Smil 2005

The Next 50 Years: Fatal Discontinuities
Author(s): Vaclav Smil

Source: Population and Development Review, Vol. 31, No. 2 (Jun., 2005), pp. 201-236
Published by: Population Council
Stable URL: http://www.jstor.org/stable/3401359
Accessed: 15/09/2010 12:11
Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at
http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless
you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you
may use content in the JSTOR archive only for your personal, non-commercial use.
Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at
http://www.jstor.org/action/showPublisher?publisherCode=popcouncil.
Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed
page of such transmission.
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact [email protected].
Population Council is collaborating with JSTOR to digitize, preserve and extend access to Population and
Development Review.
http://www.jstor.org
The Next 50 Years:

Fatal Discontinuities
VACLAV SMIL
ANY ONE OFUS may indulge in speculations about global futures tailored to
particular moods or biases, from Francis Fukuyama's (1992) ahistorical end
of history (to be delivered by the universal triumph of liberal democracy)
to Paul Ehrlich's latest lament that sees the very same liberal democracy
soon to be one with Nineveh (Ehrlich and Ehrlich 2004).1 This never-ending parade of grand forecasts has one thing in common: their outcomes are
preconceived, and their arguments are predetermined by strongly held visions, be they of inexorable progress or unavoidable collapse. And then there
is a burgeoning field of specific point forecasts, supposedly the outcomes of
disinterested exploration, that quantify numerous attributes of populations,
environments, techniques, or economies. The internet has made it a matter
of seconds to find the requisite data for particular years: total number of
females in Yemen in 2040, CO2 concentrations in the atmosphere in 2030,
share of electricity generated by wind turbines in Denmark in 2020, the
aggregate US national debt in 2010.
The only sensible way to appraise the reliability of such forecasts is to
look back and see how well their counterparts foretold yesterday's and
today's realities. Such backward-looking exercises are particularly valid because during the past generation most of these specific point forecasts have
relied on the same suite of intellectual approaches and (often computerized) forecasting techniques as do today's prognoses that look five to 50 or
more years ahead. These retrospectives reveal that most of the truly longrange quantitative forecasts (spanning roughly one generation, or between
15 and 25 years) turn out to be useless within years, even within months,
of their publication. I have demonstrated these failures by a detailed examination of more than a century of every possible category of long-range
energy forecasts (Smil 2003).
Trend forecasts fail so rapidly because they tend to be unrealistically
static. But trends are not infinite: they weaken or deepen suddenly, they
can be reversed abruptly. Population forecasts provide pertinent examples
POPULATION
AND
DEVELOPMENT
REVIEW
31(2):
201-236
(JUNE
2005)
201
202
THE NEXT 50 YEARS: FATAL DISCONTINUITIES
of these repeatedlyfailedanticipations.The unprecedenteddecline of Russia's

fertility was obviously taken into account when the medium variant of the
1998 revision of the UN WorldPopulationForecastput the country's total at
121.25 million people in the year 2050 (UnitedNations 1999); but that trend
had deepened so rapidly that a mere four years later the 2002 revision offered only 101.45 million (United Nations 2003), and then it weakened a
bit so the 2004 revision raised Russia's 2050 population total to 111.8 million (United Nations 2005). In this single instance these rapidly obsolete
forecasts first lost 20 million and then found 10 million people within just
six years.
Failed forecasts can also bring good news. For example, an unexpectedly rapid decline of Iran's total fertility rate (it fell from more than 6.5
duringthe early 1980s, the firstyears afterthe fundamentalistmullahs overthrew the Pahlavi dynasty, to about 2.5 by the late 1990s) forced the UN
forecasters to lower the country's 2050 level to 1.85, that is, well below
replacement. And a comparison of the 2004 revision with the UN's 1990
forecast (United Nations 1991) that ended in the year 2025 shows the difference of about 600 million people, the reduction about 10 percent greater
than today's entire population of Latin America. Thus, even forecasts that
deal with given biophysicalrealities (most of the females who will give birth
during the next 20 years are already alive) and that are issued only a dozen
years apart can differ by continent-size margins. I have no desire to add to
this almost instantly irrelevant mountain of specificpoint forecasts.
