DEDICATION
To my children, Sarah, Erin, Felicity and Michael,
and my grandchildren, Sophie, Angus, Eliza, Grace
and Andrew, with love and in the hope that your
Australia will remain ‘one and free’, and also
sovereign, safe and prosperous.
CONTENTS
Dedication
Foreword by Peter Dutton
Introduction
Prologue: Out of the Blue?
1. Rise to Power
2. Playing Catch-up
3. Wolf Warriors
4. In the Grey Zone
5. Where and When?
6. The Response: The US and Its Allies
7. The Response: Australia
8. Concepts of War
9. Current Policy
10. The Right War
11. Can the US Win?
12. Australia, Defend Yourself!
Epilogue: A New Reality
Acknowledgements
Endnotes
About the Author
Copyright
FOREWORD
by Peter Dutton, Minister for Defence
It is a pleasure to write this foreword to Danger on Our Doorstep. Although
I’d been aware of Jim Molan for quite some time, I first got to know him
well during the Rudd–Gillard–Rudd government of 2006–13. We were in
opposition. After we assumed government, against all professional advice
and media commentary, and working with Immigration Minister Scott
Morrison, Jim provided the means whereby a simple political policy of
‘Stop the Boats’ was converted into the strategy that became ‘Operation
Sovereign Borders’. This was no easy accomplishment. The Rudd and
Gillard governments had proved over six years how difficult it was to make
such a conversion, and not one public voice was of the opinion that it was
possible. The combination of Scott Morrison and Jim Molan was most
effective, not just in formulating policy and strategy, but also in explaining
the policy and then making it work in government.
Jim has brought the same insight to the Parliamentary Joint Committee
on Intelligence and Security (PJCIS). Again, he’s been ahead of his time,
arguing for the formation of the Home Affairs portfolio. Jim was proved
right.
Danger on Our Doorstep is a serious work by Jim Molan. It is a
manifestation of all Jim’s qualities. With clarity, Jim deals with some of the
most complex and sensitive issues facing our country – and he has drawn
on a lifetime of experience. As a member of the Australian Defence Force
(ADF), Jim has thought and talked about war for 40 years. He has run a real
war as Chief of Operations, primarily for the US forces in Iraq in 2004–05.
He has participated in several other operations or activities in our region
over those decades, in and with our regional neighbours. And, as a
politician and a diplomat, he has been exposed to the complexities of
foreign and domestic policy. I have to say, I do not agree with all of Jim’s
assessments in this book, but even this cursory review of his significant
experience and demonstrable capabilities in the defence and national
security spheres is enough to show that we should at the very least listen to
and debate his arguments.
To deter war, the first step is to understand the problem. Jim makes the
argument that due to a reliance on the US for security and prosperity,
Australia is lacking resilience in a period of increased strategic competition
in our region. I will let Jim defend his assessments, except to say that, in
launching the Defence Strategic Update in July 2020, the government
outlined how rapidly Australia’s strategic environment was deteriorating. It
was the recognition of this deterioration and the rapid militarisation of our
region that reaffirmed the need for Australia to invest in high-end
capabilities that bolster our deterrence and better prepare us to respond in
the event of conflict. This in turn led the government to explore the
feasibility of acquiring nuclear-powered submarine technology, which
culminated with the establishment of AUKUS, the trilateral security pact
between Australia, the UK and the US.
In this book, Jim Molan describes his view of the military options that
both China and the US would have available should China move to achieve
regional dominance. As a military tactician, Jim is aware that it is only
through understanding the nature of modern warfare that Australia can
prepare correctly to defend its sovereignty and regional stability. Uniquely,
Jim links high strategy, concepts of how to fight campaigns, and low-level
tactics, all of which must be aligned to be successful in war.
Danger on Our Doorstep reflects on the challenges of international
relations within the Indo-Pacific and does so in some detail. A key recent
influence on our thinking in this regard has been the Russia–Ukraine War,
which has shown that very bad things can happen in this world and that
they can happen quickly and with great violence, and that all countries have
an obligation to be prepared for them. Jim is able to explain these complex
issues in the simplest terms, which is a valuable skill for any politician or
author and one recognised by the media, who compete for his valuable
commentary, and by the Coalition’s base.
Given the sensitive and classified nature of the issues Jim addresses
from his unclassified base of knowledge and experience, it will always be
difficult for a minister, especially a defence minister, to express an opinion
on the merits of a work like Danger on Our Doorstep. But I will conclude
this foreword by asking, first, is there anyone as credible as Jim to address
these issues? And, second, how important is it to appreciate that we live in a
country that not only tolerates different views and opinions, but exuberantly
defends our right to express them? In that context, this book also represents
some of the priceless values we believe are worth fighting for in Australia.
Peter Dutton, April 2022
INTRODUCTION
We had a saying when I was in the military that bad things happen at night,
in rain and on the corners of four maps. It refers to a time when wars were
fought by commanders looking at maps on paper. In fast-moving
operations, you often moved off one map and onto another very quickly, so
you needed lots of them. If you found yourself at a corner of one map, it
was almost guaranteed that the enemy would have the temerity to be on one
of the three adjoining maps, and commanders sometimes had to join maps
together with sticky tape, so they ended up about the size of a bedsheet.
I spent a lot of my military life as a light infantryman, often travelling to
battles by foot. Everything I needed I carried on my back, so too many
maps were a real burden. The only thing worse, though, was not knowing
where you were, especially if you had to call for support.
Back in the ‘dark ages’ – the 1980s – I did an exchange posting to a
mechanised British battalion on the NATO front line in northern West
Germany. We could see Russian and East German troops across the border
and hear them on our radio frequencies. We travelled in tracked armoured
personnel carriers, in a total lack of comfort, but our view was that a bad
ride was far better than a good walk. As second-in-command of a company
of these armoured vehicles, I had hundreds of paper maps. We had to be
prepared to move just about anywhere in northern Europe.
Times have changed. The digitisation of just about everything except
the rifle and the bayonet has meant that maps can be carried electronically
or downloaded to individual commanders when required. Nowadays they
are displayed on the equivalent of a mobile-phone-sized personal computer,
worn on a soldier’s body. All information is captured from space in digital
form, then disseminated through the internet or an intranet.
Over 20 years after my time in West Germany, when I was Chief of
Operations for the Coalition Forces in Iraq, all maps were digitised and
displayed on screens. Through sheer habit, I still tended to print certain
maps out and jam the paper copy in the ‘map pocket’ of my pants, just in
case the system in my vehicle, known as Blue Force Tracker, failed.
Invariably, I would find the paper map unused in my pocket after a day or
so, throw it away and replace it with another paper copy.
To produce the maps we worked with in Iraq, every part of the country
was put through a process of ‘mensuration’. I pretended to know what
mensuration meant when speaking to smart young US military surveyors,
but I took it to mean that every digitised part of the surface of Iraq was
fixed by us in space. From my headquarters I could launch a bomb or
artillery round or helicopter raid, and everyone involved would know
exactly where they were in relation to everyone else. Knowing where
everyone is becomes almost the most important thing when you start
throwing explosives around.
In Iraq we were lucky. We could move information around our
battlefield or between any two points on the surface of the Earth with mindnumbing speed. We had no highly technical enemy opposing us. Our
biggest enemy in doing this was normally ourselves. If some lowly staff
officer decided that a video of a football match should be seen by all his
mates across the operational area, most of our precious bandwidth was
consumed. Orders and intelligence would then be distributed by our system
so slowly that on several occasions operations had to be delayed or
postponed. Each week the US commanding general would announce the
‘Bandwidth Pig of the Week’ to try and discourage irresponsible use. The
hallmark of every headquarters was the large number of satellite dishes,
pointing at some satellite in the heavens.
That was 2004 to 2005, a long time ago in terms of data wars. Iraq was,
comparatively speaking, a very small and manageable war. The next war
will not be like that, because we cannot guarantee our next enemy will be so
accommodating. Not only satellites will be used, but also the underground
and undersea cables that currently carry our internet, distributed through
vast networks of router centres, each the size of a city building. And in the
next war they will be one of the first targets of our enemies, as they are the
target now of cyber hackers.
We have potential adversaries who are planning and training for exactly
these kinds of attacks. In our part of the world, the Indo-Pacific, an
aggressive China has the capacity to do this right now. When China talks
about the incorporation of the nation of Taiwan into the People’s Republic
of China by force if necessary, this is shorthand for war. Most of us think
this will be a geographically limited war, involving parts of China’s coast,
the Taiwan Strait and the island of Taiwan itself. If we do ever think about
it, we Australians anticipate reading about this war in our newspapers or
listening to talking heads on TV. The conflict will take place far from our
shores – and life will go on as normal.
But this is not necessarily so. It is more than likely that the first thing
we will know about a Taiwan war is that we will not know much about it at
all. All our depended-on means of receiving information through the
internet and satellites will fail in a single event. And if China decides to act
very seriously, at the same time as it attacks us in real space and through
cyber space, it will use its extraordinary rocket and missile forces to attack
US forces in the region. Having done this, China will have pushed the US
out and will become the dominant power in the Western Pacific. As for
Australia, for the first time since 1942, we will face a region dominated not
by our great and powerful friend America, but by a superpower that has
since 2016 openly indicated its dislike of us, and twice indirectly threatened
us with nuclear attack.
Such a surprise attack has occurred before – at Pearl Harbor in 1941.
The US recovered after that but, especially in the early years of World War
II, the Allies were also very, very lucky. We will always need luck in
wartime but the more we prepare for the right war, the less luck we will
need. War with China is possible, and more likely than most leaders are
saying, but it is not inevitable. And the more we prepare for the right war,
the greater our chance of deterring it, or, in the worst case, minimising its
impact. But unless we prepare for the right war, not for the one we would
prefer to fight, it might be a very good idea to keep a few of those old paper
maps handy. In a fight with China, a lot of the electronic devices that drive
our plans and our weapons will become as useful as paperweights.
The aim of this book, then, is to answer these questions: what might a
regional war with China be like, and how should Australia prepare? Faced
with such frightening notions, many Australians will adopt the hope that
something as disastrous as this will never happen. They might feel that if it
does, Australia should stay aloof. Regional war represents a threat that is
recognised by few in the government or bureaucracy, and even less by the
Australian people. But given the recent war in Ukraine, which has made
stateon-state warfare seem much more possible, and China’s increasingly
aggressive stance, recognition by Australians of the threat is certainly
expanding. After 75 years of relying on the US, Australia has paid for our
amazing prosperity by overindulging in globalisation and sacrificing much
of our security, and the last thing any of us wants to do is think and talk
about war.
I have thought and talked about war for 40 years as a member of the
Australian Defence Force (ADF). That was my profession. I have run a real
war as Chief of Operations in Iraq and spent years as a soldier in Asia and
the South Pacific. I believe I have an intimate knowledge of the US at war
and at peace, having trained with the US military at the highest level and
fought side by side with the US in Iraq. I also spent five years as a diplomat
in Australia’s embassy in Indonesia, during the most difficult period of our
relationship with that extraordinary country. As I write this, I am a
backbench member of the Liberal–National Party Coalition government,
which allows me to call myself a partial insider. I have been engaged in the
defence and security debate internally with my parliamentary colleagues,
and externally, through the media, with the people of Australia.
In this book I aim to tackle themes that are not being addressed
elsewhere or widely acknowledged by Australians. Our country may be
dangerously weak, but the road to strength does not lie in tackling only one
military deficiency after another – submarines, ships, fighters, cyber
technology, industrial bases, fuel supplies, military culture – no matter how
good the solution to each of these challenges might be. There is danger on
our doorstep, and our salvation as a nation lies in aligning high-level
strategy with the details of how to fight campaigns (called ‘operations’ in
the military), all the way down to low-level tactics. And the place to start is
with what is called a national security strategy, a process of strategic
thinking that is used by many of our allies. The first step in such a strategy
is to identify the problem, and in national security strategy thinking, the
problem is the kind of war that a nation is preparing to fight. Get that wrong
and just about everything else is wrong.
My approach here is to describe what some may consider to be a worstcase scenario of war in order to explain how and why I think this ‘worst’
case is not only possible but likely, and to suggest in this real-politik
assessment what Australia’s best options are – to avert through deterrence
and, if that fails, to respond effectively. This book attempts to answer the
question that the Australian government and military refuse even to ask
publicly, much less resolve: what war are you preparing for? Only by
asking and answering this question can Australians decide whether their
government is preparing effectively for the right war.
PROLOGUE
OUT OF THE BLUE?
It’s about 10pm on the east coast of Australia when the news services across
the nation start to drop off. Within a few minutes, they’re gone. Many
Australians are already in bed, with few venturing out on this wet, windy
night. The year has seen an exceptionally warm and wet summer, but at
least COVID-19 and its variants have kept their distance. The last few years
have been memorable for all the wrong reasons – drought, fires, floods,
plagues, war in Europe, an aggressive superpower in the Western Pacific,
even a volcano in the South Pacific, a weak leader in a partisan US, and
autocratic leaders across the world becoming more aggressive. At least
Australians can comfort themselves that the economy, given the
circumstances of world inflation and very high energy prices, has boomed
initially after COVID and is still performing quite well, with energy
resource prices staying very high. At this hour, in this weather, bed is the
only place to be, and as a result very few see the unusual cascade of flashes
or ‘shooting stars’ in the heavens in those places where locally the sky is
clear.
Newsroom updates on radio and free-to-air TV include early reports of
earthquakes somewhere off the Chinese coast or the nearby Ryukyu Islands,
detected by seismographs across the world. The aftershocks are confusing
geologists because they show no correlation with previous volcanic activity.
Certainly, China’s coast is part of the Ring of Fire around the Pacific, so
geological activity is to be expected, but still there is no pattern, no
consistency, no leadup, just three very loud disturbances. Staff in
newsrooms wonder whether a tsunami warning will be issued by the
Chinese or Japanese governments; that certainly would be news.
When the emergency generators kick in and the lights come back on,
journalists switch on their computers, allow them a few seconds to boot,
then enter their IDs and passwords. Nothing happens. This is not a routine
outage. The bulletins that can’t be broadcast that night won’t be issued for
several weeks, and by then geological activity will be the least of anyone’s
concerns. No one, not even in the most secret corridors of government,
where backup generators quickly restore the lost connection, can see what
the disruptions are, or how far they will go.
At first no one suspects this is an act of war. A few unconnected reports
of unusual military and other occurrences on the Chinese mainland have
been received by intelligence and media, but most of these are from the
previous year. If anything, things in China have quietened down. Those
looking for ‘the deep, still waters of universal peace’,1 as Australia’s
Robert Menzies once described them, are even saying that China has been
walking back on some of its inflammatory demands. Others think that China
might still be absorbing the lessons of the Ukraine War, of which there are
many, especially for autocracies. The aggressive form of ‘wolf-warrior’
diplomacy China practises in its embassies across the world, designed to
intimidate its neighbours, has recently been tempered to a mild form of
undiplomatic abuse. Australia has even been sent a new ambassador from
China, who in his first public speech was more than reasonable and raised
the hopes of many for eternal peace and unlimited trade.
The US, like Australia, is fairly relaxed about security, although its
surveillance satellites are still constantly monitoring China and Russia
(when they are not dodging space junk caused by China and Russia). Nor is
the US particularly fazed by reports of civilian Chinese ships taken off
routine schedules; of more than the usual number of missile firings on
ranges in China’s west; of a slight increase in the number of incidents in
space, such as the movement of Chinese satellites into orbits that disrupt
other satellites; or by an abnormal number of cyber-attacks, probably by a
national actor (a country such as China, rather than a criminal group or
individuals). These developments can all be explained away.
No one has detected any unusual movements of strategic missiles or
nuclear warheads. As far as Western intelligence can tell, there is no
worrying number of China’s submarines at sea, and China’s navy seems to
be spending a greater amount of its time in port. North Korea is doing what
it always does, blustering and threatening and every now and again firing a
test missile into the Sea of Japan. The Russians in Ukraine and a nucleararmed Iran are considered the biggest current problems for the US.
Artificial intelligence and quantum computing, still in the development
stage as predictive tools, have been reporting nothing of great interest, and
what has been reported has been discounted as teething problems with
these new technologies. Much the same thing had happened with a then
brand-new device called ‘radar’ that was installed on the hills overlooking
Pearl Harbor at exactly the time of the Japanese attack in 1941. That new
device detected the Japanese attacking aircraft at great range but the
warning was rationalised away; the idea of a Japanese attack was beyond
comprehension. Just as everyone thought, until Ukraine, that war between
great powers, especially in Europe was a thing of the past.
By the time the world’s internet starts going down as a result of massive
inexplicable failures and unprecedented levels of cyber-attacks, even the
surveillance satellites capable of reporting anything have been destroyed by
Chinese space-based weapons or other weapons blasted into space from
China’s mainland and ships at sea. Some of the defence intranets around
the world survive, and through these networks it becomes clear that space
and internet assets are under purposeful attack; that maybe this is more
than just a technical hitch. On the Australian east coast a glance outside
military bunkers or media studios shows a sky filled with the ‘shooting
stars’ of satellites burning up. There are lights in the sky, but no light at all
in our computer world.
Around the world, control centres watch as the internet tries to revive
itself through the automatic rerouting of data. This is supposed to be the
strength of the internet, designed as it was to survive a Russian nuclear
attack during the Cold War. Yet it still depends for 97% of its traffic not on
satellites but on the undersea cables that thread their way through the
depths of the world’s oceans. Within a few hours, many of these have also
gone down, broken by specialist submarines or ships, in addition to space
communication and land links. Those cables that remain are considered
unreliable; suspicion is high that they have been rerouted, possibly to
China, and everything important on them is likely to be read by super
computers and foreign analysts.
In national emergency centres and strategic military command centres
around the world, the conclusion is very soon reached that some form of
major attack has been carried out. But even if this is the case, who has done
it, and what can be done in response?
On whatever communication circuits are still operating from London
and Brussels to Delhi and Sydney, local officials and military officers are
asking questions. Regardless of the answers, action is very limited. In
nations where Cold War nuclear attack has been expected in past decades,
such as in the UK and the US and in former Eastern Bloc countries, old
plans are dusted off and old procedures reinitiated, essentially with
everything reverting from centralised to local control. The US nuclear force
puts itself on the highest state of alert, but again, no real action can be
taken because only the President can initiate the use of nuclear weapons in
a counterattack. Only the President has total visibility of the situation – or
that is how it is supposed to happen. In this case, when every action so far
detected cannot be confidently attributed to a particular country, who
should the President defend against or attack?
As communication systems across the world progressively go down, the
US national command system with the President at the top is not even
aware that, by midnight, Chinese missiles are being readied for launch in
their fortified hiding places across China, to be fired at a selection of
targets throughout the Western Pacific and elsewhere, such as Diego
Garcia, a US military base in the Indian Ocean, and possibly even
Australia. These missiles are not detected at launch because most of the
assets that would normally do so, such as early-warning satellites, have
been rendered inoperable or, if they are still working, cannot communicate
with their controllers. The launch of the missiles against disparate targets
has been coordinated by China so that they impact almost simultaneously
on the US air force and naval bases in the Western Pacific, including
armouries in those locations holding US nuclear weapons, and on Diego
Garcia.
The Chinese missiles are carrying only high-explosive warheads, not
nuclear ones. But they have an accuracy of 5 to 10 metres through mid-
course correction and do not require a large number of hits on each base to
be effective. Ten or so missiles targeted at key points at each base are
enough to render these critical facilities inoperable for a long time into the
future, and to destroy aircraft and ships there at the time of the strike. Just
as in 1941 in Pearl Harbor, US aircraft, including fighters, tankers,
surveillance aircraft, bombers and intelligence-collecting aircraft, are
mostly parked in neat straight lines clearly visible from space. The Chinese
targeteers – the military operatives who coordinate complex attacks – do
not make the same mistake as the Japanese when they attacked Pearl
Harbor, by failing to destroy the fuel storage tanks which later refuelled the
replacement aircraft carriers that ultimately defeated the Japanese Imperial
Navy.
With no expectation of an immediate attack the US bases have been
operating to a peacetime routine. Their commanders have been screaming
for years about how vulnerable they are to such an attack, but Congress
and the President denied them the funds to build an adequate number of
shelters to protect the aircraft, to ring their airfields, ports and facilities
such as armouries or fuel storage tanks with defensive missiles, or to
disperse their aircraft to other locations – in other words, to ‘harden’ their
bases. What good is an aircraft carrier without aircraft? What good are
fighter or bomber aircraft on local bases without tanker aircraft? What
good is shipping in replacement cargoes of fuel, spare parts and weapons if
there are no port facilities? What good are air refuelling tankers without the
fuel to fill them? And finally, the biggest question of all, what good is US
military power in the Western Pacific without the bases from which to
project that power?
Once hit, the ‘flight lines’, as the aircraft parking spaces are called,
burn for days and make reinforcement of the Western Pacific by the US
from its worldwide forces almost impossible, or at least immensely complex,
slow and vulnerable. Just to make sure that any essential targets do not
survive, a second wave of cruise missiles is fired from ships, submarines
and large bombers. These missiles are soon winging their way at just below
the speed of sound and just above the surface of the sea towards what is left
of the US bases. Their job is to prevent any chance that the bases can be
brought back to a functioning state. As the waves of missiles approach their
targets at sea level, they climb suddenly so that their radars and optical
systems can look down and lock onto their targets for the final few seconds
of their flight. In most cases they smash into burning piles of wreckage, but
their psychological effect, where there is enough daylight to see them as
they rise in waves and then attack with merciless accuracy, is as significant
as the physical damage.
In or near bases, local commanders, staff and emergency services are
completely consumed for the next few days with handling casualties, reestablishing basic communications back to Hawaii or Washington, and
considering what they might be able to do next. What can they do? They
have no aviation fuel to pour into any new refuelling tankers that might be
flown in from somewhere else, such as the US or Europe. The massive
amount of fuel that these bases consume can be supplied by local civilian
refineries but only for a short period of time, and only if the piping and the
storage is at least partly working. Most bases depend on shipping to bring
in bulk supplies of petroleum products, and that will take weeks or months,
not days.
Nor do the bases have repair facilities now for the military equipment
they support. Modern aircraft demand a high level of maintenance by very
skilled maintainers with a guaranteed supply of spare parts, and ships are
even more demanding. Even if the destroyed attack aircraft (also known as
fighters) can be replaced from other parts of the world, they will need their
own maintenance crews and spare parts, which will have to be flown in by
transport aircraft before the attack aircraft arrive. And nothing at that scale
and across the mighty Pacific can happen fast.
Moreover, the bases’ stocks of bombs and missiles have all been
destroyed, so even if by superhuman effort they were able to obtain the fuel,
the aircraft and the maintenance staff, they have no armaments. And unless
the US can establish anti-aircraft and anti-missile units around the bases to
defend against the kind of missile attacks they have just suffered, the
Chinese can simply re-attack the bases when they see them approaching a
revived operational state.
***
The US has 11 nuclear-powered attack submarines active in the Pacific at
the time of the devastating attack. A few of these are in the vicinity of
Taiwan, and several others are tracking Chinese ballistic missile
submarines (known as ‘boomers’) in the East and South China seas. The
‘boomers’ have not tried to launch nuclear missiles, so no action is taken
against them by the trailing US submarines. One of the US submarines in
the navy fleet, armed with a large number of missiles originally designed to
attack targets on land but now also capable of attacking ships, is on patrol
near Taiwan and monitors strange rumblings in its vicinity. The listening
devices on board this submarine are so sensitive they can detect whales
talking to each other oceans away, so, unlike the geologists, the crew know
that the noises they hear are not volcanoes or earthquakes because, when
the initial explosive noises die away, the submariners can hear the sounds
of other ships or submarines breaking up under water. But even though the
crew have an idea of what is happening, there is no quick way to report it;
the sophisticated comms systems, which depend on space communications,
very low frequency radio waves and then various cyber systems, are so
degraded as to be inoperative.
What the US crew know from what they can hear is that China has
probably detonated nuclear depth charges under water (the one exception
to China’s self-imposed non-nuclear rule) and that the other three US subs
may have been destroyed. The detonations occurred a little while before the
main attack on the US bases and were coordinated as much as possible with
the attacks on space satellites and cyber-attacks on the internet. While
China knew roughly where the US submarines in the vicinity of Taiwan
were, they did not know precisely. The massive blast power of the nuclear
depth charges meant any submarine in the wider area would be destroyed.
These nuclear depth charges are a killer weapon that the US, the UK
and the Russians developed to use against each other during the Cold War.
The US and the UK apparently stopped their development – at least
unclassified sources said so. The Chinese secretly took up the idea, and by
this stage have a number of working models deployed either by ship or
aircraft or linked to seabed sensor arrays that listen to all the noises of the
ocean and can detect even the quietest submarines.
The US Pacific Fleet has only two groupings of ships deployed in the
region at the time of the attack, which is fairly normal for the past few
years. The first is a carrier strike group based around the USS Gerald R.
Ford, which consists of seven warships and one submarine plus a logistic
support ship. This group entered the South China Sea through Luzon Strait
in the Philippines, intending to split up and make visits to various ports in
the Philippines to celebrate the re-establishment of the defence pact
between that nation and the US. It then planned to exit the South China Sea
and head back to more distant waters – not because it might be under
greater threat from China than normal, but because the US government did
not want to annoy China while ongoing negotiations on climate agreements
were in progress.
The second group consists of seven ships carrying a US Marine Corps
unit of several thousand soldiers, all their vehicles, aircraft and equipment
– known as the expeditionary strike group. After six months away on a
deployment in the Persian Gulf, living on board their ships as marines do,
they are transiting back to their Japanese home base and, after a brief stop
in Perth, are threading their way through the Indonesian Archipelago. Like
the carrier strike group, the expeditionary strike group is proceeding in a
peaceful manner, unaware that its location is being intensely monitored by
China’s satellites and that its fate is sealed.
All the ballistic and cruise missiles that China launches on this night to
attack the US ships have maximum ranges of between 1000 and 5000
kilometres. They can reach out into the Pacific as far as Guam
(approximately 4000 kilometres away) in the south-west, and as far south as
parts of Papua New Guinea (approximately 5000 kilometres away). They
include a missile referred to as the ‘carrier killer’, which the Chinese claim
can hit and destroy a US aircraft carrier – a particularly difficult target as
it moves and manoeuvres on the ocean. All the missiles attacking the ships
in these two battle groups are conventionally armed with non-nuclear
warheads and up to a tonne of high explosives.
The missiles launch against the ships just as the US command system
across the world loses communications. Carrier-killer missiles from China’s
east and south coasts are fired at the larger ships in both battle groups, with
backup from smaller cruise missiles from Chinese ships and submarines in
the vicinity, and from China’s old but usable H-6 bombers, which each fire
two of the enormous anti-ship cruise missiles they haul into the air under
their wings.
The conflagration among these two battle groups is reminiscent of naval
battles in the Pacific during World War II. There are no mushroom clouds,
but crimson explosions bloom as the missiles penetrate the flight decks of
the carriers, fuel is ignited and weapons magazines explode, tearing these
gigantic ships and their thousands of crew members to bits. Several of the
accompanying cruisers are also hit and break apart. The wrecks burn red
until they sink, as smaller supporting ships move in to try and rescue troops
or train hoses on the fires. But these ships too are soon hit by missiles and
lost. The cost in human lives is appalling.
The aim of the Chinese is not to destroy every ship in the two groups, or
even a large number of ships in the Pacific Fleet, because they know that
the US can bring more ships into the Pacific. But they have sent a clear
message: we are the dominant power in the region. We can destroy the US
Navy.
***
It is not just an attack on US forces that occurs on this disastrous night,
however. Strikes are also launched against US allies Japan, South Korea
and Australia. China’s attacks on US bases in Japan kill tens of thousands
of Americans and many thousands of Japanese on the bases or in their
vicinity. China does not attack Japan’s military forces, which are not
insignificant and of high quality, reasoning that its own missile stocks are
not infinite and that some have to be held in reserve, and that the US will be
so hurt in the Western Pacific by the initial attacks that Japan will not fight
on while the bases are burning and the Japanese first responders are still
digging out their casualties. It is a critical assumption that remains to be
tested. But even if Japan does not accept the inevitable – Chinese
dominance of the region – China still has the ability and the opportunity to
use force later if it is needed.
South Korea, another key US ally, is not considered by China to be a
force that needs to be dealt with at this stage, because north across the
border there are one million North Korean troops who represent the most
imminent threat to that country. Both Japan and South Korea will be docile,
China reckons, if the US forces in the Western Pacific are destroyed.
Australia is a slightly different matter. The Chinese hope – indeed
expect – that the US will now acknowledge China as paramount in the
Western Pacific. This is the most logical course of action from China’s point
of view. The US can withdraw all the way to its hemisphere and redefine its
sphere of influence as east of Hawaii. The American people, China reasons
and hopes, will not be prepared to shoulder the burden of world leadership
as they did during World War II, and pay the price of hundreds of thousands
of dead.
Yet China cannot blindly rely on this hope. It has to consider the
possibility that the US might fight back, and the most likely way to do this
(without using nuclear weapons) is by assembling both its allies and its
military forces from around the world. Trade sanctions and other USinitiated actions in the UN will of course occur, but the US will have to
prepare itself for offensive action sometime in the future or surrender world
power to China. Mobilising military forces and shifting them to the Pacific
from the US and from Europe will take time and will need a firm base, a
trusted ally somewhere in the region closer than Hawaii. Australia is likely
to be that base, China has reasoned, and it will therefore have to be dealt
with in this first round of attacks.
To this end, eight Chinese submarines have been deployed to four
critical locations around the Australian continent. Cruise missiles launched
from the subs are easily able to destroy Australia’s fleet of F-18F Super
Hornet fighters, Growler electronic warfare aircraft and most of the F-35
fighter aircraft sitting on criminally unprotected bases in the north and the
east of the country. The attack is of a different scale from that prosecuted
against US bases, but it is as effective against Australia’s small but very
technologically advanced air force as the larger missiles are against US
bases. As Australians try to work out what is going on in a world deprived
of its normal means of communications, and as its defence forces try to
switch to backup systems that are slow and intermittent, the government
receives information that Australia is under attack as well, and its small but
beautiful air force has been destroyed.
Over the next day or so, the Chinese submarines silently move to new
attack positions. Lurking just off the coast, they dispatch automated and
self-propelled sea mines to the vicinity of key military facilities in the
harbours of Darwin, Sydney and Garden Island south of Perth. These
advanced, unmanned undersea weapons, with sensors able to identify any
type of ship passing by, travel to a predetermined position and stay there.
They are programmed to destroy any warships that attempt to move out of
the harbours, effectively locking them in. No one has any idea how many of
these sophisticated mines have been deployed and it takes some time to
sweep the harbours clear of them, and free the Australian fleets.
Joint US and Australian communication facilities in Australia remain
physically untouched; the Chinese logic is that so many other parts of the
US early-warning and submarine-communications system have been
destroyed that it is not worth wasting longer-range missiles or cruise
missiles from submarines on other targets, especially those far inland on the
Australian continent, such as the sites of the over-the-horizon earlywarning radar collectively known as Jindalee.
China has taken a step from which there is no retreat. It knows that after
its space and cyber-attacks, after striking US bases in the region, destroying
two US battle groups and several nuclear submarines, but particularly with
the killing of a large number of Americans, it has redefined the power
struggle in the Western Pacific and perhaps across the world.
CHAPTER 1
RISE TO POWER
It is my belief that a war with China – perhaps resembling the scenario
outlined in the previous pages – is possible, and that it is more likely than
most Australian commentators, and indeed most world leaders, are saying.
These are not just my views but the views of US and other military
commanders and regional defence ministers.1 And a growing number of
Australian politicians subscribe to these views too.
War is not inevitable, but whether it occurs or not is likely to depend on
the strength and resolve of the US and its allies, including Australia. This
question of likelihood must not be dismissed by the false hope that war is so
appalling that it could never occur, or the probability is small enough to be
ignored. We need a strategy based on the real world, and not on how we
hope the world is.
China and other regional powers are armed and are talking tough, and
many have competing interests.2 But why is China most likely to be our
foe, how has it attained a level of military power capable of threatening the
global balance of power, and what would it be seeking to achieve by
triggering a regional conflict?
Like Germany and Japan before it, the military power of the People’s
Republic of China (PRC) has increased exponentially under its more
modern authoritarian leaders,3 and China is now searching for spheres of
influence in Asia and beyond and trying to control critical technologies and
resources. It is not only the leaders who have created this militarisation of
the state but many causes working together: strong centralised leadership by
presidents and the Chinese Communist Party (CCP); strong memories of
recent experiences of aggression towards China, which reduced it to an
effective colony for over 100 years; an ideology around the CCP that put
the party and common good before the rights of the individual; technology
and industry, some of it stolen from its competitors, that is as good as
anything in the world; the ability to control populations through technology
rather than through coercion; and the willingness of the people to accept
that the military and the civilian spheres of their lives are one. Prussian
militarism and Japanese imperialism had nothing on the militarisation of
China under the CCP, especially in the years since Xi Jinping became
president in 2013.
Chinese soldiers on parade in the Forbidden City in Beijing. China has
the world’s largest army, navy, militia, non-nuclear rocket force and
homeland air defence. (Getty)
China under the CCP claims special legitimacy as a many-thousandyear-old nation with exceptional culture and traditions, and regards the PRC
as merely an extension of ancient China. BBC journalist Bill Hayton in his
2020 book The Invention of China addresses these points, and characterises
the political ideology of the CCP as ‘national-socialism with Chinese
characteristics’.4 How else, he asks, to describe a system featuring a ‘core
leader, insistent demands for natural homogeneity, intolerance of difference,
rule by party not by law, corporativist economic policies, a focus on
discipline and an ideology based on tactical exceptionalism – all backed up
by a massive surveillance state’?5 Others have continued the Nazi Germany
analogy. Journalist Daryl McCann notes in a review of Hayton’s work, that
‘the unbridled ambitions of Xi Jinping’s foreign policy reverberate with the
hubris of the Third Reich in the lead up to the second world war’.6
Like Nazi Germany, communist China is a modern construction, not a
unique 5000-year-old civilisation, although Han chauvinism is certainly not
identical with Nazi Aryanism. ‘Exterminationist antisemitism’ (a hatred of
Jews involving extermination of the race) does not lie at the heart of
President Xi Jinping’s so-called Chinese Dream. That said, notions of racial
superiority are key to the CCP’s modus operandi.7 Just ask the Uighurs,
and the Tibetans, both peoples whose rights have been severely impinged.
China is a revisionist nation that sees its place in the family of nations
as the dominant power at least in this region, and ultimately in the world.
China does not accept the status quo that has made it rich and powerful, a
status quo built on the security provided to the world economy and to its
trade routes by US power and influence, including military power. China
has demonstrated that it has no respect for international law and
conventions through its claims in the South China Sea, the militarisation of
islands and reefs contrary to international law, and its treatment of its
populations in its Muslim areas, in Tibet and in Hong Kong.8 Being
dominant is strongly related to the continuance in power of the CCP. And
this is a zero-sum game for China. China must reduce US power and
influence to increase its own.
Modern China is a highly sophisticated, technologically advanced,
militarised, industrialised nation of 1.3 billion people, with the biggest
sense of insecurity in the world. Part of this insecurity is based on the
appalling abuse inflicted on China not only by the West, especially the
British during the Opium Wars, but also by Japan during its murderous
invasion before and during World War II. The years between the
unravelling of the Qing Dynasty, beginning in 1839, and the ascent of the
CCP are known as China’s 100 years of national humiliation. This sense of
insecurity, shared by the Chinese people as a whole, is reinforced regularly
by CCP propaganda. China will never forget or forgive.
China’s technological and industrial base has produced a military
machine of daunting capability, supported by civilian efforts. The West won
the Cold War against the Soviet Union because its industrial and
technological base was far greater. But China has been so successful in
modern industrialisation that the difference between soldier, scientist,
technologist and civilian is hardly perceptible. China’s national aim is to
ensure it is never in such a position of weakness that it can be so humiliated
again. And this seems reasonable to most Chinese.
Of course, the party still has to consider the mood of the people, and it
has done this fastidiously. The State has given the people as much
prosperity as it can afford, while, with few exceptions, its citizens have
gratefully and proudly accepted this prosperity at the cost of certain
individual liberties. They have done this because they, their parents or their
grandparents remember how much suffering the Chinese people endured in
earlier times, or because, as with most younger people, they have been
constantly told of this suffering by the party. Like good people everywhere,
the Chinese just want to live in peace, enjoy life, educate their children and
increase their prosperity. The price for doing this, they believe, is to
faithfully support the CCP.
Although today’s nation of China – as reinvented by the party – is as
modern a creation as any other nation in the world, the Chinese people
believe in the nation’s myths and legends. Perhaps the most enduring of
those myths is that the island of Taiwan is rightfully part of China. The CCP
has staked part of its legitimacy on the question of ‘reincorporating’ Taiwan
into the PRC and has been saying for decades that Taiwan will be returned
to China, even if force must be used.
For thousands of years, Taiwan was a home to local tribes and was seen
as a source of natural resources including gold and sulphur, which were
mined by Chinese interests and, in more modern times, various colonising
powers. In 1662, after the Manchus overthrew the Ming Dynasty on the
Chinese mainland, the Ming loyalists retreated to Taiwan and threw out the
foreign miners, mainly Dutch, but were themselves defeated 20 years later
by the Manchu Qing Dynasty. In 1885, the Qing empire designated Taiwan
as China’s twenty-second province. In the First Sino-Japanese War of
1894–95, China was defeated by Japan and ceded Taiwan, which the
Japanese then occupied until 1945.
After Japan’s defeat in World War II, the island was taken over by the
Chinese Nationalist Party and returned to China’s control. But after the
Chinese Communists defeated the Nationalists in the Chinese Civil War, in
1948, the Nationalists retreated to Taiwan and established the island as a
base for their plan to reconquer the Chinese mainland. The PRC, led by
Mao Zedong, was determined to liberate Taiwan from the Nationalists but
was unable to do so mainly because of the presence of US forces in the
region, especially US naval forces in the Taiwan Strait. Taiwan, which was
never part of the PRC, has been an independent country since that time –
much to the chagrin of China.
Another strongly held belief, and one of the most contentious
international issues for China, is that it should control the South China Sea.
