflected energy of a ship’s radar. The longer-range variants cover a
spectrum from about 30 km to over 100 km. Sea Dart, the British missile
designed to give long-range AAW protection, gave excellent service in the
Falklands War against well-defined aircraft targets detected in good time against
a clear background. It destroyed six Argentine aircraft and (by mistake) one
British. It was predictably less effective, however, against low-flying targets
coming out from the coast, as HMS Coventry found to her cost.47 Another
general problem with weapons of this type is their ability to engage only the
number of targets for which they have illumination radars. The capacity of the
system can be improved by giving a mid-course guidance capability to the
missile, as the Americans are doing with the long-range variants of the SM-2
missiles deployed on their older missile cruisers. This, however, once again
implies expensive modifications to the ship’s missile control system.48
Whatever their capability of dealing with multiple targets, such systems force
enemy aircraft to attack from low altitudes, so lowering range and weapons
carrying capacity. Low-altitude targets that get through the longer-range missile
envelope are dealt with by shorter-range systems. These may be advanced
command-guided systems using radar and or TV trackers, like the British Sea
Wolf. The latter proved successful in the Falklands (four aircraft and a
contribution to a fifth), although prone to confusion at first due to narrow
programming of its software. It will provide the vital self-defence capability of
British warships into the next century. With the advantage of combat testing it is
now probably as effective a close-range defence system as any in the world. The
early variant, however, was only intended for short-range use (up to 5-6 km) and
could only be used in support of other units at some danger to the platform due
to the close manoeuvring involved, and at some risk of tactical confusion.49
Moreover, its fit to all ships was constrained by cost and relatively high
topweight demands. The latter problem, at least, has been dealt with by the
creation of a lighter launcher and also the new variant for vertical launch, which
also, by giving it extra boost, significantly increases its range and flexibility and
converts it into something that can protect something else as well as just the
platform carrying it.50
Sea Wolf was designed to deal with the Soviet missile threat. Less capable
point defence systems provide adequate defence against low-flying aircraft.
Missile technology designed for land-based surface-to-air or air-to-air use is
often used with various forms of guidance: semi-active radars, infra-red,
command-to-line-of-sight. They can give vessels down to fast-attack craft a
useful self-defence capability. They are produced for export and domestic use by
most of the arms manufacturing nations, including Israel, and should increase in
effectiveness incrementally over the years.
It is possible that, in the future, missiles might acquire longer ranges and
greater effectiveness still by the adoption of some kind of external control
system whereby a surface vessel is only the launch platform and the engagement
is controlled by an airborne or even satellite platform. This has been discussed in
the United States as, in American football parlance, a ‘forward pass’ or SLAT
(surface-launched air-targeted) missile. Developing such a system would be very
expensive and problematical, especially in terms of adequate data links. The link
between a missile and its own launching platform is much more secure.
Nevertheless, work continues on such concepts, and the increasing adoption of
vertical launch systems with the potential for loading large, long-range missiles
gives the idea some credibility.51
The last way of dealing with air threats at maximum range is to use aircraft of
one’s own, which raises bureaucratic and organizational as well as command and
control difficulties (see Chapter 6). The use of aircraft as missile launch
platforms makes air interception especially lucrative. Carrier-based fighters,
combined with air-to-air missile technologies, are still vital weapons if one can
afford them. One central lesson of the Falklands War, however, was the
effectiveness of relatively crude and simple fixed-wing platforms (Sea Harriers),
when combined with advanced infra-red homing missiles (AIM-9L
Sidewinders). Possession of the superior missile deprived Argentina of any hope
of dealing with the main defence of the British Task Force from air attack.52 The
development of the new types of active radar guided missiles, like AMRAAM
(advanced medium-range air-to-air missile) will enhance the capabilities of such
relatively unsophisticated aircraft as the Sea Harrier still further.
The main constraints on the operations of the Harriers in the Falklands were
their minimal radar capability and, more importantly, the lack of airborne early
warning radar coverage. Land-based aircraft can provide the latter within range,
but adequate results can be obtained by radars mounted on helicopters. It would
seem prudent not to deploy short take-off vertical landing (STOVL) fighters
without such AEW helicopter support. The combination is obviously less
capable than a fleet air defence envelope of US proportions, but only the US
Navy is in that kind of league – a circumstance that will not change for many
years.
The level of air defence capability provided by the combination of Phoenix
missiles, F-14 Tomcat fighters and E-2 Hawkeye AEW aircraft provides
protection that stretches well over 500 km around the centre of any carrier battle
group.53 Each aircraft can launch up to six Mach 4 Phoenix missiles, each of
which can be directed on to a separate target at ranges of 120 km using a
combination of semi-active and active radar homing using the aircraft’s own
radar as illuminator. Phoenix was primarily designed to deal with Soviet missile-
carrying bombers. Lesser missiles, both radar and infra-red guided, are more
appropriate for close-range encounters with fighter type aircraft, and can be
carried by the F-14 or its less capable F-18 stablemate. The ability to fight for air
superiority is formidable, and US carrier battle groups can think in terms of
controlling the air environment in which they find themselves to a considerable
degree, even to the extent of taking on the Soviet Naval Air Force close to its
bases.54
It is in AAW that the naval warfare of the future may get into the realms of
the exotic. Ultra-high-velocity railguns might well have an application given
their potential hitting power, and speed of response and engagement. The
availability of electrical power from the ship’s machinery should make their
mounting in naval platforms easier than in the tanks for which they have been
proposed. As for directed energy weapons, relatively low-powered electro-
optical lasers of sufficient power to blind and/or damage sensors, including
pilots’ eyes, are currently practical weapons and are already deployed at sea.55
As work is carried out by both superpowers on more powerful directed energy
weapons as part of anti-ballistic missile (ABM) research and development, these
very advanced technologies might well become available for the defence of
warships. In the short term, however, the American Strategic Defense Initiative
may have slowed down the development of exotic naval weapons. The US
Navy’s Chair Heritage programme to defend carriers with particle beams was
absorbed into the SDI and developed a new ABM emphasis. All directed-energy
weapons have one enormous advantage. By working at, or close to, the speed of
light they require few sophisticated fire control calculations, just accurate
pointing.

Operations against the shore


Admiral Gorshkov has pointed out that perhaps the main revolution in modern
naval affairs has been the greatly enhanced utility of navies for operations
against the shore.56 Bombardment is a longstanding role, either for its own sake
or in support of armies. Guns, aircraft-launched weapons, and missiles will all
continue to have a part to play in this form of warfare.
Coastal bombardment, even more than ASUW, will be the major role of the
medium/heavy calibre artillery carried in ships. Vietnam confirmed the US Navy
in its continued requirement for such capabilities, as did the Falklands War for
the British. Guns can deliver fire more accurately than aircraft, in greater
quantity and with less risk to personnel. This is particularly true in limited war
and gunboat diplomacy situations as it is more likely that the object of the
exercise of force will be better protected against aircraft than seaborne long-
range artillery.
The US Marine Corps found especially welcome the reactivation of the Iowa
class battleships, as their 16 inch guns make formidable shore bombardment
weapons. The accuracies obtained with the old propellent available initially
proved to be less than expected but these problems have now been solved by
remixing. Even the old armour-piercing rounds might have a role against some
targets, such as airfield runways, and cluster munitions offer considerable
potential in both the anti-personnel and anti-armour role.57 Nevertheless, it is
hard to see new guns of this calibre being deployed given the adequacy of
modern medium artillery in the 5-8 inch bracket to cover realistic bombardment
tasks. US ships still mount 5 inch (127 mm) guns, and the Soviets also mount
nothing larger than 130 mm (5.1 inch) guns in their new ships, even their new
battlecruisers. Calibres may well creep up, however: the British are developing a
155 mm weapon, based on army field howitzer technology in order to provide
greater explosive firepower and, more importantly, an improved facility for
indirect fire in mountainous (fjord-like) terrain. It would also not be surprising if
the Americans eventually returned to something like their 8 inch Major Calibre
Light Weight Gun project, abandoned in 1978 (in part to help the battleship
case), together, perhaps, with a specialist shore bombardment ‘fire support ship’
to carry it.58
For shorter range, multiple rocket launchers offer many advantages in their
ability to cover broad areas with fire. They are already in use to some extent and
there would seem to be few problems in applying the new guidance and warhead
technologies being considered in connection with NATO’s new multiple-
launched rocket systems to shipboard variants. This could allow even more
formidable fire support to be delivered in support of amphibious forces.
Missiles can be used to carry out deeper conventional bombardment of land
targets. The American Tomahawk cruise missile has two conventional variants
TLAM/C and TLAM/D for this task. The former is coming into service, the
latter will be available by 1990. These missiles have a range of around 1,000 km
and navigate themselves over land using terrain contour matching (TERCOM)
guidance, updating a basically inertial system by comparing a digital picture of
the ground over which the missile is flying with a ‘map’ in the missile’s
computer memory. On closing the target a digital scene matching area correlator
(DSMAC) system compares an optically-scanned picture of the target area with
another digital picture in the computer memory. This increases accuracy to about
30 feet to give TLAM/C’s single 972 lb blast fragmentation warhead a chance of
inflicting serious damage on a precise target. TLAM/D’s submunition warhead
will add to the Tomahawk’s conventional effectiveness, and experiments have
been carried out with one missile making a number of bomblet attacks on
separate targets. Nevertheless, conventional Tomahawks will still probably have
to be fired in large numbers to be sure of achieving significant military effect.
This means that platforms with large magazine spaces, i.e. surface ships, are
necessary for it. Submarines simply cannot carry enough missiles. It is hard to
imagine suitable targets for individual missiles, especially the current TLAM/C,
which is a relatively costly way of delivering a very limited amount of
ordinance. Pictures have, however, been published of successful attacks on
parked aircraft and bunkers, and it might be useful in making exemplary ‘anti-
terrorist’ punitive strikes, especially in circumstances when loss of pilots would
be embarrassing. Used in adequate numbers the conventional Tomahawk’s major
role is defence suppression in support of more conventional air strikes. Radar
and missile sites, and control centres, would be lucrative targets and even
airfields might be thrown into confusion, if not knocked out physically, by a
series of massed TLAM/D strikes, especially if missiles were fitted with
specialized submunitions, perhaps with time fuses.59 Mike MccGwire has
suggested that the new Soviet SS-N-24 large sea-launched cruise missile
currently being tested on a converted Yankee class SSBN (Nuclear Powered
Ballistic Missile Submarine) may be a conventional long-range bombardment
system for use against command centres and nuclear assets in Europe in the
early days of a conventional war.60
Sea-based aircraft provide the final and most important means of carrying out
conventional ‘naval bombardment’. The capabilities of the US Navy to carry out
attack missions with a wide range of conventional munitions (‘smart’ and
unguided) is as good as any land-based air force’s, although limited numbers of
all-weather strike aircraft can be a constraint in certain circumstances, e.g. in the
attacks on Libya in 1986, when attacking the full target would have required two
sequential waves, both politically and operationally undesirable. Stealth attack
aircraft are under development for the US Navy. All navies with organic air
components have some ability to mount bombing raids, although no other navy
is anywhere near to US capabilities. In the Falklands War, for example, the
limited bomb-carrying capacity of the British Sea Harriers forced the use of
land-based, long-range bombers, at considerable expense and to dubious effect.
STOVL aircraft, however, can still be extremely useful, especially for the close
support of ground operations.

Command and control


The twentieth-century revolution in command and control – the conversion of
individual weapons into integrated weapons systems – is perhaps the least
appreciated aspect of modern naval operations. Ships and weapons are too often
compared by their outward appearance, and their performance parameters
narrowly and simplistically defined. The less tangible elements may, however,
be the most crucial. As early as World War I the performance of the British
Fleet’s gunnery was far more affected by the defective performance of its
mechanical analogue fire-control computers and the interface of those computers
with accurate but low-data-rate sensors, i.e. rangefinders, than it was by marginal
advantages in numbers of guns, their calibre or even the quality of their
projectiles.61 In World War II the advent of truly three-dimensional naval
warfare and electronic sensors that increasingly replaced the eye as the main
means of identifying the enemy’s position enforced a fundamental development
in the way in which naval warfare was managed. Spaces in the ship had to be set
aside for what in the Royal Navy was called the Action Information
Organization (AIO), where the data from the ship’s electronic sensors were
integrated into a coherent picture. British practice was to call the surface warfare
room the Operations Room. American warships had Combat Information
Centers (CIC).62 All this was space intensive and it put strains on the internal
volumes of ships. Sometimes, this meant less obvious armament but what was
left was much more effective. The ‘heavy’ armament of the ships of the past was
a function of inaccuracy and inefficiency rather than real fighting power.
The data handling and collation initially had to be carried out by human
means, but the advent of the electronic computer allowed mechanization and
automation. The Americans began to computerize their CICs with Naval Tactical
Data System (NTDS) in the late 1950s. This has been progressively upgraded
since. In the 1960s the British followed suit and the carrier Eagle was fitted with
ADA (Action Data Automation System). This ‘co-ordinated air defence
information from the ship’s own radars, from other ships and aircraft by data
link, and from visual sightings. This information was either stored in the
computer’s memory, or displayed. The computer was capable of evaluating
threats, calculating interceptions and passing out information required by
fighters, guns and missiles. It was able to perform tasks that normally took some
sixty men.’63 This kind of capability spread to smaller ships the following year
with the ADAWS (Action Data Automation Weapons System) of the County
class large destroyers. ADAWS was provided to various subsequent classes
especially, but not exclusively, to those with an air defence role. It remains in
service in various versions: two in HMS Bristol; four in the Type 42 destroyers;
five in the Ikara Leanders; six in the first two ASW carriers; and ten in Ark
Royal. A simpler system, known as CAAIS (Computer-Assisted Action
Information System), was developed for frigates, and this has been replaced by
CACS (Computer-Assisted Command System) introduced into the later Type
22s. CACS has not had a happy history. The first variant had software problems
and CACS 4 intended for the Type 23s has had to be re-tendered, which will
delay its entry into service until after its platform has been completed. The first
Type 23s it seems will thus go to sea without their electronic ‘brains’ and hence
much of their ability to fight.64 History sometimes repeats itself.
The function of these systems has been described as follows:

They aim to gather and collate information and present it intelligibly for
decision. . . . They enable targets to be designated, and the two more
comprehensive systems, ADAWS and CACS, directly control certain
weapons, taking over the sort of function which otherwise is carried out by
separate fire control systems.65
The captain of a modern warship now fights it not from the bridge, but from a
centralized operations room in the bowels of the ship where officers and men
pore over cathode ray tube displays making keyboard injections to input data or
decisions. This revolution has made necessary structural changes to the naval
profession, e.g. in the United Kingdom, where a new Principal Warfare Officer is
now the master tactician responsible under the captain for the fighting of the
ship.
Earlier AIO systems tend to be over-centralized and thus subject to battle
damage. The revolution in microelectronics has, however, helped produce
improved systems with spare main computers and operations decentralized to
miniprocessors and microprocessors in the consoles. Capacity is increased,
which speeds up processing and allows more inputs like passive sonar or
electronic intercept information to be integrated with all the other tactical data.
The use of colour screens and of light pens to interact with the screen makes
systems easier for sailors to operate, and reduces the training load. The trend is
to increase decentralization still further with smaller – and cheaper – computers
in ‘autonomous intelligent consoles’ spread over the ship linked by protected
data highways. Each part of this system could operate independently if required.
Such modular systems also lend themselves to installation in small surface
combatants and MCM vessels.66
Computerized AIO systems are now being designed for submarines whose
sensors now produce so much data that automation of its handling is a necessity.
The new British Submarine Command System (SMCS) now under development
is designed to

relieve the human link in the command of many of the operations currently
necessary to provide the tactical picture. The Command is thus freed of the
more routine and tedious operations which can be carried out by the
computer, enabling the Command to concentrate more on the decision-
making process which, in underwater warfare at least, is a much more
complex and vital problem than in the surface or air environment.67

That very complexity, however, mandates a powerful and complex system,


especially if a high degree of automation is required. The Americans have been
facing serious problems with their SubACS (Submarine Advanced Combat
System). The problem has been to integrate sonar information both in terms of
automating passive detection and putting together the outputs of different sets
operating on different frequencies over different acoustic paths, all in a system
that has to operate in a restricted space.
The manifold limits to the human capacity to respond adequately on a
consistent basis, the sheer data problems created by effective modern sensors,
and the inherent complexities of modern three-dimensional naval warfare,
mandate ever greater mechanization of tactical and command functions.
‘Intelligent’, knowledge-based systems displaying options to commanders will
allow them to concentrate wholly ‘on decision making rather than procedure
management’.68 Even certain decisions may be increasingly taken out of their
hands. Already a degree of automatism is built into weapons control to cope with
the rapidity with which threats can be presented. The human role will be
increasingly to veto rather than to take the initiative. A major problem here will
be that the ship may well be in a situation where the degree of discrimination
required is beyond what can be fed into a computer, however advanced. Only
when unlimited hostilities have begun can the man be left out of the command
loop. Even then there might be inflexibilities in the command system’s
programming that an enemy can exploit. The human commander with his
‘human faculties, such as complex forms of recognition, scepticism, realising the
significance of apparently unrelated occurrences and exercising initiative and
judgement’ will be required at sea for a long time to come.69

Nuclear weapons
So far this analysis of technical trends in naval warfare has ignored the question
of nuclear weapons. This is deliberate, as the use of nuclear weapons as an
element in naval warfare lacks a certain plausibility. Beginning in the 1950s, the
United States, Britain, and the Soviet Union all developed a wide range of
nuclear weapons for use at sea: nuclear depth charges and nuclear-armed
torpedoes to help solve the problem of the high speed submarine and make up
for sensor inadequacies; air-dropped nuclear bombs and warheads for cruise and
ballistic missiles to deal with large enemy surface combatants such as cruisers
raiding NATO supply lines, or aircraft carriers threatening the Soviet Union
itself.70 The USSR even developed plans to call down long-range, land-based
nuclear missile ‘fire’ on NATO naval concentrations, the most useful Soviet
intercontinental ballistic missile of the 1960s, the SS-11 being originally
designed for this purpose by the Chelomei naval missile design bureau.71
Many of these weapons are still available (although the United States has
unilaterally announced the withdrawal of its nuclear ASW and AA missiles) but
their use raises a number of serious problems.72 It is hard to see their relevance
in other than a general nuclear war, in which naval operations would probably
cease to be very important (see Chapter 9). Second, the weapons themselves
would have most undesirable side effects that would impact on the user as much
as the enemy. Nuclear depth charges would probably cause so much
reverberation that they would seriously degrade sonar performance for some
time after an explosion. Nuclear air defence would create considerable problems
of electromagnetic pulse and ionization around a task force just at a time when
clear sensor information was at a premium. These drawbacks are combined with
the problem of political acceptability that nuclear weapons face when surface
ships tend to be turned away from ports. The continued deployment of nuclear
weapons in surface ships, if not carriers or submarines, is thus highly
questionable. The conventional munitions of the future will be even more
capable of carrying out the ASW and anti-shipping tasks than they are today,
without resort to nuclear weapons. Indeed, to argue that nuclear weapons are
necessary for certain tasks is to say that those tasks are impossible, as nuclear
release will in all likelihood not be granted purely as a results of the dynamics of
the war at sea. As long as one side in the superpower naval balance deploys
nuclear weapons it would be imprudent for the other to get rid of all means of
retaliation at sea, but this should not require a stockpile of any great size or the
wide deployment of nuclear-armed systems that we have today.
Only two kinds of deployment of nuclear weapons at sea makes much sense.
One is operations against the shore as part of one’s overall deterrent posture. US
carrier aircraft can deliver a range of nuclear bombs accurately over long ranges.
TLAM/N variants of the Tomahawk cruise missile can convert any American
nuclear attack submarine, or even a destroyer, into a potent strategic launch pad.
TLAM/N is able to deliver a 200 kiloton warhead with an accuracy of 30 metres
or so over a range of 2,500 km. The Soviets are developing a similar system, SS-
N-21, as well as the larger, possibly dual capable SS-N-24 submarine launched
cruise missile. Whether, given the problems of arms control verification, such
systems are desirable strategic strike systems is another matter. They could have
a special role as part of NATO’s theatre nuclear forces in the aftermath of the
mutual withdrawal of land-based intermediate nuclear forces (INF), but sea
deployment makes them symbolically much less tied to Europe’s security than if
they were physically in the theatre themselves. Even if put under the command
of NATO’s Supreme Allied Commander Europe (SACEUR), they would
probably be seen as strategic and therefore as incredible of early first use as
those submarine launched ballistic missiles already allocated to SACEUR.
Moreover, the US Navy has always argued against any limitation on SLCM (sea-
launched cruise missile) carriers’ movements being involved in specific targeting
plans, especially the overall strategic SIOP (Strategic Integrated Operational
Plan). TLAM/N is the US Navy’s own strategic reserve which it will remain
most reluctant to share. It would be most ironic if converting their surface
combatants into nuclear launch pads came back to haunt the US Navy, putting
unacceptable constraints on their freedom of action. Navy attitudes to TLAM/N
might well then change.73
The other naval nuclear role with some logical stature is the French one of
use in pré-stratégique strikes to demonstrate resolve to initiate the use of
national strategic forces. Large Soviet naval targets might be more suitable
candidates for such desperate measures than land targets, and there is thus some
rationale for plans to procure in 1988 twenty-four ASMP 300 kiloton, 250 km
range, inertially guided, ramjet powered missiles for the Aeronavale’s carrier-
borne Super Etendards. The latter are already configured to carry the 20-25
kiloton AN-52 free-fall bomb. The light carrier Foch came into service as an
ASMP capable ship in the summer of 1988, and the new nuclear-powered carrier
will also field the weapons.74 One wonders whether the nuclear bomb capability
announced for Britain’s Sea Harriers reflects similar thinking; it is hard to think
of other reasons for so configuring these short-range aircraft in this role.75
Most important in general strategic terms, and the main reason that at least
one superpower (the Soviet Union) needs to be able to use the sea, is the
deployment of submarine-launched ballistic missiles (SLBMs). Solid, or in the
Soviets’ case, storable liquid fuelled, long-range ballistic missiles carried in
submarines are acquiring ranges almost comparable to land based ICBMs (over
7,000 km or more). They can, therefore, be fired by one superpower at the other
from waters close to home. Their throw weights are also becoming sufficient to
give them a generous loading of multiple independently targetable re-entry
vehicles with a large footprint, i.e. an area in which an individual missile can
attack separate targets. All five current nuclear powers deploy SLBMs and, if not
capable of any other modern long-range nuclear delivery mode, deploy them
alone.
Traditionally, SLBMs have not had a sufficiently high combination of
accuracy and yield to attack hardened targets, such as missile silos or launch
control centres. They have, therefore, been directed against soft military targets
and urban industrial centres, sometimes whole areas. The apparent British plan
to direct all sixteen of their deployed Polaris missiles, each with 2-3 warheads, at
one area, Moscow, is a somewhat extreme variant of this thinking.76 Now,
however, the combination of yield and accuracy of SLBMs is approaching the
point where they will be usable to attack hardened targets. The US Trident D-5,
planned for the 1990s for deployment by the US and Royal Navies, will be able
to place its eight 475 kiloton MIRVs accurately enough to take out the hardest
target. The CEP (circular error profitability – the radius of a circle in which 50
per cent of the warheads are expected to fall) is about 250 metres, similar to
current ICBMs.77
This enhanced accuracy is not just a result of missile improvements,
including updating the inertial guidance system using star sights (the latter
technique is also used by the USSR), or NAVSTAR (navigation system using
tuning and ranging) satellites, but also by providing the launching submarine
with means of updating its own inertial navigation system. Trident submarines
will be able to scan the ocean floor, rather as a cruise missile scans the ground
over which it is flying, and, to the same purpose, to update the on-board inertial
navigation system to achieve pinpoint accuracy. Naval forces will thus soon be
physically capable of carrying out the ultimate operation of modern war, the full
range of strategic nuclear attacks against the homeland of the other superpower.
Chapter 6

Platforms

By naval ‘platforms’ we mean the vehicles by which military power is exercised


at and from the sea. These have for a long time involved much more than ships.
Aircraft and submarines are now as important parts of the maritime environment
as surface ships, sometimes more so. Space platforms are already making an
impact at sea, and may do so to an increasing extent in future. But before we
move to assess the importance of each of these means of exerting force at sea,
we must look at the projected evolution of the main mode of sea use that these
platforms will be protecting – namely the ships, designed to carry materials of
greater human and economic wealth than the instruments of destruction.
Shipping development sets the context for naval development.
There can, as argued elsewhere in this book, be little doubt that the sea will
remain the world’s most important long-distance cargo transport medium and
that sea transport will continue to be man’s primary purpose in ‘going down to
the sea in ships’. It seems unlikely that for the foreseeable future those ships will
be anything else but the familiar ‘displacement’ surface vessels, floating at the
sea/air interface. The now familiar mix of tankers, bulk carriers, container ships,
and general cargo ships will continue to carry the world’s seaborne trade.
Economic factors more than technical ones will probably prevent any major
revolution in design or characteristics that will fundamentally affect the current
situation. The displacement ship is simply the most efficient way of carrying
material, and only in certain specific cases where quite literally ‘time is money’
(such as carrying people) is it worth paying more for the displacement vessel’s
major drawback, restricted speed.
As we have already seen,1 ships have been getting slower and it seems
unlikely that there will be a major increase in ship speeds, at least as long as
diesel oil remains relatively expensive, but not so expensive as to justify
investment in alternative fuels. It seems unlikely that nuclear power will be
generally adopted in merchantmen for cultural reasons as much as technical and
economic. The main speed bracket for ships will probably, therefore, remain
between 15 and 25 knots.
Although more novel types of vessel might appear, it seems unlikely that they
will become a significant form of commercial sea use. Cargo-carrying
submarines have never gone beyond the early German experiments of World
War I, experiments that reflected Germany’s inability to use the surface of the
sea rather than any advantages of submerged transport. Nuclear power has made
submersible merchantmen more practical technically, if not economically, and
concepts have been suggested, especially for use in the Arctic as polar oilfields
are discovered and suspected. But it seems unlikely that the high cost of
construction and operation of commercial nuclear submarines will see any but
the most limited construction of such vessels. Indeed, submarine barges drawn
by ice-breaking surface tankers would seem to offer a much more cost-effective
solution to the problems of arctic oil transport if pipelines are impossible or
inadequate. Civilian submarines may proliferate in the future, but their role will
probably be connected with the growing research into exploration of the ocean
depths rather than as long distance transport.
Surface skimmers of various kinds will continue to be significant in short sea
and riverine transportation, but, once more for economic reasons, it is difficult to
foresee any use for them in commercial oceanic crossings. It is possible that
some skimmer technology, e.g. hydrofoils or side-wall air cushion vehicles
(surface effect ships) might be useful on short and medium-length routes with
especially high value cargoes. A 15,000 ton ‘Atlantic Ferry’ has even been
suggested, capable of carrying trucks (or 2,000 cars and their passengers) say
from the USA to Europe, cruising at 130 knots! Power, however, would be a
problem for such a vessel, as nuclear energy, with all its problems, would
probably be essential. It is also hard to see who would produce the necessary
investment. ‘Wing in ground’ (WIG) machines which use the ground effect to
skim across the waves also might have greater potential for fast oceanic
crossings. Gas turbines might be used for propulsion, although fuel storage
would be a problem. Again, initial investment could well prove an even more
insuperable difficulty.2
Indeed, all novel forms of sea transport have the overwhelming problem that
there are cheaper and more established alternative means. Only if states wish to
ignore economic factors for political, including strategic, reasons can this
financial strait-jacket be overcome. It is well known that the Soviet Union, given
its socio-economic system, has been able to put such factors high in its priorities
for the impressive build-up of its mercantile marine. The United States, however,
provides a more interesting example of the interplay of ‘market’ forces and
politico-strategic factors. Now ‘Sealift’ has been officially defined as a naval role
it is just possible that noncommercial considerations may come more into play in
the development of fast vessels to carry supplies and fuel. Multi-hull vessels
might lend themselves to such tasks, such as catamarans or even SWATH (small
waterplane twin-hull) designs. The latter, while much less economic than
conventional ships at normal operating speeds, offer less resistance at speeds
above 37 or so knots.3 They might, therefore, offer possibilities for ultra-fast
cargo transports. The investment implications, however, are so enormous as to
make it difficult to foresee the circumstances in which such ships would be built.
For the foreseeable future, therefore, naval warfare is going to continue to
revolve around relatively slow ships. This means that platforms will be required
of similar endurance and performance, i.e. military surface ships. The inherent
load-carrying advantages of the surface ship also make it an excellent military
platform in its own right. The surface ship’s endurance, as Admiral Hill called it,
its capacity for ‘being there’ in a wide range of conditions, give it an unrivalled
capacity to maintain a patrol when a military or constabulary presence is
required.4 The surface ship also has unrivalled capacity to operate in three
dimensions, and to communicate with other platforms.5 It can even act as a
mobile platform for other platforms, notably aircraft. Navies still cannot do
without surface ships and will not be able to for a long time, if ever.

Surface warships
The term ‘warship’ covers a very wide range from the small patrol boat to the
huge super-carrier. Naval planners must choose which are necessary for their
strategic and operational requirements. For navies whose main role is
constabulary, relatively small but high endurance vessels with a very limited
armament and probably some helicopter capability are the best solution. There
are a range of such vessels around the world’s navies and coastguards today as
the constabulary responsibilities of nations at sea have expanded. Archetypal
vessels are the 1,150 ton Icelandic ‘gunboats’ of the Aegir class, diesel powered
with a small 22-man crew and a speed of 20 knots. Some coastguards need much
larger vessels (in the 3,000 ton bracket) for higher endurance, e.g. the US
Coastguard and Japanese Maritime Safety Agency. For states with very short
range responsibilities, e.g. the Bahamas, a 100-ton patrol vessel will do.
Armament need only be very limited, the Icelandic vessels carry ninety-year-old
Danish six-pounders. Toughness, seakeeping and manoeuvring characteristics
are much more important for such vessels even in clashes with quite major naval
powers over fisheries, as the British found out to their cost in the Cod Wars of
the 1970s. Such patrol vessels, however, stand little chance against more
aggressive opponents, as the Bahamians found out in 1980 when their patrol
vessel Flamingo was sunk by Cuban Mig-21 fighters. If patrol duties move into
shallows or shoals then hovercraft offer some advantages as patrol craft,
combining speed and ubiquity with a certain level of endurance and an ability to
land as well as just inspect.6
Many more combatant navies use their smaller MCM vessels for constabulary
duties, as the characteristics of an MCM vessel are similar to those of a
specialized patrol craft. The British coastal MCM vessels, built to counter the
Soviet mining threat of the 1950s, found their major role in policing and
constabulary duties all over the world. The problems here are that the increasing
sophistication of MCM vessels and the need for intensive training in their
primary role can make diversion of these valuable assets to other tasks difficult.
A contemporary coastal mine hunter, with its computer-assisted action
information system, sophisticated sonar, acoustic magnetic and wire-sweeping
equipment, and guided remote-controlled mine locators, is an expensive warship,
not available in large quantities. For this reason the United Kingdom, for
example, has turned to simpler, more specialized, vessels for new MCM
construction. Trawlers can also be pressed into service in emergencies for certain
simpler MCM duties, e.g. the sweeping of deep-laid ASW mines. Given the need
to keep signatures at a minimum and the coastal nature of the work,
minesweepers are seldom much over 1,000 tons.
Serious fighting against other surface craft or aircraft is not the function of
any of the above vessels. For this duty much heavier armament and high speed is
required, at a cost in endurance. The world is now full of fast-attack craft armed
with guns, missiles, and torpedoes for use against other fast-attack vessels as
well as larger ships. They complement aircraft in coastal defence and operations
in restricted waters, being especially useful at night. They can also project power
over relatively short distances, especially as their size and armament increase, as
the Israelis have shown in the coastal bombardments carried out by their 415 ton
Reshef class boats. The Israelis have indeed become masters of fast-attack craft
operations, even giving some of their latest vessels helicopters for over the
horizon targeting of long-range Harpoon missiles. Fast-attack craft can be very
heavily armed indeed, e.g. a Reshef can carry a 76 mm gun, a Vulcan Phalanx
close range weapon system, 2 x 20 mm guns, four to six 40 km range Gabriel
missiles, and two to four longer range Harpoon missiles. Hydrofoil technology
also lends itself to the fast-attack role, especially for the carriage of missiles and
torpedoes. Such vessels are produced by a number of countries, including China,
whose navy deploys hundreds of fast-attack craft of all types.7
The ‘military sex appeal’ of fast-attack vessels is enormous and they make
junior officers’ mouths water. They will certainly remain a most important part
of the naval scene, making operations in coastal waters dangerous even for more
powerful navies. The attention paid by the Americans to the Libyan fast-attack
craft in the Gulf of Sidra confrontation was noteworthy, even to the extent of a
jumpy USS Yorktown firing a Harpoon missile at a dubious contact, and sinking
it without even being sure of what it had hit.8 The Libyan fast-attack craft, built
in France and deploying western Otomat surface-to-surface missiles, were
probably the most formidable anti-shipping systems available to the defenders of
the Libyan leader’s claims. But aircraft equipped with the same missile might
have been even more useful. Moreover, despite an undoubted ability to conduct
long ocean transits, even quite large fast-attack craft in the 500 ton bracket are at
the margins of seaworthiness for sustained and efficient oceanic operations.
Moreover, their size prevents them from carrying the suite of sensor and control
systems essential for fully effective ASW and AAW duties.
Only larger surface combatants can cope with these demands, the degree of
size and sophistication reflecting the interaction of the duties that the ship is
likely to be called upon to perform, the range at which it is expected to perform
them, the weather conditions in which it is expected to operate, the threat it is
likely to face, its likely uses, and last, but not least, the amount of money the
national finance ministry can be persuaded to make available. Obviously,
different nations can come up with different answers and will continue to do so.
Sometimes the same navy comes up with a complementary set of answers, a
hi/low ‘mix’ as Admiral Zumwalt called it in the 1970s, reflecting the balance of
capabilities it perceives it requires.9 It must also be stated that all these rational
factors are affected by the nature of warships as tokens of national prestige.
Sometimes this dictates a level of capability (and sometimes even an existence)
that a more rational decision might have thought excessive.
If we look at the types of surface combatants being built for the world’s
navies we see an interesting array of corvettes, frigates, destroyers, and cruisers
of different shapes and sizes. First, some words are necessary on nomenclature.
For some time now, navies and naval analysts have been struggling
unsuccessfully to come up with an agreed nomenclature to cover the surface
units of the three-dimensional world of naval warfare that began in World War
II. The old battleship/cruiser/destroyer distinction born at the turn of this century,
a rationale embedded in the surface warfare of the gun and the torpedo, was
already breaking down even before the World Wars saw the development of
ASW and, later, AAW escorts for merchantmen which could not be fitted into
the standard framework. Mercantile escorts were called attractive and often
inappropriate names culled from the days of sail: first sloops, later corvettes, and
then, as the latter got larger, frigates. Destroyers got larger too as their builders
tried to acquire more margins of advantage over fellow navies’ destroyers;
indeed, they began to approach the size of the cruisers of old, especially as they
acquired primarily ASW and AAW roles to deal with the new threats to the
fleets. The only difference between destroyer and frigate became the higher
speed required for operation with the new carrier battle groups, and as slow
mercantile escorts increased in speed themselves to cope with newer submarines,
confusion became inevitable. Different navies used the same word for a totally
different type of ship, e.g. the Americans and French chose ‘frigate’ for their
large destroyer/small cruiser category, a rather different ship from the British
‘frigate’, a less capable ‘sub-destroyer’ primarily mercantile escort category born
in World War II. The USSR ignored the problem and began a new system of
‘large’ and ‘small anti-submarine ships’. Some countries almost deliberately
produced misleading ‘small’ titles for new construction, in order to ease them
through the bureaucratic process, notably the British with their County class
destroyers and Invincible class cruisers.10
The tendency of Western analysts to call Soviet ‘large antisubmarine ships’
cruisers, and the demise of the traditional large cruisers of the Dreadnought era,
led the US Navy to reassess its nomenclature and come up with perhaps the best
solution yet to the problem. In a major reclassification in 1975 it sensibly
redesignated most of its frigates as cruisers and the rest as destroyers. It joined
the rest of NATO in classifying its ‘destroyer escorts’ as ‘frigates’. Yet even this
system is now beginning to go fuzzy around the edges as the US Navy plans to
build destroyers that will be as big as, and more capable in every way than, most
current cruisers. Attempts to produce a comprehensive and widely acceptable
system of nomenclature seem doomed to failure given the various customs of the
world’s navies. However, if the terms ‘frigate’, ‘destroyer’, etc. are to be saved,
they need some content if they are to have any meaning at all. The following is
offered as one solution which will be followed for the rest of this chapter; it will
be noted that displacements overlap between the various types.