Nor do I want to become an inventive fabulistand profferassortedscenarios, a practicethat is now so popularamong think tanks, long-term strategy departmentsof major corporations,and consultants of every stripe.2Although a few of these tales make interesting reading, most of them are
memorablebecause of their catchy titles (CivilizationsCollide;Uneasy Peace;
Biotech Bonanza) ratherthan because of their content. The principalreason
why even the cleverestand the most elaborateexplorativescenariosare ultimately so disappointingis that they may get some components of future realities approximatelyright, but they will inevitably miss other ingredients
whose dynamic interactionwill createprofoundlyalteredoutcomes.
Supposethat in 1975 (yearsbefore the adoptionof the one-child policy)
a group of scenario writers correctly forecast China's population total in
2005. It is a safe conclusion that no group, however learned and imaginative, would have set that number (during the last phase of the Maoist Cultural Revolution) in the context of a more than quadrupledquasi-capitalist
economy that is absorbingtens of billions of dollars of direct foreign investment annually, is now the prime workshop for the world, and is the second
largestbuyer of US government debt.
I offer no quantitative point forecasts, then, no alternative scenarios:
instead, my intent is to explore those key variables (whether physical, so-
VACLAV
SMIL
203
cial, or economic) whose impact is likely to be large enough to shape the

course of world history during the next 50 years. My firm belief is that
looking far ahead is done most profitably by looking far back. Naturally,
there are no specifics to be learned from such an exercise, yet those extended retrospectives impress with one key lesson: history advances as
much by saltations-sudden discontinuities-as it does by a gradual unfolding of long-lasting trends. In this respect history mirrors, in a much
contracted fashion, the record of life's evolution on Earth that is marked
both by very slow (Darwinian) transformations and by relatively sudden
(discontinuous) changes (Simpson 1983; Eldredge and Gould 1972).3 As
just explained, the increasingly frequent attempts at long-range projections (ranging from dynamic modeling to scenario building) are of a gradualistic variety, resting largely on following and embellishing a number of
critical trends.
A later essay will examine a number of these gradual, evolutionary
processes that may profoundly influence the size of human populations and
the basic contours of modern history. Its main topics will include the new
population realities (resulting mainly from differential growth, regional redistributions, aging, and migration), socioeconomic trends with capacity for
long-lasting global impacts (marginalization of Japan, Islam's role, Russia's
possible reemergence as a major power, China's rise and its checks), changing global leadership (resulting from America's likely retreat), and worrisome environmental trends (dominated by relatively rapid global warming
and the unprecedented loss of biodiversity). Here I focus on those unpredictable discontinuites whose consequences-in terms of lives lost and disrupted, economies destroyed and transformed,outlooks dashed and alteredcould change humanity's collective fortunes during the next 50 years.
Fatal discontinuities
These events fall into three broad categories.4 In the first one are known
catastrophic risks, events whose probabilitiescould be assessed in meaningful ways given the reasonably well-known biophysical realities and historical precedents. Both their near- and long-term recurrence can be quantified in a revealing, in some cases even satisfactory(though always imperfect),
manner. This category includes such disparateevents as the Earth'sencounters with large extraterrestrialbodies and extraordinarilyvirulent pandemics.
In the second category are the events that have never taken place,
whose likelihood thus eludes any meaningful quantitative assessments both
as concerns their occurrence (some have been widely anticipated for decades) and their impact but that should not be excluded from the assessment of future fatal discontinuities. Who can quantify the chances of an
accidental nuclear war or (a more recent worry) the probabilityof Pakistan's
204
nuclear arsenal ending up in the hands of jihadis? And who can then estimate the casualties of a launched missile and of a devastating retaliation?
The third category includes entirely speculative events as well as the
risks that remain completely unknown. A clear example of the former is
Bill Joy's vision of new omnivorous "bacteria"capable of reducing the biosphere to dust in a matter of days (Joy 2000). Obviously, no one can give
examples of the latter, but the likelihood of such unknowable surprisesincreases as the time span under consideration lengthens.5 Still it is worthwhile to comment on key speculative unquantifiable risks and assign them
to two basic categories of more and less worrisome events: this division can
be based on the best relative ranking of (guess)timatedprobabilities,on the
most likely overall impact of such developments, or, best of all, on the combination of these two factors.