This stretch of water, along with the East China Sea, is considered to be of
crucial importance to the country’s strategic security. Since the late 1940s,
China has claimed all of the South China Sea as its sovereign territory. It
bases its claim on maps produced by the Chinese themselves in the 1920s
and 1940s, one of which showed a nine-dash line running from Taiwan
south along the coast of the Philippines, swinging west past the island of
Borneo, then turning north past the Vietnam coast to Hainan.
The nine-dash line encompasses many island groups claimed by other
nations, including the Spratly Islands and the Paracel Islands. Using the line
to justify its actions, China has in recent decades constructed military
facilities, including harbours and airstrips, on some of these islands. The
validity of China’s claim over this maritime area was legally tested in 2016,
when the Permanent Court of Arbitration in The Hague ruled that China
had no claim. China strongly rejected that determination and continues to
use the nine-dash line to validate its ongoing attempts to control these
territories and support its view that what the rest of the world calls
international law is irrelevant.
CHAPTER 2
PLAYING CATCH-UP
Until the twenty-first century, China had limited military tools that it could
use to achieve its geopolitical goals. But recently it has made astonishingly
rapid progress in developing and expanding its military, and it now has
vastly greater firepower and far more strategic options at its disposal.
Tracking this development in a secretive and strongly authoritarian
country like China from outside is no easy task. To assess progress, it’s
essential to constantly monitor press reports and official statements and
deduce from these and the appearance of new rocket systems, nuclear
submarines, stealth technology and the like when such programs might have
started, how they progressed, who ran them and so on. I’ve been doing this
for the last three decades and the following account is based on that long
experience.
China’s level of technology in the late twentieth century meant it could
control the air and sea approaches to China and Taiwan only to the limits of
its ability. That in turn meant that if it wanted to retake Taiwan it would
have to send its forces by air and sea across the Taiwan Strait to disembark
on the island’s shores, in a similar way to the D-Day landing of 1944, then
secure the island and wait for the US and the rest of the world to react.
Military leaders assumed that the US response would come initially
from nuclear submarines in the vicinity of Taiwan sending up flights of
conventional (non-nuclear) missiles from below the surface, as far as 1000
kilometres away from their targets, or from long-range torpedoes fired from
the vicinity of the Taiwan Strait. China’s analysis showed that this might
cost China up to one-third of its invasion force over a couple of days. And
for some time, as China put together a military that had the capacity to
invade Taiwan, it was prepared to accept this loss; there was no other
choice. It also accepted the losses that would come to such an invasion
force when US air bases in Japan, South Korea, Guam, Diego Garcia,
Alaska and Hawaii, or even further away, mounted the expected retaliatory
strikes on Chinese bases and air and sea invasion fleets.
However, through the 1990s and early 2000s, as China’s mastery of
space, cyber, missile and rocket technology enhanced the warfighting
capacity of all branches of the People’s Liberation Army, as Russia’s
willingness to provide weapons that China could copy increased, and as US
power did not expand in line with China’s, broader military options in
relation to Taiwan came to be considered. Furthermore, as China’s industry,
technology, infrastructure and military expanded, and its cashed-up
businessmen and entrepreneurs spread throughout the world, buying and
trading and giving ballast to the emerging superpower and to its military, so
did the country’s confidence rise.
Yet even as its military capabilities grew, there were still some
assumptions to which China remained committed. The first was to avoid
war if possible, but to prepare every part of the nation, not just the military,
for war if it became necessary. The second was that if force were to be used,
it would not involve nuclear weapons, again unless absolutely necessary. A
third consideration, however, was that if China was to be dominant, at least
initially in the Western Pacific, the US would have to be dealt with there.
The fourth was that war, if required to deal with the US and regain Taiwan,
was likely to cause the largest geopolitical upheaval in a century. The
biggest question for the Chinese leadership in considering contingency
plans for a regional war was how to incorporate these four theoretical
considerations within a practical plan for military action.
Over the course of decades, the Chinese had reasoned that the best way
to ensure any war would stay non-nuclear was to strengthen its own nuclear
rocket force as a deterrent. This was gradually achieved. It took the US
intelligence services several years to determine that China’s nuclear
warhead tally had passed 400, something they announced only in 2021. In
all probability, China by then could have been at 1000 warheads, a figure
that the Western media had previously predicted China would not attain
until 2030.
Chinese military leaders studied the war methods used by the US in the
Persian Gulf during 1990 and 1991 and reached some very important
conclusions. The first was to acknowledge the decisive US victory over
Iraq’s forces, who relied on Russian weapons and tactics, as did China at
that time. The US achieved its victory using its own Cold War–era weapons
and tactics, which were far superior to anything China possessed. China
also noted how long the US took to build up its military strength in the
region – about six months – and that, when it did finally act, its victory was
almost total. It stalled only because the US did not want to risking losing its
allies, and because US leadership objected morally to the final destruction
of the Iraqi armed forces as they streamed helplessly out of Kuwait.
China concluded it could imitate and improve on US tactics when the
time came for its own attacks. It admired the coordination of military action
that the US almost took for granted, and also saw how dependent the US
was on the so-called revolution in military affairs – that is, the harnessing of
civilian information technology for military use. China realised it could
learn these operational techniques, or it could fight in a way that would
render such techniques less important, especially in a war with the US in
China’s backyard. The US had achieved great things in its reliance initially
on technology and industry and more recently on cyber and space
technology, and the Chinese would have to do the same to exploit US
vulnerabilities.
In terms of military hardware, initially China concluded that it could not
compete with American aircraft technology as demonstrated in the Gulf, at
least not for a decade or so, but that it could compete in the areas of missiles
and rocketry. So China focused on these areas, producing a world-leading
rocket and missile force. Soon after the year 2000, China turned its
attention back to its air force and began to develop aircraft that were
competitive with US combat aircraft, especially in relation to the reliability
of engines, computing technology and invisibility to radar (known as
‘stealth’), often based on information technology stolen from the US and
Europe.
Following its success in developing and modernising its industrial base,
infrastructure and human capital, China soon had much more to protect than
in revolutionary times. It had long been confident that its massive tunnel
systems would safeguard its military assets until they were deployed for
action. Now it also hardened its east-coast military bases, making them as
impervious as possible to missile, rocket or bomb attack, and enabling them
to recover quickly if such an attack took place. Its submarines were safe in
their pens and could enter and leave their base on the island of Hainan
under water. There even arose a belief that, because of the size of its
population and its preparations, China was the only nation on Earth that
could sustain a nuclear attack and, as a nation, survive.
The only real vulnerability that China had in this period when its
military power was still somewhat limited was its surface fleet, which could
not be shielded and would have to be deployed and manoeuvre to protect
itself. So the country developed an air-defence system, based on short- and
long-range missile systems initially procured from Russia, which it copied
then manufactured locally. These missiles could cover some of the most
important parts of China and its eastern cities, while mobile units could
dominate Taiwan and its approaches from as far away as the Chinese
mainland. In the event of an attack on Taiwan, this would at least weaken
the impact of US air power and provide some protection for China’s fleets
when they were in their ports or even while transiting the Taiwan Strait.
As China steadily increased its firepower, its military planners must
have been astonished that the US increasingly dismissed China as a
potential threat and dedicated almost all its research and development and
military innovation to counter-terrorism and counter-insurgency, rather than
pouring money into the rocketry, fighters, naval self-defence, strikehardened regional bases and large military forces that were necessary to
maintain its advantage in the Pacific sphere. The real deficiencies for the
US in the post-Reagan era started with the Gulf War of 1990 and carried
through the Afghanistan War of 2001 and the Iraq War of 2003, which wore
out both national resolve and military equipment at great national cost. The
development of key future technologies was delayed in order to fight
‘today’s wars’ in Iraq and Afghanistan. The decline of US military capacity
continued under President Obama, who reduced the defence budget
severely, and it was not helped by President Trump, who talked a strong
game but never quite got around to spending effectively on defence. There
is no sign of a Reagan-like president in the wings, who is willing to focus
the US on what is really important.
All through this period the US military was aware of its weaknesses,
revealed in wargame after wargame. A few generals broke their silence, and
a few politicians tried to get Congress to focus, while the US military
attempted to do even more with even less. In its inimitable manner, the
military did its best to solve the problems essentially created by US civilian
leadership. It re-examined the way it might fight a war with China with
inadequate forces by making changes to its operational concepts – that is,
how it fights. But clever concepts and tactics need to be based on even
smarter strategy, and that has been missing in the US, and in many allied
countries, in recent times.
One thing that Chinese military strategists could see as they studied
their adversary towards the end of its involvement in Iraq and Afghanistan
was that the US just did not have enough attack aircraft. The limited range
of most US aircraft, almost all of which were older than the best Chinese
aircraft and designed for short-range wars in Europe, negated the ability of
the few US stealth planes that could avoid radar detection. Even if a small
number of US stealth aircraft could get through the Chinese missile screen,
bomb a target and get out, the most important Chinese targets were now
either impervious to such attacks because they had been hardened, required
repeated attacks to breach, or were strongly defended by missiles. Such a
situation meant that the risks that China faced in achieving its aims in the
Western Pacific were becoming less daunting.
China’s military leadership also recognised that Pacific geography and
the long distances US aircraft had to travel from regional bases to targets
made them totally dependent on aerial refuelling. The US air refuelling fleet
had expanded as a result of the need to prepare for a potential war in Europe
or South Korea and was truly enormous, consisting of over 700 adapted
civilian airliners. In an air battle close to China’s shores, these lumbering
tanker aircraft would sit in a line a few hundred kilometres from the targets
of the US attack formations in order to refuel aircraft going into China and
then coming back out. China realised that the refuelling tankers were a
weakness and that it had the missile technology to attack them. China’s
regiments of older fighters could transport the latest long-range missiles to
launch points and fire them as the tankers flew lazy circles in the sky,
waiting for returning fighters and bombers.
If China were able to destroy US air tankers during a conflict in the
Western Pacific, US fighters would soon run out of fuel and plummet
into the ocean. (Getty)
So Chinese wargames concentrated on this vulnerable tanker line, rather
than on the stealthy attack aircraft. Even the stealthiest of US aircraft
needed refuelling to get back to base after an attack, and if many or all of
the tankers had already been hit by China’s air-to-air missiles, these superb
US fighter aircraft, designed for shorter-range wars against Russia in
Europe, would splash harmlessly into the ocean after they ran out of fuel.
There would be few follow-up attacks, and the US stealth technology –
which China soon gained after stealing it from the Americans – would rot at
the bottom of the sea.
The Chinese were aware that if they invaded Taiwan they would be
subject to attack by aircraft from US aircraft carriers. It’s publicly known
that only one US carrier attack group, with its ‘air wing’ of 70 aircraft, is
likely to be in the Pacific at any one time during routine peacetime
deployments. But if the US is given warning, it can deploy up to three
aircraft carriers relatively quickly and, over time, even more. Chinese forces
therefore needed to be strong enough to deal with at least one carrier strike
group immediately, and maybe one or two more in the short to medium
term. As Chinese missile and rocket technology became more and more
sophisticated, this goal appeared attainable.
Carrier strike groups suffer from the limits imposed by Pacific
geography on all forces. China gradually built up enough accurate missiles
to destroy most of these ships – as long as it could defend its space-based
targeting system. It does not matter how accurate a missile is if the location
of its target is either not known, or the target, possibly an aircraft carrier
underway on the high seas, moves after the missile is launched. The initial
location of a target would be detected by surveillance satellites, or by very
sophisticated long-range land-based radars located in China itself, and this
data then passed to the launch sites. Such targeting data is sufficient for the
missile to be fired, but if the target is moving, the aim point of the missile
has to be continually adjusted while the missile is in flight. This is
technologically very difficult to do, and because it depends so much on
space systems, any such technique is vulnerable to weapons that might be
used against it in space.
If US aircraft carriers moved to points from where they could launch
attacks by air against mainland targets, they would be well within the range
of China’s missiles. In surface-fleet action, the advantage goes to whoever
fires first, and who has the longer-range weapons. And the key advantage in
missile range has been moving in China’s favour for some years.
For some time, the greatest dangers to the US surface fleet have been
two types of Chinese missile that have gained fearsome reputations around
the world. The DF-21D ‘carrier killer’ carries up to a tonne of conventional
explosive, and has an accuracy of 20 metres over thousands of kilometres,
even if the target is moving. It can be fired above the Earth’s atmosphere
from the Chinese mainland and travel at many times the speed of sound,
and its flight path can be corrected mid-course. By the time a fleet-onfleet
engagement is possible, US ships within range will have been attacked
decisively. The second kind of missile, the DF-26, or ‘Guam killer’, is the
largest air-launched cruise missile in the world and is carried by a Chinese
aircraft to its launch point, from where it ‘cruises’ to its target, staying
within the atmosphere. The US Navy, the greatest navy in the world, is now
vulnerable if it chooses to fight within range of these missiles – if the US
wants to become involved in a Chinese invasion of Taiwan.
China’s military planners have also grown increasingly confident that
they have the technology to attack US surveillance and communication
systems in space, at least during critical periods. At the same time, their
massive cyber capability can be focused on preventing the US and its allies
from conveying critical warfighting data around the world. Even if some
US or allied systems in space survive a Chinese attack, once China gains
control of ground links through cyber-attacks or other actions, the
surveillance systems will be rendered incapable of conveying what they see
or hear to terrestrial command centres. China also has the ability to secretly
redirect the seabed communication cables that carry most of the world’s
supposedly secure internet data to monitoring centres that study and collect
useful information. In addition, China has planned for specially designed
submarines to simply cut the cables at vulnerable points. The advantage that
the US has enjoyed for many years through its advanced technology and
industry no longer exists.
The last challenge for China’s military planners with regard to potential
wars in the Western Pacific was to align how it would fight – its
‘operations’ – with the tactical capability of the specific weapons and
techniques that progress had delivered. Once that was resolved, the generals
could confidently offer China’s political leadership war plans to achieve
their national strategic objectives. No longer did China have to accept that it
would have to absorb US attacks on its mainland bases and its deployed
invasion forces during a D-Day-style invasion of Taiwan. Instead, it could
link two of the most potent techniques of war – deception and surprise –
using its missiles and other weapons, giving it the ability to spring a
devastating attack on US regional bases that would remove US power from
the Western Pacific and leave China free to resolve the Taiwan issue in its
own time, without any interference.
Even after these goals were achieved, Chinese generals were unlikely to
make a case for war to the president; that’s not the culture of this
authoritarian nation. But it might be different if a near-deified leader asked,
‘Why attack Taiwan first and then wait for the US, the region and the world
to react? Why not attack US bases, which we know to be vulnerable, as
well as the US fleet if it approaches? And only once US bases are put out of
action and the US fleet is damaged, if not destroyed, then attack Taiwan?’
If they were given the opening they were secretly hoping for, the
generals would begin coordinating their complex plans. One of their first
objectives would be to create the kind of national resilience that China
would need if it were to fight a war. That would require China to build up at
least a year’s reserves of strategic materials, including microchips, or to
establish alternative sources of supply that could be trusted.
For commercial reasons, China has already commenced this process,
amassing natural resources such as coal, oil and iron ore. It has opened up
rail lines into Russia and is shipping iron ore and coal across the Amur
River border to reduce its dependence on resources from US allies like
Australia. And instead of being imported via vulnerable shipping routes
through international waters or US-controlled territories, oil from the
Middle East and from the Central Asian nations known as the ‘Stans’ is
now flowing through pipelines, straight into China.
CHAPTER 3
WOLF WARRIORS
No other country in world history has moved up the power ladder as fast as
China has over the last decade or so. From coming into being in 1948 at the
bottom of the ladder as a primarily agrarian nation and economy with next
to no industry, China has lifted the majority of its people out of poverty and
has created a sustainable economic society with a level of industry and
technology that is the envy of the world. It is far from perfect but stands in
contrast to so many countries in Asia and Africa that have started from a
higher base yet have not reached the heights of China’s achievements.
Today China can comfortably sustain a military that is regionally dominant.
And now, disturbingly, this extraordinary nation is becoming irrationally
aggressive.1
When President Xi arrived on the scene as paramount leader in 2013,
China was already melding its civilian and military worlds into one. As a
powerful dictator in an authoritarian system, Xi could ensure that if the
Chinese state wanted something to happen, it would happen with the
greatest of rapidity. And while for much of its history the CCP had relied on
fear and coercion to achieve its authoritarian aims, now it could place
greater emphasis on incentives such as access to a prosperous life for a
large number of people. The China Xi has created is far different from that
of the nation headed by the pragmatic Deng Xiaoping, who opened China
up to the outside world, and promoted modernisation of the economy and
what was essentially a form of capitalism – ‘socialism with Chinese
characteristics’, as it was known.
Xi has made a profound difference to China by enacting an aggressive
foreign policy, supported by so-called wolf-warrior diplomacy, the effects
of which have been seen already in China–Japan relations, in Africa and in
Eurasia, and in the trade sanctions that were applied to Australia after its
leaders supported an inquiry into the source of the COVID-19 pandemic.
We saw Germany wage war in 1914 to prevent its hegemonic
aspirations from being crushed by a British–Russian–French entente, and
Australia was drawn into that conflict at the cost of over 50,000 killed, even
though we were from the far side of the world. In 1941, in our
neighbourhood, Japan waged war to prevent the US from choking its
empire, and this war was a direct threat to Australia. Japan conducted what
was supposed to be a lightning war to gain access to Dutch East Indies’ oil,
which it did successfully. Japan then tried to establish an economic bloc in
the area of its conquered states known as the Greater East Asia Coprosperity Sphere, which at that time included most of Manchuria and
China.
In recent times, China has been acting in ways that are frighteningly
similar to Germany and Japan in those two world conflicts. In both those
cases, either individual leaders and/or their parties provided decisive
leadership that led to a national desire for empire. In China’s case, the
appearance of President Xi as the paramount leader was decisive. Under Xi,
China has claimed the South China Sea as its own, citing dubious historical
justification; encouraged the advance of Chinese interests into Africa and
other parts of Asia and the Pacific; increased social control of ethnic
minorities such as the Uighurs in Xinjiang region; and clamped down on
democratic movements in Hong Kong. Furthermore, since 2014, gambling
on the weakness of any US-led opposition, China has engaged in a truly
massive nationwide construction effort in the South China Sea, building
islands that it has then duplicitously fortified, militarising the area and
pushing China’s controlled territory another 1000 kilometres out from its
coast. Not only has this improved China’s coastal defences, but it also
means China can control all civil shipping through this area when it wants
to. And it has started to apply strictures to international shipping already,
demanding a reporting system and more information on individual ships
than it has any right to collect.
China keeps testing the waters to see how much can be gained short of
the massive use of armed force. As well as its total control of Hong Kong
against every assurance given and its illegal militarisation of islands in the
South China Sea, examples of it trying to have its own way include its
bullying and its sanctioning of trade restrictions on other countries, as well
as its actions on the Indian border. China has also opposed a Philippines-led
push for a review of its 1951 Mutual Defense Treaty with the United States,
claiming that it is an effort to contain China’s rise. The threat to use force is
ever present. According to a Reuters news report: ‘The push for clarity on
Washington’s commitment [to the treaty] comes amid a rapid build-up of
Chinese maritime assets in contested areas of the South China Sea,
including what the Philippines says is a militia disguised as a massive
fishing fleet near Beijing’s militarised manmade islands.’2
The vast majority of Chinese people, who have some knowledge of the
recent troubles in Hong Kong involving the ruthless suppression by Chinese
security forces of the democracy movement and the trashing of
international agreements, nonetheless support China’s intervention and
cannot understand why the privileged people of that special province are
not thankful to the party for the riches they enjoy. The Chinese people are
proud that the State is taking back the South China Sea. Of course, the
people do not want war, but if the State needs war to ensure their security,
they will trust and support the State.
***
China understands deterrence and is not unhappy that information on its
nuclear weapons program has been made public across the world. To deter a
first strike by its enemies, China has to be capable of a credible second
strike, and likely adversaries need to know this. Initially, the nation’s
nuclear force was focused on the US, but, thanks to their shared border, the
Chinese have long considered Russia the most threatening enemy, despite
the spring of friendship that bloomed at the time of the Ukraine War.
Until the Russians found themselves bogged down in Ukraine, they
were prepared to say anything to encourage China to take on the US but
were actually ambivalent about who would win. Russia simply wanted US
power reduced across the world, particularly in the Baltic and within
NATO. The economic synergies between Russia with its vast reserves of
natural resources and China with its manufacturing power, plus the
autocratic government systems of each nation, were, however, natural
attractors. Declarations of everlasting friendship between Russia and China,
and personally between President Xi and President Putin, were made early
in 2022 while President Xi was presiding over the Winter Olympics in
China – and just as Putin was moving troops into the border area of Ukraine
and issuing a series of demands to the US and NATO. After a postOlympics summit, Russia and China released a joint statement warning
Europe and the US that NATO forces should not approach the borders of
the Russian Federation and asserting that Ukraine lay within Russia’s
sphere of influence.
At the Winter Olympics, early in 2022, President Xi Jinping and
President Putin made declarations of everlasting friendship and loyalty
– as Putin was positioning his troops to invade Ukraine. (Alamy)
When the mighty Russian army crossed Ukraine’s borders on 24
February 2022 and attempted to decapitate the Ukraine government and so
control the whole nation, as it had succeeded in doing in Georgia in 2004,
most of the world expected a quick victory. But this time the decapitation
strategy failed and left the conscript Russian army clumsily fighting on at
least four fronts. There was a breakdown of logistical support for the
masses of troops because the Russians had banked on a short, sharp
campaign, and the West was astonished by the valour of the Ukrainian
military and people fighting the Bear. The fact that the war turned out to be
neither short nor sharp meant that NATO was able to ship weapons to
Ukraine, including the most sophisticated anti-tank guided missiles and
portable air-defence missiles, which, at least in the north and northeast of
Ukraine, stopped the Russian army. This delay in a conclusion to the war
also meant that the democratic nations of the world could not only put
together the most daunting set of sanctions ever seen outside of the world
wars but could also assess whether the sanctions were working or not, and
then adjust them as required. As the Russian people started to feel the
consequences of President Putin’s rash actions on their standard of living,
the Russian army reverted to a massive and indiscriminate use of firepower,
the traditional floundering way of war for Russia. At the time of writing, the
war grinds on with little sign of a victory for either side, or any kind of
negotiated settlement. The question for the Western Pacific nations, and
especially for Australia, is what lessons will China choose to learn from
Ukraine in relation to the use of force?
The Russian attack on Ukraine has made the US starkly aware that it
has at least two major enemies in the world, China and Russia. Of the two,
China is still considered the most threatening in the medium to longer term,
given its power in the Pacific. Russian aggression in Eastern Europe was
always something the US hoped could be handled by Europe and NATO.
During the Cold War, the US had accepted that it had to maintain armed
forces capable of fighting and winning two major wars and one minor war
simultaneously. But since the end of the Cold War in 1991, by the US
government’s own admission, its armed forces have been manned, equipped
and funded to win only one major war, and to ‘hold’ in one minor war.
After the Russian invasion of Ukraine, when the US cast its glance back to
China, Taiwan and the Pacific, it found itself facing two enemies with a
one-enemy military.
***
Despite its increasing influence on the global stage, not everything looks
rosy for China. Annual economic metrics show that its growth has begun to
slow, which is inevitable as it develops. China feels itself strategically
encircled by the US, India, Japan and South Korea, and by new alliances
such as AUKUS (the strategic agreement between Australia, the United
Kingdom and the United States), and to some extent sees Australia as part
of that threat. There is some argument that China might be reaching the
peak of its power, which will decrease over time or at best plateau for an
indefinite period.3
The question is whether this makes China more dangerous or less
dangerous. Michael Beckley and Hal Brands write in Foreign Affairs
magazine:
If there is a formula for aggression by a peaking power, China
exhibits the key elements . . . Beijing is a strong revisionist power
that wants to remake the world, but it’s time to do so is already
running out. This realisation should not inspire complacency . . .
just the opposite. Once-rising powers frequently become aggressive
when their fortunes fade and their enemies multiply. China is
tracing an arc that often ends in tragedy: a dizzying rise followed by
the spectre of a hard fall. Revisionist powers tend to become most
dangerous when the gap between their ambitions and their
capabilities start to look unmanageable. When a dissatisfied
power’s strategic window begins to close, even a low-probability
lunge for victory may seem better than a humiliating descent. When
authoritarian leaders worry that geopolitical decline will destroy
their political legitimacy, desperation often follows.4
If China has peaked as a power, what will it now do? Will this make it more
dangerous, and how should the world react? These are critical questions for
Australia and for our allies.
Few question China’s worldwide ambitions, but in his book Unrivaled:
Why America Will Remain the World’s Sole Superpower, Michael Beckley
certainly doesn’t think China is in a position to dominate the world.
Beckley, a fellow in the International Security Program at Harvard’s
Kennedy School, delves into the available economic, demographic and
military data on the Chinese regime to determine whether it can really hope
to become a global power in the same league as the US. His answer? Such a
scenario is extremely unlikely. Rather than heading in the direction of
rivalling the US, China is a rapidly ageing, inefficient, conflict-ridden and
relatively poor country that simply is not on the road to seriously
challenging the US’s hegemony.5
A view like this, expressed in general terms, that China is not likely to
threaten US dominance, is likely to be misunderstood. It allows us the
luxury of reinforcing a strategy of hope that regional war will not occur. But
this is not what Beckley is saying.
In the military, when we are planning for an uncertain future, we define
the ‘bounds of feasible planning’. Events that are within those bounds can
be predicted with some degree of confidence and can thus be planned for.
Events outside those bounds are so dependent on the success or failure of
earlier events that, except in the most general sense, planning for them is
impossible.
World dominance by China, as addressed by Beckley, I consider to be
outside the bounds of feasible planning and will leave for others to
examine. However, regional dominance by China, in my view, is well
within such bounds and must be examined. It must form the contingency
against which those involved in the region should plan. Regional war
cannot be dismissed because at present China is not in a position to
dominate the world.
Regardless of whether the world may consider that China’s power is
peaking and plateauing, a vital assumption made by Xi Jinping’s
administration is that China is on the way up the power ladder while
America is on the way down. Daniel Russel, a top State Department official
under President Obama, has said, ‘The strongest driver of increased
Chinese assertiveness is the conviction that the Western system, and the US
in particular, is in decay.’6 China may not be world-dominant yet, but it is
quite capable of a regional test of strength in its own backyard. And if
China is able to be dominant in the region, this might increase its view of
whether or not it could ultimately be dominant across the world.
Furthermore, President Xi has been talking tough with regard to local
geopolitics, as Wall Street Journal columnist Walter Russell Mead noted in
October 2021:
Tension over Taiwan has been mounting for months. In a major
speech commemorating the 100th anniversary of the founding of the
Chinese Communist Party, delivered in Tiananmen Square in July,
[Xi Jinping] promised to ‘utterly defeat’ any attempt toward
Taiwanese independence . . . Over the weekend of China’s Oct. 1
National Day, a record 149 Beijing military aircraft crossed into the
island’s air-defense identification zone.7
To up the ante, Beijing hawks in the Global Times newspaper, the CCP’s
mouthpiece, have focused on the presence of a very small number of US
troops in Taiwan involved in training the Taiwanese military. The paper’s
editor-in-chief referred to the presence of US troops in Taiwan as ‘a red line
that cannot be crossed’ and warned that in the event of war in the Taiwan
Strait, ‘those US military personnel will be the first to be eliminated’.8 The
Global Times has also stated, in its attempts to keep the tensions high, that
if Japan defends Taiwan, then China will make Japan an exception to its
rule about never using nuclear weapons against a non-nuclear nation.9
What China says, even if it is through its Global Times mouthpiece,
should not be dismissed as just rhetoric or paranoia. If you are paranoid –
and China is exhibiting strong indications that it might be – it is not difficult
to find people or organisations that look like they hate you and against
which you eventually feel you should act. Even if this is rhetoric based on
deep paranoia, there is no reason to disbelieve that at some stage it will
develop into violent action. Statements such as this one from the Global
Times must at least be a basis for future contingency planning.
China’s paranoia is manifest at the moment in the expansion of its
military to a historic level, and at a historic rate. The CCP has said exactly
what it intends to do – in certain circumstances China will use military
force – and it has made the judgment that it has the military power to act at
any time. China can afford to focus its military plans on the island chains to
its east; in contrast, its chief rival, the US, accepting the worldwide security
responsibilities that have fostered a secure and prosperous world since
1945, must spread its military capability across the world. The key strategic
question now is: if China has the military power to act, what exactly can it
do, and when might it do it?
CHAPTER 4
IN THE GREY ZONE
At present, it is generally acknowledged that China is involved in what is
known as grey-zone conflict. This is conflict above the level of normal
competitive activities such as trade and lawful influence, but below the
level of war. These grey-zone activities have consisted of punitive trade
actions, irrational demands made through wolf-warrior diplomacy, unlawful
influence over politicians and institutional leaders, intimidation of ethnic
groups, denial of international law, physical occupation of disputed
territory, militarisation of contested regional terrain, and general coercion,
threats and bluster.
China has been making increasing use of non-military means of
conflict, particularly coercive diplomacy and cyber-attacks. Australia has
experienced more than its fair share of such activities, but we are not the
only target. According to the Australian Department of Defence, over the
past 10 years there have been 152 cases of such coercion affecting 100
governments and 52 companies, with an exponential increase in such tactics
since 2019.1
It is not only the resources of China that can be brought to bear as this
grey-zone conflict continues. China has so-called friends and, as mentioned
earlier, Russia, as a main China ally and also a Pacific power, would like to
see China diminish the power of the US in this or any part of the world, if
for no other reason than it will distract the US from Russia’s activities in
Ukraine. We have not yet seen a major power, much less two major powers,
conduct the highest level of cyber-attack possible on another country. The
cyber-attacks by Russia on Ukraine have not been as disabling as predicted,
and the reason for this may not be known for some time. But Russia and
China combined have massive cyber influence, and their common friends,
Iran and North Korea, are active in this field as well.
It is now widely known that in 2015 Russian hackers tunnelled deep
into the computer systems of the US Democratic National Committee, and
subsequent email leaks had a significant impact on US politics. But to
interpret the impact of such activities as only political is to miss the bigger,
more important story. In that same year, as David Sanger describes in his
book The Perfect Weapon: War, Sabotage, and Fear in the Cyber Age, the
Russians not only broke into networks at the White House, the State
Department and the US Joint Chiefs of Staff offices, but also placed
implants in American electrical and nuclear facilities that potentially give
them the power to switch off vast swathes of the country.2
This was the culmination of a decade of escalating digital sabotage
among the world’s powers, in which Americans and their allies became the
collateral damage as China, Iran, North Korea and Russia battled in
cyberspace to undercut one another in daily just-short-of-war conflict.3
Prime Minister Scott Morrison indicated in June 2020, in a very public
statement, that all levels of Australian government, critical infrastructure
and the private sector were being targeted in a ‘sophisticated state-based’
cyber-attack, and confirmed it had been conducted by ‘a sophisticated statebased cyber actor’, assumed by commentators to be China.4 In the past
decade or so, cyber-attacks have displaced terrorism and North Korea’s
nuclear missiles in the public’s mind as the most critical threat to
Australia’s vital interests.5 The Coalition government reacted to this by
progressively increasing Australia’s offensive and defensive cyber
capabilities and, in a massive leap forward, nearly doubled those
capabilities through an AUD$10 billion investment over ten years in the
2022 budget.
China has long been known for attempting to steal intellectual property
and industrial secrets through cyber intrusions. Chinese authorities and
‘netizens’ routinely conduct low-level attacks against information
infrastructure: they silently penetrate systems, map their security and
connections, and prepare for further penetrations in times of conflict. A
recent shift, though, has been towards using cyber technology to inflict
physical damage. In a notable recent example, China mounted a cyber
campaign against India’s electrical power grid to coincide with a clash
between the two nations on the border in Ladakh.
Cyber technologies have given China some major advantages when
undertaking coercive diplomacy, and its grey-zone activities can now have
an impact on distant nations, not just those in its region. Chinese grey-zone
activity even extends into space, where China has tested weapons and
created significant debris fields that make some near-Earth orbits unusable.
According to Bill Gertz in the Washington Times:
The PLA [People’s Liberation Army] plans to use outer space to
project power far from its shores and defeat adversaries, including
the US, in a future conflict . . . Its space warfare arms, dubbed
counter-space weapons, include a wide range of systems. The PLA
has begun to transition the arms from testing to deployment. The
weapons include ground-launched ‘kinetic kill vehicles’ – precisionguided missiles that slam into satellites at extremely high speeds.
Other space weapons are orbiting robot killer satellites, directed
energy weapons, electronic jammers and cyberwarfare
capabilities.6
In July 2021, The Australian reported on assessments by UK and US
military leaders that China and Russia are carrying out threatening actions
against Western satellites daily, as part of a ‘reckless’ pattern of behaviour.
According to General Sir Patrick Sanders of the UK’s Strategic Command,
Moscow has been manoeuvring satellites ‘close to what they know is a
sensitive or important satellite to us’ to force a Western satellite to move to
a less strategic area.
The Chief of the UK Air Staff, Air Chief Marshal Sir Mike Wigston,
has stated that a Russian spy satellite has been gathering intelligence on
commercial and military satellites.7 Wigston said that the UK cannot yet
see everything that Russia has been doing, but what they have been seeing
is ‘questionable’. Wigston also accused China of developing missiles to
target satellites, laser-dazzle weapons to ‘blind’ targets, and electronic
jammers. Further, Wigston revealed that Russian weapons can ram other
satellites: they ‘are practising against their own redundant satellites . . .
When we talk about reckless behaviour in space, these are the things we
see.’8 Wigston’s assessment was complemented by reports from the Deputy
Commander of US Space Command, Lieutenant General John Shaw, who
said: ‘If we were in a potential conflict scenario, we would be relying on
space, the United Kingdom would be relying on space. If the Chinese take
out our space capabilities, it could affect our ability to wage war in the
terrestrial domains.’9
The same military chiefs suggest that a future war would be won or lost
in space, because the ability to destroy another country’s satellites could
have an immediately devastating effect. This is a critical judgment for these
military leaders to make in this specialist area, and their views are doubtless
well known to China. Given this, and the fact that Australia is part of the
Five Eyes alliance, which shares intelligence between the US, the UK,
Canada, New Zealand and Australia (a grouping commonly known as ‘the
Anglosphere’), any attack on a satellite belonging to one of the other four
members of the alliance would have to be regarded as a direct threat to
Australia.
***
China has also employed its maritime militia as part of its grey-zone
activity. Properly known as the People’s Armed Forces Maritime Militia
(PAFMM), it is a very large, irregular force of fishing and other civilian
boats controlled by the State. Some crew members are armed, but with
concealed weapons. They are sometimes referred to as ‘little blue men’, a
phrase borrowed from the militia forces used by Russia during its 2014
occupation of parts of Ukraine, who were known as ‘little green men’.10
A Philippine coastguard ship sails past a much larger Chinese
coastguard ship off Scarborough Shoal, a triangular reef in the South
China Sea that is claimed by the Philippines, China and Taiwan. (Getty)
China’s maritime militia has been widely used to intimidate other
nations when the use of identifiable military vessels might be too
provocative. In 2012 the militia participated in China’s illegal seizure of
Scarborough Shoal from the Philippines. In 2014 it was used to repel
Vietnamese ships from an oil rig that China had stationed near the contested
Paracel Islands. And in 2015 the militia attempted to prevent a US destroyer
from exercising its right to free passage in international waters, known by
the US as a freedom-of-navigation operation or FONOPS, near an artificial
island illegally constructed by China, called Subi Reef.
All these activities are clear signs that China wants to impose its will on
other nations, but not through violence just yet – mainly because there is a
chance of prevailing through other means. Grey-zone conflict is likely to
increase in the short term. Possible activities include even more severe trade
embargoes, unattributable biological attacks, anonymous cyber-attacks and
mysterious interruptions to space-based capabilities.
But while China has achieved a number of objectives through grey-zone
conflict, such as making the nine-dash line a realistic territorial boundary,
this strategy is unlikely to bring about the main prizes of securing Taiwan
and pushing the US out of the Western Pacific. China will need to do more
to achieve those objectives, and we should be very concerned about that.
CHAPTER 5
WHERE AND WHEN?
The most obvious war scenario for Australians to consider is what is
commonly called the Taiwan scenario.1 Most commentators seem to
misinterpret this scenario as one that is geographically limited to the region
around the island of Taiwan and the Taiwan Strait, and concentrates on
whether a Chinese invasion force could move across the strait by sea and air
in order to invade, given Taiwan’s defences and the ability of the US to
intervene.2 I am not saying that this scenario could never occur; what I am
arguing is that it is dangerously simplistic and portrays Chinese strategists
and tacticians as fools, which would be a silly assumption for planners to
make.
If China were to move on Taiwan only, leaving US air and naval power
in the Western Pacific untouched, it would risk a long and damaging
counteraction by the US. Given time and the willingness of its allies, the US
might be able to assemble its not insignificant regional military power and
mount immediate and sizeable counterattacks from local US and allied
bases. Then, over time, the US and its allies might be able to concentrate
their world power by assembling additional forces in the region and apply
that power against China. China might still be able to win this kind of
limited D-Day-style Taiwan conflict, but absorbing US air and naval attacks
over a period of time, as Saddam Hussein found out twice in Iraq, would
involve enormous costs to its military and to its infrastructure on the
mainland.
A more realistic option for China, which addresses both the Taiwan
question and another of China’s most important strategic objectives, the
expulsion of US power from the Western Pacific, is the one I have
presented as a hypothetical scenario in the prologue. If China could push
US forces out of the region and deter them from attempting to approach its
coastline, then China would become the dominant regional power. At the
same time, this would simplify almost all the problems of ‘reincorporating’
Taiwan into the PRC.
If China decides to invade Taiwan and limits itself to that objective
only, it will certainly attract the enmity of the US and the rest of the world,
as well as military attacks from the US and others that have military forces
in the region. But China might just consider, as Japan did at Pearl Harbor in
1941, that if you are going to be hated, it is worth going all the way and
removing the regional ability of the US to use force against you. Then you
can deal with Taiwan in any way you want, and it becomes irrelevant how
much you are hated by the US – as long as the US does not use nuclear
weapons.