Cruiser: Large combatant with a comprehensive range of capabilities including


area air defence: usually about 7,000 tons in standard displacement to give
sufficient room for all required systems.
Destroyer: Medium-sized combatant of between 2,750 and 7,000 tons with the
speed and capabilities of the most demanding combat operations, including
participation in fast carrier battle groups. Large ‘destroyers’ differ from ‘cruisers’
of equivalent size in that they lack comprehensive capability, e.g. no area
defence AAW system. Large ‘frigates’ usually differ from small ‘destroyers’
primarily in terms of lower speed and more austere engineering arrangements,
e.g. single as opposed to twin screw.
Frigate: Combatant of about 1,750 to 3,000 tons usually optimized for ASW but
with general purpose capability; essentially intended for the escort of non-
combatant shipping, although useful for patrol and limited offensive operations;
capable of ocean-going transits and tasks; usually air capable.
Corvette: Small sea-going combatant of 500 to 1,750 tons of limited endurance
and capability; best employed in coastal escort and patrol work and relatively
low-level operations, although with some potential for higher endurance
operations in the absence of more capable units; not usually air capable.
Only the largest navies, effectively the superpowers, can afford cruisers
which combine all three capabilities, AAW, ASW, and ASUW, in adequate
proportions. The Soviet Union, only just joining the carrier business, has
produced instead latter-day ‘battle cruisers’. These interesting Kiev class ‘heavy
aircraft carrying cruisers’ of 36,000 tons could lay claim to being the most
comprehensively equipped warships in the world. They carry a mix of
helicopters, missiles and rocket launchers for ASW, VTOL fighters, missiles,
and guns for air defence (and limited ASUW) and long-range missiles for
ASUW. They are clearly intended to form a centrepiece of any grouping of
Soviet surface combatants. Equally impressive, if not more so to some eyes, are
the four 24,000 ton Kirov class ‘missile cruisers’ with better long-range missile
capabilities for both ASUW and AAUW than the Kievs’, but only three
helicopters and a generally more limited ASW potential. These would seem to
be primarily fighting mobile command platforms that provide a capacity for
sustained oceanic sea-control operations if required. Although there is evidence
that these large vessels are something of an embarrassment to the post-Gorshkov
navy, all eight of these ships will remain in service for some time as status
symbols as well as instruments of Soviet naval power.
In the absence of carriers the Soviets naturally give a greater emphasis than
the West to anti-surface warfare in the capabilities of their larger vessels. An
impressive example of a modern ASUW missile cruiser of more moderate
dimensions is the 10,000 ton Slava class with serried ranks of SS-N-12 anti-ship,
long-range cruise missiles. It is protected by the Soviets’ latest long-range
surface-to-air missile (SAM) systems, and with torpedoes and rocket launchers
to give adequate ASW protection (although the helicopter is primarily for
surface-to-surface missile (SSM) targeting). A ship of this class will provide a
potent and visible anti-carrier warfare capability of some sustainability: 130 mm
guns even allow the means for some ‘gunboat diplomacy’. Other cruisers, like
the older Karas, emphasize ASW with missiles, torpedoes, and helicopters.
American cruisers reflect their primary role of providing anti-air defence to
carrier battle groups. All carry long-range air defence missile systems as primary
armament and the latest ships are the first to carry the ultra sophisticated Aegis
system. All can also participate in ASW, although only to a limited extent in
some ships due to lack of helicopters; even if a hangar is fitted an aircraft may
not be. Some ships have sacrificed helicopters for cruise missiles but the Aegis
class ships are fully helicopter-capable. SSMs have allowed the US surface
sailors to enhance the capabilities of their vessels in ASUW, and both Harpoon
and Tomahawk missiles are being fitted to the US Navy’s cruiser force. The new
vertical launch system will allow a comprehensive mix of AAW ASUW, and
ASW missiles to be carried, although one suspects AAW missiles will
predominate. It will be difficult, however, to stop the United States increasingly
thinking of its cruiser fleet as a force of self-sufficient ships capable, like the
cruisers of old, of a wide range of duties.
In the 1970s the United States thought of producing larger, more impressive
ships still: nuclear powered ‘strike cruisers’ along the lines of Soviet vessels,
albeit rather smaller. It was hoped that Aegis would give sufficient self-defence
capability to operate independently. The ships would also be armed with SSMs
and an 8 inch gun and would also be armoured.12 Cost, and the supremacy of the
air lobby, helped defeat the concept. Another factor was the reactivation of fast
battleships of the Iowa class. These heavily armed and armoured vessels were
built during World War II, with the high speed of cruisers and the range to
operate with carrier task forces. They now provide large and impressive
centrepieces for surface action groups, supplemented by a screen of cruisers and
destroyers. They can be given SSMs capable of dealing with targets far beyond
the range of their guns, but not only expense advises against turning them into
truly modern warships. Their invulnerability depends to a considerable degree
on their ammunition and other flammables being kept deep inside the ship
behind armour. An aircraft hangar and/or large topside missile magazine, might
well compromise this kind of protection, probably to a decisive degree. As they
stand, battleships may well remain for some time useful as coastal bombardment
and ‘firework display’ vessels; they will also have useful long-range capabilities
at night and in poor weather, especially in conditions of electronic silence. They
will also remain useful symbols of naval strength. There is perhaps still a role for
such symbols but one hopes that the battleships’ active service will not ruin them
as the historical relics they essentially are. The US Navy’s surface sailors are
already pitching for missile-armed battleship replacements which may possibly
see the strike cruiser resurrected in a new, more powerful form. Whether such
ships would be built, however, remains a very open question.13
Other old warships that may well see more service are the surviving large
gun-armed cruisers. The United States retains in reserve two huge 17,000 ton
Salem class ships which it might be premature to write off. Latin America,
however, is the traditional home for these vessels with their imposing appearance
and considerable surface bombardment potential. Peru has spent considerable
sums (that she has had problems raising) on modernizing her two 10,000 ton, 6
inch gun ships purchased from The Netherlands. One has been modified to carry
missiles. Chile, which still retains an old US-built Brooklyn class ship half a
century old, is talking of converting her into some kind of aircraft carrier. Chile,
however, like Pakistan, is ‘trading down’ to purchase very large destroyers as the
‘light cruisers’ to carry her into the next century. She has acquired no less than
four of the British County class destroyers of over 5,000 tons, complete with a
stock of Seaslug SAMs. The latter are probably nowadays more useful in a
surface-to-surface mode (and were used for shore bombardment in the Falklands
War). They add, therefore, to the ship’s Exocet SSMs and 4.5 inch guns and the
relatively modern action information suite to produce quite a capable modern
light cruiser for the old fashioned South American naval environment. One ship,
which had already lost its Seaslug, has been fitted with improved helicopter
facilities and this is the near-term future for half the class.
Modern destroyers are, indeed, hard to distinguish from small cruisers, the
key perhaps being increased role specialization. The superpowers deploy large
and impressive examples of the ‘destroyer’ type, the different operational
priorities of the two navies affecting the types of ship they will take into the
1990s and beyond. The Soviet emphasis on more ‘offensive’ uses of groups of
surface ships in the ASUW and ASW roles, rather than the screening of higher
value units, is still reflected in the new generation of 6,000-7,000 ton destroyers
of the Sovremenny and Udaloy classes. Both are optimized for one form of
warfare only, and neither has anything much beyond a point defence anti-air
capability. Effectively the Udaloy is a smaller platform for the ASW capability
of a Kirov (medium/low frequency, hull-mounted and variable depth sonars;
Helix ASW helicopters; SS-N-14 missiles; and short-range rocket launchers),
and the Sovremenny, a highly specialized ASUW vessel, with 130 mm guns and
modern SS-N-22 100 km range sea-skimming missiles.
Again, USN ships as ‘task force escorts’ show a considerable difference in
emphasis, with greatly enhanced AAW potential in most ships. The new DDG51
Arleigh Burke class vessels are in some way ‘cruisers’ as they will carry Aegis
in a slightly reduced form: their lack of a helicopter hangar, however, denies
them the superior classification. More simple AAW missile systems, although of
medium rather than long range, will remain deployed for some time in older US
destroyers. The large and fast Spruance class ships provide fast, more specialized
ASW task force escorts. Again the tendency to fit ASUW potential is noticeable,
with Harpoon in service and Tomahawk due to appear as ships are refitted with
vertical launch tubes. In terms of general capability, in fact, the US destroyer
force of the future will, to some extent as now, be distinguished from the larger
cruisers purely by size. They will, however, all lack something in comparison to
their larger brethren be it in terms of AAW (e.g. Aegis) or ASW (no proper
helicopter facilities).
Both Britain and France deploy slightly smaller and cheaper destroyers than
the superpowers. This makes task specialization even more essential. The British
solution has been to build the Type 42 destroyer carrying the Sea Dart system for
area AAW and to procure the Type 22 frigates (with low frequency, hull-
mounted sonar, towed arrays in some ships, and better helicopter facilities). The
Type 22 has been developed into a more general purpose vessel in its final
version by enhancing the ASUW potential, and these four fine ships, Cornwall,
Campbeltown, Cumberland, and Chatham will be important core units of the
Royal Navy of the next twenty years. Sea Dart is an excellent long-range missile
but the lack of point defence capability in the Type 42s caused serious problems
in the Falklands. The Royal Navy had to abandon plans in 1981 for more
impressive Type 43 AAW ships which would have carried a mixed area/point
defence missile system.14 The Type 42s are now belatedly receiving Vulcan
Phalanx CIWS that will allow them to cope better with high-threat situations.
Their replacements look like being air defence variants of the NATO ‘frigate’
fitted with some kind of Aegis type air defence system procured under
international auspices.
These ships will be similar in size to the 3,500 ton Duke class frigates of Type
23 now under construction. Much criticized for cost escalation, in truth these
ships possess probably the minimum ‘fleet destroyer’ capability required for the
Royal Navy’s role as an integral part of the NATO striking fleet and ‘first
division’ superpower naval warfare. With very long-range 2031 towed array
sonar as their primary sensor and a hull-mounted, 2050 advanced active sonar
set, the Type 23s will carry a large search/strike helicopter to prosecute sonar
contacts, torpedo tubes for close range ASW, a 4.5 inch gun, and Harpoon
missiles for ASUW and vertically launched Sea Wolf missiles for self-defence.
Their advanced CODLAG (combined diesel electric and gas turbine) power
plant makes them relatively slow (28 knots) but very quiet, an essential for
effective towed array operations.15
France also follows the specialized route with AAW and ASW variants of its
3,800 ton C70 class escort (the C stands for ‘corvette’, a logical but confusing
extension of France’s previous ‘frigate’ category; however, the ships carry
NATO ‘destroyer’ pennant numbers). The capabilities of these ships are not as
great as those of their British counterparts, the Crotale point defence systems of
the ASW ships are much less capable than Sea Wolf, and the American Standard
medium-range system used in the AAW ships is not so long-ranged as Sea Dart.
Crotale is, however, being upgraded to cope with the missile threat and to
increase its range. The ASW ships have low-frequency, hull-mounted and
variable depth sonars, and towed arrays are beginning to be used. The limited air
defence potential reflects the French Navy’s emphasis on operations outside the
NATO area against more limited threats, as well as budgetary constraints caused
by the nuclear emphasis of her strategic policy: the Standard missile fits in the
two AAW ships are from older destroyers. Older, larger frigates also remain: the
two 5,000 ton Suffrans with French-built long-range Masurca AAW missiles
used primarily to escort France’s operational carrier, and the three 4,600 ton
F67s optimized for ASW with Lynx small search/strike helicopters and the
Malafon ASW missile (also carried in the Suffrens). These ships will be in
service until 1998-2003.
Pulling ahead of the European ‘second rank’ navies in the destroyer category,
and in terms of the individual quality of her units, is Japan. One must expect the
future to see her deploying perhaps the most impressive surface combatants
outside the superpower navies. The relatively limited reach so far required for
her forces, and the physically small stature of her personnel, have already
allowed her to produce some outwardly very impressive destroyers, four near
cruisers of about 5,000 tons, which were some of the first to carry large
search/strike helicopters as opposed to just weapons delivery machines. Japan
plans, when funds are available, to procure full Aegis capability for its new
generation of 6,000 ton destroyers. She has just built a pair of 4,450 ton
Hatakaze destroyers optimized for AAW with SM-1MR missiles; guns and
missiles provide supplementary ASW and ASUW capabilities, but there is no
helicopter. Further down the Japanese destroyer scale are ASW ships in the
3,000 ton bracket, a dozen Hatsuyuki class ships recently completed, and eight
slightly larger Asagiris are entering service between 1988 and 1991. All these
fast and rakish ships, powered by Rolls Royce gas turbines, combine a low-
frequency, hull-mounted sonar, large search/strike helicopter, ASROC missiles
and torpedoes for ASW, a 76 mm gun and Harpoon missiles, and Sea Sparrow
and Vulcan Phalanx for AAW self-defence. The Asagiris also carry a towed
array, and the older ships are to be fitted with this vital addition to their
capabilities.
NATO’s major navies deploy destroyers of various shapes and sizes. Canada,
Italy, and the Netherlands have modern ships in the 3,500 ton bracket that will
run on for some time. Canada’s current destroyers are platforms for long-range
search/strike helicopters intended for Atlantic operations. Six new units are
planned to enter service between 1989 and 1992. Italy has four air defence
guided missile destroyers of 3,200-3,600 tons with two 4,400 ton ships
projected. The Dutch have two new 3,665 ton frigates that have capabilities in
all the dimensions of naval warfare, medium and short range AAW, missile
defences, helicopter, ASW torpedo tubes, SSMs, and 120 mm guns. They
provide excellent flagships for groups of smaller escorts, but even the Dutch
could not afford one for each of their escort groups and have had to produce
SAM-equipped versions of their 3,000 ton Kortenaer class frigates as substitutes,
minus a helicopter capability. The standard Kortenaers, ten of which are in
service, are primarily ASW vessels with a medium frequency, hull-mounted
sonar, provision for two Lynx helicopters, and torpedo tubes. Equipment
includes a 76 mm gun, Harpoon, SSMs, and point defence SAMs plus a
Goalkeeper close-in weapons system complete the weapons suite. Towed array,
currently fitted to two older frigates, may be fitted later, and this is also due to be
fitted to the new, slightly larger Karel Doormans that will also carry a large Sea
King type search/strike helicopter.
Related to the Kortenaers are the German Meko frigates of the Bremen class,
which are powered by diesels as opposed to the gas turbines in the Dutch ships.
The Germans plan to procure an improved Type 123 frigate to replace their old
gun-armed destroyers in the early 1990s, while they are currently hoping the
NATO AAW frigate will eventually replace their current guided missile AAW
destroyers. The Kortenaer/Meko design has shown some potential for export
with the former being purchased by Greece and the latter, in heavily modified
form, by Argentina with an armament optimized for ASUW (5 inch gun, Exocet
missiles, 4 × twin 40 mm AA guns) and gas turbine engines of, remarkably,
British manufacture.
As we have already seen, large general purpose ships in the destroyer
category have the ability to be the ‘poor man’s cruiser’, allowing smaller navies
to deploy surface combatants of significant size and prestige and some military
utility in less sophisticated environments against less difficult opponents. An
interesting example is the new Romanian 5,000-6,000 ton ‘battle cruiser’,
effectively an ASUW vessel with Soviet type SS-N-2C missiles, 76 mm and 30
mm guns, 533 mm torpedo tubes, and a helicopter hangar. Along the same anti-
surface lines, although rather smaller, there are China’s fifteen 3,250 ton Luta
class ships, commissioned between 1971 and 1985, which grafted Chinese
copies of the SS-N-2 onto a destroyer hull of Soviet/Italian 1930s conception.
India has been able to take an alternative path, purchasing from the USSR six
newly-built versions of the 1960s vintage 4,000 ton Kashin class, which at least
have some SAMs as well as SSMs. As a domestically produced supplement the
Indians have enlarged the Leander class frigate design into the 4,000 ton full
load displacement ASW frigate Godavari, actually an impressive destroyer, with
the ability to operate two helicopters. Three of these ships are already in service
and the design is being enlarged into the 5,000-6,000 ton gas turbine powered
Project 15 frigate, the first of which was laid down in 1987. Pakistan showed an
interest in purchasing British Type 23s to provide her with something
approaching equivalent capability, but for the time being has settled for old,
former Royal Navy and US Navy frigates to supplement her second-hand
American destroyers.
One of the most interesting features of the next decade or two will be the
extent of the destroyer development of the Chinese and Indian surface navies.
The Indians seem to have abandoned their plans to buy Soviet Kresta IIs, but
instead hope to purchase six to eight modern destroyers of the Udaloy and/or
Sovremenny classes from the USSR, as well as build six of her own Project 15
destroyers in the 1990s. The Chinese are modernizing the Luta to make up for its
manifold deficiencies. Vulcan Phalanx CIWS will provide last ditch air defence,
while American ASW torpedoes and, in some ships, a helicopter, will give ASW
potential. The Chinese have so far failed, for technical reasons, to produce an
adequate SAM of their own and, for economic reasons, to find a foreign
substitute. But a new SAM system seems to be under development now. A
slightly enlarged, modernized Luta design has been exhibited, apparently with
exports in mind. It mounts an American 5 inch gun, eight modern surface-to-
surface missiles (American or Chinese), an unidentified SAM system, Vulcan
Phalanx CIWS, ASW torpedoes, and a helicopter. China will need to replace her
own destroyer fleet with large numbers of this kind of ship before she possesses
a true ‘blue water’ naval capability. This will take well into the next century.
Many of the smaller modern destroyers in the world today are often classified
as ‘frigates’ by their owners. Where to draw the line between the large frigate
and the small destroyer is very much a matter of national taste. Nigeria deploys
an identical ship to Argentina’s German built Meko 360s: in the former navy she
is a ‘frigate’, in the latter they are ‘destroyers’. For the purposes of this analysis,
however, the criterion of suitability for main fleet open ocean operations dictates
a lower destroyer cut-off point of about 2,750 tons standard displacement, at
least for twin-screw ships without some significant limitation on their capability.
Ships marginally larger than this may remain ‘frigates’ due to some significant
limitation or other, e.g. single-screw propulsion.
The largest frigates, rather than destroyers masquerading as such, are those
deployed by the superpowers. American frigates are in the 2,750-3,000 ton
category but only have single-screw machinery which restricts their
performance. They are intended to screen underway replenishment groups and
military convoys rather than act as fleet escorts (although they are often used in
the latter role). The latest US frigates, the gas turbine powered FFG7s, have
comprehensive sonar equipment, multi-purpose missile launchers, large
helicopter hangars, and 76 mm guns, and have been exported to Australia and
Spain as well as procured in large numbers (51 units) by the US Navy itself.
Having already procured 46 very large 3,000 ton Knox class steam powered
ASW frigates between 1969 and 1974, the US Navy, paradoxically for an
essentially power projection force, now has the largest fleet of convoy escorts in
the world.16
The Soviet approach to the larger frigate is also a 3,000 ton vessel with four
long-range ASW missiles, shorter-range ASW rockets and torpedoes, and 76.2
mm, or 100 mm automatic guns. The Krivak class ships, which might be
destroyers but for their limited potential, were subject to a great deal of
unwarranted admiration when they appeared in the early 1970s. A new variant
has now appeared carrying a helicopter instead of missiles. This seems to be a
modification for coastal patrol operations as the ships are operated by the border
guards rather than the Navy. Paradoxically, however, the escort capacity of these
vessels is higher than their more combatant predecessors, which are not suitable
for much more than patrol or picket duty, as implied by their new post 1977-78
classification of storozhevoy korabl (patrol ship).
More classical frigates come in two size bands, one of about 2,500-2,750
standard displacement, and the other between 1,750 and 2,250 tons. The
archetype of the larger class is the classic Leander, developed by the British from
the Type 12 fast convoy escort frigate built with all the lessons of the World War
II Battle of the Atlantic in mind, the campaign that spawned the original modern
‘frigates’. At the end of the 1950s, the British developed the Type 12 concept
into a ‘general purpose’ vessel that was used as a fleet ASW escort and as a
colonial sloop for presence and peacekeeping duties: the first examples were
commissioned in 1963 and no less than seventy-two ships of the class were built
in three countries. The Royal Navy received twenty-six, many of which are still
in service, some in much modified form to enhance their capability as ‘ersatz
destroyers’. The ‘Leander package’ – displacement of about 2,500 tons, medium
gun armament, and/or SSMs, point air defence system, relatively powerful hull-
mounted sonar and weapons delivery helicopter, sufficient radar fit for fighter
control, and aerial and surface surveillance – is the shape of the modern frigate, a
ship of great operational flexibility.17 It can be seen in the British Type 21s built
in the 1970s, and the Italian Lupos built for both domestic consumption and for
export in the 1980s. The latter provide an interesting example of the impact of
operational demands on ship size. Originally conceived as a 2,200 ton convoy
escort, an extra 300 tons were added to make the frigate into more of a small
destroyer for fleet operations with improved hangar, variable depth sonar, and
improved accommodation. A large frigate like this, rather than a full-scale
destroyer, is adequate for higher-level operations over the limited distances and
in the shallow waters of the Mediterranean.
British naval shipbuilders have been constrained by their domestic navy being
more in the market for small destroyers than ‘real’ frigates. Various private
venture designs are currently on offer, however. Vosper Thorneycroft have taken
the Leander concept and modernized it into a 2,800 tonne, general purpose
frigate with diesel or combined diesel and gas turbine propulsion with electric
slow-speed drive. The ship can engage submarines with the helicopter or ship-
launched torpedoes; surface targets can be engaged with Harpoon missiles, 5
inch guns or SSMs carried by its helicopter; and it can defend itself from the air
with Sea Sparrow missiles, and its gun and Phalanx close-in weapons systems.18
The Greek Navy showed some interest in procuring a batch of these vessels but
instead returned to a requirement for large Dutch destroyer sized ‘frigates’. The
2,000 ton frigate bracket is, however, the one usually preferred by export
customers and it is a highly competitive market. The small Italian Lupo design
has been purchased by several foreign customers, Iraq, Peru and, Venezuela.
France sold four 2,000 ton F2000S ships to Saudi Arabia in the mid-1980s. The
latter are general purpose vessels with a small helicopter, ASW torpedo tubes,
3.9 inch and 40 mm guns, Crotale short-range SAMs, and Otomat medium-range
SSMs. The West Germans are building slightly larger MEKO 200 frigates for
Turkey, and Greece has been tempted by these ships as its smaller alternative to
the larger Dutch vessels.19
The 1980s marked something of a slump in the warship market due to
difficult economic conditions in the developing world.20 This made second-hand
warships attractive once more. Indonesia, with a requirement for relatively long-
distance cruising ability due to her paucity of bases in the large archipelago,
purchased 1960s vintage frigates from Britain and the Netherlands. Such vessels
are often recently modernized by their previous owners, and are capable of some
years of extra service: their withdrawal from their home navies is as much due to
financial pressures as technological obsolescence.21 Older ships like the 1950s
vintage Dutch gun-armed destroyers sold to Peru in 1980-82, or the old
American destroyers of wartime construction distributed in large numbers
around the world, can even now be attractive frigate substitutes for navies whose
primary ASUW requirement reflects the naval priorities of an earlier age. Such
vessels may, indeed, have the ‘military sex appeal’ that a more modern ship
lacks. Although negotiating for a batch of domestically built FFG7s, Taiwan
currently relies completely on ex-US destroyers of various wartime classes for
its major surface combatants in the 2,000 ton category. She deploys no less than
twenty-four such ships, the most recent transfers having taken place in 1983.
Several of Taiwan’s ships, like the later transfers to other navies, had been
modernized by the US Navy with ASROC ASW missiles and a hangar for a
small helicopter. The Taiwanese have continued the modernization process,
fitting their locally produced 40 km range Hsuing Feng version of the Israeli
Gabriel SSM, as well as additional light guns and the simple Sea Chapparal
missiles to provide point defence AAW. Sea Sparrow and Standard surface-to-air
missiles are planned to convert ten ships into specialized AAW vessels.22 It is a
tribute to the quality of American wartime design and construction that these
hulls will still be in service over half a century after their construction in the
mid-1940s.
Such old hulls as these, and more modern frigates of similar size, give a navy
the opportunity to carry out sustained operations at sea. This, however, may not
always be necessary. If one is willing to accept serious limitations in seakeeping
and military capability in one or more dimensions, then the corvette of 500-
1,750 tons has some utility, even for superpowers. The USSR which still gives
coast defence a very high priority, deploys a large number of rocket and torpedo
ASW corvettes in the 500-1,000 ton class, as well as missile corvettes that are
really large fast-attack craft. Both types of corvette are capable of limited sea
duty, e.g. in the Mediterranean, but for those states who might need a little
greater sustainability a larger 1,700 ton Koni class is built for export to friendly
states such as Cuba, East Germany, and Yugoslavia. These ships have limited
abilities above, on, and below the seas with ASW rocket launchers, light 76 mm
guns, SA-N-4 point defence missiles and sometimes anti-ship missiles. Such
large corvettes rather than small frigates, notably lacking in air facilities, provide
their owners with a very limited, but possibly sufficient, ability to deploy a little
force a little way from home, but not too far.
Second rank navies with foreign bases and a need for a basic presence in
distant seas also see corvette-type vessels as attractive. With their ‘out of area’
priorities, the French deploy a crude 1,100 ton ‘aviso’ or ‘sloop’ with some
capabilities in all three dimensions, which has also been exported. Their larger
1,750 ton predecessors were also little more than colonial gunboats, even to the
extent of having ASW launchers that could double as shore bombardment
weapons. The Iberian navies, with their colonial background, have found
corvette vessels of this type attractive, but also find difficulties in breaking the
habit, despite such ships’ limited capabilities. A modern Spanish Descubierta
1,720 ton corvette can shoot at surface and air targets with guns and missiles,
and engage submarines with torpedoes and rockets, but all at only the closest
ranges. They can also carry thirty troops each which gives more than a clue to
the type of duties at which they are best. Yet even nations such as Japan feel that
for their shorter ranged and shallow water operations a small corvette type of
ship with ASW rocket launchers (and Harpoon surface-to-surface missiles) is
useful. This DE (destroyer escort) concept is being enlarged with ASROC in
new construction to 1,900 tons, which may better qualify it for the ‘frigate’ title,
despite its lack of a helicopter. With its rather similar Baltic requirements,
Denmark has built similar, but rather smaller, British-designed vessels. The little
1,320 ships were designed as classic shallow-water ASW vessels intended to
work in home waters with shore-based air support. A hull-mounted sonar and
ASW torpedoes would provide the primary systems while ASUW potential came
from Harpoon missiles and a 76 mm gun; Sea Sparrow would provide close
AAW cover. In service the ASUW role seems to have come to dominate: the
torpedoes have not been fitted and the only ASW weapons available are depth
charges.
Helicopters can be added to the large corvettes in the 1,500 ton bracket, but
the aircraft must be small and the operating conditions are always marginal. The
FS1500 large corvettes/small ‘frigates’ built by West Germany for Malaysia and
Colombia have helicopter facilities and the latter even have hangars. Hangars are
being added to Argentina’s similar-sized Meko 140 ships, but the latter vessels in
particular seem very crowded and cramped. Indonesia has fitted a helicopter
hangar to one of its Dutch built corvettes and the owners contemplate landing
and taking off helicopters from the 670 ton corvettes purchased by Ecuador from
Italy in the early 1980s. Middle Eastern navies contemplate helicopter operations
from very small vessels. Indeed, Israel has even added a hangar to two of her
500 ton large fast-attack craft and plans the same for her new 1,000 ton corvettes
proper. Such helicopter operations must be severely limited by weather and are
probably more practical in the Mediterranean than the South Atlantic.23
Corvettes provide a suitable format for the newly industrialized countries to
try their hand at building sophisticated warships. South Korea has produced a
1,600 ton Ulsan design for its own navy; they are basically ASUW ships, but
more than a match technically for the crude 1,200 ton vessels built by North
Korea. China has also long produced ‘frigates’ in this class, enlargements of the
Soviet coastal escort design of the 1950s. China is enlarging this concept into the
2,000 ton class, and South Korea may well follow, both for domestic
construction and export. The latter’s greater access to Western technology gives
its products a distinct technical edge, although at a price. Indeed, as South
Korean labour costs increase its products might well begin to cost almost as
much as European built vessels.24
Warship building outside the traditional naval shipyards of Western Europe,
the Soviet Union, and America may well increase. Already there are clear
economic advantages in contributing as much as possible to the construction of
one’s own ships; doing so is considered an important way of creating and
developing a shipbuilding industry. However, warship design is a specialized art
and the design work still tends to be concentrated in the industrialized world, as
does the provision of advanced weapon systems. This means that the domestic
shipyard may receive little of the total cost paid by its government for the ship.
Domestic construction can also have other disadvantages, e.g. sabotage. The
optimal current compromise seems the co-operative technology transfer deal
between developed and developing countries, with lead ships built in the
originating yard and the rest by the purchaser. India, however, has an ingenious
combination of Soviet and Western technologies. With the Godavari class
destroyer, India shows what can be achieved by the Third World, albeit with
some limited design help from foreign consultants. This seems the obvious and
natural next step beyond building a ship completely to someone else’s design.25
International co-operation may well also become more normal among the
navies of the developed world. No navy, not even the richest, can afford as many
surface combatants as it would like. Costs can be limited be engaging in fully
international projects to spread costs over a number of countries. This latter
technique also has the advantage of making the programme much more difficult
to cancel or reduce. Given all the major European navies’ problems of
maintaining numbers as their older ships approach obsolescence it would seem
that the 1990s might well finally see the adoption of a common ‘NATO frigate’
(actually more a small destroyer), perhaps with different weapons fits for
different countries. There were some difficulties in getting all the nations to sign
up for the project, but it is being pursued with more seriousness than earlier,
similar, but in the end abortive, endeavours.
As for the basic form of future surface combatants, it seems likely that
monohulls will predominate. There is still much room for development in the
surface warship. The application of advanced technologies to monohulls can
improve carrying capacity and combat ability for a given size of vessel. Vice-
admiral Joseph Metcalf, as Deputy Chief of Naval Operations (Warfare) in the
US Navy in 1986-7, supervised the setting up of machinery within the US Navy
to study the impact of the ‘Revolution at Sea’ wrought by both the new weapons
and sensor technology that has increased battle space ‘up, out and down’, and
the control technology that allows highly automated systems. Metcalf’s concept
of the warship of the future is a long, low ‘strike cruiser’ with a speed of 31
knots and a ten decibel radar signature. Most of the ship would be composed of
vertical launch missile cells. Gas turbines with superconducting electric drive
would provide a compact power source for propulsion and, perhaps, beam
weapons. Phased array radars mounted round the whale-like hull would be able
to control missiles, which with command guidance, autopilots and active
terminal homing, would not need illumination of the target from the moment of
launch. A range of missiles would be carried for all tasks, ASW, ASUW as well
as AAW. These ships would devote much more volume to weapons than has
been customary in recent years. Metcalf has been willing to ask the difficult
questions:

... the matter of navigation and signal bridges . . . why [do] we have one on a
modern warship[?] Both add space and weight. Bridges contribute to our
ships’ high radar cross section. Manning both bridges requires people who
would be better used elsewhere in a ship’s limited personnel allowance. The
function and location of Navy ships’ bridges are examples of policies based
on history. Over the centuries, ship control has moved from the quarter deck
on a sailing ship – located well aft, near the helmsman – to its present
location in the bridge. The bridge’s location high in ships is a throwback to
the days when the captain watched the fall of shot. Today, he absorbs the
output of countless electronic devices as he fights his ship from the combat
information center (CIC) .... Could we co-locate ship control with the CIC,
where the captain fights the ship and put both deep inside the ship?26

Like their immediate predecessors these futuristic surface ‘flat iron’ combatants
might well be somewhat broader than classical warships. Contemporary
warships emphasize beam in their designs as a result of the need to combine
maximum topweight with the enhanced survivability gained from the use of
heavier steel building materials. Both the new British Duke class frigates and the
USN DDG51 have length/beam rations in the 7.6-7.7 range, compared to over 9
for British and American equivalents of the late 1970s. Some in Britain
suggested that this should be further reduced to about 5, and a major controversy
broke out between the advocates of ‘long thin’ and ‘short fat’ ships. Broad-beam
vessels usually suffer from inherent problems of higher resistance and inferior
seakeeping qualities. The supporters of the ‘short fat’ ship claimed, however, that
their design overcame most of these difficulties and, indeed, reversed the balance
of advantage. They claimed that their experiments, which included extensive
tests with scale models, demonstrated that, contrary to the conventional wisdom,
a considerable increase in beam is possible for any given speed without
increasing power per ton of displacement. The short fat ship had, they argued, a
superior propulsive coefficient (effective horsepower divided by installed
horsepower) compared to conventional vessels. The reasons for this were,
however, unclear and seemed to be related to dynamic lift. However, such lift
only gives significant advantages at high speeds that would require very large
amounts of power to reach. As far as seakeeping is concerned, a ‘short fat’ hull,
while less wet forwards in heavy seas, is still probably prone to rapid rolling,
creating operational difficulties, notably the launching and recovery of
helicopters, and making life unnecessarily unpleasant for the crew. Even the
small patrol craft already built to the ‘short fat’ design have not, apparently, had
entirely trouble-free careers in difficult sea conditions.
By far the weakest claim for the ‘short fat’ supporters was that a ‘short fat’
ship would be cheaper through its ability to carry more weapons for a given
displacement. If other disadvantages of the design are ignored, it seems possible
that a ‘short fat’ ship should be able to carry more equipment and, indeed, place
its sensors at a higher and more effective level; but there is no way in which such
a ship would be cheaper. Ship steel usually accounts for less than 25 per cent of
the cost of a modern warship (in a British Type 23 it is 17 per cent). Propulsion
represents about another quarter. Weapons and command and control systems
account for over 55 per cent of the cost. If a ‘short fat’ ship can carry more
systems than a ‘long thin’ one it must be more expensive, not less. Eventually
the British Government was forced to commission an independent inquiry which
compared enlarged versions of the original ‘short fat’ frigate design with the
Type 23 frigate. It found that the former were, in fact, more costly than the Type
23, and that overall the more conventional design had better seakeeping. The
inquiry, therefore, found ‘no reason to disagree with MOD’s preference for the
Type 23 and it seems unlikely that major warships of “short fat” configuration
will obtain general acceptance’.27 Indeed, the trend could go the other way, with
‘slender’ vessels having advantages in certain applications, e.g. as a fast ocean
escort for ‘sprint and drift’ submarine detection duties with subsidiary ASUW
capabilities, or as an off-shore patrol vessel with good sprint capability. One less
extreme 2,500 ton ‘modern monohull’ ASW/AAW ship is slightly longer in
relation to beam compared to modern frigates and destroyers and has a
maximum proposed speed of 41 knots with gas turbines and superconducting
electric drive.28
Given the low relative ship steel costs the option of a deliberately ‘enlarged
ship’ might be taken up. A larger ship has better speed and seakeeping qualities
than a smaller one, and if the tendency to fill the spare space with expensive
equipment is vigorously resisted, the result would be a cost-effective answer to
the problem of maintaining the operational effectiveness of a given sensor and
weapons suite at high speeds and in poor weather conditions. In some ways the
current US Spruance class destroyer is such a vessel.29
Double-hulled vessels, catamarans, have certain advantages for specialized
work, e.g. as small, fast patrol boats. Catamarans with their slender hulls are fast
and economical of power, and seakeeping is also good up to a certain sea state. A
planing catamaran is intrinsically more stable than a planing monohull, and foils
placed between the hulls improve both seakeeping and fuel consumption. A
German designed 32 ton Hyuscat-18 was built in Thailand in 1987 along the
latter lines, armed with a Sea Vulcan rotary gun like a helicopter gunship. It is
hoped to be useful for operations both in rough seas and shallow rivers. As for
slower displacement catamarans, their lack of sensitivity to large amounts of
topweight means that propulsion machinery can be mounted high up, away from
underwater shocks and where it gives the least possible acoustic magnetic
signature. This, together with shallow draught, makes for an excellent MCM
vessel. The Australians have pioneered this concept with their Carrington
minehunter now in service, and France is pursuing the idea with its Batiment
Anti-Mines Oceanique due in the 1990s.30
Returning to monohulls, modern design gives planing craft some greater
utility in more difficult sea conditions. It is possible that ‘supercritical’ planing
hulls of the Sea Knife kind may be able to compete with the hydrofoil as a
means of achieving high speeds in heavy seas.31 For the moment, however, the
hydrofoil still offers the best possibilities for high performance vessels required
to operate in a wide range of sea conditions. The seakeeping conditions are such
that ocean-going hydrofoils can be contemplated in the near term. As Michael
Eames has argued: ‘Hydrofoil development should proceed in the direction of
simplification, cost reduction and improved range, even with some sacrifice in
speed. This simplified hydrofoil will be attractive up to about 1,000 tons. In
these smaller sizes the concept offers its unique capability of matching the
seakeeping and exceeding the speed of major combatants.’32
Other types of skimming technology, such as hovercraft and wing in ground
machines, have obvious potential in specialized roles, where a direct interface
with the water is a drawback. Such duties as MCM, amphibious operations, and
patrol work among sandbanks (e.g. in the Gulf) are obvious applications for
hovercraft, and they are already used quite extensively for such duties, especially
amphibious landings which promise a ‘dramatic impact’ operationally.33
Hovercraft can be quite heavily armed: the British Hovercraft BH-7 is offered in
fast-attack craft form with surface-to-surface missiles, point defence SAMs, and
two 30 mm guns. An ASW version with dunking sonar is also available.34
Such craft are, however, only suitable for coastal duties. Of much higher
endurance is the hovercraft’s offspring, the ‘surface effect ship’ or SES. Edward
Butler has summed up the advantages of SES: ‘As compared to an aircushion
vehicle (ACV), the SES hull, which pierces the water (hence non-amphibious)
has less air leakage, better longitudinal stability, and an acceptable form of
utilising water propulsion systems, which, at speeds to about 120 knots are more
efficient than air propulsion.’35 (Normal hovercraft use aircraft-like propellers).
The Americans seemed set on utilizing SES technology to produce a ‘100 knot
navy’ in the 1970s. Two 100 ton prototypes were tested extensively but in 1979,
an 80 knot, 3,000 ton prototype was cancelled only three weeks prior to its
construction beginning. One problem was finding a mission for such a large SES
with such a different operational envelope from the rest of the surface fleet.36
Development continued in the 1980s with SES-200, a 200 ton trials vessel. The
Coast Guard purchased three 110 ton Sea Bird vessels for coastal patrol, while
the Navy ordered a class of 400 ton MSH mine countermeasures vessels for in-
shore duties. The latter programme ended in fiasco when, after the first unit had
been laid down, the design proved too vulnerable to shock. MSH-1 was
abandoned in 1986, the SES design being replaced by a conventional Italian
designed monohull.
Hardly more fortunate were plans to procure a 100 ton surface effect ‘special
warfare craft medium’ (SWCM) for the US Navy’s SEALS (sea-air-land special
forces). ‘Technical and scheduling problems’ intervened and the contractor went
into bankruptcy. The status of the SWCM programme has been described as
‘tenuous’, and again foreign monohull alternatives have been suggested.37
Despite all these tribulations, consideration has been given to a larger Coast
Guard cutter with a helicopter deck and a 600-800 ton Navy patrol craft for
continental shelf ASW, surveillance operations ‘associated with national interest
and activity in hemispheric security’,38 and a fast replacement for current
landing craft.
Norway has chosen the small SES catamarans for its new MCM programme
and is considering the concept for a fast-attack craft. Given, however, the
troubled development history of the SES concept so far, the application of the
technology to fully ocean-going platforms is much more uncertain. Such
surface-effects ships must (a) have the potential to be integrated tactically and
logistically with existing fleet units, and (b) be an incremental development of
current designs to minimize both technical risk and development costs. As it is
few navies, if any other than the American Navy, have sufficient marginal
expenditure to find such development attractive. Edward Butler, a leading
exponent of SES technology, argues that the most desirable and practical large
SES would be an escort/picket vessel in the 1,500-2,000 ton class. This would
possess the seakeeping qualities of a larger ship and would enjoy the tactical
advantages of being able to work in quiet waters listening on towed array sonar
at normal fleet speed, but being able to work up to twice that speed to reposition
itself in relation to the convoy or task force being protected. It would also be
able to move quickly as a sensor or weapons platform as part of the outer AAW
defence perimeter of either formation. Such a variant might carry Aegis and up
to 80 missiles in vertical launch tubes. As ASUW derivative has also been
suggested as a stealthy platform for high speed cruise missile attacks at ranges of
up to 1,000 miles from the main formation. Such a vessel would be ‘the least
expensive missile attack mode that could be put together should this mission be
needed. Its survivability should be high.’39 Finally, a relatively small SES might
be useful as a highly mobile aircraft platform given both the increasing use of
aircraft and the ability of an SES to ‘provide more flight deck area per
displacement ton than any other’ hull form.40 Such SES ships could be made of
one basic form interchangeable between missions.
A more suitable ‘exotic’ ship for the larger surface warship categories is the
SWATH vessel. SWATH offers better performance in terms of the surface
warship’s main essential attributes, its capacity to operate for long periods in a
wide spectrum of sea states and its capacity for effective presence close to the
form of sea use being safeguarded. Compared to a monohull, SWATH offers
greatly improved seakeeping and higher speeds in rough conditions. Rough seas
act as a severe constraint on the operations of conventional warships. A 1983 US
Navy study of its cruisers, destroyers, and frigates demonstrated that they could
achieve full speed only up to low sea state 5 (moderate conditions with six-foot
waves). In the North Atlantic in winter, frigates could only achieve full speed 30
per cent of the time, and destroyers and cruisers 55 per cent of the time: on a
whole year basis the respective figures were 45 and 70 per cent. Low speed was
not the only problem: helicopter operations, hull-mounted sonar searches, and at-
sea replenishments were all deemed equally impossible by the USN beyond low
sea state 5, conditions which occur almost 40 per cent of the time north of the
equator. SWATH vessels would be much less limited, especially for air
operations which would also be facilitated by the larger and higher deck area
available.41
For ASW operations SWATH vessels provide excellent sonar platforms. They
are perhaps the optimum towed array vessels, being able to move steadily in any
direction in the widest possible range of sea conditions. The low deck motion,
dry decks and inherent quietness of SWATH ships are also fundamental assets in
this role. As platforms for hull-mounted sonars a SWATH ship’s transducers
would not be prone to coming out of the water: moreover, the twin cylindrical
hulls would be excellent mountings for the wide aperture conformal sonar arrays
that are expected in the future. SWATH’s suitability for signature reduction, as
well as improving offensive ASW potential, also enhances survivability,
especially as a broad SWATH vessel is also less prone to catastrophic damage
from a hit on one side of the waterline. The SWATH layout does have
disadvantages, e.g. its inherently higher frictional drag leads to lower speeds in
calm waters for a given power output and hence higher fuel consumption at
normal cruising speeds, excessive draft, higher construction costs, and inherent
difficulties in finding arrangements for weapons and machinery. Although
steadiness is good, stability, the ability to return to normal after being disturbed,
is poor. SWATH’s vulnerability to flooding also seems ‘excessive’ to some.42 On
balance, however, SWATH’s advantages seem substantial for ships designed to
operate in rough conditions. Such vessels could be smaller and cheaper than
conventional monohulls; advanced electric propulsion systems, coupled with
contra-rotating propellors, might solve the fuel economy problem. The US Navy
has ordered SWATH ocean surveillance ships (T-AGOS) for use in especially
heavy sea conditions, and a SWATH frigate has been seriously considered.
Studies demonstrated that on a calm water basis a SWATH FFG was 13 per cent
more expensive than a monohull but that on a seakeeping basis she was 9 per
cent cheaper. Moreover, ‘incremental improvements in seakeeping are more cost
effective using a SWATH hull form’.43 As the United States is currently
following a co-operative policy for FFG procurement and a low-risk monohull
has already been chosen for this craft, the near-term prospect for a USN SWATH
frigate seems low. Nevertheless, such a ship may well have been built
somewhere by the end of the 1990s or in the early part of the next century, and
the application of SWATH to small carriers or amphibious assault ships is a
realistic possibility in the twenty-first century.44
More visionary proposals still have appeared to exploit SWATH to give
surface units radically new shapes: Captain S.E. Veazey, ex Chairman of Naval
Systems Engineering at Annapolis, has proposed a family of ellipsoid saucer
shaped vessels, the optimum ‘shape in which to distribute antennas, sensors and
defensive and offensive weapons to obtain the most effective hemispheric
view’.45 Perhaps the greatest problem with such vessels is their unconventional
‘science fiction’ shape. This more than other considerations will probably tend to
rule them out. Attracting scarce defence resources is perhaps going to be an
insuperable problem to such radically new vessels, although they might evolve
over the next century as SWATH vessels of more conventional layout become
more common.
One recently produced analysis of the ‘most promising options’ for various
warship designs has balanced the various factors for each hull form.46 Table 6.1
summarizes the situation.

Table 6.1 Most promising future seaborne platform options


Ships’ hull Small Escort General AA ASW Search Ocean Inshore Assault
forms aircraft carrier purpose ocean ocean & patrol patrol landing
carrier cruiser escort escort ASUW vessel vessel craft
(ocean ocean
combat escort
ship)
Conventional 6,000- 1,500-
monohull 12,000t 3,000t
Slender 800-
monohull 1,500t
Semiplaning 200- 200-
monohull 400t 400t
Enlarged 6,000-
monohull 12,000t
Swath 12,000- 6,000- 3,000-
24,000t 12,000t 6,000t
Hydrofoil 400-
800t
Hovercraft 100-
200t
Surface 1,500-
effect ship 3,000t

One important factor in future warship design will be the need to economize
on manpower. This affects all countries as trained manpower will be a scarce and
expensive resource, even to states with high rates of population growth. Modern
ships are much less manpower intensive than their predecessors. A World War II
destroyer, or even a 1960s vintage Leander class frigate, required 250-275 men
as ship’s company; the new Type 23s require just over 150. This may be taken
further still. A British consortium (CAP Scientific Ltd. and YARD Ltd.) has
produced a design study for an FCF-50 4,200 ton, 28 knot ‘frigate’ manned by
only fifty people. The ship is heavily armed and carries a large helicopter. The
very low ship’s company was obtained by minimizing the at-sea maintenance
requirement, within a sixty-day patrol cycle for the ship. On-board systems are
fully automated and designed for the maximum availability and maintainability.
Administration and support would be shore based with crews being rotated
between patrols. The ship would be highly automated with data fusion as well as
decision support at all command levels. Faults and damage would be diagnosed
and repaired automatically, and duplication of systems and control centres would
assist with damage control. The project shows what might be achieved in
manpower savings in the near to medium term, at least for navies that do not
need to operate over global distances and too far away from land.47
One factor in reducing manpower demands has been the adoption for
warships of the internal combustion engine in its two main forms, the gas turbine
and the reciprocating diesel. These are often combined in various ways to exploit
the best characteristics of each type of engine to optimize fuel economy over a
range of speeds. A diesel might provide cruising power while a gas turbine
provides high speed capability. This is combined diesel or gas turbine
(CODOG). A more complex arrangement with either diesel or gas turbine power,
or a combination to obtain the highest power outputs, is called CODAG –
combined diesel and gas turbine. Ships with only gas turbine plants usually have
small cruising turbines (e.g. the Rolls Royce Tyne) and large full-power turbines
(e.g. the Rolls Royce Olympus). This is known as COGAG (combined gas and
gas) if the two can be used in combination, or COGOG (gas or gas) if the two
work separately. Gas turbines may well become more fuel efficient but they will
remain expensive. It is significant that the Royal Navy, which has built some of
the largest gas turbine powered warships in the world (notably the Invincible
class carriers), is returning to diesel as its main cruising power plant for its new
generation of frigates, although with gas turbine boost for high-speed dash (an
installation called CODLAG). The new French Cassard class guided missile
destroyer, on the other hand, is all diesel (CODAD). Without extra machinery,
diesels are now able to produce adequate power levels for high service speeds
and with their less technically demanding maintenance requirements will remain
attractive, especially to less sophisticated navies. Dispensing with turbine air
intakes and exhausts also helps save space in smaller vessels. Electric drive now
offers excellent quietness, even to the inherently noisy diesels, and will be
adopted more widely. Low-temperature conducting techniques offer future
reductions in weight and volume together with a power storage facility for beam
weapons or pulse radars. Electric drive might also allow propulsion to be put in
external pods with advantages of redundancy and dynamic efficiency.48
Oil-fired external combustion steam turbines may be abandoned in new naval
construction, although there are possibilities, especially in the Third World, for
new remotely-controlled, high-technology, low-maintenance systems reversing
this trend given the ‘unreliability and high levels of maintenance with associated
high operating costs’ of gas turbines and higher-powered diesel installations.
Fuel for the latter types has to be of high quality and specific consumption is
also high.49 Many countries would be willing to continue to pay this price for the
flexibility of high-powered internal combustion engines but others may not. The
Soviets may continue to employ oil-fired boost boilers in future nuclear-powered
vessels, as they do in the current Kirov class ships. As for nuclear power itself,
this looks like being restricted for the foreseeable future only to the largest
warships, and these are almost all aircraft carriers. Even the United States, which
in the 1960s and 1970s built limited numbers of nuclear-powered cruisers in the
9,000-10,000 ton bracket for carrier screening duties, seems to have little interest
in building any more of the ultra-expensive and (for their role) somewhat over-
sized units.
Nuclear propulsion brings us to aircraft carriers, the most impressive and
powerful vessels of the present time and the near future. Their demise has often
been prophesied, but the US building programme and their service life means
that the twenty-first century will still see monster carriers of almost 100,000 tons
full-load displacement steaming majestically across the world’s oceans. Such has
been the achievement of former US Secretary of the Navy John Lehman, a
carrier aviator himself. Before achieving his office in the Reagan
Administration, Dr Lehman wrote an academic study setting out the arguments
in favour of these impressive behemoths.50
Size, he argued, gives significant economies of scale. Not only does it require
less power per ton to drive a supercarrier at speed through the sea than a smaller
vessel, but the volume of any container increases as the cube of the ratio of
enlargement of its dimensions. The lower tendency to pitch and roll of larger
ships and their greater seaworthiness means both lower rates of operational
accidents and a narrower band of conditions in which aircraft cannot be operated
for safety or physical reasons. Only a larger carrier can operate a comprehensive
air wing of fighters, day and all-weather strike, airborne early warning,
electronic warfare, ASW, and tanker aircraft. This is due to the big carrier’s
ability to operate a wide range of aircraft types, as well as aircraft in large
numbers. With four catapults and large lifts and a large parking area, a
supercarrier can also generate the very high sortie rate required in the vital early
stages of operations in high-threat environments.
On the question of the apparent vulnerability of the large ship caused by its
size and visibility he countered:

It is one thing to put a hole in the deck of a carrier, it is another to damage it


sufficiently to prevent it from launching aircraft, and quite another still to
sink it. The carrier is the least vulnerable ship to destruction or sinking
because of its large size in relation to the warheads constituting the threat,
greater relative compartmentalisation, massive protection of propulsion and
ordnance, and its structural strength necessary to handle the stress loads of
catapulting and arresting 35 ton aircraft.51

He went on to assert that ‘short of a direct nuclear hit, the carrier is the least
sinkable ship ever built, and the larger its size the less its vulnerability’.52 Given
the fact that any carrier is going to be a relatively obvious target, the greater
toughness of the large ship, as well as its greater fighting capacity, makes it the
better platform. The supercar-rier’s Achilles heel is its vulnerability to
underwater damage from mine or torpedo. If not sunk, it might well be put out of
action and prevented from taking any part in further operations. A naval force’s
supreme asset would have become its greatest liability. This means that even
supercarriers, formidable weapons systems though they undoubtedly are, must
be deployed and used with a certain level of prudence.
Only one navy has both the resources and expertise to operate this level of
carrier capability in the foreseeable future. The Soviet Union seems to have
evolved its large cruiser into an interesting new 60-70,000 ton carrier concept to
operate both conventional aircraft and STOVL machines. The first two ships
being built (the first of which is to be called Tblisi) retain many of the
characteristics of their cruiser forbears, i.e. reliance on SSMs as their major
ASUW capability and advanced SAMs for air defence. A ski-jump bow will
allow the efficient operation of improved STOVL aircraft and a Yak-41 Harrier
type aircraft is under development, but not without serious problems which may
have enforced a reversion to experiments with more conventional fighters using
catapults and/or ramps. Now a third larger carrier has been laid down. It is too
early to say what will become of this programme but there seems little doubt that
even if conventional fighters are carried the main roles of the ships will be AAW
and ASW, and they will have only limited power projection compared with US
ships.53
Although the American view seems to be that 40,000 tons is the minimum
size of carrier for the operation of normal high-performance aircraft by means of
catapults and arrester gear, other countries disagree. France plans to build new
ships that are reported to have an intended full-load displacement somewhat
below this figure. These vessels, nuclear powered like the new US and Soviet
ships, will enter service around 1996 and 2001 to replace France’s current pair of
22,000 ton carriers. The new design tries to achieve maximum utilizable deck
area on the limited displacement. The new ships will not be able to launch and
land simultaneously, due to the layout of the catapults, and they will not be
capable of very intense operations. They will carry between 35 and 40 aircraft, a
mix of the projected Rafale based ACM fighter bombers (the first ship might
carry an interim air group of Super Etendards and F-18s), and helicopters. They
will not be in the same class as US carriers but, in the many low-intensity
operations in which France thinks of using its naval power, such an air group
may well prove to be more than enough. As at present, demonstrative ‘pré-
stratégique’ use of nuclear weapons will be another important role for these
ships.
For navies of similar, or lesser, means and facing relatively limited threats the
smaller carrier still offers a most useful list of assets, as well as the prestige of
having carriers in the fleet. A surprising number of navies operate small carriers
and both the evolution of the helicopter and the development of the STOVL
aircraft has enhanced this trend. Old aircraft of the 1950s and 1960s that could
relatively easily be operated from modified light carriers of World War II
conception can be replaced at similar levels of capability by new STOVL
aircraft, notably Harriers and Super Harriers. Smaller carriers come in a variety
of shapes and sizes. They might be light carriers of an earlier age, soldiering on
for as long as their refurbished engines or solid hulls can carry them. India’s
addition to her existing Vikrant of the Viraat, the 24,000 ton former British
Hermes, fully modified for STOVL and helicopter operation, was a considerable
vote of confidence in the longevity of even a ship built largely to low quality
British wartime standards. Her 1986-7 refit gave her another ten years of active
service. She possesses a modern combat information and C3I capability: not for
nothing was she used as flagship in the Falklands War rather than HMS
Invincible, the first of Britain’s three new ‘aircraft carriers’. This was ironic as
the latter ships had evolved as essentially air capable command/cruisers. Their
aircraft-carrying capacity is about twenty, which is quite limited for ships of their
size and expense, although attempts are being made to increase it. Ski-jumps
have been added to increase the performance of embarked Sea Harriers.
Although very useful vessels, especially with airborne early warning helicopters
added to their air groups, the political factors that made a cruiser with its own
integral SAMs, sonar, and command facilities grow into a carrier has produced a
ship that is perhaps unnecessarily expensive compared to a vessel built purely
for the operation of aircraft. It was lucky for the British that they still had a real
carrier available to fight perhaps their last colonial war. The Invincibles are fully
stretched in high-threat areas and would benefit from additional ASUW missile
capability to allow their eight Sea Harriers to concentrate on combat air patrol
and reconnaissance duties. Both Spain and Italy are going to enter the 1990s
with new STOVL/helicopter carriers in the 10,000 ton (standard) class that will
have more or less the air carrying potential of a British Invincible, but with
fewer ‘ship’ features. Spain used the American ‘sea control ship’ design for the
Principe de Asturias built in Ferrol. The slightly smaller Guisseppe Garibaldi
was a wholly domestic product. It will be interesting to see if other current
operators of old ex-British carriers of the same size, notably Argentina and
Brazil, will find similar ships attractive or, more importantly, affordable; or
indeed if other naval powers will show interest in small STOVL carriers. India
has announced she will build her own new carriers in future but Japan still
suffers from a political reluctance to procure such vessels.
It is possible to produce very basic carriers indeed by using fast merchant ship
hulls. Such unsophisticated vessels lack much appeal to naval planners, who
have many reasons (some rational, some less so) to build ‘real’ warships. It
usually takes the pressure of actual operations to overcome these problems, and
interestingly, Britain during and after the Falklands War made perhaps the most
progress in modifying mercantile hulls for the operation of aircraft. Operated for
some time afterwards was the helicopter carrier Reliant, using American
Arapaho conversion gear and manned by the mercantile marine personnel of the
Royal Fleet Auxiliary. She was eventually disposed of but has been replaced by
RFA Argus, an extremely impressive ship with a large hangar, a protected flight
deck, and high capacity. At a quarter the cost of an Invincible she is primarily
intended as a training ship, although she would be operational in an emergency
primarily as an ASW carrier in the Channel. Her ability to operate STOVL was
limited mainly to save money and design complications but also perhaps partly
for political reasons. A ski-jump fitted to Argus, although more expensive, would
have raised some disturbing questions about relative capabilities given that the
last Invincible class vessel, HMS Ark Royal, cost considerably more still.54
It would be idle to pretend that ships such as Argus offer a complete
replacement for aircraft carriers of higher capability, especially in terms of
sensors and command and control, but they do provide a means for a small
power to provide an economical, if limited, air capability if it can afford no
other, or for a larger naval power, perhaps even a superpower, to proliferate air
capable platforms in a crisis, relying on the electronic equipment of escorting
surface combatants to manage the battle. More capable carriers cannot be
everywhere, indeed, as the US Navy’s main instruments of power projection,
they may well be sent on operations that have little to do with the essential sea
control task.
Possibly the US Navy may itself move down the carrier scale and produce a
ship of more moderate dimensions, as it came close to doing in the 1970s. The
election of a president committed to military reform might mandate smaller
carriers. These could be 45,000 ton mid-size carriers (CVV), 20,000 V/STOL
support ships (VSS) or small 10,000 ton sea control ships (SCS). These ships
would obviously supplement rather than replace the current leviathans.55 As well
as saving money their construction might reflect a conceptual judgement about
roles and missions, and an emphasis on sea control in the Atlantic rather than
power projection around the world. But for a whole range of reasons, not least
the essential nature of US sea power, sea control has never struck too many
chords among US naval planners. Given this professional ideology, coupled with
the understandable self-interest of the naval airmen, it becomes much less
surprising that the great bulks of the USS Abraham Lincoln and George
Washington have been taking shape at Newport News. Financial constraints on
US Navy procurement plans could easily force concentration on these ships at
the expense of small carriers rather than the abandonment of the supercarriers
themselves. It is interesting that the Lehman study showed more interest in
smaller ships as additions to US capabilities rather than as replacements.
In one sense, however, the US Navy currently deploys many lighter carriers.
These are the various amphibious warfare vessels that are capable both of
transporting US Marines ashore by helicopter, and also supporting them there
with STOVL fighters and attack helicopters. This relatively little publicized
fleet, any one member of which might transform the purely naval equation in
many parts of the world, provides the alternative dimension of the USA’s naval
power projection ability. There are seven 11,000 ton LPHs (landing platform
helicopters) and five newer 25,000 LHAs (landing helicopter assault). The LPHs
in particular have proved useful beyond the amphibious role, being used as
experimental ‘sea control ships’, i.e. limited capability ASW/AAW helicopter
and STOVL aircraft carriers, and also as mother ships for MCM helicopters.
This flexibility is being exploited in the new class of 26,000 ton LHDs (landing-
ship helicopter docks), whose lead ship name, Wasp, a traditional carrier name,
reflects a greater emphasis on flexibility as an aircraft platform. The new ship
will be more easily convertible into an ASW carrier or, alternatively, an ‘assault
carrier’ with short range STOVL ground-support aircraft. Some compromises
have, however, had to be accepted between ‘fixed wing’ and amphibious
helicopter assault capability, e.g. no ski-jump, in order to maximize deck space
so that the maximum number of helicopters can be launched in an assault wave.
The LHDs, like the current LHAs, will have full docking capability, i.e. the
capacity to flood themselves to act as sheltered docks for landing. Ever since
World War II the US Navy has operated a huge fleet of such landing ship docks
(LSDs) and landing platform docks (LPDs), the latter having a higher personnel
carrying capacity than the former. The US Navy is complementing its new LHD
programme by building ten new LSDs, each just over 11,000 tons in
displacement. These will be capable, like the new LHDs, of operating the new
LCAC (landing craft air cushion) hovercraft that are taking a more important
role in amphibious landing operations. The hovercraft, with its ability to move
with no difficulty from water to land, and to skim over the shore line, has
obvious utility in the landing role and is finally being taken up by both
superpowers for this duty, as well as by other states such as China.
The USSR has nowhere near the long-range oceanic amphibious lift and
power projection capability of the USA. She has a large force of small and
medium landing ships in the less than 1,000 ton bracket, but these are only
suitable for short-distance operations. They lend themselves to export and are in
service with a number of friendly navies. Even her large landing ships are
usually relatively simple tank-landing vessels, half the size of their US
equivalents. The USSR only has two naval vessels capable of docking, the pair
of Ivan Rogov class vessels. Given limited numbers, these ships have had to be
made very versatile. They are capable of landing a battalion by hovercraft (or
landing craft) and a small number of helicopters, and giving fire support with
rockets and gunfire. The ship can beach if necessary to land tanks. Given the
continued supremacy of the airborne forces in the Soviet Union’s power
projection capabilities, one cannot expect much extra development of the Soviet
Navy’s amphibious warfare capability.
Several other countries have LPD/LSD-type specialized landing vessels. Such
ships are in fact probably essential if a country wishes to be able to land troops
over beaches. Merchantmen can be most useful in amphibious operations as
transports, as demonstrated in the Falklands War. Many Soviet roll-on roll-off
cargo vessels seem to have been built with possible use as landing ships in mind.
Yet exercises show that simple unmodified ferry-type vessels cannot be safely
beached. Also, transfer to landing craft can be hazardous to men and ships alike.
Military features, like beaching capability, must be built in at construction if the
ship is to have operational functions beyond that of a mere transport.56 This can
be done with merchantmen built with rapid mobilization in mind, but the need
for specialized command and control equipment mandates some specialized
naval manned ships, as does the ability to deploy at short notice.
Unfortunately amphibious vessels tend to be relatively expensive. Even
middle ranking states, like the United Kingdom which has deployed two LPDs
since the 1960s, has to consider very carefully the question of replacement. She
has decided in principle to replace these ships with rebuilds or new construction,
and in addition to acquire at least one vessel for helicopter landings. Again the
precise configuration and origin of the new ship has not been settled. She might
be a converted merchantman, and British Aerospace and Sea Containers have
produced an interesting 28,000 tonne ‘aviation support ship’ planned on the
container ship Contender Argent, which would lend itself to the role. A ‘through
deck’ ship (without ski-jump), she would be able to embark, transit, and deploy a
Marine commando.57 France plans to complete three new LPDs in the 1990s and
Italy is constructing two new 5,000 ton San Georgio helicopter-operating landing
ships. Interestingly, funds were obtained to build the second vessel, San Marco,
by building for the Ministry of Civil Protection as a disaster relief ship. There
are also smaller general-purpose landing ships on offer, e.g. the 2,000 ton, 76
mm gun armed vessels built by Brooke Marine of Lowestoft in 1979 or the 900
dwt LS100 design currently offered by Harland and Wolff of Belfast for Oman.
Amphibious warfare vessels, with their large accommodation spaces, ability to
operate smaller craft, and possession of helicopter flight decks that make
excellent stowage spaces (and parade grounds), lend themselves also to use as
training ships.
They also make good command and headquarters vessels. The United States
has gone to the extent of building two specialized command ships especially for
the control of amphibious operations. This pair of 20,000 ton vessels, based on
LPHs, have a bewildering array of electronic command and control equipment,
and are perhaps the most electronically sophisticated vessels afloat. They are
beyond the reach (or the needs) of any other navy and tend to be used in
peacetime as fleet flagships.
Before leaving surface warships, a word must be said about fleet support. If
any navy wishes to have the capacity for sustained operations at sea it must be
able to take supplies of fuel and other stores with, or out to, the operating forces
and transfer them there to the combatant units. This means significant
investment in relatively fast (20 knot) tankers and supply ships specially fitted to
engage in replenishment at sea (RAS) operations. These can be supplemented by
slower, less capable mercantile hulls in time of need. No navy can claim to be
fully ocean-going if it does not have such ships, which can double as ASW
helicopter platforms and aircraft maintenance facilities. There is a tendency to
increase the combatant utility of these vessels thus integrating them with the
whole force. This also implies a level of point defence armament such as Sea
Wolf missiles fitted to the new British Fort Victoria class.
British Shipbuilders has recently unveiled a ‘Multi-Role Naval Auxiliary’
(MRNA) which is basically a replenishment ship but which has the potential for
use as an amphibious warfare vessel with helicopters and landing craft, or
disaster relief ship, oceanographic vessel, or MCM support ship. The ship has
various extra survival and protection features compared to a normal mercantile
hull and could be attractive to various navies. As its builders put it, given an
MRNA’s capacity to be configured in a number of various ways: ‘a navy can
procure a vessel tailored to its own particular needs. For some countries the
MRNA can provide a navy with a new capability, for other navies it can provide
an increase in capacity for undertaking various operations without the need for
extra ships and crews.’58 It will be interesting to see if such multi-role auxiliaries
become more common.
China has recently built her first four fleet tankers of more or less
conventional layout, the most important sign of her aspiration to longer-range
surface operations. India needs to improve her capacities in this regard, a third
fleet tanker is under construction but she, like the other two, will be chartered
not owned. France, who has traditionally relied more on fixed bases, is also
somewhat belatedly improving her capabilities in fleet supply ships, both by new
construction and charter. Britain, who built up her supply vessel capabilities
impressively in the 1960s to sustain her East of Suez peacekeeping role in a
context of few shore bases, is allowing her supply fleet to run down, but is
investing in limited numbers of new and expensive fleet replenishment ships, the
first of which should be in commission in 1990. Soviet specialized capabilities
look like remaining relatively limited, although the Soviet merchant fleet is
always available to supply the navy when required. Inevitably, the United States
has a supply of fleet auxiliaries second to none.
The final matter to be discussed is the question of surface ship vulnerability.
Surface combatants are indeed inherently vulnerable to the widest range of
enemy weapons. Thus they are not very good offensive platforms, except at the
long ranges that carrier aircraft allow. However, they are not so vulnerable as to
be swept from the seas by anti-surface missile threat. Hitting them at all is not so
easy and little real damage may result even when a hit occurs. Modern surface
ships have evolved from the traditionally non-armoured vessels of the past. Even
the modern cruisers are the descendants of older destroyers rather than the older
cruisers that used to carry a little armour against gunfire. The electronic
sophistication of three-dimensional warfare and the need to maximize space for
sensors, command equipment, and space-intensive missile systems have all
emphasized ship designs that continue to optimize volume for a given
displacement. Putting armour all over such spaces is a more difficult task than
protecting the armament, mechanized vitals, and watertight integrity of ‘old
fashioned’ Dreadnought era warships.
Much can be done to stop ships being unnecessarily vulnerable. Surface ship
design is going increasingly to emphasize signature reduction, acoustics, radar,
infra-red, and magnetic means. The physical shape of future navies is going to
be increasingly conditioned by the need to make both electronic detection and
identification difficult. Separate modular ventilation and fire-fighting systems
can also greatly reduce vulnerability to fire and smoke threats. Aluminium,
which if not quite so flammable as sometimes argued, is certainly much more
vulnerable than steel to heat damage and which is also prone to splintering, will
be used much less in warship construction. Plastics may, however, be used more,
because of their ‘stealth’ characteristics. Less flammable materials can be used
for wires, which themselves can be replaced by more modern forms of highway
for electronic data distribution. Only the largest ships, however, notably carriers,
can be armoured in anything approaching a traditional kind of way. In smaller
vessels armouring would necessitate substantial extra increases in volume and
power that cost would soon make prohibitive. Alternatively, the fighting power
of the ships would be reduced, thus reducing their potential offensively and
defensively.
It is likely, however, that the warships of the future will be given some small
level of armour protection to limit excessive vulnerability to damage. Extra
metal plating or other substances such as Kevlar can be added to especially
sensitive areas, such as combat information centres. The increase in cost and
weight is small, about 50 tons on a modern frigate. Even radar antennae can be
given Kevlar covers for there is little point in protecting the ship if the means by
which it can engage in above water warfare are disabled. The level of protection
is not intended in any sense to protect the ship from the structural effects of
direct hits, but it does limit collateral damage and the effects of near misses. It
also protects against small-arms fire, important in a range of low-level
situations.59