Firstthe discontinuities whose very occurrence remains purely speculative and that, as many criticswould argue, should be more appropriately
relegated to the realm of science fiction. The rationale for addressingthese
matters was aptly describedin Tom Wolfe's (1968) bemused description of
the way America's business leaders of the late 1960s reacted to the quasiprophetic (and often utterly incomprehensible) statements of Marshall
McLuhan:what if he is right? Several of these highly speculative concerns
were widely popularizedby Bill Joy's (2000) lengthy paper about the danger of three powerful twenty-first century techniques-robotics, genetic engineering, and nanotechnology-that are threatening (fairlyimminently?)
to make humanity redundant or outright extinct.
Joy's overblown, curiously self-absorbed,and poorly conceived piece
was largely a derivative effort based on the work of two artificial intelligence enthusiasts, Hans Moravec (1999) and Ray Kurzweil (1999), who
maintain that robotic intelligence will soon rival human capability. Joy's
most sensational claim was reserved for the aforementioned claim about
new omnivorous microbesthat could swiftly dispose of the entire biosphere.
Joy might have been less agitated had he acknowledged some fundamental
ecological realities and considered the necessary resource and interspecific
competition checks on such a runaway scenario:microorganismshave been
around for some 3.5 billion years and evolutionary biologists have difficulty envisaging a new one that could do away, almost instantaneously,
with all other organismsthat have survived, adapted,and prosperedagainst
such cosmic odds.6
As for the armies of superintelligent,omnipotent robots, we have been
promised their advent for several generations ((apek 1921; Hatfield 1928).
There are no such machines today; even the most "intelligent"software
installed in IBM'sDeep Blue II in order to beat GarryKasparovin 1998 did
not show the coming triumph of machines but merely, as John Casti
(2004:680) noted, that "world-classchess-playingcan be done in ways com-
VACLAV SMIL
205
pletely alien to the way in which human grandmasters do it." And while
computers have been used for many years to write software and to assemble
other computers and numerous machines, such deployments do not indicate any imminent self-design and self-assembly capability, as all of those
processes require human actions to initiate them, raw materials to build
the hardware, and, above all, energy to run them. I find it hard to visualize
how those machines (particularlyin less than a generation) would launch,
integrate, and sustain the exploration, extraction, conversion, and delivery
of the requisite multitude of energies.
In any case, there is little we can do about the frightening (or liberating:
no human worries anymore) events in Joy-Kurzweil-Moravec visions. If the
emergence of superior machines or all-devouring nanospecies is only a matter of time, then we are just passive spectators waiting to be eliminated. If
such developments are possible but not certain, we have no rational way to
assess the risk:is there a 75 percent or 0.75 percent chance of self-replicating
robots taking over by 2025 or of nanobots being in charge of the Earth by
2050? And if such threats are more than pretentious, upscale science fiction,
then they have a massive amount of company in print, film, and television
and are good for little more than their intellectual frisson effect.
By far the most catastrophic unquantifiable risk is the possibility of
accidental nuclear war, a fear that has been with us since 1951.7 During the
height of the Cold War, casualties of an all-out thermonuclear exchange
between the two superpowers (including its aftermath) were estimated to
reach hundreds of millions (Coale 1985), and on several occasions we came
perilously close to such a civilization-terminatingevent. Nearly four decades
of the superpower nuclear standoff were punctuated by a significant number of accidents that involved nuclear submarines and long-range bombers
carrying nuclear weapons, and by hundreds of false alarms caused by malfunctions of communication links, by errors of computerized control systems, and by misinterpretation of remotely sensed evidence. Many of these
incidents have been detailed after a lapse of time in Western publications
(Sagan 1993; Britten 1983; Calder 1979), and there is no doubt that the
Soviets have had a similar (and likely even larger) number of such momentarily terrifying experiences. Probabilitiesof such mishaps escalating out of
control rose considerably during the periods of heightened crises, when a
false alarm was much more likely to be misinterpreted as the beginning of a
thermonuclear attack.