Taiwan is important to China in a nationalistic sense because the CCP
has staked its reputation on regaining its claimed lost province. To do this in
the limited scenario most Australians think of as a Taiwan conflict, China
needs either to so intimidate the US that it concedes regional power to
China without fighting for Taiwan, or to entice US naval forces within the
range of Chinese weapons so that they suffer significant casualties and the
US then withdraws its surviving forces and removes its bases that sustain
air and naval activities. To act so decisively, China would have to be as bold
as the Japanese were at Pearl Harbor and take even more risks than them.
For the Japanese it did not pay off. For the Chinese – who knows?
If China threatened Taiwan and the US was not able to resolve the
situation in its own favour, then it would be a severe blow to US credibility
and the region would perceive the US as a diminished power. If the US
refused even to try to fight China because it did not believe its forces were
strong enough, that would confirm US weakness to every regional nation.
From the most recent statements of President Biden, it does not look as
though the US is going to succumb peacefully and hand power to China
without a fight of some kind.
An even worse result for regional stability, however, would be if the US
tried to resolve a Taiwan scenario with force and was tactically defeated
while Taiwan fell into China’s hands. Then the US would be weakened
militarily and in terms of prestige. This option could have frightening
consequences for Australia.
There are other possible conflict scenarios – a Chinese attack on Japan’s
Senkaku Islands, or serious conflict on the China–India border, or conflict
with Vietnam or the Philippines, for example. But if Australia was prepared
for the Taiwan scenario as I have described it, and not as it is generally
misinterpreted, then we would be prepared for all other non-nuclear
scenarios. Or, if we discussed such a scenario and even then elected not to
prepare and to take a risk, at least the magnitude of the risk would be
understood by the Australian government and people.
Really, what we are doing here is trying to predict the future when
almost nothing is out of the question. Sadly, if we have prepared for four
options, the real world might still find a fifth.
If nothing is impossible, then China could continue on its current way,
regularly rumbling about Taiwan and issuing threats to all and sundry, but
not taking military or other actions that the international community would
consider illegal. Life in Australia would not change much. China might
increase or decrease its use of trade sanctions against Australian goods,
leading us to complain to international bodies, but generally the situation
would be manageable.
China could even be overtaken by a popular reaction to authoritarian
rule. The Chinese people could rise up and democracy could break out, as it
did in Eastern Europe in the 1990s, and after a period of internal violence
this might bring about the demise of the CCP. China might become a
normal nation, expressing its power in terms of international competition
for trade and legitimate influence. China might even retain the CCP and
become something like communist Vietnam: socialism or communism with
a real Chinese face, not a CCP face.
China could decide to take military action that is geographically limited
to the island of Taiwan. The naval component of China’s invasion could be
smashed as it set out from the Chinese coast by a conventional missile
counterattack from US nuclear submarines lurking in the Taiwan Strait, and
by Taiwanese US-supplied weapons. Meanwhile, the air element of China’s
invasion force could be shot out of the sky by Taiwan’s air defences, backed
up by US, Japanese and South Korean air attacks from still-intact bases in
the region or from US carrier battle groups. The political ramifications of a
failed invasion could bring down at least Xi Jinping if not the entire CCP.
Alternatively, on a dark night, China’s military, led by an independently
thinking general, might move against Xi, with a similar result for the CCP.
If China changes course as a result of any of the options given above, it
might become even more authoritarian, if that is possible, or possibly even
democratic, or revert to any of the many governance arrangements between
those two extremes. If there is change in the governance of China, as there
was in Indonesia and in Vietnam, and that change is for the better, the world
might enjoy 10, 20 or 50 years before the question of Taiwan’s
independence or a regional war needs to be again addressed.
Or, perhaps China will remain the same aggressive actor on the
international scene yet suddenly the US and its allies decide to change their
own strategy, and Xi is the one surprised as strong allied forces are placed
on the Taiwanese front line as a serious part of the everyday defence of that
prosperous and democratic nation. (More on this in the next chapter.) No
policy of strategic ambiguity there, and the next move would be up to
China.
In the worst case, someone makes a mistake, or for some unfathomable
reason one party initiates a nuclear strike. No one can predict where a
nuclear exchange between China and the US would end, except to say the
deaths would be well into the tens of millions and the world would never be
the same again. Modern nuclear weapons with less fallout than their
predecessors may not destroy the whole world, but the use by either side or
both, and the consequences of such use, are almost beyond comprehension.
China, of course, might suddenly invade Taiwan in some way that we
have not thought of, catching the rest of the world napping. US leadership is
weak and vacillating, so Taiwan might be incorporated, leaving Australia
and the world to wonder where in the waters of the Pacific the democratic
nations of the world will draw the next line.
Or, of course, China may act as I have described in the prologue of this
book. China may give no indication of its intentions and catch out US
military forces in their peacetime deployments in a surprise attack. In a
coordinated offensive over only a few hours, China could destroy US and
allied nations’ space resources, thus blinding the US and limiting its
gathering of intelligence and passing of data so essential to modern military
operations. It might launch a massive cyber-attack supported by its friends
Russia, Iran and North Korea, thereby closing down infrastructure and
communications across the world. It might destroy or re-route undersea
cables, thus switching off or monitoring the internet, which is critical for
passing encrypted military and government data around the world. And it
may use its world-class rocket and missile force to attack local US air and
naval bases as well as US Navy ships at sea, thus destroying US military
power in the Western Pacific, after which it might make Taiwan an offer
Taiwan cannot refuse.
Regardless of the many possible contingencies, it is prudent from a
planning point of view for Australia to look at a situation that is as difficult
as reality can produce, rather than one that comfortably matches our
preconceived ideas. The tendency is to plan for the future war that we
would like to fight, not the one that an adversary is likely to initiate. History
is replete with examples of countries doing exactly that – preparing for the
war they wanted to fight, and not the one that an enemy was able to fight.
***
For Australians to assess whether we are prepared for the Taiwan scenario
as I have outlined it, it is important that we understand how and where such
a war would occur, which requires us to look more closely at the geography
of the area.
The Western Pacific, where I believe this war could be fought, is part of
what is now called, in a geostrategic sense, the Indo-Pacific. Until recently,
the term used was ‘Asia-Pacific’ but the need to consider the area from the
Middle East all the way to the US west coast – covering the Indian Ocean
as well as the Pacific Ocean – as a single strategic entity, means the term
Indo-Pacific is now used widely, even by the US military, who changed the
name of their Pacific Command to the Indo-Pacific Command. ‘IndoPacific’ was a term first used by Australian strategists because Australia
borders both the Pacific and Indian oceans and has an understanding of how
the two oceans are linked in terms of trade and strategic interests.
The Western Pacific is a small part of the Pacific Ocean and consists of
an awful lot of water and some major island chains, including Japan, the
Philippines and New Guinea, many smaller islands and the coast of much of
South-east Asia.
Among the smaller islands out to the east are the US-controlled Mariana
Islands, which include the separate territories of the Northern Mariana
Islands and Guam, where there are massive US air and naval bases and
armouries; the Federated States of Micronesia; and the Marshall Islands,
site of a US naval base on Kwajalein. There is a US Air Force base on
Wake Island, and additional US bases may be established on Palau and
Manus Island and even in Australia in the future. These mid-ocean islands
played a vital role for both sides in World War II.
Closer to China’s coast, in Japan, there is the Okinawa marine base and
Iwakuni marine aviation base, with naval forces at Atsugi, Okinawa,
Yokosuka, Sasebo and Misawa, plus US Air Force bases at Kadena,
Misawa and Yokota, outside Tokyo. South Korea has several bases
(Kunsan, Osan, Yongsan, Chinhae Navy Base and a number of US Army
bases and camps) and of course a permanent deployment of US troops.
Far out in the Pacific, in Hawaii, lies the US Joint Base Pearl Harbor–
Hickam, one of many in Hawaii, and far to the west, off the coast of India,
there is a logistics base and a base for longer-range US bombers on Diego
Garcia.
The China Seas (East and South) define the Chinese continental littoral,
lying between the Chinese coast and Japan, the Philippines and the Malay
Peninsula and extending eastward as far as the island of New Guinea.
Within and beyond the Chinese littoral are three island chains that are
particularly important to visualising a future war.
The first island chain is linked by China to the historical ‘nine-dash
line’, discussed in Chapter 1, and runs from Russia’s Kamchatka Peninsula
to the Malay Peninsula. It consists of the Kurils in Russia, parts of the
Japanese Archipelago, including the Ryukyu Islands, Taiwan, the Northern
Philippines and Borneo. It also includes small islands within the Spratly and
Paracel island groups that have been stolen and militarised by China,
specifically four main militarised island bases and approximately 16 lesser
islands, reefs or artifices on which radar and communications bases have
been built. The first island chain is China’s littoral, and if you elect to fight
China in this area, you fight on China’s chosen battlefield, with China
holding all the advantages.3
The second island chain stretches from the main Japanese islands to
well east of the Philippines, touching Guam, then swings west and south
towards the Indonesian territory of West Papua.
The third island chain, a term rarely used, starts imprecisely in the
vicinity of Alaska, crosses the Pacific to Hawaii and then runs down into
the South Pacific.
Depending on where you stand, these island chains can be defensive as
well as offensive. China experts Toshi Yoshihara and James R. Holmes
wrote in 2018:
If anything, Chinese strategists see [the first island chain] as an
American defence perimeter meant to channel, constrict and
perhaps even block Chinese sea and air movements along the Asian
seaboard and from the China Seas into the Western Pacific. If so, it
is a hostile fortification to be punctured, not a friendly fortification
to be defended.4
Whether these island chains, the first and second in particular, are
defensive or offensive in any future war will depend on who fortifies them
first, when they are fortified and for what purpose. If Chinese forces can be
rapidly located in small groups at key locations, particularly the land edges
of sea straits, with anti-ship missiles to dominate passages between the
islands, then China can push its defence and warning screen out another
1000 to 1500 kilometres. This would give the Chinese an immense
advantage, allowing them to hold a US fleet even further out from its
optimal weapons and aircraft ranges. Conversely, if US forces were able to
be deployed first and hold similar positions with similar weapons (which, as
we’ll see, now appears to be part of the developing US operational
concept), then the US could stop the Chinese from deploying their surface
fleet from Chinese bases and seas to the second island chain, should it want
to do so.5
Though technological advances may mean we now live in a ‘shrinking’
world, one of the toughest challenges in facing down China remains the
tyranny of distance in the Indo-Pacific. This is illustrated when we look at
the unrefuelled ranges of certain aircraft and the ranges of common cruise
missiles as being in the vicinity of 1000 to 1500 kilometres. These are
relevant for the tactical fight, but the distances for strategic reinforcement
are even more daunting. Just look at the distances to Taiwan (Taipei) from
various US bases: from Guam, 2750 kilometres; from Okinawa, 650
kilometres; from Tokyo, 2100 kilometres; from Seoul, 1485 kilometres;
from Alaska, 7505 kilometres; from Diego Garcia, 6422 kilometres; and
from Honolulu, 8114 kilometres.
***
To further visualise a future war between China, the US and its allies,
consider the three island chains in relation to China’s rocket and missile
ranges.
You can see that, with its current short- to medium-range rockets and
missiles (DF-17, DF-21A, DF-21C and DF-21D, the ‘carrier killer’), China
has the capacity to reach out at least into the second island chain and almost
as far as the main Indonesian islands and perhaps even Australia. When you
add in intermediate-range ballistic missiles (DF-26, the ‘Guam killer’),
submarine-launched ballistic missiles (JL-2 surface-launched ballistic
missiles), and intercontinental ballistic missiles (DF-31A, DF-41 and DF5), nothing in our region is out of range. Chinese long-range, nuclearcapable bombers may have a reach of 9000 kilometres, according to some
reports, though the reality is probably closer to half that range.6
A model of the DF-21D ‘carrier-killer’ missile, on display in Nanchang,
China. The DF-21D has a range of over 2000 kilometres and is accurate
to within 20 metres. (Getty)
China has the most active and diverse ballistic missile development
program in the world. It is continually upgrading its missile force in terms
of numbers, types and capabilities, modernising its intercontinental ballistic
missiles, and developing multiple independently targetable re-entry and
manoeuvring vehicles that can dodge missiles sent to intercept them. China
has also begun deploying a new fleet of nuclear ballistic missile
submarines. Short- to medium-range cruise and ballistic missiles form a
critical part of what China calls its regional anti-access and area-denial
capability, a term used to indicate that China can keep US forces out of its
‘area’ – that is, its littoral and at least the first island chain. For example, the
2000-kilometre-range DF-17 rocket can move towards a target at Mach 10
– some 12,200 kilometres per hour. A moving US aircraft carrier would be
unlikely to evade such a missile or to survive a direct hit and a US base hit
by several of these missiles would be taken out of action for days or even
weeks.
China is keen to let the world know it has developed this advanced and
powerful weaponry, a product of widespread theft of intellectual property,
excellent Chinese innovation (especially in copying weapons) and the
allocation of resources that an authoritarian state can make in the areas of
technology and industry. To this end, China’s air force released a video in
September 2020 showing H-6 bombers making simulated strikes on a
runway that looked like Andersen Air Force Base on Guam, a key staging
area for US support for Taiwan.7 According to Sampson Ellis, the CCP’s
Global Times reported that China’s intermediate ballistic missiles, such as
the DF-26, could take out American bases while China’s air defences shot
down incoming US aircraft, rockets and missiles. In August 2020, China
fired four test missiles into the South China Sea that it is assumed would be
capable of destroying US bases and aircraft carriers.8
The intermediate-range DF-26 is often showcased during military
parades in Beijing, for example at the seventieth anniversary of the PRC’s
founding in October 2019. Since the DF-26 can be armed with both nuclear
and conventional warheads, arms-control experts worry that any sign China
is preparing to fire one could trigger a pre-emptive US nuclear strike –
potentially leading to an uncontrollable conflict.9
Whether China will use these weapons to achieve its twin aims of
reducing US power in the Pacific and reincorporating Taiwan, and how they
will be used, have become key questions.
***
Another essential question for Australians to consider is the timing of any
potential war. My estimate is that conflict over Taiwan could occur at any
time within the next three to five years. This was originally based on my
expectation that President Biden, when he was elected, would raise US
military spending and it would take three to five years for this to
significantly increase US military power. If China saw that its window of
opportunity was closing, it might precipitate aggressive CCP action to
achieve its strategic aims before US forces could grow in strength. Under
the Biden administration, however, despite the tough talk, and based on the
2021–22 US defence budget, there is no indication that a significant
increase in US military power is to occur before the next election; in fact,
US support for the Ukraine War may mean that, overall, US military power
diminishes in the very short term. This may make China think twice about
the use of force, and delay any such move, or precipitate an early move.
The unpredictability means that the worst-case scenario – any time in the
next three to five years – must still be the basis of prudent planning.
The timing of such a war might also depend on how ready China is, and
how unprepared and weak China perceives the US and Taiwan to be.
Authoritarian governments have a distinct advantage in being able to
provide resources for their military without having to consider the will of
the people to the same extent as a democratic country.
As Reuters reported, on 6 October 2021, Taiwan’s defence minister
Chiu Kuo-cheng voiced his opinion that China already has the capacity to
invade his country:
Chiu said China already has the ability to invade Taiwan and it will
be capable of mounting a ‘full-scale’ invasion by 2025. ‘By 2025,
China will bring the cost and attrition [of an invasion] to its lowest.
It has the capacity now, but it will not start a war easily, having to
take many other things into consideration.’10
China might act in the aftermath of another failure, such as was recently
seen in Afghanistan, or take advantage of a conflict that diverts US military
attention to a place like the Black Sea, the Baltic, the Persian Gulf or, of
course, Ukraine.
***
The bottom line is that China should not be underestimated. It has the
world’s largest army, navy, militia, non-nuclear rocket force (referred to as
sub-strategic) and homeland air defence. It has not been involved in a war
requiring sophisticated joint warfighting since 1979, when it arrogantly
attacked Vietnam to teach that country a lesson and was soundly beaten and
humiliated. That failure is often seen as highly relevant when assessing the
modern warfighting capability of China and is often used to dismiss China
as a threat. But the US and most of its allies have also not been involved in
the kind of war that might be fought in the Western Pacific as portrayed
here. The US has not fought a peer adversary since Japan in World War II,
or China in Korea in 1950. The US and its allies will have some advantage
in joint warfighting from experience gained in a series of recent counterinsurgency and counter-terrorist wars, but that advantage should not be
overstated in relation to the kind of war China could fight on its littoral and
in the island chains.
It would appear that, following a century of shame forced on it by its
own weakness and the aggression of the colonising world, this time around
China has noticed what strategy and tactics are all about. China is by no
means militarily perfect and does have its vulnerabilities. But no Australian
should ever think that US military power is infinite compared with China’s,
as we have grown to believe over the last 75 years. No Australian should
think that as a Taiwan scenario starts to unfold, our nation need not prepare.
And if the US is defeated in a fight with China or decides to not engage and
so surrenders its credibility and influence in our region without a fight, no
Australian should ever think that the implications for Australia will not be
immense.
US Iraq War commander David Petraeus’s famous comment ‘Tell me
how this ends’ is as relevant here as anywhere.11
CHAPTER 6
THE RESPONSE: THE US AND ITS
ALLIES
In 1941, the world was aware that Japan was on the verge of military action
in the Pacific. Knowing that Japan had only one year’s worth of oil
reserves, the US had cut it off from its oil supplies because of Japan’s
aggression towards China. In order to deter military action on Japan’s part,
in mid-1940 the US had moved a major part of its Pacific fleet from other
bases to Pearl Harbor, thus providing the Japanese with a clear Pacific
target.
Not only did the US arrange its fleet as a target in an eye-pleasing
manner in Pearl Harbor, but it also lined up its aircraft on Hawaii’s airfields
to be destroyed in detail. It was only luck that saved US aircraft carriers
from destruction as well. And all of this occurred at a time when the US
thought it knew exactly where the Japanese fleet was, thanks to access to
some of Japan’s secret coded messaging.
The very day after the infamous 7 December 1941 attack on Pearl
Harbor, the Japanese attacked and destroyed US air forces in the Philippines
while they were still lined up wingtip to wingtip. And just days later, to top
it all off, the Japanese sank two of the best battleships the British had off
Malaya, with losses in the thousands.
The assumption that we will act logically and our enemies will act
stupidly is a very dangerous one indeed. And it is certainly not proven by
history.
Could it happen again?
Arrogantly disregarding the Taiwan scenario as a geographically limited
war between China and Taiwan is reminiscent of the West’s attitude to
Japan before World War II. An unwillingness on the part of US and allied
planners and politicians to accept that China will use all the weapons
systems that it has spent decades developing against US bases and naval
units in China’s backyard is a similarly appalling error of judgment.
The US and its allies must not make the Pearl Harbor error again. Only
by understanding the likely nature of war with China in the Pacific, and
only by planning for the worst while hoping for the best can they deter a
devastating conflict or, if deterrence fails, as it does so often, have some
chance of surviving.
***
Understanding what kind of conflict may lie ahead is particularly important
for Australia, given that we are now more vulnerable than we have been
since the 1930s, years before Pearl Harbor. It is not only the Chinese we
should be concerned about, but also the willingness of our major ally, the
US, to confront them. Even if the Americans are willing to confront China
when the time comes, will they and their allies be able to harness enough
military force to prevail?1
The question arises time and time again: if there is not a high level of
confidence in the US that it could be successful in even a simple Taiwan
scenario, then why would it commit itself militarily? Not stepping in to
defend Taiwan, having provided the Taiwanese with weapons for decades
and warned China against a military solution, would represent a serious
lessening of US influence in the region, and would be a severe blow to
Australia’s security.
To deter a foe, a military force cannot just posture, it must be able to
fight and so impose an unacceptable potential cost on its adversary. If the
US government is not able to look its allies in the eye and say that a
coalition the US might propose to lead has at least a fair chance of success,
why would the US propose such a coalition, and why would anyone join it?
In particular, why would Australia commit to such a deployment given our
paucity of military forces? And why would any coalition partner commit to
such a serious undertaking unless they had a say in the strategy to achieve
success?
Where in the vast Pacific waters would we draw the line on China’s
power, if we do not draw it around Taiwan? Yet, Taiwan’s external
supporters must be prepared to back Taiwan with the lives of their sons and
daughters, because that will be the cost. For most countries, and particularly
the US, at this stage that is a politically difficult notion. The US is war
weary after years of fighting in the Middle East and believes that it was left
unfairly by its allies to carry most of the cost of those long years of war. Its
involvement in support of Ukraine against Russia is yet another drain on
US attention and on US weapons. The last thing it wants to do after its
recent experiences in Iraq and in Afghanistan and its ongoing experience in
eastern Europe is to take over the defence of Taiwan.
As was explained earlier, it is traditional in intelligence circles when
assessing possible adversaries to look at both intent and capability. This is a
very powerful intelligence framework. Despite the importance of the US to
Australian security, I don’t believe that we have looked at our great ally, our
great friend, in those terms with sufficient scrutiny.
Who cares how big and aggressive China becomes in our region if the
US is powerful enough and willing enough to lead its allies in deterring
Chinese actions, militarily or otherwise? Yet since the end of the Cold War
in the early 1990s, even as the US continues to accept its worldwide
military responsibilities, to expand on an earlier point, its effective military
power has reduced by a frightening 30 to 50%. This is clear from its own
National Security and Defense strategies.2
As early as 1998, the US National Security Strategy illustrated this
frightening decline in US power and the magnitude of the task of recovery.
Referring to its plan, for obvious reasons, as the ‘two and a half war
strategy’, it stated: ‘We must maintain superior military forces at the level
of readiness necessary to effectively deter aggression, conduct a wide range
of peacetime activities and smaller-scale contingencies [‘half-wars’], and,
preferably in concert with regional friends and allies, win two overlapping
major theatre [sic] wars.’3
US war weariness from winning the Cold War and from the long
inconclusive wars in the Middle East manifested in a period of defence
budgets that were significantly less than was necessary to maintain US
forces as superior to their peer adversaries. Critics of US defence spending
asked for and received a ‘peace dividend’ – sadly the peace was ephemeral.
During the first two decades of this century, US defence budgets were in the
vicinity of US$700 billion per year (in today’s dollars). A better comparison
during that period was that the budgets were between 3% and 4.9% of GDP.
During the peak of the Cold War the US spent 5.9% of its GDP on defence.
This amount goes up and down depending on whether the cost of the
current war is included in the total figure. In 2005, as the Iraq War was
increasing in magnitude, the budget was US$493 billion, which was 4.09%
of US GDP. When both wars in Iraq and Afghanistan were running down,
President Obama inherited a budget of US$738 billion (4.9% of GDP) and
then ran it down to US$646 billion (3.31% of GDP). The first budget signed
into law by President Biden at the end of 2021 was US$777 billion. What
suffered most from these budget cuts was not just personnel numbers, but
investment in the most modern weapons, aircraft, ships and equipment. It is
often stated that the US spends more on defence than the next at least 16
countries combined, but, although it is a staggering amount of money,
fundamentally what the US spends on defence is irrelevant. This is because
money spent on defence is an input. As the Ukrainians are showing the
world, what is important is not what a defence force has, but what it can do,
what it gets for what it spends – its output, in other words. And the most
important output of a defence force is its ability to win wars. For this is the
definitive judgment and the military is required by US law to state this
judgment once every two years in its National Security Strategy.
By 2018, and as a comparison to its judgment in 1998, reflecting both
the further demise of US military power and the growth of military might in
China, Russia, Iran and North Korea, the US National Security Strategy
only required that ‘in wartime, the fully mobilised Joint Force will be
capable of: defeating aggression by a major power; deterring opportunistic
aggression elsewhere; and disrupting imminent terrorist and WMD
[weapons of mass destruction] threats’. Critics of this strategy call this a
‘one-and-a-half-war strategy’.
If, in the period from 1998 to today, the effective military capability of
the mighty US has decreased from being able to win two and a half wars
simultaneously, to being able, once fully mobilised, to win one major war
and maybe hold in another (half) war, then Australian intelligence
organisations had better start paying very close attention to the friend on
whom we depend so entirely for our security. Australia needs to wake up to
the fact that US military power is not infinite, and may become still weaker
through budgetary decisions, industrial limitations and waning intent.
Australia should study the US to the same extent as I hope we study our
potential adversaries. Not because the US is likely to become an adversary,
but because as a dependent nation we cannot afford to misread either US
intent (which is hard) or capability (which is relatively easy). At present it
seems we read neither in a formal way.
We should watch very closely the budget decisions that the US makes,
and carefully assess the implications for Australia. We should be very
sensitive to a US that is tired of war, that sees most of its allies free-riding
on the US, and to any ally that talks of smaller defence budgets. And we
need to know what a substantially smaller US defence budget would look
like, because for Australians, that is the trillion-dollar question. The US
decision about whether to support Taiwan will be based on a judgment as to
how successful it is likely to be at a reasonable cost.
The issue of the defence budget was recently taken up by US
Congressional Budget Office (CBO) analysts in a report that looked at how
a smaller defence budget could be realised with a cut of $1 trillion over the
10 years from 2021 to 2031. This could be achieved, the authors wrote, ‘by
phased budget cuts over the first five years . . . and growth with the rate of
inflation thereafter’, which could yield a 2031 defence budget ‘about 15
percent smaller than DoD’s [the Department of Defense’s] 2022 funding
request’.
Three very confusing options were proposed, none of which really
looked at whether the US could win or lose a war or wars against a peer
adversary. The first was called ‘Proportional Reductions’ and is essentially
‘deterrence by denial’, which involves ‘denying or reversing military gains
in regional conflicts’, and eventually reducing personnel numbers by 20%.
Option 2 is called ‘Coalition Defense’, described as ‘deterrence through
punishment’, which ‘would call for reductions in conventional forces, such
as brigade combat teams and fighter aircraft, and increases in long-range
strike capabilities, such as cruise missiles, anti-ship missiles, and airdefense missiles’. The third option is called ‘Command of the Commons’
and emphasises ‘freedom of navigation in sea, air, and space around the
world’ and ‘avoids the use of large ground forces to seize and hold territory
in regional conflicts in favour of engaging enemies at standoff ranges’.
‘Although substantial,’ the CBO report continues, ‘a reduction that
reached 15 percent by 2031 would be smaller than both the 1990s’ budget
reductions (a 30 percent decline in annual budgets between 1988 and 1997)
and the 18 percent decline in annual budgets between 2012 and 2015 that
followed enactment of the Budget Control Act of 2011’.4
It appears that this report has had little impact, but just the fact that it is
out there illustrates how reluctant the current US administration is to spend
big on defence. Since 9/11, the US has been continually at war in the
Middle East, with only limited support from its allies, and simultaneously
challenged by an empowered China, Russia, North Korea and Iran. To
increase spending, groups within Congress are trying to add US$24 billion
to a Pentagon budget proposal that far exceeds US spending at the peak of
the Korean and Vietnam wars, probably because the cost of both volunteer
personnel and state-of-the-art equipment today is many times more than it
was during the 1950s and 60s. The Congressional Budget Office analysis,
so say the opponents of large defence spending, ‘offers an opportunity to
step back and take a closer look at how much is actually necessary to
protect the US and its allies . . . At a time when the greatest risks to our
lives and livelihoods are not military in nature, saving a trillion dollars that
could be devoted to preventing pandemics, addressing climate change, or
reducing racial and economic injustice is no small matter.’5
This statement is realistic in relation to the many challenges facing the
US, but it is totally unrealistic to ignore the one big challenge to US power,
and that is an aggressive China. The ramifications for Australia if the US
fails to stabilise its own power in our region are immense.
In terms of potential adversaries, China, of course, is not the only focus.
The 2018 US National Defense Strategy, which is a subset of the US
National Security Strategy previously described, identified five major
military threats for the US: China, Russia, Iran, North Korea, and terrorist
groups. Since 2018, each of those threats has only grown more serious.
The much-publicised ‘Pivot to Asia’ during the Obama and Trump
administrations was a policy to recognise the importance of Asia and in
particular China, compared to Europe and the Middle East, for US
diplomacy and the US military. Despite much talk, the policy pivot
produced very little in terms of diplomacy and even less in terms of moving
military resources to the Pacific. At the same time, there are encouraging
signs that other parts of the US government are homing in on China in
particular. One area of national security that is, critically, focusing on China
is intelligence. In October 2021, the security website Defense One reported:
The CIA’s new China Mission Center sends a clear signal to the
intelligence community that it’s time to shift its focus to near-peer
competitors after 20 years of tracking terrorist threats, two analysts
said. The center is part of the administration’s broader effort to
pivot the national security community’s focus towards competition
with great powers, such as China, and away from the
counterterrorism operations that dominated the past two decades,
including the war in Afghanistan that ended this summer. CIA
Director William Burns announced on Thursday that the new China
Mission Center will bring together capabilities from around the
agency to better respond to the threat posed by Beijing.6
The next US National Security Strategy is due and preliminary documents
have been published. The full and final version was to have been out early
in 2022, but there is a history of this important document being outpaced
and delayed by the fast-moving real world and, in this case, probably the
situation with Russia, Ukraine and NATO has impacted on the policy
writers. As governments slowly recognise the importance of the threat from
China, despite not yet taking significant action to counter it, it is ironic that
four experts on the website War on the Rocks refer to the not yet completed
National Security Strategy and write:
The strategy should acknowledge that China has the world’s largest
standing army, navy, coast guard, maritime militia, and substrategic missile force. But China’s advantages are not only
quantitative.
As the likely home team in a confrontation with the United
States, the People’s Liberation Army has pursued capabilities
specifically designed to frustrate the US military’s ability to project
power into the Western Pacific. The Defense Department’s own
military power report on China judges that this strategy has placed
the Chinese military qualitatively ahead of the US military in landbased missiles and integrated air defenses, including new
technologies such as hypersonic and directed-energy weapons. In
other areas like AI [artificial intelligence] and energy storage,
Beijing is exploiting its policy of military–civil fusion in innovation
to gain an edge on US forces.
The US military is emerging from a decade of delayed
modernization and insufficient funding, whereas China grew its
defense spending by at least 8% a year for the last decade.7
As this article suggests, and to repeat a point made earlier, the US is
surfacing from decades of war in the Middle East with worn-out equipment,
understandably having allocated a lot of its funding to ‘today’s wars’ rather
than investing in the future. During the Iraq War, for instance, Secretary of
Defense Bob Gates wanted more drones to carry on the day-to-day fight in
Iraq and found himself in conflict with the US Air Force, which wanted to
continue building the fighters and bombers that it thought would be needed
in the future. Gates sacked the chief of the US Air Force and restricted the
production of aircraft such as the stealth F-22 fighter and the B-21 bomber,
in order to build the drones and other aircraft he needed. The result was that
only a limited number of the extraordinary F-22s were built and the B-21 is
still not in production. The impact of diverted spending and focus will be
felt for a long time to come. The likely war with China, if it is ever fought
by weapons of this type, is going to be fought by a very small number of
modern stealth fighters, but mainly by US fighters and bombers that are 20
to 30 years old.
The result of all this is that the US will not be able to marshal sufficient
military power to deter China in the Western Pacific, possibly for years. A
University of Sydney study warned in 2019 that America ‘no longer enjoys
military primacy’ over China and that US bases, air strips and ports in the
region ‘could be rendered useless by precision strikes in the opening hours
of a conflict’.8
Nor should the US expect much help from its current regional allies. It
doesn’t appear that nations in the Western Pacific or the Indo-Pacific have
the attitude towards mutual defence that was prevalent in devastated Europe
after 1945, when NATO was formed. Such a binding defence alliance is
unlikely in the Western Pacific. South Korea will always be too consumed
by the threat from North Korea’s massive land forces to undertake any
large-scale manoeuvres that take its forces too far away from its own
northern border.
Japan is changing in its attitude to the constitutional limitations imposed
on it by Article 9 of its post-war constitution, which prohibits anything but
military action in self-defence, and is now at least talking about pre-emptive
strikes on adversary bases that may threaten it. Japan has made the
significant declaration that an independent and free Taiwan is so important
for Japanese security that it is to be regarded as an area covered by Japan’s
self-defence. This is still a far cry from Japan deploying its very capable
military forces beyond its home islands to assist US forces in the seas
around Taiwan. And, of course, Russia also represents a threat to Japan,
which claims sovereignty over the Kuril Islands, occupied by Russia since
1945. This threat may, however, have decreased with the Ukraine War,
which has forced Russia to move significant military forces from its eastern
provinces to more central positions closer to Ukraine.
Vietnam is in conflict with China over a number of issues, particularly
its maritime borders, and has suffered frequent Chinese intrusions.
Vietnam’s military today is not the military that defeated France, the US
and China in the past, and as well as sharing a land border with China,
which is a continuing threat, Vietnam has a navy and an air force which are
small and totally outdated.
In our world, it is dangerous to be considered weak. If you are perceived
by an adversary as weak, or you are in fact weak, especially with regard to
conventional (non-nuclear) forces, and you are one of the nine countries in
the world with nuclear weapons, you might face a dilemma. Because your
conventional forces are weak, you may be tempted to consider using
nuclear weapons, not just as a deterrent but in desperation. President Obama
was considering declaring a ‘no first use’ nuclear policy for the US, and it
seemed that the Biden administration was also considering such a policy.
Many US allies have found the idea of a ‘no first use’ policy by the US
hard to accept. The US nuclear umbrella was something that emerged
during the Cold War and is still seen as a guarantee that the US, as a
nuclear-weapons state, will defend certain non-nuclear allied states, to
discourage them from developing their own nuclear weapons. It is
commonly said to apply to Australia, NATO, Japan and South Korea. Japan
has been threatened with nuclear attack by China if it becomes involved in
defending Taiwan, so for Japan this is a real issue.9
The US has historically maintained a policy of ambiguity about whether
it would carry out a first strike with nuclear weapons, but the new policy
that was being contemplated would expressly rule it out. According to
Senator Jim Risch of Idaho: ‘It gives more comfort to the enemy that they
can plan an attack and do whatever they want to and not worry about us
[carrying out] a first strike . . . Nobody wants to use a first strike, but there
are scenarios where you can imagine a first strike, and the best thing you
can do is keep [adversaries] off balance.’10 One of the very few upsides of
the appalling Ukraine War, apart from the sanctions and the increase in
readiness and armaments within NATO, is that President Biden has now
stated that, given the repeated threats of the use of nuclear, biological and
chemical weapons by Russia against Ukraine, the US will not adopt a ‘no
first use’ nuclear policy as part of the US Nuclear Posture Review.
Meanwhile, to overcome at least the perception of weakness, the US is
considering a very different policy to deter Chinese and Russian aggression.
The four experts from War on the Rocks quoted earlier advocate that the US
should establish a ‘strengthened forward defense posture’ in vital regions –
that is, locating forces permanently, especially in the Indo-Pacific, rather
than having to deploy them over very long distances in a period of tension
or war.
The idea of a ‘strengthened forward defense posture’ is a complicated
concept and is military-speak to explain that in the Pacific, Europe and
South Korea, it has been the normal state of affairs that not all the US forces
needed to fight a war are based close to where the war might be fought, in
the so-called theatre. Instead, most of the forces required are located in the
continental US, which is the source of manpower, equipment and training
facilities and where it is normally less expensive to maintain large forces.
Only a smaller number of troops are deployed permanently close to the
various likely war theatres. In previous theatres such as Europe, South
Korea and the Middle East, it was not uncommon to find that the forward
deployed troops numbered in the order of perhaps 30,000 to 50,000, a
relatively small number, and the troops that were deployed from their home
bases once tension increased or war broke out numbered as many as several
hundred thousand.
So, if tension increases or actual war breaks out, the very large forces
have to be deployed from the US to the theatre where the war is occurring.
It can be a very long and complicated process to move what might be
hundreds of thousands of troops, with their equipment, vehicles, ships and
aircraft, literally from one side of the world to the other. It might be slightly
easier to deploy combat aircraft or warships than it is to deploy large
numbers of soldiers, but there is still a delay and in the case of a surprise
attack, or a fait-accompli attack, as it is sometimes called, this delay can be
disastrous. And, of course, while those forces are being deployed in troop
ships or transport aircraft, they may be vulnerable to being attacked if the
enemy has the capability to do so.
The US used to practise the deployment of forces from the continental
US to either South Korea or Europe in massive exercises every few years,
so that it was confident it could readily carry out such a deployment and
potential enemies could also see that it was feasible. The balance between
the smaller permanently forward-deployed troops and the larger number of
those to be deployed into the theatre when necessary was based on risk
assessments conducted regularly.
US soldiers from Camp Humphreys in Pyeongtaek, South Korea, on
exercises in May 2022. Camp Humphreys is the largest overseas US
military base, with a population of more than 45,000. (Getty)
When the threat from Russia against Europe was very high or the threat
from North Korea was more than the South Korean army was capable of
handling, the majority of US troops required to win the war in Europe or the
Korean peninsula were actually stationed in the relevant theatre. As the
threat diminished, or the local forces were assessed as adequate, the bulk of
US troops were sent home and only a small force was kept ‘forwarddeployed’, as the military says.
One function of the permanently forward-deployed US forces was to act
as a tripwire. That is, if an enemy invaded the country or area concerned,
the forward-deployed troops would immediately be involved, which would
in turn guarantee that the US, with all its military might, would assist. A
second function of such forces, if the local forces were not strong enough,
was to gain time for the US to deploy more forces into the theatre, to bring
US strength up to the full number required for combat.
As explained earlier, the US maintains significant forward-deployed
forces in the Western Pacific, in Japan, South Korea and in the US territory
of Guam. Because the Western Pacific is mostly ocean, the US has more
navy and air forces in this theatre and fewer army forces (the US Marines in
Japan being part of the US Navy). A notable feature of Western Pacific
theatre deployments is that the US does not maintain any forces on the
island of Taiwan, even though this is usually seen as the most likely focus
of an attack by China. This is because the US has long believed that in a
conflict with China, the US could control the Taiwan Strait, between the
Chinese mainland and Taiwan, with navy and air forces alone. (These
forces were sometimes referred to as ‘blocking forces’ because they would
block a Chinese invasion fleet.) What changed this belief was the
development of Chinese military capabilities.
In its 2018 report, the bipartisan National Defense Strategy Commission
(a Congressional body that makes suggestions as to what might be in both
the National Security Strategy and its subordinate National Defense
Strategy) expressed concern that ‘China’s missile, air, surface, and undersea
capabilities’ would grow and potentially make it too costly for the US to
respond to Chinese military aggression in the Taiwan Strait. Since 2018, the
Chinese military has only improved the capacity and range of these
capabilities, especially within the Taiwan Strait.