Submarines
Of course the major way in which a naval platform can protect itself from above
water damage is to submerge deliberately beneath the waves. The twentieth
century has seen the ever-increasing importance of the submarine, especially as
nuclear propulsion has solved the fundamental problem of providing safe,
reliable propulsion without the need for external air. Submarines have become
the primary seaborne platforms for high-level anti-surface warfare. They are also
extremely effective ASW platforms, being able to exploit the changing
characteristics of the sea environment to optimize the performance of their sonar.
Submarines do, however, suffer from four very serious limitations. First, the
kind of damage they can inflict with their primary torpedo or missile weapons is
almost always fatal and catastrophic. This rules them out as weapons of much
utility in operations at the lower levels of intensity. Submarines can be used as
cover for more valuable forces in such situations and they can threaten
unwelcome retaliation if the rules of limited conflict are broken; if they are
unleashed then it usually marks a major escalation of the conflict. The Belgrano
affair demonstrates the problem clearly.
Second, submarines are only relevant to the ASW and ASUW battle, they
have very little AAW potential beyond a limited early warning role. Attempts
have been made, and will be made in the future, to give submarines anti-air
missiles and they might have some utility to a cornered, detected submarine,
especially one being harried by helicopters. Submarines cannot, however,
provide air defence for ships, either in terms of sensors or weapons, without
giving away their position and thus losing their main advantage of stealth. Third,
the submarine is unsuitable for roles where the presence of the unit, and the
interest and the capability it implies, is the very point being made. Even when
surfaced, submarines have the appearance of stealthy, silent killers which
alienates rather than attracts. Only small submarines are good for port visits and
there are obvious problems in the social side, although the ships’ companies
make valiant efforts to overcome them. Nuclear-powered submarines, whether
armed with nuclear weapons or not, are rarely welcome visitors anywhere.
Submarines do have excellent utility in providing a latent naval presence that can
be revealed if required and denied if not, as skilfully demonstrated by the British
Government in 1977 in relation to the Falklands, but they cannot substitute for
surface warships in most elements of diplomacy.
Finally, and perhaps most importantly, submarines suffer from very severe
command and control limitations. For precisely the reason that a submarine is so
effective, its use of the sea to block electromagnetic radiations, it is extremely
difficult to use the main communications medium, the same electromagnetic
radiation. (Sound communication is only possible at the shortest of ranges.)
Radio waves do not penetrate sea water to any extent, except at the lowest
frequencies. Even with very low frequencies (VLF) the submarine has to put
itself, or an antenna, close to the surface. Such low frequencies, however, have a
very low traffic capacity and are only one way from the base complex and its
vital transmitter to the submarine. Range is limited to about 1,500 miles. If a
submarine wishes to have a dialogue it must raise its antenna above the surface
and use higher-frequency radio. This, and the act of transmitting, makes it
vulnerable to detection. The use of satellites and ultra-high frequencies (UHF)
and super-high frequencies (SHF) can allow more private directional exchanges
of information at very high speeds, and with much reduced chances of detection,
but even this only works best on a routine basis as the boat must come shallow
and extend an antenna. Expendable radio buoys can be used as alternatives to
coming close to the surface. These can be released from submarines with pre-
recorded signals or dropped by aircraft to allow UHF communication via sound
signals. The latter, however, again makes the submarine vulnerable if the
communication is to be two-way.60
Some of the problems of low frequency communication can be overcome by
the use of extremely low frequencies (ELF) which increase sea penetration from
10-15 fathoms to about 100 fathoms. ELF allows staying deep, freer from
detection by sonars or infra-red detection which can pick up buoys or wires.
ELF, however, has a very low data rate, three characters in twenty minutes, thus
making it merely an alarm to get a submarine to rise to receive more
information. It also required considerable tracts of land, limiting its acceptability
to the American public. An alternative means of communicating with
submarines is to use blue-green lasers. These have the property of being able to
penetrate sea water’s ‘optical windows’. They can be mounted in satellites and
work has been going on in the USA on a Submarine Laser Communication
Satellite System (SLCSAT) since 1978. This could allow much more specific
and comprehensive communication from bases out to submarines, especially to
ballistic missile firing boats. The cost, however, is considerable, some $2 billion
at current prices, and the US Navy is proceeding relatively slowly with the
programme. During the 1990s this method of communication could
revolutionize submarine communications in terms of the degree of real time
controllability of the submarine force, both SSN and SSBN. Given the
uncertainty of the sea, however, and the availability of the right ‘windows’,
SLCSAT would only supplement other systems: it would not be the sole means
of submarine communication, VLF and ELF would remain. The US Navy has
been accused of dragging its feet over SLCSAT and one reason may well be the
instinctive service suspicion about the degree of control if would allow over its
traditionally independent submariners. As one senior American official pointed
out: ‘the submariners of the USN must now accept that they were no longer “a
law unto themselves”’.61
Poor communications mean that submarines operate best as ‘lone wolves’.
They receive basic orders and intelligence and then must be left to their own
initiative, informed by general rules of engagement. This constrains matters at
all levels, from the lowest, where it confirms the submarine’s nature as a
somewhat blunt instrument, to the very highest where ballistic missile
submarines cannot easily or quickly be retargeted. It also makes close co-
operation with surface units difficult, which is a basic problem given sea power
still being largely about the use of the sea by surface platforms. Submarines can
sink them but they have drawbacks as means of defending them. They are
essentially weapons of sea denial rather than sea control.
For reasons explored at the outset of this chapter, it seems unlikely that
submarine transports will drive naval warfare wholly into the depths for some
time to come, if ever. Submarines will not, therefore, replace surface ships but
they will remain potent naval weapons that fewer and fewer navies will feel able
to do without.
Nuclear power, as mentioned above, has been the key to the submarine’s
quantum leap in capability. Not only does the nuclear reactor confer
independence of the surface and make crew endurance rather than mechanical
limitations the major constraint on a submarine’s range, but it also provides
massive power outputs that allow the most extensive and powerful sonar and
command and control suites. Nuclear submarines are, therefore, vessels of some
considerable size, about 3,500-6,000 tons for current American, British, and
Soviet torpedo attack submarines (SSNs). Future boats will probably be in the
8,000-9,000 ton bracket. A need to carry bulky missiles causes elephantiasis to
set in. Modern Soviet cruise missile submarines intended primarily for long-
range anti-carrier warfare displace about 12,000 tons and the latest Soviet
ballistic missile firing submarines are over twice as big. Double hulls make such
‘boats’ tougher to attack. The latest American ballistic missile firing submarines
(SSBNs) are in the same large size bracket, being 17,000 ton vessels given the
names previously reserved for battleships.
Smaller nuclear submarines have been built but they suffer from quite serious
technical drawbacks. The French have limited their SSNs to a displacement of
2,500 tons, but the exceptionally small natural circulation reactor plant results in
a lower maximum speed than that attained by most SSNs. This is a serious
tactical disadvantage. Alternatively, the Soviets have built small, titanium-hulled
Alfa class boats of about 3,000 tons with dense metal-cooled reactors and
remarkable speed (45 knots) and diving capabilities. They were probably
constructed as a temporary admission of defeat on the quietness front but
construction seems now to have stopped. The boats were exceptionally noisy:
this did not matter so much if the sole aim was to avoid destruction, but to do
anything positive the Alfa would have to slow and/or come to a shallower depth
to listen on sonar or to engage a shallower diving or surface opponent. The
development of better Western weapons could also not long be delayed. Hence
construction of the Alfa class boats ceased after six production units had been
constructed and much consternation created amongst some more alarmist naval
commentators. The Soviets seem to have paid considerable attention to
developing sophisticated skins for their boats to minimize drag and
hydrodynamic signature. It is clear, however, that the Russians have seen little
alternative but to attack the problem of noise once more. Helped by spies and
Western firms willing to sell advanced constructional technologies they seem to
be essentially copying the West, with relatively large boats with advanced
propellers of non-cavitating design, propulsion machinery on ‘rafts’ to minimize
sound transmissions, other machinery on flexible mountings, quiet reactor
coolant pumps and anechoic coverings. The new Akula class submarine seems
very much in the same size (8,000 tons dived) and technical bracket as a modern
American boat, with the same kind of sound quietening techniques. Both the
latest Sierra class boats, incremental developments of the previous Victors and
the doomed liquid metal reactor fitted Mike (the latter apparently optimized for
ASUW with torpedoes) were also quieter than their predecessors.62
Quietness is, of course, the crucial parameter in submarines, with an obvious
impact on other performance characteristics. A submarine’s various quiet speeds
for optimum transit, search, area coverage, approach, and attack are much more
important than an absolute maximum figure. The West still has the major
advantages here, indeed the Americans now claim that their latest nuclear
submarines are quieter even than conventional submarines. As well as ‘rafting’
machinery spaces, pump jet propulsion is used in new British SSNs. It is planned
for the projected American Seawolf class of SSNs and can be used to diminish
propellor noise still further.63
Mechanically, the pressurized water cooled reactor seems set to retain its
supremacy for some time to come. Alternative types of reactor have severe
safety drawbacks. Soviet style liquid metal reactors have been described by the
head of the US Navy’s Nuclear Propulsion Directorate as containing ‘more
radioactivity than all the reactor coolant in our whole navy’.64 It has been
suggested that the Soviets have operational magnetohydrodynamic waterjets on
their latest submarines, but these reports have been discounted by authoritative
sources elsewhere:65 Even more futuristic propulsion concepts such as rippling
skins seem equally unlikely to emerge from the Soviet production base. Electric
drive of a more conventional kind may have more of a future however: it is
inherently quiet and superconductivity techniques could greatly increase its
efficiency.
Steam turbines, however, are likely to remain the major near to medium term
form of transmitting the power of nuclear reactors to the driving shafts of new
submarines, especially as new countries tend to develop SSNs. China, after
considerable problems with development, has produced nuclear-powered
submarines in limited quantities, both SSNs and SSBNs. This has no doubt been
a factor in India’s decision to lease nuclear-powered submarines from the Soviet
Union: delivery of the first, a Charlie I class anti-ship missile firer took place at
the beginning of 1988. At the same time, Canada was in negotiations with
Britain and France for no less than ten to twelve nuclear-powered boats for
service in the Arctic beneath the ice. Both Britain and France offered designs but
the cost of the programme led to its eventual abandonment.66 Other countries
have expressed an interest in acquiring nuclear submarines. Brazil has started
planning and spending money on a project to begin construction in the 1990s.
Pakistan, spurred by India, is examining the possibility of conventional boats
with small nuclear power plants to improve underwater performance. Such
hybrid submarines may be a good compromise for nations of limited means.67
There is also the problem that ‘nuclear’ is still a word with enormous political
connotations, not least in Japan, a country sophisticated enough to build nuclear
submarines if she wished.
Other than greater political acceptability, and despite their operational
disadvantages, conventional submarines will always have one major advantage
over nuclear boats: cost. They have other advantages too, standard of quietness
better than all but the most advanced nuclear submarines. Diesel electric
propulsion is standard for such boats with snorts to allow operation of diesels at
periscope depth. Batteries provide limited range under water and propulsion with
no tell-tale surface activity. Batteries of greater efficiency, and possibly,
eventually, superconductivity techniques in motors and generators, will increase
the underwater efficiency of such boats still further, although not in any way that
would fundamentally change the nature of conventional operations, which
emphasize low speeds and short dashes. Under development, however, are novel
non-nuclear engines that could give longer ranges and higher speeds under
water. Fuel cells, flywheels, Stirling external combustion engines using liquid
oxygen, and the closed-cycle Krieslauf diesel are all under consideration.
Sweden and West Germany, both countries with important political inhibitions
about nuclear submarines, are leading the way in these developments, the former
with Stirling engines, the latter with fuel cells.68
Constructing and manning submarines is probably more demanding than
constructing or manning frigates, but the advantages of the submerged platform
make submarines increasingly popular in the world’s navies. West Germany, in
particular, has done well with the export of completed boats and the technology
to build them locally. The Soviet Union provides rather larger and less
technically advanced boats to its friendly states, and China exports submarines
also. A few countries operate large 2,000 tonners, usually obtained from the
superpowers, but in the Dutch case such boats are constructed domestically,
despite enormous cost over-runs and disastrous fires. Significant political
pressure did not prevent The Netherlands exporting advanced submarines to
Taiwan which hopes to build them itself in the future. The other conventional
submarines in the world tend to come in three size brackets, about 1,300-1,600
tons, about 700-900 tons, and about 400-500 tons. The last named are only
suitable for use in coastal and restricted waters and are preferred in the northern
European navies, but even here the tendency is to larger 900 ton boats as the
usual compromise in requirement, capability and numbers. One detects a
tendency for countries to be moving up a band in their conventional submarine
procurement policies. Britain, previously satisfied with 1,600 tonner, is now
producing a 2,400 tonner. Argentina is moving from 900 to 1,700 tons. One
thousand now seems to be considered the minimum for open ocean operations,
but Sweden prefers this bracket for its own impressive submarines specialized
for use in the Baltic. Midget submarines are also making a comeback. Colombia,
Pakistan, Taiwan, and Yugoslavia all have modern examples and the surveillance
activities of the Soviet Union’s varied fleet of small submersibles are notorious.
Submarines of all shapes and sizes provide excellent surveillance platforms,
as well as means of attack. The latter have been enhanced by the advent of
compact cruise missiles compatible with the tubes of existing submarines. These
give submarines a potent new weapon against surface ships, and also against
targets ashore. Towed sonar arrays can be used to give target acquisition
potential. One must not underestimate, however, the power of the torpedo, which
carries a much heavier warhead than any missile and attacks its target at its most
vulnerable point, i.e. beneath the waterline. If submarines use their torpedo tubes
to fire missiles, then missiles eat into already limited torpedo stocks. For this
reason the USA prefers to fit extra missile tubes to its latest SSNs. The Soviets
have had the choice made for them by the relatively large size of their missiles,
although they have the extra potential given by the large torpedo tubes fitted to
submarines to carry the large wake-homing torpedoes and the Soviet family of
underwater ASW weapons. The Soviets also continue to use separate missile
tubes for their large anti-ship and anti-shore cruise missiles that will be in service
in submarines for many years to come. Certainly some cruise missile capability
will be a vital part of the armament of the submarine in the future: reportedly the
French have even given their latest SSBN a subsurface launched Exocet
capability as a self-defence weapon.69

Aircraft
Sea power and air power are indivisible. Seaborne platforms provide some of the
most flexible, and least vulnerable ways of deploying air power, either in the
form of aircraft or missiles. This has created bureaucratic problems, and
accounts for the almost universally uneasy, if not downright bad, relationship of
a nation’s navy and air force. Such problems have often been compounded by
the undoubted utility of land-based aircraft to operate over the sea in all three
modes of naval warfare. Lines of demarcation have never been easy to draw and
these disputes have sometimes had serious operational effects. The full potential
of land-based air power over the sea has often not been fully exploited. Navies
have been limited in their abilities to operate aircraft. The situation is usually
happiest when the sea/land boundary is used as the basic organizational divide,
i.e. aircraft designed to operate over or from the sea being operated by the navy:
those over land by the air force. Historical accident, however, has often dictated
otherwise. Certainly the aircraft specially intended for maritime operations are as
much a part of the nation’s ‘navy’ as any ship or submarine, an air force can also
have a considerable relevance at sea as part of its general capability in attack or
defence.70
Aircraft are inherently fast: they can cover great distances in short time spans,
and strike rapidly and surprisingly. Aircraft can carry sensors to a height where
they are not limited by anything other than their inherent technical
characteristics, either to find targets for other platforms to address if
communications are adequate, or to attack targets themselves. They can carry a
wide range of weapons for the latter role: long-range missiles and bombs for
ASUW, depth charges and lightweight torpedoes for ASW, missiles and guns for
AAW. In the ASW battle they can even ‘look’ beneath the waves with the help
of sonobuoys or magnetic anomaly detectors or, in the case of aircraft that can
hover, active dipping sonar.
Their drawbacks are, like those of the submarine, the reverse side of the coin
of their assets. Speed demands power and power demands fuel which takes up
much of the available payload. Large aircraft can achieve substantial feats of
endurance but this is measured in hours rather than days. Moreover, such aircraft
tend to be inherently vulnerable. Very high endurance aircraft can be obtained by
the use of light gases rather than aerodynamic lift. From the earliest days of
naval aviation airships have lent themselves to operations over sea rather than
land, and modern technology has overcome most of the problems which raised
doubts over the safety and controllability of earlier examples. It is quite possible
that airships will be used for certain naval tasks in the future, especially
constabulary/patrol duties in economic zones. The airship’s high endurance,
coupled with its ability to stop and inspect, or even board, would be especially
useful for constabulary duties, but it would also be very suitable as an airborne
early warning platform with the radar in the airship envelope itself. Secretary
Lehman was a great supporter of the LTA (lighter than air) programme, and in
May 1987 an Anglo-American consortium was selected to build a series of 129
metre long helium filled, diesel powered ‘patrol air ships’ with a 30-day
endurance (with refuelling). The programme became vulnerable with Lehman’s
departure, however, and without powerful support from an existing ‘community’
within the service it remains doubtful whether airships really will be appearing
over US task forces after 1992, as originally scheduled.71
More conventionally, large fixed-wing aircraft will certainly remain a
valuable asset, both for surveillance and warning duties, against three-
dimensional threats, and for attack against surface and underwater targets.
Although the combination of a relatively simple airframe with a relatively
unsophisticated radar is adequate for constabulary surveillance duties, capability
at a higher level of maritime warfare demands much more sophistication and
expense. The specialized maritime patrol aircraft has evolved as a highly
expensive combination of electronically co-ordinated and analysed radar, sonar
and electromagnetic sensors, combined with the ability to deliver depth charges,
lightweight homing torpedoes and anti-ship missiles. These aircraft are currently
best represented by the American P-3 Orion and British Nimrod, both based on
airliner airframes, the Electra and the Comet respectively. The United States
evaluated three contenders for its new ‘long-range air ASW-capable aircraft’
(LRAACA): Lockheed, Boeing, and McDonnell Douglas each offered different
concepts. Boeing used turbofan engines in a derivative of the 757: these power
plants now offer comparable fuel economy and range to the improved turboprops
used by the developed Lockheed P-3 that was eventually successful in the
competition. (In fact the Nimrod solved the problem of jet propulsion for
maritime aircraft many years ago by its facility of cutting engines in and out
depending on the speed required.) The McDonnell Douglas design utilized the
new unducted fan concept that combines the speed of the turbofan with economy
approaching that of the turboprop. The US Navy insisted that its future
LRAACA should be able to stay on task for at least four hours, with a minimum
radius of action of 1600 nautical miles and played safe with an advanced
turboprop for its new aircraft that will be in production until 2001.72
Whatever the layout, such ASW aircraft are complex and expensive items.
The simpler anti-surface task can best be addressed by long-range bombers
equipped with less extensive sensor suites. The characteristics required are those
of any long-range strategic bomber, and the Soviets have made this form of long-
range maritime striking power very much their own. The USAF is, however, also
demonstrating greater interest in a maritime role for its strategic bombers as
penetrating the Soviet Union with them becomes more problematical. All such
large aircraft, however, given their inherent vulnerability, must rely on stand-off
missiles of various ranges to allow them to strike a well-defended target. One
future concept that has been suggested to exploit these weapons is the long-range
Missileer, a large 747-type aircraft crammed with offensive missiles of various
ranges, and with laser weapons for self-defence.
Smaller aircraft facilitate greater penetration as well as being much more
suitable for anti-air duties. Given the relatively low payload of such machines,
range is limited unless substantial tanker support is provided. Sea basing is
probably a better answer to this problem, although modern and future combat
aircraft of the highest available capability will require large and expensive
aircraft carriers of conventional type. The United States will remain the only
navy with the ability to deploy such vessels and is developing an advanced,
stealthy attack aircraft, the A-12, to take over the all-weather attack role in the
mid-1990s. This may well become a common aircraft with the USAF, whose
Advanced Technology Fighter may replace the F-14. For the other carrier
aircraft roles – AEW, ASW, and electronic warfare – a common advanced
tactical support aircraft (ATSA) is due for development in the 1990s.
All the above are conventional take-off and landing (CTOL) concepts. If
reduced performance is acceptable, however, then more unconventional layouts
can be used to facilitate sea basing. Vectored thrust jets have already allowed the
deployment of relatively low-performance STOVL jet fighters which, when
combined with high technology missiles, have demonstrated an ability to defeat
even a relatively well-equipped Third World air force. The Harriers that
achieved this feat remain the only effective STOVL aircraft of fighter-type
performance. Produced in the USA as a clear weather fighter bomber, the
concept has been developed there into the higher performance AV-8B. Britain
has produced the radar-fitted Sea Harrier with lower flight performance but more
all round avionic capability: the Sea Harrier is currently in process of upgrading
to FRS-2 standard with better radar and long-range radar homing AMRAAM
missiles. Sea Harriers are in service with the Royal Navy and India; AV-8s in
various forms are in service with Spain and the US Marine Corps. It is not clear
which branch of the family Italy will purchase. Japan is also interested in a
Harrier-type purchase, as is Brazil, and the widening market is encouraging the
joint manufacturers, British Aerospace and McDonnell Douglas, to consider a
Multi-Mission Harrier II which will combine an uprated engine with radar. If
decided upon it will enter service in the early 1990s.73
Such aircraft put usable air power within the means of a widening range of
middle-level states. For the superpowers, STOVL technology in various forms is
also probably going to be exploited more. The British developed the ‘ski-jump’
take-off aid, which allows vectored thrust aircraft to take off with higher
payloads of fuel and/or weapons. It is now to be fitted to the new Soviet nuclear-
powered aircraft carrying supercruiser Tblisi, currently under construction. The
crude VTOL Forger aircraft used from the Kiev class cruisers must be hard to
use with ski-jump, but it seems that the new Yak-41 may overcome this problem
if it conquers its own difficulties. The Soviets have persevered with experiments
to convert conventional take-off and landing (CTOL) fighters to carrier
operations (using both catapults and arrester gear and ski-jumps). The fate of
these experiments will be governed by economic and political factors as much as
technical ones.
On the American side, US Marine Corps interest has allowed the diversion of
some of America’s massive resources into limited STOVL development.
Aviation journalist Bill Sweetman has, however, summed up the problem
succinctly and not too misleadingly as regards a more extensive US espousal of
STOVL technology:

The world of naval aviation is divided into the U.S. Navy with its mighty
fleet of supercarriers and the rest of the world with its handful of ships per
country. The rest-of-the-world market is too small and scattered to sponsor
extensive development programmes; the U.S. Navy is so terrified by the
prospect of armchair admirals promoting STOVL as a substitute for the
supercarrier that it finds it difficult to consider it as anything but a threat.74

The problem is not only bureaucratic. Supersonic STOVL designs have been
considered for many years but there are real technical problems giving vectored
thrust aircraft afterburning engines of sufficient power. In addition, large and
heavier aircraft require more complex control systems. Alternatives have been
offered to vectored thrust, but none is yet practical. The most likely outcome
seems to be an afterburning (or ‘plenum chamber burning’) development of the
Harrier concept, followed in the next century by the exploitation of second-
generation technologies like the ‘tandem fan’ which splits the airflow from a
conventional augmented turbofan. Vought have produced a naval fighter design,
the TF-120, based on this principle. As advanced STOVL types approach the
performance levels of CTOL aircraft the limits on the latter’s development may
well help STOVL types to ‘catch up’. This will probably make deployment of
significant ship-borne aviation an available option to all the major naval powers
of the twenty-first century.75
Propeller-driven STOVL and VTOL technologies with considerable potential
are already becoming available. The most important is the American Bell-
Boeing V-22 Osprey tilt rotor aircraft which was to be delivered to the US
Marine Corps for amphibious landings and to the US Navy for search and
rescue, special warfare, and fleet logistic support. It could have also been
developed to provide useful ASW and airborne early warning aircraft to operate
from a wide range of platforms. Very sadlly the whole project has been cancelled
to save money, it is to be hoped only temporarily, as such aircraft offer greater
speed, range, fuel efficiency, reliability, and maintainability as compared to
helicopters. The V-22’s twin turboshaft engines allow it to cruise at 275 knots,
and as a troop transport it has a range of 500 nautical miles. Although initial
procurement costs are high, operating costs should be lower. In the longer term,
jet-powered VTOL multi-mission aircraft have been suggested. Again these will
make the development of a militarily highly capable STOVL air group a
practical possibility.76
Some see the future belonging to STOVL or VTOL aircraft operating from all
ships. A Skyhook technique has been developed in Britain to allow Harrier
operations from destroyer type vessels. Skyhook is a ‘clever crane capable of
capturing a Harrier in the hover and placing it in a prescribed position at its
base’.77 The technique is technologically both practical and original, but it seems
unlikely that destroyers will soon acquire STOVL aircraft as part of their
armament. Such fighters need considerable logistic and manpower support. They
must also be carried in numbers if sustained operations are required: and the
most cost-effective solution, both logistically and tactically, is to group them in
one ship with a large hangar and, therefore, a large flat deck for a roof. It would
seem sensible, therefore, to develop specialized carriers for such aircraft rather
than trying to equip the entire fleet with them. Such carriers, however, might be
of novel design. Conversions of mercantile hulls have been discussed above.
Indeed Skyhook technology might allow small container ships to operate as
forward operating bases or as escort carriers in their own right. Alternatively
Skyhook might make small new construction ‘air capable destroyers’ in the
5,000-6,000 ton class more practical. Effectively small escort carriers with short
‘through decks’ and ‘ski-jumps’, they would allow countries with political
problems like Japan to obtain the benefits of STOVL capabilities. Their
complement might be five Sea Harriers plus a large helicopter. Both the coupling
of air capabilities with relatively small displacement and the imaginative
adoption of novel foreign technology are things which fit Japanese culture and
experience extremely well.78
This is not to say that low-performance VTOL aircraft will not allow virtually
any platform to be used as a forward operating base. Already the helicopter has
achieved a remarkable degree of ubiquity at sea. Conventional rotary-winged
machines provide VTOL platforms of sufficient range and performance for ASW
surveillance and attack, limited ASUW surveillance and attack, and a wide range
of utility and logistical duties. It is now possible to operate even large helicopters
on ships of frigate size and above, and any sacrifice of such abilities is a
considerable degradation in the ship’s capability. The new importance of active
sonar will emphasize the helicopter’s usefulness as a sonar platform still
further.79 Helicopters will continue to be a vital part of the naval scene, acquiring
new roles as target designators for laser-guided ship-borne artillery. They will
also continue to be used for amphibious operations, although the tilt-wing
aircraft and the hovercraft may cause a long-term decline in the uniqueness of
the helicopter’s current ability to carry troops and stores inland from offshore
platforms.
Helicopters can also provide airborne early warning facilities, although their
vibration and endurance characteristics make them less than optimal for this
work. They cannot carry the sophisticated electronic computers necessary to
back up a modern AEW radar. Probably, Osprey-type aircraft, with more load
carrying capacity, will have greater potential in this role, providing a more
capable AEW to the STOVL carrier equipped navy. Certainly, as the Falklands
showed, some AEW is vital to the viability of any use of the sea against the
threat of air attack, especially by small, fast, low-flying aircraft. Conventional
carriers, like the US Navy’s, can carry fixed wing AEW aircraft of maximum
capability, but again this is one resource that is perhaps uniquely relevant.
Otherwise, land-based, high-endurance AEW aircraft can be provided, if
available, and can be useful to any maritime commander: they could be
optimized for naval purposes, being fitted with weapons for ASUW strike.
Here, however, we return to the question of the allocation of ‘air power’
assets. Aircraft can be extremely effective over the sea, but only if properly
allocated, committed, and trained. When any of these three latter attributes has
been missing the result has often been disastrous. Given the ever-growing
expense of aircraft, especially those with the greatest potential and utility (e.g.
long-range AEW machines), there must be clear agreement between the various
operators on roles and rigorous training in them. Even so a limited insurance
policy (e.g. the British solution of AEW helicopters) is advisable to provide
adequate assurance of some AEW support in the case of unexpected events,
which tend to be inherent in the maritime environment.

Unmanned platforms
The ‘unmanned platform’ term covers remotely controlled vehicles operating
above, below, or on the water’s surface. Remotely piloted aircraft (‘drones’)
have not been exploited much for maritime operations, due perhaps to the fiasco
of the American DASH (Drone Anti-Submarine Helicopter) system in the 1960s.
Manned helicopters proved a more practical answer to the particular problem of
long-range ASW weapon delivery. Reconnaissance drones were also
experimented with in the Vietnam War but recovery at sea proved difficult. More
recently, however, after positive experience with Israeli-built drones as spotters
for naval gunfire support, the US Navy has begun development of drones for a
number of naval applications.80
The most important application of such remotely-piloted vehicles (RPVs)
seems to be the form of obtaining over-the-horizon targeting data for the long-
range weapons systems now becoming available. Such weapons also demand
more lofted sensors to deal with incoming threats. Although the technology for
such devices is now increasingly available, recovery is still a crucial limiting
factor. Recapturing an RPV at sea is a tricky business even in good weather
conditions. As RPV weights increase, the possibility of ship damage goes up in
proportion. Ditching RPVs in the sea usually causes an undue level of damage to
the vehicle. Unless an RPV is to be considered expendable (a possibility for
certain applications) then a solution to the problem must be sought. One is to use
tethered systems, such as lighter-than-air ‘aerostats’ already used by the US
Coast Guard for surveillance. More mobile remotely-piloted lighter-than-air
surveillance vehicles are also conceivable, with an air endurance measured in
weeks between refuelling and maintenance periods carried out while hovering
over surface ships.81
It seems likely that remotely-piloted aircraft of various kinds will overcome
their technical problems in the next few years. In navies with significant air
lobbies (either fixed or rotary winged) there might well be opposition to their
deployment, but the USN is already deploying the Pioneer heavier than air RPV
system for fire control in its battleships and some of its destroyers. The ever
increasing need for relatively long-range surveillance and target acquisition and
designation should force the increasing adoption of such vehicles at sea.
On and beneath the sea, MCM requirements are already causing various
nations to develop and deploy remotely piloted surface and underwater vehicles
of various types. More general deployment of robot submarine platforms also
seems likely. One British proposal, the Seicon Patrolling Underwater Robot
(SPUR) concept, foresees a vehicle with considerable autonomy guided by
artificial intelligence. It would be used both for information gathering and
‘suicide’ attacks. The US Navy has concepts for ‘small mobile sensor platforms’
deployed from torpedo tubes for target detection and assessment. An American
‘unmanned underwater vehicle’ (UUV) prototype 40 feet long and 4 feet in
diameter is being developed, which is too big for submarine deployment.
Norman Friedman has speculated that this might be carried by a surface ship or
helicopter and make a suitable bi-static receiver to give a ‘three dimensional
sonar picture. Alternatively it might be used to service or to plant successors to
the SOSUS fixed sea bottom arrays perhaps in enemy waters’.82 Successor UUV
systems might also be used as transmitters for multi-static sonar: this would not
risk the parent vessel, especially if it were a submarine. The usual problem of
submarine communications creates difficulties for remote interaction with these
subsurface platforms, but these are not insurmountable. Fibre optical cables
would be useful over relatively short distances while sound signals (perhaps via
air-dropped UHF buoys) and/or VLF radio might give longer range
communication. This would be tricky, however, in the under-ice operations often
mentioned in connection with these vehicles. Much would be expected from
their on-board computers and data-banks.

Space platforms
Increasingly, space platforms will interact with naval operations to provide an
extra dimension to the already complex sea power situation. As the last US Chief
of Naval Operations put it: ‘Sea control means space control.’83 Orbiting
satellites provide global UHF and SHF communications coverage that is less
susceptible than normal radio to detection, jamming, and atmospherics. These
communications are absolutely essential for the execution of modern multi-
platform naval operations and their further improvement in terms of resistance to
detection and jamming remains a high priority. The use of laser beams for
communication will help solve these problems as well as allowing penetration
into the sea to submerged submarines. Navigational satellites, also, will give
increasingly precise fixes to naval forces. Such satellites can also be used to
provide guidance data to missile systems, both cruise and ballistic.
Surveillance satellites can already help solve navigational, and even
operational, problems by giving meteorological information, but the most
important way in which these devices will impact on naval operations is by
watching for, and keeping track of, enemy naval forces themselves. There are
two ways in which this can be done, either passively watching for
electromagnetic radiations (e.g. radars) from ships (Electronic Ocean
Surveillance Satellites or Eorsats), or by the use of active radar to detect
seaborne objects (Radar Ocean Surveillance Satellites or Rorsats).
The US Navy has access to various intelligence-gathering satellites operated
by the Central Intelligence Agency and the US Air Force. These, however, tend
to concentrate on fixed points on land rather than moving targets at sea, and the
US Navy, therefore, began in the early 1970s to experiment with its own
surveillance systems, at first attached to USAF platforms. These culminated in
the deployment of the USN’s own Naval Ocean Surveillance Satellite Eorsat
system. This is based on a set of White Cloud satellite clusters integrated with
five major receiving stations and into a system of ground processing known as a
whole as Classic Wizard. The satellites pick up not only radio and radar
emissions, but also millimetre wave and infra-red radiations. The latter can be
used to give images of the ocean’s surface and indications of what is happening
underwater. The elements of the system interact with other forms of electronic
intelligence gathering to produce in the Naval Ocean Surveillance Information
Centre a comprehensive picture of what is the situation at sea, or more
accurately what it was recently. This lack of real time data is a major defect in
the system. Ships can move away in any direction from the position in which
they were originally detected. Other problems are the arbitrariness of some of
the signal processing. Targets picked up on the edge of a cluster’s area of interest
are sometimes placed at the wrong point in the final output. The information
received from the sensors is insufficient to detect submarines with any degree of
reliability. All the system can currently achieve is the provision of basic data
from which a part of the strategic picture can be made up for transmission as
general guidance to naval units. It can be of great help to the naval commander,
but the danger is that the magical information from the Eorsat satellite is
sometimes believed rather than the visual data of the man on the spot!
In order to provide tactically useful data more promptly, in 1977 the US Navy
began developing a system called Clipper Bow to give seaborne commanders
‘all-weather/day-night active global coverage for surveillance and targeting of
hostile ships in the critical area around the task force’.84 The aim was to produce
a high-resolution active radar satellite system, but the project ran into trouble in
Congress, partly due to the technical limitations, notably an inability to deal with
aircraft as well as ships. A revised requirement for an Integrated Tactical
Surveillance System that can integrate satellite data with other sources was
issued in 1981. This requires satellites that can deal with both ship and high-
speed air targets. Such Rorsats should be in orbit by the 1990s. High resolution
should be encouraged by the use of synthetic aperture radar (SAR) techniques,
as was experimented with by NASA’s SEASAT-A experimental ocean survey
satellite in 1978. If the Strategic Defence Initiative leads to the orbiting of a
world-wide surveillance system the US Navy might take advantage of such a
facility.85
The Soviet Union has already orbited radar ocean surveillance satellites,
although not without its own problems. Soviet Rorsats use nuclear power plants,
and this caused particular embarrassment when Cosmos 954 came down over
Canada in 1978 and Cosmos 1402 broke up in late 1982 and crashed into the
Indian Ocean early the following year. Given the Soviets’ requirements to find
and target large Western ship targets and their doctrine of mass stand-off missile
attack, they were less concerned than the Americans with providing high
resolution for their Rorsats. The Soviets also deploy passive electronic
intelligence satellites, sometimes in pairs with Rorsats. Satellite surveillance is
especially useful to the USSR to provide targeting data for the latest Soviet high-
speed, long-range anti-ship missiles.86
As well as providing increasingly comprehensive intelligence information,
satellites may be used to participate more directly in maritime operations. The
development for ABM purposes of directed energy weapons in satellites, or
orbiting mirrors to direct ground emitted laser energy, opens up possibilities for
the use of such systems against naval targets, if not to damage platforms
themselves, then at least to impair their sensors. At a slightly exotic level, the
increased dependence on satellites for naval communications, navigation, and
surveillance add a new edge to the debate on the development of anti-satellite
weapons (ASATS). A campaign against satellites could have a serious impact on
the battle at sea. The naval commanders of both superpowers might feel they had
an interest in maintaining a capability to knock out such key systems in the
direction of the other side’s naval forces. Alternatively, an ASAT ban might
allow both sides to exploit space even more and in greater security in order to
increase the effectiveness of their naval forces.
Other nations, such as the United Kingdom, have deployed their own naval
communications satellites, and the diffusion of space technology promises
eventually to make some ocean surveillance capability available to navies other
than those of the superpowers. The increased transparency of the future is bound
to diminish the surprise and concealment traditionally associated with naval
operations. The nature of the maritime environment, however, with all its
parametric variations, makes it unlikely that even the superpowers will always
be aware of everything that is occurring over, on, and beneath the vast spaces of
the oceans of the world. And even if they are, many others will not be.

Interconnections among the platforms


As Norman Friedman put in in 1986, ‘it may be more accurate to characterise
the capabilities of the Navy not only in the traditional terms of platforms, but
also in the interconnections among the platforms’.87 The growing complexity of
twentieth-century naval warfare has progressively emphasized the synergy of the
various naval force elements. Ever since World War II, surface warships have
been able to use aircraft under their direction and control as an integral part of
their ‘armament’. Now the electronic revolution in the technology of information
and distribution allows still more intimate integration. As essential feature of the
automated AIO arrangements, described in Chapter 5, is their ability to converse
with each other using data links. This allows one ship to target its weapons on
the information of another, perhaps some distance away: thus ‘a fleet can
become a unified weapon system in which different capabilities are distributed
among its units’.88
This is not quite so straightforward as it sounds. Not all ships are equipped
with tactical data systems, or equipped with link equipment. The interface
between different navies is sometimes difficult to achieve. Jamming is a constant
threat, at all frequencies. Nevertheless, the various platforms of the most
advanced navies will increasingly be capable of fighting in a remarkably
integrated way. This might well make up for shortfalls in the provision of the
most expensive types of naval assets, e.g. Aegis-type systems could be used to
fire the missiles of a much less well-equipped ship. Even more so will it be
impossible to assess the naval power of a state by simply assessing the
individual members of its naval forces. The total force posture, and its ability to
act together as integrated units on the basis of the widest possible foundation of
available sensor information, will be the more reliable guide.
Part IV

The evolving environment


Chapter 7

The international and legal context

In Chapters 7 and 8 we examine the evolving environment in which sea power


will operate over the next few decades. Chapter 7 deals with the political
interactions of states which at sea increasingly involve a legal framework.
Inevitably economic questions cannot be excluded. One does not have to be a
Marxist to accept that economic factors are the foundation of national power and
policy. The more sheer resources possessed by a state the more foreign policy
instruments, both military and non-military, there are available. This chapter
must, therefore, concern itself with a broad assessment of the economic
foundations and prospects of the nations of the world and how they will affect
the international politics that will be its major subject matter. Chapter 8 will
address itself more narrowly to the economic factors that stimulate nations to
have interests in the oceans, interests which may be worth defending militarily,
as well as the factors which constrain their maritime military postures.
The world international system has been dominated since 1945 by two
superpowers, the USA and the USSR. The former has constantly outperformed
the latter in all areas except, arguably, the military. The United States looks like
remaining the possessor of the world’s largest gross national product: by the
mid-1980s it stood at $4,000 billion, about twice that of the USSR. Japan,
however, has caught up, and, depending on one’s dollar assessment of the Soviet
GNP, may well have overtaken the lesser superpower. Japan’s high growth rate
should confirm her as the world’s number two economy in the 1990s, although
some analysts estimate that China may have overtaken her by about 2010.1
Generally in industrial Asia, economic growth rates are high; European countries
continue to perform relatively well.
Both current superpowers seem to be in serious trouble economically. The
United States is facing high balance of payments and trade deficits. From the
world’s leading creditor she has become the world’s leading debtor. Curing the
USSR’s problems is the essence of Gorbachev’s Perestroika policies. Relatively,
therefore, other countries look like having a greater share of the world’s
production and income: they will, therefore, become more important.
Superpower hegemony will be mitigated and political multi-polarity enhanced.
Another effect of this process is that on the Soviet–US confrontation itself,
especially in Europe. As Jonathan Dean has pointed out: ‘The peak of the East–
West confrontation in Europe has probably passed. . . . There is a possibility over
the next twenty years of the gradual decline or attrition of the confrontation
under the combined impact of arms control, political measures and budgetary
shortages.’ Ambassador Dean goes on to conclude that

European confrontation and the rival military alliances that organise it will be
with us for many years to come. But there is no requirement for the
confrontation to be at the present level and size, cost and risk. To the
contrary, now that the high point of the confrontation has been passed, the
task is to move deliberately but also cautiously towards dismantling it to the
extent possible.2

Dean argues that the descent from this watershed must be managed carefully, a
theme taken up by Stan Windass and his colleagues in their work on ‘Common
Security in Europe’. They argue that a safe descent requires two ‘skis’, a strong
defence policy and an imaginative co-operative security policy, the latter being
negotiations between the adversary partners on arms control, confidence-
building measures, and crisis control.3 This process is liable to place an
increased emphasis on the relative importance of naval forces; sea-based nuclear
systems will probably remain the least vulnerable and most stabilizing strategic
delivery vehicles. They will probably tend to be emphasized in post-treaty
balances. Similarly, as deterrent strategies move towards increasing reliance on
their conventional component, naval forces become more important to sustain
the potential operations of theatre conventional forces (see Part V).
The increasing emphasis on co-operative security will, however, pose
something of a challenge to navies. Although naval arms control has a long and
relatively distinguished history, having played a significant part in ‘facilitating’4
the transfer of naval supremacy from Britain to the United States in the inter-war
period, the West has been markedly reluctant to accept agreed constraints on
naval forces, deployments, or activities since 1945. Only SSBN forces have been
affected by SALT Agreements and they will be included in a new START Treaty
when it eventually materializes.5 Conventional naval forces and operations have,
however, remained largely unconstrained. Indeed, the West went out of its way
to exempt ‘independent’ operations from the confidence-building measures
(notification and observation of exercises and other significant military
activities) discussed at the CDE Talks in Stockholm in 1985-6, and has refused
to discuss naval forces in the new negotiations on conventional forces in Europe.
The only naval operations included in the CDE Agreement were amphibious
landings. If amphibious exercises include 3,000 ground troops or over they have
to be notified forty-two days in advance: if they include more than 5,000 troops,
observers have to be invited.6
John Borawski, Director of the Political Committee of the North Atlantic
Assembly and a leading expert on CDE and CBMs in general, sums up the
Western view on independent naval activities thus:

Not all naval activities surrounding Europe affect European security directly
(the Falklands War or U.S. operations in the Mediterranean directed towards
Middle Eastern states) and ... in any event, . . . defining naval ‘manoeuvres’
would be difficult because ‘manoeuvring’ is a routine aspect of naval
operations. The threat to Europe, it was further argued, arose primarily from
ground forces and thus they should receive primary attention. Another
concern was that the Soviet notification measures would set a precedent for
constraint proposals, such as for a nuclear free Mediterranean or geographic
bounds on aircraft carrier deployments – a presumed Soviet objective in
pursuing this line of proposals.7

The Soviets recognized Western reluctance and at the beginning of 1986


announced the ‘deferral’ of consideration of independent naval operations at the
CDE. The USSR, however, which seems to perceive it has more to gain from
such proposals than the West, will not abandon the demand for naval co-
operative security measures. Indeed, from 1986, Soviet spokesmen, including
Gorbachev himself, produced an impressive ‘menu’ of such proposals. These
included the mutual withdrawal of Soviet and American naval forces from the
Mediterranean, the placing of limits on the movements of ships equipped to
strike coastal areas with nuclear weapons, and a ban on ASW measures in
ballistic missile submarine ‘sanctuaries’. All these measures ran counter to
Western naval strategic thought and also raised considerable problems of
verification for the West. Nevertheless, the Soviet ‘menu’ also contained serious
proposals which stand more chance of forming the basis for agreement. Soviet
spokesmen have called for agreements not to conduct naval exercises in
international straits or other agreed areas, limits on the numbers of naval
exercises held annually, and the advance notification of exercises and the
invitation of observers. Various threshold figures have been suggested for
notifiable or constrained activities. At Stockholm the Soviets suggested 30 ships
and 100 aircraft as the size of exercise beyond which notification would be
mandatory. In November 1986 Mr Lygachev, then number two in the Politburo,
put the level at 25,000 men (a large carrier battle group with two carriers might
contain about 15,000). It was suggested that an exercise greater than this figure
should not be held ‘more than once or twice every one or two years’.8
Such proposals as these might well form the basis of a regime of naval
confidence and security building measures that could evolve over the next
decade. Indeed, unless the West shows some (albeit necessarily limited)
recognition of Soviet maritime concerns, the whole conventional arms control
process might make but slow progress. There are useful precedents for purely
naval co-operative security measures. Both the United States and the United
Kingdom have signed bilateral ‘Incidents at Sea’ Agreements with the Soviet
Union, which regulate operations when the ships of those states are in close
proximity so as to avoid the risk of either collision or accidental combat.
Weapons are not to be trained at each other, nor objects launched in the other’s
direction, nor searchlights shone to dazzle those on the opposing bridge. In order
to reduce the risks of collision both treaties include agreements not to ‘conduct
manoeuvres through areas of heavy traffic where internationally recognised
traffic separation schemes are in effect’.9 This is an interesting precedent for
wider CSBM (confidence and security building measures) application.
As East and West become further entwined in this web of common security
their confrontation may well become gradually less significant as the primary
focus of both their naval rationales and their naval deployments. The focus
might well shift to the utility of naval forces in sustaining national interests on a
more global scale.
It seems unlikely in a world of sovereign states that global conflict will
become any less endemic in future decades. As Ken Booth has put it: ‘The
periodic explosion of international conflict is one of the few things that can be
taken for granted in an unpredictable world.’10 Sir James Cable would put things
a little more strongly: ‘The world is not becoming a more peaceful, orderly or
law abiding place. When comparing conditions from 1920 to 1939 with those
after 1945, the growth of lawless violence is striking and progressive. More than
ever before governments today are prepared to use force not only against their
own nationals, but also against foreigners. And these are not the governments
which started most of the trouble in the past.’11
Many conflicts, especially those not directly involving both superpowers, do
not have much perceived potential to raise the dreaded nuclear stakes.
Intervention operations by the superpowers to restore a favourable situation
within their acknowledged spheres of influence seems still to possess a
legitimacy which shows little sign of going away. This takes a maritime form in
the American Caribbean sphere of influence, where the Grenada landings of
1983 provide a classic near contemporary example. This American sphere is less
sharply delineated than the Soviet land empire and the Americans show
somewhat greater restraint in ‘sending in the Marines’ than they did before 1939,
a restraint from which contemporary Nicaragua benefits. Nevertheless, it would
not be surprising to see future Grenadas, given the right circumstances of chaos
and power vacuum ashore. When things are not quite so chaotic, naval forces
may still exert pressure, even if direct intervention may be more difficult both
politically and operationally.
Sir James Cable is right to place the degree of organization of modern
governments above the sheer military capability provided by modern weapons as
the major factor making applications of great power naval force more difficult.