A series of such incidents took place during the most dangerous moment of the Cold War, the October 1962 Cuban Missile Crisis (Blight and
Welch 1989; Allison and Zelikow 1999). On 24 October a Soviet satellite
exploded shortly after reaching its orbiting height; on 25 October a sabotage alarm triggered by a bear led to the takeoff of nuclear-armed F-106A
interceptors; on 26 October a US U2 spy plane strayed into Soviet airspace
206
above Chukotka;on 28 Octobera New Jersey radarstation mistook a satellite pass for a missile launch as did a newly activated radar site in Laredo,
Texas. The outcome of all of these incidents is well known: there was never
any accidental launch, either attributableto hardware failure or to misinterpreted evidence. One of the architects of the (in retrospect remarkably
stable) Cold War regime in the United States concluded that the risk was
small because of the prudence and unchallenged control of the leaders of
the two countries (Bundy 1988).
How small depends entirely on the assumptions made in order to calculate cumulative probabilitiesof avoiding a series of catastrophicmishaps.
Even if the probabilityof an accidental launch were just 1 percent in each
of some 20 known US incidents (i.e., the chance of avoiding a catastrophe
being 99 percent), the cumulative likelihood of avoiding an accidental
nuclear war would be about 82 percent, or, as Philips (1998: 8) rightly concluded, "aboutthe same as the chance of surviving a single pull of the trigger at Russian roulette played with a 6 shooter." But these are meaningless
calculations. As long as the time available to verify the real nature of an
incident is shorter than the minimum time needed for a retaliatorystrike,
the latter course can be avoided and the incident cannot be assigned any
definite avoidance probability:if the evidence is initially interpreted as an
attack underway but a few minutes later this is entirely discounted, then in
the minds of decisionmakers the probability of avoiding a thermonuclear
war went from zero to 100 percent within a brief span of time.8
The demise of the Soviet Union undoubtedly diminished the chances
of accidental nuclear war thanks to a drastic reduction of total warheads
deployed by Russiaand the United States:in January2005 Russiahad some
7,200 warheads comparedto its peak total of 45,000 in 1986, and the United
States had about 5,300 warheads compared to its peak of 32,500 in 1967.
Further cuts lie ahead: the StrategicOffensive Reductions Treaty signed in
May 2002 will reduce total warheads to fewer than 2,200 on either side by
the year 2012 (Norrisand Kristensen 2005a and 2005b). But Forrow et al.
(1998) argued that because of the aging of Russian weapons systems, the
risk of an accidental nuclear attack had actually increased. They also calculated that an intermediate-sized launch of warheads from a single Russian
submarine would kill nearly instantly about 6.8 million people in eight US
cities and expose millions more to potentially lethal radiation.
Moreover, with more countries possessing nuclear weapons, it is possible to argue that chances of accidental launching and near-certainretaliation have been increasing steadily: since 1945 an additional nation has acquired nuclear weapons roughly every five years, North Korea being the
latest entry and Iran the next most likely candidate. The North Korean case
introduces another dimension of the unknown since we cannot be certain
whether we are dealing, as many facts indicate, merely with an exceed-
VACLAV SMIL
207
ingly devious leadership that is engaged in a risky game of serial extortion

or with men who are willing to pull down their peninsular temple on their
own heads as they launch a suicidal attack on the United States (Albright
and O'Neill 2000; Cha and Kang 2003).
Terrorismhas been a recurrent form of unconventional warfare in the
modern world, and murder by suicide has deep roots in Muslim history
(Andriolo 2002), but only the attacks of September 11, 2001 elevated this
kind of violence to the class of catastrophic events whose recurrence could
profoundly change history. In the wake of that attack, modalities and consequences of possible future terrorist actions have been explored-at once
excessively yet insufficiently, with exaggeration as well as with inadequate
appreciation of possibilities-but their probability remains beyond any useful quantification. Their list ranges from launching cyber attacks on modern electronic infrastructures and poisoning of urban water or food supplies to the decapitation of a national leadership and, of course, explosion
of a dirty bomb and releases of old or new pathogens. Public policies and
precautionary actions dictate that none of these incidents, no matter how
low its probabilitymight seem, can be simply dismissed as too unlikely. But
this very reality makes it highly difficult to manage the challenge because
we are unable to assess the relative likelihood of many different modes of
attack.