If the US is not able to stop a Chinese attack on Taiwan via blocking
forces in the Taiwan Strait, then it has only a limited number of other
options. It would have to switch to relying on US and allied aircraft based
in South Korea, Japan and Guam; to utilising one or two nuclear
submarines that are almost permanently present in the vicinity of Taiwan,
and others that could be called upon quickly to attack an invading force;
and, possibly, increasing the number of US surface fleets that could
approach Taiwan from the second island chain. These measures, perhaps
backed by long-range bombers based in more distant places such as Diego
Garcia in the Indian Ocean, provide only limited confidence that China
could still be deterred.
The 2018 Report advocated steps:
to strengthen the reach, agility, and survivability of American and
partner blocking forces already forward positioned in and around
the first island chain. By raising the costs and increasing the
uncertainty of success for Beijing, a more robust posture would be
more likely to deter aggression. Ready and capable blocking forces
can also provide valuable time for surge forces to arrive from
outside the region.11
In summary, because of the strength of the People’s Liberation Army (PLA)
in the Taiwan region and its ability to dominate this region and interfere
with the deployment of US forces from outside the Western Pacific, the
Commission recommended in 2018 that US forces be permanently located
within the region so they don’t have to expose their vulnerabilities in
moving there during a period of tension or actual conflict. Such a move
would be a significant change in policy by the US and its allies and would
be seen by China as an aggressive move; indeed, China has said that
forward deploying US troops to Taiwan would be considered an act of war.
The commission went on to stress:
Defeating an attempted fait accompli attack (surprise attack) by
Beijing will not be easy. Once hostilities commence, US surge forces
trying to get to the region can reasonably expect to be inundated
with a range of attacks before they arrive and likely even before they
depart the United States. Airlift and air-refuelling challenges will
impair any effort to get assets to the region quickly. That puts a
premium on strengthening US and partners’ military capability prepositioned along the first island chain. The next National Defense
Strategy, therefore, should prioritise this effort and support projects
and activities such as those promoted by Congress’[s] Pacific
Deterrence Initiative.12
What the National Defense Strategy Commission is advocating as a feature
of the next US National Defense Strategy is a ‘strengthened forward
defence posture’, that is the permanent location of a much larger number of
US and allied troops, ships and aircraft permanently deployed either on
Taiwan or in the islands around Taiwan, referred to here as the first island
chain. This is intended to overcome the problems that would be faced if
China conducted a surprise attack and if China had the ability to reach out
and interfere with the deployment of US forces from more distant bases.
Such a recommendation is only valid as long as those forward-deployed
forces are not themselves subject to the surprise attack referred to. Many of
the US bases in the first island chain are now well within range of Chinese
missiles and rockets, and so are vulnerable.
The Pacific Deterrence Initiative mentioned in the above quote is not a
new idea, and is based on the 2014 European Deterrence Initiative,
stemming from the perceived weakness of the US military in Europe and its
inability to deter Russia’s annexation of Crimea in 2014. Members of
Congress specifically asked the Commander of the US Indo-Pacific
Command, Admiral Phil Davidson, to submit a costed list of capabilities
that would be necessary to deter Chinese military action in the region for
which he was responsible. What they received in early 2020 was a US$20
billion wish list for the next six years. It envisaged a $1.6 billion defensive
ring around Guam, millions for partner nations, and increased stockpiles of
long-range weapons. Davidson called his plan ‘Regain the Advantage’,
which indicates how the key US commander in Australia’s region sees his
position: he has lost the advantage.
Congressional commentary supporting Davidson’s plan reflected
‘bipartisan frustration with the Pentagon’, which, it was claimed, continued
to fall short of providing answers. One senator noted: ‘The National
Defense Strategy is more than just about how many planes, ships and tanks
we buy. It’s also about making sure our forces can be in the right place, at
the right time, with the right stuff.’13
The issue that is emerging in the Indo-Pacific is that the US is deficient
in both the weapons of war and the operational concepts of how to use
them. Australia’s major ally cannot afford to be deficient in both forces and
fighting concepts, but Australia can hardly criticise, because (as I shall
reveal in the next chapter) our deficiencies are even greater.
The question is: can any operational concept for the use of military
forces that will not come fully into being for another six years deter China,
especially if it is based on a bluff concerning its overall weak forces? And
even if the US does send more permanent troops to bases in Japan, South
Korea, Guam, Diego Garcia or even Australia, it must also ensure the bases
that host those forces are hardened and defended. This means that fuel
storage, maintenance facilities for ships and aircraft, and headquarters are
dispersed and duplicated in places where they can be attacked by rockets,
missiles or aircraft; that aircraft and ships are not lined up in the open,
inviting attack; and that all such bases are ringed by anti-aircraft and antimissile defences. Few of these bases are currently hardened or defended,
and to send more forces to vulnerable bases is no better than the US
deploying its battleships from US west-coast bases and lining them up for
the Japanese to attack at Pearl Harbor in 1941.
For military power to deter, it must be real. Before any serious
rethinking of Beijing’s current policies of repression at home and
aggressive competition abroad, China’s leaders would need to be deterred
by real evidence that the US is more resilient than they thought, and that the
US could win a military confrontation. The Chinese leadership must be
convinced, and they are smart enough to see through any bluff or bluster.
***
The most basic objective of the democratic nations of the world should
surely be to avoid war with China if at all possible. This objective could of
course be achieved by allowing China to incorporate Taiwan, even if it did
it by force: a modern form of the policy of appeasement. War would
certainly be avoided, but at the cost of the freedom of 23 million Taiwanese
and encouraging China to use intimidation to achieve other goals. Dealing
with the issue of an illegally aggressive China in our region could then be
put off, at least until China wanted to use force in some other illegal way.
Yet the rhetoric would suggest that the aim of the US and its allies is to
secure Taiwan against illegal coercion, aggression and invasion, as they did
in post–World War II Europe through the Cold War (1948 to 1991) and in
various Asian countries, such as Malaya, where they opposed a communist
takeover during the Malayan Emergency (late 1940s till the early 1960s);
South Korea, where they fought in the Korean War in the 1950s; and
Vietnam, where they unsuccessfully attempted to help South Vietnam fight
off North Vietnam in the 1960s and 70s.
If the aim is to secure Taiwan against a Chinese takeover, there may be
a number of ways of doing it. First, the Taiwanese should be assisted in
every way to defend their own nation, which has been happening to some
extent for decades now through Taiwan’s purchase of advanced US
weaponry, authorised under the US Taiwan Relations Act of 1979. In
relation to this, questions have been asked in the US Congress about
whether Taiwan is doing enough itself. These are legitimate questions if
there is an expectation that friends across the world will come to help the
Taiwanese in the event of an attack.
Taiwan’s ground forces need to reform. This started early in 2021 with
the imposition of a new army structure more suited to wartime, in which
operations are decentralised through regional commanders. A submarine
production facility was opened that will manufacture eight new dieselelectric attack submarines, with assistance from the US, and in September
2021 the Taiwanese launched a new warship with air-defence capabilities
and anti-ship missiles. Taiwan also plans to spend $1.4 billion on more
fighter jets (F-16s) and more air-defence missiles to counter China’s
fighters.
A column of Taiwanese armoured vehicles on the move during a
military exercise simulating a Chinese invasion of Taiwan, in November
2021. (Getty)
These are admirable improvements, but it may all be too little, too late.
Taiwan’s military budget for 2022–23 is US$16.9 billion, only a small
increase on this year’s US$16.2 billion, whereas China has said it will raise
its defence spending next year by 6.8%, still impressive but a little less than
its long-term average of 8% a year. Taiwan has stated that it will commit an
additional US$9 billion to its military over the next five years. But five
years is a long time. It is certainly on the edge of many predictions of a
likely invasion timeline, including mine, and longer even than the
predictions of Taiwan’s defence minister, who has an interest in portraying
the shortest threat period possible.
Due to the confidence that a US fleet in the Taiwan Strait would
physically block such an attack, for decades no one really worried about
Taiwan’s ability to defend itself from invasion. The Taiwanese became
complacent because the US Navy carried the burden. Nowadays Taiwan has
plans to buy long-range cruise missiles that could be used against an
invasion fleet while it is on the water or just after it lands, but the
improvement of China’s amphibious capability means that China could still
land despite these missiles, so the deficiencies in Taiwan’s army must be
further improved. Taiwan has a small fleet of US F-16 fighters, which are
too few to stop the Chinese air force’s jets dominating the island. As noted
above, it plans to buy more, but even if the number of fighters was doubled,
Taiwan could field only half the number with which China would likely
attack. Simply put, Taiwan is nowhere near where it needs to be.
Of particular interest, and reflecting the sympathy that Taiwan has
within the US Congress, Senator Josh Hawley, a member of the US Senate
Armed Services Committee, recently introduced an Arm Taiwan Act which,
if passed, would ‘ensure Taiwan has the . . . defenses it needs to deter a
Chinese invasion’.14
The big issue is that Taiwan’s defence spending is only 2.1% of its GDP,
an amount far too small to counter China’s military and not proportional for
a very prosperous nation. The proposed Arm Taiwan Act requires that
Taiwan spend at least 3% of its GDP on defence, ensuring that it plays its
part if its allies are prepared to play their part. The Act quite logically
focuses on Taiwan’s reserve personnel. Taiwan has a population of 23
million and a claimed reserve of nearly 2.2 million soldiers, but US
assessment of the military reserve force is that it is both declining and
woefully underprepared. To ensure that Taiwan’s reservists are effective and
ready, the bill demands that its equipment be modernised and a review of
how this reserve force will fight if an invasion occurs, so the reservists can
better operate with active forces and with allies.
The Arm Taiwan Act favours weapons such as land-based anti-ship
missiles (referred to as asymmetric defences because supposedly for a
smaller outlay they give a greater return) over expensive submarines that
would take years to develop, or fighter jets that would be inferior to
China’s. It proposes that the US match Taiwan’s improvement in its military
resolve with what is called the Taiwan Security Assistance Initiative. This
would provide US$3 billion from US coffers each year from 2023 to 2027
to the US Department of Defense for equipment, training and other support
to build US capability to assist Taiwan.
In a stinging but constructive opinion piece in Newsweek, defence
expert Sam Abodo concluded: ‘From reserve numbers to fighter jets,
Taiwan is playing catch-up in a military build-up race it will never win . . .
Senator Hawley gets Taiwan right: The Arm Taiwan Act solidifies the
United States’ commitment to its democratic partner, reorients its defense to
more reasonable priorities and conditions assistance on matching
investments. For these reasons, Congress must waste no time passing it.’15
The bill was introduced in January 2022, referred to the Committee on
Foreign Affairs and, at the time of writing, has not been passed.
A second way to secure Taiwan and deter China from military action
would be, as discussed previously, for a relatively small number of foreign
troops to be deployed to Taiwan to act as a tripwire force, which would, in
certain circumstances, guarantee the subsequent arrival of US
reinforcements. The current US policy of having no troops at the likely
point of action on Taiwan is one of ambiguity and does not compel the US
to become militarily engaged should Taiwan be attacked.
A third way of guaranteeing Taiwan’s security – again assuming there is
a change to the US policy of ambiguity – would be to locate a significant
number of foreign forces, doubtless mainly American, on the ‘front line’
beside Taiwanese forces, to form an integrated part of the defence of
Taiwan. As previously discussed, this was the strategy that guaranteed
European security for decades against the Warsaw Pact. Even with US and
other troops permanently located on the inner German border, the US
regularly practised the movement of even more troops from the US to
Germany as reinforcement. Even Russia did not have the ability to prevent
such massive sea and air movement at the time.
A fourth method of deterring Chinese aggression, according to thinking
at the time when battles in the Taiwan Strait were envisaged, would be to
use a small number of the most highly capable weapons systems, such as
nuclear-powered attack submarines lurking somewhere close to Taiwan,
which if the Chinese mounted a D-Day-style invasion, could interfere with
or even deter the maritime component of a Chinese invasion. China of
course would know that the submarines were in the region because the US
admits to them being there and so would have to take them into account in
their battle plans. In addition, US air power mounted from regional bases in
South Korea, Japan and Guam, and more distant bases such as Diego
Garcia or Australia, or from US Navy carrier battle groups at sea, could be
used to support Taiwan without being located on Taiwan. China’s reaction
to these strategies, and how it might negate them, will be discussed later.
All of these strategies are relevant to the Taiwan scenario only if that
scenario involves a Chinese fleet being assembled in mainland seaports and
airports then moving across the Taiwan Strait, protected by China’s
warships and aircraft, and carrying out a sea and air landing. This would
then be opposed by Taiwanese forces and by US air and sea power based
out of regional bases and aircraft carriers that were not themselves subject
to attack. As Chinese military capability increases, however, as we have
seen, other options are becoming available to China, most notably removing
the US from the Western Pacific and then ‘reincorporating’ Taiwan into the
PRC. So, unless the US and its allies begin to think much more broadly and
start looking at other options available to them, we might all be preparing
for the wrong war.
All of this complicates the Taiwan issue from Australia’s perspective,
and it becomes legitimate for Australians to ask: if the US is ambiguous as
to whether it will assist Taiwan and its 23 million citizens in the event of a
Chinese invasion, might it feel the same way about 25 million Australians?
CHAPTER 7
THE RESPONSE: AUSTRALIA
There is change in the wind for Australians. After 75 years of one of the
most secure existences in human history, most Australians have either
become deeply complacent about issues of national security or have never
even considered them. Those attitudes are changing, along with government
policy, but not fast enough and not comprehensively enough.
In the recent past, there seemed to be a high degree of confidence that
national security was something the government could manage. The
government would tell the people when they should begin to worry. But in
the face of a highly aggressive China, and with the real possibility of war
now looming, many Australians have already begun to worry.
Australia is a prosperous nation, yet a deeply insecure one. This is a
situation that has developed over many decades, and at the hands of many
governments. Over the entire post–World War II period, Australia has been
mainly a strategy-taker, not a strategy-maker. We have had the luxury of
this contrariness because we have been allied with the US,1 whose world
role over those 75 years has been to provide security and prosperity to many
countries, not just Australia.
As a result, we have prioritised prosperity over security. We have been
remarkably successful as regards everything except our national security. It
has been an easy ride along with the US and we have produced an
extraordinary nation, if a complacent one. We have been very happy to live
with whatever the US strategy of the day was, while having very little input,
if any. As a substitute for strategy, we have expressed ourselves as having
interests and values and, in some ways, tried to act accordingly.
Our dues to the US for 75 years of security have been paid by military
involvement in many conflicts at America’s side, leading to our
identification as a close US ally. But all of those conflicts have been distant
from our shores and have not involved critical, existential issues for
Australia. We who fought in them classified them as ‘wars of choice’
because Australia could choose whether to participate, choose what forces
we sent, choose when we sent those forces, choose what they did when they
arrived in the operational area, and choose when we brought those forces
home. At no stage did Australia have to commit to actually win. At no stage
did we have to seriously pay heed to what the enemy would do, because we
had so much choice, and when things were not going well, or when the US
had decided it had had enough, we brought our forces home. Because we
had these options, Australia did not need to have a military with high levels
of readiness to go to war (which is expensive) nor did we need the most upto-date equipment or even a wide range of forces, because what we did not
have, the US had and so could support us logistically or on the battlefield.
Compared with the strategic environment that Australia faces now,
where a major power could dominate the region at very short notice and
where our major ally may not be able to provide the assistance we always
took for granted, past wars of choice were easy. In those wars, failure was
impossible above the tactical level because we did not decide which
operations were to be conducted or how success was to be defined. In some
of those wars, the size of our military contingent or how ready they were to
fight did not even matter because the presence of the Australian flag beside
the US and other countries’ flags gave legitimacy to US claims of wide
support. What was of importance to Australia was to be regarded by the US
as an important ally, so that if in the future Australia needed US help in a
more serious war, there was a greater chance of Australia receiving it.
Preparation and participation in a comfortable, distant and cheap war of
choice is far easier than one in which Australian forces must commit to
being prepared to win. Because of the importance of commitment to
winning possible wars, and not just participating, those of us in the military
who were keenly aware of the deficiencies in our defence forces, called
these new forms of warfare ‘wars of commitment’. Behind this view, held
widely in the thinking parts of the ADF (Australian Defence Force), lay a
deep knowledge of both the strengths and the weaknesses of Australia’s
defence force, especially if it had to be used without as much US support as
we were used to, and against an adversary more demanding than the East
Timor militia, the insurgents in Iraq or the Taliban in Afghanistan. Having
fought those wars, we could see the problems Australia could have if we
had to actually win.
Australia has performed very well at the lowest tactical level in the wars
in which we have fought because our soldiers are well trained and generally
well equipped for tactical warfighting, and we were never rushed in
deploying forces because of the presence of the US or other allies, already
in place. In all our wars and operational deployments except East Timor –
and East Timor was a UN-endorsed stabilisation operation in which there
was very little combat – we have fought within a warfighting and logistics
system provided by allies, mainly the US. In wars of choice, Australia has
not needed to supply the more difficult parts of full combat support, such as
heavy artillery, helicopters and attack aircraft, or a complete logistical
system with mass sea, air or protected road transport, large-scale storage
facilities or even the full range of ammunition. We merely plugged into an
ally’s system.
Winning a war consists of achieving the war aims. Each war is different,
but what is consistent in wars of commitment is that the enemy has a far
greater say than in our wars of choice. Because of that, Australia must be
able to design and conduct military operations to achieve a strategy by
ourselves, even if it is within a coalition of some kind. This is far more
difficult than it sounds, unless there is experience of conducting military
campaigns and aligning them with political needs. And it is my contention
that Australia is incapable of conducting and supporting such a coherent
regional campaign of its own – a war of commitment – or, at present,
fighting at all except as a small part of a strong US force. This reflects
appallingly on our sovereignty as a nation and our view of national security.
In the early 2000s, Professor Paul Dibb, director of the Strategic and
Defence Studies Centre, made a damning forecast about the nature of the
defence force Australia was in the process of creating: ‘If we are not careful
[we] will produce a one-shot ADF with nothing left over after we have
protected such a small and vulnerable force.’2
The results are clear to see today. The ADF, a force I grew up in and
that I love and support, lacks lethality (its weapons are not powerful
enough), sustainability (it cannot fight for long enough) and mass (it is not
big enough). As the professor predicted, it is a ‘one-shot’ defence force,
appropriate for the wars of the last 75 years but almost impotent for the
coming decade.
Most Australians think that Australia’s experience of war over the last
100-plus years is epitomised by the ANZAC spirit, and the view that
despite failures we always win in the end. I don’t agree. I have long
believed that our experience of war as a nation has been one of
unpreparedness for each and every war we have participated in, and that
still holds today. It was massive failure after massive failure that created the
need for the ANZACs to suffer, and it is to their everlasting credit that they
overcame these disadvantages.
The greatest failure of any government is to not make reasonable
preparations for the defence of the nation, and Australia has had a series of
governments of all political persuasions who have failed in this respect. It is
a moral failure of the highest order to expect that the spirit and blood of the
nation will act as a substitute for proper preparations to face evil in this
world.
In industrial-age warfare, such as World Wars I and II, it was possible to
be unprepared and, with a bit of luck, to survive. In the 1920s and 30s, it
was possible for Australia to have a minuscule regular army backed up by a
larger militia, a small navy as an adjunct to the Royal Navy, and a
‘weekend’ air force. In those days a nation could mobilise itself relatively
quickly and could lose battle after battle but still put the nation on a war
footing and prepare and equip armies. Australia had created sizeable war
industries and developed the fourth-largest air force in the world by the end
of World War II, and an army so big that at a certain stage of the war parts
had to be disbanded to make workers available for growing food.
Australians could fight and fail, but learn and win.
The wars we face today are overshadowed by nuclear weapons, and
although conventional weapons are still predominantly industrial-age, using
chemical energy to create kinetic impact or blast on soft human bodies, they
are enhanced by information-age technology. Today computer digits can
propel bullets and warheads accurately towards targets. The next
information-age war will probably occur with unprecedented violence and
speed, and it is unlikely, if not impossible, that Australia will be able to
conduct even modern mobilisation as it was able to do the last time the
existence of this nation was at risk, in World War II.
Luck is important in all wars. In World War II it was luck and bravery –
as well as preparation by the US – that allowed us to survive in the naval
battles of the Bismarck Sea, the Coral Sea and Midway, and the land, sea
and air battles of Milne Bay, Guadalcanal and Papua. We all need luck, but
when we over-rely on luck or on hope, we go a long way towards
demanding that a new generation of ANZACs suffer. The more we prepare,
the less luck we need to have.
Australian governments were told time and time again during the 1920s
and 30s that the next war would be against Japan. Imperial war plans
envisaged the deployment of significant elements of the Royal Navy to the
Far East to defend its Singapore base, which was to be strengthened and
garrisoned as the pivot of the defence of Australia, New Zealand and
Britain’s Far East colonies. However, it was always clear that if the German
navy was active in the North Sea, the Royal Navy might be unable to send a
naval force to Singapore. Yet with full knowledge of this, Australia was
willing to send its under-equipped and under-trained army to North Africa,
its navy to the Mediterranean, and its young men to North America to be
trained as air crew then sent to Europe or the Middle East, even though the
German navy was indeed active in the North Sea.
What caused the dislocation in our strategic thinking that led us to
ignore information that was provided to us time and again over two
decades? Australian decision-makers knew that Japan could attack at any
time, and that we could be on our own. Why was it that the Australian
government of the day, still bearing the scars of World War I from just 20
years earlier, was willing to dispatch the only forces we had to the other
side of the world? Then, having done this, based on a strategy of hope that
the Japanese would not attack our region – and if they did the Royal Navy
would save us – we put ourselves to the task of creating a new generation of
ANZACs to fight against an enemy we considered, in our bigoted
ignorance, to be an inferior race.
For the last 75 years, Australian governments have been able to
maintain defence spending at very low levels, affecting the purchase of new
equipment and enabling only minimal training. My father, who was a
soldier in World War II, would probably have recognised the structure and
equipment of the infantry battalion that I commanded 40 years later.
It was embarrassing to be part of a nation that built warships after the
Russians invaded Afghanistan in 1980 that we proudly admitted were fitted
for, but didn’t have, weapons! The intention was that the weapons could be
added at some future time, thereby saving money in the short term. It was
embarrassing to be in the military of a nation that maintained our land force
at such low numbers of personnel and levels of equipment-readiness that
critical tactical commanders could not be trained except at the most basic
tactical level because their instruction was always theoretical and never
practical, and where units and equipment had to be rebuilt to be deployed
even to minor conflicts post-Vietnam, as we had to do with armoured
vehicles over several months before our troops could be deployed to
Muthanna Province in Iraq in 2005.
Our air force in the 1960s and 70s had no balance between combat
aircraft that did the fighting and critical support aircraft, such as refuelling,
command and control, intelligence collecting, and transport aircraft. It held
almost no stocks of ammunition or missiles, due to our inability to produce
anything much above small-arms ammunition. Our ability to deploy air
force fighting units outside their peacetime airfield was severely limited.
We ignored (and continue to ignore) our lack of production and storage of
liquid fuel for civilian as well as military needs, relying mainly on overseas
production of crude oil or refined product (aviation fuel, petrol and diesel)
and pretended that our lines of communication by sea and air, by which
liquid fuel was brought to Australia, would always be open – the classic
example of a strategy of hope. We assumed that anything we had been too
blinkered to provide for ourselves could be purchased and transported to
Australia across hostile oceans in times of tension or war, in ships that we
did not own.
Despite some improvements, those strategic chickens have now come
home to roost.
***
Although we have been at war many times since 1945, Australian society
has hardly been affected. As I have said, this has led to the highest
imaginable degree of complacency within the general population on the
issue of national security. Although some views have changed in recent
times, Australia lacks both resilience and self-reliance, where resilience
means the ability to take a hit, such as the reduction or cessation of shipping
due to international tensions or war, and self-reliance refers to the ability
that this country has to domestically manufacture the goods and services
that we need – not what we ‘want’ but what we need.
Australians have long been consoled by the fact that we are far from
regional trouble spots. If war ever occurs, this theory goes, it will be fought
by the ADF in some distant theatre and, just like before, will not affect the
everyday lives of Australians.
Most Australians who bother to think about the possibility of a regional
war with China probably believe that the ADF can defend us. After all,
governments tend to talk up the ADF at every opportunity and seem to
spend a lot of money on it. (This, in my observation, is also a view held by
most federal politicians.) And, returning to the point I have made
repeatedly, most Australians also think that US military power is infinite
and that, regardless of what we do, our great friend will always come to our
rescue. Yet, as I pointed out in the previous chapter, since the Cold War
ended in 1991 this has not been a true depiction of US military power.
Comparisons with US success in facing down Russia no longer apply.
There is still an element of truth in our belief in a very powerful US, but it
has now become a very dangerous view, encouraging complacency.
It is unreasonable to expect that today’s Australians have the same
attitude to and knowledge of war as the immediate post–World War II
generation, because most of us have no relevant lived experience. Since
1945, the essence of what has passed for grand strategy in Australia has
been to deter conflict by being allied to the US. This has made us one of the
richest countries in the world, because we did not have to exorbitantly fund
our own defence, nor did we have to secure the sea lanes along which our
exports were carried. We have one of the highest standards of living in the
world, we have the world’s 13th largest GDP,3 we are a liberal democracy,
we are a member of two of the most influential economic and security
world groups, the G20 (Group of Twenty) and the g7+, and we are in a
number of treaty alliances with the US and other countries. With a
population of 25.8 million, Australia occupies a continent rich in exportable
natural resources at the strategically important intersection of the Indian and
Pacific Oceans.4 Yet as a country with a strong economy, good diplomacy
and leadership (all part of national power), we are mind-numbingly weak in
terms of hard (military) power.
Under the reassuring umbrella of US power, initially we traded over
secure sea lines of communication, then in the information age we
embraced globalisation, assuming we could purchase products at cheap
prices from overseas industries forever, and so we exported much of our
manufacturing base to countries that produced the items more cheaply. But,
as we are now discovering, having passed through the COVID pandemic,
our desire to buy everything cheaper from overseas and so to ‘export’
critical industries and even our shipping, has frightening implications for
our national security. We have become over-reliant on one trading partner –
China – and it is now a potential adversary that is using this dependence
against us.
China’s attitude towards Australia has deteriorated markedly since
about 2016, but particularly after we had the gall to suggest in 2020 that an
inquiry into the source of COVID should be conducted. Much of this
hostility was expressed through CCP mouthpieces in the media, such as the
Global Times, but the rhetoric indicates an attitude. Australia was certainly
targeted by China in this manner earlier and more than most other countries,
but our strong reaction to this bullying, and the backing we received from
many other countries, has been an example to all. It is likely that our close
relationship with the US, a relationship that has grown stronger as a result
of China’s bullying, was one of the reasons that China targeted Australia.
But it has not worked as China intended and it did not flow on to the US
through Australia. China has not brought Australia to our knees. But
China’s actions and our embrace of globalisation do have long-term
implications for Australia in that our prosperity has been achieved at the
expense of our security, making Australia vulnerable to the kind of coercion
and bullying that is now a hallmark of the CCP under President Xi Jinping.
And it makes us very weak if China’s actions move into conflict.
We have long believed in the myth that everyone loves Australia, but
many Australians’ views of international relations are shaped by a very
limited experience of the world, extending no further north than the holiday
island of Bali. China’s actions have come as a shock because we have not
been the target of direct coercion by a large regional power since World
War II. Polls indicate that Australians are adjusting to the CCP view that
Australia represents everything China is not, and that therefore Australia
should not seek to prosper from our relationship, or even have a close
relationship with China.
From my own informal polls of friends and political colleagues, I have
learned that most now share an awareness of the threat from China, and
most know that, in the extreme, it may manifest itself as war resulting from
Chinese moves against Taiwan. According to successive Lowy Institute
polls, since 2017 Australians’ trust and confidence in China and its leaders
have noticeably declined. In 2021, 63% of Australians saw China as ‘more
of a security threat to Australia’, a 22-point increase on 2020. In the same
poll, 93% of Australians saw China’s military activities in our region as
having a negative influence on their views of China. On 4 October 2021,
Newspoll claimed that 75% of those polled believed that China posed a
significant threat to Australia’s national security. Over the period of 23–27
February 2022, at the start of the Russian invasion of Ukraine, Newspoll
ran a poll on both Russia and China, finding that 74% of Australians
believed that China posed a threat and 64% of Australians believed that
Russia posed a threat to Australians.
If Australia is to take that threat seriously and acknowledge the
possibility of a war with China, it is essential for us to understand the
probable nature of that war. The consequences of preparing for no war, or
the wrong war, or even of overpreparing for an unlikely war, are either the
disaster of losing when an adversary executes its war, or a massive waste of
scarce resources. History is replete with examples of preparing for the
wrong war or the wrong battle: Pearl Harbor and other allied bases in the
Pacific in 1941, the German attack on France through the Ardennes and the
Maginot line in 1940, the Chinese attack across the Yalu River in 1950, the
German attack on Russia in 1941, the Iraqi attack on Kuwait in 1990, and
the US invasion of Iraq in 2003 are just a few.
At present, the war Australia seems to be preparing for is one where
China instigates a major amphibious and airborne attack across the Taiwan
Strait while the attacking force is subject to US air and navy counterattacks.
We seem to believe that there will be time for Australia to prepare a force of
ships and aircraft capable of fighting a daunting adversary and dispatch
them from Australia to assist a US force somewhere in the region, then
support the US in whatever it decides to do. There is also an expectation
that like all the wars that Australia has been involved in over the last 75
years, casualties in personnel and equipment will be (comparatively) light.
Given the lack of action by governments over many years, we
erroneously believe that in such a war Australian society is unlikely to be
impacted any more than it was by our previous post–World War II conflicts.
There seems to be an implicit belief, too, that the US coalition of forces will
win, and our forces deployed to that war will return victorious to Australia.
But, as we have seen, a D-Day-style attack on Taiwan is only one option for
China, and it could be disastrous if Australia is not prepared for other
scenarios too.
It is essential for us to make a national commitment to winning and
have significant forces that are highly ready at all times – solid forces that
have the lethality modern war demands. These forces may need to fight
over a long period of time and sustain significant losses – because war often
involves learning through failure, with costs in lives and materiel – and will
need to be substantial enough not just to fight one battle in one place, but to
take part in campaigns consisting of a series of battles within an overall
military strategy.
Even more importantly, the nation that backs up its defence force must
be resilient and self-reliant. In the Westminster system this makes the
policy, direction and coordination of national security the responsibility of
the Prime Minister through Cabinet. Australians will need to feed and
provide for themselves when cut off from outside support, and to establish
an industrial base that can build and maintain military equipment,
innovating to a world standard to support the military materiel that we have
to purchase from overseas.
It is my judgment – military and otherwise – that Australia cannot do
this at present and will need considerable time to develop the capability to
do so. Unless we can dramatically change our ways, the question we will
need to address as a nation is whether we can achieve much at all in the
three-to-five-year predicted timeframe for a regional war. At present we
cannot deter either a regional war or a collateral attack on Australia as part
of that war, and we can do nothing to minimise the impact of such a
collateral attack. This is a terrifying national weakness.
Of course, diplomacy is our first line of defence and our diplomats work
very hard at this. We were in probably the most reliable alliance in the
world, ANZUS – the Australia, New Zealand, United States alliance, which
was activated for the first time on the occasion of 9/11 by Prime Minister
John Howard and then renewed in a different form in 2021 by Prime
Minister Scott Morrison in the form of AUKUS, the agreement between
Australia, the United Kingdom and the United States. But we do need to
consider the fact that being in such a strong alliance with the US may have
made us a target for bullying in the past from China and may in the future
make us a target for attack if a war occurs.
National security requires a national approach, and, as I say often, it
takes a nation to defend a nation. This in turn requires government to lead
the way in every aspect of national life. If we cannot solve every national
security deficiency at once, we need to decide on our priorities. But how
can we ever decide on priorities unless a comprehensive scan of all likely
threats, and what Australia needs for the necessary responses, is carried
out?
US President Joe Biden, flanked by Australian Prime Minister Scott
Morrison and UK Prime Minister Boris Johnson, announces the
formation of the new AUKUS trilateral security initiative, in
September 2021. (Alamy)
There is no indication that this has occurred, even at the highest level of
classification. If it had, the results would be observable at lower,
unclassified levels. And if in fact there is a comprehensive national security
strategy, what is the benefit of keeping its existence secret? The only
alternative explanation is that there is no strategy, and this is reinforced by
the patently inadequate state of national security in Australia.
A comprehensive process of national security that improves our
defences and resilience is surely worth advertising. If we have serious
national and military vulnerabilities, as I maintain, and if many in the
government and the population are nourishing the belief that we have
somehow reached national security perfection when we have not, then
something is deeply wrong and should not be hidden.
Certainly, the last Coalition government began to display a healthy
concern about our strategic environment, and many Australians are showing
a change of attitude towards China. Yet a better knowledge of what a
modern regional war might be like would give both government and people
a much more realistic view of Australia’s national security strengths or
weaknesses.
In my parliamentary committee duties I have often asked defence
officials what war they are preparing for, and I have never received an
answer. Most retreat behind a ‘classified’ barrier. Either they do not know
or are unprepared to commit themselves. Yet, by comparison, when I asked
the Japanese military attaché the same question in a defence committee
hearing, his answer was straightforward: war against China and Russia.
As I have discussed, war is more likely than most leaders admit,
because of the movement in the Western Pacific of sizeable military forces
in relative proximity to each other on a regular basis in disputed areas, as
well as the often professed readiness of China to use force.5 If war occurs
suddenly, or if there is some unpredictable incident or even an accidental
clash that does not get out of hand and is contained, Australia may or may
not get involved immediately if the clash involves other nations’ forces.
Given the small size of Australia’s military, its current low state of
readiness and our distance from the likely combat areas, Australia may take
longer to get to the fight than the time it takes for the fight (or at least the
first stages of a war) to be resolved, especially if China is successful.
If we are somehow involved, we go to war as a highly vulnerable
nation. Australians in general, but officials and leaders in particular, have
little or no conception of how such a war might be conducted, or how to
prepare for it. Those with a military background have little direct
experience, because we have not seen a strategic situation like this for
decades.
I would like to see the views expounded in this book being refined by
the knowledge and opinions of others through constructive debate. But that
debate is not occurring. Talking about something is the first step towards
action, so let’s at least start talking.
CHAPTER 8
CONCEPTS OF WAR
Modern military theory links a nation’s overall strategy, the operational
concepts required to achieve that strategy, and the tactics that make the
operational concepts possible, all of which must be aligned to achieve
success in war. This is a well-known idea, but very hard to do. A great
aphorism, which is usually attributed to the famous Chinese general,
military strategist and philosopher Sun Tzu, from around the fifth century
BC, says: ‘Strategy without tactics is the slowest route to victory. Tactics
without strategy is noise before defeat.’1 What this means is that it is more
important in war to get your overall strategy right first and then over time
you can develop the tactics to be victorious. If you try to do it the other way
around, you can be as good as it is possible to be in low-level tactical
fighting, and all you are doing is making noise before you are defeated.
For Sun Tzu, strategy sat directly above tactics, but as war became more
complex, the need for an operational concept to link tactics and strategy
evolved, particularly after World War II. For Australia in the twenty-first
century, facing a strategic environment dominated by the threat from China,
we must get all three right before the war begins, because modern war
might provide very little time to fail, learn and improve. It might be all over
too fast.
Strategy is about ends, ways and means in a very general sense. For a
military force, it is critically important to know, to the maximum extent
possible, how an operation or a campaign or a war is going to be conducted,
and how, where and who the force will fight. Predicting this is a large part
of the ‘art’ in The Art of War. It is about intuition, where intuition is about
making judgments from a base of deep experience of a subject. And, finally,
it is about the willingness to take risk, but calculated risk based on facts and
judgment.
A successful strategy must not stifle initiative, but by defining
boundaries and objectives you can open up initiative at every government
level and focus a nation’s efforts. Leaders who claim that strategy stifles
initiative reveal their lack of experience with good strategy, and they are
perhaps displaying a need to control everything centrally and personally,
which very quickly becomes impossible. There are severe limits to a
centralised approach, and I believe we are seeing those limits in Australia at
present, where we are addressing elements of national security one by one –
modern manufacturing, storage of liquid fuel, nuclear submarines,
sovereign missile production, cyber warfare – without an overall strategy
that indicates priorities, timing and where and when risks can be taken.
The need for strategy might be an esoteric point for some people, and
those who advocate for strategy might be accused of mere process and
pedantry, but strategy is just as important now as it was 2500 years ago.
However, just having a national security strategy, as the US does, will not
guarantee success. There is no point in developing a national strategy that
the nation or the military cannot achieve at an operational or a tactical level.
The strategy must be as appropriate as possible, it must apply not just to the
defence force but across the nation, it must influence the actions of every
government minister and every department without stifling them, and it
must be executed in a way that holds us all accountable. Do we have
enough industry with which to produce weapons and with which to
innovate? Can we feed our people? Can we keep a secret? Can we provide
the one thing that is still necessary for industry, agriculture and warfighting
– liquid fuel? Can we mobilise the nation in a modern sense, to move the
nation from a peacetime focus to one of war in a reasonable period? Are we
resolved as a society to prosecute a war?
If you start with a good national security strategy – what you want to
achieve overall across the nation and the ways and means of doing it – you
can derive subordinate strategies (or what the strategy purist might prefer to
call ‘plans’) for each area of government (generally in Australia’s case
corresponding to government ministries), and then you have started down
the road to success. Even the process of deriving the strategy will almost
immediately pay dividends. In the case of the military, as an example, a
national defence strategy should be defined from the national security
strategy, which is then refined into one or more operational concepts stating
how a nation will fight, and then solutions like the number and type of
weapons needed will be much more obvious and defendable, and so
procurement priorities can be set.
A serious attempt must be made by government to get the strategy right
to begin with, and hopefully a good strategy lasts, but if it needs to be
changed, it must be changed. However, the operational concepts and the
tactics employed must be infinitely variable in modern, fast-moving and
decisive conflicts, and in the period leading up to a war as the situation
changes.