Few nations can claim an increase in political wisdom during this century,
but many governments now have more control over people and resources. By
the narrow and sinister test of their ability to sustain organised combat, the
governments of older and more civilised countries have lost much of their
earlier advantage. In terms of political and military discipline, they are closer
to a rough equality with lesser developed nations.12

This higher degree of organization goes hand in hand with a greater degree of
regard for national sovereignty. In a world where at least lip service is paid to the
UN Charter, the political legitimacy at governmental and popular levels of one
country using blatant force or threats of force against another has significantly
diminished. This does not mean, however, that states will always deny
themselves the option to intervene by sea. Ken Booth sums up the balance of
factors well:

International politics is always changing, but more kaleidoscopically than in


a linear direction. ‘Intervention is dead’ proclaimed the war in Vietnam.
‘Long live intervention’ proclaimed some Africans in the 1970s despite the
rhetoric of the OAU. Colonialism had gone, but a form of tacit neo-
colonialism seems to have arrived. The weak will always look for protectors;
and the strong will always look for gain. Interventions will therefore continue
to take place. But if they will not be as frequent as in the historic ‘age of
imperialism’, because of changing cost-benefit calculations, they will
certainly be more frequent than was generally assumed during the anti-
intervention mood which characterised Western thinking in the years
following the American tragedy in Vietnam. The continuing ‘agony of the
small nations’ will provide frequent opportunities in the future for powerful
states to intervene for ambitious purposes, to intervene as a result of feelings
of insecurity, to intervene in response to requests, to intervene to pre-empt
another’s intervention and to intervene to maintain the principle of non-
intervention.13

The United States has been the major recent practitioner of maritime
intervention and may well remain so.14 Her network of world-wide
commitments and interests, and the simple fact that these are truly overseas
commitments and interests will mandate a forward power projection role. Even
though American naval power may begin to be constrained by the twin factors of
deficit reduction and a slightly less forward strategic posture in the post-Reagan
years, the fact of America’s traditional superpower position and the perception
that the Western buck stops in the White House, will continue to pull her into
various uses of the US Navy and Marine Corps when they are considered
appropriate. Such uses could escalate into large-scale military operations in
which shipping would be vital for logistic support.
The Soviet Navy, even at the height of the Gorshkhov era, was always less
ready than its superpower rival actually to fire its guns or missiles in anger. It
tended to be used more to neutralize American options than act as the main
instrument of Soviet intervention, although this is not to underestimate its role
in, for example, the Angola crisis in facilitating the insertion of friendly Cuban
forces to sustain a newly established revolutionary regime.15 Such activity
remains conceivable in future, although there will be fewer opportunities to act
given that there are no more traditional formal empires, except perhaps the
Soviet Union’s own, to break up. Moreover, the USSR does seem to have revised
its approach to the utility of this degree of support for the Third World
‘progressive’ regimes. Countries sustained by Soviet and Cuban arms and aid
have not proved necessarily to be loyal satellites. Even more importantly, the
Soviet political leadership has come to recognize the dangers of pursuing the
revolutionary process in an armed manner, except where absolutely
unavoidable.16 One may expect the Soviets to be even more reluctant than before
in using its fleet. Nevertheless, in its world-wide deployment the Soviet Navy
remains a vital factor in demonstrating direct Soviet interest in world events. As
one study of Soviet naval diplomacy has put it: ‘The Soviets are now universally
understood to be fully fledged participants in the diplomacy of the Third
World.’17 The feathers of that plumage are in large part the ships and auxiliaries
of the Soviet Navy.
The bird itself, however, must be sufficiently large and imposing if the
feathers are to be of real meaning. As Edward Luttwak has made clear: ‘It is the
overall political context that will determine the political effectiveness of Super
Power navies, and not their tactical capabilities, since it is this context that
governs the application of national power, the true determinant of which the
naval element is only one constituent.’18 It is the overall power of the Soviet
Union that is called into play when the Soviet Navy puts in an appearance. This
makes up to a very considerable extent for any defects in capability on the spot.
Less powerful nations cannot hope to approach this level of peacetime naval
diplomatic weight, even if they deploy navies of significant strength. Currently
the two main ex-imperial powers, Britain and France, still deploy large and
imposing navies, indeed the latter’s gives a particular emphasis to operations
outside Europe. From the mid-1970s, when France’s carriers were redeployed to
the Mediterranean, the French have been involved in a range of operations,
notably off Lebanon, in support of their contribution to the multinational forces
ashore. This involved two carrier air strikes, one against a Syrian battery and the
other against a Palestinian camp. France’s navy contributes to guaranteeing
Djibouti’s security and her Indian Ocean Squadron there allows her flexibly to
operate in the Gulf and Red Sea as required by the perceived threat to French
national interests and shipping. France has also quietly supported the Seychelles
against South African mercenaries and has demonstrated resolve to protect her
remaining possessions in the Indian Ocean, e.g. Mayotte from the Comoros, and
Tromelin from Mauritius. France has similar requirements in the Pacific where
her use of her Pacific islands as nuclear weapons testing sites has created a
particularly sensitive context. Until the New Caledonia problem is settled the
uneasy situation there is also one which will probably require continued and
frequent naval deployments.19
France has chosen to integrate her far-flung relics of empire into the national
state, something for which Britain has shown a traditional reluctance. Indeed,
misleading signals of determination to hang on to the Falkland Islands tempted
Argentina into an attempted island grab in 1982 that turned into the most notable
naval event of the decade and the most significant traditional ‘fleet versus fleet’
battle of the post-war era. The war also showed that one should not
underestimate the residual power of an ex-colonial state in possession of a few
imperial relics. There is still a capability to embarrass those who go too far to
exploit apparent weaknesses. Yet, a medium to long term view must share Sir
James Cable’s doubts about the long-term ability of those ‘medium navies with
ambitions above their economic station’ in life to continue to play the national
world power role.20 When the case for a world-wide or ‘out of area’ role for a
European maritime force is sustained by lists of past interventions, mostly
carried out in a period when the state in question actually owned a long
dismantled colonial empire or was following a responsible policy of making sure
a pro-Western regime inherited the fruits of imperial withdrawal, the temptation
to reach for one’s intellectual pistol is especially acute. British naval spokesmen,
both official and unofficial, were recently prone to such indulgences in imperial
nostalgia, almost as if the sun had never set on a red coloured globe or the
withdrawal from East of Suez in 1968-71 had never taken place.21
There are, however, two factors which may well provide extra substance to
the ‘out of area’ pretensions of Europe’s navies, both great and small. The first is
the desire to maintain good relations with the United States. As former Secretary
of Defence Caspar Weinberger made clear to his European colleagues at the
Sealink Conference in Annapolis in 1986, some Americans, at least, feel the
Europeans have a duty to support their American ally in dealing with the ‘global
maritime challenge’.22 ‘We must have leverage for action,’ he argued, ‘in corners
of the globe far from the North Atlantic, places where developments can
mightily affect NATO’s security. We need to protect all of our NATO flanks.
And we can only protect all of NATO and its flanks if we are fully prepared to
take actions outside the NATO theatres. “Out of Area” can no longer be the
automatic veto some feel that outworn geographical tag fastens on NATO.’23
Weinberger went on to define as vital interests the ‘critical sea lanes’ in the
Pacific, Indian Ocean, and the Gulf.24
The outbreak of war between Iraq and Iran in September 1980 had indeed
seen British and French ships along with American patrolling off the entrance to
the Gulf. These were purely national efforts on the part of the two European
nations, although they did give comfort to the United States and were clearly
appreciated. In 1987, shortly after a major British exercise in Oman, the naval
dimension of the war, previously largely an Iraqi air offensive against Iranian
shipping, was significantly stepped up. Increased Iranian attacks on ships in
international waters and a build-up of Iranian capabilities to attack shipping
serving Iraq’s Arab allies, notably Kuwait, led to the American declarations that
it would use force to maintain rights of passage for merchantmen in the Gulf.25
During 1987 the British and French stepped up their own naval operations in
the Gulf alongside those of the Americans, but were careful to maintain their
respective efforts as national enterprises and within certain constraints. The
British, for example, notably refrained from accompanying other than Red
Ensign ships, and went no further north than Bahrain. During July the mining
threat became serious but the British and French risked US displeasure, and
showed their own reservations about certain aspects of US policy, by initially
refusing to send MCM vessels to operate directly with the US forces. With,
however, the extension of the mining threat on the south-eastern side of the Strait
of Hormuz, in the middle of the Royal Navy’s main operational area, the two
European powers reversed their position in August and announced plans to send
minehunters to operate in support of their existing forces. The British, indeed,
also took the lead in trying to interest other European governments in a
multinational MCM force.26
The Americans were very happy with this development. As a contemporary
British press put it ‘there was glee in Washington that, finally, one of the Western
allies was doing something to ease the U.S.’s burdens in the Gulf’.27 Initially, the
other Europeans had reservations about joining the British but, assured of the
military, technical and logistic support of British warships, the Dutch and
Belgians decided in September to send out MCM vessels. Following an attack
on one of their merchantmen the Italians went further and decided to send out a
purely national full patrol of three MCM vessels and three frigates in order to
give protection to their ships all the way to Kuwait.
The latter decisions had much more to do with national and European
perceptions of interest than the desire to contribute to a US presence and respond
to American pressure, or even control US action (although the latter may have
been achieved). Moreover, the deployment of forces into a European multilateral
framework played a vital part in the decision-making processes of some
countries which were unused to sending their naval forces so far afield. Without
such a framework, no deployment decision might have been possible. The Dutch
used their tenure of the Chairmanship of the Western European Union (WEU) to
bring together in August 1987 senior officials of all seven member states to
discuss concerted naval action. There was little desire to set up a ‘WEU
Squadron’. The British, and even more so the West Germans, did not want to
appear to be ‘ganging up’ on the Americans. France, pro-WEU at a general level,
proved equally anxious to preserve freedom of action on the spot. She was,
therefore, reluctant to enter into too integrated a force. The Italians, for their own
political reasons, chose to emphasize independence also. Nevertheless, after the
first August meeting a process of ‘concertation’ was established, by which the
WEU framework was used as a forum for concerted action working at three
levels: ministers and their senior representatives, experts, and the officers on the
spot. Even Luxembourg gave financial support to the Dutch and Belgians,
creating a Benelux flotilla which operated closely with the British, who provided
surface ship cover. In 1988 the relationship was formalized and a joint Anglo-
Benelux unit eventually formed under British command in order to allow a
reduced presence of MCM vessels to be used to optimum effect. The WEU
concertation process was used to achieve joint MCM operations in the summer
of 1988 and also provided a context for wider rationalization. The West Germans
helped cover for the absence of Italian and US ships from the Mediterranean by
deploying forces there in late 1987 and early 1988. Finally, after the Gulf War
ended, in the most obvious show of European concertation yet, the WEU took
upon itself the role of clearing up the remaining mines in the southern Gulf Area
in Operation Clean Sweep.28
The Europeans, almost despite themselves, crossed a number of thresholds of
co-operation. The progressively closer Anglo-Benelux arrangements, first on
mutual support and then on a joint force, were fundamental steps forward given
previous doubts about even the legality of one navy’s warships protecting
another. The use of WEU to discuss co-operation was also a significant moment
in the rebirth of this organization. The Europeans, dependent on Gulf oil for
much of their supplies and interested in the freedom of navigation of their own
vessels, were being drawn into a new kind of ‘out of area’ commitment, one that
reflected real and modern necessities, rather than imperial echoes and one that
was, slowly and diffidently, but nonetheless significantly extending the bounds
of European defence and foreign policy co-operation. The European naval force
in the Gulf might be a real foretaste of things to come as Europe’s further
integration inexorably continues and the United States’ perceptions of the world
may begin to diverge more and more as the latter looks West rather than East as
its primary area of concern.
It has become increasingly fashionable to emphasize the growing strategic
importance of the Pacific, both in itself and as a preoccupation for the United
States. As Colin Gray told the audience at the 1986 Sealink Conference: ‘We
may debate the when and the how, but East Asia is becoming a – and perhaps
even the – predominant center of world power. The security architecture of this
region and the geostrategic significance of the region for other regions – NATO
Europe in particular – will naturally change as China modernises and Japan
adjusts to China’s modernisation.’29
Currently, Japan is the economic giant of the region, although she has yet to
transfer that dominance into military or naval terms. Her 1945 aversion therapy
to strategic ambitions remains strong and effective. Her heavy dependence on
Gulf energy supplies has not shaken her traditional reticence about military
involvement far from her shores. Japan shows every sign of continuing to be
relatively cautious in adding military fledging to her new economic superpower
capacity, but her navy will soon be deploying some of the world’s most
sophisticated non-superpower surface combatants. Given her extraordinary
dependence on maritime trade and her possession to a greater degree than any
other state of the traditional attributes of seapower, a large mercantile marine and
a merchant shipbuilding infrastructure, it would be surprising indeed if her
military potential at sea, and her willingness to use it if necessary, did not
increase as World War II disappears increasingly into the mists of history.
Japan’s declaration of an interest in the defence of shipping a thousand miles
from her coastline along the ‘shipping lanes’ to Taiwan and towards Iwo Jima is
no doubt only a first step.30
It also seems likely that China will build up, albeit slowly, her maritime
capabilities and her capacity to engage in naval diplomacy at increasing
distances from her shores. The Chief of the Chinese Navy has stated that ‘as the
strategic position in the Pacific is becoming more important . . . and as China is
gradually expanding the scale of her maritime development, the Chinese Navy
will have to shoulder more and heavier tasks in both peacetime and war’.31 The
main confrontation is in the South China Sea with Vietnam and her Soviet ally,
but from 1980 Chinese ships began to move further afield. In that year there was
a naval cruise to the South West Pacific to recover rocket re-entry capsules, and
in 1981 Chinese ships sailed round the Philippines. In 1984-5 Naval auxiliaries
visited Chile and Argentina on their way back from establishing an Antarctic
scientific station, and in 1985-6 a destroyer and an auxiliary cruised into the
Indian Ocean to visit Bangladesh, Sri Lanka, and Pakistan. The latter
deployment was of particular interest, especially as the small Chinese squadron
exercised with the US Navy on the way home. One American analyst has
summed up the significance of this cruise as follows:

By Western standards the PRC naval cruise to the Indian Ocean was merely a
routine training exercise. However, within the context of the Third World
navies it was a significant accomplishment. Chinese sailors further proved
their ability to refuel ships at sea and demonstrated their ability to maintain
communications over great distances for an extended period of time. The
political advantages accrued to Beijing were substantial and may prove to be
of lasting benefit. Overall the cruise was a notable achievement in China’s
naval development and will probably be a stimulus for more distant visits
over the next year or two.32

China’s actual capability to project real military power much beyond the South
China Sea will remain very limited for some time, but as her latent economic
potential is unlocked there is little reason to doubt that she will use the growing
reach of her navy as at least an expressive token of that strength and potential.
This potential growth of Chinese naval power acts as another stimulus to
India to make her own the major Indian Ocean navy. China is not the only
‘threat’ however; India still remembers the US deployment of the Enterprise
carrier battle group as an exercise in coercive diplomacy in the 1971 Indo-
Pakistan War. With a navy centred around two light aircraft carriers (and with
plans for a third of her own construction) operating modern STOVL fighters, a
much more capable surface combatant fleet than China’s, the beginnings of a
nuclear submarine force and long-range maritime patrol aircraft, India possesses
‘the largest and most balanced fleet in the Indian Ocean’.33 She can lay claim to
a respectable capacity to project military power into the seas around her shore, a
capability that is more than one of mere defence. The main targets (or sometimes
potential beneficiaries) of Indian naval power projection are her neighbours,
Pakistan (the adversary above all), Bangladesh, Sri Lanka, the Maldives, Burma,
and Indonesia. The main threat to Indian naval hegemony in the area will remain
the presence of the navies of the superpowers, especially that of the United
States. It is thus hardly surprising that India has been a major sponsor of ‘zone of
peace’ proposals which might shut out outside actors and confirm her own naval
hegemony. Until such measures come into effect, India will seek to use her navy
as a deterrent to the projection of power into the Indian Ocean from outside.
India’s powerful navy, however, might be a two-edged sword. As one informed
analyst has warned:

In so far as it complicates the force planning of a particular extra regional


threat it will have served the defensive purposes for which it was created. Its
inherent capability to unnerve India’s regional neighbours, however, is a
political issue that merits greater attention. Unless India’s military expansion
is accompanied by a greater willingness to be more sensitive to the concerns
of smaller neighbours, its force superiority will never adequately translate
into political pre-eminence. In these circumstances, the naval forces acquired
for the general purpose of greatness may become the very source of larger
political discomfort and discord.34
An Indian Ocean ‘zone of peace’ measure may indeed be a contribution to the
solution of this thorny problem. This could legitimize some significant
‘peacekeeping’ role for India in the area which is clearly acceptable for countries
like Sri Lanka and the Maldives, which have already accepted Indian
intervention. Such a role, however, would be clearly unacceptable for Pakistan
unless Indian ambitions and capabilities were themselves significantly curtailed.
Other, more distant Indian Ocean powers, like South Africa and Australia, would
also be concerned to see India’s wings clipped by such a measure.
A world where superpower rivalry has had the edge taken off it by a new and
more substantial era of détente, and where other naval powers, including Europe
in new political forms, are becoming more active, also allows other states to play
a more leading role in the local naval balance. They may indeed be able to assert
a local naval leadership. Such a state is Australia, currently reassessing her
defence policy in a more local context and de-emphasizing her traditional role as
an expeditionary contributor to British or American campaigns. Although
Australia has not abandoned her commitments on the Asian mainland, she has
been taking a careful look at the defence priorities of ‘one of the most secure
countries in the world . . . distant from the main centres of global military
confrontation, and . . . surrounded by large expanses of water which make it
difficult to attack’, especially a country with neighbours possessing ‘only limited
capabilities to project military power against it’.35 In his wide-ranging report
presented to the Australian government, Paul Dibb identified as an important
security interest ‘a strong stable region free from external pressures’. The area
for exerting ‘independent power’ was defined as stretching ‘over 4,000 nautical
miles from the Cocos Islands in the west to New Zealand and the islands of the
South West Pacific in the east, and over 3,000 nautical miles from the
archipelago and island chain in the north to the Southern Ocean. It represents
about 10% of the earth’s surface’.36 And most of it is sea.
Dibb went on to define Australia’s national role as follows: ‘In the South
Pacific we are perceived as being by far the largest power. Our fundamental
security interest is to maintain the benign strategic environment that currently
prevails, free from unwelcome external pressures. Our foreign policy, aid
programmes, and defence policy should be co-ordinated carefully with other
regional states, in particular New Zealand, so as to discourage Soviet naval
visits, or other unwelcome military access in the South Pacific.’ In other words,
local actors of significant power prevent outsiders taking advantage of power
vacua.
This policy of ‘maintenance’37 implies powerful maritime forces, at least by
local standards. Land-based, over-the-horizon radars and aircraft, along with
conventional submarines, provide useful surveillance and strike capabilities as
well as defensive air cover, but reach and presence can only be provided by a
surface fleet. Dibb recommended twenty-four ‘platforms’, a three-tier force of
up to nine destroyers (currently American guided-missile destroyers and large
frigates), and up to ten ‘light patrol frigates’ with a similar number of smaller
patrol vessels in the 200 ton class.38 Preliminary work on the new frigates is
being pursued jointly with New Zealand. The latter, currently isolated from the
USA because of its anti-nuclear rhetoric, places much emphasis on its continued
defence connections with Australia at the various levels within the
Australian/New Zealand Defence Consultative Committee (ANZDCC)
framework. Such consultations reflect New Zealand’s recognition that ‘it shares
vital strategic interests in the South Pacific with Australia’.39
Given the limited opposition in Southern Pacific waters, developed countries
like Australia and New Zealand are regional powers of some importance with a
long reach measured in thousands of miles. In many other areas, however, albeit
over shorter distances, local powers will feel the need to assert their sovereignty
and control in areas for which they feel responsible, and in which they feel
legitimately interested. This may well cross the line between mere constabulary
duties and ‘gunboat diplomacy’. As Sir James Cable has put it: ‘Because more
nations than ever before now have sea coasts, navies and maritime interests to
quarrel about, political considerations make it likely that threats and/or limited
uses of naval force will grow rather than decline during the rest of the century.’40
What seems to be emerging is a system where the superpowers will remain
for a time supreme in the area of what Luttwak calls ‘naval suasion’.41 They,
especially the Americans, will use navies to compel smaller nations to do
something, to make them desist from some activity, or to deter them from
carrying out certain actions contrary to the perceived interests of themselves and
their clients. A second group of powers, some on the upward slope, some on the
downward, will attempt with more difficulty to act as assailants in exercises of
limited naval force. Those on the downward slope may, however, play a
significant role in superpower confrontations, and they may also learn the
usefulness of acting collectively. A mass of medium and small states will always
have the motivation and means to defend or assert their national interests at sea
either collectively or, more usually, nationally. This is especially so given the
connected developments in both resource exploitation and the law of the sea that
give all littoral states substantially increased off-shore areas of responsibility and
control.
As the sea has been used more it has been more necessary to regulate its use
through the medium of international law. The attempt to create a universally
accepted modern legal framework that was the Third United Nation’s
Conference on the Law of the Sea (UNCLOS III) has not yet been fully
successful. Despite eight years of negotiation, from 1974 onwards, complete
agreement proved impossible on one outstanding issue, the future of the deep
seabed. President Reagan’s USA loyally followed by its two most faithful allies,
Britain and the German Federal Republic, refused to sign the resulting United
Nations Convention on the Law of the Sea that opened for signature at Montego
Bay in 1982. Although no less than 159 countries had signed the Convention by
the deadline of December 1984, the opposition of the Americans, British, and
West Germans continued.42
This hostile attitude still prevents the 1982 Convention from forming the
formal basis for the international law of the sea. Nevertheless, its main
provisions, which were already becoming the norm anyway – a twelve mile
territorial sea over which littoral states have full sovereignty (saving rights of
other’s innocent passage), a 200-mile exclusive economic zone in which coastal
states have the sole right of exploitation of living and non-living resources, and
similar monopoly rights in certain circumstances to manage the same resources
on adjacent continental shelf beyond that – have become accepted as the
customary law of the sea. Non-signatory nations, like the United Kingdom, have
extended their territorial seas to twelve miles, and it will not be long before all
states have unilaterally conformed to the new rules, both in the rights they assert
and others’ rights they accept.
Even the 1982 Convention’s greatest enemy, the United States, has stated that
most of its provisions ‘reflected prevailing international practice’ and that,
therefore, ‘their substance . . . could be invoked by non-parties as representing
new customary international law’.43 As Ken Booth has pointed out: ‘The history
of the law of the sea has been predominantly a story of the evolution of
customary rules through state practice. The factors contributing to order at sea
have always been relatively strong.’44
Nevertheless, especially in a context where there is no universal signed and
sealed agreement on what the law of the sea actually is, there is still scope for
conflict. In his excellent paper, ‘A Sea of Troubles’,45 Barry Buzan analysed the
‘sources of dispute’ that could arise over law of the seas questions. These came
in four groups, one concerning disagreement over national boundaries, the
second concerning the rights within national boundaries, the third over ‘rights in
the ocean beyond national jurisdiction’, and the fourth ‘disputes arising from
non-ocean sources’. Among the first are (a) ‘delimitation of opposite or adjacent
boundaries’, (b) ‘ownership of, or rights pertaining to, islands’, and (c)
‘delimitation of the outer limits of national jurisdiction’. The second includes
disputes over (a) fishing rights, (b) non-military navigation rights, (c) rights to
carry out military activities, (d) rights to carry out marine scientific research, and
(e) rights to exploit the seabed. The third general category is not subdivided and
it has indeed turned out to be the major problem in preventing universal adoption
of an agreed law of the sea regime. The attempt to make the distant oceans ‘the
common heritage of mankind’ brings us to the fourth major category which
includes disputes over (a) ‘the rights of land-locked states’, (b) ‘the distribution
of power or rights within states’, and (c) ‘territorial disputes involving coastal
states’.46
In viewing the ‘law of the sea’ problems that have arisen during the 1980s one
can see examples of several of these categories, often in an overlapping form.
Disputes between the Soviet Union and its Scandinavian neighbours over
maritime boundaries led to the cutting of a Norwegian seismic research ship’s
cable and the boarding and expulsion of various Swedish, Danish, and West
German fishing boats in the Eastern Baltic. The United States deliberately
exerted its right to sail warships through territorial sea limits it did not accept
and to exert the right of innocent passage of warships in territorial waters over
the protests of the sovereign power. (Sweden and the Soviet Union were the
respective targets.) The conflict over Libya’s claim to internal waters sovereignty
over the Gulf of Sibra came to blows in 1986, as it had in 1981. The United
Kingdom very successfully established a 150-mile fishery protection zone
around the Falkland Islands which directly conflicted with Argentina’s 200-mile
Exclusive Economic Zone (EEZ). Canada asserted internal waters claims to the
North West Passage which the United States had opposed to the extent of sailing
a Coast Guard icebreaker through the disputed area without asking Canadian
permission. The uninhabited islet of Rockall and its associated continental shelf
rights remained a bone of contention between the United Kingdom, Ireland,
Iceland, and Denmark. France had problems with the USSR over seabed
exploitation rights, and asserted claims to forbid research in its EEZ except by
permit, while Norway prevented the Greenpeace ship Moby Dick entering its
territorial sea without permission.
One should not overstate the significance of legal disputes such as these.
Where there is a political will, peaceful settlement is both possible and probable.
One notable case is the Beagle Channel settlement between Argentina and Chile
ratified in 1985. Thanks to Vatican mediation the dispute, which had caused
considerable tension between the two states concerned, was settled on the basis
of the sovereignty of three islands going to Chile with an adjacent twelve miles
of territorial waters, but Argentina retaining a 30-40 mile EEZ in the Atlantic.
The International Court of Justice proved useful in arbitrating other disputes, e.g.
between Tunisia and Libya, and Malta and Libya. Indeed, only where wider
political factors created additional tensions, notably in the US–Libyan affair, was
conflict pushed to the level of armed confrontation involving the two sides’
naval forces.48 The wider political factors were clearly more important. The
United States, so anxious to uphold the law against Libya, totally ignored the
strictures of the International Court of Justice when it was criticized for laying
mines in Nicaraguan territorial waters. A Nicaraguan Security Council motion to
enforce the judgement was simply vetoed by the United States.
As usual in international politics, international law is only useful when two
states wish to settle disputes by agreement. These, happily, cover the majority of
cases but some have been disturbed by ‘a broader willingness on the part of the
major powers to disregard the territorial sovereignty of other states and, as a
matter of deliberate policy to place military or police expediency above
compliance with international law’.49 The superpowers are major offenders. The
USA’s action in using carrier aircraft to seize an airliner carrying the Achille
Lauro hijackers was one such example, along with the reported violations of the
territorial waters of Sweden and other neighbouring states by Soviet submarines
of various shapes and sizes. This is a dangerous and slippery slope. Other
aspects of this dangerous trend to lawlessness are the continued prevalence of
piracy and the beginnings of maritime terrorism, and the equally lawless great
power responses to it just discussed (see Chapter 10).
Nevertheless, one should not be too pessimistic. It is likely that the dominant
trend will in fact be the opposite, towards an increasingly firm framework of
maritime law at sea. As the disadvantages of noncompliance with the UN
Convention became clear, e.g. the non-access to UN procedures for settling
seabed exploitation rights (as agreed within the framework of the 1982
Convention by the USSR and Japan), the small rejectionist front might crack.
Whatever the precise status of the 1982 Convention, however, as Ken Booth has
pointed out, UNCLOS III might well not ‘be the end of the conference circuit
but merely the first lap’.50
This brings us to the phenomenon of Creeping Jurisdiction, defined by Booth
as ‘the extension of national or international rules and regulations, and rights and
duties over and under the sea, in straits and coastal zones, on and under the
seabed and in the vast stretches of the high seas’.51 It is hard not to agree with
Booth when he goes on to argue that ‘In future, technology, interest and the will
to govern seem set to fill out large chunks of the map of the sea with appropriate
forms of national and international administration as inexorably as railways, the
industrial revolution and nationalism spread control throughout the land masses
in the last century.’51
The exact nature of the EEZ jurisdiction has already led to disagreements, and
these will continue. Moreover, foreign naval forces may well come in for
unwelcome attention and attempted restraint. Booth concludes that

the secular trend is adverse to the naval interest – as traditionally and


conventionally understood – despite the rights enshrined in the 1982
Convention ... the long term future of naval mobility is under challenge. It
may take half a century but unilateralist drives to parcel parts of the ocean
will continue and will be legitimised by the territorialist mood of the
international community. As this development unfolds, and as state control
intensifies over larger patches of sea, greater meaning will be invested in the
new boundaries which are inevitably the outcome of the process. Nations will
feel protective and sensitive – indeed patriotic – about these patches of
ocean. Hence it is appropriate to talk about the spread of what might be
called ‘psycho-legal boundaries’ at sea.52

The major naval powers will inevitably oppose the creep of ‘territorialisation
and psycho-legal boundaries’ but analysts such as Booth and Alexander53 put
forward convincing cases that this will be a rearguard action. Increasingly, legal
questions will interact with political ones in exercises of naval power, especially
at the lower levels. Indeed, creeping jurisdiction will create naval opportunities
as well as problems. Having boundaries that are such but which are ‘less
immediately sensitive and further from the national nerve endings than those on
land’54 could well provide new and useful options for assertive and forceful
activity without the same risks of similar confrontations on land.
The greater territorialization of the oceans may well act as an extra impulse to
explore more international uses of naval power. As the relations of the two
superpowers improve, the barriers to international peacekeeping may well
decline. Soviet ‘new thinking’ in the mid-1980s has dramatically changed Soviet
approaches to UN peacekeeping in a positive direction. Indeed, it was the West
that strongly opposed a UN naval force being set up in the Gulf. Although there
were other more practical problems, some in the West feared such a force might
legitimize the existing Soviet presence in an area of very significant Western
interest.
Such concerns, however, may diminish in the future. As they do, the technical
and legal problems that have traditionally been held to prevent UN maritime
action – problems of mutual defence if acting outside Chapter 7 of the Charter,
problems of common communications and rules of engagement, problems of
command – might well suddenly seem soluble. Indeed, the more general
possession of sophisticated naval capabilities by non-aligned countries such as
India might allow a wider spectrum of potential participants in joint
peacekeeping operations. There are useful operational precedents for joint action
by a large number of different navies in a UN framework, e.g. the Korean War.55
Whatever the political framework, the evolving political and legal context
will allow wide scope for the use of maritime military power. It would be fair to
allow Ken Booth, upon whose work this chapter has relied so heavily, to be
allowed the last word, especially as it helps clear up a slightly naughty case of a
quotation out of context. James Cable took Booth to task in a recent article for
implying that traditional ‘gunboat diplomacy’ was dead.56 The full statement
reads as follows and sums up the findings of this chapter very well: ‘While there
is no prospect of the revival of gunboat diplomacy which characterised the age
of imperialism in the last century’, naval forces ‘will remain instruments of
diplomacy for several countries and the defenders of the maritime sovereignty of
many more’.57
Chapter 8

Economic stimuli and constraints

Mahan argued that military sea power grew out of the economic uses of the sea
and the importance of those uses to the state. The seas, however, are now
arguably less central to the power position of the great powers than they once
were. The modern Mahanian can quote persuasive figures of greatly expanding
world trade volumes, more than the lion’s share of which goes by sea (see
Chapter 3). Certain countries are indeed more dependent on overseas trade than
they once were, e.g. Japan which exported about 10 per cent of its GNP in 1960
and exported about 14 per cent a quarter of a century later. Japan also depends
entirely on seaborne trade to keep all her economic activities going in terms of
energy and raw materials. Britain still exports about 20 per cent of her GNP and
depends on the seabed for the energy supplies of oil and gas that are vital for her
agriculture as well as her industries. Other states for whom seaborne exports are
as important as they are to Britain and Japan are states such as Israel, The
Netherlands, New Zealand, Norway, Nigeria, Sweden, and Venezuela (all 20 per
cent or over), and Argentina, Australia, Canada, Chile, Egypt, France, the
German Federal Republic, Iran, Italy, Saudi Arabia, and South Africa. Oil-
producing states are especially dependent on seaborne trade and oil is also a
particular item of export dependence in the developed world. It is no coincidence
that The Netherlands, dragged somewhat against her will into a far-flung naval
commitment, is almost two-thirds dependent on Gulf supplies of oil.1
The continued economic interest of such states in using the seas is clear,
although maritime policy is shaped in interesting ways by extraneous factors. In
Japan’s case, for example, her post-1945 reticence about exerting military power
to parallel her economic strength has limited her military navy in both size and
capability. These restraints are steadily and slowly loosening, but it will be a
long process. In the United Kingdom’s case the civilian aspects of sea use have
been allowed to wither to a remarkable degree. British governments, especially
the current one with its fierce ideological commitment to the free market, both
national and international, leave Britain’s maritime industries to be squeezed by
states with less liberal ideals and practice and/or less expensive ships and crews.2
Britain’s military naval forces also have become primarily orientated to power
projection in both its nuclear and conventional forms as they have been given
both the custody of the national deterrent and become locked into the United
States’ side of the superpower balance.
Neither major naval superpower relies on the economic uses of the sea. The
United States exports between 4 and 5 per cent of its GNP, and over a quarter of
those exports go to Canada and Mexico. Given the USSR’s connection by land
to her major markets her figure is even lower, perhaps about 3 per cent, although
her merchant marine is a significant ‘export’ in itself, especially as a hard
currency earner. The Soviet Union is, however, self-sufficient in terms of
dependence on overseas imports. She has to import significant amounts of food
products to sustain domestic living standards but she does not rely on overseas
supplies. She is also very well off for domestic supplies of oil (with the largest
reserves outside the Gulf) and non-fuel minerals, although she does import
significant quantities of the latter from non-CMEA countries, especially tungsten
and bauxite/aluminium.
With the United States the matter of import reliance is much more
controversial. Much is often made of American ‘dependence’ on overseas
supplies of fuel and raw materials. As President Reagan has put it: ‘The United
States is a sea power by necessity, crucially dependent on the trans-oceanic
import of vital strategic materials. For this reason our navy is designed to keep
open the sea lanes world wide . . . maritime superiority is for us a necessity.’3
The figures to sustain this theory look superficially impressive. The United
States still imports about 30 per cent of its oil, and is heavily dependent on
imports for a wide range of non-fuel minerals, many of crucial industrial
importance, such as manganese, cobalt, and chromium. For all these substances
import dependence is 90 per cent or over and the suppliers are all overseas.4
Understandably, American navalists make much of these figures but the real
situation is rather more subtle. The United States itself is well supplied with
deposits of all minerals but chooses not to mine many of them because cheaper
supplies are available elsewhere. US vulnerability to a mineral cut-off is often
overstated. As one attempt to put the other side of the picture has put it: ‘A
combination of “safe” production, substitution, domestic production,
conservation and recycling could readily compensate for a loss of imports. The
few minerals for which this is not the case are non essential.’5 Moreover, the
USA maintain significant stockpiles of critical materials, enough to deal with
short- to medium-term disruptions.
Neither superpower is therefore a small island critically dependent
economically on the use of the sea as the British Empire was at the height of its
power. Both are essentially continental superpowers. The United States is linked
into a complex, international, and interdependent capitalist economic system
where it chooses to buy in the cheapest markets rather than face the cost in living
standards and prosperity dictated by a self-imposed blockade. The USSR
operates as the centre of an alternative economic world, but one increasingly
conscious of its need to interact with the rest of the globe economically and
politically. Both, therefore, rely on the sea as a matter of economic well-being
rather than survival. The one’s use of the sea is in this sense no more or less
‘legitimate’ than the other’s.
It must be stated, however, that in any kind of protracted conventional
conflict the United States would require substantial imports of raw materials
from overseas to exert its maximum strength. Moreover, the USA has
economically related reasons to deploy more military force at sea than its Soviet
adversary partner; the OECD world as a whole is more dependent on sea
communications than CMEA (Comecon). The former contains, as we have seen,
a highly sea-dependent island medium power and an economic giant, weakened
in military terms. Part of the bargain that the allies of the United States make
with their patron includes an American commitment to maintain freedom of
navigation by force if need be through the ‘maritime superiority’ of which
President Reagan spoke above. This ‘bargain’, however, is far from being a
straightforward one. Legally, the United States cannot even protect directly that
majority of US-owned ships that does not fly the stars and stripes. Moreover, the
diminishing relative economic power of the USA vis-à-vis the rest of the
Western world argues for some changes in the relative share in the provision of
defence for the West’s maritime trade. Those states that are more sea dependent
may well have to do more for themselves, either individually or collectively.
Two tables of the relative economic interest that states have in using the sea
have been prepared, one by Alistair Couper and his Cardiff University team for
the Times Atlas of the Sea, and the other by Admiral Richard Hill for his
excellent study Maritime Strategy for Medium Powers. The Cardiff University
team were only concerned with interest in seaborne trade and took three criteria:
(1) total seaborne trade import and export, (2) seaborne trade as a percentage of
GNP, and (3) size of merchant fleet. Different indices were prepared for
parameter (3), one based on beneficial ownership, the other on flag of registry.
On the former criteria the indices were as follows. The data upon which
calculations were based were for 1980.6
Interest 9– Japan
index 10
8– Saudi Arabia
9
7– Kuwait, The Netherlands, USA, United Kingdom, Singapore
8
6– Italy, Norway, Libya, Iran, Greece, France
7
5– Venezuela, Spain, USSR
6
4– West Germany, Australia, Algeria, Qatar, Canada, Denmark,
5 Brazil, Sweden
3– Belgium, Nigeria, India
4
2– Poland, Liberia
3
1– Tunisia, Gabon
2
0– Lebanon, Syria, Panama, Trinidad and Tobago
1

Adjusting parameter (3) for flag registry produced some interesting changes.

Interest 9– Liberia, Japan


index 10
8– Saudi Arabia
9
7– Kuwait, The Netherlands, United Kingdom, Singapore
8
6– Italy, Norway, Libya, Iran, USA, France
7
5– Greece, Venezuela, Spain, USSR
6
4– Australia, Algeria, Qatar, Denmark, West Germany, Brazil,
5 Sweden, Canada
3– Nigeria, India, Panama
4
2– Poland
3
1– Tunisia, Gabon
2
0– Lebanon, Syria, Trinidad and Tobago
1

The Hill indices brought in both the question of dependence by relating


degree of interest to GDP and population, and broadened the scope of the study
to cover shipbuilding, fish catches, and off-shore zones. The data is now a
decade out of date but the general results probably still have some validity. On
the basis of GDP, Norway emerged as the most sea dependent nation by a
substantial margin, followed by Denmark and Argentina and then Japan, Greece,
and Brazil. Next came Chile, The Netherlands, and Spain leading a large group
of Australia, Egypt, Finland, Iran, Italy, Nigeria, Sweden, and the United
Kingdom, all at about the same level. Canada and India led the next group with
Belgium, France, Mexico, New Zealand, South Africa, and Venezuela all about
on a par with the USA. Finally, there came West Germany and Israel.7
When dependence on maritime economic activities was compared with
population the results were a little different. Norway still led the table, but
Denmark and Japan were closer behind than they were on the basis of GDP, as
were The Netherlands, Spain, Sweden and the United Kingdom, and then
Canada, Australia and Finland. France then led the United States and Belgium
followed by Italy, Argentina, Greece, Chile, Venezuela, South Africa, West
Germany, and New Zealand. Then came Mexico with a gap before Brazil which
led India, Iran and Israel, and finally Nigeria and Egypt. The spread of degrees
of dependence seemed to be larger on a basis of population than it was on that of
GDP.8
Admiral Hill sums up his ‘unspectacular but useful’ results as follows:

The most consistently sea dependent nations are either islands or states with
long coastlines and limited land frontiers. Of those in the tables Japan, the
Scandinavian countries and the United Kingdom stand out; the pattern is
repeated for smaller states like Singapore, and probably would be for a state
like Cuba if its figures were available. Other nations may be less consistently
sea dependent but still have a very marked dependence on one aspect or other
of the sea affair. Liberia on merchant shipping, Saudi Arabia and Kuwait on
shipborne trade, Iceland on fishing, Gabon on the offshore zone; for all these
dependence is extremely marked and . . . their importance to the world
economic and strategic situation makes their vulnerability of some
significance.9

Certainly almost all the states mentioned above would seem to have some
interest in deploying maritime forces of some kind, although there is no direct
and necessary connection between the extent of dependence and the degree of
capability of those forces. Open registry states do not, on the whole, feel a
responsibility to defend their merchantmen. One can, however, protect
shipbuilding by ordering warships; and offshore zones always need policing. The
latter factor has probably fostered more naval improvement than any other in
recent years. There is thus still a connection between economic dependence and
interest and military naval capability, although the reach of a maritime nation’s
maritime forces reflects both a country’s absolute level of economic power and
its perception of its political role more than it does its degree of sea dependence.
In some ways sea dependence is steadily increasing. The decline of domestic
heavy industries in Europe and the USA has led to a remarkable degree of import
dependence for certain items, e.g. motor vehicles. Most important of all,
however, is the dependence on imported energy of much of the developed world.
In the late nineteenth and early twentieth century the realization that Britain was
dependent on overseas supplies of food acted as a potent spur to naval
expansion. Now Europe has more food than it can eat. Nevertheless, import
dependence has only been pushed back one remove: modern farming techniques
are heavily dependent on imports of energy.
The most important single component of maritime trade is, as we have seen in
Chapter 2, crude oil and petroleum products. H.W. Maull in his major study,
estimated that ‘the industrialised countries will remain dependent on oil imports,
although in terms of absolute quantities there will be little, if any, increases and
perhaps even a reduction’.10 Indeed, import dependence may well actually
increase, with the Middle East asserting a more important position as the main
source.11 The oil nexus may well draw the producing states into increased naval
roles just as it will emphasize the need for sea control with long reach in the
hands of the importers. It is thus hardly surprising the the Gulf has already been
the context for the most significant innovations in naval operations in defence of
shipping, i.e. re-flagging and, more importantly, European co-operation.
In the longer term, into the next century, the oil trade may go into absolute
decline as reserves are depleted and the prospect of more expensive oil
stimulates synthetic production. Alternatives, like natural gas and coal, will
become more important. The former sometimes comes from the sea, but the
main producer is the USSR with its large underground fields. Gas also lends
itself to long pipeline transit, both overland and across short stretches of the
seabed (e.g. from Algeria to Western Europe). Japan will, however, continue to
rely on sea transport for its natural gas supplies.12
Coal may well become an important cargo. It is currently shipped in large
quantities by sea and will be transported in even greater quantities over longer
distances in ships from the United States, Africa, China and other East Asian
states to countries, like those of Western Europe, less well endowed with cheap,
easily exploited seams. The next century may well be the age of the ‘super
collier’.
As oil becomes more expensive, and the exploitation of undersea oilfields
becomes more attractive once more, the scope for maritime conflict over access
to the seabed may increase. This could be especially true in Asian waters, where
there remain serious disputes over the ownership of various archipelagos sitting
above undersea oilfields. ‘Very few islands in the South China Sea are
uncontested.’13 Both China and Vietnam claim the Paracel (or Hsi-Sha) Islands,
occupied forcefully by the former in 1974, Taiwan and the Philippines claim an
interest in these islands also.
The Paracels are also significant in terms of fishing. Demand for fish is
expected to rise to between 113-125 million tonnes by the end of the century. As
the ‘sustainable harvest’ is estimated to be no more than 100 million tonnes the
excess demand for fish may well fuel disputes over management of stocks and
access to fishing grounds.14 Where exclusive economic zones are not currently
properly policed there will be increased efforts to maintain rights to control and
license exploitation of national resources. Asian waters (including the Indian
Ocean) will once again be a focus for this activity.
Economic pressures may well thus militate in favour of disorder at sea, and
an enhanced role for naval forces in both their military and law enforcement
roles. Nevertheless, one need not be too pessimistic. Both in terms of oil
exploration and fishing an agreed ‘pax’ is necessary before the seas can be
economically exploited. Oil men in particular are easily scared away by their
inherently expensive and dangerous activities at sea being thrown into even
more jeopardy by human opposition. This will act as a potent factor to achieve
peaceful settlements, albeit settlements that will have to be backed up by
continuous patrols by maritime law enforcement forces.
One maritime industry that has not yet been mentioned as a stimulus to naval
power is shipbuilding. This industry has naturally suffered from the supply of
shipping exceeding demand and has been in heavy recession, although signs of
an upturn were noted in 1987. The most important shipbuilders, accounting for
over three-quarters of world output, are Japan and South Korea, virtually first
equal with about a quarter of world tonnage each, with the other quarter divided
between Yugoslavia, Italy, Brazil, Poland, China, Romania, and Taiwan.
Subsidies help to maintain Italy’s shipbuilding industries, and Poland had been
sustained by orders from the USSR in the non-capitalist CMEA framework: few
successful shipbuilders today operate a fully commercial environment.15 This
means that noncommercial shipping, i.e. warships and support vessels, are often
the key to the survival of any significant shipbuilding industry in certain
countries, notably in the UK and the USA, which had no major merchant ship
under construction in domestic yards at the beginning of 1988. Shipbuilding is
secure only by becoming part of the defence industries that such states sustain
for security reasons. The specialist skills required are still assets that give a
competitive edge to states uncompetitive in the construction of simple
merchantmen. Even free market governments can feel happy about using naval
shipbuilding as a form of economic relief, as long as there is an overall national
security interest in obtaining the vessel in question. Thus can economic factors
become a factor in sustaining naval strength.
As the sea gets more important in various ways this may have an effect on the
internal political debate in various states that will make governments more sea
minded and therefore more likely to fund maritime forces. There are, however,
countervailing factors. The traditional industrial nations (and sea powers) seem
to be moving away from production and transport. As Admiral Sir James Eberle
pointed out to the Sealink Conference in 1986: ‘International financial flows,
rather than trade flows, now form the dominant part of international activity.
Despite increases in the volume of international trade and its importance to
manufacturing and industry only some 2 per cent of international money
transactions now relate to payments for international trade.’16 Electronic
connections are the arteries of much modern commerce, not ships. If financial
connections are the important factors then ownership of fleets and the
wherewithal to protect them may not seem important any more. As Sir James
Cable has pointed out with reference to the United Kingdom, ‘the complacent
regard merchant shipping as a primary industry that can safely be abandoned to
the Third World as long as the vessels are still insured at Lloyds’.17 One can
sympathize up to a point with the nationalistic criticism implied by the above
statement. Certainly, an increasingly internationalized world shipping industry is
coexisting increasingly uneasily with the established naval and legal maritime
order based around national flags and national navies. As Sir James Eberle
succinctly put it to the delegates in Annapolis: ‘If you sink a ship owned in
Britain, financed in the Europe market, registered in the Bahamas, crewed from
Hong Kong, insured in London, carrying Japanese cars whose engines were
assembled in Germany, whose interests are you mainly damaging?’18 The
answer is many peoples’, most of whom will expect someone, but not
necessarily themselves, to sort the situation out.
The world in general, and the seas in particular, have never been so
internationalized, a reflection of a wider economic phenomenon: ‘Modern
industrial societies, and indeed the present world economy, are built on reliance
and interdependence, on free flows of goods, capital and technology and thus
generally on the exploitation of benefits which can be reaped from the
international division of labour.’19 As this international division of labour
becomes ever more dispersed, however, the provision of naval protection for its
maritime dimensions is going to have to be reconsidered. Warships flying one
flag are going to have to become much more flexible in their attitudes to
protecting shipping flying another. This can be done by national decision; both
France and the USA announced that it would defend all Gulf shipping regardless
of flag in January 1988;20 but perhaps a resurrection of the UN political
framework for such activities might help these politico-legal problems to be
fully solved on a more permanent basis. Sadly, however, the decline of national
maritime industries is likely to lead to a widening of a phenomenon noted by
Richard Hill in his typically pungent style: it is a ‘malign neglect that can most
kindly be characterised as sea blindness’.21 This could have serious negative
effects on naval capabilities, especially as warships are not getting any cheaper.
This brings us to economic constraints on naval capabilities. No navy is free
of such problems, least of all the world’s largest, the United States Navy.
Although in some ways the massive budgetary deficits run up by the Reagan
Administration in the 1980s have been a novel and creative way of mobilizing
Japanese and Western European financial resources to pay for US naval power,
the consensus view both nationally in the United States and internationally in the
world financial community is that ‘the deficit’ should be cut. Of the two
influences, domestic US opinion is by far the most important. ‘Americans are
convinced that federal spending generally – and defence spending specifically –
must be brought under control.’22 The result of this conviction, the Gramm-
Rudman-Hollings Act of 1985 has created a new environment in Washington,
‘forcing the hard questions’ on defence spending.23
In a context where a 3 per cent increase in real terms was required to maintain
the 600 ship/15 carrier battle group fleet that emerged from the John Lehman
era, funds began to be cut in real terms from after 1985. In 1987 one sympathetic
observer expressed ‘great concern that Gramm Rudman and its related trends
will make it impossible for the Navy to meet all of its requirements, especially as
the competition intensifies with the other services for increasingly scarce
funds’.24 Sure enough, February 1988 saw the 600-ship navy target falter, with
the decision to cut the American frigate force and even consideration being
given to the possible paying off of the three oldest carriers.25
It will have been noted above that increasing expenditure was required to
maintain US naval strength. ‘Level funding’ does not solve a navy’s financial
problems, as the current British Royal Navy finds. Naval costs have an
inexorable tendency to spiral upwards under inflationary pressures all of their
own. These have been analysed most recently by Philip Pugh.26 He points out
that increasing capability and increasing cost are an inevitable result of trying to
keep up in the military competition by increasing capabilities to match those of
an opponent. Naval power, depending on expensive platforms like ships and
aircraft, is especially vulnerable to such cost escalation. He also demonstrates
that ever since industrialization unleashed the prometheus of technological
development, naval budgets have been tending towards the ‘absurdity’ of only
being able to afford one ship. The British naval budget from 1860 to 1885 and
the American naval budget from 1960 to 1985 both display this frightening
characteristic. If the lines of unit costs and new construction budgets are
extrapolated for two decades or so they cross. This absurdity, however, was
avoided through a mix of three measures: increasing the budget thanks to
economic growth, a reduction in numbers of units, and alterations of structure
towards the procurement of smaller and cheaper vessels.
Pugh bases his case on the post-1950 Royal Navy, a fleet specially afflicted
by the conflicting requirements of force enhancement and inadequate resources.
Yet it seems that other navies less prone to pressures to live beyond their means
have responded to the same stimuli in similar ways. ‘Escalating costs,’ Pugh
argues, ‘have demanded change towards smaller navies equipped with smaller
ships; but that has not of itself meant any change in the balance of naval power.
Alterations in that follow from changes in the balance of defence budgets and,
hence, from differences in the economic performances of nations.’27
Optimizing the naval budget, however, as Pugh goes on to point out, is a
tricky business. The natural tendency to try to exploit new technology to the full
leads to ever higher development costs and another potential reductio ad
absurdum, all money going on development and none on actual procurement.
Spacing out development programmes and updating existing weapons limit
development costs in their different ways, especially for missiles and aircraft that
incur especially high development costs. It is indeed surprising how long lived
basic missile systems are in an age of rapid technological progress. In some
cases their capabilities have been improved, sometimes transformed, by
relatively small changes to the basic airframe.
Another balance that has to be struck is that between individual ship
capabilities and the quantity of units required. This dilemma operates both
within the same class of ships and between different classes and sizes of vessel.
This is a most difficult balance to achieve, especially as smaller vessels may not
necessarily be the more cost effective. Smaller vessels may indeed by less
efficient, given the higher relative costs of the hull relative to the weapons
systems carried. A 6,000 ton combatant devotes about 12 per cent of its cost to
floating, 14 per cent to moving, and 74 per cent to fighting. A 750-ton vessel
devotes 13 per cent to floating, 29 per cent to moving, and only 58 per cent to
fighting.28 Thus ‘restraining the size of warships is not a wise policy unless there
are overriding reasons for doing so. ... Reasons for setting a lower limit to the
number of ships are: a recognition that no vessel, however good, can be in two
places at once or, else, anxiety to ensure that accident or misjudgment cannot
deprive a fleet of too great a proportion of its strength.’ Pugh argues
convincingly that it is ‘fundamentally mistaken’ to make

small size and/or low unit cost into a primary goal of procurement policy ... if
halting the rise in unit cost is made the first priority then costs will cease to
rise. But the penalty of such an achievement will be a disproportionate loss in
capability and hence worse value for money spent on the fleet as its
formation and the specification of its ships drifts further from their optimum
... the rise in unit costs is not inevitable; but it is unavoidable if limited
budgets are to be spent rationally and to the best effect. Sailing with this tide
has obtained the best value for money that was to be had. If, as a result,
cherished roles and missions have had to be given up then the harsh logic of
the situation is that attempts to retain them could only have done greater
injury to the overall capabilities of the navy in question.29

Navies cannot just keep their existing vessels forever, given both the fact that
machinery wears out and manpower is an ever more expensive resource. The
latter factor may even prevent continued procurement of new examples of the
same designs. Navies usually lack men more than they lack ships or aircraft.30
Trained manpower of sufficient quality for modern naval operations is a scarce
commodity the world over, and its price tends to increase at a higher rate than
other commodities. There is thus a powerful incentive, as we have seen in
Chapter 6, to reduce the ships’ companies of modern warships. Modern vessels
may be required to prevent units being laid up for lack of crews. Navies have to
run even to stand still.
Although the quantities of resources allocated to naval defence by the states
of the world will not increase as quickly as their naval staffs would like, it seems
unlikely that there will be a substantial dismantling of naval capabilities in a
world that will still rely to a considerable, and quite possibly increasing, extent
on the various uses of the sea. ‘Getting the best value for money ... to be had’
will remain a fundamental of naval policy making. In this context Pugh argues
that, ‘continuation of present trends will eventually force all but the superpowers
into essentially regional capabilities and, finally, only the economically strongest
of these will be left as the one naval power with global reach. However, . . . these
developments will, at present rates, take many decades and almost a century,
respectively.’ However, he also concedes that ‘both political and economic
circumstances might change before then’.31 They may in fact be changing now.
As we saw in Chapter 7, the economies of the superpowers are declining
relatively to other states. As we have seen, the most important naval power in the
world is considering swingeing cuts in capability. As the superpowers decline in
relative terms the lesser states will inevitably rise in relation to the big two. This
may well make it easier for them to deploy power at a distance. This, again, as
we said at the outset of this section, is the economic foundation of the future
promised multipolarity. Sea power is not about to price itself out of the market,
indeed it looks like becoming more widely dispersed. As Pugh concluded, ‘a
broad spectrum of navies will long continue to exist with each being of a size
and strength commensurate with the ability of their national economies to pay
for them’.32 These navies will be of considerable use to their owners in both
peace and, if deterrence at various levels fails, war.
Part V