A skeptical appraisal must point out at least two facts: many of these
attacks are not as easy to launch as the sensationalizing media coverage
would lead us to believe, and many of them, even if successfully executed,
would have relatively limited consequences and would not rise to the level
of developments capable of transforming history.9 This is particularly true
about the attacks that Homer-Dixon (2002) defined as ingredients of complex terrorism:actions that are much easier to launch than spectacularmass
murders but that might prove ultimately both costlier and deadlier. Terrorists would exploit growing complexities of modern societies and deploy their
"weapons of mass disruption"by attacking nonredundant nodes of critical
infrastructures, be they electricity networks, other energy supply systems,
chemical factories, or communication links.
A skeptical riposte to this scenario is to ask, given that many of these
attacks are so trivially easy to launch, why we are not seeing scores of them
every month. After all, it is impossible to safeguard against explosives every
one of hundreds of thousands of steel towers that carry a large nation's
high-voltage lines, or to protect round-the-clock every one of tens of thousands of transforming substations (or, facing a different threat, to detect
every poisoned kilogram among millions of tonnes of harvested crops or to
secure thousands of reservoirs and river water intakes that furnish drinking
water). A relatively rich experience with accidental large-scale electricity
supply outages (caused by weather, human error, or a technical problem)
208
demonstratesthat similar,or even larger,failureswould not rise to the level

of historicmilestones.'0Indeed, the use of weapons of mass disruptionwould
be most likely to amount to a spatially and temporarilylimited (albeit possiblyfairlyexpensive) disruption,remainingfarbelow the thresholdof events
capable to appreciablymodify the course of global history.
Those who argue that a greater complexity generates increased vulnerability should also keep in mind that it also brings greater resilience.
Evolution of complex life is based on this very principle:otherwise the terrestrialbiosphere would consistjust of simple bacterialmats that would preclude the emergence and complexification of climactic grasslandsand forests. And the success of our species makes it clear that humans, unlike all
other organisms, have evolved not to adapt to specific conditions and tasks
but to cope with change (Potts 2001). This ability makes us uniquely fit to
cope with assorted crises and to transform many events from potentially
crippling milestones to resolved challenges. This capacity would be most
severely tested by a nuclear war, whether through a failure of national controls, a madly deliberatelaunch by an irrationalleadership, or a takeover of
nuclear weapons by a terroristgroup: the unknown risk of this catastrophe
may be very low but no other anthropogenic act comes close to causing as
many instant casualties.
Threatsfrom the weapons of mass disruption (includingany truly dangerous deployment of pathogens) would rank much lower than a nuclear
holocaust, and I would relegate the prospects of life crippled by entirely
new pathogens and devouredby nanobots for science fiction accounts.Earlytwenty-first-centurysocietieswill have to live with all of these unquantifiable
threats, and the risk of terroristattacks will be particularlyprominent owing to media attention and the nature of the American political process.
But we would be much better off if we were to treat this threat with broad
statisticalrisk assessments in mind. Even if unavoidable, many of these attacks represent much lower dangers (be it from an individual's or a
collective's perspective) and have less profound and long-lasting consequences from the standpoint of national stability, economic damage, and
standardof living than do a number of voluntary risk exposures (drinking,
driving, smoking, overeating) and deliberate yet deleterious policy actions
(rangingfrom enormous budget and tradedeficitsto subsidiesand unchecked
environmental destruction).
Indeed, Chapman and Harris(2002) argued that the disproportionate
reaction to the attacks of 9/11 was as damaging as the direct destruction of
lives and property. Relative risks are best shown in terms of fatalities per
person per hour of exposure, a risk assessment approach that was originally developed by Chauncey Starr(1969). I have used it to calculatea number of relevant US rates for a 15-year period between 1989 and 2004. Their
plot in Figure 1 shows the risk of dying because of terrorismin the United
States (including all 1993 and 2001 World Trade Center fatalities and the
VACLAV
209
SMIL
FIGURE1 US fatalities per person per hour of exposure plotted

against the average annual number of fatalities during the 15 years
between 1989 and 2004 (both axes are logarithmic)
10-12
10-11
* California earthquakes
* blizzards
10-o
* terrorism (US only)
* terrorism (US citizens)

0
(L
0
10-90
s1
C-.