Sadly, though, it is rare that a nation’s strategy, operations and tactics
are aligned at the start of a war. Normally they align the hard way, through
the impetus of defeat, as we have seen in our long world wars, where there
were opportunities to try, fail and improve. But in modern warfare, in the
era of attacks in space and cyberspace and with missiles and rockets that are
hyper-accurate over vast distances, defeat may be so devastating and quick
in the first instance that, unlike in past wars, learning and recovery may not
be possible. This is the risk that Australia and its allies are taking at this
very moment.
A nation that does not first set out to get its strategy right, while putting
significant energy into totally implausible tactics regarding how the nation
will react and how the military will fight, risks defeat. Success will not
come to a nation that cannot sustain its warfighting capability in the ways
required to achieve the strategy it adopts. If it does not have the tactical
equipment and the training for its military, if it lacks resolve as a nation or it
cannot feed its people or make and maintain the tools of war, it is not in the
game.
Since 1976, Australia has produced defence and foreign affairs white
papers every few years to summarise our security situation. Labor prime
minister Julia Gillard bears the distinction of having introduced Australia’s
first national security strategy, just months before Labor’s electoral defeat
in 2013, but it was too little too late, was not even debated by parliament
and never had the chance to become institutionalised. Apart from that, we
have never produced an overall national security strategy from which all
other subordinate national strategies might be derived. We have never really
needed to, because our great and powerful ally the US has always been
there.
It is my opinion that the strategy stated or implied in every defence
white paper since 1976 – a time when, as a serving soldier, I was a very
interested customer of defence policy – has never been achievable by the
ADF. That, I think, is a damning accusation.
Because we do not have a comprehensive national security or defence
strategy, it is very hard to test, say, any proposed equipment purchases
against an operational concept, much less something as vague as the
resilience or resolve of the nation.
For years we pretended to have a ‘Defence of Australia’ strategy, which
was designed in 1987 by the Hawke Labor government to justify cutting
defence expenditure. The intelligence community and bureaucracy fell into
line. Every experienced person exposed to the Defence of Australia strategy
knew that it could only be achieved by the small, impotent military we were
permitted to have if our enemy was as weak as it was in the exercises we
conducted.
Our navy and air force just ignored the strategic guidance over this
period and kept doing what they could do with the funds they had. But the
army, sadly, not only listened to it but, like good armies everywhere, started
to believe it, with the consequences that it reduced it from a good
conventional army that had fought well in several vastly different wars
(Korea, Malaya and Vietnam) to what was essentially a police field or
paramilitary force. We then equipped the army with weapons and vehicles
totally unsuited to modern war. Our strategy and our tactics were seriously
misaligned, and a generation of military leaders and military nous was lost,
until we were forced by East Timor, Iraq and Afghanistan to think, equip
and train for a more modern form of warfare. It was only recently, under the
last Coalition government, that we started to bring the three services into
the twenty-first century, but at its current pace it will take a very long time,
time we may not have.
Average but narrow policy, with no real-world outcomes or
accountability, has put Australia at great risk. The endpoint of national
security policy, despite what many bureaucrats believe, should never be just
policy itself, but should be a secure and strong sovereign nation. And
Australia is weak. For so many years now, the leaders of our nation, and the
bureaucrats in and out of uniform, have often been more interested in
producing paper policy than examining whether the nation’s diplomats or
soldiers could actually achieve the policy in question.
The danger of this disconnect between policy and reality is now being
seen, because an existential threat to this country has appeared in the form
of an aggressive China. But still Australian ‘strategists’, in and out of
government, revel in clever words spoken and written in their alternative
reality and attend to the national security of this nation with glacial
slowness, as though today’s China did not exist and there were no possible
threats for the next 10 or 20 years.
If there had been a culture of comprehensive strategy-making within
Australian government over the last few decades, via some kind of national
security mechanism or council, I wonder whether someone might have
spoken out about the caution that needs to be exercised if the market is left
to decide what industries we give up on in Australia and send overseas? But
those concerns were never raised and now, when we need industries and
shipping and other strategically important national capabilities, we are left
in the invidious position of trying to recreate them. We are reaping the
consequences of decades of decision-making failure.
Much of the commentary at the moment in Australia is totally captive to
an out-of-date and dangerous view that no more money needs to be spent on
defence and, instead, Australia should sacrifice one area of defence,
decreasing the current level of expenditure on land warfare, for instance, to
subsidise maritime, air, space or cyber warfare. Such thinking belongs to
the Australia of the last 75 years, a time when we could choose when to
participate in a war, and only send the forces we had and were prepared to
deploy. The current force is not a force that we can commit to a serious war
we have to win.
I do not question the motives of recent governments, but in the national
security area we are a product of our strategic upbringing. For most of the
last 75 years, Australia and its governments have expressed what they did
primarily in terms of ‘inputs’ to security, not in terms of the ‘outputs’ that a
nation needs. For most of the last 75 years, we judged ourselves by whether
we had met a certain percentage of gross domestic product (GDP). The
figure we settled on was 2%, though it has dropped below that. This figure
was derived from the demands made by the US to its NATO partners, who
were free-riding on the US by making a lesser contribution to European
defence because the US was prepared to make such a large contribution.
Two per cent of GDP might have been the right amount to spend on
defence for NATO countries while the US, paying up to 6% of GDP at
times, was the most dominant power in the world. But now the question
needs to be asked by all Australians: even if 2% was the right amount in the
past, is it the right amount for Australia to pay for defence now, at a time
when China, Russia, Iran and North Korea are rising to challenge the US,
and US military power is declining?
When in government, we in the Coalition boasted that we had finally
returned defence expenditure to 2% of GDP, after attacking our opponents
for having allowed it to drop in 2013 to about 1.6%. Our Prime Minister
and Minister for Defence then began saying that, if anything, 2% was a
floor for spending. My view is that 2% has become not just irrelevant but
dangerous.
Referring to our national security credentials in terms of defence
spending might have been acceptable for most of the last 75 years, when the
amount of money spent on defence was mostly decided by what was left
over after everything else in the budget had been allocated. But today the
dominant question for national security should not be how much do we
spend on defence, but what war or wars are we preparing to win and what
would that cost?
Money itself does not throw back the enemy from the gates or defeat the
danger on our doorstep. In our strategic environment it only matters what
we do with that money. Will 2% of GDP allow us to build sufficient modern
ships to assist our allies in deterring conflict in the Western Pacific, and at
the same time provide missile defence to vulnerable points on the
Australian continent? Will 2% allow us to provide enough airborne fuel
tankers to support our full range of fighter planes? Will there be enough
liquid fuel at allied bases to supply these air tankers? Will we have enough
missiles to support a modern war any time short of 10 or 20 years from
now? If not, then something is desperately wrong with how we are thinking
about defence.
The measure of what this nation does in terms of defence and national
security is whether it meets the need. And my intention in this book is to
define the need that must be met. At present, in the absence of a clear
defence strategy, Australia has ‘strategic objectives’, as expressed in the
2020 Strategic Update of ‘Shape, Deter, Respond’, to guide defence
planning, though sadly not national planning. The idea is to shape
Australia’s military strategic environment prior to conflict or war, to deter
actions against Australia’s interests, and to respond with credible military
force when required.
Using Shape, Deter, Respond is admittedly slightly better than referring
to our defence achievements in input terms – that is, the now meaningless
2% of GDP. At least Shape, Deter, Respond comes close to being an
operational concept for an unstated war against an unstated enemy at an
unstated time using unstated operational concepts and tactics. The problem
is that there is a deep reluctance in government and the bureaucracy to go
from these three well-intentioned words to stark reality, to add detail that
lays out what is going to be shaped, what is going to be deterred and what
will be responded to. Such words remain just words without appropriate
critical detail. Without any detail, how can anyone – defence planners or the
general public or even our potential adversaries – determine whether we
have allocated enough resources to these three objectives, or whether we
are deficient in any of them (and I believe we are deficient in at least two –
Deter and Respond), and by which point we are supposed to have achieved
them?
The essence of Australia’s strategy must indeed be to deter war, but if
we base that strategy on a bluff, then we can only bluff for so long, and
China is not buying it. It must be assumed that China knows more about our
national security, and particularly our defence potential, than 99.9% of
Australians. In fact, China has called us out on our defence weakness, and
laughs at us through the Global Times.
Deterrence is only successful if a nation can convince an adversary that
in the event of war they will incur debilitating costs, and the adversary
refrains from action. For the US, what must be deterred is relatively
obvious: war with China. But Australia must be very careful about what we
are trying to deter or we will look ridiculous. We cannot deter a war
between China and the US; we are too small and weak. We can contribute
to regional deterrence by committing our armed forces to a US-led
coalition, but we have so few forces in our one-shot defence force that if we
lose them – which we must accept as a genuine risk – we are in deep
trouble.
If we are clever and lucky and we have enough time, we might be able
to achieve alignment of Australia’s national and defence strategies with
how we fight the next war. But, of course, the whole process must start with
a strategy, and any other approach is likely to be guided by the worst form
of strategy: a strategy of hope.
***
Those outside the intelligence and policy communities often lack accurate
information about important matters like military and civilian security. So
we also need to ask ourselves: should we be talking about these matters at
all? Some have criticised me for doing so, and claim I am exposing
vulnerabilities.
My intention in this book is indeed to expose vulnerabilities, none of
which is classified. I have been running the argument that Australia is
inadequately defended since at least 2008, when my first book, Running the
War in Iraq, was published.2 But, really, I have held these views for the
second half of my 40-year military career, which ended the same year – in
2008.
The main motivation behind my decision to originally offer myself for
preselection as a Liberal Party senator for New South Wales was to address
the issue of national security by joining the government with the best record
on defence achievement, the centre-right federal Coalition government. I
stated this clearly during each of my four Liberal Party preselections, when
I offered myself to the party faithful to be chosen as a candidate. National
security was a central theme in my first speech in the Senate.
On any number of occasions, I have confronted ministers and
government officials and I have even been (very generously) given the
privilege of addressing the National Security Committee of Cabinet on
these issues. At no stage was I able to convince my colleagues of the merit
of my argument, I think because the concept of war for Australians is so
unfamiliar. In my last interaction, when I pointed out the need for a national
security strategy, I was asked if having a national security strategy was
likely to make the bureaucracy in Defence move faster than their current
glacial pace – a 10- or 20-year time period to change major capabilities, as
the progress of most of our major policies (infrastructure, submarines,
missiles, cyber, etc.) illustrates. I answered that a strategy would not
necessarily make things happen faster but that the right strategy might make
the right things happen, which is better than rushing to failure by building
the wrong capabilities for the wrong war, in a period that produces
irrelevant results after we have lost the next war.
CHAPTER 9
CURRENT POLICY
Many Australians across this nation are uncomfortable with what Australia
is doing in defence and national security, but, understandably, few if any
have the experience to make a detailed judgment. Instead, they depend on
me and others who are similarly qualified to provide what they are not
getting from the government. What they do want is for the government to
act, and to act effectively. Many are just not sure what ‘effective’ looks like.
Governments in Australia frequently say that they have no greater duty
than to keep our people safe and protect our way of life for future
generations. As a member of the recent Coalition government, I was
impressed by the way prime ministers Tony Abbott, Malcolm Turnbull and
Scott Morrison each improved upon their predecessors’ approach to
defence. Australia under the Coalition showed distinct signs of resolve.
More money was allocated to defence and there was a plan as to how that
money should be spent. In the past, the Coalition had reflected the
complacent view of the Australian people towards defence in some ways,
but more recently it led the Australian people much more effectively in this
area. It does not reflect a partisan position to say that only the Coalition can
be trusted on defence; this is what the record, at least of defence
expenditure and projects started or completed, shows. But we will have to
see how the new Labor government responds.
The aberration of Gillard’s security strategy aside, Labor during the
Rudd–Gillard–Rudd governments generally used defence as a cash cow to
pay for its social programs, reducing the expenditure on defence to pre–
World War II levels, 1.6% of GDP. This has meant that many of the
programs that should be in place now, when Australia is facing an
identifiable threat, are not there. The prime example is shipbuilding: the
Labor Party did not lay down one keel during the entire time it was in
power. As of May 2022, the Coalition had completed or started the building
or purchase of 70 ships for the Royal Australian Navy.
The Australian political party known as the Greens has an ill-considered
defence policy, essentially aimed at disarming Australia at a time when the
region is more uncertain than it has been since 1945. If the new Labor
government needs to rely on the Greens in government, then a suboptimal
approach to defence might be the price it is prepared to pay.
Because we must prepare for the kind of war that I have described in
this book, I believe Australia needs the Coalition. The last Coalition
government did more for defence than any previous government since the
end of the Vietnam War. But far, far more needs to be done, and I fear that
only the Coalition has the attitude and knowledge to do this. Yet Australia
now relies on a Labor government that will need to prove its credentials
with regard to national security.
One of the real strengths of the Coalition is that backbench senators and
members of parliament are permitted, in fact encouraged, to speak their
minds. If they have experience and expertise in a particular area, they are
given a hearing and listened to. Within the Coalition I have been given a
hearing and listened to. I have made my points strongly and without
prejudice. As I have said publicly, the biggest problem I faced as a member
of the recent Coalition government was that because it had done so much
for defence compared with previous governments, a request to do more
could sound like whining. ‘What more do you want us to do, Jim?’ was not
an unusual response to my suggestions.
The last Coalition government responded very well to the increase in
the terror threat since 9/11, through often world-leading legislation,
additional funding for security agencies, addressing cyber and online safety
and restructuring government departments. It also responded very well by
any international standard to the national security challenge of COVID-19
and its economic consequences. The enormous impact of COVID on our
health and economic resilience was without doubt the most immediate
national security threat for Australia in recent times, and could have
remained so if the government’s economic and health reaction had not been
so effective. The Morrison government paid in excess of AUD$331 billion
in COVID support payments to Australians during 2020 and 2021,
payments that I fully supported, at a time when such an amount could have
been used more directly to encourage self-reliance across the entire nation. I
supported them because they were needed to keep workers in jobs and
businesses running so that the economy could recover quickly, which it did.
A healthy economy is the basis of national security.
The management of the pandemic left the Coalition government deeply
experienced at governing during a crisis. Recovery from COVID is our
most immediate challenge during 2022, but China’s coercion and
aggression in our region could lead to a far more dangerous crisis that the
new Labor government will now need to manage.
***
On 1 July 2020, Prime Minister Scott Morrison gave a well-crafted and
very tough speech to accompany a more formal strategic update of our
defence policy. He also announced defence spending of AUD$270 billion
over 10 years – far, far too long a period. This new funding may marginally
increase the ADF’s lethality and sustainability once various types of
missiles are procured and brought into service. Unfortunately, the
investment will not increase the mass (size) of the ADF: $270 billion is a
very large amount of money, but it will not be enough to create a military,
or a nation, that can face the emerging strategic environment with
confidence.
Nevertheless, the speech itself was a good reaction to the aggression
and coercion of China over the four years from 2016 to 2020. At the time, I
totally supported the Prime Minister’s approach, mainly because I had not
heard anything like it for the 40 frustrating years that I had spent in the
military. At the time of writing, just after the election of the new Labor
government, it stands as our current defence policy, and so should be the
subject of some discussion here.
Prime Minister Morrison spoke of ‘Tensions over territorial claims
rising across the Indo-Pacific region, as we have seen recently on the
disputed border between India and China, and the South China Sea, and the
East China Sea. The risk of miscalculation and even conflict is
heightening.’ As a result of this, he said, ‘my government is making a
further commitment to better position Defence to respond to rapid changes
in the environment that I’ve noted’. He admitted: ‘The ADF now needs
stronger deterrence capabilities, that can hold potential adversaries, their
forces and critical infrastructure at risk from a distance, thereby deterring an
attack on Australia and helping to prevent war.’
The Prime Minister justified the approach and the extra resources by
saying that ‘The strategic environment and the heightened risk of
miscalculation in the region [make] this a necessity. There’s much more
tension in the world these days. We need an ADF that is ready now, but is
also future-ready.’ This was an admission, I believe, that war is far more
likely than most leaders are saying. I agree with the Prime Minister’s
assertion that war could occur at any time by mischance, but such an
accidental conflict might at least be limited in some ways. A planned and
intentionally executed war by China might not be limited in either time or
space.
The Prime Minister, however, went on, saying that ‘responding credibly
to threats doesn’t simply come down to the ADF. It’s about the system that
surrounds it, supports it – the ecosystem that it is a part of – and this is the
hard bit, it’s about the support and structures that [have] to do with the job.’
This was the closest the Prime Minister came to asserting that it takes a
nation to defend a nation.
Critically for this examination, the Prime Minister then said:
The strategic challenges of today and tomorrow call Australia in
many ways, as we’ve been called before at difficult times . . . 2020
has demonstrated once again the multiple challenges and radical
uncertainty we face, eerily haunted by similar times many years ago
in the 1930s . . . Our Defence Force will need to be prepared for any
future, no matter how unlikely, and hopefully not needed in the
worst of circumstances.1
In this reference to the 1930s, which Prime Minister Morrison made in a
number of other forums, he was most likely referring to the threats that
appeared to peace in Europe and Asia during that period, and to the weak
policy of appeasement adopted by European governments, which achieved
nothing, was morally bankrupt as it divided up smaller nations across
central Europe, and encouraged further intimidation, coercion and military
action by Germany, Italy and Japan.
This was a very good speech by any measure. It would have been
interesting to see how different a strategic update speech might have been
only one year later, given that our relationship with China deteriorated even
more in that period, and there were other developments, which I shall
discuss below.
What the speech missed, though, except in a single obtuse reference –
‘responding credibly to threats doesn’t simply come down to the ADF’ –
was that, regardless of how well prepared the ADF might be, unless the
Australian nation is also prepared for war, the ADF’s efforts will come to
not much at all. What the speech also lacked was a comprehensive view of
national security, and a specific statement as to the nature of the war we are
to prepare for, by when we must be prepared for it, and the changes needed
in government to initiate those preparations. There was also no mention of
the specific roles of each cabinet minister, because in our modern nation,
every minister has national security responsibilities for which they should
be held accountable through the National Security Committee of Cabinet.
In summary, there appears to be no coordinated national security
strategy from any government that aims to produce an ADF that is lethal,
sustainable and large enough in any timeframe. Neither does there appear to
be a plan to produce a resilient and self-reliant Australia in anything like a
reasonable period.
It is the job of governments to lead on national security issues. Much of
the national security threat is seen as military, but the solution in Australia’s
case concerns the resilience and self-reliance of the whole nation. The
Prime Minister’s speech was aimed at the initiated, at the defence and
thinktank communities and at policy wonks. But if the Australian people
are to be carried along and support any government, then a clear and
unclassified strategic assessment must at some stage be released to them. It
is the people’s security, and the people are paying for it. This is not
something that governments or the intelligence community are likely to be
comfortable with. But the Australian people deserve some straight talk
about the security threats they face.
Regardless of what Australia decides to do, we need to begin with a
comprehensive strategy. And most of that strategy should be made public.
***
The Coalition, however, had more important defence initiatives to come. On
16 September 2021, Prime Minister Morrison, along with the US President
and the UK Prime Minister, announced the biggest development in
Australian security since the 1950s. It was declared by the three leaders that
an enhanced trilateral security partnership to be known as AUKUS
(Australia, United Kingdom, United States) would be formed, and its first
initiative would be Australia’s acquisition of nuclear-powered submarine
technology – the consequence, of course, being that Australia would no
longer proceed with its existing French submarine program. Australia
would also gain access to other US technologies (such as artificial
intelligence and quantum computing), purchase missiles for current use, and
be given assistance in developing a future missile manufacturing enterprise
in South Australia.2
On 24 September, following the AUKUS announcement, Prime
Minister Morrison joined the Minister for Foreign Affairs and the Minister
for Defence in the US for the first ever in-person meeting of another
significant grouping, the Quadrilateral Security Dialogue, known as the
Quad, involving the US, Japan and India. There is no talk at this stage of a
defence treaty between these nations. Although the Quad is not as attentiongrabbing as AUKUS, it is probably as important, and the Quad meeting,
attended in turn by former prime minister Morrison and recently by newly
appointed Prime Minister Albanese, was an indication that China’s
aggressive attitude is coalescing the region, which is in Australia’s interests.
Much of the significance of the better-known AUKUS agreement has
been overshadowed by two issues: Australia’s acquisition of nuclear
submarines, and the fallout between Australia and France after the
cancellation of the French submarine contract. There has been much
criticism of the way our relationship with France, an important ally with
common interests in our region, has been handled. In general, though, the
announcement of the acquisition of nuclear submarines sometime in the
future has been extraordinarily well received in Australia, indicating the
unfocused concern Australians now have about their strategic environment.
The AUKUS agreement certainly caught Beijing’s attention and
resulted in a burst of hateful and threatening CCP rhetoric, even though the
AUKUS leaders never mentioned China directly. Ironically, AUKUS
increases both the likelihood that China can be deterred from taking
military action, and the likelihood that a war will occur sooner. The new
alliance must make China consider the balance of power, both in the region
and across the world. But it might also make the CCP think that if it wants
to take Taiwan by force, it will have to act soon or it will not be able to act
at all.
AUKUS came about as a result of the Coalition government’s belief that
our strategic environment had deteriorated to such a degree that we needed
to reaffirm our relationship with the US in particular, but also with the UK,
our traditional partner in security. It is not just our environment that is
deteriorating, but that of the whole region, and we are not the only ones to
see it. Other recent individual one-to-one agreements in our region, such as
various agreements between the US and Japan, and South Korea and the
Philippines, may also have the same contrary effect on China – that is,
encourage an early decision by China to go to war.
In October 2021 – at the same time as it became public knowledge that
a very small number of US Marines and special forces have been training
Taiwanese forces for more than a year – a six-nation joint naval exercise in
the Philippine Sea was intended to signal growing allied resolve. Journalist
Peter Hartcher wrote in the Sydney Morning Herald:
Seventeen naval vessels from six nations . . . conducted joint naval
manoeuvres in the Pacific . . . They represented Australia, Canada,
Japan, New Zealand and included US and UK aircraft carrier strike
groups . . .
The senior British officer, Commodore Steve Moorhouse, said
the joint exercise was ‘an important message for those here that
nations like ourselves really do believe in the freedom of navigation,
in the freedom of trade and really are alarmed at the militarisation
of the area. Chinese intelligence-gathering vessels were in the
region, so I have absolutely no doubt the message would have been
relayed back to China quite quickly.’3
In peacetime it is fairly easy for allies, especially from Europe, to agree to
conduct these kinds of joint deployments. This jolly attitude to allied
solidarity and resolve might not be so obvious if the US’s allies were being
faced with DF-21D carrier-killer missiles emerging from space at many
times the speed of sound, with accuracies of 20 metres, or massive anti-ship
cruise missiles fired from H-6 bombers. Or if the Suez and Panama canals
had been mined and their infrastructure entirely destroyed.
***
Without downplaying the significance of all that is contained in the
AUKUS announcements, I do think much more seemed to be publicly
ascribed to them than they actually contained. Wishful thinking plus media
sensationalism clouds the evidence that a huge amount still needs to be
done. The AUKUS agreement is indeed a momentous event for Australia,
but despite the popular and media reaction to it, Chinese aggression cannot
be considered solved by this announcement alone.
What is exciting with regard to AUKUS is that we saw decisive national
leadership on defence. The submarines, which may not be in the water for
20 years, are more symbolic of a significant change of attitude than they are
an answer to the China problem: a view also put up by many US
commentators and serving US naval officers.
But the significance of AUKUS is even wider. It represents a refreshing
of the 1951 ANZUS Treaty in that it reminded our major ally, the US, that
we are a serious player in Pacific national security, and so refocused the US
administration on Australia. It gave us access to US technology, the
potential for local missile production, and the possibility of stationing US
troops, aircraft, ships and submarines on Australian soil. Overall, it is a
mighty signal to China that Australia and its allies are not going to be
pushed around.
But there could be a big cost for Australians. As a result of these
decisions, which reinforced the togetherness of the three countries involved
in AUKUS, it could be considered that Australia is now even more locked
into our relationship with the US than we were under ANZUS, which
required us to consult in the case of a threat or an attack. AUKUS does not
work the same way but it does remind everyone in the world how close we
are to the US. This is not something I am uncomfortable with, as long as we
realise the limitations of US power and accept our responsibility over the
medium to long term to make ourselves self-reliant within that relationship,
and do not use the relationship as an excuse not to be self-reliant.
Even after the AUKUS announcements, I stand by my long-held
assessment that Australia as a nation remains dangerously vulnerable, and I
took Prime Minister Morrison’s statement that there were still more things
to do as acknowledgement of that situation.
Australia, as a key ally of the US in the region, wields significant
diplomatic and political power, and we have seen that in action throughout
2021. Respect for Australia was apparent when Japan asked Australia to
lead in the region, and it was Prime Minister Morrison’s advocacy that
prompted the Quad and AUKUS agreements and meetings.
But in the end only real power matters, which must be the most
important lesson from the Ukraine War. Australia’s defence potential is
significant. We inhabit a single island continent and have no threatening
close neighbours. We have a large population that is healthy, well educated
and technically oriented. We have natural resources that could make us selfreliant if we chose to be. We are rich and our population is generally well
motivated. And we are well governed in comparison with the rest of the
world.
Yet our defence potential is far from being realised. Unless Australia
adopts emergency measures, such as declaring a defence emergency and
then using the powers given to the Commonwealth under the constitution to
override, for example, the states’ refusal to exploit oil resources under the
ground or to ensure the continuation of other energy for industry, we will
need years to develop enough military and national power to become selfreliant within these alliances. This is despite the significant steps that the
Coalition government took after it came to power in 2013. Every indication
is that we may not have years. If this is the case, as I believe it to be and, as
we have an obligation to assume it to be, then we should recognise that we
are in the early or mid-stages of an emergency situation, and it is legitimate
for the Commonwealth to approach the states to cooperate in security
aspects and, if they refuse, to claim federal emergency powers. The
likelihood that we do not have time to become self-reliant before China
challenges the Western Pacific is why I support an even stronger alliance
with the US as our only option, particularly in the short to medium term.
As James Curran in Defense One argues, ‘AUKUS represents the death
knell for strategic ambiguity in Australian foreign policy’, and that may be
a good thing:
Although Canberra does have a record of shrewd alliance
management in the past – of roaring loud in allied solidarity but
committing the minimum muscle up front, as in the case of Iraq
2003 when Australian special forces were pulled back after the
initial assault on Baghdad – it is fanciful to suggest that in any
future military conflict with China, especially over Taiwan, the
United States will not expect Australia to play a role in the battle.
Will a future Australian government of either political
persuasion be able to resist US pressure to be part of any such
conflict? History suggests not. Australia, as we are constantly
reminded in the speeches, has been by the US’s side at every major
conflict since the First World War.4
According to the South China Morning Post, a new report released by the
Sydney-based United States Studies Centre argues that ‘Australian and US
military forces should integrate further under a “collective deterrence
strategy” aimed at China’s rise, giving Canberra access to American
operations in the Philippines, Singapore and Guam . . . The allies should
look at new “combined access arrangements” among a number of ways to
strengthen “integrated deterrence” against Beijing’s growing assertiveness
in the region, according to the report.’5 So, an Australian government may
at some stage be asked to contribute to a force in the Western Pacific, which
may initially be intended to deter China from acting militarily.
In probably the clearest statement on this subject by a senior member of
the previous Coalition government, the then Minister for Defence Peter
Dutton said that his overriding strategic vision was to make provisions for
‘the threat of conflict’ in the region and deal with an assertive China by
addressing our ‘lack of preparedness . . . [China’s] been very clear about
their intent to go into Taiwan and we need to make sure that there is a high
level of preparedness, a greater sense of deterrence by our capability, and
that is how I think we put our country in a position of strength’.6
Mr Dutton noted that China is an economic and military superpower
that admits to spending 10 times more a year than Australia on defence, and
to producing more military assets by tonnage every 18 months than the
Royal Navy has in its current fleet. So the thought that Australia could
compete with China ‘is of course a nonsense,’ he said. ‘That’s not the
question before us; the question is: would we join with the US?’
Mr Dutton added: ‘It would be inconceivable that we wouldn’t support
the US in an action if the US chose to take that action. And, again, I think
we should be very frank and honest about that, look at all of the facts and
circumstances without pre-committing, and maybe there are circumstances
where we wouldn’t take up that option, [but] I can’t conceive of those
circumstances.’7
Mr Dutton’s comments came after former Labor prime minister Paul
Keating told the National Press Club that Taiwan was ‘not a vital Australian
interest’ and not recognised as ‘a sovereign state’, and that Australia should
not be drawn into a conflict over the island.8 Mr Dutton described Mr
Keating as a ‘grand appeaser’.9
Mr Dutton is right when he speaks of our ‘lack of preparedness’. What
he means by ‘preparedness’ is complex. A military can possess certain
items of equipment (ships, planes, tanks) which, when they are combined
with personnel, training and leadership, produce what is called ‘capability’.
Capability refers to the ability to actually do something and for a military,
that is to fight. For example, a country may have a certain number of
warships but, in order to save money in peacetime, the warships may not
have their full complement of crew to operate 24 hours a day in a battle
situation. The ship may not have the full number of protective weapons to
stop missiles because those weapons are expensive; instead of having five
missile defence weapons per ship, the ship may only have one which the
crew can train on, with the intention of fitting the other four as war
approaches. There may also not be enough money in the defence budget to
fully train all the crew and to maintain that level of training. So the ship
lacks preparedness, as the former minister says.
And that applies across the entire Australian military for every type of
equipment and the personnel in every unit. To save money, only some
elements of the ADF are kept at full war or operational readiness. This is a
legitimate policy whereby the money saved can be spent on buying more
equipment with the intention of increasing the readiness of the force as war
approaches. The key to this policy is the ability to assess the risk of war
early enough to raise the preparedness of all elements of the force. In times
of wars of choice, Australia could maintain almost all its force on lower
preparedness levels because we knew we could determine when we went to
war. In these days, with the possibility of wars of commitment that we must
win, where an enemy determines when the war occurs, it is much more
difficult to assess risk, and much more important to get preparedness right.
As I have maintained, a war with China is likely to occur any time from
now out to three to five years in the future. This is the time Australia and
our allies have to develop the capability to counter China, to build more
military capability and to raise the level of preparedness of the capability
we already have. Three to five years may not be enough. China can be
deterred, but that will take real military strength by the US and its allies in
the immediate future.
The question that government and military planners must ask is what
needs to be done in Australia in the short to medium term – that is, the next
five years, the outer limit of my estimated timeframe. Having long-term
objectives is commendable, but unlikely to produce results for the next war.
Australian national security will be best served if the government takes the
right war as the standard it must meet, examines what it needs to do to
prepare for that war, and only then decides what risks it can afford to take
by not preparing.
As I have repeatedly stressed, if there were an understanding of the
likely nature of such a war, and a risk analysis were conducted and
preparations commenced, Australia should be able to handle anything that
the real world throws at us, short of a nuclear war. Sadly, it does not work
the other way if Australia prepares only for lesser contingencies like the
kind of wars we have been fighting for the last 75 years. In modern war,
where decisions are likely to be reached quickly, if we prepare for a lesser
contingency we may never get the chance to adjust upwards.
Compared with the likely timeframe for war in our region, Australia’s
ability to increase our military strength is severely limited. The previous
Coalition government claimed that the first of the eight planned nuclear
submarines would be ‘in the water’ by 2039. The US Chief of Naval
Operations agreed that the process will take perhaps decades. What has
been definitely decided is that there will be an 18-month study, possibly
later reduced to 12 months, to identify the details, and that is in train now.
The key to making the submarine agreement relevant any time in the next
20 or so years is to identify the proper procurement strategy, and that is
what the study will do.
Many, including former prime minister Tony Abbott, are clamouring for
nuclear submarines from either the US or the UK to be delivered much
earlier than 2039. While the American Virginia class submarines are
excellent, the US is in no position to shift production efforts and help
Australia get started any time soon. Among the biggest challenges is that
the US Navy is critically short of submarines itself. It was assumed that
certain parts of the Virginia attack submarines would last the life of the
vessel, but as in-service Virginia submarines arrived for maintenance, the
navy found it had to replace parts that weren’t supposed to need replacing.
Since there weren’t spares available for these unexpected replacements,
maintainers began taking the parts off the production line, thus slowing
down production at the only two shipyards building these submarines,
General Dynamics Electric Boat and Newport News Shipbuilding. Even if
the US could increase nuclear attack submarine production at these
shipyards, Washington would likely serve America’s needs before
Australia’s.10
The UK, on the other hand, has almost completed its equally excellent
Astute-class attack submarines. That might leave UK shipyards free to start
building Astute subs for Australia. The specifics of the nuclear reactor and
the front of the submarine where the weapons systems and sensors are –
where we may prefer US technology to ease the transition for Australian
submariners who primarily operate US systems on the current submarines –
could be worked out over the next year.
There is no reason why the Australian government should not try to
speed up delivery. Nuclear submarines in two decades’ time will be of no
value to anyone if war occurs within the next five years. The question is: is
it possible to speed up delivery? The answer looks like being no!
As a result, we are likely to fight the China war not with nuclear
submarines but with our six diesel–electric, Australian-made Collins-class
submarines. Given their age and plans for a mid-life update or two, we will
probably have only two initially available in the short term, and three or
four in an emergency over time.
The decision following the AUKUS announcements to extend the life of
the Collins class by a decade, from 2030 to 2040 (when the nuclear subs
will be in service), revealed in a statement in Senate Estimates by the
Australian Chief of Navy, seems to have confused commentators and the
submarine lobby alike.11 Certainly, any government needs to explain why it
was decided or agreed to in a 2009 defence white paper that we needed 12
submarines then but now we can make do with only six submarines, in a
worsening strategic environment, until 2040, at which point we can get
away with eight nuclear ones.12
Given the long lead time for the delivery of Australia’s new nuclear
submarines, we are more likely to fight a war with China using our
existing Collins-class submarines. (Getty)
Clearly Australia must cover the gap between now and when the nuclear
submarines are ready in one or two decades, in part with the Collins, but
there are other options. It must be remembered that our aim should not just
be to have submarines. Our aim should be to have the capacity to sink
enemy ships and submarines, to lay sea mines, to insert and deploy special
forces and to conduct surveillance. All of these tasks can be carried out by
aircraft, shore-based missiles or air-dropped sea mines. The only job
submarines can do that other forces cannot is a very important one: lurking
in an area to establish a persistent presence and make an enemy think that a
submarine could be anywhere. A logical plan, derived from an overall
defence strategy prepared as a subordinate strategy to a national security
strategy, would show the need for more forces to supplement the Collins
submarines, probably aircraft initially, because they can do most things that
a submarine can do, and can be procured much faster than anything else.
This raises the question of what military forces we do have and how
effective they are. We have a highly competent and modern military force in
Australia which has a good international reputation based on our fighting
performance over our history as a nation and, in particular, on our more
recent experience in Vietnam, Somalia, Cambodia, East Timor, Iraq and
Afghanistan.
It is only possible to judge a defence force against the kind of threats it
may have to face, and the Australian military has performed well in all of
our wars of choice since the end of World War II because we are good
soldiers, sailors and airmen and women, and we have always fought beside
our great and powerful ally. I often observe that the ADF has never been
better than it is at the moment, particularly with the support and funding it
received from the Coalition government from 2013, but that is judging it
against the wars of the last 75 years and not against what the ADF may
have to do in the future. I stand by the case I made earlier in this book that,
despite the advanced nature of our equipment and our moves into cyber and
space warfare, the ADF is a one-shot defence force because it lacks lethality
(it cannot fight hard enough), sustainability (it cannot fight for long enough)
and mass (it is not big enough).
Overall, our navy is very small but attracts a disproportionate amount of
examination because of the issue of Australian-built submarines and ships.
The previous Coalition government supported the upgrade of navy surface
ships and the procurement of destroyers, amphibious ships and logistic
support ships, as well as offshore patrol vessels. But the size of the navy
needs to be examined, not according to what we think we can afford, but
what we need. Regardless, it is very difficult to increase the size of a navy
with complex modern ships quickly, as the US is finding out as it tries to
grow its navy to match China’s navy.
Following their recent upgrades, the Australian navy’s eight Anzac-class
frigates are very capable ships, especially as regards their sensors,
especially their radars, but some believe that they need many more missiles
on board, in the form of a second eight-cell vertical missile-launching
system on each ship. This system, known as the Mark 41 Vertical
Launching System, has missiles preloaded into ‘canisters’, which are then
loaded into the individual ‘cells’ of the launcher, four to each cell, giving it
a total of 32 missiles. These missiles (known as Evolved Sea Sparrow
Missiles) protect ships against attacking aircraft or missiles, even
supersonic missiles that can manoeuvre to protect themselves as they attack.
The role of these ships is to escort larger vessels in a fleet, convoy or
battle group and defend them against short- to medium-range attacks. In the
event of hostilities in the Western Pacific, where they are likely to face
many Chinese subsonic, supersonic and hypersonic cruise missiles as well
as rockets, this small missile-launching capacity is risky to say the least,
particularly as the ships have to return to a harbour to reload missiles.
Australia’s three Air Warfare Destroyers (known as Hobart class after the
name of the first ship) have 48 vertical launch cells each, but modern US
and Chinese ships have twice that number.
In a very good move, the Coalition government planned to acquire the
longer-range (some versions exceeding 1000 kilometres) Tomahawk
missiles for use on the six Collins-class submarines, the three Air Warfare
Destroyers and the eight future Hunter-class frigates, which will come into
service early in the next decade to replace the Anzac-class ships. These
missiles will all come into service sometime in the future, but unless the
new Minister for Defence can work miracles, Australia will not be selfreliant in missile production for the next war.
Our army, like our navy, remains very small. The Coalition government
decided to strengthen the army with very capable armoured vehicles,
missiles, drones, helicopters and command-and-control IT. But, again, the
size of the army is a result of the needs of previous wars and is not related
to the wars we may have to fight in the future. In particular, Australia does
not have a reserve force, as many other countries have, which, while not as
well equipped or as well trained as a full-time army, could be prepared for
mobilisation in times of tension.