Navies in peace and war


Chapter 9

Navies in peace

Although navies are built primarily for war they find their main utility in peace,
in deterring the outbreak of major and minor conflicts, in exerting influence and
pressure as part of normal ‘peaceful’ diplomatic activities, and in enforcing the
remarkable developed corpus of international law and regulation that prevents
the maritime environment dissolving into chaos. Indeed, it might be argued that
the world’s navies are nowadays primarily configured for peace rather than war.
Ballistic missile submarines pose formidable deterrent threats that if carried out
could bring nothing but disaster on their owners. Even the largest navy in the
world, that of the United States, concentrates its resources on forces that make
more sense in a ‘power projection’ role in peace or limited war than in the less
likely conditions of a major conventional war with the Soviet Union. The new
American Maritime Strategy is an attempt to deal with the conundrum of how to
use such a peacetime navy to achieve the immediate wartime objectives of
safeguarding sea use.
If there is a tension between peacetime and wartime requirements in the
configuration of military navies, the differing priorities within the peacetime
roles themselves create extra complexities still for states deciding the type of
maritime capability they ought to deploy. There can be little doubt that the
constabulary and regulatory naval roles will grow steadily in importance as the
forms of sea use become more diversified, and as states become more aware of
their rights and duties in the various offshore zones. This diversity of naval
constabulary duties, defined by Richard Hill as information gathering, the giving
of instructions and orders, inspection, detention and disaster control, requires a
sufficient set of skills for it to be considered to be a specialized task in its own
right.1 The United States, as is well known, has a totally separate naval
organization for this role, the US Coast Guard, and this pattern has been
followed by other countries, e.g. Argentina, Canada, Japan, Taiwan, and India.
The creation of the 200 mile Exclusive Economic Zone acted as a potent
stimulus for the creation of such specialized forces in the 1970s. Very small
countries, e.g. most Caribbean states, only bother with the maintenance of a
‘coast guard’ or self-consciously styled maritime police force rather than a fully-
fledged ‘military’ navy. Even those countries which tend to give the ‘navy’ the
main constabulary role may have small additional ‘navies’ of various types, e.g.
in the United Kingdom the Department of Agriculture and Fisheries for Scotland
operates a small flotilla of ships and aircraft on fishery protection duties, and
there are almost a dozen small craft operated by the Customs and Excise Marine
Division. Other non-official organizations, e.g. The Royal National Lifeboat
Institution, have an important part to play in maritime rescue. Although small
specialist organizations may have advantages in concentrating on one task and
doing it very well, too many groups can create problems of co-ordination and
control.
The particular bureaucratic framework of maritime law and order will
probably remain a function of local tradition and political culture, as well as the
resources available. Probably greater coordination will be necessary in those
states that do not have ‘umbrella’ coastguard or ‘maritime safety’ agencies. It is
likely, however, that the competition for resources in most states between the
‘constabulary’ role of navies and the more ‘military’ role will grow more intense.
The characteristics of the police vessel or aircraft are not the same as those of
equipment optimized for higher level combat. Navies are increasingly going to
be caught on the horns of a dilemma. Hiving off less attractive naval roles to
other services might well mean a loss of resources in the short term or, as time
goes by, a loss of national relevance. If the navy retains a comprehensive set of
constabulary duties, however, it may end up, like the Mexican Navy for
example, looking and acting like a coastguard anyway.
Even if a separate coastguard is set up the line of demarcation may not be
clear. Canada provides an interesting example of this, a case study with
particular piquancy in that the main ally in the military area is a major adversary
in the other. The Canadian Navy is primarily oriented to contributing ASW
surface escort groups to American sea control forces in both the Atlantic and
Pacific (but largely the former). Given the relatively low levels of defence
provision in Canada in recent years, this has not been an easy task but currently,
despite design delays, the Canadians have an impressive programme of naval
modernization under way. In 1985, however, the Americans sailed a Coast Guard
icebreaker through the Northwest Passage on a ‘freedom of the seas’ operation,
similar in intention, though somewhat less dramatic, than those carried out about
the same time in Libyan waters. This led to such a public outcry that the
Canadian Prime Minister had to announce planning for greater Canadian ‘naval
activity’ in the Arctic to safeguard national sovereignty. The eventual proposal to
purchase nuclear-powered submarines was a direct result of these sensitivities as
the threat to Canadian sovereignty had subsurface and military overtones, but
there have been controversies over the interface between constabulary and
military roles, especially over the ownership, equipment, and role of the large,
new Polar 8 icebreaker, the most powerful in the world, currently earmarked for
the Coast Guard. It was clear that the Canadian Armed Forces had serious
reservations about being forced to take this ship with its significant manpower
requirement and repertoire of ‘civilian’ roles.2
As the use of the sea grows, even in inhospitable northern and southern
latitudes, this kind of choice will become more and more common. Navies may
well have unwelcome and unfamiliar roles thrust upon them. To an extent this
has already happened at the other end of the scale of peacetime naval roles, the
provision of ballistic missile submarines for strategic deterrence. Perhaps only in
the Soviet Union, where alternative naval traditions were weak, was this role
wholeheartedly welcomed by the naval leadership. One can detect even in the
United States a distinct prejudice against the SSBN role, marked by its relative
low priority in the Lehman naval building programmes, and the development of
alternative forms of nuclear delivery, cruise missiles, more easily integrated with
the navy ‘proper’. In Britain, where the potential opportunity costs were even
greater, the anti-SSBN feeling is commensurately deeper. There was
comparatively little naval desire to acquire Polaris, and there have been distinct
signs of anxiety over the effects of the Trident programme on more ‘traditional’
general purpose naval capabilities. There have been persistent demands for some
kind of separate funding arrangements for the SSBN force, as was achieved in
the 1960s and 1970s.3
That admirals and their supporters should argue this is as understandable as
that their political and financial masters should argue the opposite. For SSBNs
are some of the largest and most imposing assets a modern navy can deploy.
Their existence reflects two fundamental facts about the modern situation: first,
that war between the major powers is so destructive as to be best avoided, and,
second, that in an age of continental superpowers sea power can only affect the
core interests of the major actors by being the least vulnerable platform for the
catastrophic application of air power against their heartlands. They are thus the
most important part of the ‘navy’ in the eyes of political leaders, and perhaps
deserve to be taken a little more to the hearts of their operating institutions.
There are, however, major disadvantages to deploying ballistic missile firing
submarines. The opportunity costs in other naval capabilities are usually
significant and the special nature of the SSBN makes it useless for other roles.
The majority of officers in the operating navies, therefore, are primarily
concerned with other ways of exerting force at sea, and usually put a higher
priority on these than on a high quality strategic nuclear deterrent. It is arguable
whether countries with close alliances to superpowers need deploy ‘nuclear
deterrents’ at all, or if they do whether they need to be much more than a few
nuclear-tipped cruise missiles mounted in submarines that would be forward
deployed anyway and which could act in desperate circumstances as a trigger for
Armageddon. It would be very difficult for a Soviet leadership to distinguish a
200 kiloton nuclear burst over a major Soviet fleet base, for example, as coming
from a US or a British submarine. Such nuclear forces would also be useful in
deterring potential Third World nuclear powers, a factor raised at times by some
supporters of medium power SSBN forces.
All nations with navies are much more interested in threats and applications
of force at a much lower level of intensity. In some cases the connection between
the use of force at the lower level to much higher potential levels is taken for
granted. In others the dangers of escalation are much less. It is in the latter cases
that violence will more probably be used, although the international laws of self-
defence and proportionality limit too easy a recourse to armed violence (if not
violent physical contact between ships).
Certain navies and coast guards, like the Soviet and Argentine, have shown a
disconcerting tendency to ignore such rules in enforcing their sovereign and/or
economic rights in off-shore zones subject to their claims. It is hard to gauge
whether this will become more the international custom or not. The degree of
restraint shown in the Anglo-Icelandic contenders during their fisheries dispute
was remarkable given the bitterness and intensity of the conflict. Possibly,
however, as lines of demarcation become more marked at sea and the legitimacy
of national claims over it become more accepted, the confidence shown by the
‘maintainers of law and order’ on it will increase. The degree to which this will
include easy resource to arms will then depend on the national culture of the
state involved: this could mean more shooting rather than less.
It is not only rogue fishermen and smugglers of various kinds who are
potential targets for this hostile fire. Ships are still subject to violent robbery,
especially in areas where ‘trading patterns bring considerable numbers of
merchant ships into contact with the developing world where there are large
numbers of people still living at low subsistence level’.4 Many of these piratical
raids occur in territorial waters where there is currently neither the capability, nor
sometimes the will, to take serious counteraction. There can be little doubt that
the growing divorce of mercantile shipping from military shipping has been a
prime factor behind this violence, which has shown some disturbing trends in
recent years. Established naval powers are nowadays rightly reluctant to act in
waters claimed by other nations. When a local actor finally desires to take a
strong line, however, the effect can be dramatic. Nigeria was able by combined
naval and police action to reduce to negligible proportions by 1984 the number
of piratical acts which had been running at eight a month in the worst period of
the preceding years. Given the likelihood of ships remaining the interface
between the developed and developing worlds and the inherent vulnerability of
vessels alone at sea to armed attack, a constant effort will be required by states to
fulfil their maritime law and order responsibilities. Sometimes, in situations of
crisis, as with the massacre of the Vietnamese ‘boat people’ in the 1980s, firm
international action may be required.5
‘Piracy’ proper is the use of violence for private gain. When the aim is
political it gains legitimacy, although not to those against whom it is directed
who tend to stigmatize its lower level acts as ‘terrorism’. Happily, there has been
more fear of ‘terrorist’ activity at sea, e.g. against oil rigs, than actual ‘outrages’.
Guerilla warfare at sea requires rather more infrastructure than that ashore. Ships
have been hijacked, notably the Achille Lauro, but, as Professor Wilkinson has
pointed out:

There are a number of reasons why ship hijacking has not caught on. It is far
easier to attack targets ashore, while the impact on the mass media is far
greater on land. . . . For those who seek more spectacular targets, there are
still enough loopholes in airport security to make aircraft hijacking possible.
If aviation becomes more difficult, terrorists will be driven to other forms of
hostage taking on land. Most terrorists are landlubbers and have no taste for
the storms and other rigours of the high seas. There are two other key tactical
considerations that would deter the average small terrorist group. It is very
difficult for a tiny group of hijackers to take and keep control of a major ship
carrying a large crew and perhaps hundreds of passengers . . . and it is in
principle far easier for the authorities to mount a counter attack and regain
control of the vessel.6

The more organized guerilla groups have attempted maritime activities, notably
the PLO, which has attempted raids with vessels from freighter size downwards,
but it usually requires state backing to acquire the resources to engage in
maritime warfare, as when the Libyan freighter Ghat unloaded mines in the Red
Sea on behalf of the Islamic Jihad in 1983. Some eighteen ships were damaged.
The international naval operation designed to neutralize these weapons is but
one example of the use of naval forces in counter-terrorist and other low-level
operations. The growth of the Israeli Navy has received considerable
encouragement from the maritime PLO threat and the alternative to bombing
that naval power provides for relaliatory actions. The US Navy has made
something of a name for itself in the 1980s as a ‘counter-terrorist’ force. When
US forces were subject to ‘terrorist’ attack in their ‘peacekeeping’ mission in
Beirut, the battleship New Jersey was called upon to wreak vengeance with her
16 inch guns. The use of the spectacular battleship had the effect of
demonstrating to public opinion back home that ‘something was being done’.
There was also reduced danger of suffering losses, a problem demonstrated
when naval aircraft suffered losses while engaged in similar missions against the
Syrians. The actual effect of the dubiously accurate bombardment was less
certain, and probably less important.
The bombing of Libya by a combined USAF and carrier air strike in 1986
was another example of naval forces being used for ‘counter-terrorist’
bombardment. More discriminating was the use of naval interceptors to capture
the Achille Lauro hijackers over the Mediterranean.7 Generally, if the main target
is the actual ‘terrorist’ rather than the opinion back home, it is the less
spectacular riposte that is the most effective deterrent. One suspects that the
training of maritime anti-terrorist squads able to deal directly with active
maritime terrorism acts as a greater deterrent than the possible attacks of an
American carrier battle group or battleship on one’s backers. With the drawing
up in 1988 of international conventions to prevent terrorist attacks on ships on
the high seas and fixed maritime platforms, there might even be some prospect
of joint international action to deal with threats that are increasingly being
perceived as universal and not part of legitimate great power competition.
Naval bombardment of weaker states that cannot hit back effectively is a
classic form of gunboat diplomacy. The latter, as we have already discussed in
Chapter 7, shows little sign of going away. It used to be fashionable to argue that
its days were numbered with the spread of relatively well organized sovereign
states and the acquisition by them of increasingly powerful defensive weapons.
Yet the record of the 1980s clearly demonstrates that if the margin of
technological superiority is sufficiently great, a first class navy should have little
difficulty in ‘punishing’ even quite a superficially strong littoral state. The
Iranian Navy was consistently overwhelmed in surface action by American ships
and aircraft. Events off Libya show clearly that an American carrier battle group
has relatively little to fear from a few isolated missile boats or badly handled
aircraft. It can operate almost at will.8 It will take some considerable time for the
margin of superiority of the US Navy at least to be eroded sufficiently for it not
to have considerable scope for activities ranging from simple bombardments to
amphibious landings like the invasion of Grenada in October 1983. Indeed, one
of the primary roles of the US Marine Corps remains, and is likely to remain, the
maintenance of US interests, by force if necessary, in the traditional US sphere
of influence in the Caribbean and Central America. The only weapons that have
seriously discomfited the Americans have been Iran’s mines, but this threat was
soon overcome and the lesson of the need to improve MCM capabilities has
been clearly learned, for the time being at least.
As the US Navy is currently the major exponent of naval diplomacy, and is
likely to remain so for the foreseeable future, perhaps it is worthwhile to look at
its own definition of the range of activities involved. They were succinctly
described by Admiral Watkins in early 1986 thus: ‘Maintaining presence,
conducting surveillance, threatening use of force, conducting naval gunfire or air
strikes, landing Marines, evacuating civilians, establishing a blockade or
quarantine, and preventing interference by Soviet or other forces.’9 Admiral
Chernavin would probably only have altered one word, although the power of
the Soviet Navy to engage in active action is significantly more restricted than
the US Navy’s. Limited carrier assets act as a potent constraint on bombardment
potential. The Soviets are slowly improving their carrier air power but its
potential is going to be very limited for some considerable time. The
concentration of the Soviet Navy on operations against other naval forces does,
however, put a very potent, perhaps potentially the most important constraint, on
the activities of the US Navy. The most dramatic demonstration of this has
probably been the Soviet anti-carrier warfare activities during the 1973 Arab-
Israeli War, when the potential price of direct US involvement was demonstrated
only too clearly.10
The Soviets on the whole have used their maritime forces sparingly in more
positive ways in their own diplomatic interests, transporting friendly troops to
reinforce client forces on land, protecting their merchant ships in danger zones,
demonstrating the USSR’s concern when her ships have been seized and even
carrying out the odd limited bombardment in support of her friends.
Nevertheless, despite the accepted wisdom of most commentators, the Soviet
Navy has been less ‘operational’ than those of the West for maritime
intervention.11 It is significantly limited by the small size of its amphibious
component. Possibly the traditional bureaucratic dominance of the ground forces
has prevented the naval infantry replacing the airborne forces as the Soviet
Union’s main means of military intervention. When the Soviet Union does feel
impelled to land forces rapidly the target may be far from the sea, e.g.
Afghanistan. The USSR is thus more willing to face the higher risks inherent in
airborne operations compared to seaborne landings.
As we discussed in Chapter 7, although the superpowers will remain the main
practitioners of peacetime naval diplomacy in the short to medium term, other
nations will also use their navies as diplomatic instruments. Both Sir James
Cable and Ken Booth have produced comprehensive analyses of the peacetime
uses of naval forces.12 Booth offers three headings for the aims and objectives of
the ‘diplomatic role’ of navies: (1) Negotiation from strength, (2) Manipulation,
and (3) Prestige. The first and last are more or less self-explanatory but the
second heading requires closer definition: ‘changing (however incrementally)
the political calculations of relevant observers. . . . Effects may be induced by
negative as well as positive actions’. Naval forces might be disposed to
demonstrate support to friends, ‘gain or increase access to new countries’, create
‘a degree of naval dependency’, or establish the ‘right to be interested’ in a
distant area.13 All of the above imply no actual action being taken. Booth lumps
all uses of force under the ‘military role’.
Cable puts the distinction in a different place. He defines ‘gunboat diplomacy’
as ‘the use or threat of limited naval force other than as an act of war’. The
distinction from ‘showing the flag’ is clearly made in a recent article:

The expression ‘showing the flag’ is often loosely employed as a synonym


for gunboat diplomacy. That practice involves the use or threat of limited
naval force for some specific purpose. Even the ‘expressive’ variety, when
warships are manoeuvred merely to ventilate emotion, is recognisably related
to a particular dispute. ‘Showing the flag’ is a more general reminder to
foreigners of the existence of the navy concerned. Any threat of force is
seldom more than remotely explicit.14

To Cable, therefore, the existence of the specific dispute is the central defining
factor, both distinguishing between gunboat diplomacy and generalized flag
showing at the one extreme and specific uses of limited force, and ‘war’ at the
other. Cable defines the distinction between gunboat diplomacy and war as
follows:

An act of coercive diplomacy ... is intended to obtain some specific


advantage from another state and forfeits its diplomatic character if it either
contemplates the infliction of injury unrelated to obtaining the advantage or
results in the victim attempting the infliction of injury after the original
objective has been either achieved or abandoned. Coercive diplomacy is thus
an alternative to war and, if it leads to war, we must not only hold that it has
failed: we may even doubt whether it ever deserved the name.15

In 1982 General Galtieri tried coercive diplomacy: the British replied with a
limited war.
Booth’s distinction seems to centre around means, Cable’s around ends,
although the latter does slightly muddy the definitial waters when he argues
additionally that ‘showing the flag becomes gunboat diplomacy when the risk of
encountering armed resistance is deliberately run’.16 The implication must be
one only runs the risk of resistance when one has a strong specific motive for
doing so.
Cable divided his ‘gunboat diplomacy’ into four kinds:

1. Definitive – when a fait accompli is created;


2. Purposeful – to induce ‘someone else to take a decision which would
otherwise not have been taken: to do something or to stop doing it or to
refrain from a contemplated course of action’;17
3. Catalytic – ‘a situation arises pregnant with a formless menace or offering
obscure opportunities. Something it is felt is going to happen, which might
otherwise be prevented if force were available at the critical point.
Advantages, their nature and manner of their achievements still
undetermined, might be reaped by those able to put immediate power
behind their sickle. These are situations peculiarly favourable to the
exercise of limited naval force’;18
4. Expressive – ‘to express attitudes, to lend verisimilitude to otherwise
unconvincing statements or to provide outlets for emotion’.19

Cable then goes on to draw up an ascending list of levels of operation: simple


ship, superior ship, simple fleet, superior fleet, simple amphibious, and superior
amphibious.20 As one moves up the levels of violence the numbers of fleets able
to carry out the required operations diminishes very significantly, at least when
one moves over any distance. When one moves even to ‘simple fleet’ operations
the effort can be considerable, e.g. at the time of writing a navy like Britain’s has
to tie up a third of its surface combatants in maintaining a simple fleet in the
Gulf. Both numbers of ships and sufficient auxiliaries are required to achieve the
sufficient ‘reach’, at least in national terms.21
The dividing line between ‘simple fleet’ and ‘superior fleet’ operations is a
function of the opposition to be expected and it is here that the growth of naval
capabilities among potential victims acts as an important factor. Deploying a
‘superior fleet’ would strain even powers like Britain in many sets of
circumstances. Simple (i.e. unopposed) amphibious operations are also by
definition dependent on the nature of the opposition but may be carried out with
surprisingly limited forces (e.g. a single ship’s contingent of marines) in the right
circumstances. Superior amphibious operations, however, require maximum
efforts from medium powers, and are correspondingly risky.
The growth of navies with significant abilities to project power, and the
increasing opportunities the future multipolar world, may provide for this power
to be projected, will widen the number of assailants in exercises of gunboat
diplomacy and the types of operations that are practical for them as well as the
number of victims. India, China and perhaps even Japan may find themselves as
practitioners of the art, as well as other local powers where appropriate. The
Europeans will also retain a surprising number of options if they choose to
operate them, especially if they pool their efforts ‘out of area’ as they do within
it.
Before leaving the peacetime roles of navies we need to examine the roles
they might play in the management of a major crisis between the superpowers
and their allies. The American Maritime Strategy puts great emphasis on crisis
response and the world-wide deterrent value of forward deployed forces.
Admiral Watkins argued that, ‘the flexibility and precision available in
employing naval forces provides escalation control in any crisis, but has
particular significance in those crises which involve the Soviet Union’.22
NATO’s Concept of Maritime Operations, as well as plans to fight (see
below) also contains a ‘family of contingency plans for use in times of
tension’.23 These are, according to official spokesmen, ‘extremely flexible in the
scale of forces they may generate, and how and where they may be applied.
They permit graduated responses to a threat, providing a wide range of political
as well as military options, with the aim of deterring war.’24 Once the NATO
Defence Planning Committee approves, SACLANT can assemble forces from
Alliance members for early deployment. Such forces might well include short
notice amphibious units (see Chapter 10). Surveillance, however, would be
crucial. In the words of Deputy SACLANT in 1987:

The presence of allied forces in a surveillance role not only allows the
collection of information on Warsaw Pact platforms, weapons systems,
operational patterns, tactics and current activities; but also provides an
efficient and effective demonstration of our own maritime capabilities. The
advantage of these contingency plans is that they are flexible and are written
to move easily into either complementary war plans or, preferably, back to
peace.25

This process of maritime crisis management might be significantly enhanced by


the increasing ritualization of crisis responses through an agreed regime of
confidence and security building measures (CSBMs) putting agreed limits on
naval exercises in normal times and providing codes of conduct for naval
manoeuvres in close proximity. As Admiral Mustin has perceptively pointed out,
this process has tacitly gone a long way already with the routine deployment on
exercise of US carriers in the sea areas relatively close to Soviet bases, in a sense
legitimizing their forward deployment for deterrence and stabilizing crisis
management purposes when circumstances require.26
Formalizing these deployments in a regime of CSBMs might enhance
stability still further. It would provide formal legitimacy to forward deployments,
establishing their ‘normal times’ nature. The onset of a ‘crisis’ could also be
formally announced by exceeding agreed limits or, at least, manipulating the
threat to do so. Once ‘normal times’ limits were exceeded, strict rules of
engagement would be even more important to prevent the situation getting out of
hand. This might include assets being exposed to risks that would not be
acceptable in a normal operational situation. There is little reason to think that
the political imperative not to be the first to fire would be any weaker in such a
confrontation than it has been in other post-war crisis situations, e.g. in the Gulf
where the USS Stark was a notable victim of necessary restraint. Indeed, with
the stakes so high restraint would be even more important – and strongly
applied.
In Europe, however, persistent and contradictory doubts have been expressed
not only about the wisdom in crisis stability terms of US forces rapidly
deploying forward in a crisis, but also whether they would actually be available
to do so given the other likely calls upon them. This gives Europe’s own navies a
vital role in early forward crisis deployment in the Norwegian Sea. Such
deployment demonstrates collective Alliance resolve by its very presence (and
the insertion of amphibious reinforcements), but it does not create such a
dangerous offensive threat that the adversary in the crisis need feel provoked.
The possibility of US reinforcement would, however, be real; indeed, the NATO
ships would be actively engaged in ‘precursor operations’ to clear the way for
carrier striking fleet. One would hope that the Soviet Union would take the hint
that they had misjudged the situation and defuse the confrontation. Such
operations are much less demanding than a full-scale battle in the Soviet Union’s
‘backyard’, but the naval forces available and the land-based air cover giving
them support need to be sufficiently strong not to pose tempting vulnerabilities.
Such a level of capability is not too difficult to achieve with present and likely
European naval assets.
Despite everyone’s efforts, things might go wrong and shooting might start. A
victim of gunboat diplomacy might surprisingly refuse to accept defeat. An
assailant might grossly misjudge the situation. Despite the ritualized manoeuvres
of warships – or perhaps even because of them – a quarrel totally unconnected
with the sea might spill over into a situation recognizably called ‘war’. What of
navies then?
Chapter 10

Navies in war

Navies would have a part to play in all forms of future war, total and limited,
nuclear and conventional. To take the least likely first, if an unlimited nuclear
war occurred naval forces would be the ultimate instruments of the destruction
of the participants’ homelands. The ballistic missiles carried in the submarines’
missile tubes of both superpowers are enough on their own to obliterate the
others’ societies as viable units. Although the United States has currently rather
neglected its SSBN forces in its rush to enhance general purpose naval
capabilities, the end of the 1990s should see forces that could be used, if
required, in counter-force strategic nuclear warfare against land-based missiles,
Soviet command centres and other ‘hard’ targets.
This might enhance the ‘strategic reserve’ role of the seven hundred plus
nuclear-tipped cruise missiles to be deployed in the US Navy’s attack
submarines, battleships, cruisers, and destroyers. Such missiles provide a
capability to strike the Soviet Union from all azimuths, compounding the
defensive problem to impossible proportions. The numbers and flight
characteristics of cruise missiles make them relatively invulnerable to a wide
range of Soviet weapons, especially the expanded ABM system that the USSR
might deploy in any break out from, or breakdown of, the ABM treaty. There are
significant costs, however. Naval forces are likely to be called away from more
immediately useful duties to pose a nuclear threat; if they are not specially
allocated to nuclear strike duties they are unlikely to be in the right place to
attack the required land targets. Movements of nuclear-armed surface units
might well be constrained, either out of prudence in a crisis or because they are
unwelcome in nuclear free areas. Nevertheless, the attractions of using the
West’s maritime predominance to enhance the strategic nuclear threat facing the
Soviet Union seem at the moment to be too strong to overcome.
The Soviet Union continues to improve its nuclear-powered ballistic missile
submarine forces with new missiles, MIRVs and boats (if boat is an appropriate
term for the huge Typhoons). These forces seem to be regarded by the Soviets as
a strategic reserve to be withheld and defended in bastions off the Soviet coast
and below the ice pack. Deployed in a more forward position, perhaps implying
earlier potential use, will be the large and small strategic nuclear cruise missiles
currently about to enter service in reply to US developments. Soviet sea-
launched cruise missiles might join Soviet land-based forces in a counter-force
first strike against the USA should the USSR’s leaders be forced to such
disastrous extremes.
The smaller nuclear powers, too, stand ready to inflict catastrophic damage on
their opponents if suitably provoked. Currently constrained to aim all its
available Polaris warheads at Moscow, the British will be able to adopt more
ambitious targeting plans when Trident D-5 comes into service. The French will
soon be able undoubtedly to destroy the œuvres vives (vital works) of their
Soviet potential opponent with their MIR Ved M-4 missile and its eventual M-5
replacement.1 China is also sending some of its strategic nuclear forces to sea
with an SSBN in service and three more are under construction, but the
programme has run into considerable and persistent difficulties and it proceeds
very slowly. All these smaller powers would no doubt continue to target their
weapons primarily on relatively soft, urban industrial targets so as to inflict
maximum damage. More importantly, however, the threat of the use of all these
ultimate deterrents would be used as the primary means of deterring all forms of
war, or should war nevertheless occur, limiting it to ‘acceptable’ levels, whether
that be very limited nuclear exchanges at a strategic or theatre level or, much
more likely, to conventional weapons alone.
Navies have an enormous interest in such limitations as they are at their most
relevant in conventional hostilities. Both superpowers, to a considerable degree,
and the British and French to a much more limited one, deploy tactical nuclear
weapons for naval warfare. It is possible to rehearse a theology of nuclear war at
sea that has some plausibility (lack of collateral damage, easy definition of
military targets, etc.), but in reality it seems unlikely that nuclear weapons would
be released for use at sea before they were released as the ultimate symbol of
resolve on land, probably in Europe or Asia. Given the prevailing doctrines of
nuclear use whose logic is more in line with the subtle French concepts of ‘pre-
strategic’ nuclear warnings rather than simple operational nuclear defence, it
would make sense not to blur the signal too much by, for example, using nuclear
weapons in an early stage of hostilities as ASW devices. There might, however,
be some utility in making nuclear warnings against naval targets if these were of
sufficient significance. The sinking of a large surface warship like Kirov or Kiev
with a large nuclear warhead, such as the 300 kiloton weapon carried by the
French ASMP missile (which is to be fitted to naval aircraft), might well be a
sufficiently ‘sensational’ event to give the Soviets pause for thought.
If such a warning failed and full-scale nuclear war ensued, a major target of
counter-force attack would be the ballistic missile firing submarines of both
sides. Given the inherent capacity of such submarines to lose themselves in the
ocean, only isolated successes could be expected even if nuclear weapons were
used (including long-range strategic missiles in barrage mode if such weapons
did not have more appropriate targets or objectives of higher priority). There are,
and there are likely to remain, too many variables operating in favour of the
ballistic missile submarine to undermine the second strike potential of even quite
a small force of these boats. They have the advantage of not having to close their
targets to be effective. This makes the ASW problem enormously difficult. A
surprise disarming strike is also probably impossible. Once an SSBN has fired
its first missile it has given away its position, but in full-scale nuclear war, with
all its confusions, it is asking a great deal for an enemy to react rapidly to
ballistic missiles rising from the ocean in a number of widely separated places.
An individual submarine, therefore, will be likely to get away all its missiles.
This, however, implies Herman Kahn’s nuclear ‘wargasm’. Submarine launched
ballistic missiles are probably not the most appropriate means of mounting
nuclear demonstrations. Maintaining the uncertainty of their positions is
paramount and their owners would probably always possess alternative means of
carrying out nuclear warnings.
It is possible that some kind of naval hostilities might continue after the
destruction of the homelands of the participants in a full-scale nuclear war. In the
1950s it was fashionable for a time, especially in naval circles, to speculate
about a kind of ‘broken backed’ hostilities, when operations would continue at a
minimal level, both to sustain survivors and to rebuild for a next intensive phase
which might even bring victory. This concept was always unattractive to the
policymakers or the public then, and is little less so today. It is possible that
naval forces would continue to fight each other in some sort of strategic spasm
until their ammunition and supplies were exhausted. Perhaps cut-off from their
infrastructure, some might become latter-day pirates preying on the remaining
shipping and coastal areas of what parts of the world remained habitable. The
control of a coherent ‘war’, however, would be impossible and only
sentimentality and instinct would call the ships and submarines of countries that
had suffered nuclear attacks back to view the destruction.
Happily, the uncertainties and dangers of such a scenario (if the
understatement is to be forgiven) make its happening extremely unlikely. Indeed,
the likelihood of limited nuclear war becoming total is going to give decision-
makers enormous incentives to keep hostilities below the nuclear threshold. The
likelihood of large-scale conventional superpower vs. superpower coalition war
is a matter of debate. Perspectives differ within NATO, for example. The West
Germans, for whom such a war would bring destruction virtually
indistinguishable from nuclear devastation, have always tended to argue against
too much emphasis on large-scale conventional war fighting. Nevertheless, even
they now seem to be accepting that the only credible response to a Soviet
conventional attack is conventional response in kind. Every trend in NATO
strategy is towards an enhanced conventional capability. The implications of this
are clear. A conventional war will not be over in a few hours: it may last weeks
or months, possibly even years. The greater the conventional emphasis in NATO
strategy, the greater the importance of sea power in its traditional role of
maintaining the flow of men and supplies so that the flight can be continued and
domestic populations sustained.
It can be argued that a long drawn out conventional war is not a very
attractive option, and that it should be prevented by making clear that nuclear
escalation is almost certain if the conflict goes on beyond a certain relatively
brief period of time. In an era of nuclear parity, however, such a threat is only
dubiously credible. There is probably no alternative to retaining at least some
options on sustaining hostilities for as long as possible. Deterrence is indeed
enhanced by demonstrating to the potential enemy, especially one with a
plausible, short conventional war winning operational doctrine, that there is a
more likely danger than nuclear war if he chooses to try such an adventure: a
long drawn out conventional conflict which his economic weaknesses mean he
cannot win.
The self-cancelling effect of both side’s nuclear forces is making large-scale
conventional war at sea more, rather than less, likely in the event of superpower
confrontation: although that confrontation itself is something both sides would
rather avoid, due to the incalculable dangers of nuclear escalation in the event of
a direct clash between them. Even as things stand today, with publicly articulated
statements emphasizing the high probability of nuclear escalation, NATO plans
to carry on a major maritime reinforcement operation of conventional forces.
The current Rapid Reinforcement Plan calls for thirty US and Canadian combat
brigades, 100 air squadrons, and 300,000 men to be transported to Europe after
the day reinforcement begins (R-Day). By R + 180 some 1½ million men, 8½
million tons of ammunition and stores, and 114 million barrels of fuel would be
transported to Europe. In addition to this about 100,000 military personnel,
25,000 tons of material, and perhaps 20,000 vehicles would have to be
transported from the United Kingdom to Europe. Ninety-eight per cent of
personnel can be airlifted but, even with pre-positioning of significant amounts
of heavy equipment in Europe, very large quantities of the latter must come by
sea. Shortages of aircraft and congestion problems at airfields put significant
constraints on reinforcement airlift.2
According to EASTLANT (Eastern Atlantic) figures, the requirement for
military reinforcement and supply shipping over the first six months is very
considerable: 3,045 shiploads.3 The pressure for ships would be highest in the
opening stages. In February 1988 a British MOD official estimated the first
month’s shipping movements at 1,000 and cargoes at 1.3 million ‘tonnes of war
stores’.4 SACLANT’s (Supreme Allied Commander Atlantic) figures are rather
smaller but this could reflect recent changes in perceived reinforcement
requirements. At the 1986 Sealink Conference SACLANT, Admiral Lee Baggett
stated that ‘our latest studies show that the requirement is about 400 shiploads a
month for military reinforcement and 400 for military re-supply: according to
1987 figures these 800 ships would carry 40-50 million tons of cargo’.5 On
given shipload figures even this pace would only be necessary for the first two
months. After R + 60 one might, therefore, expect the military requirement to
tail off somewhat, perhaps to about 300 shiploads a month or so.
In addition to all this, economic shipping would have to continue to arrive to
maintain economic life in embattled NATO. SACLANT currently estimates the
total number of monthly shiploads that need to arrive in NATO’s ports at
between 1,500 and 2,000.6 Unofficial estimates of NATO’s ‘civilian’
requirements in war are 365 million tonnes per annum of crude oil, bauxite, iron
ore, phosphates, cement, sugar, salt, fertilizer, gypsum, non-ferrous ores, and
wood products for North America, and 948 million tonnes per year of similar
materials plus coal and grain for NATO Europe. More official SACLANT
figures estimate the economic requirement at ‘95–110 million tons of cargo’. As
Vice-Admiral Sir Geoffrey Dalton, Deputy SACLANT, told his audience in Oslo
in April 1987:

For the economic shipping ... the routes will be quite different from previous
wars. No longer will the great majority of cargoes move from North America
to Europe, instead they will originate mostly in other parts of the world. The
reason is that, although large quantities of items like, grain, coal, iron ore and
general cargo will continue to be transported to Europe from North America,
by far the bulk of the economic shipping will travel to Europe from the
Arabian Gulf, West, North and South Africa, South America and Australia.
Moreover, as an additional difference from previous experience, the
destination of a significant part of the shipping will be North America.7

The shipping requirements implied by the above needs are hard to calculate
exactly given different ship sizes, speeds, and transit times. The NATO
reinforcement and resupply requirement is officially put at 600 American and
600 European merchantmen, but there are doubts over whether these figures can
now be met.8 According to the US Joint Chiefs of Staff’s own figures, the total
number of ‘funded strategic sealift resources’ available is 533 ships. Reserves
requiring two to three months to get to sea and, hence, only useful for replacing
attrition, add another 197 hulls. These must cover all contingencies, both in the
Atlantic and Pacific. Korea can add another 45 ships, and 600 vessels are
expected from European NATO.9
Sadly, however, the latest figures produced by NATO’s Planning Board for
Ocean Shipping show that only 448 of the requisite dry cargo ships are available
on the NATO ‘Sealift Shipping List’.10 The situation is not quite so desperate as
it sounds. The 600 figure included a 50 per cent margin but that margin is now
only just over 10 per cent and falling: NATO lost some 50 dry cargo ships in
1986-7 alone. It is now being recommended that beneficially owned flag of
convenience vessels be added back to the list in order to restore its size. This is
especially important, as the sealift requirement is expected to increase.11
Whether or not the ships are available, however, their defence plus the defence
of all the economic shipping required for the West’s survival raises all the old
questions of battlefleet versus direct protection as discussed in Part I.
This debate has been sharpened in the 1980s by the American development of
the forward Maritime Strategy, its clear articulation in open publications and its
adoption as the basis of NATO’s maritime defensive concept.12 Although, as we
have seen, World War II was a triumphant vindication of convoy and escort, the
predominance of this operational doctrine was soon questioned. The major
factors were technological developments both in submarines and anti-submarine
warfare, the advent of nuclear weapons, and the bureaucratic domination of the
aircraft carrier in the dominant navy of the Western Alliance. New types of high-
speed submarines, both conventional and nuclear powered, rapidly undermined
many of the tactics of the World War II convoy escort. The development of new
long-range surveillance sonar networks seemed to provide one answer to this,
giving ASW some of the characteristics of air defence in allowing the vectoring
of ‘interceptor’ assets on to fast-moving threats. Hence the 1950s saw a major
new emphasis on ‘hunter killer’ operations. Even these, however, seemed less
effective than the ‘attack at source’ on the submarines in their bases. These
operations had been a miserable failure in both world wars but in the 1950s
nuclear weapons, including special naval earth penetrating devices developed for
the role, accurately delivered by carrier aviation, could open up the most
formidable submarine pen. As soon as NATO’s Atlantic Command was set up, a
nuclear-armed carrier-striking fleet was established as its primary arm, with the
‘attack at source’ on the Soviet submarine fleet as a high priority. This fitted in
well with an understandable tendency of the USN to emphasize the carrier battle
group operations with which it won the Pacific War, rather than the convoy
escort work that it had largely left to its British allies.13
There is thus little really new in the US Navy’s Maritime Strategy and its
proposals for forward carrier offensive, although that offensive is now designed
to be primarily conventional rather than nuclear. NATO has always had a carrier-
centred battle fleet with an important perceived role in gaining the command of
the North Atlantic in any future war. With the new-found confidence of the
1980s, however, the US Navy now openly discusses how such an offensive
would force the mass of Soviet naval assets on to the defensive, so drawing them
away from NATO’s Atlantic shipping. Not only would the Soviet Navy be
pinned down, but it might be rapidly defeated and destroyed in a latter day
version of Mahan’s decisive battle.
It had tended to be assumed in the 1960s and 1970s that carrier battle groups
(CVBGs) would keep their distance from heavy concentrations of enemy
submarines and strike aircraft due to the inherent vulnerability of the carriers and
their escorts to torpedo and missile attack. The authors of the Maritime Strategy
have turned this on its head. They now argue that a CVBG would positively
welcome the enemy coming out to fight with his major offensive assets
(effectively a three dimensional battle fleet) as the technologically superior
NATO forces would use this opportunity to destroy them. The Battle of the
Philippine Sea, when US carriers effectively destroyed the Japanese carrier-
based Naval Air Arm as an effective force, is a powerful precedent. The long-
range threat posed to Soviet bases by the carrier-based aircraft (and to a lesser
extent cruise missiles) of the latter-day American battlefleet would mean that the
option of mounting a ‘fleet in being’ might not be so open to the Soviets as it
was to the Germans in both wars. They would have to come out and fight or face
destruction in harbour. The strategic ‘pay off’ from the naval offensive might be
greater still. What is new in recent formulations of US naval strategy is the focus
on the Soviet ballistic missile firing submarines. Some American naval planners
seem to have detected in missile firing submarines a key form of Soviet sea use
which the Soviets would not only hazard their fleet to defend, but which might
even be a point of such sensitivity that it could be exploited to achieve war
termination on favourable terms, i.e. victory. This campaign could lead well
under the polar ice cap. The Soviets have understandably chosen to exploit one
of the advantages geography has given them of a large area of ocean cut off from
aerial or space surveillance by solid masses of ice. Modern Soviet PLARBs are
specially configured for under ice operations, and the Americans have taken up
the challenge by modifying existing and future SSN designs to go in after them.
The cold arctic depths may well be a major combat theatre in any future conflict.
The ‘war termination leverage’ achieved by inflicting attrition on Soviet
ballistic missile firing submarines is one of the most controversial aspects of
Maritime Strategy.14 Some, like Barry Posen, have seen it as a formula for
premature and accidental nuclear escalation.15 The Soviets might see the loss of
their SSBNs (in Russian, PLARBs) as tantamount to a pre-emptive strategic
strike. Alternatively, it can be argued that the Soviets themselves expect to
destroy NATO nuclear targets on land in their official conventional offensive
without necessarily leading to NATO nuclear use. There would apparently be no
automatic ‘use ‘em or lose ‘em’ assumptions in the minds of the USSR’s
strategic planners.
A major offensive into waters close to the USSR is fraught with dangers and
the US Navy has been at pains to point out that it foresees no simple ‘charge of
the light brigade’ with its carriers. Examination of recent NATO exercises clearly
demonstrates that in the North Atlantic, at least, the aim is to take the initiative
early and place carriers inside the fjords north of Bodo, where the mountains and
water conditions help the NATO forces exploit the traditional advantages of the
defender. Soviet attack options are constrained by geography, and available
NATO assets can be concentrated on certain threat axes; Soviet attempts to
neutralize the potent threat these forces pose to Kola bases are thus more easily
defeated. Once the Soviet three-dimensional battlefleet has been neutralized then
moves forward might be possible.
This brings in a further danger, however. If the forward deployed NATO
Striking Fleet did succeed in achieving a Marianas Turkey Shoot against the
Northern Fleet’s aircraft and inflicting serious casualties on submarines and
surface ships, it might only leave Soviet commanders with nuclear options with
which to defend the homeland and their vital nuclear strategic reserve. Soviet
political leaders, faced with dire and extreme choices, might well feel that a
definitive nuclear strike against the NATO Striking Fleet might not necessarily
lead to all-out nuclear war, although this assumption would be more difficult to
make with the fleet operating close to land. The American reaction to the loss of
tens of thousands of naval personnel might, however, be more violent. A
disastrous spiral of escalation would probably ensue.
Of course, American confidence might be misplaced. Even a conventional
naval battle might go badly for the Striking Fleet in its enemy’s ‘backyard’. A
single lucky hit from Soviet long-range homing torpedoes could have
catastrophic effects on the operational viability of an American supercarriers.
The level of missile threat posed by Soviet aircraft and submarines (and possibly
surface ships) is such that any air defence, even the F-14/Aegis combination of
the Americans, would be stretched to the uttermost, especially as it would also
have to cover the British ASW forces, vital to give protection against the
subsurface threat but less well-equipped with organic air defences.
Naval warfare of this intensity is a desperately risky business, with very high
stakes, given the concentration of resources in each individual modern naval
unit. It is likely, therefore, that prudence and caution will prevail. Western carrier
battle groups are not going to go charging off into the Barents Sea. Here the
initial Western offensive will be limited to nuclear-powered submarines. This
will achieve the vital objective of pinning down Soviet assets, although there
would still be various pulls for escalation. Submarines in these circumstances
would be very hard to control if intra-war bargaining produced a cease-fire. In
such a submarine offensive those lesser naval powers which have invested
heavily in SSNs (at considerable opportunity cost) could, however, play a
significant role.
Whatever the level of forward deployment it is unlikely that NATO could
neglect ‘barrier’ operations to take advantage of its favourable geographical
position, notably the Greenland–Iceland–UK ‘gap’. All kinds of force could be
used on barrier operations although mines, conventional submarines, and shore-
based aircraft seem the most attractive options, leaving surface units for use
elsewhere, where their unique characteristics are of greater use. The utility of
barriers is, however, only limited and far from certain. It is hoped that five may
be set up, but their efficiency might vary from one kill in twenty boats to one in
five.16 Moreover, barriers could, of course, only become operational once
hostilities had commenced. A large number of hostile submarines could well be
able to break out into the open ocean therefore.
It is often argued that the Soviet doctrine seems to indicate that they do not
put a very high priority on interdicting reinforcement shipping in the allocation
of their submarine and air assets. The naval forces would be tied down
protecting their precious PLARBs in their defended ‘bastions’. Certainly the
Soviets seem prone to such doctrinal assumptions and this argues for NATO
doing as much as it prudently can to foster such concerns in order to direct
Soviet activities away from what could hurt the West the most. Nevertheless, the
central importance of reinforcement and supply shipping to NATO’s capacity to
continue with conventional fighting should imply as high a level of capability
being allocated to the direct defence of shipping as practical. Indeed, a good case
exists for direct defence operations still being allocated the highest priority. Only
in direct sea control operations can naval assets achieve the primary strategic
aim – the safe and timely arrival of cargoes – even if they do not actually sink or
shoot down enemy units. The latter’s life can be made so difficult and dangerous
that they cannot carry on with their missions. If it actually attacks a valuable
cargo vessel an enemy submarine has inevitably to give away its position,
making its destruction much more likely, especially if a friendly naval unit is
nearby. Even if its initial attack is successful it will thus have considerable
difficulty in carrying out a second or third. Modern ideas combining close escort
and more distant support using modern passive sonar techniques have been
suggested and seem very practical.17
One major argument against convoying, one with a very long pedigree, is
shortage of escorts. It is now usual to say that NATO must adopt a forward
strategy as she has insufficient escorts to protect directly all her shipping. The
evidence produced to sustain this thesis, however, cannot but be very uncertain.
A relatively recent study found it ‘difficult at best to determine Atlantic escort
level requirements more accurately than somewhere between 18 and 273’.
Persian Gulf requirements were between 53 and 370 and a ‘middle the road
guess’ for Pacific convoys ‘about 175’. Non-US NATO escorts usable in an
Atlantic battle were estimated at between 50 and 100. The whole question was ‘a
matter of judgement rather than mathematics’.18 The current compromise seems
to be, as far as the Atlantic is concerned, to send most ships individually to the
Azores where they will be escorted in convoy to the coast of Europe, being
defended in depth by surface and air forces.19
There may be little alternative but to return more wholeheartedly to convoy as
the declining capability of very long-range detection makes such direct
protection more necessary than less. As we have seen, the long-range sonar
capabilities that allowed the authors of Western naval operational concepts to
downplay the disturbing lessons of experience seem to be coming to an end. Not
even at the height of passive sonar capabilities were naval commanders able to
do without defensive screens for their most valuable assets about which they
cared most, their major warships. Long-range active sonar capabilities in
helicopters, and the availability of cheap floating helicopter platforms, might
well improve the effectiveness of escorts to the level that the inherent economy
of the convoy system in grouping assets at risk with protective forces is still
further enhanced. Even Norman Friedman, an informed analyst usually quite
hostile to convoys, has argued that ‘the advent of long range helicopter dipping
sonar would be a way of achieving something resembling effective screening
even though NATO cannot really hope to achieve ship numbers comparable with
those of, say World War II’.20 It was also interesting to see convoying
immediately adopted to protect US-flagged shipping from aerial, mine, and
surface threats in the Gulf in 1987.
Convoy has never been as profligate of escorts as its opponents have
consistently claimed. Indeed, under the conditions of the two world wars it
proved to be by far the most economical and efficient use of limited naval assets.
What has often been the case, however, has been the starvation of mercantile
escort groups to feed the apparently insatiable appetite of warship escort and
‘main fleet’ operations. Troop convoys have usually received a higher priority
somewhere in between normal ships and warships; indeed, main fleet units have
been used to screen sensitive troop convoys on many occasions. This would lead
one to speculate that in any future conventional NATO war there might well be
alternative pressures for the allocation of NATO’s precious naval assets.
NATO’s Concept of Maritime Operations (CONMAROPS) is in fact very
flexible. It was designed as ‘a conceptual basis for Nato’s maritime planning in
the 80s and 90s’.21 It contains three principles:

The first one is containment – which in war translates into offensive action to
prevent Warsaw Pact forces deploying into open waters. This is particularly
important in areas such as the Norwegian Sea – if we can prevent Soviet
maritime forces from steaming into the Atlantic, the antisubmarine battle,
particularly with our limited forces, will be less difficult.
The second principle is defence in depth – the deployment of forces to
defend in depth is designed to put Soviet ships and submarines at risk at
every turn. For this to be achieved, we must be prepared not only at the
forward edges of the NATO area, but throughout all our area.
The third principle is keeping the initiative – vital in tension to ensure forces
are brought early to a high state of readiness and properly positioned, but
equally important in war to ensure they are used where they are needed
most.22
This clearly implies forward operations of some kind, the readiness ‘to fight the
Soviets at the forward edge of the NATO area, along their exit routes, and in
defence of the allied war and merchant shipping’, and to adopt an aggressive
operational and tactical approach. It does not, however, prescribe any particular
action: available forces would be shuffled as required between three
‘campaigns’, Norwegian Sea, Atlantic, and Shallow Seas on the basis of a set of
‘general defence plans’.23 NATO’s Atlantic naval exercises are ambiguous, as
they are bound to be, as they are merely testing and rehearsing options.
Sometimes operations in the Norwegian Sea follow strike fleet operations in
close defence of ‘sea lines of communication’. Sometimes the Striking Fleet
heads straight for Norway.24 The actual war deployment would depend on ‘the
prevailing circumstances . . . what Soviet forces are deployed in the area? What
is the status of the land battle? And who has the necessary air superiority?’25 The
key to the Norwegian sea battle is early deployment. If powerful forces are not
inserted early as part of NATO’s crisis contingency plans (see Chapter 9) it will
prove much harder to insert them against a prepared enemy. World War III may
well begin, like the active part of World War II, with a race to the fjords.
In order to bring ground forces to bear in Norway, the Americans seem to
envisage a three-phase approach, taking advantage of the varying strengths of
pre-positioning, air mobility, seaborne logistics, and more traditional amphibious
operations. Equipment and supplies have been pre-positioned in Norway for a
specialized cold weather marine expeditionary brigade to be air-lifted in within
days of alert, with thirty days of ‘initial sustainability’. Supplies must, however,
be sea-lifted from Trondelag up to the likely northern theatre of operations.26
Secondly, there is a heavy mechanized marine expeditionary brigade that can be
rapidly inserted and supported by the four ships of the Atlantic Maritime Pre-
positioning Force. As the Joint Chiefs of Staff define their role: ‘The MPS
program is designed to combine the responsiveness of air-lifted marines with
sea-lift of pre-positioned equipment’.27 The troops are air-lifted ‘to the vicinity
of the objective area where they join up with the ships and their equipment. . . .
This brigade . . . can be combat capable and ready to move to designated
objectives in as little as five days of arrival in the assembly area.’28 Finally, the
Striking Fleet brings in the rest of Second Marine Expeditionary Force in its
Amphibious Striking Force Component, as well as the carrier air power and
other naval support that would probably be crucial to eventual success.
In addition to American forces the UK/Netherlands Amphibious Force, a
brigade of 6,500 men, also has reinforcing Norway as its primary role. It is
rapidly available in theatre at a nominal week’s notice. The force can be ‘poised’
as required, i.e.