(
10-8
homicides
._
.(
* falls
motor vehicle accidents

A
10-6
10-55
10?
II
II
II
II
102
10
103
104
II
o05
smoking
* all mortality
II
106
II
107
108
Fatalities / year
SOURCES:
Calculatedfromdatain CDC(2005) and froma varietyof reportson naturalcatastrophesand
terroristactivity.
1995 fatalities from the Oklahoma bombing) to be only slightly higher than
the risk of dying in a blizzard somewhere in the country. When the deaths
of all US citizens attributable to terrorism are included-those from 1996
Saudi Khobar Towers and 1998 East African embassies bombing, from the
Yemeni attack on the USS Coleand even from combat in Afghanistan and
Iraq-the rate about doubles to 2x10-~0,still an order of magnitude below
the annual risk of homicide and three orders of magnitude below the annual risk of fatal car accidents.
210
During the first five years of the twenty-first century the US highway
death toll exceeded the 9/11 fatalities every single month; at times it was
higher in just three weeks. And even one of the worst cases from recently
leaked terroristattack scenarios preparedby the Department of Homeland
Security (Jakes 2005) does not imply an extreme risk. Sprayingof anthrax
from a truck driving through five cities over two weeks was estimated to
kill 13,200 people; if these actions were to take place in metropolitan areas
with populations of at least 2 million people each and be repeated every
ten years, then even such an unlikely recurrence (and high fatalityassumptions) would prorateto only 1.5x10-8 fatalities per person per hour of exposure, a risk lower than the risk of dying from an accidental fall and less
than 1/30 of the risk due to driving. Many more deaths could be prevented
while spending much less per life saved by investing in extreme weather
education, safer stairways, and enforcement of lower speed limits and nodrinking-and-drivinglaws. But disproportionatereactions to fatalities arising from rare, involuntary, and spectacularrisks will not be changed easily.
The rest of this essay is devoted to those unpredictablediscontinuities
whose comparative probabilitiescan be quantified with a degree of accuracy that is useful for assessing relative risks and allocating resources for
preventive actions or for eventual mitigation. Three kinds of such events
fulfill the combined requirement of an extraordinarymagnitude, global impact, and long-term consequences: natural catastrophes,pandemic illness,
and transformationalwars. I will look only at those natural catastrophes
that do not have a vanishingly low probabilityduring the coming 50 years,
that is, those that recur with intervals no longer than 100,000 years and
that could shape world history because of their directly lethal effects or the
long-term changes they might inflict on the biosphere.
This is why I leave aside such very rare events as the Earth'sexposure
to supernova explosions, its head-on collisions with massive comets (which
are much less common than encounters with nearby asteroids), or periods
of enormous lava flows (such as those that created India's Deccan Trapsby
piling up about one million km3 of basaltic lavas over a period of about 5
million years beginning some 65 million years ago). And because neither
an abrupt climate change nor an altered intensity of ocean circulation (two
events that have been commonly posited as potentially major environmental crises of the coming decades) can unfold as rapidly as the natural
discontinuities examined in this essay, I will deal with them in the aforementioned subsequent essay on globally important changing trends.
Natural catastrophes
Recurrent natural catastrophes can claim hundreds of thousands, even millions, of lives per event. The most recent case in this category, the Decem-
VACLAV
SMIL
211
ber 2004 Indian Ocean tsunami, was a convincing illustration of the fact
that while these catastrophes may elicit worldwide humanitarian response,
they do not alter the course of world history." Indeed, one of half a dozen
similarlydevastatingevents that took place during the latter half of the twentieth century remained an entirely internal affair as xenophobic China, just
seven weeks before Mao Zedong's death, did not ask for international help
following the Tangshan earthquake of July 1976 that killed, officially,
242,219 people in that coal-mining city and in its surroundings but whose
toll was estimated to have been as high as 655,000 (Huixian et al. 2002;
Chen et al. 1988).