The air force, meanwhile, is in the process of finishing its purchase of
the advanced F-35 fighter aircraft (an aircraft I shall discuss in the next
chapter), but of course what our F-35s, our FA-18F Super Hornet fighter–
attack aircraft and our Growler EA-18G electronic-warfare aircraft can do
is severely limited by the lack of sufficient numbers of refuelling,
command-and-control and other specialist supporting aircraft. If we were
preparing for the right war, the balance between supporting aircraft and
attack aircraft would have been addressed. This imbalance is not a problem
for an air force going to a war of choice, but it is a problem for an air force
that has to deploy all its assets in an all-out attempt to defend a nation.
Given the size and capability of our military, the question Australian
military planners must address in the event of war with China in the
Western Pacific is: if Australian ships and planes are sent to a serious fight
in the Western Pacific alongside the US, can we afford to lose those ships
and planes that are so critical to the defence of our continent, especially if
the US is forced out of the Western Pacific for some period? A historical
analogy is that in 1940 Australia sent our army to North Africa, our navy to
the Mediterranean and our air force to Canada to be trained and then to
Europe to fight, when the likelihood of a Japanese attack in our region was
well known and subsequently occurred.
So, Australia has enormous defence potential, but fully realising that
potential will take time and a total change in national attitude, if we decide
to do so, or if we have to do so.
***
Going to war is a monumental decision and, as such, should not be rushed.
The merits of a military contribution by Australia should be very seriously
examined as tension builds in the region. In fact, they should be examined
by Australia in the greatest detail long before tension occurs. They should
be examined now, side by side with the US and other likely allies – Japan
and South Korea to begin with – with all options and expectations of our
allies on the table.
Our involvement in such a war in China’s backyard along with the US
should only be agreed upon by Australian authorities if there is more than a
fair chance of success against China, and at present the US may not inspire
confidence in its allies that there is such a fair chance. The most basic
lesson we can learn from our recent involvement in a range of wars is that
we, even as a minor ally, need to influence US warfighting strategy, and that
needs to occur now, before the build-up to a war begins. Australia must
demand a seat at the strategic warfighting table. After all, this is the way
NATO works.
Australia also needs to be very clear about why we would go to war
over Taiwan. The final decision will be taken by the government of the day,
but it will not be able to take that decision responsibly at the time unless
there has been significant preparation, essentially for years, beforehand. If,
of course, we have years.
That preparation would mainly be to create, in the first instance, the
physical capability to go to war, that is, weapons and trained personnel
backed up by resilience and resolve in the nation. But it would also be
necessary to develop a strong mental capacity to go to war, that is, effective
and ethical decision-making at government level, supported in execution by
resolve among the population.
If we do not have the military capability to go to war effectively, yet do
so, that would be truly immoral. Effectiveness in defence in this kind of war
is about both protecting our homeland and deploying forces in the Western
Pacific along with our allies in order to deter China from attack, or in order
to fight if deterrence fails. If Australia were so vulnerable that sending our
forces away to war risked bringing down the wrath of a superior enemy on
an unprepared nation, that would be very hard to justify. As noted, what
Australians generally consider at the moment to be a war in the region of
Taiwan might just be the opening battle of a much wider regional war. Later
aggression might even be aimed directly at the Australian continent.
Because the ADF is essentially a one-shot force, Australia should be very
careful about where it deploys that one shot.
Moreover, we only have an obligation to become involved in war only if
the war itself is just, and only if we commit to prosecuting such a war in a
moral and ethical way. This means we need to make assumptions about the
nature of this war and determine whether our involvement would be right.
Having run a war in Iraq on that basis, I can only say that the ability to win
need not be hampered by prosecuting the war in an ethical way. In fact, to
run a war unethically risks losing both the physical and the moral high
ground. You can win ethically.
To defend Taiwan after China launches a war of aggression is a
relatively straightforward moral issue. Opposing aggression is acceptable.
To defend Australia if this continent were attacked as part of or after a
Taiwan war, which is a distinct possibility, is likewise morally
straightforward. Australia is permitted to act in self-defence, even if that
self-defence is manifested in offensive ways, such as joining our allies to
fight China in the Western Pacific.
But what if the Taiwan war were instigated by an irresponsible attempt
by elements in Taiwan to declare independence? Here the issues start to
become complex, which is why it is so important to have discussed all
aspects of the war far in advance of a time of rising tension, when national
leaders may be under extreme pressure.
A response to a war of aggression, past a certain immediate point of
self-defence, must be approved by the UN. But what if the aggressor, in this
case China, so dominated the UN Security Council that approval by the UN
became impossible? Once again, this dilemma should be considered far in
advance of the start of any war.
If a nation has decided that it has the capacity to deploy forces to a war
while at the same time protecting the homeland because it has created those
forces over time, and it has satisfied itself that it is morally justifiable to
become involved, a rational nation then needs to ask: is the war winnable?
This is particularly important if the war being considered is one involving
an existential threat to that country. Again, if there is a lack of confidence
that the battles are winnable, much less the war, and that even a stalemate
would be too risky for a small nation like Australia, our participation might
be morally questionable.
Because of this, the next issue that a responsible government should
address is whether to send a token force to such a war to fight beside our
allies or send a serious combat force that is committed to winning and is
prepared to fight and even to die? Does Australia envisage doing roughly
what we have done with our US allies for years – sending a minimal force
surrounded by directives and rules of engagement that prevent it from doing
any serious fighting, so that the local US commander sees that force as a
burden rather than a benefit? Or, alternatively, does Australia send a
maximum force to join our allies, one that can fight to win and is prepared
to take casualties, unlike every commitment by Australia post-Vietnam?
Given our general relationship with the US, and the new AUKUS
alliance in particular, there will be great pressure on Australia to stand
beside our ally to the maximum extent, even with the limited military
resources we have. We will probably be asked to contribute at least the
largest possible air and naval units to a coming war. This pressure will be
hard to resist.
If a regional war comes to pass, Australia cannot get away with what we
have got away with for decades: sending a minimal force and our national
flag, but not sending hard, strong warfighting forces. To deter China – if
there is a chance to deploy forces before China attacks, which may not be
the case – sending only one token frigate and a few token non-combat
aircraft, as we usually do, would make this country a laughing stock in the
eyes of our allies. In the next regional war, if there is time, Australia will
find that it must commit forces with a view to winning, not just
participating as has been our practice for far too long. An existential threat
from China requires a committed response by Australia. If Australia does
send a substantial force that is prepared to fight and not just show the flag,
it is likely that there will be severe casualties. And if the scenario that I
predict in this book comes to pass, then many ships and planes will be lost.
It is unlikely that any action Australia might take will ever deter a war
between China and the US – we are simply not a big enough player. But if
we are to be involved to the extent that we suffer even just collateral
damage, we must prepare carefully for two possible outcomes.
The first is the likelihood of an attack on Australia as a US ally – be it
by rockets, missiles, sea mines, cyber technology, biological warfare or a
trade embargo. Our aim should be to at least reduce our vulnerability to
these types of attacks. If we carried out a full strategic assessment, we
might discover a large number of deficiencies in this regard. Consequently,
we might decide to invest in an ability to defend either the whole nation or
the most important parts of the nation against attack by, say, ballistic
missiles – we have some relatively important targets on our shores,
although missile defence is a very difficult thing to do in a country as big as
Australia and might be disproportionately expensive for the return in terms
of security.
Alternatively, or in addition, we might increase our capacity to manage
a sea-mine attack on our ports, particularly because it might be important
for China to limit the ability of Australia to be used as a base for US forces,
or it might be of value to China to stop the Australian navy from deploying
from its harbours. A consequence of such a sea-mine attack would be that
no commercial shipping would be willing to use our harbours because of
the danger of losing a ship to a mine or the inability to obtain insurance for
a voyage to Australia. Australia is extraordinarily dependent on sea trade
even for basic commodities such as liquid fuels, pharmaceuticals, fertilisers
and IT, yet it currently has limited ability to clear its ports of modern sea
mines. It’s said, anecdotally, that only one or two of the seventeen
nationally significant ports could be kept open by the Royal Australian
Navy.
We could of course improve our resistance to economic attack by
producing what we need domestically, such as liquid fuels, by building up a
sizeable reserve of fuel or other critical items within the country, and by
expanding domestic production of those items in a realistic period of time.
That would not mean withdrawing from the global marketplace, but it
would mean producing domestically everything we need and not
necessarily everything we want. This would require the government to
identify critical items and manage the market with controls and subsidies.
Rather than assume that our military forces, mainly navy and air force,
would deploy to the vicinity of the first and second island chain to join with
a US fleet because that is how we have thought of a Taiwan war for
generations, we might decide on a different contribution to a coalition
operational plan. We might suggest to the US that Australian forces, along
with regional allies such as Indonesia, could control sea and air access
through the straits in the Indonesian Archipelago and the South Pacific
while at the same time protecting the Australian mainland. This would be
achievable, is within the bounds of feasible planning and would be a
sensible use of scarce Australian resources. And if the allied war effort were
well coordinated, such an Australian operational concept would contribute
significantly to the overall war aims.
The second contingency that Australia should prepare for is that the US
is forced out of the region by the kind of massive and coordinated surprise
attack that I outlined at the start of this book, and Australia then stands
alone as an ally of the US in the face of a dominant China, but without US
support. To face such a contingency, we would need to be self-reliant, and
capable of withstanding China’s direct intimidation – even threats of
military occupation – or risk becoming a tributary state of China, one that
loses its sovereignty because it is required to obey all of China’s directions.
***
Given the destructive power of modern weaponry, the aim of every person
on Earth must surely be to deter war wherever possible. This is why the
moral aspect of war is critical. To be able to wage war that involves extreme
violence, yet not lose the moral high ground as an individual or as a nation,
is an essential way of surviving the most awful of human experiences.
Sadly, over the full expanse of history, war has been one of the most
common of human experiences. In the last century, it is estimated that 17
million people died in World War I and then another 60 to 80 million in
World War II, and that is not counting the myriad of other wars in places
like Korea and Vietnam and Iraq.
War is so appalling that all Australian governments have a solemn
responsibility to prepare for it in order to deter it or to minimise its impact
should it occur. And the people of Australia should be vitally interested in
what their government is doing.
Our world is a tough world, but there are still things in it that are worth
making great sacrifices for.
CHAPTER 10
THE RIGHT WAR
What is the right regional war to prepare for, and how should Australia
prepare for it? I fail to see how any assessment of Australian defence
policies can proceed without a focus on answering these two questions.
Preparing for the right war is critically important, because until we
understand what a regional war might entail, we have no way of
determining whether Australia is effectively defending its sovereignty, the
ultimate aim of national security.
As I have been arguing, China has the capability to go to war to achieve
its aim of dominance, first in the region and then perhaps across the world,
and may do so, and Australia must prepare. The war will be initiated by
China at a time of China’s choosing. It will be as widespread and as violent
as China chooses it to be. China has been preparing for this war for decades
and it intends to win.
If a major power like China gets to choose when to initiate hostilities
because its major objective is to be the dominant power in the region, it has
an enormous, even overwhelming, advantage simply because it has the
initiative. China can exercise the advantage that this initiative gives it in one
of three main ways:
1. The incremental model. This model involves using a wide variety of
techniques to pick an issue to advance China’s influence and interests
– whether it be the building of islands in the South China Sea and the
militarisation of those islands; the incorporation of other, mainly
developing, countries into China’s infrastructure building plans across
the world, known as the Belt and Road Initiative, and the use of that
influence for military purposes, as in the South Pacific; the suppression
of dissent in Hong Kong; skirmishes on the Sino-Indian border in the
Ladakh; cyber activity and the theft of intellectual property from
foreign companies and universities; the use of social media to create
distrust of institutions in other countries; the weaponisation of trade;
and attempts to extend influence by bribing decision-makers in foreign
countries. The essence of the incremental model is that no one activity
is enough to cause a significant reaction from the target countries, but
put together they have a real benefit for China. Should any of the target
countries react to any of the increments, then China denies
involvement, lies about the issue and/or rapidly backs off. In many of
these spheres, China achieves its objectives without causing a dramatic
reaction from any of the target countries and, when there is a reaction,
it is usually mainly rhetorical. Behind the incremental moves there is
always the implied threat of direct military action by China, which is
one reason why target countries are cowed.
2. The sudden-attack model. This model involves the sudden and
violent use of military might, with the intention of achieving a decisive
result, at least in the short to medium term. This model only became
viable for China in the past decade or so because for many decades
prior to the early twenty-first century, the power of the US effectively
deterred such a move. China’s advances in space technology, cyber
warfare, nuclear weaponry, rocketry, missiles and conventional
military forces (ships, planes and tanks) mean that this option can now
be considered. What has also made this model viable is the reduction
in US military power since the end of the Cold War in 1991 (discussed
previously), and a perception that the West is in decline and US
leadership is weak. This model depends very heavily on surprise and
every effort would be made to not signal that a sudden attack was
imminent.
3. The combination model. As the name suggests, this is a combination
of the incremental and the sudden-attack models. China would
continue to use the incremental approach, taking smaller, apparently
unconnected steps to either achieve its objectives without the costs of a
major war, if that were possible, or, over a longer period, it might act
incrementally to create the conditions for a sudden attack if the
incremental approach was not bringing about regional dominance.
When China was ready or when there was an opportunity, then it
would suddenly attack. The incremental model was used by Germany
prior to World War II, when it employed both diplomacy and coercion
throughout the 1930s to take advantage of European naïveté and post–
World War I war weariness. Violent armed force was not employed by
Germany until it had built up its military to a necessary strength, and
when the likelihood of achieving its objectives by incremental methods
had diminished. This resulted in an incremental gaining of Germany’s
strategic objectives, always with the option of backing off if it
overreached and met with more opposition than expected.
A comparison of China with Nazi Germany serves to illuminate this
argument. China has already embarked on a version of the incremental
model, building over time on its grey-zone activities as a means of
weakening and intimidating the US and its regional allies, while making
real territorial gains, just as Germany did. All of the grey-zone methods
China has utilised are incremental techniques based on bullying and
coercion, backed up by the threat of China’s massive armed forces.
China has positioned itself well through these grey-zone operations and
thinks that the US and its allies – which as I have said it believes are weak
and in decline – have been successfully cowed. (This is why the AUKUS
agreement is so important: to demonstrate to China that there is resolve in
the region.)
In the vicinity of Taiwan, China has recently been carrying out various
intimidatory activities as part of its grey-zone strategy, such as regularly
flying large numbers of attack aircraft across the centre line of the Air
Defence Identification Zone (ADIZ) between Taiwan and China. Chinese
air force flights towards or past Taiwan increased through 2021 and 2022,
with periods of near-daily flights, usually involving a small number of
planes, though in mid-2022 even attack helicopters were used. In addition
to air incursions against Taiwan, China has been conducting amphibious
exercises in the area as well naval live-firing exercises. It has also
conducted almost daily transgressions of Japan’s airspace, requiring the
dispatch of Japanese fighter aircraft to intercept the Chinese aircraft and
escort them out of the area. While these activities have been increasing
generally, large incursions usually appear to be in response to particular
events, for instance US arms sales to Taiwan, or military activity by Taiwan
in or near the Taiwan Strait.
Perhaps the upper limit of the incremental option for China, just short of
the use of armed force, would be to declare a no-fly, no-sail exclusion zone
around Taiwan based on some pretext, to be subsequently enforced by
Chinese air and sea forces. Such an exclusion zone would prevent all
aircraft and ships, military and civilian, from entering or leaving. China
would have to deploy a credible military presence prepared to use force,
including the sinking of ships and the shooting down of aircraft, to maintain
the zone. Academic Peter Layton writes:
The final trend is that having undertaken apparently successful greyzone activities, China may use the technique elsewhere including
against Taiwan . . . President Xi Jinping will not leave the
unification of the mainland and Taiwan to future generations. While
large-scale military conflict is improbable . . . a more likely scenario
is a protracted and intensive campaign by Beijing, using ‘all means
short of war’. A grey-zone operation could be commenced that aims
to destabilize Taiwanese society and force its government to enter
into unification talks.1
A no-fly, no-sail exclusion zone around Taiwan would technically constitute
an act of war and would permit the US and its allies to use force to counter
it if they thought themselves strong enough. Given this, China could wait to
see how Taiwan and the world reacted while slowly bankrupting Taiwan. If
China encountered more resolve in its opponents than expected – which at
the moment does not look likely – it could either back off or, if it were
sufficiently prepared and wished to do so, rapidly move to the full use of
armed force.
If the US decided to do nothing except protest, boycott, sanction and
attempt resolution through the UN, China would use its influence at the
UN, where it controls many of the international bodies, to delay any
counter-action. US credibility would be increasingly diminished the longer
it failed to act, while Taiwan would come slowly under China’s control.
And if Australia did not insist on being part of the US decision-making
process in the leadup to such events, we would deserve the insecurity we
would then face.
The sudden-attack model was used by the Japanese when they attacked
Pearl Harbor: a surprise assault out of a clear blue sky, delivering a
devastating blow, from which it was hoped the US would not be able to
recover until imperial Japan had consolidated and fortified its gains as far
south as the Dutch East Indies. If China decides to take sudden, direct and,
it would hope, decisive military action in the immediate future rather than
rely on incremental grey-zone action over time, it is likely to do so only in
its own backyard, where its military strength is most pronounced and where
the US is weakest. China would want to have all the advantages, especially
short lines of supply and the ability to quickly magnify its military
strength.2
Based on its assessment of its opponents, China could act with little or
no warning, as Japan did in 1941, or it could manufacture a reason to act.
For example, it could take advantage of arrangements such as the 20-yearold Sino–Russian Treaty of Friendship, addressing ‘defence of national
unity and territorial integrity’, which was extended for a further five years
in mid-2021. Meeting just prior to the February 2022 Winter Olympics,
President Putin and President Xi reaffirmed their relationship, declared their
opposition to any expansion of NATO, and affirmed that the island of
Taiwan is part of China. Their 5000-word joint statement also highlighted
what they called ‘interference in the internal affairs’ of other states.
Ironically, this declaration was made as Russia assembled its military forces
around the border of Ukraine, and just before the invasion of that country
by Putin’s forces. Reports claimed that Xi asked Putin to delay military
action against Ukraine until the Beijing Winter Olympics had concluded.
The Winter Olympics finished on 20 February. On 21 February Putin sent
his troops into the two disputed provinces in the Donbas, and on 24
February the Russian invasion began in earnest. The message to the world,
especially to the US, was clear: don’t interfere in Ukraine or, by
implication, in Taiwan if something should happen, and be aware that China
and Russia stand shoulder to shoulder.
A sudden Chinese attack in the Indo-Pacific region could feasibly be
coordinated with Russia and/or other allies such as Iran, North Korea and
Venezuela; however, there is no indication of that being likely at the
moment but it must be planned for.
Any reduction of the US presence in the Western Pacific might also
encourage China to launch a sudden attack. Such a reduction could stem
from events almost anywhere in the world – for example, if the US
redeployed Pacific forces to support Ukraine in its ongoing war with
Russia, or if, say, Iran closed the Persian Gulf, thereby suspending all oil
shipments to US allies such as Australia. The US and its allies might then
send military forces to open that seaway.
The sudden-attack option could gradually come to be seen as a
preferable alternative to the incremental model if China observes continued
weak US and allied responses to its grey-zone activities, or if it becomes
aware that it cannot achieve its objectives by incremental methods. Such a
course of action would be very high risk, but it could potentially bring very
high rewards for China, so it will always be enticing – just as the Japanese
considered the planned attack on Pearl Harbor very high risk, but still went
ahead with it. Given the potential for a sudden attack, Australian planners
must focus on this worst-case but logical scenario, regardless of how much
more comforting it might be to dwell on less-demanding scenarios.
***
In the vast Indo-Pacific region, the US military depends on a complex of air
and naval bases, a wide range of support aircraft and ships, intelligence
gained from satellites, and a communications network capable of receiving
and distributing large quantities of data to and from headquarters and allies.
Combined with the warships and combat aircraft that carry the attack to an
enemy, or defend US and allied interests in the Pacific and the Indian
oceans, this results in an enormous US military force of 375,000 personnel,
known as the Indo-Pacific Command, or USINDOPACOM. It is this force
that makes the US the dominant regional power.
To defeat this kind of military capability even just in the Western
Pacific, an enemy such as China would have to attack the bases, the aircraft
and ships, but also critical infrastructure such as fuel and communications
to reduce the ability of the US and its allies to reinforce the Western Pacific
from the continental US or anywhere else in the world. And China would
have to attack in a way that did not cause the US, if it suffered such a
defeat, to use its overwhelming number of nuclear weapons as a first
(nuclear) strike against China.
Something as complex as this military command inevitably has a
number of vulnerable nodes. The Indo-Pacific Command is of course,
dependent on the intelligence that it gets from space through a range of
satellites and also from space-based communications, through which it
receives directions from Washington and disseminates its own orders to
subordinate elements across half the world. This is backed up by civilian
seabed cables which carry encrypted military data, plus there are backup
systems often based on high-frequency radio, to be used if all others fail.
Because of the nature of radio waves, these backup systems have very
limited bandwidth and so are slow in transmitting data. The internet, which
is carried primarily through seabed cables, is very important for the
military, but it is not just the cables that are vulnerable, it is the ability of an
enemy to get into and distort all parts of the cyber system.
The regional US bases, which are airfields, harbours, armouries (missile
storage facilities), fuel storage and other facilities, being geographically
fixed and poorly hardened and defended, are particularly vulnerable. Most
vulnerable are the ships in harbour and the aircraft on the airfields. It is
much easier to destroy these assets when they are in their bases rather than
at sea or in the air, if of course you can reach them with something accurate
enough and powerful enough. Navy fleets at sea used to be able to hide in
the vastness of an ocean and manoeuvre to protect themselves. Nowadays
they may have lost some of that protection given the development of longrange missiles whose paths can be continually corrected mid-flight and
which, it is maintained, have a degree of accuracy that can hit and kill a US
aircraft carrier.
So, in summary, the most vulnerable parts of US military power in the
Indo-Pacific are: space-based communications; seabed cables; backup
systems; cyber systems; bases; aircraft and ships in or on their bases; and
naval fleets at sea. Should China be able to destroy or neutralise US power
in the Western Pacific by attacking these vulnerable points, it is fair to say
the US would lose its power and influence, and China would become the
dominant regional power. If the US wanted to make a fight of it, as it did
after Pearl Harbor, then the US would need to re-establish its power
somewhere in the region, use that as its base and then fight back, as it did
with Australia in World War II.
Because of the critical role that allies of the US play in providing bases
for US forces in the Western Pacific, China would need to somehow assure
itself that it could convince these allies not to fight on once the US was
unable to back them militarily. In my view, Japan and South Korea are far
too close to China to fight on by themselves without the US, as well as
having other enemies in the region who would like nothing better than to
see them both reduced by China. It can therefore be assumed that these
countries would at least listen to a Chinese proposition such as ‘Let’s not
fight, let’s trade and all of us Asians be prosperous together now that the US
is gone.’
Taiwan is even closer to the source of Chinese military power than
Japan or South Korea, and without US backing is hardly defendable at all,
so any proposition from China would at that stage be attractive to any
Taiwanese government. ‘Return to the fold,’ China might say, ‘and look
forward to a prosperous future as a prosperous autonomous region within
the PRC. If you do not return, we will destroy you and reduce anyone who
is left to poverty, and then occupy you.’
Australia is further away from China than the other allies, but its lack of
military power in the absence of the US makes it just as vulnerable. If the
US decided to fight back against China following the kind of defeat in the
Western Pacific I have described, Australia, along with Hawaii and Alaska,
would be ideal bases from which to mount a counter-offensive. The critical
questions would be how long it would take for the US to build up its
strength in the Pacific again and could Australia last as an independent
country for that period with China rampant in the region?
So, if China decided to go to war, it would have to do a number of
things simultaneously and completely.
Launch offensive operations in space (possibly assisted by Russia or
others) to ensure maximum destruction of US and allied
reconnaissance and communications satellites (both military and
civilian), either by jamming them or by destroying them by denying
entire orbits to US satellites, smashing them to bits or burning their
electronic parts with space- and ground-based lasers. These operations
would probably have to be repeated as soon as the US and its allies
managed to relaunch replacement satellites. At the same time, China
would need to protect its own satellites against US and allied attacks.
Launch massive cyber-attacks on the civil and military networks of the
US and key allies around the world, again assisted by its few allies,
closing down industry, logistics and communications. The US and its
allies would try to recover their internet systems and networks in the
days after such an attack, new satellites would be launched – and in
fact some are kept standing by for just this purpose – but they would
be few, and Earth-based systems would be repaired where possible.
China would hope to achieve at least a few days’ worth of massive
disruption before the networks could be even partially re-established.
Launch maximum missile and rocket attacks on US air and naval bases
in Japan, South Korea, Guam, possibly Diego Garcia, and even
command-and-control facilities in countries such as Singapore and
Australia. China would hope to achieve the destruction of all the ships
and planes on those bases and in the harbours and also, by attacking
the key infrastructure described above, prevent the US using those
bases for many weeks to support replacement aircraft and ships.
Strike US fleet units at sea in the first and second island chains with
onshore carrier-killer missiles at longer ranges or with land-based antiship missiles for US vessels closer to shore, as well as missiles
launched from aircraft.
Activate its own air-defence system on the mainland in anticipation of
counterattacks on its bases by any regional enemy forces that might
have survived. If China’s surprise attack was even partially successful,
then very few allied attack aircraft or tankers would remain. This
system would include long-range air-to-air missiles launched against
any surviving US strike or tanker aircraft, any navy aircraft that might
have been launched off aircraft carriers that were able to operate and
were within range, or against aircraft that might have been rapidly
brought into the region.
Attack the US nuclear submarines that are normally stationed in the
vicinity of the Taiwan Strait and openly acknowledged by the US
because the role of these submarines, ironically given our scenario, is
to deter Chinese action.
Order its own submarines to conduct cruise-missile attacks on key
regional targets and sea-mine warfare against key shipping lanes and
regional harbours, including places as distant as Australia, to slow
down the ability of the US and its allies to redeploy forces. China
would also have the option of blocking transit routes such as the Suez
Canal using civilian ships, which would hamper the US’s European
allies or European-based US forces from redeploying to the Pacific,
and take weeks to be cleared.
Once the success of China’s strikes in space, in cyberspace and on US bases
in Japan, South Korea and Guam was confirmed, and US submarine
strength in the Taiwan area was eliminated or reduced so that it posed no
more than a reasonable risk to Chinese ships, China would probably move
small groups of forces by air to the edge of the first island chain and occupy
key points with land-based anti-ship and anti-air missiles and early-warning
forces, to preempt the seizure of similar points by US forces and prevent the
ingress of a US fleet to within the first island chain, from where effective
air and missile attacks could be mounted. A major advantage of this move
for China would be that any ensuing naval battle – if there were one, and it
would appear in this scenario to be unlikely – would start several thousand
kilometres off China’s coast.
As soon as it was relatively safe to do so, China would move its surface
fleets from their secure bases on the mainland to within the first island
chain, essentially into the East and South China seas. From there, China
would use every opportunity to attack any US ships that approached from
within the second island chain (of which there should not be many left),
using missiles from these ships; shore-based missiles now located on the
key points within the first island chain; submarines; long-range aircraft
firing cruise missiles; and long-range rockets and missiles launched from
the mainland.
With its defensive perimeter consolidated in the first island chain and
giving it control of adjacent seas, and the US submarine threat reduced,
China would be in a position to begin a full air and sea invasion of Taiwan
if it still needed to, and, if it did invade, China would be able to limit the
casualties and destruction that might previously have been involved. In its
new position of strength, it might then move to offer other countries in the
region a prosperous and secure future, protected by China’s military and its
economic strength – as long as they did not resist.
CHAPTER 11
CAN THE US WIN?
China has spent decades preparing for a war in its backyard, and at present
it is likely to win – at least in the short to medium term. This is admitted by
the US military and informed commentators, but the logical consequences
have not yet entered Australia’s consciousness.
Unless we have an understanding of the nature of a likely regional war –
that is, timings, speed and activities – how is it possible to prepare? Would
Australian warships be able to clear mines from Sydney Harbour before the
most critical phase of such a war was over? Would NATO be able to make
ready and send forces halfway across the world before everything was
over? Would the US be able to send more aircraft or ships from distant
bases in Hawaii and Alaska, and would it still have the regional bases to
send them to? Realistically, if China succeeded with a surprise attack like
the scenario I have outlined in this book, all current operational concepts
that exist in the Australian military, and in the minds of our national
leadership, would be useless.
The scenario I have put forward will be disputed by those who find it
intimidating because of set beliefs, based on ignorance and well-intentioned
hope, in the limited nature of a war between the US and China over Taiwan.
These people desperately hope that China will act stupidly.
It may be that China would try to incorporate Taiwan by force in a
militarily unsound way, by leaving US forces in place in regional countries
and on the high seas, but China has put an enormous amount of its national
treasure and decades of effort into building a rocket and missile force
designed to do exactly what my scenario envisages. Why would they not
use it? China could launch its amphibious fleet across the Taiwan Strait into
the teeth of missile attacks not just from Taiwanese forces, but also from US
submarines, while fending off US air and surface fleet attacks. But where is
the logic in that?
China might very well decide to signal its intent to use military force by
aggressive rhetoric or by assembling its forces around mainland bases that
can be surveilled by satellites, just as the Russian forces were continually
surveilled as they assembled around the Ukraine border for weeks before
the invasion, thus allowing Ukraine and its supporters to prepare military
force and sanctions. If China assembled conventional military invasion
forces, air forces and amphibious ships along the mainland coast opposite
Taiwan, or as close as it could get to Japan and South Korea, and so allowed
the US and its allies to concentrate their forces to counter an attack, or
temporarily move their vulnerable forces in vulnerable bases out of harm’s
way, then China would deserve to suffer the best that the US could throw at
it.
Some argue that China might very well decide that it wants to avoid the
level of US enmity that would result from an attack on US bases in the
region and the casualties that would occur, and instead would be more
likely to conduct a conventional sea-borne D-Day-style attack. But how
much less would the US hate China for such an attack on Taiwan compared
to the scenario I propose in this book? And even if the US’s hate for China
in response to a sudden attack on US bases and the removal of US power
from the Western Pacific was far greater, sadly, the US could still do
nothing militarily about it because the US and its allies would no longer
have forces intact in the region.
China may act stupidly when it actually does not have to do so, when
there is an alternative that delivers it both Taiwan and regional dominance.
Then again, it may not.
***
Should my suggested scenario occur, or something very like it, it is hard to
see that the UK at least would not support the US, at least diplomatically.
But the question is whether European countries would be prepared to send
armed forces halfway around the world to fight against a China empowered
by its successful sudden attack, beside a weakened and humiliated America
which has lost its bases in the region, some ships, and many of its aircraft,
and left regional allies questioning how the war might end.1
As I have discussed, why would China give the world time to assemble
US, and particularly NATO forces, to confront it? The whole purpose of a
surprise attack would be to deny the target of that attack time to prepare. As
suggested above, if China acted in a logical manner from a military point of
view, it would have consolidated its hold on the first and second island
chains long before European ships had passed through the Suez Canal. And
why would they leave the Suez Canal open to shipping, knowing how easy
it is to close the canal, particularly after seeing how, in March 2021 a
container ship accidentally jammed itself across the canal and blocked the
main channel of the canal for three months?
The military involvement of European or even regional allies like
Australia would require a build-up of tension that prompted the early
deployment of allied sea and air fleets from various parts of the world. If
China is half smart, which it is, it will avoid putting the world on notice.
Even Australia, regardless of our intent, might miss the war, given the
current unpreparedness of our military, and the distance from Australia to
likely combat zones.
***
There are probably good reasons why word has leaked from classified US
wargames addressing the Taiwan scenario. The US military wants at least a
chance to win, and undoubtedly thinks they are not getting it from
Congress.
These wargames are very serious activities, which the Americans run
very well. They are not scripted as some training wargames might be in
order to achieve certain training aims but are played against a ‘live’ enemy,
called a Red Team, who use innovative but real-world tactics and are
unpredictable, just as war is and just as our enemies are likely to be.
Apparently, the way the US has tried to fight these in these wargames
has not given commanders any confidence that the US could prevail against
China. In crude terms, on each occasion – so the leaks go – the US ‘has
been handed its arse’. We have not been informed of the detail of these
games. We do not know exactly what scenario the wargames were based on.
Regardless, the US military does not believe it can win by fighting a war in
China’s backyard with the forces it currently has.
These wargame losses convinced the Vice Chairman of the US Joint
Chiefs of Staff, General John Hyten, in mid-2021, to abandon the
warfighting concepts that had guided US military operations for decades.
He reported: ‘Without overstating the issue, it failed miserably. An
aggressive Red Team that had been studying the United States for the last
20 years just ran rings around us. They knew exactly what we were going to
do before we did it.’2 Among the game’s lessons: first, the US doctrine of
massing forces is a recipe for 21st-century defeat; second, the information
networks that define modern American warfare will disappear almost
instantly once combat begins. Massing forces might, in a Taiwan scenario,
refer to the concentration and dependence of warfighting forces on and
around a few US bases in the region, which if they were destroyed would
represent a massive loss for the US. The information networks that Hyten
refers to are heavily dependent on space and on cyberspace, which in turn
are vulnerable to a sudden Chinese attack. This is the essence of the correct
war that we should be preparing for.
In response, since October 2021, the Joint Chiefs of Staff have been
shifting the US military to a new concept of warfighting operations they
call ‘Expanded Maneuver’. Unfortunately the concept has been described
using the worst kind of military jargon, making it hard to understand what it
consists of. Essentially, it seems to involve four ideas: keeping war fighters
supplied with fuel and other items no matter where they are nor how
dispersed they might be (referred to as Contested Logistics); being able to
bring the fires of widely dispersed units, whether part of the army, navy or
the air force, onto a single target (Joint Fires); connecting all devices to find
targets (referred to as sensors) through superior communication, to enable
better decisions to be made faster (Joint All-Domain Command and
Control); and sharing real-time information and data across the entire force,
including allies (Information Advantage).
The US wants its military to be ready to fight under the new operating
concept by 2030, using many of today’s weapons, aircraft and ships, but
pulled together in a new way that avoids the pitfalls revealed in the
wargames. But can it do it in China’s backyard, and does it have until 2030
to fix its problems?
It is only prudent to consider in any planning scenario that China will
probably not play the game in accordance with our preconceived ideas. It
will probably not give warning of an attack by raising tensions and so
allowing forces to be assembled against it from across the world. It will
probably not visibly assemble any invasion forces that can be detected by
foreign surveillance and provide a reason for us to act against it. Yet,
current military thinking, particularly in Australia, assumes that there will
be a rise in tension, China’s forces will assemble somewhere on the
mainland, the defences of Taiwan will be worn down to a certain extent, airdefence zones will be established around Taiwan and around China’s
mainland bases to protect them from US attacks, the invasion of Taiwan
will occur by air and sea, the US will intervene from its bases and from
aircraft carriers, and someone will win and someone will lose or there will
be a stalemate, and all of this without a nuclear exchange.
This is too convenient and comfortable ever to happen.
Even before the wake-up call of recent defeat in the wargames occurred,
and the search for a new US fighting concept was endorsed at the highest
US command level, the US had considered a range of tactical concepts that
might work to assist in the defence of Taiwan. Some of these have been
touched on earlier, but I will revisit them here, so that we can then focus on
preparing for the right war.
The US has long considered using so-called tripwire forces located in
Taiwan, which would report if an attack were imminent and call in
reinforcements. But while several decades ago, the US might have been
able to move hundreds of thousands of troops unhindered across the
Atlantic to Europe or across the Pacific to Korea, today China’s dominant
capability in certain classes of rockets and missiles, as well as the reach of
its rapidly developing submarine and air forces, makes reinforcing Taiwan
in the leadup to a war, or once a war has started, an activity of great risk,
and perhaps even impossible.
The deterrent capability of US submarines in the vicinity of the Taiwan
Strait used to be considered the classic asymmetric defence. This is because
for just one or two boats, a small commitment, each of which can carry up
to 50 land or ship attack missiles and can be located somewhere within
about 1000 kilometres of the strait, the US could deter a Chinese invasion
fleet. But, as I shall discuss in more detail later, the value of these
submarines is likely to become limited by China’s steadily improving
antisubmarine capability.
The US has also considered relying on US or allied land forces
stationed permanently on Taiwan, as part of Taiwan’s permanent defence
structure. But there is no desire now to take over the defence of Taiwan
from the Taiwanese, especially after the US experience in Iraq and
Afghanistan and with Ukraine fresh in everyone’s minds. So other solutions
need to be found.
China’s anti-access, area-denial strategy (called A2/AD) – its strategy to
prevent US forces from penetrating the first island chain – would
predominantly use its rocket, missile and submarine forces, stationed in the
first island chain, to target the major surface elements of the US fleet: its
aircraft and helicopter carriers and its Aegis destroyers.3 The US
understands that if the US fleet in the first and second island chains loses
major ships or submarines, and US regional bases are attacked at the same
time, the US will have little choice but to leave the Western Pacific and
relocate any forces that are left to Hawaii, Alaska or as far south as
Australia.
To target so accurately, China would depend on its space-based
surveillance satellites and its over-the-horizon radars (more on them later),
plus the ability to correct the course of certain missiles during their flight.
This applies especially to the DF-21D, the nuclear or conventionally armed
missile referred to as the ‘carrier killer’, which has a range of over 2000
kilometres and an accuracy of 20 metres,4 and to another longer-range
missile, the DF-26, referred to as the ‘Guam killer’. Bases closer to China
such as those in the Japanese Ryukyu Islands and the Philippines would be
particularly exposed.
Given the formidable power of China’s rocket and missile force, the US
military and Congress are right to be concerned about the vulnerability of
their bases in the Western Pacific. But, according to a report on Defense
News, the US has not yet advanced past rather pathetic ‘interim solutions’:
The US Army will send to Guam one of the two Iron Dome air-andmissile defense batteries it recently purchased as an interim solution
for cruise-missile defense, according to an Oct. 7 statement from the
94th Army Air and Missile Defense Command.
The deployment, dubbed Operation Iron Island, will test the
capabilities of the system and further train and refine the
deployment capabilities of air defenders, the statement notes. It will
also fulfill the requirement in the fiscal 2019 National Defense
Authorization Act that an Iron Dome battery be deployed to an
operational theater by the end of 2021.5
The US is trying to be innovative at the tactical and operational levels,
adopting tactics such as loading F-15E fighter jets up with cruise missiles to
see if they can be used like bombers. But every single innovation depends
on secure regional bases from which such power can be projected. US
regional bases are the Achilles heel in a regional war, and Australian forces
would be just as vulnerable as those of the US if we elected to station our
forces in the same places.