held in readiness . . . rather than committed prematurely. . . . Depending on


the situation ashore and recent intelligence the decision on where to land can
be delayed till the last moment (provided that contingency plans have been
made for alternatives) ... the entire landing force while embarked can be
moved a distance of up to 300 nautical miles in 24 hours, and does not
require large sophisticated beaches over which to land.29

European naval forces, operating under land-based air cover would be required
to support and escort this operation, either in addition to those committed to the
Striking Fleet or as a prelude to moving westwards to join up with the main
force.
Before the main Striking Fleet arrives allied maritime forces, probably very
largely European, must engage in ‘precursor operations’, clearing the fjords of
enemy mines and submarines. The Norwegian Navy will, of course, have a
major role in coast defence, mine countermeasures, coastal convoy escort (as
stated above, sea communications will be vital in moving even pre-positioned
equipment), and generally providing the ‘socket’ into which outside
reinforcements can be ‘plugged’. Nevertheless, the Norwegians keep their
options open. As Vice-Admiral Rein, the Naval Commander Northern Norway
diplomatically put it: ‘The uncertainties related to the time aspect of deploying
NATO forces into the Northern area imply that our national forces may get
heavily involved in crisis management, or even combat operations, before allied
forces could arrive and provide the desired support.’30
The defence of Norway would be a joint operation between Supreme Allied
Commander Atlantic (SACLANT) and Supreme Allied Commander Europe
(SACEUR). The Shallow Seas Campaign brings in CINCHAN (Command in
Chief Channel) also, and covers the Baltic, North Sea, Channel, and South West
Approaches. In the Baltic the emphasis would be on local fast-attack craft,
coastal submarines and aircraft, notably the Bundesmarine’s powerful missile-
equipped Tornado force. These light, relatively unspectacular forces of a three-
dimensional kind have the vital task of safeguarding the Central Front’s northern
flank. In the North Sea and close South Western Approaches the emphasis is
once more on MCM and ASW, and the safeguarding of both ships in the final
stages of their oceanic voyages to Europe and those engaged in short sea
passages transporting reinforcements from the United Kingdom to the Continent
and (in the early stages) families and other civilians the opposite way. Again
shore-based air, surface and sub-surface assets have to act in close cooperation,
e.g. the use of the aviation training Royal Fleet Auxiliary Engadine as a forward
base for shore-based ASW helicopters.
In the Mediterranean, SACEUR via the Admiral who is CINC-SOUTH has
two maritime campaigns of his own, Eastern Mediterranean and Mediterranean
Lifelines. Here the 6th Fleet has traditionally provided one of the NATO’s core
air-striking forces. This might still be important, although the carriers might well
have other calls on their attention to the east as well as to the north. Amphibious
operations might also play a part on the southern flank, either in the classic
scenario in the defence of Turkish Thrace or elsewhere. It would also be
necessary to use sea denial submarine, air and fast-attack forces, as well as
minefields, to deny the Soviets any amphibious options, e.g. the Black Sea.
Local navies would come into their own here.31
As for the Central Front itself, the NATO Striking Fleet has traditionally been
allocated on air support role. This reflected NATO’s early weaknesses in land-
based air power, as well as the Striking Fleet’s lack of targets in its early days
once the Soviet cruiser or two that had broken out into the Atlantic had been
dealt with. Nowadays the Central Front role seems less important, although it is
still exercised. One suspects that now the US Navy has got a worthy opponent in
the Soviet Fleet it will tend to concentrate its major surface assets on the latter’s
defeat, or at least in defence of major NATO assets from an enhanced Soviet
threat. SACEUR will have to rely on much more appropriate land-based aircraft,
leaving the carriers to cover targets to which they give special access.
Nevertheless, if NATO air power does suffer catastrophic losses US carriers
might well prove useful to plug gaps or to mount desperate nuclear escalations.
In such circumstances sea-based cruise missiles might play a useful role.
So far the discussion has concentrated on a European scenario but it is quite
possible, if not likely, that any coalition war would spread round the periphery of
the USSR. The Maritime Strategy indeed welcomes the probability of horizontal
escalation, the application of maritime pressure against the Soviet Union’s all
round periphery. The Far East PLARB bastion and Soviet Far Eastern naval
bases are as fair game as those in the Arctic, especially as one of the main roots
of the current Maritime Strategy was the Sea Strike plan for ‘a multi-carrier
battle group operation directed against Soviet Pacific naval ports such as
Petropavlovsk’, prepared in the mid-1970s by Admiral Thomas B. Hayward
(later CNO 1978-82).32
The maritime battle in Pacific waters would be of even greater long-term
significance than that in the Atlantic, for it is here that the key industrial base of
the capitalist world is increasingly concentrated. In a future protracted war
between US and Soviet-dominated coalitions many of the main munitions
industries of the US coalition would be in Japan, the leading OECD producer of
iron and steel and motor vehicles, and to a lesser extent in South Korea and even
Taiwan. This would require a great deal of shipping, both to take in fuel and raw
materials, and to bring out finished products. Japan depends on external supplies
for almost all of her iron ore and crude oil: she even exports 80 per cent of her
coal. Some stockpiling has taken place in Japan but her near-total import
dependence, and the need to continue to use the massive industrial facilities both
there and in Asia, adds up to a formidable naval problem. One wonders how far
the Japanese Navy, for example, has learned the lessons of its poor record in
shipping defence operations in 1941-5. Barrier patrols can now be expected,
given the very unfavourable Soviet geographical position, and there are few
reasons for the United States to forego its predilection for a carrier offensive
against Vladivostok or, more especially, isolated Petropavlovsk. Increasingly
sophisticated Japanese vessels should be able to give assistance in coping with
the air and ASW threats. Again, however, force of losses might well dictate
something of a reallocation of assets to direct shipping defence duties as time
went on. Once again, forward operations into the heart of the Soviet bastions in
the Sea of Okhotsk are probably best left to submarines, leaving the US carriers
to knock out the various isolated forward Soviet bases, e.g. Kam Ranh Bay or
any other future North Korean facilities. Amphibious landings on Sakhalin, or
similar islands, might well have a part to play in an American Far East maritime
offensive. It must also not be forgotten in this context that even such an
important Soviet base as Petropavlovsk on the Kamchatka Peninsula is
effectively an island.33
A third front in a superpower war would be the Middle East, especially the
Gulf area. Indeed, a Soviet American confrontation in this highly sensitive and
economically important area might be the spark that set off the conflagration in
the first place. The level of military activity in this region is crucially dependent
on the level of US forces diverted from other areas, but the sea inevitably plays a
major role in any US deployments. The US Central Command obtains a quarter
of its allocated strength in fighter strike aircraft from US Navy carrier forces,
both those in the Mediterranean as well as those in the Indian Ocean. Carrier-
based aircraft would be absolutely vital to a US air effort if the Americans were
finding it difficult in obtaining access to local bases. Carrier based A-6 (or A-12)
aircraft could reach into Northern Iran from both the Eastern Mediterranean and
the Northern Gulf, although the capacity of US carriers to operate in the latter
waters at least is a little dubious given lack of sea room and potential mining
threats. Such forces (plus USAF B-52s from Diego Garcia) would demonstrate
US opposition to any Soviet moves on land, and be effective against Soviet
airborne operations. For a complete air offensive against a full-scale land attack,
however, land-based aircraft flying from friendly states in the area would be
imperative.34
A divisional-sized Marine Expeditionary Force (MEF) is earmarked for the
Gulf area, but most of its troops would probably come by air. Supplies for the
MEF are carried in the three Maritime Pre-positioning Ship (MPS) squadrons.
Only one MPS squadron is at Diego Garcia, the other two are in the Atlantic and
Pacific, but it is considered that all three could be in the area within ten days,
each carrying thirty days’ supplies for one of the MEF’s brigades. This, of
course, assumes there are not other calls on MPS assets. Other PREPO (pre-
positioning) ships are at Diego Garcia; they carry about 55,000 tons of supplies
for the air force units initially deployed in the area.35
The aim, as in Europe, is to use sea, air, and land assets in optimum synergy.
Until the deployment of the three MPS squadrons, the whole MEF was to have
come by sea in amphibious lift, which would have taken between 3-4 weeks.
Now both the MEF and the 82nd Airborne Division are fully deployable within
fourteen days of mobilization. This, however, also requires extra airlift
capabilities. Until all 50 C-5Bs were purchased by 1988, the 82nd took as long
to deploy by air as the MEF coming by sea. Moreover, in initial deployment
time, sea and air are surprisingly equal. It takes 48 hours both to deploy the
initial battalion of the 82nd to an airfield in the Gulf area and to deploy a marine
battalion from the Indian Ocean. The 7th Marine Expeditionary Brigade can be
in the Gulf with amphibious ships in a week, by which time the 82nd would have
been built up to a roughly similar level. Only then does enhanced airlift begin to
give the advantage, and only on the basis of heavy equipment coming by sea.36
The purchase of the eight Algol class, 30 knot vehicle cargo ships allows one
of CENTCOM’s (Central Command) Army divisions to be in the Gulf with
supplies within four weeks, about the same time it takes the USA’s enhanced
1988 airlift capability to carry an infantry division there. Slower sealift brings in
another Army division in about 35 days, as well as supplies to keep operations
going.37
Whether such forces would be capable of successfully defeating a Soviet
army with all the advantages of land communications and internal lines is
another matter, but the terrain of Northern Iran is not very favourable to a rapid
offensive.38 After all, the major aim initially would be to demonstrate American
interest and commitment. The sheer possibility of the deployment of forces of
this size also must give the Soviets some pause for thought. They must calculate
that the Americans might well oppose a thrust to the Middle East in a way that
either could prevent any early conventional success, or that would provide the
conditions for a nuclear war. Four or five divisions is a force that no US
president could see destroyed without seriously considering nuclear escalation.
The synergy of nuclear and conventional forces, like that of land and sea power,
creates conditions where Soviet military adventurism is probably successfully
deterred. The problem of forces and lift assets being diverted elsewhere by
horizontal escalation is, however, significant. Marines fighting in Iranian passes
cannot storm ashore on Sakhalin; fast cargo ships racing for the Gulf cannot
simultaneously reinforce Europe.
Having some potential for action may be the most significant factor. For the
dangers of nuclear escalation would be a constant concern in any conventional
conflict on the scale just described, especially if one side or the other scored
successes approaching the decisive. These dangers will probably continue to
enforce the considerable caution in avoiding direct clashes that has generally
characterized superpower relations to date. This will, however, still leave much
scope for hostilities between smaller states, where the stakes are not so high,
except for the immediate participants.
There have, of course, been many such ‘limited’ wars in the period since the
end of the last great coalition war in 1945, sometimes with an external power
intervening on one side to help. These conflicts emphasize what are often called
‘power projection’ capabilities of navies. Naval air forces and, to a lesser extent,
naval artillery, have proved to be useful and flexible means by which external
powers can bring fire power to bear against less sophisticated opponents. Indeed,
from about 1950 to about 1973 and the end of the Vietnam War, this was the
major war-fighting role of the West’s navies.39 Such firepower helped stave off
Western defeat on a number of occasions, although it could not ensure victory.
Smaller powers can also use sea-based fire power where appropriate, e.g. Israel
against its neighbours. The limited nature of the opposition also makes
amphibious landings in limited war even more possible than in higher-level
hostilities, with potentially decisive results, e.g. against the North Koreans in
1950 and the Argentines in 1982.
‘Sea power’ can only be decisively and definitively influential, however, in
certain circumstances. Unsurprisingly, when the aim is to capture or recapture an
island, it can be the most important factor, but even in these cases other
circumstances have to be favourable. In 1982, a few hundred miles or so of extra
proximity to the Argentine coast might have made a great deal of difference.
When the target is physically connected by adequate communications to a major
land mass the problems are even greater. The Inchon landings in 1950 were
suddenly reversed by a massive application of continental power in the shape of
the Chinese People’s ‘Volunteers’. In major wars between Third World states
(which given their ferocity are better termed ‘local wars’ than ‘limited’), e.g. the
Indo-Pakistan War of 1971 or the Iran–Iraq War of the 1980s, high levels of sea
fighting can take place, but often only to limited effect. In 1971, Indian control
of the waters off what was soon to be Bangladesh isolated the area from the rest
of Pakistan, and made external military intervention troublesome. But the Indian
victory might have conceivably taken place without any maritime operations at
all. Only if the Pakistanis had been able to blunt the remarkable Indian blitzkrieg
into the disaffected province and create a defensible bastion would sea power
have been able to come into play, and this begs certain questions about
Pakistan’s capacity to mount a maritime reinforcement or relief operation.
In the war between Iraq and Iran there has been much more scope for the
traditional application of maritime power. Iraq rapidly made herself much less
vulnerable to this kind of pressure, being supported by her Arab friends and
being able to export oil by pipe through Turkey and Saudi Arabia to help sustain
her war effort.40 Iran, however, is dependent on ships for her vital oil exports,
and these were vulnerable to Iraq’s ‘naval’ forces in the shape of missile-firing
aircraft.
Although ships had been attacked by Iraq previously the campaign began ‘in
earnest’41 in Spring 1984. The Iraqis were hobbled by limited capabilities in
aircraft range, reconnaissance support, and lack of missile effectiveness against
mercantile targets, especially large tankers. The Iranians used convoys to protect
their shipping and introduced a shuttle tanker system requiring large tankers
(some VLCCs), equipping them heavily with ECM equipment and running them
down to new trans-shipment terminals around Sirri and Larak at the southern end
of the Gulf. Only slowly were the Iraqis able to acquire the aircraft range to
attack this lifeline effectively along all its length, and it took two years before the
Iranians began seriously and significantly to be hit by the attacks.42
The Iranians began retaliation against this Iraqi guerre de course in 1984
using aircraft-carrying weapons like the Maverick electro-optically homing
missiles, and even Sidewinder air-to-air systems.43 As well as being even less
effective than those of the Iraqis, Iranian attacks were also much less numerous
until the pressure of 1986 caused significant escalation by Iran. In 1985 there
were forty Iraqi attacks on shipping and only thirteen Iranian: in 1986 the figures
were sixty-five and forty-one respectively.44 Iran also began to make much
greater use of surface forces, including her Sea Killer equipped frigates. She also
began to acquire shore-based anti-ship missiles from China and perhaps most
significantly began to deploy mines from small motor boats.45
Iranian strategy seems to have been to exert pressure on Iraq’s Arab allies,
notably Kuwait. It also, however, had the effect of widening the naval
involvement both of the superpowers and of the Western Europeans. All these
powers had significant interests in maintaining freedom of shipping to the Arab
States, and the stage was set for the increasing naval deployments and
reflaggings of 1987.46 The involvement of outside naval actors, however,
illustrates another significant factor in limiting the role of the sea power of the
participants in local wars. Outside powers with an interest in maintaining the
flow of maritime trade, especially in vital materials such as oil, may well not let
local sea power prove more effective. Three factors then, (a) limited
vulnerability to maritime pressure, (b) limited capability to inflict it, and (c) the
likelihood of intervention by outsiders, given the international nature of the
maritime environment, all militate against major local naval impact in the
majority of wars between Third World states.
It is conceivable, however, that local navies might play more of a part as
Third World states develop their maritime interest and their forces. One or two
maritime conflicts might escalate into situations where force was used and
replied to on a high enough level, and consistently enough, to be characterized as
war. The main area of potential naval conflict probably lies in South East Asia,
where conflicting areas abound in an area ripe for hydrocarbon exploitation. A
more favourable economic context for these activities could possibly, with other
factors, increase the stakes to the point where one of the non-possessing
claimants felt impelled to use force, as when China seized the Paracels in early
1974: a counter-attack might lead to more sustained hostilities. But even here the
probability of the USA, USSR, and China becoming involved would probably
act to limit the extent of hostilities.
Once outside actors do appear, their impact on a highly sea-dependent
participant in a local war can be very important. The escalating activities of the
US Navy in the Gulf in 1987 and 1988 clearly showed Iran that there was
nothing she could do to protect her vital oil exports should the USA decide to
interdict them. Mining failed to drive US forces away and when the Iranian
Navy and Revolutionary Guards finally gave battle following the retaliatory
attacks on the oil platforms at Sirri and Sassan on 18 April 1988, the result was
the loss of a frigate, a fast-attack craft and two small gunboats, and damage to a
second frigate. A shore-based Silkworm missile was easily decoyed away.47
Iranian naval impotence had been clearly demonstrated. In combination with
other factors – the missile ‘war of the cities’ and reverses on the battlefield –
outside maritime pressure (added to Iraq’s demonstrated capacity to devastate
Iran’s far-away oil terminals) obtained Iranian acceptance of Resolution 598. As
ever, sea power did not work alone, but it must have been a key factor, given the
high stakes. The US Navy was threatening Iran’s very economic survival.
Thus, although they may not actually participate in the fighting, navies can
play a highly significant, if not crucial, role in limited or local wars. Their
presence can act to set the boundaries for a conflict by demonstrating the
willingness of larger powers to intervene if things get out of control. They may
be used to exert pressure on one or both sides to accept a ceasefire, especially in
a protracted conflict in which neither side can clearly win. Similarly, the
presence of one major power’s naval forces may neutralize the activities of the
other by the demonstration of the potential capacity to disrupt any planned
intervention. The paradox of this situation is that the very presence of the
outsider’s naval forces may well be the major factor in preventing the local navy
exerting greater influence on the outcome of the war itself. Military sea power,
therefore, will continue to play a key role in the maintenance of peace and
security, just as the civil uses of the sea will continue to be crucial to the world’s
economic prosperity and well-being. The connection between the two will,
however, be ever more subtle and complex. Can one produce some new theory
to provide a framework for thought in this developing situation? It is to this task
we shall finally turn.
Part VI

Conclusion
Chapter 11

The future of sea power – three


theoretical frameworks

This final chapter will attempt to develop three theoretical frameworks that it is
hoped will assist those who wish to think about sea power over the next few
decades. None are very original: indeed all three are based on existing ideas,
albeit amended in significant ways. The first will cover the foundations of sea
power, why nations will seek to maintain naval forces and other capabilities to
exert power at and from the sea. The second will cover the roles of navies, what
they will seek to achieve at sea, how they will do it and how those tasks will
change in emphasis in the future. The third will try to construct a typology for
navies, putting them into different analytical categories.

The foundations of sea power


Mahan is a name often taken in vain by commentators on sea power; in vain,
because they sometimes reveal a certain lack of knowledge of what he actually
wrote. Some contemporaries, notably Corbett, doubted the American’s depth of
analysis at the time and, as we saw at the outset, his fundamental thesis has been
seriously questioned since.1 Nevertheless, given its seminal nature, Mahan’s
‘Discussion of the Elements of Sea Power’, Chapter 1 of The Influence of Sea
Power Upon History, which provides the basis of his reputation as the leading
analyst of the subject, deserves close examination in the context of developments
since he wrote it and likely development in the future.2
The sea remains, as Mahan wrote, ‘a great highway ... a wide common over
which men may pass in all directions’.3 Despite the growth of other forms of
economic sea use (see Chapter 3) and the concomitant creeping assertion of
greater sovereignty and/or jurisdiction over what were previously high seas (see
Chapter 7), it seems unlikely that the principle of the free passage of merchant
shipping (if not warships) will be significantly constrained. Indeed, the wider
distribution of merchant shipping and the likely increase in world sea-borne
trade will give more and more states an interest in allowing it freedom of transit.
As in Mahan’s day, it will still be easier and cheaper to transport heavy and
bulky items by water.
Where Mahan is on less firm ground, however, is his bland assertion that ‘it is
the wish of every nation that the shipping business should be done by its own
vessels’.4 This is clearly not the case in the contemporary era of economic
liberalism. As we have seen, some countries, notably the United Kingdom, show
surprising carelessness in the fate of their merchant fleets, choosing to pay the
price in balance of payments losses rather than the cost of maintaining ships at
higher than ‘market’ rates through subsidy or other support. It is too early to say
yet whether states such as Britain will appreciate the full implications of this
policy before terminal damage is done to their national capacity to transport
goods by sea in crisis or war. The Government argues that there will be enough
available ships for such circumstances. Others – especially interested parties –
are far less sure and argue for special measures to maintain a core fleet.5 One
suspects that most states will take special measures to protect some minimum
seaborne carrying capacity for a combination of strategic and economic reasons.
Even that bastion of free market economics, the United States, has little
hesitation in doing so and has even made ‘sealift’ a naval priority. It will all
depend, as Mahan himself pointed out, on ‘the character and clear sightedness of
rulers’.6
One crucial factor affecting governmental attitudes towards merchant fleets
will be Mahan’s ‘spirit of the age’,7 the general trend towards or away from
protectionism. The signs are currently ambiguous. Greater protectionism will,
however, imply more measures, in certain countries at least, to protect mercantile
marines from foreign competition. A major break-down of the current liberal
world economic order would probably also reforge some of the broken links
between military navies and mercantile marines. If states are locked in overt and
bitter economic competition, merchant ships may well depend, as they did in the
age of classical mercantilism, on ‘the protection of their country throughout the
voyage’.8 Yet, despite danger signs, the trend, in both the western world and the
Soviet bloc at least, seems to point more firmly in the direction of greater
interdependence and economic liberalism. In this context, as now, reliance for
naval protection will have to be placed on certain states’ more sophisticated
analysis of the need to protect freedom of navigation as an act of enlightened
self-interest for themselves, their allies, and their trading partners, and for the
whole economic system from which they benefit. The states which carry out this
duty may well not have large mercantile marines but their interest in overseas
trade and general good order will be great nevertheless.
That naval power might be powerful globally, regionally, or just locally. The
potential mismatch between flag of merchant ship and flag of warship may, for
the time being, create certain legal difficulties. It may, perhaps more importantly,
prevent the interest in defending a particular trade being perceived. But the
evidence of recent events in the Gulf demonstrate that warships will, eventually,
come to the aid of all merchantmen threatened in a particular area, e.g. by
combatants whose belligerent rights are in dispute. States who are the
beneficiaries of maritime trade – albeit carried in other peoples’ ships – will deny
themselves the capability to do something in defence of that trade at their peril.
In the future multipolar world, it seems likely that the responsibility to protect
shipping in peace and war will be spread more widely. Local powers will
increase in importance, powerful regional actors like India will emerge, and the
growing great powers of Japan and China will probably match and surpass the
existing medium naval powers of Britain and France. Despite all the problems,
the latter will almost certainly eventually combine together with their
Community partners into some kind of loosely constructed European naval
force. It seems unlikely, however, that any of these states will require what
seemed a vital attribute of distant water operations in Mahan’s day: colonies and
foreign bases. Not only are formal colonies now unnecessary to safeguard
economic contacts, but the striking feature of sea power in the latter half of the
twentieth century is its independence of permanent shore bases. The United
States and Soviet Union both have some ‘bases’ but these are forward outposts
for fleets largely maintained from home dockyards by means of afloat support.
Admiral Hill’s vital attribute of ‘reach’ is inherent in any navy willing to invest
in afloat support, tankers, and repair ships.9 Local friends will, in all likelihood,
provide harbours in which such support vessels can reside (as the Gulf states
quietly gave the European navies house room in 1987-8).
Mahan summed up the keys to sea power as ‘production, with the necessity of
exchanging products, shipping, whereby the exchange is carried out, and
colonies which facilitate and enlarge the operations of shipping and tend to
protect it by multiplying points of safety’.10 All these ‘keys’ are now
internationalized to a degree that Mahan would have found surprising and even
shocking. Asia now produces cars, steel, and consumer electronics for the world.
Huge merchant fleets now carry the flags of minor African and Asian states.
Almost all the ex-colonies now have at least national sovereignty and seats at the
United Nations. This maritime background still creates interests that states
perceive as requiring defence by military force deployed on, over, and under the
oceans. But that perception requires and will require even more of the
aforementioned ‘character and clear sightedness’11 from governments than that
required in Mahan’s day.
Mahan argued, deterministically, that the ‘shrewdness and foresight of
governments’ had been less important in determining ‘the history of seaboard
nations’ than certain ‘natural conditions’ that shaped their development. These
‘principal conditions affecting the sea power of nations’ were: ‘I Geographical
Position. II Physical Conformation, including as connected therewith, natural
productions and climate. III Extent of Territory. IV Number of Population. V
Character of the People. VI Character of the Government, including therein the
national institutions.’12 Let us critically examine these in turn from the
perspective of the late twentieth century and beyond.

Geographical Position
Mahan argued that ‘a nation ... so situated that it is neither forced to defend itself
by land nor induced to seek extension of its territory by way of the land . . . has,
by the very unity of its aim directed upon the sea, an advantage as compared to a
people one of whose boundaries is continental’.13
The invention of the aircraft and the rocket, which Mahan could perhaps not
have foreseen back in 1890, have enforced strategic unities between continental
land masses and offshore islands that have done much to undermine this basic
Mahanian tenet. Britain, for example, can ignore the occupants of the coast
twenty miles across the Channel less than ever. As one of the wisest First Sea
Lords put it in relation to the commitment of land and air forces to the continent
of Europe in March 1950, a Britain facing a hostile continent in modern
conditions would be ‘menaced as never before in our history’.14 Nations,
however insular, require defences and/or deterrents against air attack (which
might, of course, be seaborne), and often armies to keep the enemy as far away
as possible, at least as much as they require navies for traditional tasks.
Geography is now much less of a defence from the need to weigh strategic
priorities. The world is in many ways a much smaller one. Few, if any, states,
therefore, can now emphasize their navies at the expense of other forces because
of mere geographical position.
Mahan regarded the position of his country on the two oceans as ‘either a
source of great weakness or a cause of enormous expense’.15 In fact it is an
enormous American advantage given an ability to mobilize huge resources. The
United States can deploy its navy easily into either ocean, Atlantic or Pacific,
and has no need to defend itself on land from its weak neighbours. The other
superpower, on the other hand, must deploy its fleets in widely separated areas
behind choke points easily defended by hostile powers and must pay due regard
to the overriding need to protect its land frontiers and maintain order in its
western buffer. Mahan need not have worried.

Physical Conformation
Mahan argued that countries with long seaboards and good harbours tend to be
more maritime than others. Richard Hill’s evidence on sea dependence in
Chapter 8 would seem to bear out this point. It is, of course, not just the
possession of a coast that is important. The Soviet Union has one of the largest
coastlines in the world but most of it is iced up and effectively unusable in terms
of access except at great technological effort. The fact that the Soviet Union’s
harbours are relatively few, and sometimes isolated, is also a source of potential
vulnerability to maritime pressure.
Islands are obviously especially sea dependent: ‘When the sea not only
borders, or surrounds, but also separates a country into two or more parts the
control of it becomes not only desirable but vitally necessary.’16 The level and
extent of that control is, however, a function of other factors.

Extent of Territory
A fundamental weakness of Mahan was that he was unconcerned with ‘total
number of square miles which a country contains’17 in his analysis of a nation’s
potential as a sea power. In this he was almost completely wrong. True,
possession of large tracts of unex-ploited territory on its own is no help, but the
current century was to prove that the extent of territory in terms of masses of
exploitable resources was a key factor in building state power to equally massive
proportions. Mahan clearly felt ill at ease with the continental preoccupations of
his own time and country: he did not realize that his compatriots were laying the
foundations for the largest expansion of military sea power the world has ever
seen. In the modern and future world of more easily exploitable land masses,
large territory often means more economic power if the wherewithal is there to
unlock it. More economic power in turn means more sea power.

Number of Population
By this Mahan meant the number of the population ‘following the sea, or at least
readily available for employment on ship board’,18 and related it to his point
about the need for coasts and harbours. Again he was wrong and for much the
same reasons. The nature of modern navies is such that the overlap with other
forms of maritime activity is relatively small. Military naval officers see
themselves as separate, perhaps increasingly separate, from other sea users. Thus
a large population ‘following the sea’ is not necessarily a prerequisite for an
ability to deploy military power on the oceans. Even in Mahan’s own USA, most
naval officers come from deep inside the American continent.
This is not to say that having little in the way of seafaring population does not
have implications in terms of the reserves that can be called upon to sustain
military navies in crisis, the perceived importance of general maritime activities
(of which military navies are a part) and the general status of a country in
maritime affairs. But the situation is not as Mahan baldly stated it. Large
numbers of suitable people may be important if there is a naval requirement for
them, just as large quantities of any other resource might be significant, but they
must be suitable for naval employment. This has much more to do with the
general skills and education of the population than it does with any physical or
professional association with the sea. Fishermen, or even merchant seamen, may
need as much training as landsmen in order to operate a modern warship. The
overlap between naval skills and other maritime skills is nowadays surprisingly
limited.
Finally, and most importantly of course, the number of population in its
relationship to the general resources and economic development of the state is a
vital factor in determining the resources available for diversion to naval
purposes. Countries with very large populations which they can only just feed
are going to find it difficult to deploy powerful navies. This is, however, not just
a question of diversion of resources into naval channels. It also reflects the
resources available for education and training, and general industrial
development. This leads us on to Mahan’s next point.

National Character
Here Mahan was at his most ‘nineteenth century’ in outlook. He detected a
natural ‘tendency to trade, involving the production of something to trade with’
as ‘the national characteristic most important to the development of sea
power’.19 Our appreciation of economics now goes somewhat further than
instinct, although cultural factors do play a part in economic growth and power.
There seems to be good reason, looking at the comparative performance of the
USA and USSR, to see certain factors beyond the material in the degree of
economic success enjoyed by the two states. Britain and Japan also make
interesting comparisons. However, the analyst is on firmer ground if he sticks to
more easily measured and explained phenomena.
It is probably better to divide Mahan’s old ‘national character’, like Gaul, into
three parts: Economic Strength and Technological Prowess, as well as the more
diffuse Socio-political Culture. The first is measured in GNP, the second in the
quality of the technological output of the state, the third is harder to quantify but
clearly visible in certain notable cases.
Only countries of considerable economic strength can afford to be major sea
powers in the modern world as the cost of naval equipment of all kinds
increases. As Chapter 8 has shown, sea power is a direct function of economic
power. Even a country like Japan, with considerable inhibitions about mobilizing
its enormous economic strength, cannot help but deploy one of the world’s most
significant navies. As touched on above, it is not just a question of absolute
strengths. Countries with low per capita GNPs, like India and China, are
currently relatively weak sea powers for their size. Their sea power will grow as
their economies get richer and they bring their populations under control.
Countries with high per capita GNPs, on the other hand, may already be sea
powers of some importance. They have the resources to invest in warships.
Technological prowess, however, is of equal significance. A well educated,
technologically literate population is essential both to man and fight modern
warships, and produce the equipment that allows them to fight. The modern
warship, or other naval platform, like the ship of the line of the nineteenth
century, is often one of the most sophisticated items of contemporary equipment
made by a nation’s industries. Its weaknesses often reflect the general
technological weaknesses of its owners.
The third factor, socio-political culture, is historically rather than ethnically
based. Certain states have, for various reasons, eschewed traditional naval
expansion. The two classic cases are Japan and the Federal German Republic.
Both have been considerable sea powers in their time and both could be greater
sea powers than they are today if they chose. Japan, especially, is perhaps the
world’s greatest maritime nation and as world number two economically should
be a major naval power. That she is not is a product of psychological and
political inhibitions firmly implanted by the traumas of defeat. The Federal
Republic is in the same state. The movement of its warships to the
Mediterranean in late 1987 to sustain NATO’s strength there was hailed as a very
special event. It was made clear that these ships could go no further ‘out of area’.
West Germany is still not that sort of country.
These inhibitions are fascinating, not least because they are self-imposed. The
United States often thinks of trying to make its allies do more militarily,
although those with a feeling for the past might be grateful for a quiescent Japan
and Germany. As self-imposed limitations they could well change over time,
depending on the external environment. Japan’s self-denial has been part of a
wider unwritten trade-off that allowed her, unlike before 1941, virtually
unlimited penetration of markets and raw material sources in a liberal world
economic order. This condition need not last. If the protectionist factors
mentioned above prevail and Japan was suddenly faced with a set of closed
trading doors the effects might be both surprising and terrifying for those with
short memories.
The Federal Republic is not the only country in Europe to feel inhibitions
about acting navally on a wider stage. The Netherlands has reacted to centuries
of imperial decline by a considerable ambivalence towards foreign
entanglements. Even when her energy supplies were threatened in 1987, the
context for sending naval forces to safeguard them had to be international if
political consensus was to be maintained. This is probably a key to the future: a
greater European rather than national dimension to naval activity might re-
legitimize and modernize uses of naval power at some distance from Europe’s
shores when European interests are threatened. By such means the economic
power and technical prowess of Europe as a whole might be applied more
powerfully and extensively at sea.
A final aspect of this factor relates more closely to Mahan’s original concept,
the effect a nation’s socio-political structure has on the qualities of initiative,
resourcefulness, and independent thought required for naval operations.
Although this can be overstated, as all naval forces must operate in a disciplined,
organized and controlled manner, the stereotyped operations of the Soviet Navy
compared to Western navies has some relationship to the negative and restrictive
aspects of Marxist Leninist society. This brings us to the last of Mahan’s factors.

Character of Government
Mahan doubted ‘whether a democratic government will have the foresight and
the keen sensitiveness to national position and credit, the willingness to insure its
prosperity by adequate outpouring in time of peace, all of which are necessary
for military preparation . . .’20 However, as Philip Pugh has shown, the rise of
democracy in Britain has not seen any decrease in the defence burden in
peacetime. It has fluctuated between minima of 2 per cent and maxima of about
5.5 per cent of GNP in every period of peace from 1698 onwards. Indeed the
figures for 1969-83 are the highest minimum over the three-century period.21
Government policy can foster sea power by fostering general economic
growth, or by specially emphasizing defence in general and the navy in
particular within the normal peacetime margins. This can be a feature of
governments as different in ideology as the Reagan Administration or
Brezhnev’s Politburo. On the other hand, equally right-wing governments in the
West to President Reagan’s may allow a devotion to free market competition to
destroy the ‘industries and sea going bent’22 that gives a country those interests
at sea to defend that Mahan thought to be the only legitimate basis for a nation’s
sea power. Britain under the Thatcher Administration is a classic example. As
military sea power operates nowadays at least semi-autonomously from
‘peaceful commerce’, however, the latter effect may not be decisive in reducing
a nation’s capacity to exert most aspects of military power at sea.
Government policies can affect to a considerable extent a nation’s capacity to
mobilize extra naval assets from the merchant marine, and sustain military
operations at sea by the same means. Nevertheless, it is an open question
whether any resultant reduced perception of ‘maritimeness’ affects first-line
military naval capabilities very much. This author suspects it might not, at least
in the short to medium term. Other factors, for example wider national
perceptions of security interests, usually have a greater impact, for example the
need to contribute to the overall military posture of an Alliance with a
superpower patron. Shipping and shipbuilding are not the only industries with a
security relationship deserving of government support when required and as
appropriate. Their defence relevance is, however, more direct than most.
What are we to make then of Mahan’s determinants? They need both re-
ordering and changing. At the top of any list of elements of sea power must go
Economic Strength. This is the foundation upon which all stands. It
comprehends the old ‘extent of territory’ and ‘number of population’. Large
countries with numerous populations and powerful economies can devote
resources to military sea power, as sheer expressions of that strength, even if they
have little other interest in using the sea. The economic strength of smaller states
also fundamentally affects the level of sea power they can afford, but it is not
just a question of sheer resources. They must be of the right kind, be put into the
right form, and deployed with a certain lack of inhibition.
This leads on to Technical Prowess. A country with considerable economic
strength may not have the capability to build a high technology navy and operate
it effectively. The Soviet Union has had considerable difficulty in building up an
effective fleet. Even after half a century of industrialization, the USSR still lags
behind the USA at sea. Much of its technical development has been derivative,
reliant on technology transfer and even espionage. Although the United States,
and to some extent Western Europe, will retain a lead in naval technology it will
slowly be eroded. More countries are beginning to build warships and even quite
sophisticated naval weapons are being produced by countries that were
previously reliant on the established naval powers for equipment. China now
produces quite sophisticated sea-skimming anti-ship missiles and puts them on
submarines of her own design. Taiwan also produces SSMs that are copies of
Israeli weapons. Brazil is developing her own SSM. The trend towards wider
production of naval technology will continue.
There is an interaction between technological prowess and the next
‘condition’, which is Socio-political Culture. Certain societies and political
systems are better at adapting to technological change than others, but even
when, like Japan, they have demonstrated a quite remarkable adaptability and
prowess they may still significantly limit their investment in the naval
dimensions of security and foreign policy. Others countries may also feel similar
inhibitions. This seems to operate almost independently of the sea dependence of
that state. Equally, countries that are not overly sea dependent can feel the need
to use the seas to express and exert military strength, both offensively and
defensively (although they will probably interpret the former in defensive
terms).
The above three conditions seem to be of a first order. Even large islands with
good harbours are of little naval significance without people, effective ships, or
aircraft. But the second order, more specific, conditions – those affected by
national situation – are none the less important. First comes geographical
position. Land-locked states will tend not to have significant navies, although it
must be admitted that water is such a universal and important transport medium
even within continents that Switzerland, Hungary, and Bolivia all appear in the
pages of Jane’s Fighting Ships.23 As we have seen above, sheer position is of
less significance in today’s smaller world. Other geographical factors are
perhaps best comprehended in Admiral Hill’s wider concept of Sea Dependence
with its various criteria, seaborne trade, merchant marine, shipbuilding, fishing
catch, and off-shore zone.24 Countries with a high degree of sea dependence will
have more reason to deploy coercive power at sea than others, although whether
they actually do so will be affected by the first order conditions.
This leads on to a final second order factor. Character of Government should
be changed to Government Policy and Perception. The way the other factors are
mediated by governments reflects the specific constellation of political forces in
a country at the time. Governments may ignore certain interests and emphasize
others. Governments may be elected or come to power in more violent ways
because of historically specific circumstances. Governments cannot change
naval policy very rapidly, given the long-term nature of naval forces, but they
can affect how it is used in the short term and take decisions that may affect
future sea power assets in significant ways. Government policy and perceptions
thus have an impact of their own on a nation’s sea power that is at least no more
dependent on the others than the rest of the members of this highly
interdependent list of factors.
Our new set of Principal Conditions Affecting the Sea Power of Nations thus
becomes:

First Order 1. Economic Strength


2. Technological Prowess
3. Socio-political culture
Second Order 1. Geographical position
2. Sea dependence in terms of:
(a) Seaborne trade
(b) Merchant marine
(c) Shipbuilding
(d) Fish catch
(e) Offshore zone
3. Government policy and perception

Putting sea power into these terms deals with one of the major problems of
Mahan’s theory. Mahan insisted that ‘purely military sea power’ can only be
built up by an aggressive despot. Otherwise, ‘Like a growth which having no
root’ it ‘soon withers away.’25 The history of American sea power over the last
century would cast considerable doubt on this assertion. A perception of
legitimate interest in how the international politico-economic system should be
ordered, indeed a sense of duty that it had to intervene as the senior member of a
set of like-minded nations, has impelled the largest democracy in the world to
send the world’s most powerful navy to sea. At times it has seemed that a
mercantile marine has been fostered to ‘legitimize’ this military sea power in a
strange process of turning Mahan on his head. Other demands, military security
and the protection of economic, political, or even ideological interests across the
globe, will impel other nations too to use the seas for military purposes.
Attempts somehow to legitimize a military navy in a large national mercantile
marine are usually doomed to failure and are simply unnecessary.
Yet Mahan might have been right, in one way, to draw some attention to the
relationship between the economic uses of the oceans and the military power
deployed upon them. As the economic forms of sea use become ever more
internationalized so the military uses of the sea may have to become more
internationalized too. Perhaps a mismatch is even now arising, one that Mahan in
a sense foresaw. In 1987-8, national navies got themselves into something of a
tangle trying to defend international shipping in the Gulf. It took time for the
situation to sort itself out. Is it too much to hope that as the superpowers become
ever more aware of their common interests, the security of each other, and that of
the other nations of the world, a new attitude to sea power may eventually be
born; one that emphasizes international duties to safeguard an international
maritime community. If it does, Mahan may have been proved right after all, but
in a way that the old imperialist and mercantilist would never have imagined.

The roles of navies


Mahan’s analysis concentrated on the use of the sea, albeit in the limited sense of
a transport medium, and this leads on to a wider theory of the functions of naval
forces. The ‘use of the sea’ was the basis of Ken Booth’s perceptive triangular or
‘trinity’ model published ten years ago in Navies and Foreign Policy.26 Booth
rightly saw the vital factor being ‘the use of the sea’. A state might wish to use
the sea for three purposes: ‘(1) for the passage of goods and people; (2) for the
passage of military force for diplomatic purposes or for use against targets on
land or at sea; and (3) for the exploitation of resources in or under the sea’.27
Booth’s ordering of these priorities reflected the traditional Mahanian emphasis
on shipping. The nature of the current and future strategic and economic world,
however, notably the continued sea-basing of the most secure nuclear deterrent
systems, would advise placing ‘the passage of military force for diplomatic
purposes or for use against targets on land or at sea’ at the head of the list, with
economic shipping next and resources last, but increasing slowly in importance.
Booth then went on to construct his ‘trinity’ of naval functions:

The unity (the one-ness) of the trinity is provided by the use of the sea: this is
the underlying consideration in the whole business of navies. . . . The
character of the trinity is then defined by the three characteristic modes of
action by which navies carry out their purposes: namely the military, the
diplomatic and the policing functions.28

Deciding which category to place activities is trickier than it first seems. For
example Booth, as we have seen in Chapter 9, separates diplomatic from military
at the ‘actual employment of force’.29 Yet on his own definition the latent as
well as the actual application of force comes under the military category.
Equally, the constabulary duties of navies must include some recourse to the
actual use of armed force, albeit limited at all times. This forces one, like Cable,
into a distinction between higher and lower level operations, peace and war. No
such definition is truly satisfactory, especially in a world where war is officially
illegal under the UN Charter, and every conflict since 1945 has been undeclared
and justified as ‘self-defence’ under Article 51.
Nevertheless, such a distinction has to be made. Strategic deterrence is not
quite the same thing as gunboat diplomacy or showing the flag. The distinction
must surely be, as Cable argues, at the point where the infliction of damage,
actually carried out or just threatened to one’s opponent, becomes an end in
itself, be it in a limited theatre or more generally.30 When on 2 May 1982 a
provoked Britain, with her forces on the spot severely menaced, altered its rules
of engagement to allow general attacks on Argentine warships, that particular
rubicon was crossed: gunboat diplomacy became war. Clear preparation for such
hostilities or worse, including nuclear city-busting, and the perception of these
preparations in the context of a potential ‘war’ if certain boundaries (physical
and legal) are crossed, is a rather different matter from using the capabilities thus
acquired for specific and limited political or law enforcement purposes, even if
the result of the latter use is the actual killing of someone.
Now, on this basis, we can reconstruct the Booth triangle. At the base go the
‘military’ roles, for which most naval forces are primarily designed. This can be
divided in relatively traditional terms: power projection, sea control, and sea
denial. Naval missions such as strategic deterrence (power projection), coast
defence (sea denial), guerre de course (sea denial), and defending shipping (sea
control), can be subsumed under one or more of these headings. All these
objectives require distinctly military forces; they are concepts of maritime
strategy properly and narrowly defined.
The ‘diplomatic’ role of navies can also be divided, à la Cable, into ‘gunboat
diplomacy’ and ‘showing the flag’ (for the distinction see Chapter 9), and
operate in a context, by definition, of inter-state diplomacy, be it on a unilateral
or multilateral basis. If the context is some universally agreed order in which
‘enforcement’ is considered to be legitimate then one moves into the
constabulary role (preferred to Booth’s ‘policing’). Currently this role is almost
entirely national. International constabulary roles await the development of
forms of enforceable and mutually agreed international law and order similar to
those claimed by sovereign states over their own territories, territorial seas, and
(within closely defined but expanding boundaries) exclusive economic zones.
Figure 11.1 shows the ‘use of the sea’ triangle.
Figure 11.1 The ‘use of the sea’ triangle I

Before leaving this, perhaps a few more words are necessary on ‘sea control’.
All that is meant by this term is the ability, when required, to deploy armed force
and/or fight a maritime campaign to enable one to pass shipping defined in the
widest possible sense for a specific purpose of one’s own. By definition the
expected opposition is serious enough to put that shipping seriously at risk. Sea
control operations are often a prerequisite of some forms of power projection but
certain types of forces may have what might be termed ‘inherent sea control’,
e.g. silent submarines or very powerful warship groups facing limited
opposition. Power can thus be projected over an ‘uncommanded sea’ without
engaging in any greater sea control operations than submerging or providing a
powerful enough escort that can brush weak opposition aside. Technology is,
however, emphasizing the need for powerful close defences around assets at risk.
As platforms get stealthier and weapons more formidable, the emphasis in
military operations at sea is going to swing much more to direct protection of the
form of sea use being contested. As one of the Ministry of Defence’s key
forward thinkers summed it up at the end of 1988, defence will have to be
concentrated much more about the ‘key sea users’, whatever they may be.31
Having moved away from Corbett’s ‘normal’ we now seem to be moving back to
it. (see Chapter 2)
Navies will continue to be used in these roles as they have in the past,
although the balance of the contexts in which they work may change. One could
extend Booth’s triangle, which is already based on contexts, into an interlocking
set of circles.32 For NATO and Warsaw Pact navies, at least, these three circles
would be: (1) East–West confrontation, (2) national interest, and (3) law and
order. If the diameter of each circle is very arbitrarily taken to be (a) military
role, (b) diplomatic role, and (c) constabulary role, we end up with a pattern like
Figure 11.2.
Figure 11.2 The ‘use of the sea’ triangle II

Sometimes a particular naval event finds itself in one circle, sometimes at the
overlap of two, sometimes all three. The precise location of any point is
sometimes more a matter of interpretation than clear definition but the three
circles diagram is interesting in its future implications. If East–West
confrontation is indeed withering away, albeit very slowly and uncertainly, there
may well be significant implications for naval forces. Certain military roles, e.g.
forward power projection against superpower opposition, will be reduced in
importance; others, like gunboat diplomacy and maintenance of good order, may
well become more important. Forces most relevant for high-level operations may
seem less important, especially nuclear-powered attack submarines, fleet
carriers, and sophisticated ASW and AAW ships. Presence, perhaps at some
considerable distance, and the capability to deploy sufficient but limited force to
make one’s mark in a basically ‘peaceful’ context may well become the essence
of naval power. In such activities surface ‘warships’ could be more important
than ever. ‘Sea power’ in a very traditional sense thus may have a secure future
in the longest of terms.

A typology for navies


A significant literature has grown up on how to rank navies in terms of their
overall power and capability.33 This is a difficult task as it must take into account
factors such as the types of forces deployed, the sophistication of their
equipment and level of afloat support as well as mere numbers of vessels. The
categorizations below originate in Morris’s work on Third World navies, as
amended by Haines, but they try to go further to produce a ‘global naval
hierarchy’34 that will form the basis of some speculative remarks on the future
balance of naval power. Note that ‘navy’ includes all those forces capable of
exerting force at sea, not necessarily just those bureaucratically organized into a
‘navy’. Also ‘force projection’ implies a capacity to engage in ‘sea control’ and
‘sea denial’ as well as ‘power projection’.

Rank 1: Major Global Force Projection Navy – Complete


This is a navy capable of carrying out all the military roles of naval forces on a
global scale. Currently, only one nation’s navy at this level of capability
qualifies, the United States, and it will maintain this lead for some time to come.
No other navy is likely to equal the capacity for force projection provided by the
numerous carrier battle groups supplemented by a powerful SSN force. Surface
action groups and convoy escort groups back up the carrier capabilities, while a
sea-based ballistic missile force makes the main contribution to the USA’s
strategic air power. Amphibious and afloat support capacities remain unequalled
and sealift capacity is being safeguarded and enhanced. The lessons of the Gulf
have led to neglected MCM capabilities being improved. It is unlikely that any
rival will seriously approach the United States in its ability to deploy sea power
by moving into the first rank: the main danger to US hegemony is that she will
decline in relative power to Rank 2 in which there may well be other members.

Rank 2: Major Global Force Projection Navy – Partial


Currently there is only one nation at Rank 2: the USSR. Although she deploys
powerful SSBN forces and highly capable sea denial forces of oceanic range in
the shape of submarines and shore-based strike aviation, the USSR does not
have the ocean-going sea control and power projection assets to dominate the
local naval environment far from her shores. Her cruiser-based forces will not
provide her with anything approaching US capabilities in this regard. A limited
number of specialist vessels and state-controlled merchant shipping provide
supplementary amphibious and sealift capacity that can carry out limited power
projection duties, but they do not approach America’s specialized amphibious
fleet in scale or sophistication. Current trends seem to imply a reversion to a
priority for adjacent force projection, but the USSR will not lose a limited
capacity to be a significant naval actor on a global scale if she so chooses. As
other nations approach the USSR’s level of economic strength the latter may find
herself with rivals at this level. The USA may also be reduced to it in the long
term if she faces serious economic difficulties.

Rank 3: Medium Global Force Projection Navy


Currently there are only two candidates for the Rank 3 category: Britain and
France. Both possess, to varying degrees, carrier and amphibious capabilities,
sea control surface forces, SSNs, SSBNs, and afloat support. Each would be
capable of conducting one major ‘out of area’ operation and each (Britain
currently much more than France) would be capable of engaging in high-level
naval operations in closer ocean areas. Other nations that might move into this
category (in the medium to long term) are Japan and China. Both may eventually
move through this level into Rank 2, but not for several decades. On the same
time-scale Britain and France may draw closer together in a more united Europe
and with the other members form the basis of a Rank 2 navy. If they do not move
forward together it is conceivable that the relative decline of British and French
capabilities in relation to those of the rest of the world might continue and one,
or both, may be reduced to Rank 4.
Rank 4: Medium Regional Force Projection Navy
A Rank 4 navy was defined by Morris as possessing the ‘ability to project force
into the adjoining ocean basin’.35 The three Asian members of this group
currently are India, Japan, and China. In Europe, Italy, The Netherlands, West
Germany, Spain, and even Belgium qualify given their overall maritime
capabilities, including land-based aircraft and/or the proven ability to operate as
integral parts of NATO’s or the WEU’s maritime forces at some distance from
their own shores. Canada qualifies currently for the same reasons – her ships are
often seen in European waters. Her possible purchase of nuclear submarines
would, however, have confirmed her status at the top end of Rank 4. The same is
true for India which will probably remain a dominant regional power even if
China and Japan eventually pursue more global ambitions. In the southern seas,
Australia is clearly a Rank 4 navy with a capacity to deploy usable and useful
maritime power far into the Pacific. South America has two Rank 4 powers,
Brazil and Argentina, each of which currently possesses carriers. Brazil will
probably be the first Latin American nation to acquire SSNs.

Rank 5: Adjacent Force Projection Navies


Rank 5 navies have ‘some ability to project force well offshore’.36 They include
NATO’s Portugal, Greece, and Turkey and the two major non-carrier South
American navies, Chile and Peru. Israel, although superficially a territorial
defence navy, has often displayed a surprising capacity to strike far from her
shores, and South Africa has some capacity to do the same. The two Koreas have
navies that are moving into this category, while Taiwan also should not be
underestimated with her new submarines and impressive surface fleet. Pakistan
and New Zealand both possess potentially long-legged fleets, while the three
major Gulf navies – Iran, Iraq, and Saudi Arabia – should soon be in this
category if they exploit the equipment they have purchased. None of the above,
however, will be capable of carrying out high-level naval operations over
oceanic distances.

Rank 6: Offshore Territorial Defence Navies


Rank 6 navies have relatively high levels of capability in defensive (and
constabulary) operations up to about 200 miles from their shores. They include
the navies of most of the remaining European nations: Norway, Denmark and
Sweden, East Germany, Poland, Romania, Bulgaria, and Yugoslavia. Most North
African navies are in this category: Algeria, Morocco, Libya, and Egypt. Nigeria
is the only West African navy at this relatively high level by African standards,
and the mid-level Latin American navies – Cuba, Colombia, Ecuador, and
Venezuela – all qualify. In Asia, Bangladesh, Indonesia, Malaysia, and Thailand
are clearly in this category, and the Philippines just qualifies. All these navies
have the sustainability offered by frigate or large corvette vessels and/or a
capable submarine force.