In contrast to frequent natural disasters, there are only three kinds of
sudden, unpredictable, but recurrent natural events whose global, hemispheric, or large-scale regional impacts can have profound influence on the
course of history: the Earth's collisions with nearby extraterrestrialobjects,
massive volcanic eruptions (some also generating major tsunami), and (possibly) voluminous, tsunami-generating slides of parts of volcanoes into the
ocean. The probability of any of these events is very low during the coming
50 years, but this comforting appraisalmust be counterbalanced by the fact
that if any one of them were to take place it would be an event without
counterpart in recorded history: the near-instant death toll would reach 1
million to 10 million people, one to two orders of magnitude higher than
for high-frequency natural catastrophes.
The Earth constantly passes through highly dispersed (but in aggregate quite massive) amounts of cosmic debris whose most common sizes
range from microscopic particles to meteoroids with diameters of less than
one meter. This constant infall poses virtually no risk to life, but the planet
is also periodically hit by much larger objects whose size and energy determine the consequences of such impacts. Hills and Goda (1993) calculated
that stone objects with diameters of up to about 150 m will release most of
their energy in the atmosphere and will not hit the surface and create impact craters. Bland and Artemieva (2003) increased the size of objects that
produce only air blast to 220 m in diameter. Largerobjects (and also heavier,
but much rarer, iron-nickel bodies) will strike the surface, but the odds are
roughly 7:3 that they will hit the ocean and will damage the land only indirectly by generating tsunami.
The destructive radius of ground impacts and the amount of debris
lifted into the atmosphere would be the function of the size and speed of
the incoming objects. Very large bodies (at least one km in diameter) would
produce longer-term global climate change regardless of the point of their
impact. Recent research has produced more reliable accounts of the numbers of near-Earth objects and of the probabilities of their collision with the
Earth. Rabinowitz et al. (2000) used improved detection techniques to conclude that there are nearly 1,000 near-Earth objects with diameters in excess of one km, about half the total of earlier estimates. Stuart (2001) nar-
212
rowed the total number of kilometer-sized near-Earthasteroidsto just over

1,200 and also found them less likely to collide with the Earth.
Brown et al. (2002) used the US Department of Defense and Department of Energy satellite designed to detect nuclear explosions in order to
identify light records of detonations of objects in the 1 to 10 m size range
in the atmosphere. From these observations they concluded that, on average, the Earth is struck by an object with energy equivalent to 10 Mt TNT
(about 50 m in diameter) every 1,000 years, about five times less frequently
than thought only 20 years ago (uncertainties widen the frequency range
to between 400 and 1,800 years). The probabilityof such an impact is thus
about 5 percent (2.8-12.5 percent) during the next 50 years, and its effect
would be similar to that caused by the famous Tunguska meteor of 1908, a
stony object that disintegratedin the atmosphere, produced a shock wave
that flattened trees over an area of about 2,150 km2, but killed nobody
(Dolgov 1984).
A similar object, were it to disintegrate over a densely populated urban area, could cause great damage: after all, its explosion (at about 15 km
above ground) would release energy equivalent to 100 or so Hiroshima
bombs and the resulting blast wave would destroy many structuresand kill
many thousands of people.12 But the chances of such an event are roughly
two orders of magnitude smaller than the probability of its hitting an
unpopulated or thinly-inhabited region; and, as was clearly demonstrated
by the contrast of casualties in Hiroshimaand Nagasaki,the actual destruction would depend on the physical configuration of the affected area.'3The
Earth is also hit roughly once a year with an extraterrestrialbody whose
airburstis equivalent to 5 kt TNT,about a third as powerful as the Hiroshima
bomb, and if this body's center of disintegration were right above the US
Capitol during the President's State of the Union address the effect would
be felt globally.But the probabilityof such an encounter is vanishinglysmall,
at least eight ordersof magnitude smaller than that of a similarobject disintegrating any time above any densely populated area.
Impacts of global consequence would be produced only by asteroids
with diameters of at least one km and by much rarercomets whose return
orbits cannot be reliably calculated long in advance and whose threat is
only about 1 percent of that of near-Earth objects. The good news is that
the assumed probabilityof their near-term impacts has been decreasing:by
2003 the international telescopic SpaceguardSurvey had already identified
about 55 percent of all such objects with a diameter in excess of one km,
and it found that none of them is on a trajectory that would lead to its
collision with the Earth during the twenty-first century (NEOSDT2003).