In any coming conflict, air power will be critical, therefore bases will be
critical.6 One thing that the US (and Australia) is very good at is the
coordination of large numbers of aircraft of different types, called packages,
to strike different targets in a coordinated ‘air war’. This is something that
the US and its allies have developed over many decades in their Middle
East wars, and which Australia has experienced and learned from. The two
allies operate common software and communications systems and share
common ways of doing things, referred to as doctrine.
If the US were to encounter a situation in which it was appropriate or
possible to attack Chinese sea and air bases on the mainland, and US
aircraft and bases had not been previously destroyed by a sudden Chinese
attack, the US Air Force would need to attack all or most of China’s
hardened coastal air and naval bases not just once, but perhaps several
times. Because the very few regional US air bases that exist are remote
compared with the many Chinese coastal air bases, a serious US strike
would require a large number of refuelling tanker aircraft waiting several
hundred kilometres off the Chinese coast to refuel strike packages as they
went in and came out.
As discussed earlier, these ‘tanker lines’ would be a prime target for
China’s air-launched long-range missiles. If US launching bases had not yet
been attacked, it might be possible for fighter aircraft to get into China’s air
bases on the first occasion, perhaps using stealth, but they would be
unlikely to make it back to their bases if the tanker aircraft were destroyed.
It would then be even harder for them to hit Chinese targets a second time.
As well as requiring tanker support within a few hundred kilometres of
their operational areas, many US jets in the Pacific need a range of other
support aircraft to operate at their maximum effectiveness. They need
commandand-control aircraft, electronic warfare and intelligence-gathering
aircraft, and heavy-lift aircraft to support the logistic train. Take away the
command-and-control and electronic-intelligence aircraft, all of which
depend on very sophisticated US bases and large amounts of fuel to travel
across the vast Pacific, and things get very difficult. Take away the tanker
aircraft to refuel the strike packages and the mission is over. Take away the
air bases through Chinese rocket and missile attack and the mission never
starts.
The question continually returns to the security of bases. There have
been calls for almost a decade for much more money to be spent on
hardening US bases to make them less vulnerable. Some work has been
done but nothing compared with the efforts China has put in. Its military is
based in up to 40 hardened air and naval bases in areas on its coast that
have good access to Taiwan, and many of its 3210 aircraft can be hidden
and protected in underground tunnels. This disparity is compounded by the
warning and strike power of the artificial islands in the South China Sea
that China has created illegally.
Particular concern has been publicly expressed about the vulnerability
of the giant Andersen Air Force Base on Guam, and the need for the US
Congress to fund an air-defence system known as Aegis Ashore, adapted
from the Aegis Ballistic Missile Defense System used by the US (and
Australian) Navy.7 Back in 2018, when one particular unclassified study
was done, China had hundreds of cruise missiles and 1400 ballistic
missiles; now it has at least 2000 ballistic missiles. Most have a range of
less than 1000 kilometres, but China also has more intermediate-range
missiles (1000 to 3000 kilometres) that can reach Japanese bases with an
accuracy of 5 to 10 metres, and a lesser number of long-range missiles. The
Chinese are more than capable of causing enough destruction to close US
air or naval bases – for example, Kadena on Okinawa, home to 20,000 US
personnel – for an indeterminate period.
***
In 2018, global thinktank the RAND Corporation assessed that China has
the advantage over the US in its ability to penetrate and attack US air bases.
In the case of valuable fighter aircraft such as the F-22 and F-35, and their
key support aircraft, it is not about technology as much as it is simply about
the numbers.
So how does US air strength stack up against China’s? Accurate
numbers of China’s aircraft are harder to get hold of than those of US
aircraft. In the US F-22 fighter class, China operates about 100 J-20
fighters, with some analysts saying the numbers may be closer to 200. This
compares with 125 US F-22 aircraft in fighting squadrons, of which 60%
are available at any one time for combat – that is, 75 aircraft.
China also has at least 300 different Soviet ‘Flanker’type aircraft of
various vintages and variants, roughly equivalent to a US F-15. The F-15 is
an older (first entered service in 1972), rugged, twin-engine, air-superiority
fighter with over 100 air victories and no losses which has been developed
most successfully as a strike aircraft capable of attacking ground targets.
The US operates around 400 of several variants of F-15s, including 165 of
the latest Strike Eagle version, the F-15E.
China has at least 300 different Soviet ‘Flanker’-type aircraft of
various vintages and variants, roughly equivalent to a US F-15. (Getty)
China’s J-10 series aircraft is the equivalent of the US F-16 class, which
is a multi-role aircraft (meaning it is able to fight other aircraft as well as
carry out ground-attack missions), smaller than the F-15, with one engine
and relatively short range. About 4600 have been built since 1976 and the
plane is in service with 25 countries. The US Air Force operates 1245 F16s, of which 700 would be immediately available for combat across the
world, with 544 in the Air National Guard and the Reserve, consisting of
part time personnel. China intends to build around 1000 J-10 aircraft.
Evidently, China’s air force is comparable in numbers of quality fighters
in the given tiers explained. With regard to a conflict in the Western Pacific,
it would also have the advantage of a concentrated air fleet, whereas US
fleets are spread paper-thin across the US and on overseas deployments, and
its F-15 and F-16 fleets are mostly around 30 years of age.
The US F-22 fighter aircraft is a key part of the US warfighting
equation, as probably the best aircraft of its type in the world. It is a
valuable warfighting tool, but it is poorly suited to very long range
operations, having been designed to operate from European bases that have
far more support assets and not as much need for refuelling tankers. The US
Air Force has been starved of funds for F-22 maintenance, upgrades and
spare part stocks over the last decade or so, so combat availability is
perhaps 60% of the 186 F-22 aircraft in existence. This would not be an
issue if fleet numbers were not already so low. The overall number of F-22s
is too small to allow for the concentration of force required to deal with the
most sophisticated Chinese aircraft. No more F-22s can afford to be lost –
much less lost on the ground, on bases that are inadequately hardened.
Of the 186 F-22 jets in existence, only about 125 of them are assigned
to combat squadrons, with the rest being set aside for various training and
test and evaluation duties. Many of the jets in the latter categories have not
been upgraded to the most recent standard. So, there are very few of these
top-line aircraft compared to the number of aircraft China operates, their
maintenance system is deficient in the region, they are not long-range
fighters and need tanker support, and when they are on the ground in
undefended bases they are particularly vulnerable. This is not a situation
designed to create confidence in warfighting leaders or in allies.
The more modern F-35 fighter aircraft, still being introduced into
service, is touted as the most lethal, survivable and connected fighter
aircraft in the world. The term ‘connected’ means the F-35 is connected to
other sources of data indicating targets and threats and is thereby part of a
much larger system rather than just a single aircraft. It is also a multi-role
aircraft, meaning it can fight against other aircraft (in air-to-air combat, as
the saying goes), carry out surveillance and attack ground targets. It has
extensive stealth characteristics too, allowing it to minimise detection by
enemy radar.
The US F-35 fighter is a very impressive aircraft, but it currently lacks
the range, performance and stealth capability that would be required
for a war in the Western Pacific. (Getty)
Only 720 of these aircraft had been built and brought into full service as
of March 2022, with a plan to produce 175 per year. At present, they are
spread over 13 countries, not all of whom would join the US in a war
against China. The majority are held by the US, with the air force planning
to ultimately have 1763 of them, the US Marines 420 and the US Navy 273.
This compares to the very large numbers of Chinese aircraft with only one
operator – the PLA air force – which operates them from hardened bases
backed up by an apparently functioning maintenance system under, one
assumes, a single plan as to how they should be used.
The F-35 comes in three variants. The F-35A is a conventional take-off
and landing aircraft using long concrete runways. (The Royal Australian
Air Force currently has 44 of these aircraft, and aims to have 72 operational
by 2023 and ultimately 100 in total.) The F-35B model is a short take-off
and vertical landing aircraft suitable for restricted airfields, smaller aircraft
carriers and helicopter carriers, and it can also be deployed away from main
bases to make it less vulnerable. (It is operated by, among others, the US
Marine Corps and the Royal Navy.) The F-35C is designed to operate from
larger aircraft carriers, such as the supercarriers operated by the US Navy.
Although a very impressive aircraft, at this stage the F-35 lacks the
range, performance and stealth capability to be as effective as may be
necessary. Like all such US aircraft designed for the European theatre, it is
unsuited to combat involving the vast distances of the Pacific because of its
short range, though there is talk of longer-range variants to come. In
addition, the fleet availability of the F-35 is below 50%, and at one stage
there were no fewer than 21 different configurations, each with
substantially unique hardware and software. Throughout the US F-35 fleet,
spare parts often have to be manufactured to order, which can take many
months. For the kind of combat operation that we may see in the Pacific
against China, the problem is compounded by the need for long concrete
runways, at least for the F-35A; deploying them out of forward-operating
bases, such as highways in remote areas, is extremely difficult.
The Next Generation Air Dominance (NGAD) fighter, which will
eventually succeed the F-22, is still in the conceptual stage and too far away
to be counted, which is why commentators speak of the ‘fighter capability
gap’ between the US and its likely adversaries, primarily China, in the next
decade or so. This is a real gap but it becomes irrelevant if the war occurs in
the next three to five years and all the F-22s deployed to the region are
smoking ruins on destroyed air bases.
The US bomber fleet does not need to be in bases as close to where the
action is likely to be as the fighter fleet does, because the bombers have
much longer range and can attack targets in China from, for example, Diego
Garcia in the Indian Ocean, or from bases in Australia. The US bombers,
however, suffer from similar issues to the fighters, mainly lack of numbers
and older fleets. In the type of war we are considering here, the bomber
fleet is much more likely to use longer-range, standoff missiles, such as
cruise missiles, than the traditional kinds of bombs that were flown over
and dropped directly onto a target during World War II, Vietnam and even
Iraq. This allows them to launch their missiles far from the target and
outside the range of most air defences. Like other US air fleets, though, the
bomber fleet is caught between older aircraft and new developments not yet
available.
The US Air Force currently has 158 bombers in total, consisting of 62
B-1 Lancer bombers, 20 B-2 Spirit bombers, and 76 older B-52H
Stratofortress bombers. It is the intention that the new B-21 will replace the
82 B-1 and B-2 bombers, while the B-52H will fly on through to at least the
2040s, if not into the 2050s, which would mean it will have seen 100 years
of service by the time it is retired!
The B-2 Spirit stealth bomber is one of the mainstays of the US air
attack force, along with B-1 Lancer bombers and older B-52H
Stratofortress bombers. (Getty)
The US Air Force Assistant Chief of Staff for Plans and Programs, in
written testimony to the House Armed Services Committee in 2021,8 called
for a fleet with only two bomber types within the 225 heavy bombers the
US currently wants or thinks it can afford – still a substantial number.
Subtracting the 76 older B-52 Stratofortress bombers, which are still
effective but are to be used only as ‘missile carriers’, that leaves room
within the desired fleet size of 225 for 149 B-21s. Apparently, because
much of this is still secret, just five prototypes exist at the moment, and the
first has not yet flown. The figure of 149 new B-21 bombers is nearly a
50% increase in demand for the B-21s by the US Air Force, up from its
previous plans to buy only about 100 of the planes. By the 2050s, when the
entire fleet of B-21s is likely to be operating, the total cost will be in the
vicinity of $203 billion – not cheap and, as production occurs, they are
likely to cost more than anticipated.
If that happens there will be a danger of ‘sticker shock’ – a negative
response on the part of Congress and the administration in response to the
final price, the sticker price, of each aircraft. This is always a risk with
expensive military aircraft designs, except in countries like China where
there is no public scrutiny. The usual result of sticker shock in the US is that
the number of aircraft ordered is then cut back by the government as
delivery approaches, with little consideration given to what might be
needed based on the threat.
This is what happened with the B-2 during its public rollout in the late
1980s. Development of the aircraft began in 1981 and the procurement of
132 aircraft was approved in 1987. Yet it was only in April 1989 that the air
force disclosed in public testimony that it had spent $22.4 billion on B-2
development. In early 1990, citing cost concerns and the end of the Cold
War, Defense Secretary Dick Cheney cut the numbers of B-2s on order
from 132 to 75. Later he trimmed that figure to just 20; Congress added one
more later, resulting in the 21 planes eventually built. If we thought we
needed 132 B-2s when the threat was Russia during the Cold War, I wonder
how many the US might assess it needs in an honest appreciation of today’s
threat from China, even without Russia being in the equation.
In peacetime, it is all about money. In wartime, and for the kind of war
that China might fight against the US, it is all about the numbers. The total
number of US aircraft does not come near the number of Chinese aircraft.
And of course, China can concentrate all its aircraft in one area, the Western
Pacific, but the US has worldwide responsibilities. Unfortunately for
regional security, the situation regarding US aircraft numbers is unlikely to
improve in the short to medium term, meaning China holds all the aces.
This is something that should be at the front of Australians’ minds if we are
to contemplate joining the US in such a war.
The ability of China to protect its mainland bases from attack by US
aircraft is also impressive. From the Chinese mainland, the effective,
modern S-300 and S-400 surfaceto-air missile systems China has bought
from Russia, as well as China’s domestic air-defence systems, have the
range to reach out from China’s mainland up to 400 kilometres, which
means that these missile systems can shoot down any non-stealthy or less
stealthy aircraft over Taiwan itself, which is only 180 kilometres off China’s
coast. To provide early warning, China has built multiple Over the HorizonBackscatter (OTH-B) radars similar to the Australian Jindalee Operational
Radar Network (JORN). This is a world-leading system in Australia which
covers much of the Pacific to Australia’s north, and there is no reason to
believe that the Chinese version is any less capable. It is said that when
operational these Chinese radars will be able to detect and track even the
very stealthy US B-2 and B-21 bombers.9
***
Fighters and their support aircraft are not the only forms of military power
that either side is likely to deploy in a more traditional Taiwan war. It was
always the US assumption that its navy would play a crucial role.
Comfortingly for Australia, because of our dependence on the US, the
US Navy Secretary’s new strategic guidance document, called One Navy–
Marine Corps Team and released in 2021, makes it clear that the service
now sees China as its number-one enemy, claiming that ‘the long-term
challenge posed by the People’s Republic of China is the most significant
for the Department. The People’s Liberation Army Navy (PLAN) has
radically expanded both its size and capabilities, growing to become the
world’s largest fleet.’10 The document also promotes the US Navy
Secretary’s ‘Four Cs’ concept: China, Culture, Climate change and COVID,
an acknowledgment that the US government has many urgent priorities
competing for its restricted funds.
US maritime power across the world is based on its eleven carrier strike
groups, its nine amphibious-ready groups and its nuclear-powered
submarine force.
Aircraft carriers like the USS Ronald Reagan are the bedrock of the US
Navy, allowing it to reach out and strike distant targets, including in
the Western Pacific. (Getty)
The basis of the US Navy’s ability to reach out and strike targets, the
US carrier strike group consists of an aircraft carrier (often referred to as a
‘supercarrier’ because of its size compared to previous US aircraft carriers),
with its ‘air wing’ of roughly 70 aircraft on each carrier; a number of
cruisers, destroyers or frigates, all missile-armed; several supply ships; and
often a nuclear submarine. There are perhaps 7500 crew in total across all
ships in a carrier strike group.
An amphibious-ready group consists of navy ships that can carry US
Marine Corps troops, equipment and aircraft to where they can conduct an
amphibious assault. Aircraft in the group include the F-35B fighter; the
MV-22 tiltrotor aircraft, capable of vertical take-off and landing and
cruising at faster speeds like normal aircraft; and conventional attack and
medium- and heavy-lift helicopters. An amphibious-ready group might be
supported by a carrier strike group, as well as land-based aircraft if they are
within range. These groups are ‘task-organised’, which means that they are
assembled as needed to carry out any specific task. An amphibious-ready
group normally carries about 5000 crew and troops.
To understand the limits of US naval power, it is important to realise
that on any one peacetime day, only three of the eleven US carrier strike
groups might be deployed: one each in the Pacific, the Atlantic and the
Persian Gulf (although another one is currently being deployed in the
Eastern Mediterranean because of the Ukraine War). Of the nine US
amphibious-ready groups, four might be deployed: perhaps two in the
Pacific, and one each in the Atlantic and the Gulf. The other carrier and
amphibious groups are likely to be in rest, training or in maintenance at
naval bases in Japan or on the US east and west coasts.
Massing forces takes time. During the Gulf War, it took six months to
assemble overwhelming US and allied military power against a second-rate
dictator. In a Taiwan war, China is not going to allow the US months.
During the Cold War era, the US Navy had confidence that, in a fleeton-fleet engagement, its ships and carrierbased aircraft would be able to
unleash missiles at a range beyond that of the Russian and Chinese navy’s
missiles. However, the Chinese have now caught up across the board, with
near state-of-the-art systems on board many Chinese vessels that outrange
the anti-ship weapons on US aircraft and ships.11
The US SM-6 missile – recently approved by Congress for sale to
Australia – has a range of up to 370 kilometres from its launch ship and
may work to remove some of this disadvantage for the US Navy if it can
survive attacks by bigger ballistic missiles fired from the Chinese mainland.
Yet the new Chinese anti-ship cruise missile YJ-18, with a range of up to
530 kilometres, would allow Chinese shooters to fire first because of the
longer range: a significant advantage in modern maritime warfare, and one
that for decades the US Navy thought it owned.
Fleet-on-fleet battles, in which a fleet of Chinese warships meets a fleet
of US warships and they fight it out with the weapons they each have on
board, may not take place early in such a war, or, if China strikes first with
its mainland missiles and sinks a significant number of US ships, they may
not happen at all. If fleet-on-fleet engagement does occur, considering the
ability of China’s missiles and rockets (with a range of 1500 to 3500
kilometres from the Chinese coast) to reach even the second island chain,
China is likely to dominate.12 As China specialists Toshi Yoshihara and
James R. Holmes write: ‘Both the defensive and offensive sides of sea
combat are stacking up in China’s favour and progressively eroding or
nullifying altogether some of the US Navy’s tactical advantage.’13
The survivability of a US fleet, even in the second island chain, to say
nothing of the first island chain, will be of critical importance to Australia.
As mentioned, given the ties that have recently been refreshed through the
AUKUS agreement, Australia is likely to be asked to send ships and aircraft
to join an allied fleet. To do this, we must have some confidence that such a
fleet can survive.
It would also be expected that the US submarine force would play a key
role in any war with China, and the US seems to be preparing for that
eventuality. The specific locations of the US Navy’s submarines are
unknown, but there was recently an exercise in the Pacific in which it was
reported that 25 US submarines were active.14 This was apparently an
exercise in preparation for surging (sending out far more than the usual
number) submarines into the Pacific if war with China eventuated.
It must be assumed, as previously discussed, that the US currently has
nuclear-powered but not nuclear-armed submarines on patrol within the first
island chain in the vicinity of Taiwan to act as a deterrent to a seaborne
invasion. One class of nuclear subs is called the Seawolf class. There are
only four of these submarines, as they proved too expensive to make, so a
cheaper submarine, the Virginia class, was designed and is already in
service. The Seawolf-class submarines carry at least 50 Tomahawk missiles,
whose range can be in excess of 1000 kilometres. This class is faster, deeper
diving and better armed than its Los Angeles–class predecessors and
Virginia-class successors. To be effective, such submarines do not have to
be actually in the Taiwan Strait, as their missiles have a very long reach. In
July 2021, the US Navy deployed three of these exceptional submarines
(USS Seawolf, Connecticut and Jimmy Carter) to the Pacific Ocean at the
same time: an exceedingly rare occurrence.15 It can only be assumed that
this deployment was also a practice run for a surge of such submarines into
the Pacific in response to a Taiwan war. But to surge anything, warning is
required and why would China give warning?
The US Navy operates 69 nuclear-powered submarines. Fourteen carry
nuclear-tipped ballistic missiles and are part of the US strategic nuclear
force that deploys for very long patrols (months on end) and attempts to
hide until they are needed to fire their missiles. Another four submarines of
the 69 total are the Seawolf class, which are referred to as cruise missile
submarines because that is their main weapon type. These cruise missiles
were predominantly for stationary land targets but now have capability
against ships at sea. This leaves 51 nuclear-powered attack submarines
equipped with a variety of weapons that could be used against a Chinese
fleet.
The US Navy may have 69 submarines at present but because US
shipyards are not producing Virginia-class submarines as fast as other
submarines are being retired, the navy’s nuclear attack submarine fleet is
smaller than is required now and is expected to be not greater than 50
submarines by 2026, despite the fact that the US Navy states that it needs
66 attack submarines. This may not be a problem the US Navy can
overcome, because they are also short of ballistic missile submarines (the
Columbia class) and are trying very hard to build more of these as well over
the next 15 years in limited dockyards.
Under the ‘Pivot to Asia’ policy pursued by recent US administrations,
60% of US submarines are supposedly available to patrol the Pacific Ocean
at any one time, which means roughly 31 attack submarines. The US can
surge its submarine forces during a crisis, but normally one-third is on
patrol, one-third is in training, and one-third is in maintenance. That
deployment model is very common around the world and explains why the
US Navy has just under 300 front-line ships but deploys only 100 at a
time.16 It means that the immediate availability of attack submarines in the
Pacific might be as low as 10 or 11, with another 10 or 11 available after a
surge period. As Yoshihara and Holmes assert, ‘Twenty-two SSNs [nuclearpowered submarines], no matter how good they are individually, constitute
a slender force to cover the vast China seas and the western Pacific in
wartime.’17 And unless fleet commanders can prepare for a surge – as the
nation that initiates a surprise war can – increasing the number of
submarines on patrol takes time.
So, as with aircraft, the US military faces a problem related to numbers.
They do not have enough ships or submarines to match China, and they no
longer have a technological edge over the Chinese navy. According to an
article in Defense One:
The declining size of America’s attack submarine fleet is
particularly problematic given that the [Chinese] People’s
Liberation Army Navy fields more than 60 attack submarines.
Admittedly, most of those submarines are diesel submarines. But
China has at least seven SSNs and is working hard to increase both
quality and quantity. To make matters worse, the United States must
deploy its SSNs (attack submarines) around the world, whereas
Beijing focuses almost all of its attack submarine deployments in the
Indo-Pacific. That provides Beijing with a numerical advantage in
attack submarines in locations where US–China conflict is most
likely to occur.18
RAND made the judgment in 2018 that US submarines pose a serious threat
to China’s navy and could possibly sink up to 40% of a Taiwan invasion
fleet over a seven-day period in an engagement in the Taiwan Strait, but the
report also observed that while this threat is still serious, it is decreasing.19
China can now place devices on the sea floor to detect the presence of even
the world’s most sophisticated submarines, and its ability to destroy them is
also growing. It is suspected that Russia has provided advanced nuclear
depth-bomb technology to China, or China may have independently
developed these bombs, as both the US and the UK have in the past.
Regardless, it must be assumed that China has such weapons.
***
Can the US Navy increase its size and capability in the limited time that
might be available to it? There is deep concern being expressed that the US
Navy is not just too small, but also that it may not have a shipbuilding plan
that would increase its size in place until 2023.
The US Navy is desperate not just to maintain its technological edge but
also to regain the numbers it considers it needs to deter China. At the same
time, it is examining every different idea that it has about how to actually
fight with the fleet it has and be able to win. To illustrate how seriously the
US is taking the need to integrate technology, fleet size and operational
concept (how to fight), the US Chief of Naval Operations says:
The fleet plan will reflect various operational concepts that are
being tested by the USN at present, such as Distributed Maritime
Operations, Expeditionary Advanced Base Operations and Littoral
Operations in a Contested Environment, as well as the results of the
Large-Scale Exercise 21 wargame conducted in August 2021. The
USN has also created a new operational unit TF 59 based in
Bahrain to refine operational concepts.20
Parts of these concepts have been exercised in the Philippines and recently
in Australia, with the US Marines using a rocket system known as
HIMARS, which Australia also hopes to possess sometime in the future. At
this exercise, the Marines, with their HIMARS system, were deployed on
Australian C-17 transport aircraft along with ground-protection elements, in
the way such forces might be deployed into key positions in the first island
chain.21 So the US seems to be developing an operational concept similar
to what is assumed to be China’s new operational concept: to transport
small groups armed with powerful anti-ship missiles very quickly by air
onto islands along the first island chain, with or without the permission of
the countries concerned, to be able to attack ships using the strait. This
would either prevent the breakout of China’s navy from the first island
chain or assist the break-in of the US Navy from the second island chain.
There would be no invasion or occupation of these territories; such
techniques of troop deployment would simply be about locating small
forces on these critical locations to act in concert with maritime or air
forces.
Both China and the US are working on refining or radically changing
how each will fight such a war. What this means is that both sides realise
that the military equation has changed and both are trying to solve the new
problem they face. In that respect, the question this book addresses is: as the
situation changes due to the relative sizes of the forces and their
technological competence, who will come to the right conclusion first?
Of course, these issues of tactics and numbers may already have been
rendered irrelevant by the advantages conferred on the Chinese by the vast
geography of the Pacific Ocean, the size of the PLA, the power of its rocket
and missile force, and the vulnerability of both US bases and major US
naval platforms such as aircraft carriers. None of the figures or tactics being
planned over the long term will matter if, within a few years, China strikes
first and is able to destroy US fleet units and bases with long-range
missiles, block seaways and prevent America and its allies from reinforcing
its Western Pacific units, and convince the likes of Japan, South Korea and
Taiwan that there is no point in fighting on without US support. At that
point, the big question for the US will be, should it respond with tactical
nuclear weapons?
***
If China were successful in a sudden attack, as described in this book, it
would then be essentially in control of the region, with little or no threat
from the US. This is when the real test for Australia would begin, as we
could be on our own for quite some time. Australia therefore has to plan not
only for a sudden attack by China, but also what China might then do to
Australia in the months or even years before the US could re-establish its
power in the region.
A regionally dominant China represents an existential threat to Australia
as a liberal democracy. It should be expected that, at the very least, China
would demand access to Australia’s resources at very favourable prices to
overcome trade embargoes or to obtain cheap materials for its industries,
even if it had also gained access to resources from other sources. China
might even establish a military presence in this country, to guarantee access
to resources and to stop the US using Australia as a military base for
launching a counter-offensive. And a dominant China might just at some
stage demand our agreement to the 14 demands made of us in 2020, and
much more indeed.
The prospect of a war in the Western Pacific is dark enough for the US,
but it is even darker for Australia, with our one-shot defence force, our
enormous vulnerability as a nation if we or the US cannot control the seas
over which we import products such as fuel, pharmaceuticals, fertiliser and
hundreds of other critical items, and over which we export the natural
resources that make us prosperous. The weakness of our once great and
powerful ally will change just about every aspect of life in Australia. In the
very short time that China might take to move from its peacetime routine to
a wartime footing, Australia will not be able to make any substantial
advancements to its military capability or its national resilience. So now is
the time for Australia to act. Or have we already left it too late?
Can anyone imagine that if there was a widespread realisation that such
a war faces us, more money for defence would not be found? But, as
American General George C. Marshall said in 1942: ‘The Army used to
have all the time in the world and no money; now we’ve got all the money
in the world and no time.’
A prudent Australia would have started to prepare 20 years ago.
Accepting now that we have to make the most of things, we must start
preparing for the right war now.
CHAPTER 12
AUSTRALIA, DEFEND YOURSELF!
If I have been asked once by my constituents, journalists or interested
Australians I have been asked 100 times: if I had my way, what is the one
thing that I would do for Australian national security? Just about everyone
adds in their views before I can answer: B-21 bombers, F-22 fighters,
conscription, reform of the Federation, nuclear weapons, a nuclear power
industry . . .
Apart from being able to go back in time and start again, there is no one
solution to the problem of strengthening Australia’s national security, but
there are things that are more important than others. If I am forced by my
interrogator to suggest a first step, my answer is always that it should be a
comprehensive, formal strategy. If national security is truly the most
important function of government, as I believe it is and as all governments
claim it to be, then the government should initiate such a strategy, write the
requirements for it and be closely involved in it.
Then, of course, there are many more steps that need to be taken before
any Australian government can say that the nation is secure. The following,
in priority order and greatly simplified, are the five steps I believe Australia
needs to take to meet the demands of our current strategic environment.
1. Create a comprehensive national security strategy
The government should create a pro-active, medium- to long-term,
comprehensive national security strategy, covering the nation’s resilience as
a whole and its defence, and making every cabinet minister aware of their
national security responsibilities and holding them accountable. The
classified version of the strategy must be brutally honest, but there must be
an unclassified version that does not just engage the Australian people but
also gives them confidence in a secure future. The objective of the strategy
must be to maintain Australian sovereignty by making the nation selfreliant and resilient in extremis, and making the ADF more lethal and more
sustainable and increasing its mass so that it really can, for as long as
possible, deter war. Each of the descriptors I use to describe the nation and
the ADF (self-reliant, resilient, lethal, sustainable and mass) have a very
complex meaning themselves in this context, and must be judged against
the kind of war (the threat) that the nation and the ADF could face over the
next five to ten years, not against meaningless generalities as has been the
case in the past. We must prepare for the right war.
As I have noted here and said often in interviews, the Australian
Coalition government of 2013–22 has been the best government on national
security that I have seen since the end of the Vietnam War. So, as a
backbench senator, I was careful not to be too critical, and made a point of
praising that government for having done so much in the sphere of defence.
At the same time I continued to encourage and occasionally demand that
the government did more to address the current geopolitical and military
situation. And the more begins with a strategy.
Where Australia can do more is by, first, minimising the potential
impact on Australia of the type of attack I have described in this book as a
‘collateral attack’ – that is, one that might be directed at Australia from
within a war between China and the US. If we start to prepare for this
eventuality, even if we cannot ultimately deter such an attack, we will at
least be in a better position to mitigate its effects.
Second, our strategy must prepare Australia to survive a possible
withdrawal of the US from the Western Pacific following a sudden
successful Chinese attack on US bases in the worst case, and the temporary
or permanent domination of the region by China.
The national security strategy that I advocate should give broad
guidance to ministers and may require the preparation by every minister,
not just the traditional national security ministers, of nested national
security strategies, for approval by the National Security Committee of
Cabinet.1 By nested, I mean that a national security strategy is the
overarching strategy, and defence, energy, social and industry strategies,
plus many others, fit neatly into it.
Some of the outcomes of a government strategy process were dealt with
by the then Prime Minister in two recent speeches. The first was on 1 July
2020 in respect of what the Department of Defence called a Strategic
Update, which explained how an extra $270 million provided by the
government to Defence would be spent. The Prime Minister’s defencerelated speech was much broader than just defence, which contributed to
the confusion in Australian minds that national security is something that
Defence does.
The reason that Australia needs such a strategy now when, supposedly,
we have not needed it in the past is because of what Prime Minister
Morrison referred to as ‘hybrid warfare’. Hybrid warfare as a theory of
military strategy has traditionally referred to a blending of various forms of
warfare into a hybrid: political, conventional, irregular and cyber warfare to
give a few examples. Prime Minister Morrison described ‘hybrid warfare’
as conflict that has ‘stripped away the old boundaries that once separated
the realms of defence, foreign policy, trade and investment,
communications and other areas, reaching deep into our domestic society’.2
This is an acceptable definition of what Australian strategy needs to face
and this awareness compares favourably in my view to the ad-hoc way that
Australian governments have approached strategy up until now, essentially
by taking one problem or issue at a time and solving it. While this has
worked in the past, we now need a much more comprehensive approach.
And although the Coalition government dealt successfully with individual
national security problems that arose, Australia still lags seriously behind
the changes in our strategic environment that now demand a comprehensive
approach.
At the risk of losing the attention of every strategy purist but
recognising the reality of politics, it matters less what such a strategy is
called than that the problem of national security is considered as a single
issue, driven by government, covering both defence and the nation in its
entirety. The end product of the process is not the written strategy, but
national security itself. We must be secure, so a strategy must be produced,
reviewed and implemented. Looking at the experiences of other countries
(the US, the UK and India, for example), a national security strategy is
unlikely to be completed successfully in the first review cycle; it may take a
few years and a few attempts (which is why I have been advocating for this
for at least a decade!). But the process itself will produce benefits for
ministers, for bureaucrats and for national leaders.
My observation as a humble backbencher is that Australian ministers do
not like this process. They would rather be left to themselves to work out, in
their own time, what they need to do to achieve what they think is necessary
for national security. This approach might work during the kind of peace
that Australia has enjoyed for the last 75 years, but it will not work as we
move into a less stable period when we need to produce results across all
aspects of the nation and government, and to produce them quickly.
I have also observed that the idea of producing a written national
security strategy is not popular with Coalition leaders, who prefer some
government policy to remain much more informal. I suspect that is because
the only experience most of our leaders have of a national security strategy
is the one produced by the Labor government in 2013. This was not a true
national security strategy, in that it did not cover all government functions,
nor did it allocate responsibilities or have an accountability process. It
achieved little except consume the bureaucracy for a period of time at the
end of a very difficult period of government. The national security strategy
that I advocate does not have to be like the previous Labor Party version.
But, given the crisis-like situation that Australia finds itself in as regards
security in our region, with danger on our doorstep, strong direction is
required, objectives must be set, coordination is needed across government,
and accountability is essential.
2. Continue addressing legacy problems with national security
The government should continue to address legacy national security issues,
such as the need for specific legislation to empower security bodies such as
the Australian Security and Intelligence Organisation (ASIO) and the
Australian Federal Police, steps to counter foreign influence, and
deficiencies in skills enhancement, naval shipbuilding, anti-terrorism
procedures, cyber technology, modern industry, liquid fuels, intelligence
and policing, biosecurity (pandemics), universities, health and, the basis of
national security, the economy. Recently, the former government’s Minister
for Defence and Defence Industry Minister addressed some enduring
problems and produced solid results, but in the area of national security, of
which Defence is only part, it is very hard to achieve quick results. Yet
quick results are what is needed.
I do not advocate that the government should stop projects or
development so that the bureaucracy can conduct a long and detailed
‘white-paper approach’ to national security. This has been the practice for
years – when a white paper–type review is announced, just about
everything in defence stops to await its completion, often for years.
Australia cannot afford that this time. The government should decide what
needs to be done and do it. It must not leave these issues to be addressed by
the bureaucracy.
3. Develop plans to fight the China war
The government, being strategic, at some stage must formulate actual plans
for how a China war can be fought. The first step in doing this, if it is not
already occurring at a classified level, is to insist on a seat at US
warfighting strategy or planning tables – that is, wherever the US is making
its strategic and operational plans, in Washington at the Pentagon or in
Hawaii, the headquarters of Indo-Pacific Command. Only by getting across
an appropriate degree of detail about how a war between the US and China
will be fought can Australia consider what role it might play in that war. It
is a waste of time for Australia to be trying to work out how we will
contribute to US efforts by readying and deploying a group of ships and
aircraft to the Western Pacific if the initial stages of the war are likely to be
over before our ships clear the heads in Sydney Harbour, or our aircraft,
which need US tanker support to deploy from Australia to Japan or Guam,
find that the tankers have already been destroyed. These are the discussions
that should be occurring now, not the day after a sudden attack.
It is never too early to start such planning, given the critical nature of
this war for the US, Australia and our region. Only by looking in detail at
US expectations of its allies, which will not become firm until the military
planning begins, will Australia become aware of how the US assesses the
risks, what the US needs from Australia, and what Australia can provide.
The government must come to terms with the lack of confidence within the
US in relation to this war and assess whether it is winnable before we
commit. Australia will be a junior partner in this war, but that does not
mean that it cannot influence US strategy at the highest level. Not only can
we, but we must have a voice in strategy and in military operations.
There are any number of operational issues related to strategy that must
be clarified with the US before Australia commits military forces. How
does the US see this war going: nuclear or conventional? Will it have a slow
build-up of tension or be a surprise attack? When might it occur? Will
Taiwan and US bases be attacked simultaneously, or just Taiwan initially?
What would a coalition expect of Japan, South Korea and India? Only once
we have agreed with the US on such issues can a feasible military strategy
emerge.
History tells us that leaders of coalitions or alliances sometimes have
quite unreasonable expectations of junior allies. Churchill certainly had an
unreasonable expectation of Australia once the Japanese had attacked Pearl
Harbor and Singapore and were bombing Darwin. He requested that
Australia’s only battle-ready and experienced army divisions, which at that
point were aboard ships on the way from North Africa to defend Australia,
be diverted to Burma, because that was his priority. The Australian
government split the difference, bringing most troops home but leaving
some in Ceylon (present-day Sri Lanka).
4. Prepare for enhanced grey-zone conflict
Australia is handling the current level of China’s grey-zone conflict well,
with measures that lessen our trade dependency, strengthen cyber defence
and attack, stop the theft of intellectual property, and recognise the link
between national security and the economy and the need to defend our
political systems and our institutions. The new national security strategy I
propose is likely to indicate that Australia must prepare for an enhanced
grey-zone conflict that is still short of war. This may require the
government to
prepare for more lethal and unattributable biological attacks, cyberattacks and attacks on satellites
be prepared for further anti-trade measures that might be deployed
against Australia, such as threats that discourage or stop companies
from shipping to and from Australia, or make foreign shipping
prohibitively expensive by dramatically increasing insurance rates
increase Australia’s liquid-energy resilience by increasing our
production of crude oil, our domestic refining capacity, our storage of
crude oil and refined petroleum products, and our ability to move
crude oil by ship, rail or pipeline from source to refinery, which may
involve the purchase of some oil tankers and rail rolling stock, and the
construction of storage facilities
encourage Australian companies that are important for national
security through policies that reward a physical presence in Australia,
rather than judging companies on their competitive advantage and
‘value for money’, as with many current government policies
create and protect systems and processes (not the actual
implementation at this stage) for the mobilisation of the nation’s
resources that would be needed if the grey-zone conflict developed
into physical conflict – in essence get the systems ready in this phase
for what might have to be done to mobilise the nation at a later time
set up an organisation that can, when necessary, direct personnel, skills
and resources across the nation to warlike needs
support regional neighbours and allies against attempts to woo them
away from Australian influence or reduce the effectiveness of
alliances, especially in the South Pacific.