Rank 7: Inshore Territorial Defence Navies


Rank 7 navies have ‘primarily inshore territorial defence capabilities’.37 This is
interpreted here to mean navies capable of coastal combat rather than
constabulary duties alone. This implies missile-armed, fast-attack craft
(sometimes backed up by short-range strike aviation) and sometimes a limited
submarine force. These navies include Albania, Angola, Bahrain, Brunei,
Cameroon, Ethiopia, Finland, Gabon, Kenya, Kuwait, North Yemen, Oman,
Qatar, Somalia, Singapore, South Yemen, Syria, and Tunisia. Some of these
navies, e.g. Kuwait, are trying to acquire more comprehensive capabilities, e.g.
MCM vessels.

Rank 8: Constabulary Navies


The Rank 8 category comprises significant navies that are not intended to fight
but to act purely in the constabulary role. They do not normally have much
beyond guns as armament. The United States, Argentina, Canada, Japan,
Norway, India, and Pakistan deploy such forces in addition to their more
combatant navy. Mexico’s large navy is effectively a maritime police force, as
are the Dominican Republic’s and Uruguay’s rather less impressive forces.
Iceland and Ireland have small but highly effective constabulary navies. In Asia,
Burma’s and Sri Lanka’s navies have few capabilities beyond a policing
function, while in Africa, Ghana and Tanzania have similarly limited forces.
Some of these latter navies may acquire more combatant capabilities and move
into Rank 7, but that will depend on the perceived level of threat.

Rank 9: Token Navies


The ‘token navy’ category covers the rest of the world’s navies. These states, the
world’s smallest and weakest, have some minimal naval capability but this often
consists of ‘a formal organisational structure and a few coastal craft but little
else’.38 These relatively poor countries cannot aspire to anything but the most
limited constabulary functions and it is unlikely they will develop into anything
more.

Conclusion
The conclusions to be drawn from this typology of navies seem to be that one
might expect the naval balance to change more towards the top than further
down the scale. The future hierarchy of the world’s navies may well see four or
five global navies, USA, USSR, Japan, China, and ‘Europe’, each with more
limited capabilities than the current US Navy. Under these will come the
significant regional navies of India, the Americas, and Australia; then will come
the lesser African and Asian navies, and the two mavericks of Israel and South
Africa; then will come the longer-ranged and the shorter-ranged coast defence
navies, the coastguards; and finally the mass of token navies.
Whatever the pecking order, however, sea power will still be more than a
mere slogan. It will be a vital factor in the world political order. Countries will
have good reason to care about what goes on at sea and they will want, within
their means, to have some way of exerting some level of force there. Maritime
forces will continue to absorb large amounts of resources, depending on the
capacity of the nation to invest them and its perception of the importance of the
various uses of the sea, military and civilian, to its overall policy. Certain
countries may choose to dismantle their maritime capabilities but others will
build them up to compensate. Both seaborne transport and seaborne military
power projection will remain of key importance. There will be plenty of scope
for the threat or use of force at or from the sea. Sea power, in short, has a sound
and secure future.
Notes and references
Chapter 1 Sea power in the modern world
1. A.T. Mahan, The Influence of Sea Power upon History 1660-1783 (Boston: Little Brown, 1890),
p. 1. Page references are from the 1965 Methuen, University Paperbacks facsimile edition.
2. Ibid., p. 26.
3. ‘General Council of British Shipping, Statistical Brief, First Quarter 1987’, Table 3.
4. Jane’s Fighting Ships 1987-88, pp. 343, 407.
5. Jean Labayle Couhat (ed.), Combat Fleets of the World 1986-87 (Annapolis, Md: Naval Institute
Press), p. 283.
6. Figures kindly provided by R.S. Woolrych of the British Maritime Charitable Foundation.
7. P.M. Kennedy, The Rise and Fall of British Naval Mastery (2nd edition) (London: Macmillan,
1983).
8. H.J. Mackinder, Britain and the British Seas (Oxford: Oxford University Press, 1925), p. 358.
9. ‘The Geographical Pivot of History’ presented to the Society on 25 January 1904, Geographical
Journal, Vol. 23, No. 4, April 1904, pp. 420-1.
10. Kennedy, op. cit., p. 183.
11. Geographical Journal, Vol. 23, No. 4, p. 441.
12. This point was and is often made by Professor Bryan McL. Ranft, late of the Royal Naval
College, Greenwich and King’s College, London.
13. A.T. Mahan, The Influence of Sea Power upon the French Revolution and Empire 1793-1812
(Boston: Little Brown, 1892), Vol. 2, p. 118.
14. Sir Julian Corbett, Some Principles of Maritime Strategy (London: Longmans, 1911). Quotations
come from the new 1988 edition edited by E.J. Grove (Annapolis: US Naval Institute; London:
Brassey’s, 1988), in this case from p. 15.
15. Lord Esher, Journals, 14 August 1914.
16. Admiralty message of 23 November 1918 quoted in D. Vander Vat, The Grand Scuttle (Leith:
Waterfront, 1986), p. 122.
17. For a stimulating account of the British Empire’s victory in 1918 see J. Terraine, To Win a War
(London: Sidgwick and Jackson, 1978).
18. Most recently in R. Hough’s The Great War at Sea (Oxford: Oxford University Press, 1983).
19. For example, Kennedy, op. cit., pp. 253-4.
20. See M. Olson, ‘The Economics of Wartime Shortage’: A History of British Food Supplies in the
Napoleonic War and in World Wars I and II (Duke University Press, Durham, NC, 1963). This is
an excellent work and deserves greater attention.
21. Kennedy, op. cit., p. 254.
22. See Marder, From the Dreadnought to Scapa Flow, Vol. 4, ‘1917 the Year of Crisis’ (Oxford:
Oxford University Press, 1969).
23. S.W. Roskill, The War at Sea, Vol. III, Part II (London: HMSO, 1961), pp. 369-70.
24. See E. Grove and G. Till, ‘Adjustments in Anglo-American strategy as a consequence of the Cold
War, NATO and nuclear weapons: 1945-67’, in Maritime Strategies of the Great Powers: Anglo-
American Perceptions (London: Macmillan, 1989).
25. For the playing out of this drama in Britain see E. Grove, Vanguard to Trident, British Naval
Policy since World War Two (Annapolis: Naval Institute Press; London: Bodley Head, 1987), pp.
270-7.
26. The Times Atlas of the Oceans (London: Times Books), p. 136.
27. ‘Memorandum for the Transport Committee House of Commons Inquiry into the Decline of the
U.K. Registered Merchant Fleet’, November 1986, British Maritime League, London.
28. F.T. Jane, All the World’s Fighting Ships (London: Sampson Low, Marston and Co., 1898), p. 8.
29. F.T. Jane, Fighting Ships, 1905-6 (London: Sampson Low, Marston and Co., 1905), pp. 15-16,
358.
30. Jane’s Fighting Ships, 1987-88 (London: Jane’s, 1987), p. 105.
31. See The Maritime Strategy, supplement issued with US Naval Institute Proceedings, January
1986, containing article of same name by Admiral J.D. Watkins, the then Chief of Naval
Operations. This brochure was the first official widely published statement of The Maritime
Strategy. It also contained articles by Secretary of the Navy Lehman and the Commandant
General of the Marine Corps. See also ‘The 600 Ship Navy and the Maritime Strategy, Hearings
Before the Seapower and Strategic and Critical Materials Subcommittee of the Committee on
Armed Services, House of Representatives, 99th Congress, First Session, 24 June, 5, 6 and 10
September 1985’ (Washington DC: US Government Printing Office, 1986). For an excellent
survey of the debate stimulated by the Maritime Strategy see the survey drawn up by one of the
Strategy’s main drafters, Captain P.M. Swartz USN, Contemporary U.S. Naval Strategy: A
Bibliography (Annapolis: US Naval Institute, 1987). An abridged summary appeared in
Proceedings February 1987.
32. Notably, N. Friedman, The Post War Naval Revolution (London: Conway, 1986). See also Grove
and Till, op. cit.
Chapter 2 Maritime strategy
1. Sir Julian Corbett, Some Principles of Maritime Strategy (London: Longmans, 1911); and edited
by E.J. Grove (Annapolis: US Naval Institute; London: Brasseys, 1988).
2. F.T. Jane, ‘Under the white ensign’, Evening Standard, 15 January 1912.
3. Corbett, op. cit., p. 9.
4. Ibid., p. 94.
5. Ibid., p. 91.
6. S. Turner, ‘Missions of the U.S. Navy’, Naval War College Review, March-April 1974.
7. Ibid.
8. Corbett, op. cit., p. 77.
9. G. Till, Maritime Strategy and the Nuclear Age (London: Macmillan, 1982), p. 16.
10. Personal experience, and conversations with my ex-colleague at Dartmouth, Evan Davies.
11. For example, see Influence of Sea Power Upon History, p. 24. Mahan could not, however, refrain
from falling into the trap of trying to invest the battle itself with special significance. The result is
an interesting inversion of logic: ‘England was saved at Trafalgar, though the Emperor had then
given up his intended invasion.’ For Corbett’s views, see The Campaign of Trafalgar (London:
Longmans, 1910).
12. For good accounts of these battles see H.R. Willmott, The Barrier and the Javelin (Annapolis:
Naval Institute Press, 1983) and W.D. Dickson, The Battle of the Philippine Sea (Shepperton: Ian
Allan, 1975).
13. S.W.C. Pack, Night Action off Cape Matapan (Shepperton: Ian Allan, 1972), p. 19.
14. N. Friedman, The Postwar Naval Revolution (London: Conway, 1986).
15. The classic exposition of the Maritime Strategy was The Maritime Strategy, supplement to US
Naval Institute Proceedings, especially the article by Admiral J.D. Watkins ‘The Maritime
Strategy’. The best explanation, however, is Norman Friedman’s The U.S. Maritime Strategy
(London: Jane’s, 1988).
16. Bound lecture notes, Corbett Papers, National Maritime Museum.
17. Corbett, Strategical Terms and Definitions Used in Lectures on Naval History, 1906; Corbett
Papers. This is reproduced as an appendix in the 1988 edition of Some Principles of Maritime
Strategy, pp. 324-5.
18. Vice-Admiral Sir Geoffrey Dalton, ‘NATO’s Maritime Strategy’, in E. Ellingsen (ed.) NATO and
U.S. Maritime Strategy (Oslo: The Norwegian Atlantic Committee, 1987), p. 42.
19. See E. Grove, Vanguard to Trident, British Naval Policy since World War Two (Annapolis: Naval
Institute Press; London: Bodley Head, 1987), p. 367.
20. Kennedy, op. cit., p. 79.
21. Lt. Cdr. D. W. Waters, A Study of the Philosophy and Conduct of Maritime War, Historical
Section, Admiralty, revd. ed. 1957, p. 3. A published version of paper appeared in RUSI Journal,
August 1954.
22. See also J. Winton, Convoy, The Defence of Sea Trade (London: Michael Joseph, 1983).
23. Especially in Sea Power in its Relations to the War in 1812 (New York, 1903). For Mahan’s
views on convoy see also R.A. Bowling’s excellent The Negative Influence of Mahan on the
Protection of Shipping in Wartime: The Convoy Controversy in the Twentieth Century, Ph.D.
thesis presented to the Graduate School University of Maine at Orono, May 1980. For a more
easily accessible and recent variant of Bowling’s well argued views see his ‘Keeping Open the
Sea Lanes’, US Naval Institute Proceedings, December 1985, pp. 92-8.
24. Waters, op. cit.
25. Marder, From the Dreadnought to Scapa Flow, Vol. 4; ‘1917 the Year of Crisis’.
26. ‘Defence of Trade’, Naval War Manual 1925, p. 35; Public Record Office reference ADM
186/66.
27. Waters, op. cit., p. 32.
28. Ibid., p. 36.
29. Watkins, op. cit., p. 11.
30. Waters, op. cit., p. 37.
31. Ibid., p. 39.
32. Quoted, ibid., p. 16.
33. Ibid.
34. Letter from Cdr. Waters to Cdr. M. Ranken, 3 October 1981.
35. D.W. Waters, ‘Seamen, Scientists, Historians and Strategy’, Presidential Address, 1978, The
British Journal of the History of Science, Vol. 13, No. 45, 1980 (footnote 29).
36. Ibid., pp. 207-9; I have made one or two small alterations of expression.
37. This formula was produced by Lord Blackett in 1943. Blackett was perhaps the most
intellectually distinguished product of the Royal Naval Colleges at Osborne and Dartmouth
during their most dynamic years before 1914.
38. Ibid., p. 210.
39. E.J. Grove, ‘The reluctant partner: the United States and the introduction and extension of convoy
1917-18’, in S. Palmer and G. Williams (eds), Charted and Uncharted Waters (Proceedings of a
Conference on the Study of British Maritime History; National Maritime Museum and Queen
Mary College, London, 1981).
40. See Corbett, op. cit., pp. 214-20, for a sympathetic account of Torrington’s activities.
41. Retired Hungarian general, in conversation with the author.
42. For the ‘Jeune Ecole’ and German thought, see T. Ropp, ‘Continental doctrines of sea power’, in
E.M. Earle (ed.), Makers of Modern Strategy (Princeton University Press, 1941), chap 18. For a
more modern assessment, see G. Till (1982) op. cit., pp. 34-8, 49-53.
43. For Soviet thinking, see B. Ranft and G. Till, The Sea in Soviet Strategy (London: Macmillan,
1983).
44. For a good discussion of this matter see G. Till, Modern Sea Power (London: Brasseys, 1987),
pp. 110-22. See also Chapter 4.
45. Andrew Lambert, ‘Maritime Strategy against Russia: An Historical Perspective’, unpublished ms.
46. Ibid.
47. Ibid.
48. Ibid.
49. Ibid.
50. Corbett, op. cit., title of Part II, chap. II.
51. Ibid., p. 107.
52. Ibid., p. 111.
53. Ibid., pp. 123-7.
54. W.H. Garzke and R.O. Dulin, Battleships: Axis and Neutral Battleships in World War II
(Annapolis: Naval Institute Press, 1985).
55. For the best discussion of Fisher and the battlecruiser see J. Sumida, In Defence of Naval
Supremacy: Technology, Finance and British Naval Policy (London: Allen and Unwin, 1989),
chaps 2, 3 and 5.
56. Fisher, Records (London, 1919), pp. 212-3.
57. Admiral Sir John McGeoch, ‘Command of the sea in the seventies’, (Waverley paper, Edinburgh
University, 1972).
Chapter 3 Economic uses of the sea
1. UN General Assembly Fortieth Session, Report to the Secretary General, ‘Study on the Naval
Arms Race’.
2. British Maritime League – Memorandum for the Transport Committee, House of Commons,
Inquiry into the Decline in the U.K. Registered Merchant Fleet, November 1986, p. 7.
3. Ibid., and the GCBS Statistical Brief, 1st Quarter 1987, p. 8.
4. GCBS Brief.
5. BP Statistical Review of World Energy, June 1987, p. 19.
6. Ibid., and Times Atlas of the Oceans (London: Time Books, 1983), pp. 146-7, 150-1.
7. Why the Ships Went: A Fully Quantified Account of the Clauses and Decline of the U.K.
Registered Merchant Fleet (British Maritime Charitable Foundation and Lloyds of London Press,
1986), p. 52. For tanker and gas carrier development see Carlett, The Ship: The Revolution in
Merchant Shipping 1950-80 (London: HMSO, 1981), pp. 24-30.
8. Why the Ships Went, p. 25; Times Atlas, op. cit., p. 138.
9. Carlett, op. cit., pp. 39-43; Times Atlas, op.cit., p. 138.
10. GCBS Brief: Times Atlas, op. cit., pp. 136-9.
11. Ibid.
12. Carlett, op. cit., p. 9 and note.
13. Ibid., p. 10.
14. Times Atlas, op. cit., p. 139.
15. Ibid.
16. Ibid., p. 13.
17. Ibid.
18. Ibid.
19. A.J. Ambrose, ‘The supercarriers’, Jane’s Merchant Shipping Review, 3rd Year of Issue (London:
Jane’s, 1985).
20. Carlett, op. cit., p. 21; for the new types of vessels; see also A.J. Ambrose, ‘The big ship market –
tankers and bulk carriers’, Jane’s Merchant Shipping Review, First Year of Issue, (London: Jane’s,
1983), pp. 32-41; and ‘British yards set to exploit the future’, Marine Engineer’s Review
Supplement, October 1986, p. 5.
21. Carlett, op. cit., pp. 17-20.
22. A.J. Ambrose, ‘U.S. lines big twelve’, Jane’s Merchant Shipping Review, 3rd Year of Issue
(London: Jane’s, 1985).
23. A.J. Ambrose, ‘The traumas of the TEU transporters’, Jane’s Merchant Shipping Review, 3rd year
of Issue (London: Jane’s, 1985), p. 163. Up-to-date lists of container facilities are also contained
in the annual editions of Jane’s Freight Containers.
24. N. Polmar, Guide to the Soviet Navy, 4th Edition (Annapolis: Naval Institute Press, 1986), p. 463.
25. M. Enquist, ‘The reefer trades in the mid 1980s’, Jane’s Merchant Shipping Review, 3rd Year of
Issue (London: Jane’s, 1985).
26. Carlett, op. cit., pp. 44-5 is a brief history of this type of vessel.
27. A.J. Ambrose, ‘Development in BASH (barge aboard ship) vessels’, Jane’s Merchant Shipping
Review, 2nd Year of Issue (London: Jane’s, 1984), pp. 132-8; Polmar, op. cit., pp. 369-73, 483.
28. G.W. Detlefsen, ‘The German Coaster Fleet’, Jane’s Merchant Shipping Review, 2nd Year of
Issue, pp. 107-12.
29. ‘British Yards’ (see note 20), ibid.
30. Carlett, op. cit., pp. 33-8.
31. UN Conference on Trade and Development, ‘Review of Maritime Transport, 1983’
(TD/B/C.4/266), p. 11.
32. GCBS, Statistical Brief No. 16, pp. 3-4.
33. Ibid.
34. Times Atlas, op. cit., p. 141.
35. Ibid., tables on pp. 141-2.
36. Figures kindly provided by R. Woolrych, the British Maritime Charitable Foundation.
37. Why the Ships Went, pp. 13-15; Polmar, op. cit., pp. 479-80. For a full survey of American
merchant shipping policies see C.H. Whitehurst, Jr., The US Merchant Marine in Search of an
Enduring Maritime Policy (Annapolis: Naval Institute Press, 1983).
38. The Naval Arms Race, p. 7 (see note 1).
39. Britannica Book of the Year 1987, p. 121.
40. The Naval Arms Race, p. 7.
41. Times Atlas, op. cit., p. 97.
42. Ibid.
43. Britannica Book of the Year 1987, p. 121.
44. Ibid.
45. Arms Race, p. 7; Britannica Book of the Year 1986, p. 136; 1987, pp. 121-2.
46. Britannica Book of the Year 1986, ‘Iceland’, p. 499; 1987, pp. 198, 454.
47. Ibid., see also 1986, pp. 137, 223.
48. Times Atlas, op. cit., p. 104.
49. Britannica Book of the Year 1988, p. 723.
50. Times Atlas, op. cit., p. 104.
51. Ibid., pp. 110-17, Arms Race, pp. 8-9; also author’s discussions with W. Ostreng at the Nansen
Institute, Oslo, September 1986.
Chapter 4 The military medium
1. A.T. Mahan, The Influence of Sea Power upon History 1660-1783 (Boston: Little Brown, 1890),
p. 25.
2. J. Sokolsky, ‘Seapower in the nuclear age: Nato as a maritime alliance’, thesis presented to the
Dept. of Government, Harvard University, Cambridge, Mass., August 1984, p. 278.
3. Ibid.
4. Organization of the Joint Chiefs of Staff: Military Posture Fiscal Year 1987, p. 67.
5. Sokolsky, op. cit., p. 268.
6. Mahan, op. cit., p. 28.
7. The Military Balance 1987 (London: IISS, 1988).
8. S. Gorshkov, Navies in War and Peace published as Red Star Rising at Sea (Annapolis: Naval
Institute Press, 1974) and The Seapower of the State (Oxford: Pergamon, 1979).
9. UN General Assembly Fortieth Session, Report to the Secretary General, ‘Study on the Naval
Arms Race’, p. 1.
10. T.A. Gibson, ‘Gallipoli: 1915’, in Merrill L. Bartlett (ed.) Assault from the Sea (Annapolis: Naval
Institute Press, 1983), p. 152.
11. Ibid.
12. Colonel Robert D. Heinl Jr., USMC, ‘The U.S. Marine Corps: author of modern amphibious
warfare’, in Bartlett, op. cit., p. 187.
13. H.G. Von Lehmann, ‘Japanese landing operations in World War II’, in Bartlett, op. cit., pp. 195-
201.
14. K.J. Clifford, Amphibious Warfare Development in Britain and America from 1920-1940
(Laurens NY: Edgewood Inc., 1983), p. 249.
15. Vice-Admiral B.B. Schofield, Operation Neptune (Shepperton: Ian Allan, 1974), pp. 54-5.
16. Ibid., pp. 54-5, 120.
17. S. Roskill, The Strategy of Sea Power (London: Collins, 1962), p. 240.
18. Ibid., p. 243.
19. M. Binkin and J. Record, Where Does the Marine Corps Go from Here (Washington: The
Brookings Institution, 1976), p. 14.
20. Ibid., p. 6; see also R.J. Heinl Jr., ‘Inchon, 1950’, in Bartlett, op. cit., this also covers briefly post-
war opposition to the amphibious mission.
21. See E. Grove, Vanguard to Trident, British Naval Policy since World War Two (Annapolis: Naval
Institute Press; London: Bodley Head, 1987), especially chaps 5, 7, 8 and 10.
22. Captain R.S. Moore USMC, ‘Blitzkrieg from the sea: maneuver, warfare and amphibious
operations’, Naval War College Review, November-December 1983.
23. Discussion of the underwater environment is based on D.C. Daniel, Antisubmarine Warfare and
Superpower Strategic Stability (London: Macmillan for IISS, 1986), chap 2. See also T.
Stefanick, Strategic Antisubmarine Warfare and Naval Strategy (Lexington, Mass.: Lexington
Books, 1987); Rear-Admiral J.R. Hill, Anti-Submarine Warfare, pp. 37-43; and R. Compton Hall,
Submarine Versus Submarine (Newton Abbot: David and Charles, 1988), p. 40. The UEP mine is
mentioned on p. 63. The Times Atlas of the Oceans, pp. 50-1, is also a useful guide to the
underwater environment.
24. Daniel, op. cit., pp. 70-84 gives a good account of non-acoustic submarine indicators; see also
Stefanik, op. cit., Appendix 3, pp. 181-215.
25. Grove, op. cit., p. 291; D. Wettern, The Decline of British Sea Power (London: Jane’s, 1982), p.
292; Navy International, July 1967, p. 244.
26. Grove, op. cit., pp. 358-9.
27. For US naval submarine intelligence activities see J.T. Rachelson and D. Ball, The Ties That Bind
(Boston: Allen and Unwin, 1985), pp. 223-4.
28. Jane’s Fighting Ships 1987-8, pp. 617-20.
29. D. Pearson, KAL007, The Cover Up (New York: Summit Books, 1987), pp. 58-9.
30. The Naval Arms Race, Dept. for Disarmament Affairs, Report of the Secretary General (New
York: United Nations, 1986), p. 66.
31. Ibid., p. 67.
32. Vanguard to Trident, pp. 366-7.
33. L. Martin, The Sea in Modern Strategy (London: Chatto and Windus, 1965), p. 87.
34. Sokolsky, op. cit., p. 138.
35. CICLANTFLT report to OPLAN 2300, quoted in Sokolsky, op. cit., p. 141.
36. Sokolsky, op. cit., p. 142.
37. Sir Julian Corbett, Some Principles of Maritime Strategy (London: Longmans, 1911); and edited
by E.J. Grove (Annapolis: US Naval Institute; London: Brasseys, 1988), p. 12.
Chapter 5 Weapon systems
1. N. Friedman, U.S. Naval Weapons (London: Conway, 1983), p. 24.
2. For submarine operations in the Falklands War see E. Grove, Vanguard to Trident, British Naval
Policy since World War Two (Annapolis: Naval Institute Press; London: Bodley Head, 1987), pp.
357-81 and David Brown, The Royal Navy in the Falklands War (London: Leo Cooper, 1987).
3. R. Scheina, ‘Where were those Argentine subs?’, US Naval Institute Proceedings, March 1984.
4. ‘Plans unveiled for ASUW torpedo’, Jane’s Defence Weekly, 5 April 1986, p. 600.
5. See ‘underwater warfare’ section, Jane’s Weapons Systems 1988-9 (London: Jane’s, 1988).
6. Donald C. Daniel, Anti-Submarine Warfare and Superpower Strategic Stability (London:
Macmillan, 1986), p. 45. For Soviet torpedoes see also N. Polmar, Guide to the Soviet Navy
(Annapolis: Naval Institute Press, 1986), p. 435.
7. For decoys see R.J.L. Dicker, ‘Decoying the homing torpedo’, International Defense Review,
2/85; for anti-torpedo torpedoes see J. Hutsebaut, ‘Still no cure for the torpedo’, United States
Naval Institute, Proceedings, October 1985, p. 147.
8. See diagrams in Aviation Week and Space Technology, 2 February 1976 pp. 12, 15.
9. Polmar, op. cit., p. 431.
10. E. Briffa and G. Williams gave a good description of Exocet’s modus operandi in their television
programme ‘In the Wake of HMS Sheffield’ broadcast twice on BBC 2 in 1986. For the problems
of long-range engagement, see N. Friedman op. cit., p. 231.
11. Briffa and Williams, op. cit; see also Jane’s Weapons Systems 1988-89 (London: Jane’s, 1988).
12. For ISAR see Friedman, The U.S. Maritime Strategy, p. 148.
13. M. MccGwire, Military Objectives in Soviet Foreign Policy (Washington: Brookings Institution,
1987), p. 411; Aviation Week and Space Technology published a mission profile of SS-NX-13 on
p. 14 of the 2 February 1976 issue.
14. For novel British weapons see L. Marriott, ‘Flights of fancy’, Warship World, Spring 1987, pp.
10-11; N. Friedman, op. cit., pp. 118-19 for Petrel.
15. Wallop Systems Ltd., Barricade Naval Decoy Systems; Marconi Defence Systems, Siren and
Towed Decoy.
16. Brown, op. cit., pp. 194-5. Even retarded bombs need a minimum height to arm and allow the
aircraft to get clear.
17. For analyses of mine warfare and mine countermeasures see C. Rouarch, ‘The naval mine: as
effective a weapon as ever’, International Defense Review, September 1984; M. Vego, ‘Soviet
mines and mine warfare’, Navy International, July 1986; Navy International, December 1986;
H.J. Bangen, ‘The impact of modern technology on the performance of the sea mine’, Naval
Forces, Vol. VI, No. I, 1985; Captain C.C.M. de Nooijer, ‘Countering the mine threat’, Naval
Forces, Vol VI, No. IV, 1985; S.E. Fawson, ‘Mine warfare – NATO versus the Warsaw Pact’,
Naval Forces, Vol. VII, No. V, 1986; R.J.L. Dicker, ‘Mine warfare now and in the 1990s’,
International Defense Review, No. 3, 1986; for a historical perspective see G.K. Hartman,
Weapons that Wait, (Annapolis: Naval Institute Press, 1979). For ARMS see Advance Remote
Minehunting System (Plessey brochure, 1987); for SAM see, MCM News (Malmo:
Arbetsgruppen, 1988).
18. Captain B.R. Longwirth, ‘The lightweight torpedo’ Naval Forces, Vol. VI, No. VI, 1985, p. 94-5.
19. Ibid., p. 21.
20. House of Commons, Twenty-eighth Report from the Committee of Public Accounts, Session
1984-85, The Torpedo Programme HC391; E.R. Hooton, ‘The British torpedo: on track at last’,
Military Technology, Vol. IX, Issue 7, 1986.
21. Hooton, op. cit., pp. 92-4; ‘Anti-submarine warfare – the torpedo’, Navy International, May
1987, pp. 274-83.
22. Ibid.
23. See Jane’s Weapons Systems for performance data on the Mk 46.
24. Longwirth, op. cit.
25. ‘Anti-submarine warfare – the torpedo’, Navy International, May 1987.
26. Polmar, op. cit., pp. 435-6.
27. For British decoys see R.J.L. Dicker, ‘Avoiding the homing torpedo’, op. cit.; for US equipment
see International Defense Review, June 1987, p. 811. For double hulls and torpedoes see N.
Friedman, ‘A survey of western ASW in 1985’, International Defense Review, No. 10, 1985, p.
1597.
28. See Jane’s Weapon Systems for details of these systems; and N. Polmar, Ships and Aircraft of the
U.S. Fleet (Annapolis: US Naval Institute, 1987) p. 474 for Sea Lance.
29. Ibid., see also Compton Hall, Submarine Versus Submarine, pp. 61-3 and the brochure Crusader
Advanced Sea Mine (British Aerospace, 1988).
30. Brown, op. cit., pp. 102, 137.
31. See Jane’s Weapon Systems for a useful annual survey of these devices.
32. ‘A new lightweight ASW system against submarine intruders in shallow coastal waters’, paper
presented to the Royal Institution of Naval Architects Conference, Anti-Submarine Warfare, May
1987.
33. Polmar, op. cit., pp. 409-11.
34. N. Friedman, ‘A survey of western ASW’, International Defense Review, No. 10, 1985; also D.
Wettern, ‘Active sonars and warship design’, Navy International, May 1987.
35. Friedman, op. cit., and The Postwar Naval Revolution (London: Conway, 1986) for a discussion
of this change. Friedman summarizes early towed array developments in ‘The evolution of towed
array sonar systems’, Naval Forces, Vol. IV, No. V, 1983. See also his The US Maritime Strategy
(London: Jane’s, 1988) for more useful information on SOSUS and its operational exploitation:
FDS is mentioned on p. 129.
36. N. Friedman, ‘Sonar and sensing in the far deep’, NATO’s Sixteen Nations, July 1987, p. 79.
37. P. Sabathe, ‘ASW sonars for surface ships’, International Defense Review, No. 10, 1985, p. 1603.
For the combined ‘active towed array sonar’ see ATAS (British Aerospace brochure, 1988).
38. Friedman (1987), op. cit.; presentation by Commodore R. Cobbold, ‘Future Maritime Means and
Needs’, Royal United Services Institute, 16 November 1988, quotation from C.C.F. Adcock,
Plessey Naval Systems, Some Aspects of Bistatic Sonar, Undersea Defence Technology
Conference/Exhibition, October, 1988.
39. Friedman (1987), op. cit., and HELRAS (British Aerospace brochure).
40. E.R. Hooton, ‘Sonobuoys’, Naval Forces, Vol. V, No. VI, 1984; and G.S. Sundaram, ‘Sonobuoys
and dipping sonars: remote sensors for airborne ASW’ International Defense Review, March
1983; M. Witt, ‘Sonobuoys’, Navy International, May 1987.
41. For an excellent description of ASW in all its facets see Rear-Admiral J.R. Hill, Anti-Submarine
Warfare (Shepperton: Ian Allan, 1984).
42. Friedman (1983), op. cit., pp. 24, 168; Brown, op. cit., p. 256.
43. For ‘Aegis’ see Captain Joseph L. McClane and Commander J.L. McClane, ‘The Ticonderoga
story: Aegis works’, US Naval Institute Proceedings, May 1985; also Friedman (1983), op. cit.,
pp. 158-9; N. Polmar, The Ships and Aircraft of the U.S. Fleet, pp. 506-7, 479, 112-16, 139-140.
44. Jane’s Defence Weekly, 19 April 1986, pp. 702-3; 3 October 1987, p. 740; 17 October 1987, p.
864; for Aster see Maritime Defence, October 1988. pp. 381-2.
45. Lt. Cdr. W.A. Weronkoo in US Naval Institute Proceedings, November 1984, p. 183.
46. Polmar, op, cit., pp. 424, 448 for details of SA-N-6 missile and radar. For Baku see Jane’s
Defence Weekly, October 1988, p. 893.
47. Brown, op. cit., p. 222; for Sea Dart’s successes see Falklands, the Air War (British Aviation
Research Group, 1986); and E.J. Grove, Vanguard to Trident, British Naval Policy since World
War Two (Annapolis: Naval Institute Press; London: Bodley Head, 1987).
48. Friedman (1983), op. cit., pp. 151-2.
49. As in Coventry’s loss; Brown, op. cit.
50. See Jane’s Weapon Systems for Sea Wolf details and variants.
51. Friedman (1983), op. cit., p. 182.
52. J. Ethell and A. Price emphasize this point in their Air War South Atlantic, (London: Sidgwick
and Jackson, 1983).
53. See map on p. 13 of The Maritime Strategy, supplement to US Naval Institute Proceedings,
January 1986. See also N. Friedman, Carrier Air Power, (London: Conway, 1981), chap 7.
54. The Maritime Strategy, op. cit. For an excellent overview of the whole air defence problem see
J.R. Hill, Air Defence At Sea (Shepperton: Ian Allan, 1988).
55. For a Soviet example see US Department of Defense, Soviet Military Power (Washington:
Government Printing Office, 1987), p. 113.
56. For example see S. Gorshkov, The Seapower of the State (Oxford: Penguin, 1979), chap IV.
57. For the potential of battleships for shore bombardment and their limitations see E.J. Grove, ‘U.S.
Navy Battleship Reactivation – A Commentary’, Naval Forces, Vol. VIII, No. III, 1987, pp. 85-6.
58. Friedman (1983), op. cit., p. 22; the new British 155 mm naval gun was displayed by Royal
Ordnance at the Royal Naval Equipment Exhibition in 1987.
59. R.E. Gottemoeller, ‘The modern sea launched cruise missile’, Naval Forces, Vol. VIII, No. IV,
1987 is a good recent survey of the cruise missile scene.
60. Mike MccGwire, Military Objectives in Soviet Foreign Policy (Washington: The Brookings
Institution, 1987), pp. 501-2.
61. J. Sumida, In Defence of Naval Supremacy: Technology, Finance and British Naval Policy
(London: Allen and Unwin, 1989), Epilogue and Conclusions.
62. For an excellent explanation of these developments see Captain F.D. Stanley, ‘Action Information
Organisation’, Navy International, December 1985; and N. Friedman, US Destroyers (Annapolis:
Naval Institute Press, 1982), especially chap. 8.
63. Admiral Sir Arthur Hezlet, The Electron and Sea Power (London: Peter Davies, 1975), p. 284.
64. See editorial in Warship World, Vol. 1, No. 10, Spring 1987.
65. Stanley, op. cit., p. 753.
66. See ‘Command and Control’ survey, Navy International, January 1987.
67. Ibid., p. 137.
68. Commodore R. Cobbold to Royal United Services Institute, 16 November 1988.
69. Stanley, op. cit., p. 754.
70. For the range of US naval nuclear weapons see Friedman (1983), op. cit., pp. 278-80.
71. R.B. Berman and J.C. Baker, Soviet Strategic Forces, Requirements and Responses, (Washington:
The Brookings Institution), pp. 54-5.
72. D. Campbell, ‘Too few bombs to go round’, New Statesman, 29 November 1985, describes the
British position.
73. Gottemoeller, op. cit., for discussion of TLAM/N and the difficulties it poses.
74. P. Langereux, ‘France’s ASMP nuclear cruise missile operational’, Military Technology, vol. IX,
Issue 7.
75. As the 1985 Defence White Paper (Cmnd. 9430-1) put it: ‘Free-fall nuclear bombs can also be
delivered by the Sea Harriers of the squadrons in service with the Royal Navy’s “Invincible”
class or aircraft carriers.’
76. This targeting policy was revealed by J. Barry in the television programme ‘Britain’s bomb – the
secret story’ first broadcast 6 April 1986, Channel 4.
77. For good descriptions of Polaris and Trident see Friedman (1983), op. cit., pp. 220-5.
Chapter 6 Platforms
The main reference for this chapter is Jane’s Fighting Ships 1987-88, and
Combat Fleets of the World 1988-9.
1. See Chapter 3.
2. For a set of novel ideas on ships of the future see R. Schonknecht, J. Lusch, M. Schelzel and H.
Obenaus, Ships and Shipping of Tomorrow (Berlin: Verlag Technik; Hounslow: McGregor
Publications, 1983).
3. Ibid., pp. 120, 122.
4. J.R. Hill, Maritime Strategy for Medium Powers (London: Croom Helm, 1986), pp. 158-64.
5. H. Moineville, Naval Warfare Today and Tomorrow (Oxford: Basil Blackwell, 1983), p. 63.
6. For a survey of offshore patrol vessels see International Defense Review, April 1985, pp. 483-
509.
7. For a good survey see C.E.R. Borgenstarn, ‘Fast attack craft’, Naval Forces, Vol. VII, No. II,
1986, pp. 69-77 and Supplement ‘Fast naval combat craft’, International Defense Review,
January 1985.
8. Aviation Week and Space Technology, 31 March 1986.
9. For the High/Low mix concept see N. Friedman, ‘Elmo Russell Zumwalt Jr.’ in R.W. Love Jr.,
The Chiefs of Naval Operations (Annapolis: Naval Institute Press, 1980), pp. 371-2.
10. See E.J. Grove, Vanguard to Trident, British Naval Policy since World War Two (Annapolis:
Naval Institute Press; London: Bodley Head, 1987), pp. 204, 217-18.
11. A.H. Lind, ‘The re-emergence of the corvette’, Naval Forces, Vol. VI, No. V, pp. 54-65.
12. N. Friedman, US Cruisers (Annapolis: Naval Institute Press, 1984), pp. 419-22.
13. For an analysis of battleship re-activation see E.J. Grove, ‘U.S. Navy Battleship Reactivation – A
Commentary’, Naval Forces, vol. VII, No. III, 1987, pp. 84-90.
14. Grove, Vanguard to Trident, pp. 343-4.
15. For an excellent and authoritative description of the Type 23’s evolution, and a survey of the
Royal Navy’s procurement practices, see Sir Lindsay Bryson, ‘The procurement of a warship’,
The Naval Architect, January 1985.
16. For a debate over the best role for these ships and a good indication of how uneasily such sea
control vessels lie in an essentially power projection navy see US Naval Institute Proceedings,
June 1983.
17. For a profile of the ‘Leander’ see C.J. Meyer, Leander Class, (Shepper-ton: Ian Allan, 1984).
18. Modernised Leander Frigate, Vosper Thorneycroft sales brochure.
19. J.W. Kehoe and K.S. Brower, ‘Small combatants – the operators choices’, in Small Warships,
supplement to International Defense Review, May 1987; for Greek policy see Jane’s Defence
Weekly, 27 February 1988, p. 335.
20. R.J.L. Dicker, ‘The frigate and corvette export market: still in the doldrums’, International
Defense Review, April 1986, p. 449.
21. Ibid., p. 450.
22. L.J. Lamb, ‘The Taiwan navy: building to a regional role?’, International Defense Review,
November 1985, pp. 1745-52.
23. Lind, op. cit., pp. 54-64 gives a useful account of modern corvettes; the FS1500s are covered in
R.J.L. Dicker, ‘HDWs FS1500: the bargain basement frigate’, International Defense Review,
November 1984.
24. Dicker (1986), op. cit., p. 456.
25. Ibid.
26. Vice Admiral Joseph Metcalf III. ‘Revolution at Sea, US Naval Institute Proceedings, January
1988, p. 37. See also T.M. Keithly, ‘Tomorrow’s surface forces’, US Naval Institute Proceedings,
December 1988, pp. 50-61.
27. For pro ‘short fat’ ship arguments see D.L. Giles, ‘Want of a frigate’, Naval Forces, Vol. V, No.
II, 1984; and N. Ling, ‘The S90 argument’, Jane’s Defence Weekly, 11 October 1986. For the case
against see A. Preston, ‘Warship controversy’, Jane’s Defence Weekly, 12 July 1986, and N.
Friedman, ‘S90: A U.S. Viewpoint’, Jane’s Defence Weekly, 11 October 1986. For the author’s
views at more length see ‘Short and fat or long and thin’, US Naval Institute Proceedings, March
1987. For the quotation see Report of the Inquiry into Hull Forms for Warships. (London:
HMSO, 1988), p. 11.
28. See Michael J. Eames, ‘Future naval surface ships’, and Clark Graham, ‘The modern monohull’,
in Modern Ships and Craft, a special edition of the Naval Engineer’s Journal, Vol. 97, No. 2,
February 1985.
29. Ibid., p. 56.
30. For catamarans in general see T.F. Carreyette, ‘Recent developments in advanced hull types for
naval ships’, Naval Forces, Vol VIII, No. I, 1987. For the specific types see International Defense
Review, February 1987, pp. 199-201 and Maritime Defence, October 1988, p. 377.
31. J. Stebbins, ‘Sea Knife arriving’, US Naval Institute Proceedings, February 1985, pp 113-18.
32. Eames, op. cit., p. 61. See also R.L. Johnston, ‘Hydrofoils’ in the same work. Carrayette, op. cit.,
also contains a useful summary of the characteristics of hydrofoils.
33. D. Lavis, ‘Air cushion craft’, in Modern Ships and Craft.
34. British Hovercraft brochure: BH7 mk20 Multi-Role Amphibious Hovercraft.
35. E.A. Butler, ‘The surface effect ship’, Modern Ships and Craft, p. 200. For SES see also
Carrayette, op. cit.
36. Butler, op. cit., p. 200.
37. R.D. Watkins and P.D. Plumb, ‘SEALS and Cigarettes’, US Naval Institute Proceedings, April
1987, pp. 103-4.
38. Butler, op. cit., p. 209.
39. Ibid., p. 211: this paragraph is based on Butler’s excellent analysis. For supplementary discussion
from a Norwegian perspective see S.G. Oystein Ronning, ‘Needs within the Norwegian
Navy/NATO towards the year 2000’, Maritime Defence, May 1988.
40. Butler, op. cit., p. 211.
41. J.L. Gore, ‘SWATH Ships’, Modern Ships and Craft, p. 94.
42. For an effective critique of SWATH see W.C. Barnes, ‘SWATH: advanced technology or
mythology’, US Naval Institute Proceedings, September 1986, pp. 119-21.
43. Gore, op. cit. For a rebuttal of Barnes see A.C. McClure and R.D. Gaul, ‘Let SWATH be
SWATH’, US Naval Institute Proceedings, April 1987, pp. 106-8. For propulsion improvements
see Carreyette, op. cit., p. 77.
44. Gore, op. cit. For a practical SWATH ASW frigate see G.V. Betts, L.D. Ferroro, S.R.
Grzeskoviak, N.A. Macdonald and P.J. Bartlett, ‘Design Study for an Anti-Submarine Warfare
SWATH Frigate’, paper presented to the R.I.N.A. ‘Warship 87’ Symposium, May 1987.
45. S.E. Veazey, ‘New shape in ships’, US Naval Institute Proceedings, p. 40.
46. Amended from an original illustrating Eames, op. cit., p. 59.
47. CAP and YARD, Project Capability, 1987.
48. ‘Naval propulsion’, International Defense Review, 8/1985, pp. 1273-94 is a good survey of
propulsion matters. See also D.J. Gates, Surface Warships (London: Brasseys, 1987), chap. 5, and
Keithly, op. cit., p. 57.
49. Babcock, GEC and Weirs brochure, ‘Modern steam propulsion for warships’, Section 1, p. 11.
50. John Lehman, Aircraft Carriers: The Real Choices, Washington Papers, Vol. VI, No. 5,
Georgetown University, Washington DC, (Beverley Hills and London: Sage Publications, 1978).
51. Ibid., pp. 39-40.
52. Ibid., p. 40.
53. Private information from Soviet sources. See also Jane’s Defence Weekly 19 April 1986, p. 702
and 27 August 1988, p. 388; and ‘New Soviet carrier more like a Kiev than a Nimitz’,
International Defense Review, March 1987. Mike MccGwire, Military Objectives in Soviet
Foreign Policy (Washington: 1987, The Brookings Institution), pp. 441-2, who argues that
changing the design from a ‘pure’ carrier to such a ‘universal ship’ allowed the ship to be
transferred to a Black Sea yard from Severodvinsk in the north.
54. For a description of Argus see Harland and Wolffe’s brochure, RFA Argus: Aviation Training
Ship. See also D. Wettern, ‘First Report: RFA Argus’, in Warship World, Vol. I, No. 7, Summer
1986, and K. Walsh and N.A.J. Wells, ‘RFA Argus: The conversion of a Ro-Ro container vessel
to an aviation training ship’, paper presented to RINA Conference, Merchant Ships to War, 1987.
55. N. Polmar, Ships and Aircraft of the US Fleet (Annapolis: Naval Institute Press, 1984), pp. 80,
101. For a critique of these ships see Lehman, op. cit., pp. 59-68.
56. See D. Brown, ‘The MILCOM vessel – a philosophical view’, and N.P. Simpson, ‘Merchant
ships to warships’, papers prepared for the RINA Conference Merchant Ships to War, 19 October
1987.
57. Brochure, Sea Containers Ltd: Aviation Support Ship: The Low Cost Warship.
58. P.D. Pound, ‘Naval auxiliaries: a multi-role naval capability from a merchant hull’, paper
presented to RINA Conference, Merchant Ships to War, 1987. Also brochure, British
Shipbuilders, MRNA.
59. For an interesting discussion of protecting a modern warship see J.R.L. Dicker. ‘Survivability and
the modular frigate – new developments from Blohm and Voss’, International Defense Review,
July 1986, pp. 961-4. See also E. Rohkamm, ‘Are armoured frigates feasible?’, Naval Forces,
Vol. VI, No. II, 1985.
60. W.T.T. Pakenham, ‘The command and control of submarine operations’, Naval Forces, Vol. VI,
No. II, 1985; see also R. Compton Hall, Submarine Versus Submarine: the tactics and technology
of underwater warfare (Newton Abbott: David & Charles, 1988), pp. 47-50.
61. For the basics of blue-green laser technology see J. Starkey, ‘The renaissance in submarine
communications’, Military Electronics, April, 1981. For a useful update and the figures on ELF
data see Jane’s Defence Weekly, 14 November 1987, p. 1097. See also N. Friedman in the Naval
Review issue of United States Naval Institute Proceedings, May 1987, p. 97.
62. For modern submarine design see R.J. Daniel, ‘Submarine design’, Naval Forces, Vol. VIII, No.
II, 1987 and N. Friedman, ‘The U.S. next generation attack submarine’, International Defense
Review, March 1984. J.E. Moore and R. Compton-Hall, Submarine Warfare Today and Tomorrow
(London: M. Joseph, 1986) gives a stimulating and idiosyncratic survey of SSN developments.
For Soviet improvements see also J.E. Moore, ‘Soviet submarine technology forges ahead’,
Jane’s Naval Review, 6th Year of Issue, pp. 84-8 and ‘Soviets score silent success in undersea
race with US’, Washington Post, 17 July 1987.
63. For pump jet references see Combat Fleets of the World, 1986-87 (Annapolis: Naval Institute
Press), pp. 185, 618.
64. The 600 Ship Navy and the Maritime Strategy, Hearing Before the Strategic and Critical
Materials Sub-Committee of the Committee on Armed Services, House of Representatives,
Ninety Ninth Congress, First Session, p. 163.
65. For the rebuttal see ibid., for the assertion see Moore and Compton Hall, op. cit., p. 153, and
Moore, op. cit. See also Compton Hall, op. cit., p. 56.
66. See N. Friedman in US Naval Institute Proceedings, August 1987, p. 133: also Jane’s Defence
Weekly, 17 October 1987, p. 853, 30 January 1988, and 13 February 1988.
67. Jane’s Defence Weekly, 19 December 1987, pp. 1399, 1445; 30 January 1988, p. 152.
68. N. Friedman, ‘Conventional submarine developments’, Naval Forces, Vol. VI, No. VI, 1985. For
good surveys see ‘Conventional submarines and their combat systems’, International Defense
Review, July 1987, pp. 927-34 and the March 1988 edition of Maritime Defence. The Stirling
system was covered well in a supplement to Naval Forces, Vol VIII, No. VI, 1987.
69. Combat Fleets of the World, 1986-7 (Annapolis: Naval Institute Press, 1986), p. 119.
70. Britain is perhaps the most notable and long-standing case of conflict over maritime air power.
See G. Till, Air Power and the Royal Navy 1914-45 (London: Jane’s, 1979); and Grove,
Vanguard to Trident. In Italy the navy was only allowed to operate fixed wing aircraft by a
change in the law in 1987. See Jane’s Defence Weekly, 17 October 1987, p. 852.
71. M.A. Libbey, ‘Airships in naval operations – a US evaluation’, Naval Forces Vol. V. No. VI,
1984. N. Polmar, The Ships and Aircraft of the US Fleet, pp. 445-6 contains a useful summary of
U.S. manned airship developments; for problems with the programme see International Defense
Review, June 1987, p. 809.
72. E.D. Cooper and S.M. Shaker, ‘Soaring into the next century, new directions for US Aerospace’,
Journal of Defense and Diplomacy Study No. 4, September 1988, p. 15. This study has also been
drawn upon for later comments in this section. Jane’s Defence Weekly, 19 November 1988, had a
section on the new LRAACA’s engines on p. 1289.
73. See Jane’s Defence Weekly, 17 October 1987, pp. 852 for the latest Harrier developments.
74. Bill Sweetman, Aircraft 2000: The Future of Aerospace Technology (London: The Military Press,
1984), p. 83.
75. For consideration of more advanced designs see ibid., pp. 83-9, and ‘Advanced naval STOVL
aircraft under study’, International Defense Review, January 1983.
76. B. Cooper, ‘The V-22 Osprey tilt rotor, an aircraft for all services’, Navy International, January
1988, pp. 22-5; The V-22 Osprey (Bell-Boeing brochures); for Osprey in the ASW role see also
Flight International, 31 October 1987, p. 29. For suggestions for more advanced VTOL designs,
see Flight International, 17 January 1987, p. 27. Sweetman, op. cit. illustrates a similar Vought
VTOL concept on p. 90.
77. British Aerospace, Sky Hook, 1987, p. 1.
78. Ibid., pp. 14-15, 18-19.
79. See note 73.
80. F.D. Kennedy, ‘RPVs at sea’, Naval Forces, Vol. VI, No. V, 1985.
81. See Polmar, Ships and Aircraft, pp. 447-8 for discussion of unmanned LTA vehicles.
82. See Compton Hall, op. cit., pp. 98-100 for SPUR and discussion of other robotic projects. The
Friedman quotation (and more on UUV) is in US Naval Institute Proceedings, August 1988, pp.
123-4.
83. Quoted in ‘Satellites and naval operations’, Naval Forces, Vol. VI, No. 5, 1985, p. 66, an
important source for the rest of this section.
84. House Armed Services Committee testimony quoted in T.J. Richelson and Desmond Ball, The
Ties That Bind (London: Allen and Unwin, 1985), p. 218; chapter 9, on ‘Ocean Surveillance’, is
an excellent source on this sensitive subject.
85. N. Friedman in US Naval Institute Proceedings, May 1987, p. 93.
86. For a survey of Soviet ocean surveillance satellites see Jane’s Spaceflight Directory, 1986, pp.
306-7.
87. In the Navy Review issue of US Naval Institute Proceedings, p. 78.
88. N. Friedman, Modern Warship Design and Development, (Greenwich: Conway, 1979), p. 45.
This is still the finest survey of the subject and required reading for anyone who wishes to
understand the dynamics of the design of modern surface warships.
Chapter 7 The international and legal context
1. Discriminate Deterrence: Report of the Commission on Integrated Long Term Strategy,
Memorandum for the Secretary of Defense and the Assistant to the President for National
Security Affairs, January 1988.
2. J. Dean, Watershed in Europe: Dismantling the East-West Military Confrontation (Lexington,
Mass.: Lexington Books, 1987), pp. xiv-xv.
3. The author has assisted Windass in this work. See the two volumes ‘Common Security in
Europe’, published by the Foundation for International Security in 1987. A book version entitled
The Crucible of Peace was to be published in 1988 by Brasseys of London.
4. C. Hall, Britain, America and Arms Control 1921-37 (London: Macmillan, 1987), p. 218.
5. L. Freedman, ‘Arms control at sea’, Naval Forces, Vol. VI, No. III, 1985.
6. J. Borawski, ‘Risk reduction at sea: naval confidence building measures’, Naval Forces, Vol.
VIII, No. I, 1987. For the best brief summary of Stockholm see the same author’s ‘Accord at
Stockholm’, Bulletin of Atomic Scientists, December 1986, pp. 34-6.
7. Borwaski (1987), op. cit., p. 24.
8. For recent Soviet proposals see J. Borawski and E. Whitlow, ‘A Nordic zone of peace?’, Naval
Forces, Vol. VIII, No. VI, 1987.
9. ‘The Soviet-British incidents at sea agreement’, Naval Forces, Vol. VIII, No. I, 1987; and
‘Agreement between the USA and the USSR on the prevention of incidents on and over the high
seas’, pp. 166-7 of J. Goldblat Arms Control Agreements: A Handbook (London: Taylor and
Francis, 1983). For the possibilities for East-West naval arms control measures see J.R. Hill,
Arms Control at Sea (London: Routledge, 1988) and J.R. Hill and E.J. Grove, ‘Naval co-
operative security’, in Common Security in Europe (Adderbury: Foundation for International
Security, 1988).
10. K. Booth, Law, Force and Diplomacy at Sea (London: Allen and Unwin, 1985), p. 209.
11. Sir James Cable, ‘Gunboat diplomacy’s future’, US Naval Institute, Proceedings, August 1986, p.
38.
12. Ibid., p. 40.
13. Booth, op. cit., pp. 181-2.
14. See succeeding volumes of the official Naval review issue of US Naval Institute Proceedings,
published every May.
15. For surveys of Soviet operations see B. Dismukes and J.M. McConnell, Soviet Naval Diplomacy
(New York: Pergamon, 1979); and B.W. Watson, Red Navy at Sea: Soviet Naval Operations on
the High Seas 1956-80, (Boulder: Westview, 1982).
16. For an excellent examination of recent Soviet thinking on this see S. Shen-field, The Nuclear
Predicament: Explorations in Soviet Ideology, (London: Royal Institute of International Affairs,
1987), especially p. 43.
17. Dismukes and McConnell, op. cit., p. 299.
18. E.N. Luttwak, The Political Uses of Sea Power (Baltimore: Johns Hopkins University Press,
1974), p. 52.
19. H. Coutau-Beggarie, ‘The role of the French navy in French foreign policy’, Naval Forces, Vol.
VII, No. VI, 1986.
20. Cable, op. cit., p. 41.
21. For a stimulating critique of traditional British navalism see C. Coker, A Nation in Retreat?
Britain’s Defence Commitment (London: Brasseys, 1986), pp. 64-75. See also the final chapter of
E.J. Grove, Vanguard to Trident.
22. C.W. Weinberger, ‘NATO and the global maritime challenge’, SEALINK 86, SACLANT,
Norfolk, Virginia, 1986, p. 3.
23. Ibid., p. 4.
24. Ibid.
25. A.H. Cordesman, The Iran-Iraq War and Western Security, 1984-87 (London: Jane’s, 1987), pp.
132-4.
26. Jane’s Defence Weekly, 22 August 1987, p. 296. Also see The Observer, 16 August 1987.
27. The Guardian, 12 August, 1987.
28. Jane’s Defence Weekly, 19 September 1987, p. 581; 26 September 1987, pp. 263, 272-3; 24
October 1987, pp. 927, 941; Daily Telegraph, 22 January 1988; The Times, 25 October 1988; see
also The War In the Gulf: Background Brief (London: RUSI, June 1988) and also E.J. Grove,
‘Towards a western European navy’, Naval Forces, January 1988.
29. Colin Gray, ‘National interests and the Pacific – implications for NATO’, Sealink 1986, p. 36.
30. I.P.S.G. Cosby, ‘Maritime Aspects of Japanese Security’, Naval Forces, Vol. IV, No. V, 1983.
31. Quoted in B. Hahn, ‘PRC policy in maritime Asia’, Journal of Defense and Diplomacy, June
1986, p. 20.
32. F. Jiardano, ‘The Chinese navy’, Naval Forces, Vol. VIII, No. II, 1987, p. 198.
33. A.J. Tellis, ‘India’s naval expansion – structure, dimensions and context’, Naval Forces, Vol.
VIII, No. V, 1987, p. 45.
34. Ibid., p. 49.
35. Review of Australia’s Defence Capabilities, Report to the Minister for Defence by Mr. Paul Dibb
(Canberra: Australian Government Publishing Service, 1986), p. 1.
36. Ibid., pp. 3-4.
37. I owe the term to one of Australia’s brightest young officers, Lt. Cdr. James Goldrick RAN, the
noted naval historian.
38. Dibb Report, p. 8.
39. ‘Viewpoint’ by New Zealand’s defence minister in Jane’s Defence Weekly, 14 November 1987, p.
1119.
40. Cable, op. cit., p. 40.
41. Luttwak, op. cit., chap. 1.
42. See Booth, op. cit., chapter 1 for a useful survey of the background to UNCLOS III and the
provisions of the Convention.
43. Ibid., p. 34.
44. Ibid., p. 35.
45. Barry Buzan, ‘A sea of troubles’, Adelphi Paper No. 143 (London: International Institute for
Strategic Studies, Spring, 1978).
46. Ibid., pp. 4-15.
47. See the ‘International Law’ sections of Britannica Book of the Year, 1985-87 for useful annual
summaries of international legal disputes.
48. For a description of these ‘Freedom of Navigation’ operations see H. Parkes, ‘Crossing the line’,
US Naval Institute Proceedings, November 1986, pp. 43-5.
49. N.M. Hunnings, ‘Law’, Britannica Book of the Year, 1986, p. 291.
50. Booth, op. cit., p. 37.
51. Ibid., p. 38.
52. Ibid., pp. 44-5. Booth introduced the notion of the psycho-legal boundary in his article ‘Naval
strategy and the spread of psycho-legal boundaries at sea’, International Journal, Summer, 1983.
53. L.M. Alexander, ‘The ocean enclosure movement: inventory and prospect’, San Diego Law
Review, Vol. 20, No. 3, pp. 561-94.
54. Booth (1985), op. cit., p. 168.
55. For a discussion of some of the perceived obstacles to setting up a UN Gulf peacekeeping force
see J. McCoy, ‘Naval mirage in the Gulf’, The Times, 5 July 1988.
56. Cable, op. cit., p. 38.
57. Booth (1985), op. cit., 170.
Chapter 8 Economic stimuli and constraints
1. Figures calculated from relevant volumes of Britannica Book of the Year’ also see R. Hill,
Maritime Strategy for Medium Powers (London: Croom Helm, 1986), p. 31.
2. See various statements by the British Maritime League. For an excellent in depth analysis of the
decline of the British merchant fleet see Why the Ships Went, (London: British Maritime
Charitable Foundation, 1986). For a well-written critique of British attitudes see Sir James Cable,
‘Closing the British seas’, in Marine Policy, Vol. 11, No. 2, April 1987.
3. Statement made in 1982. Quoted in ‘Resource wars, the myth of American vulnerability’,
Defence Monitor, Vol. XIV, No. 9, 1985.
4. The US Navy League has produced such figures; see for example The Almanac of Sea Power
1983, p. 196.
5. See note 3.
6. Times Atlas of the Oceans, (London: Times Books, 1983), p. 144.
7. Hill, op. cit., p. 42.
8. Ibid., p. 42.
9. Ibid., pp. 43-4.
10. H.W. Maull, Raw Materials, Energy and Western Security (London: Macmillan, 1984), p. 110.
11. ‘Japan’s energy security to 2000 – the dawn of the age of multiple energies’, in R. Belgrave, C.K.
Ebinger and H. Okino, Energy Security to 2000 (London: Gower; Boulder, Colorado: Westview),
1987, p. 57.
12. E. Solem and A.F.G. Scanlon, ‘Oil and natural gas as factors in strategic policy and action’, in
A.H. Westing (ed.) Global Resources and International Conflict (SIPRI and Oxford University
Press, 1986).
13. Times Atlas of the Oceans, p. 189.
14. S.R. Peterson and J.M. Teal, ‘Ocean Fisheries’, in Global Resources and International Conflict,
supra., note 11.
15. ‘Shipbuilding’, Britannica Book of the Year 1986, pp. 256-7 gives the world ranking.
16. ‘The maritime role in strategic deterrence and crisis management’, Sealink 86, (Norfolk, Virginia:
SACLANT, 1986).
17. Cable, op. cit., p. 94.
18. Sealink 86, op. cit.
19. Maull, op. cit., p. 9.
20. The War in the Gulf: Background Brief (London: RUSI, June 1988), pp. 20-1.
21. Hill, op. cit., p. 47.
22. S.C. Truver, ‘Gramm-Rudman and the future of the 600-ship fleet’, US Naval Institute
Proceedings, May 1987, p. 123.
23. Washington Post, 8 August 1986, p. A14, quoted Truver, op. cit., p. 120.
24. Ibid., p. 123.
25. Jane’s Defence Weekly, 20 February 1988, p. 285.
26. Philip Pugh, The Cost of Seapower: The Influence of Money on Naval Affairs from 1815 to the
Present Day (London: Conway, 1986). A good summary of his major points appeared in Naval
Forces, No. III, 1987.
27. Pugh, op. cit., p. 314.
28. Ibid., p. 363.
29. Ibid., pp. 386-7.
30. See E.J. Grove, Vanguard to Trident, British Naval Policy since World War Two (Annapolis:
Naval Institute Press; London: Bodley Head, 1987), for repeated demonstration of this basic rule
with regard to the post 1945 Royal Navy.
31. Pugh, op. cit., pp. 387-8.
32. Ibid., p. 388.
Chapter 9 Navies in peace
1. R. Hill, ‘Control of the exclusive economic zone’, Naval Forces, No. VI, 1985. See also the same
author’s Maritime Strategy for Medium Powers, chap. 6.
2. J.D. Harbron, ‘Who shall defend the Canadian Arctic?’, Jane’s Naval Review, 5th Year of Issue,
(London: Jane’s, 19??), pp. 151-7.
3. See E.J. Grove, Vanguard to Trident, British Naval Policy since World War Two (Annapolis:
Naval Institute Press; London: Bodley Head, 1987), especially chaps 6 and 10.
4. R. Villar, Piracy Today: Robbery and Violence at Sea Since 1980 (London: Conway, 1985), pp.
10-11.
5. Ibid.
6. ‘Navies in a terrorist world’, Jane’s Naval Review, 5th Year of Issue, p. 172.
7. For descriptions of these operations see US Naval Institute Proceedings, especially for annual
reviews of naval operations published in May of every year. For the major Libyan raid see also
Aviation Week and Space Technology, 31 March 1986.
8. Ibid.
9. J.D. Watkins, ‘The Maritime Strategy’ in The Maritime Strategy supplement to US Naval
Institute Proceedings, January 1986, p. 8.
10. S.S. Roberts, ‘The October 1973 Arab-Israeli War’, in B. Dismukes and J. McConnell, Soviet
Naval Diplomacy (New York and London: Pergamon, 1979).
11. For a survey of the Soviet Navy’s operations see Dismukes and McConnell, op. cit.; B. Watson
and S.M. Watson, The Soviet Navy Strengths and Liabilities and B.W. Watson, Red Navy at Sea:
Soviet Naval Operations on the High Seas, (Boulder, Colo.: Westview, 1982 and 1986,
respectively).
12. K. Booth, Navies and Foreign Policy (London: Croom Helm, 1977); and Sir James Cable,
Gunboat Diplomacy 1919-79 (London: Macmillan, 1981).
13. Booth, op. cit., pp. 18-20.
14. Sir James Cable, ‘Showing the flag: past and present’, Naval Forces, No. III, 1987, p. 38. This
important article is effectively a supplement to the book, Gunboat Diplomacy.
15. Cable (1981), op. cit., p. 38.
16. Cable (1987), op. cit., p. 38.
17. Cable (1981), op. cit., p. 57.
18. Ibid., p. 67.
19. Ibid., p. 81.
20. Ibid., p. 117.
21. For more discussion of reach see J.R. Hill, Maritime Strategy for Medium Powers, chap 10.
22. Watkins, op. cit., p. 8.
23. Vice-Admiral Sir Geoffrey Dalton, ‘NATO’s maritime strategy’, in NATO and US Maritime
Strategy: Diverging Interests or Co-operative Effort? (Oslo: Norwegian Atlantic Committee,
1987), p. 42.
24. Ibid., pp. 42-3.
25. Ibid., p. 43.
26. Vice-Admiral H.C. Mustin, ‘Maritime strategy from the deckplates’, US Naval Institute
Proceedings, September 1986, p. 35.
27. For further development of these ideas see E.J. Grove, ‘The maritime strategy and crisis
stability’, Naval Forces, Vol. VIII, No. VI, 1987.
Chapter 10 Navies in war
1. For French nuclear doctrine and the role of submarines within it see D.S. Yost, France’s
Deterrent Posture Part 1 – Capabilities and Doctrine (London: IISS Adelphi Paper, 1984). A full
description of Britain’s targeting all its warheads on Moscow was broadcast in J. Barry, ‘Our
bomb, the secret story’, Channel 4 Television, April 1986.
2. For an excellent survey of the history of NATO’s sealift requirement see J.J. Sokolsky, ‘Seapower
in the nuclear age’, unpublished doctoral dissertation presented to Harvard University, August
1984. Figures for current sealift requirement from unclassified EASTLANT briefing given at
Northwood, 30 January 1987.
3. Northwood briefing, ibid.
4. The 1,000 shipping movements in the first month were discussed by Assistant Under-Secretary
(Programmes), R.C. Mottram, in testimony to the House of Commons Defence Committee, 3
February 1988, House of Commons Session 1987-88, Defence Committee, Sixth Report, The
Future Role and Size of the Royal Navy Surface Fleet, p. 8.
5. Sealink 86, (Norfolk, Va.: SACLANT, 1986), p. 47. Cargo figures from Vice-Admiral Sir
Geoffrey Dalton, ‘NATO’s maritime strategy’, in E. Ellingsen (ed.) NATO and US Maritime
Strategy: Diverging Interests or Co-operative Effort (Oslo: Norwegian Atlantic Committee,
1987), p. 44.
6. Sealink 86, op. cit.; Dalton, op. cit.
7. C.H. Whitehurst, The U.S. Merchant Marine (Annapolis: Naval Institute Press, 1983), p. 131;
Dalton, op. cit., pp. 44-6.
8. Sealink 86, op. cit., p. 47.
9. The organization of the Joint Chiefs of Staff, United States Military Posture, Fiscal Year 1987, p.
68.
10. ‘Analysis of the September 1987 sealift ship list’, unclassified section of 1987 PBOS Report.
11. Ibid. For a longer discussion of sealift issues see E.J. Grove’s paper ‘The merchant fleet and
deterrence’, prepared for the British Maritime Charitable Foundation, 1988.
12. See notes 11 and 31, chapter 1.
13. See E.J. Grove, Vanguard to Trident, British Naval Policy since World War Two (Annapolis:
Naval Institute Press; London: Bodley Head, 1987); N. Friedman, The Post-War Naval
Revolution; and G. Till and E. Grove, ‘Anglo American naval strategy 1945-67 – the impact of
the Cold War, NATO and nuclear weapons’, in Maritime Strategies of the Great Powers: Anglo-
American Perceptions (London: Macmillan, 1989).
14. ‘Finally seizing the initiative opens the way to apply direct pressure on the Soviets to end the war
on our terms – the new goal of our strategy once deterrence has failed.’ Admiral J.D. Watkins,
‘The Maritime Strategy’, in The Maritime Strategy, supplement to US Naval Institute
Proceedings, January 1986, p. 11.
15. B.R. Posen, ‘Inadvertent nuclear war? Escalation and NATO’s northern flank’, International
Security, Vol. 7, No. 2, Fall 1982, reprinted in S.E. Miller (ed.) Strategy and Nuclear Deterrence
(Princeton, NJ: Princeton University Press, 1984).
16. P.H. Nitze and L. Sullivan Jr., Securing the Seas; The Soviet Naval Challenge and Western
Alliance Options (Boulder Colo.: Westview, 1979), p. 373.
17. J.R. Hill, Anti-Submarine Warfare (Shepperton: Ian Allan, 1984), pp. 98-103.
18. Nitze and Sullivan, op. cit., pp. 370-1.
19. The scenario of the ‘Ocean Safari’ Exercise in 1987.
20. ‘A survey of western ASW in 1985’, International Defense Review, No. 10, 1985, p. 1597.
21. Translated from Vice-Admiral Sir Geoffrey Dalton, Atlanterhavet: NATO’s Strategiske
Planlegging (Norwegian Atlantic Committee, 1986), p. 7.
22. Dalton, op. cit., p. 40.
23. Ibid., p. 42.
24. See for example the comprehensive report on ‘Ocean Safari 83’ by Martin Horseman in Armed
Forces, October 1983. The author observed Exercise ‘Teamwork 88’ in person.
25. Dalton, op. cit., p. 43.
26. Major-General D.J. Murphy, ‘Role of the U.S. marine corps’, and Vice-Admiral Torolf Rein,
‘Reinforcing the northern flank. The case of Norway’, in E. Ellingsen (ed.), Reinforcing the
Northern Flank (Oslo: Norwegian Atlantic Committee, 1987) pp. 57 and 34 respectively.
27. Military Posture, FY1987, p. 69.
28. Murphy, op. cit., p. 54.
29. Brigadier R.J. Ross, ‘Organisation of UK/NL landing force’, in Ellingsen, Reinforcing the
Northern Flank, op. cit., pp. 66-7.
30. Vice-Admiral Torolf Rein, ‘The situation in the Norwegian Sea and the Norwegian naval
interests’, Ellingsen, NATO and U.S. Maritime Strategy, op. cit., p. 57.
31. Dalton, op. cit., pp. 42-3. For recent discussion of Mediterranean strategy see the proceedings of
the 1987 Conference of the International Institute for Strategic Studies held at Barcelona,
Prospects for Security in the Mediterranean (Adelphi Papers 229-231).
32. H.K. Ullman, ‘The Pacific and U.S. naval policy, Naval Forces, No. VI, 1985, p. 42. For the
importance of Pacific thinking in the evolution of the Maritime Strategy see J.B. Hattendorf, ‘The
evolution of the Maritime Strategy’, Naval War College Review, Summer 1988.
33. See P.X. Kelley, ‘The amphibious warfare strategy’, Maritime Strategy brochure, p. 26.
34. This analysis is based on T.L. McNaugher’s excellent study Arms and Oil: U.S. Military Strategy
and the Persian Gulf (Washington: Brookings, 1985, pp. 55-61).
35. Military Posture, FY1987, p. 69. There are thirteen modern MPS ships currently on charter, five
in the Indian Ocean, four in the Pacific and four in the Atlantic. See Combat Fleets of the World,
1988, pp. 803-4.
36. Ibid., and McNaugher, op. cit., p. 67-8.
37. Ibid.; see also pp. 78-9.
38. Ibid.; chapter 2 gives a good assessment of the Soviet threat.
39. Professor L. Martin’s analysis in The Sea in Modern Strategy (London: Chatto and Windus,
1967) was heavily conditioned by this perspective.
40. A.H. Cordesman, The Iran-Iraq War and Western Security (London: Jane’s, 1987), pp. 46-7.
41. Ibid., p. 68.
42. Ibid., especially pp. 46-56.
43. John Wise, ‘Maritime electronic warfare and its applicability to the protection of fleet support
units or merchant vessels operating in hostile waters’, Proceedings of ATI European Naval
Forecast Conference, London, 14-15 May 1987.
44. Cordesman, op. cit., p. 116 quoting US Congressional testimony.
45. Ibid., pp. 132-4.
46. A.H. Cordesman, ‘Western seapower enters the Gulf’, Naval Forces, II, III and IV, 1988 gives a
comprehensive and most useful account from the American perspective of the growing outside
naval involvement. He understates, however, the importance of European ‘concertation’; see
chapter 7.
47. N. Friedman in US Naval Institute Proceedings, June 1988, p. 119.
Chapter 11 The future of seapower - three theoretical frameworks
1. For Corbett’s and other contemporary views on Mahan see E.J. Grove’s introduction to Corbett’s
Some Principles of Maritime Strategy (Annapolis: US Naval Institute; London: Brasseys, 1988).
2. For the scope of Mahan’s writings see John B. and Lynn C. Hattendorf, A Bibliography of the
Works of Alfred Thayer Mahan (Newport, Rhode Island: Naval War College Press, 1986). The
Hattendorfs list over fifty English editions and printings of The Influence of Sea Power Upon
History 1660-1783, (first published by Little Brown and Co., Boston and Sampson Low Marston,
London in 1890) and fourteen translations.
3. Ibid., p. 25.
4. Ibid., p. 26.
5. See the House of Commons Defence Committee’s Report, The Defence Requirement for
Merchant Shipping and Aircraft, June 1988, and the author’s report for the British Maritime
Charitable Foundation, The Merchant Fleet and Deterrence.
6. A.T. Mahan, The Influence of Sea Power upon History 1660-1783 (Boston: Little Brown, 1890),
p. 23.
7. Ibid.
8. See for example Graham Searjeant, ‘The rising tide of protectionism’, Britannica Book of the
Year 1988, pp. 172-3.
9. J.R. Hill, Maritime Strategy for Medium Powers (London: Croom Helm, 1986), chap. 10.
10. Mahan, op. cit., p. 26.
11. Ibid.
12. Ibid., pp. 28-9.
13. Ibid., p. 29.
14. Lord Fraser at a meeting of the Chiefs of Staff, COS(50) 46th Meeting, DEFE4/29.
15. Mahan, op. cit., p. 29.
16. Ibid., p. 40.
17. Ibid., p. 43.
18. Ibid., p. 45.
19. Ibid., p. 53.
20. Ibid., p. 67.
21. Philip Pugh, The Cost of Sea Power: The Influence of Money on Naval Affairs from 1815 to the
Present Day (London: Conway, 1986), p. 23.
22. Mahan, op. cit., p. 82.
23. Jane’s Fighting Ships, 1987-88, respectively, pp. 494, 240, and 49.
24. Hill, op. cit., pp. 42-3.
25. Mahan, op. cit., p. 88.
26. K. Booth, Navies and Foreign Policy (London: Croom Helm, 1977), especially p. 16.
27. Ibid., pp. 15-16.
28. Ibid.
29. Ibid., p. 16.
30. Sir James Cable, Gunboat Diplomacy 1919-1979 (London: Macmillan, 1981); see chap 9 in this
volume.
31. Commodore Richard Cobbold, Director, Defence Concepts to the Royal United Services
Institute, 18 November 1980.
32. I owe the ‘three circles concept’ to my colleague in the Foundation for International Security,
Stan Windass.
33. See, for example, K. Booth, Navies and Foreign Policy (London: Croom Helm, 1977); Hill, op.
cit.; M.W. Janis, Sea Power and the Law of the Sea, (Lexington, Mass: Heath, 1976); and M.A.
Morris, Expansion of Third World Navies (London: Macmillan, 1987). Two important recent
articles on this subject are Nien-Tsu Alfred Hu and J.K. Oliver, ‘A framework for small navy
theory’ Naval War College Review, Spring 1988 and S.W. Haines, ‘Third World navies: myths
and realities’, Naval Forces, April 1988.
34. Haines, op. cit., p. 24.
35. Morris, op. cit., p. 25.
36. Ibid.
37. Ibid.
38. Ibid., p. 33.