The above measures would also be applicable to wartime and would
significantly increase the self-reliance and resilience of the nation. The
government would be very wise during this period to demand access to our
allies’ contingent strategies and operations, similar to the arrangements
between NATO member states during the Cold War.
5. Fund defence according to need
Australia should never forget that diplomacy and alliances are our first line
of defence. Australia has a fine diplomatic record and has built a network of
alliances and is now working to improve them. We need to ensure, however,
that our hard economic and military power matches our diplomatic clout,
which in turn would make us a much more valuable alliance partner.
Australia’s defence capability should not be defined by our materiel, by
the amount we spend on defence, or by government or bureaucratic spin.
Our deteriorating strategic environment now demands that our national
security be defined by what the nation and the ADF can actually do to deter
or to win a future war with China. The emphasis in defence, and in
government statements about defence, must change from concentrating on
dollars and percentage of GDP spent, and on the purchase of submarines,
ships, planes and tanks, to determining the kind of war that the defence
force, backed by a resilient and self-reliant nation, can resolve in Australia’s
favour. This involves aligning strategy, operational concepts and the
military equipment the nation is buying, but always begins with a definition
of the threat.
Stemming from the analysis in this book, the following, although not an
exhaustive list, might be seen as government priorities in this area:
establishing a ballistic missile defence of key points in Australia
accelerating the development of Australia’s lethal longer-range strike
capability, which should cover at least the Indonesian Archipelago and
the South Pacific, and commence the development, over a sustained
period, of a strike capability that can reach even further
increasing our ability to keep key Australian ports open while under
threat from modern sea mines
ensuring that the essential needs of the nation can be provided
domestically for a realistic period in the absence of commercial sea
transport, which may require setting aside the previous Coalition
government’s attachment to liberal ideologies of ‘value for money’ and
‘the pre-eminences of the market’
ensuring all existing weapons systems can be operated at maximum
combat rates for extended periods, backed up by adequate manpower
and logistics – the essence of modern mobilisation and sustained
warfighting
hardening key air bases to increase their survivability against strikes
by the full range of China’s weapons – especially if critical US aircraft
are going to be stationed on our bases – and preparing for the quick
dispersion of vulnerable assets as protection against surprise attack
increasing our ability to move a limited amount of shipping around the
coast of Australia for transportation of key bulk commodities and
justifying the current and future construction of inland railways on
national security grounds
understanding, in the context of deciding the size, timing and type of
our support for a contribution to a regional alliance, how disastrous it
would be for homeland defence if Australia deployed our best
warfighting forces from our one-shot ADF to support the US in the
Western Pacific and they suffered severe casualties, and China
subsequently became the dominant power in our region
ensuring that the ADF can function logistically, in terms of liquid fuel,
weapons, missiles, spare parts etc, at combat rates, for an extended
period of time (six months or a year) if Australia is denied access to
commercial sea or air transport
planning to increase the size of the ADF (in the first instance aircraft,
including drones, and warships) in accordance with the conclusions of
the national security strategy, by procurement or by retaining assets
that would in the past have been retired when new materiel was
introduced, to match the tasks the ADF must conduct concurrently and
to make the ADF less of a one-shot force
significantly increasing the voluntary reserve element in each service,
to increase mass and sustainability and to help Australians recognise
that defence is the responsibility of the whole nation
acknowledging, with a view to gaining public support, that the cost of
defending our nation should no longer be based on what a government
thinks it can afford, but on what the strategic environment demands.
***
The time for complacency for Australians is over, and now the task is to
build real strength. While those engaged in strategic and military policy are
fond of talking about an ‘uncertain strategic environment’, we now have a
better understanding than ever of our potential strategic threats. China –
either in concert with others or alone – is capable of challenging the US,
dominating the Western Pacific and harming Australia’s interests to the
extent that Australia as a liberal democracy is in danger, so preparing for a
China war must be the standard we judge ourselves by. And we have a good
idea of the strategic objectives that China might fight for: promoting the
CCP’s leadership by diminishing US power and retrieving the ‘renegade
province’ of Taiwan.
China’s recent actions in the South China Sea, its use of trade
embargoes and cyber-attacks, and its increasingly authoritarian leadership
mean that we are already in a conflict that does not yet involve violence. In
the leadup to a war, China will use the grey-zone tactics we are currently
experiencing in stronger and more potent ways. If it does proceed to war, it
will likely have the strategic advantage of being the ‘first mover’, using the
greatest military weapons in world history: deception and surprise. There is
no saying how long we have before it exercises this advantage.
In the face of China’s aggression, Australian governments must
continue to lead decisively on national security, and the Australian people
must support moves to protect national sovereignty and promote resilience.
This requires action on many fronts, especially using diplomacy to
encourage US strength and deter China’s adventurism. It must also include
strengthening our economy so it can absorb the shock of conflict and, most
importantly, making the ADF larger, more potent and more capable of
fighting for an extended period.
Only a complete package, as the result of a national security strategy
that creates self-reliance and resilience, will give Australia a real chance of
deterring the conflict that we do not want. War is a terrible thing. But the
only thing that is worse is being involved in a war and losing.
EPILOGUE
A NEW REALITY
The world still turns the day after the satellites tumble from the sky, and it
will keep on turning. Empires come and go in world history and the gods
that insignificant humans invented to explain the world look down and
shake their heads. With the Chinese attack in the Pacific, another empire,
the US ‘empire’, has been challenged for primacy on the third rock from the
sun.
For the gods this is hardly noteworthy. Every previous empire, dynasty
or caliphate, whether it be the Abbasid, the Portuguese, the Spanish, the
Qing, the Mongol, the Russian, the Roman or the British, thought it would
last forever, and their leaders and people began to take that eternal
existence for granted. But they were not special and they did not last.
The wonderfully inventive human brain, which has brought so much
good to the world, seems incapable of avoiding the pitfall of complacency.
The British philosopher Thomas Hobbes, living in the early stages of the
British Empire, described the life of man as ‘solitary, poor, nasty, brutish
and short’ and so in many ways set the objectives of what good nations
should strive to improve. Of all the nations on the face of the Earth through
world history, few have made greater progress in overcoming, for the good
of its citizens, the five characteristics of Hobbes’s world. Few have come as
close to perfection as Australia. But now Australians have to understand
how and why their world is being challenged.
Since Hobbes’s time, both good and evil have marked the progress of
the world through empires, and Australians must remember that wars come
and go, that it was only 80 years ago that Italian, German and Japanese
fascism, totalitarianism and militarism was defeated by our parents or
grandparents – just as an elderly Ukrainian woman looking into a TV
camera in a ruined street reminds us that she has seen war before and
wonders why Russian President Putin does not just go home.
In the absence of the ever-demanding hum of servers and computers in
a world cut off from its accustomed means of communications by the rise of
yet another empire seeking primacy at the expense of its neighbours,
Australians are perhaps contemplating that there is evil in the world, and
what the world democracies have created for the betterment of man can
only in the end be sustained by vigilance and by strength.
But philosophy is the last thing on the minds of Australia’s citizens and
their leaders on the days after the sudden and vicious attack by China. Most
people are merely aware that the world had changed forever. The post–
World War II era is over. The mighty United States of America, whose
physical presence and power has facilitated security and prosperity, has
suffered a defeat far worse and far more sudden than any in its history. The
most powerful of its military organisations, the Indo-Pacific Command, has
been smashed, and tens of thousands of its members have been killed or
injured. The likelihood of the US being able to fight back in the short to
medium term is very much an unknown.
China has made its move and that move has been almost totally
successful because of the complacency of the world towards the potential
malice of an authoritarian power. Few thought that Russia would move on
one of its European neighbours, yet it did. And few applied the lessons of
Ukraine to other parts of the world. After one authoritarian nation under its
brutal dictator mounted a war of aggression in a supposedly ‘civilised’ part
of the world, there was no reason why another equally brutal dictator
would not do the same in our part of the world. The bravery of the
Ukrainians in fighting the Russians, the surprising incompetence of the
Russian military, the willingness of the world to apply sanctions against
Russia and to supply weapons to Ukraine meant that Russia encountered
far greater resistance than it expected. Could these lessons not be
transferred from Russia to China? Might not China draw a completely
different set of lessons from Russia’s experience in Ukraine? Would it take
advantage while the attention of the democracies was fixed on Ukraine?
Was Sino arrogance so marked that they believed they would never make
the errors that Russia had made?
Not only has China smashed the US military in the Western Pacific, it
has almost totally prevented the return of US forces by securing the first
and second island chains and removing any bases that the US might have
relied on to support its future moves. China has sent small groups of troops
with missiles to every strait in the first island chain without the permission
of any of the countries involved. Backed up by its intact air force and navy,
it is able to prevent any US ship from entering the first island chain and
approaching the Chinese mainland or Taiwan, and using judiciously
positioned commercial ships, it has blocked the Suez and Panama canals to
slow the redeployment of US and allied forces, if in fact they try to move
from Europe or other places to the Pacific. China has even located small
forces in Papua New Guinea and one area of the South Pacific, the
Solomon Islands, to allow it to monitor Australia and prevent it being used
as a base for US forces.
China is now the only nation in the world which still has fully
functioning space operations and cyber systems. It is willing to permit
others to re-open their communications to a limited extent but only as far as
it suits China’s needs, for instance to communicate with China or to spread
concern and confusion. For the time being, President Xi is happy to let the
rest of the world stew in its fears about this new, uncertain world.
***
The impact of the attack on Australia’s military and infrastructure is
enormous. Australian governments and agencies learn very quickly of the
substantial damage to our modern and expensive air force, which were left
vulnerable, with planes sitting on open tarmac or inside hangars designed
to keep off the weather but not cruise missiles fired from submarines. The
Australian navy tries twice to move ships from inner harbour mooring
points, once in Sydney Harbour itself and once at its Fleet Base West near
Perth, and both ships are struck by mines inserted by Chinese submarines
from outside the harbour in the days following the attack on the air force
bases. From then on not a single navy ship can move until the slow process
of clearing the harbour of mines is completed. Only one harbour at a time
can be cleared because the navy does not have the resources to do more.
Ships that were at sea at the time of the attack are ordered to stay at sea
until the safe use of the harbours can be guaranteed.
Some Australian officials have experience of crisis management, having
recently dealt with drought, floods, pandemics and widespread fires, but no
Australian government has faced a situation like the current one since 1941
and there is absolutely no national memory. The so called ‘War Book’, a
guide to what should happen in a defence emergency, has not been reviewed
for decades. It’s still an actual book, a hard copy so out of date and so full
of generalisations from another era as to be of no use at all. And of course,
the days after the massive attack on our major ally and a significant attack
on this country do not lend themselves to the calm study of historic
documents. The federal government knows enough to adopt the total
emergency defence powers that are clearly in the constitution and which
enable the federal government to take any action that it sees fit, but of
course with most of the normal communications still down, and without a
clear picture of what is happening overseas, and without a tradition of
strategy-making itself and with no preparation for such an event, what
action needs to be taken by the federal government and with what priority is
unclear. The big question as the weeks go by is whether there will be
another Chinese attack, either on the US or on Australia. Is this the end of
the violence or only the beginning?
The most immediate needs of the population are the most pressing,
regardless of what happens overseas, or whether there is another attack or
not. People need food and transport, and they need leadership.
For a period of days, what is on the supermarket shelves on the day of
the attack will be all there is for some time. Recalling the panic buying that
occurred during the early days of the COVID pandemic, shoppers rush to
the supermarkets, though many shops remain closed since the attack.
Australia is never going to be short of food, but it is certainly going to be
some time before it can be moved from fields to factories and then to
markets, given that Australia is dependent on diesel truck transport, roads
and supermarkets. Australia had less than 30 days of liquid fuel (diesel,
petrol and aviation fuel) in the country at the time of the attack. Successive
governments spoke of having twice that number of days of fuel because
another 30 days’ worth was always on its way in tankers. But after the
attack not one commercial oil tanker docks at an Australian port, due to
fear of mines in ports and a belief that insurance will not cover acts of war.
Even tankers close to Australian ports turn about and go back to a safe
harbour at the direction of their owners or on their captain’s initiative. The
remaining two refineries in Australia are commissioned to increase
production to the maximum, but they were already working at 90% capacity
to produce only 10% of Australia’s needs. There is no spare capacity. States
are directed to permit the exploitation of crude oil still in the ground or just
offshore, but this will take months at least, and Australia will have no
refined fuel in less than 30 days. The government has no answer because it
assumed it would never have to face this problem. It relied on a strategy of
hope.
The governments’ priorities become food, fuel and governance (law and
order). Backed by its defence emergency powers, the federal government is
able to act relatively decisively in regard to directing state governments and
other agencies. It is not prepared to establish a war cabinet including the
opposition so early in this emergency, but it makes attempts to consult with
the opposition parties. Leadership itself is difficult. Although most of the
domestic mass-communications channels are re-established within days,
without international news feeds that give context to what a prime minister
is expected to say in these circumstances, it is hard for the head of
government to speak authoritatively, and what news there is, mostly isn’t
good.
Limited communications with the world are restored as permitted by
China but Australia has no backup satellites to relaunch despite the obvious
need for them, and the sea cables will take years to repair.
As time goes on, Australian standards of living collapse. Without petrol
or diesel, many people cannot get to their places of employment. While
many can work from home using the internet where the internet has been
restored, the need for such service workers diminishes. People need
physical goods, produced by workers in factories or on farms, that you can
either eat or put into a vehicle, not lifestyle services.
Health services take some time to collapse, as the reserves of most
pharmaceuticals last for up to six months. But because spare parts for the
nation’s generators of base load power come from overseas, the supply of
electricity to the cities and the little industry Australia has becomes
unreliable. The lights across the nation do not go out, as solar power and
hydro power keep some of them on, but the extraordinary lifestyle
Australians have enjoyed is only a dim memory.
Key cargo is flown in on commercial cargo aircraft, which have all been
requisitioned by the government for this purpose. But how can you find the
critical spare parts for a modern society from overseas when the
coordination of the internet is missing? How can you send a cargo aircraft
to the right place in North America or Europe to find the right item? And
what will happen when there is no aviation fuel?
***
Diplomacy, deprived of modern communications, cannot achieve very much
quickly. Governments are used to having instantaneous control over
ambassadors due to very fast communications, and leaders had been able
to speak directly to each other in ways unknown before in history.
Communications have been limited yet the need to communicate has never
been more important.
When President Xi feels ready, China’s post-attack diplomacy begins. It
enters negotiations with Japan and South Korea first, then Taiwan, then the
US, then South-east Asia and, finally, Australia. China has achieved its
aims with almost no damage to its cities, its industries, its people or its
military, and now it has every desire to secure its strategic objectives of
removing the US from the Western Pacific and reincorporating Taiwan
without the use of force.
China’s message is now one of reasonableness, peace, growth and
stability. It only wants good things for all the countries in the region, it
maintains, now that the US is gone. And not only good for the region but for
the world. China needs to trade with the world to maintain its own internal
social cohesion, but on this occasion it had put its strategic objectives
ahead of trade for the short term. Now it is back to playing the long game,
one backed by overwhelming military force.
It promises Japan and South Korea that if they accept the new regional
power situation, then China will forego any territorial claims it may have
against either country and will respect the sovereignty of both. The quicker
both countries accept the new situation, the sooner trade and normality will
recommence. Then, says China, with the US out of the region, Asia will
once again be for Asians.
The message to the US is blunt. You are out of the Western Pacific and
we will not let you re-establish your bases in Japan, South Korea or even
Guam. From Japan to Australia and out to Hawaii, the Western Pacific is
now a Chinese sphere of influence. The US’s regional allies are not going to
fight on without the US, so the US must accept the new situation. Even
Taiwan will now willingly joint the PRC.
To Taiwan, the message is even more blunt. Now is the time to think of
where your future best lies, the Chinese diplomats say, as an autonomous
province of the PRC, your home. If you now willingly join the PRC, you will
continue to enjoy prosperity and security and China combined with Taiwan
will be a global powerhouse. The alternative is too awful to contemplate. If
Taiwan does not accept reincorporation, China will achieve it by force.
Japan and South Korea will not fight and, without the US, Taiwan will be
defenceless. Your autonomy will be guaranteed under the conditions of your
reincorporation, say the diplomats. Come willingly.
The message to Australia is also threatening. China had assessed that if
they were successful with the sudden attack, the only way for the US to reestablish military power in the region would be by projecting power from
Hawaii, Alaska, the west coast of the US and from Australia. There was
nothing that China could do about preventing US power being assembled in
US territories (except for Guam and other mid-ocean territories), but it is
determined that Australia will not be used as a base for a US counteroffensive against China. US forces must not be permitted under any
circumstances to use Australia as a base. If Australia allows this, or any
other action that supports US influence in the Western Pacific, the South
Pacific or the Indian Ocean, then China will be obliged to take direct action
against Australia, which might involve sanctions, blockade and even
occupation.
China tells Australia that it regrets it had to act against the Australian
military and is sorry for the loss of life. But Australia must now recognise
that its region has changed and if Australia wants to maintain any shred of
sovereignty, if it wants to open its sea routes to commercial shipping and
import and export goods, then it has to reject any involvement with the US
military. In effect, Australia must become a tributary state of China.
Australia has lost its air force and the ability to use its navy and its
ports, its army is irrelevant and it has no sovereignty now except as
permitted by China. Australia faces an enormous decision – does Australia
accept China’s direction and do what it has been told in relation to the US,
or does it defy China and stand beside the US, impotent though both
countries are, with the awful consequences that might entail?
As time goes on, the anarchist part of Australian society takes to
demonstrating on the streets. Riots at food distribution points become
frequent and cannot be controlled by normal policing methods. The worst in
Australia’s fringe society comes to the fore. Australian society is no
different from any other society in relation to the observance of law and
order. Even during natural disasters, the mateship that Australians talk
about so often has strong elements of myth about it. Of course, some people
help others and show bravery, but others steal and loot their neighbours’
properties. On occasions, law and order have to be enforced with draconian
methods. This is our brave new world.
Australian society is as fragile as any other society in the world and
that this would surprise many Australians is only because our society has
been spared for so long the effects of the worst kind of unnatural disaster –
war. The dominant thought in most Australian minds by this point is that
anything that is required to restore normalcy as it has been understood
since 1945 should be done. There is enormous pressure on the government
to take Australia and Australians back to the so recent past. Against the
practicality of day-to-day living, as the economy and society collapse, the
notion of liberal democracy, by now only defendable through a war against
China, is a difficult idea to sell.
And what in reality can Australia do? Its greatest ally is licking its
wounds. Its neighbours have rolled over and accepted China as the regional
hegemon. Its greatest adversary has re-presented the grievances that it
dumped on Australia back in 2020: accept Huawei; remove foreign
interference laws; forget about the origins of COVID; permit no US forces
to use Australian bases; refrain from criticising China on the South China
Sea, human rights and cyber-attacks; join the Belt and Road Initiative; and
finally, re-price (downwards and severely) your exports of coal and iron ore
to China.
The Australian government looks seriously at these demands. Because
that is what happens to tributary states.
ACKNOWLEDGEMENTS
The impetus for this book was my limited success in convincing people that
national security needed more attention. In late 2020, I realised that my
challenges were due to a lack of national awareness of the likely nature of
any war, much less a regional war involving China. To overcome this
deficiency, I decided I would have to describe such a war; this book
includes that description.
The period during which I wrote was both personally and nationally
tumultuous. Early on, I was diagnosed with an aggressive form of cancer,
for which I am still receiving treatment. At the same time, COVID
lockdowns were disrupting many lives and livelihoods, and I had to
perform most of my senatorial duties remotely. The final stages of the book
were completed during the May 2022 election campaign, which of course
had an unsuccessful outcome for the Coalition. But the time I spent
campaigning, and the many conversations I had with everyday Australians
about their priorities and concerns, only served to reinforce the value of
completing this work.
I would particularly like to thank the Hon. Peter Dutton MP, the former
Coalition Minister for Defence, who not only encouraged me to be
outspoken on defence but agreed to write the foreword.
I also thank former prime minister the Hon. Tony Abbott AC and
former deputy prime minister the Hon. John Anderson AO for the many
discussions we have had and the guidance they have given me.
My gratitude to Mary Rennie, Publisher, Non-Fiction and Commercial
Fiction, and Senior Editor Scott Forbes at HarperCollins for guiding me
through this often-lonely process, and to David Connery for assistance with
local editing.
I also wish to mention my Senate staff, who have provided stellar
support during a difficult and highly unusual term of government. While
they were not involved in writing this book, we all still argue as to who
thought of the title!
But, as always, my deepest and most enduring gratitude is reserved for
my wife, Anne. She has supported this project in the same steadfast,
wholehearted way she has supported all my professional passions, and has
shown more love and endurance than any husband deserves. I could not
have survived cancer, much less the process of writing a book alongside my
political activities, without her.
ENDNOTES
Prologue: Out of the Blue?
1 Commonwealth, Parliamentary Debates, House of Representatives, 6 July 1950, p. 3 (Robert
Menzies, Prime Minister).
Chapter 1: Rise to Power
1 Davidson, P., ‘Statement of Admiral Philip S. Davidson, U.S. Navy Commander U.S. IndoPacific Command before the House Armed Services Committee on U.S. Indo-Pacific command
posture 10 March 2021’, pp. 31–40.
2 Medcalf, R., Contest for the Indo-Pacific, La Trobe University Press, Melbourne, 2020, pp. 46–
48, 221, 265.
3 Medcalf, op. cit., pp. 30 and 372–3.
4 Hayton, B., The Invention of China, Yale University Press, New Haven, Connecticut, 2020, p.
352. See also Garnaut, J., ‘National Socialism with Chinese characteristics’, Foreign Policy, 15
November 2012.
5 Hayton, op. cit., p. 352.
6 McCann, D., ‘National Socialism with Chinese Characteristics’, Spectator Australia, 21 February
2021.
7 Hayton, op. cit., pp. 350–2. Also McCann, op. cit.
8 Mastro, O.S., ‘How China is bending the rules in the South China Sea’, The Interpreter, Lowy
Institute, 10 February 2021; Maizland, L., ‘China’s repression of Uyghurs in Xinjiang’, Council
on Foreign Relations, 1 March 2021.
Chapter 3: Wolf Warriors
1 Kelly, P., ‘Australia has shown great resilience in the face of China’s aggression’, The Australian,
10 July 2021.
2 Lema, K., ‘Philippines defence chief says was urged by China to drop review of U.S. pact’,
Reuters, 30 September 2021.
3 For example, see Brand, H., ‘China is a declining power – and that’s the problem’, Foreign
Policy, 24 September 2021; and Patey, L., ‘China is an economic bully – and weaker than it
looks’, Foreign Policy, 4 January 2021.
4 Beckley, M., and Brands, H., ‘The end of China’s rise’, Foreign Affairs, 1 October 2021.
5 Beckley, M., Unrivaled: Why America Will Remain the World’s Sole Superpower, Cornell
University Press, Ithaca, 2018.
6 Ellis, S., ‘Taiwan remains China’s fatal attraction’, Australian Financial Review, 8 March 2022.
7 Mead, W.R., ‘Xi Jinping’s two-track foreign policy’, Wall Street Journal, 11 October 2021.
8
9
Ibid.
Talbot, J., ‘CCP sanctioned video threatens China will nuke Japan in a “full-scale war”’, Sky
News, 19 July 2021.
Chapter 4: In the Grey Zone
1 Hanson, F., Currey, E. and Beattie, T., ‘The Chinese Communist Party’s coercive diplomacy’,
Australian Strategic Policy Institute, Canberra, 2020, pp. 4–5.
2 Sanger, D., The Perfect Weapon: War, Sabotage, and Fear in the Cyber Age, Scribe Publications,
Melbourne, 2018.
3 Sanger, op. cit., pp. 118–130.
4 Morrison, S., ‘Morrison’s messages to the “sophisticated state-based cyber actor”’, Lowy
Institute, 19 June 2020.
5 Lowy
Institute
Poll,
comparing
results
for
2006
with
2021;
see:
https://poll.lowyinstitute.org/charts/threats-australias-vital-interests.
6 Gertz, B., ‘Chinese military seeks to dominate from space, deploys war-fighting tools into orbit’,
Washington Times, 13 October 2021.
7 Brown, L., ‘Military chiefs warn of space bullying by Russia and China’, The Australian, 30 July
2021.
8 Wigston,
Air
Chief
Marshal
M.,
UK
Space
Command,
www.facebook.com/royalairforce/videos/uk-space-command-the-chief-of-the-air-staff-air-chiefmarshal-sir-mikewigston-o/896042440948263/.
9 Brown, op. cit.
10 Tidsall, S., ‘Little blue men: The maritime militias pushing China’s claims’, The Guardian, 16
May 2016.
Chapter 5: Where and When?
1 While there is no single ‘official’ government scenario, there is significant speculation in
Australia about the possibility of conflict over Taiwan and its implications for Australia. For
examples, see: Grieber J., Smith, M. and Tillet, A., ‘Canberra prepares for Taiwan conflict as
tensions escalate’, Australian Financial Review, 16 April 2021; Medcalf, op. cit., p. 373;
Roggeveen, S., ‘Will ANZUS make it to 80?’, The Interpreter, Lowy Institute, 1 September
2021; and Burshtein, D., ‘The Taiwan tango: a delicate dance for Australia’, Spectator Australia,
4 February 2021.
2 Funaiole, M.P. and Glaser, B.S., ‘China’s provocations around Taiwan aren’t a crisis’, Center for
Strategic and International Studies, 19 May 2020.
3 Tsirbas, M., ‘What does the nine-dash line actually mean?’, The Diplomat, 2 June 2016.
4 Yoshihara, T. and Holmes, J. R., Red Star over the Pacific: China’s Rise and the Challenge to
U.S. Maritime Strategy, second edition, Naval Institute Press, Annapolis, 2018, p. 73.
5 Moseley, T.M. and Corley, J.D.W., ‘The next DC tanker battle could determine who wins the next
war’, RealClear Defense, 7 October 2021; Holmes, J., ‘Defend the first island chain’,
Proceedings, April 2014.
6 Missile Defense Project, ‘Missiles of China’, Missile Threat: Center for Strategic and
International Studies, 14 June 2018; see also Shugart, T., ‘Australia and the growing reach of
China’s military’, Lowy Institute Analyses, 9 August 2021.
7 ‘Chinese air force video shows major H-6 bomber strike on Guam-resembling island’, Military
Watch, 22 September 2020.
8 Ellis, S., ‘As fears of a PLA invasion grow, analysts offer possible scenarios’, Taipei Times, 11
October 2020.
China Power Team, ‘How is China modernizing its nuclear forces?’, ChinaPower: Center for
Strategic and International Studies, 10 December 2019.
10 ‘Taiwan defence minister pushes new arms spending, says China tensions worst in four decades’,
Reuters, 6 October 2021.
11 Robinson, L., Tell Me How This Ends: General David Petraeus and the Search for a Way Out of
Iraq, PublicAffairs, New York, 2008.
9
Chapter 6: The Response: the US and Its Allies
1 Copp, T., ‘“It failed miserably”: After wargaming loss, joint chiefs are overhauling how the US
military will fight’, Defense One, 6 July 2021.
2 Ullman, H., ‘America’s flawed war strategies’, The Hill, 15 June 2021; Mattis, J., ‘Summary of
the 2018 National Defense Strategy of the United States of America’, US Department of Defense,
2018, p. 6.
3 A National Security Strategy for a New Century, The White House, October 1998, p. 7.
4 Congressional Budget Office, Illustrative Options for National Defense under a Smaller Defense
Budget, October 2021.
5 Cordesman, A.H. and Hwang, G., ‘The Biden administration: Strategy and reshaping the national
security budget’, Center for Strategic and International Studies, 16 February 2021.
6 Feldscher, J., ‘CIA creates China center to shift to great power competition’, Defense One, 8
October 2021.
7 Spoehr, T., Bowman, B., Clark, B. and Eaglen, M., ‘What to expect when you’re expecting a
National Defense Strategy’, War on the Rocks, 27 September 2021.
8 Townshend, A., Thomas-Noone, B. and Steward, M., ‘Averting crisis: American strategy,
military spending and collective defence in the Indo-Pacific’, United States Studies Centre,
University of Sydney, 19 August 2019.
9 Gould, J., ‘Debate on “no first use” of nukes mushrooms in Washington’, Defense News, 7
October 2021.
10 Ibid.
11 Spoehr, Bowman, Clark and Eaglen, op. cit.
12 Ibid.
13 Mehta, A., ‘Inside US Indo-Pacific command’s $20 billion wish list to deter China – and why
Congress may approve it’, Defense News, 2 April 2020.
14 ‘Senator Hawley continues to stand with Taiwan, introducing new bill to help nation arm itself’,
Josh Hawley, US Senator for Missouri, 21 November 2021, quoted in Abodo, S., ‘Congress must
pass Senator Hawley’s Arm Taiwan Act’, Newsweek, 11 November 2021.
15 Abodo, op. cit.
Chapter 7: The Response: Australia
1 Church, N., ‘The Australia–United States defence alliance’, Parliament of Australia.
2 Dibb, P., ‘Radical new defence policy or Hill’s smoke and mirrors?’, The Australian, 16
December 2005, cited in ‘Australia’s national security: A defence update 2003, 2005 and 2007’,
Parliament of Australia.
3 Department of Foreign Affairs and Trade, ‘Trade and investment at a Glance 2019’, Australian
Government, 2019.
4 Dale, T., ‘The G20: A quick guide’, Parliament of Australia, 4 March 2014.
5 For examples of China’s emphasis on readiness and the use of force, see Lie Zhen, ‘Xi Jinping
orders China’s military to be ready for war “at any second”’, South China Morning Post, 5
January 2021; Yiallourides, C., ‘Is China using force or coercion in the South China Sea? Why
words matter’, The Diplomat, 11 July 2018; and Medcalf, op. cit., pp. 175–9.
Chapter 8: Concepts of War
1 It is unlikely that Sun Tzu actually said this, but it fits his overall philosophy well (in this, I agree
with Gian P. Gentile in his article ‘The accidental coindinista: A historian’s journey back from the
dark side of Social science,’ Infinity Journal, October 2012). For a detailed introduction to Sun
Tzu’s work, see Sun Tzu, The Art of War, translated by Samuel B. Griffiths, Oxford University
Press, 1963.
2 Molan, J., Running the War in Iraq, HarperCollins Publishers, Sydney, 2008.
Chapter 9: Current Policy
1 Morrison, S., ‘Launch of the 2020 Defence Strategy Update’, Australian Journal of Defence and
Strategic Studies, Vol.2, No.2, 2020.
2 Morrison, S. and Payne, M., ‘Australia to pursue nuclear-powered submarines through new
trilateral enhanced security partnership’, Australian Government: Defence, 16 September 2021.
3 Hartcher, P., ‘How Australia has shaped up to Xi’s aggression’, Sydney Morning Herald, 12
October 2021.
4 Curran, J., ‘AUKUS is the death knell of Australia’s strategic ambiguity’, Defense One, 20
September 2021.
5 Power, J., ‘US should give Australia access to operations in Singapore, Guam, Philippines:
Report’, South China Morning Post: This Week in Asia, 15 October 2021.
6 Bramston, T., ‘Defending Taiwan against Beijing is a must, says Peter Dutton’, The Australian,
12 November 2021.
7 Bramston, op. cit.
8 Keating, P., ‘Address to the National Press Club of Australia’, 10 November 2021.
9 Bramston, op. cit.
10 Eckstein, M., ‘US Navy reorganizes submarine enterprise to address challenges in construction,
maintenance’, Defense News, 28 September 2021.
11 Greene, A., ‘Australian Defence Force may use Collins Class submarines for another 30 years
while waiting for nuclear replacements’, ABC News, 15 October 2021.
12 ‘Defending Australia in the Asia Pacific century: Force 2030’, Defence White Paper 2009,
Department of Defence, Canberra, 2009.
Chapter 10: The Right War
1 Layton, P., China’s Enduring Grey-zone Challenge, Air and Space Power Centre, Commonwealth
of Australia, Canberra, 2021, p. 46.
2 De Luce, D. and Dilanian, K., ‘China’s growing firepower casts doubt on whether U.S. could
defend Taiwan’, NBC News, 27 March 2021.
Chapter 11: Can the US Win?
1 Wickman, T., ‘COMPACAF focuses on threats, Airmen efforts in Pacific at AFA Warfare
Symposium Conference’, 4 March 2022.
2 Copp, T., op. cit.
3 Davis, M., ‘Towards China’s A2AD 2.0’, Australian Strategic Policy Institute: The Strategist, 24
November 2017.
4 Missile Defense Project, ‘DF-21 (CSS-5)’, Missile Threat: Center for Strategic and International
Studies, 13 April 2016.
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
Judson, J., ‘Iron Dome heads to missile defense experiment in Guam’, Defense News, 8 October
2021.
Axe, D., ‘Anticipating war with China, the U.S. Air Force is fanning out across the Pacific’,
Forbes, 7 June 2021.
Missile Defense Project, ‘Aegis Ashore’, Missile Threat: Center for Strategic and International
Studies, 14 April 2016.
Richardson, D.Z., Nahom, D.S. and Guastella, J.T., Air Force, Force Structure and
Modernization Programs: Presentation to the Senate Armed Services Committee Subcommittee
on Airland, US Senate, 22 June 2021.
Peck, M., ‘U.S. stealth fighter jets are a problem for China – but Beijing says it has the answer’,
The National Interest, 8 July 2021.
Del Toro, C., One Navy–Marine Corps team: Strategic Guidance from the Secretary of the Navy,
October 2021, quoted in Katz, J., ‘Navy Secretary’s new guidance puts the target on China’,
Breaking Defense, 8 October 2021.
Office of the Secretary of Defense, Military and Security Developments involving the People’s
Republic of China, Department of Defense, 2021, pp. 67, 81 and 120–1.
China Power Team, ‘How is China modernizing its nuclear forces?’, China Power: Center for
Strategic and International Studies, 10 December 2019, updated 28 October 2020.
Yoshihara and Holmes, op. cit., p. 150.
Batchelor, T., ‘U.S. deploys one-third of Pacific submarine fleet for major naval exercise’,
Newsweek, 4 June 2021.
Axe, D., ‘All three of the U.S. Navy’s most powerful submarines were under way at the same
time’, Forbes, 1 August 2021.
Ibid.
Yoshihara and Holmes, op. cit., p. 154.
Bowman, B. and Montgomery, M., ‘AUKUS: Good goals, bad implementation’, Defense One, 27
September 2021.
‘An interactive look at the U.S.–China military scorecard’, RAND Project Air Force,
www.rand.org/paf/projects/us-china-scorecard.html.
Peniston, B., ‘US Navy’s latest plan for its future may not come until 2023, says top admiral’,
Defense One, 24 September 2021.
‘U.S., Philippines: A rocket deal that will make waves in the South China Sea’, Stratfor, 3 April
2019.
Chapter 12: Australia, Defend Yourself!
1 A similar approach has been adopted in the US. As commentators have noted: ‘The new US
National Defense Strategy that the Biden administration is writing should assess core U.S.
strategic objectives and delineate the necessary Department of Defense capabilities, capacities,
and forward posture required. This new strategy should be adequately resourced, or it will be
destined for irrelevance.’ See Spoehr, Bowman, Clark and Eaglen, op. cit.
2 Morrison, S., ‘Launch of the 2020 Defence Strategy Update’, Australian Journal of Defence and
Strategic Studies, Vol.2, No.2, 2020.
ABOUT THE AUTHOR
JIM MOLAN is an Australian Liberal senator and retired army major
general. He has been an infantryman, a helicopter pilot, a commander of
army units (from a 30-man platoon to a division of 15,000 soldiers),
commander of the Australian Defence Colleges, and commander of the
evacuation force from the Solomon Islands in 2000. He has served in Papua
New Guinea, Indonesia, East Timor, Malaysia, Germany, the US and Iraq.
In April 2004, he deployed for a year to Iraq as the Coalition forces’
chief of operations, where he controlled the manoeuvre operations of all
forces across Iraq, including maintaining the security of Iraq’s oil,
electricity and rail infrastructure. Described as ‘the ADF member most
directly involved in fighting the insurgents’, he was awarded the
Distinguished Service Cross by the Australian government for
‘distinguished command and leadership in action in Iraq’, and the Legion of
Merit by the US Government.
Since leaving the military, Jim has been a commentator on defence and
security issues and has written regularly for a number of journals and blogs.
He was part owner of an Australian company facilitating access for
Australian industry to defence technology grants and working with other
high-tech industries, and he was nominated as chairman of two companies
attempting to commence trading in Australia. He was a consultant to
Deakin University, BAE Systems Australia, and Israeli Aerospace
Industries.
As well as being involved in formulating the Coalition’s defence policy
while in opposition (2012–13), Jim was a coauthor, with Scott Morrison
MP, of the Coalition policy on border control. Following the 2013 federal
election, Jim was appointed to the full-time position of Prime Minister’s
Special Envoy for Operation Sovereign Borders. He has been active in
speaking out on defence issues, particularly Australia’s preparedness
against an aggressive China. He lives on a property outside Canberra.
COPYRIGHT
HarperCollinsPublishers
Australia • Brazil • Canada • France • Germany • Holland • India
Italy • Japan • Mexico • New Zealand • Poland • Spain • Sweden
Switzerland • United Kingdom • United States of America
HarperCollins acknowledges the Traditional Custodians of the land upon which we live and work,
and pays respect to Elders past and present.
First published in Australia in 2022
by HarperCollinsPublishers Australia Pty Limited
Gadigal Country
Level 13, 201 Elizabeth Street, Sydney NSW 2000
ABN 36 009 913 517
harpercollins.com.au
Copyright © A & J Molan Pty Ltd 2022
The right of Jim Molan to be identified as the author of this work has been asserted by him in
accordance with the Copyright Amendment (Moral Rights) Act 2000.
This work is copyright. Apart from any use as permitted under the Copyright Act 1968, no part may
be reproduced, copied, scanned, stored in a retrieval system, recorded, or transmitted, in any form or
by any means, without the prior written permission of the publisher.
ISBN 978 1 4607 6260 8 (paperback)
ISBN 978 1 4607 1522 2 (ebook)
A catalogue record for this book is available from the National Library of Australia
Cover design by George Saad
Cover image by Clem Onojeghuo on Unsplash; soldier silhouette by shutterstock.com
Maps by Map Illustrations, www.mapillustrations.com.au