Acronyms and abbreviations

AAW anti-air warfare


ABM anti-ballistic missile
ACV air cushion vehicle
ADA Action Data Automation System
ADAWS Action Data Automation and Weapons System
ADC acoustic device countermeasures
AEW airborne early warning
AGI auxiliary ships
AIO Action Information Organization
AMRAAM advanced medium-range air-to-air missile
ANS Anti Navires Supersonique
ANZDCC Australian/New Zealand Defence Consultative Committee
ARMS Advanced Remote Minehunting System
ASATS anti-satellite weapons
ASROC anti-submarine rocket
ASUW anti-surface ship warfare
ASW anti-submarine warfare
ASW-SOW Anti-Submarine Warfare Stand-Off Weapon
ATSA advanced tactical support aircraft
BASH barge aboard ship
Bo-Ro general purpose Ro-Ro bulk carrier container ship
CAAIS Computer-Assisted Action Information System
CACS Computer-Assisted Command System
CBM confidence building measures
CCTS Closed-Cycle Thermal System
CDE Conference on Confidence and Security Building Measures and
Disarmament in Europe
CENTCOM Central Command
CEP circular error profitability
C3I command, control, communications and intelligence
CIC Combat Information Centre
CINCHAN Commander-in-Chief Channel
CINCSOUTHCommander-in-Chief South
CIWS Close-In Weapons System
CLGP cannon-launched guided projectile
CMEA Council for Mutual Economic Assistance
CODAD combined diesel and diesel turbine
CODAG combined diesel and gas turbine
CODLAG combined diesel electric and gas turbine
CODOG combined diesel or gas turbine
COGAG combined gas and gas turbine
COGOG combined gas or gas turbine
CONMAROPSConcept of Maritime Operations
CSBM confidence and security building measures
CTOL conventional take-off and landing
CVBG carrier battle groups
CVV mid-size carriers
DASH Drone Anti-Submarine Helicopter
DE destroyer escort
DIFAR directional low-frequency analysis and recording
DSMAC digital scene matching area correlator
EASTLANT Eastern Atlantic Command
ECM electronic countermeasures
EEZ Exclusive Economic Zone
ELF extremely low frequency
Eorsat Electronic Ocean Surveillance Satellite
FFG guided missile frigate
HELRAS helicopter long-range sonar
ICBM intercontinental ballistic missile
INF intermediate nuclear forces
ISAR inverse synthetic aperture radar
ISO International Standardization Organization (based in Geneva)
LASH lighter aboard ship
LCAC landing craft air cushion
LCAT low-cost acoustic torpedo
LHA landing helicopter assault
LHD landing-ship helicopter dock
LNG liquid natural gas
LOFAR low-frequency analysis and recording
LPD landing platform dock
LPG liquid petroleum gas
LPH landing platform helicopter
LRAACA long-range air ASW-capable aircraft
LSD landing ship docks
LTA ligher than air
MAD magnetic anomaly detection
MCM mine countermeasures
MEF Marine Expeditionary Force
MESAR multi-function electronically scanned array radar
MIRV multiple independently targeted re-entry vehicle
MRNA Multi-Role Naval Auxiliary
NATO North Atlantic Treaty Organization
NAVSTAR navigation system using tuning and ranging
NTDS naval tactical data system
OAU Organization for African Unity
OECD Organization for Economic Co-operation and Development
PBV post-boost vehicle
PLARB Russian equivalent of SSBN
PREPO pre-positioning
ROE rules of engagement
Ro-Ro roll-on roll-off
Rorsat Radar Ocean Surveillance Satellite
RPV remotely piloted vehicle
SACEUR Supreme Allied Commander Europe
SACLANT Supreme Allied Commander Atlantic
SALT Strategic Arms Limitation Talks
SAM surface-to-air missile
SAR synthetic aperture radar
SCEPS stored chemical energy propulsion system
SCS sea control ships
SDI strategic defense initiative
SEALS sea-air-land special forces
SES surface effect ship
SHF super high frequencies
SIOP Strategic Integrated Operational Plan
SLAT surface-launched air-targeted
SLBM submarine-launched ballistic missile
SLCM sea-launched cruise missile
SLCSAT Submarine Laser Communication Satellite System
SMCS Submarine Command System
SOSUS Sound Surveillance System
SPUR Seicon Patrolling Underwater Robot
SSBN nuclear-powered ballistic submarine
SSM surface-to-surface missile
SSN nuclear-powered submarine
START Strategic Arms Reduction Talks
STOVL short take-off/vertical landing
SubACS Submarine Advanced Combat System
SWATH small waterplane twin-hull
SWCM special warfare craft medium
T-AGOS ocean surveillance ship
TERCOM terrain contour-matching
teu twenty-foot equivalent units
TLAM Tomohawk cruise missile
TOAD towed offboard active decoy
UEP underwater electrical potential
UHF utlra high frequencies
ULCC ultra large crude carrier
UNCLOS III Third United Nations Conference on the Law of the Sea
UUV unmanned underwater vehicle
VLAD vertical land array DIFAR
VLCC very large crude carrier
VLF very low frequencies
VSS V/STOL support ships
V/STOL vertical/short take-off and landing
VTOL vertical take-off and landing
WEU Western European Union
WIG ‘wing in ground’
WTO Warsaw Treaty Organization

Index

acoustic device countermeasures (ADC) 75


Action Data Automation System (ADA) 90
Action Data Automation Weapons System (ADAWS) 90-1
Action Information Organization (AIO) 90-1
Adcock, C.C.F. 80
advanced medium-range air-to-air missile (AMRAAM) 86, 141
Advanced Remote Minehunting System (ARMS) 72
advanced tactical support aircraft (ATSA) 141
Aegis system 83
air cushion vehicle (ACV) 118
airborne early warning (AEW) 86, 141, 144
aircraft 138-45;
advantages 139;
and airborne early warning facilities 144;
and anti-submarine warfare 140-1;
drawbacks 139-40;
and maritime strategy 14;
short take-off/vertical landing 141-3;
and skyhook 143;
threat of 81;
vertical take-off and landing 142-4
aircraft carriers 124-8;
maritime strategy 14;
mid-size (CVV) 127
Albania 240
Alexander, L.M. 170
Algeria 175, 239
Amery, L. 5
amphibious force 211
amphibious vessels 129-30
amphibious warfare 49;
maritime strategy 22-3;
and military power 48-51
Angola 240
Anti Navires Supersonique (ANS) 66
anti-satellite weapons (ASATS) 148
anti-air warfare (AAW) 61, 81-7;
future of 235;
and surface ships 100-28 passim;
weapons for 139
anti-ballistic missile (ABM) 87, 148, 199
anti-submarine rocket (ASROC) 76, 107, 113
anti-submarine warfare (ASW) 15, 72-81;
future of 235;
and nuclear warfare 200-1;
and surface ships 99-132 passim;
techniques of 53;
weapons for 63, 65, 139
Anti-Submarine Warfare Stand-Off Weapon (ASW-SOW) 76
anti-surface ship warfare (ASUW) 61;
and surface ships 102-26 passim;
techniques of 74-5;
weapons for 61-4, 139
Argentina 176, 239, 240
Australia 165-6, 175-6, 238
Australian/New Zealand Defence Consultative Committee (ANZDCC) 166
auxiliary ships (AGI) 55

Baggett, Admiral L. 203


Bahrain 240
Bangladesh 239
barge aboard ship (BASH) 37
battlefleets 25-6
battleships 103-4
Belgium 175-6, 238
Berlin Crisis 57
blockade 15, 16
Booth, K.: and international conflict 156-7, 171;
and order at sea 167, 170;
peacetime uses of navies 194;
and UNCLOS III 169-70;
and use of the sea 232-5
Borawski, J. 155
Brazil 175-6, 178, 239
Britain: confrontation with Union of Soviet Socialist Republics 156;
economic use of the sea 172, 175-6;
foundation of sea power 224, 229;
as global force 238;
merchant shipping 4, 40-1;
and mine countermeasures 99;
navies in peacetime 188;
shipbuilding 178;
surface vessels 105-6, 110-11, 126, 130-1;
torpedoes 65, 73-4;
World War II strategy 50
Brunei 240
Bulgaria 239
Burma 240
Butler, E. 118-19
Buzan, B. 168

Cable, Sir J.: and international conflict 156-7, 166, 171;


merchant shipping 179;
peacetime uses of navies 194-5;
and smaller navies 160;
and use of the sea 233-4
Cameroons 240
Canada 107, 136-7, 175-6, 188-9, 202-3, 240
cannon-launched guided projectile (CLGP) 62
Cape Matapan 14
carrier battle group (CVBG) 207-8, 213
Central Command (CENTCOM) 214
Chernavin, Admiral S. 193
Chile 104, 176, 239
China 40-1, 43, 108-9, 131, 163-4, 178, 238
circular error profitability (CEP) 95
Clauswitz, K. von 18
Close-In Weapons System (CIWS) 82, 106, 109
Closed-Cycle Thermal System (CCTS) in torpedoes 74-5
co-operation: in Persian Gulf 162;
in security 154-5
Coast Guards 187-8
coastal bombardment 87-8
coastlines and foundations of sea power 225
Colombia 239
Combat Information Centre (CIC) 90, 115
Combined diesel and diesel turbine (CODAD) 123
combined diesel electric and gas turbine (CODLAG) 106, 123
combined diesel and gas turbine (CODAG) 121
combined diesel or gas turbine (CODOG) 121
combined gas and gas turbine (COGAG) 123
combined gas or gas turbine (COGOG) 123
command and control of weapons systems 89-92
command of the sea 12-13, 19
Commander-in-Chief Channel (CINCHAN) 211
Commander-in-Chief South (CINCSOUTH) 212
communications: satellites 146, 148-9;
submarines 133-4
Computer-Assisted Action Information System (CAAIS) 90
Computer-Assisted Command System (CACS) 90-1
Concept of Maritime Operations (CONMAROPS) 15, 209-13
confidence and security building measures (CSBM) 156, 197
constabulary navies 240
containers see utilization
conventional take-off and landing (CTOL) 141-2
convoy 11-12;
in future wars 209-10;
operational laws of 19-21;
and warfare 208-9;
in World War I 18;
in World War II 17-18, 19
Corbett, Sir J. 6, 10, 19, 221;
and amphibious warfare 49;
and constitution of fleets 24;
and convoy 18;
principle of maritime strategy 11-15, 23, 58;
and uses of the sea 235
corvettes: definition 102;
role of 112-13
Council for Mutual Economic Assistance (CMEA) 9, 174, 178
Coupar, A. 174
crisis management 196-8
cruisers: definition 102;
role of 102-3
Cuba 239
Cunningham, Admiral A.B. 14
Cyprus 40

Dalton, Sir G. 203


Dean, J. 154
Denmark 175-6, 239
depth charges 77-8;
nuclear 93
destroyers: definition 102;
destroyer escort (DE) 113;
role of 104-9
deterrence 187, 196
Dibb, P. 165-6
digital scene matching area correlator (DSMAC) 88
diplomatic role of navies 233-4
directed energy weapons 86-7
directional low-frequency analysis and recording (DIFAR) 80
Dominican Republic 240
Drone Anti-Submarine Helicopter (DASH) 145

Eames, M. 117
East Germany 239
Eastern Atlantic Command (EASTLANT) 203
Eberle, Admiral Sir J. 179
economic shipping during wartime 203-4
economic use of the sea 31-45;
coal transport 34;
dry bulk cargoes 34-5, 36-40
fishing and whaling 41-4;
and flags of convenience 40-1;
merchant shipping 41;
passenger traffic 39;
petroleum transport 31-3, 40, 44-5;
roll-on roll-off 36-8;
sea routes 38;
trade 31-8;
unit cargo vessels 37-8;
unitization of cargo 35-7, 40;
world trading fleet 39-40
economy: dependence on the sea 174-7;
financial flows 179;
strength of and sea power 226-8;
of superpowers 153-4;
and trade by sea 172-3;
and use of the sea 172-83
Ecuador 239
Egypt 176, 239
electronic countermeasures (ECM) 216
Electronic Ocean Surveillance Satellite (Eorsat) 147
Esher, Lord 6
Ethiopia 240
Exclusive Economic Zone (EEZ) 42, 168, 170, 187

Falklands Islands 55;


aircraft 85-6;
missiles 68-9, 84;
torpedoes in 63
fast-attack craft 99-100
ferries see passenger traffic
Finland 176, 240
Fisher, Admiral J.A. 26
fishing and dependence on the sea 178
fishing and whaling 41-4
flags of convenience 40-1
fleet support vessels 130-1
forward operations in warfare 210-11
foundations of sea power 221-32;
coastlines 225;
geographical position 224-5;
and government 228-9;
keys to 223-4;
and merchant shipping 221-3;
and national character 226-8;
population 226;
ranking of factors 229-32;
territorial extent 225
France: economic use of the sea 175-6;
as global force 238;
maritime intervention 159;
merchant shipping 4;
submarines 135;
surface vessels 105-6, 112, 125;
torpedoes 75
Friedman, N. 149, 209
frigates: definition 102;
role of 109-11

Gabon 175, 240


Gallipoli Campaign 48-9
Galtieri, L. 195
gas see petroleum
geographical position: and foundations of sea power 224-5;
and military power 47-8
Ghana 240
global conflict: increase in 156;
intervention 157-9
global force navies 237-8
Gorbachev, M. 154-5
Gorshkov, Admiral S. 12, 22, 24, 47, 87, 158
government and foundations of sea power 228-9
Gramm-Rudman-Hollings Act (1985) 180
Gray, C. 162
Greece 3, 40-1, 175-6, 239
Grenada, US Marines landings 157, 193
‘guerre de course’ 15-16, 21
Gulf, Arabian 68, 70, 160
gunboat diplomacy 192-6;
kinds of 195;
levels of 195-6
guns: anti-aircraft 81-3;
naval 61, 62-3

Haines, S.W. 237


Hayward, Admiral T. 213
helicopter carriers 128
helicopter long-range sonar (HELRAS) 80
Hill, Admiral J.R. 98, 174, 176, 180, 187, 223, 225, 230
Hong Kong 40-1

Iceland 240
India 108-9, 131, 164-5, 175-6, 238, 240
Indonesia 239
integrated weapons systems 89-91
intelligence-gathering satellites 147-8
intercontinental ballistic missile (ICBM) 94
intermediate nuclear forces (INF) 93
internationalization of sea 181-3, 233
intervention: by post-imperial powers 159-61;
by superpowers 157-9, 161
inverse synthetic aperture radar (ISAR) 67
Iran 175-6, 239
Iran-Iraq war 160-1, 216-17;
submarines 133
Iraq 239
Ireland 240
Israel 176, 239
Italy 75, 107, 175-6, 178, 238

Japan: amphibious warfare 49;


economic use of the sea 175-6;
fishing 42;
foundation of sea power 227;
merchant shipping 3, 40-1;
as naval power 163;
as regional force 238, 240;
shipbuilding 178;
surface vessels 106-7, 113

Kahn, H. 201
Kennedy, P. 4, 15
Kenya 240
Kuwait 160, 175, 240

Lambert, A. 23-4
landing craft air cushion (LCAC) 128
landing helicopter assault (LHA) 128
landing platform dock (LPD) 128-9
landing platform helicopter (LPH) 54, 128
landing ship dock (LSD) 128-9
law of the sea 167-9, 170
Lebanon 175
legal disputes 168-9
Lehman, J. 124, 140
length-beam ratio 115-16
Leyte Gulf, battle of 26
Liberia 3, 40, 175
Libya 175, 239
lighter aboard ship (LASH) 37
Littwak, E. 159, 166
long-range air ASW-capable aircraft (LRAACA) 140
low-cost acoustic torpedo (LCAT) 64
low-frequency analysis and recording (LOFAR) 80
Lygachev, Y. (Russian politican) 156

MccGwire, M. 89
Mackinder, H. 4-5, 8
magnetic anomaly detection (MAD) 53
Mahan, A.T. 10, 15, 221;
and convoy 16;
definition of sea power 3, 4, 6;
and economic uses of the sea 172;
foundations of sea power 222-6, 228-9, 231-2;
maritime strategy 22;
and military power 46-7;
and naval battles 13, 205
Malaysia 239
Marine Expeditionary Force (MEF) 214
maritime strategy 11-30;
and aircraft 14;
aircraft carriers 14;
amphibious warfare 22-3;
battlefields 25-6;
blockade 15, 16;
and command of the sea 12-13, 19;
constitution of the fleet 24-6;
and convoy 11-12, 16-19;
definition of 11;
‘fleet in being’ 21-2;
‘guerre de course’ 15-16, 21;
and naval battles 13-14;
and nuclear weapons 14;
operational laws 19-21
Martin, L. 57
Maull, H.W. 177
merchant shipping and foundations of sea power 221-3; see also economic
use of the sea
Metcalf, Vice-Admiral J. 114
Mexico 176, 240
Middle East: and petroleum exports 31-2;
in superpower war 213-15
Midway, battle of 13
military power 46-60;
advantages of sea 54-5;
amphibious warfare 48-51;
efficiency of transport 47;
geographical position 47-8;
intelligence 55;
and role of navies 233-4;
and rules of engagement 56-7;
and submarines 53-4, 55;
in World War II 49-50
mines 61, 70-2;
anti-submarine warfare 76-7;
in Persian Gulf 161
mine countermeasures (MCM) 71-2, 99, 118, 128, 146, 161-2, 237
minesweepers 99
missiles 61, 65-70;
anti-aircraft 81-3, 84-5
Morocco 239
Morris, M.A. 237-8
Multi-Role Naval Auxiliary (MRNA) 130
multi-function electronically scanned array radar (MESAR) 83-4
multihull craft 98, 117
multiple independently targeted reentry vehicle (MIRV) 200
Mustan, Admiral H.C. 197

naval battles and maritime strategy 13-14


naval diplomacy 193-5
naval exercises 155-6
Naval Tactical Data System (NTDS) 90
navies: in peace 187-98;
role of 232-6;
typology for 236-40;
and use of the sea 232-3;
in war 199-218
navigation system using tuning and ranging (NAVSTAR) 95
Netherlands, The 107, 175-6, 238
New Zealand 165, 176, 239
Nigeria 175-6, 239
North Atlantic Treaty Organization (NATO) 10, 15, 57, 228;
areas of interest 160, 162;
Concept of Maritime Operations 196-7, 209-10;
convoys 208-9;
economic shipping in wartime 203-4;
and maritime strategy 205-8, 212, 235;
and Rapid Reinforcement Plan 202
North Korea 239
North Yemen 240
Norway 41, 175-6, 239, 240
nuclear-powered ballistic submarine (SSBN) 89, 136, 154, 189-90, 201,
237-8
nuclear-powered submarine (SSN) 135-6, 238
nuclear warfare 199-202, 215
nuclear weapons 92-5;
ballistic missiles 94-5;
and maritime strategy 14;
pré-stratégique strike 94;
strikes against shore 93

ocean surveillance ship (T-AGOS) 78, 120


oil see petroleum
Oman 160, 240
Organization for African Unity (OAU) 157
Organization for Economic Cooperation and Development (OECD) 9, 174,
213

Pakistan 239, 240


Panama 3, 40, 175
passenger traffic 39
patrol vessels 98-9
peacetime uses of navies 187-98;
and choice of vessels 189-90;
coastguards 187-8;
constabulary duties 187-9;
escalation of threat 190;
and gunboat diplomacy 192-6;
and piracy 190-2;
and terrorism 191-2;
and wartime requirements 187, 189
Peru 239
petroleum: economic use of the sea 31-3, 40, 44-5;
gases, transport of 33;
and ocean trade 177
Philippine Sea, battle of 14, 26
Philippines 40-1, 239
piracy 190-2
PLARB see nuclear-powered ballistic submarine
platforms 96-149;
aircraft 138-45;
interconnections 149;
kinds of 96-8;
space 146-9;
submarines 132-8;
surface warships 98-132;
unmanned 145-6
Poland 175, 178, 239
population and foundations of sea power 226
Portugal 239
Posen, B. 206
pre-positioning (PREPO) 210, 214;
of military forces 46-7
protection navies 239
Pugh, P. 180-3, 229

Qatar 175, 240

Radar Ocean Surveillance Satellite (Rorsat) 147


Rapid Reinforcement Plan 202-3
Reagan, R. 173, 174, 229
regional force navies 238-9
Rein, Admiral T. 211
remotely-piloted vehicle (RPV) 145
rocket launchers 77-8;
against shore 88
role of navies, triangles 233-6
roll-on roll-off (Ro-Ro) 35
Romania 178, 239
Roskill, S. 50
rules of engagement (ROE) 56-7

Saudia Arabia 175, 239


Schofield, Admiral B.B. 49
sea control see command of the sea
sea control ships (SCS) 127
sea-launched cruise missile (SLCM) 94
sea power: definition of 3;
foundations of 221-32;
future of 221-41;
and merchant marine 3-4;
in the modern world 3-10;
role 232-6;
typology for navies 236-40
sea routes 38
sea water, properties of 52-3
sea-air-land special forces (SEALS) 118
security, cooperative 154-5
Seicon Patrolling Underwater Robot (SPUR) 146
shipbuilding 178-9;
budget for 180-1
shore operations 87-9;
and nuclear weapons 93
short take-off/vertical landing (STOVL) 86, 89, 125-7, 128, 141-4, 164
Sidra, Gulf of 100
Singapore 41, 175, 240
‘ski-jump’ take-off aid 142
small waterplane twin-hull (SWATH) 98, 119-21
socio-political culture and foundations of sea power 226-8
Sokolsky, J. 46
Somalia 240
sonar 79-81
Sound Surveillance System (SOSUS) 78, 80, 146
South Africa 176
South Korea 41, 42, 178, 239
South Yemen 240
space platforms 146-9
Spain 112, 126, 175-6, 238
special warfare craft medium (SWCM) 118
speed of ships 96-7
Sri Lanka 240
stored chemical energy propulsion system (SCEPS) 74
Strategic Arms Limitation Talks (SALT) 67, 154
Strategic Arms Reduction Talks (START) 154
Strategic Defense Initiative (SDI) 87
Strategic Integrated Operations Plan (SIOP) 94
Submarine Advanced Combat System (SubACS) 91
Submarine Command System (SMCS) 91
Submarine Laser Communication Satellite System (SLCSAT) 134
submarine-launched ballistic missile (SLBM) 94-5
submarines 132-8;
advantages of 137-8;
communications 134-5;
and depth charges 77;
detection of 79-81;
in Iran–Iraq war 133;
limitations 132-4;
and military power 53-4, 55;
and mines 76-7;
nuclear, in peacetime 189-90;
nuclear powered 135-7;
quietness of 135-6;
and rocket launchers 77-8;
and torpedoes 72-6;
warfare 199-201
superpowers: confrontation 154-6;
and economic use of the seas 173-4;
intervention by 157-9;
and ‘naval suasion’ 166
Supreme Allied Commander Atlantic (SACLANT) 196, 203, 211
Supreme Allied Commander Europe (SACEUR) 94, 211-12
surface effect ship (SES) 118-19
surface-launched air-targeted (SLAT) missile 85
surface warships: aircraft carriers 124-8;
amphibious vessels 129-30;
battleships 103-4;
corvettes 112-13;
cruisers 102-3;
destroyers 102, 104-9;
fast attack craft 99-100;
fleet support vessels 130-1;
frigates 102, 109-11;
fuels 121-3;
future form of 114-17, 122;
helicopter carriers 128;
length-beam ratios 115-16;
and manpower 121;
minesweepers 99;
multi-hull 98, 117;
nomenclature 100-2;
patrol vessels 98-9;
as platforms 98-132;
second-hand sales of 111-12;
surface skimmers 97, 117-20,
vulnerability 131-2
surface-to-air missile (SAM) 103-4, 107, 109, 111, 126
surface-to-surface missile (SSM) 103-4, 108, 110-12, 230
surveillance: in peacetime 196-7;
satellites 147-8
Sweden 175-6, 239
Sweetman, W. 142
synthetic aperture radar (SAR) 148
Syria 175, 240

Taiwain 111, 178, 239


Tanzania 240
technology, strength of and sea power 226-8
terrain contour matching (TERCOM) 88
territorial defence navies 239-40
territorial extent and foundations of sea power 225
territorialization of oceans 170-1
terrorism 191-2
Thailand 239
Third United Nations Conference on the Law of the Sea (UNCLOS III)
167, 170
threats, escalation of 190
Till, G. 13
token navies 240
Tomahawk cruise missile (TLAM) 88-90, 93
torpedoes 61, 63-5;
anti-submarine warfare 72-6
Torrington, Earl of 22
towed offboard active decoy (TOAD) 69
trade: and economic use of the sea 31-8;
exports by sea 172-3
Trafalgar, battle of 13
Trinidad & Tobago 175
Tunisia 175, 240
Turkey 239
Turner, S. 12
typology for navies 236-40;
constabulary navies 240;
global force navies 237-8;
protection navies 239;
regional force navies 238-9;
territorial defence navies 239-40;
token navies 240

ultra large crude carrier (ULCC) 33


underwater electrical potential (UEP) 53
Union of Soviet Socialist Republics: Berlin Blockade 57;
confrontation with United States of America 154, 156;
diplomacy 193-4;
economic use of the sea 153-4, 175;
fishing 42;
foundation of sea power 225, 230;
as global force 237-8;
Marine Corps 87;
maritime intervention 158;
Maritime Strategy for future wars 208;
merchant shipping 38, 40-1;
missiles and rockets 65-6;
as naval power 4, 5;
nuclear weapons 94;
space platforms 148;
submarines 135, 138, 199-200;
surface vessels 102-3, 110, 112, 129, 131;
torpedoes 64, 75-6
unit cargo vessels 37-8
United States of America: aircraft 142-3;
Berlin Blockade 57;
confrontation with Union of Soviet Socialist Republics 154, 156;
diplomacy 193;
economic use of the sea 153-4, 173-6;
foundation of sea power 225, 226;
as global force 237, 240;
Marine Corps 50-1;
maritime intervention 158;
Maritime Strategy for future wars 205-12;
merchant shipping 4, 40-1;
missiles and rockets 78;
as naval power 4, 5;
Rapid Reinforcement Plan 202-3;
shipbuilding 178;
space platforms 147;
submarines 26, 134;
surface vessels 103-5, 109, 119-21, 125, 128;
torpedoes 73, 76
unitization of cargo 35-7, 40
unmanned platforms 145-6
unmanned underwater vehicle (UUV) 146
Uruguay 240
Veazey, Captain S.E. 120
vehicle carriers 37
Venezuela 175-6, 239
vertical line array DIFAR (VLAD) 80
vertical take-off and landing (VTOL) 102, 142-4
very large crude carrier (VLCC) 33, 216
V/STOL support ship (VSS) 127

warfare: amphibious force 211;


carrier battle groups 207-8, 213;
convoys 208-9;
and economic shipping 203-4;
forward operations 210-11;
in limited (local) wars 215-17;
in Middle East 213-15;
and NATO area 209-13;
navies in 199-218;
nuclear 199-202, 215;
in Pacific 213;
Rapid Reinforcement Plan 202-3;
strategy for 204-6;
submarines 199-201;
in Third World 217
Warsaw Treaty Organisation (WTO) 57
warships, costs of 181-3
Waters, D.W. 16, 19
weapons systems 61-95;
anti-air warfare 81-7;
anti-submarine warfare 72-81;
anti-surface ship warfare 61-72;
command and control 89-92;
and nuclear weapons 92-5;
shore operations 87-9
Weinberger, C. 160
West Germany 108, 175-6, 227-8, 238
Western European Union (WEU) 161-2, 238
Windass, S. 154
‘wing in ground’ (WIG) 97
world trading fleet 39-40
World War I, naval strategy 6
World War II: Britain’s strategy 50;
military power in 49-50;
naval strategy 7, 50

Yugoslavia 178, 239

Zumwalt, Admiral E.R. 100

You might also like