Once the survey is complete (by the end of 2008) the residual hazard (averaged over a very long period of time) might be only on the order of 100
annual fatalities worldwide (that is, causing 100,000 fatalities once every
VACLAV SMIL
213
1,000 years). The greatest land-impact hazard will come from smaller bodies (with more than a 1 percent chance that such an impact will kill about
100,000 people during the twenty-first century) while somewhat largerbodies (150 to 600 m in diameter) will pose the greatest tsunami hazards
(Chapman 2004).
The general size-frequency distribution of near-Earth objects is now
fairly well known (Figure 2), but, not surprisingly, there are different assumptions regarding the most likely frequency of impacts. Ward and
Asphaug (2000) assume that an object with a diameter of 400 m hits the
Earth once every 10,000 years and that a one-km body hits once every
100,000 years. In contrast, Brown et al. (2002) would expect a 400-m body
every 100,000 years and a one-km object every 2 million years; Chapman
(2004) would have the 400-m body hit once in a million years; and Jewitt
(2000) opted for 400,000 years. If an asteroid with the diameter of 400 m
were to hit the ocean at 20 km/second, the maximum amplitude of a tsunami generated by this impact would be about 50 m at the distance of 100
km (and nearly 250 m only 20 km away): a near-shore impact offshore
California or off eastern Honshu would thus instantly devastate a core region of one of the world's two leading economies and, unlike with tsunami
generated by a distant earthquake, it would not give sufficient time for mass
evacuation of affected regions.
Global probabilitiesof an impact of this magnitude are, depending on
the recurrence interval, 0.05 to 0.5 percent, and Ward and Asphaug (2000)
calculatedspecificprobabilitiesof a 5-m tsunami wave hitting Tokyo and New
Yorkat, respectively,4.2 percent and 2.1 percent duringthe next 1,000 years,
or roughly 0.2 percent and 0.1 percent during the next 50 years. In contrast,
Bland and Artemieva (2003) estimate the frequencies of bolides that would
be most likely to cause hazardous tsunamis at only about 1/50 of the rate
calculated by Ward and Asphaug. Chesley and Ward (quoted in NEOSDT
2003) concluded that the highest risk of tsunami-generated fatalities comes
from smaller but more frequent impacts of objects with diameters of 200 to
400 m and calculatedthe overalllong-term casualtiesat fewer than 200 deaths
a year (or fewer than 10,000 during the next half century).
The impact energy of a one-km body would be equivalent to the release of close to 100 Gt TNT,nearly an order of magnitude more of energy
than would have been expended by an all-out thermonuclear war between
the two superpowers in 1980 (Sakharov 1983). If its average recurrence
interval were 400,000 years, then the probability of an impact during the
next 50 years would be 0.0125 percent (with available data justifying an
unhelpfully wide range of 0.002 to 0.05 percent). If it were to enter the
ocean, such an impact would obviously generate tsunami that would hit
even distant shores with high-amplitude waves, and the principal global
effect with a continental impact would be due to an immense mass of shat-
214
THE NEXT 50 YEARS:
FATAL DISCONTINUITIES
FIGURE2 Size and frequency distribution of near-Earth asteroids (all

three axes are logarithmic). Band in the left half of the figure
indicates the range of uncertainty regarding the numbers and impact
intervals of objects with diameters less than one km.
Log impact energy (Mt TNT)
-3
-1
-2
-1
l 0I
10
10
1-
4-
50
6-
7-
8-
9
0.01
0.1
10
Diameter (km)
SOURCE:Based on a graph in NEOSDT (2003).
tered material that would be lifted high into the atmosphere, resulting in a
drastic drop of temperature, extensive deposits of dust, and long-term reduction of plant productivity.
Unusually large volcanic eruptions are the only other category of natural phenomena whose frequencies and global effects are similar to those of
major impacts of extraterrestrial objects. The largest one during the Quaternary period was the megaeruption that created the giant Toba caldera
(an oval roughly 30 by 100 km filled by a lake) in northern Sumatra about
75,000 years ago (Rose and Chesner 1990): it produced about 2,800 km3 of
ejecta compared to about one km3 during the Mount St. Helens eruption in
May 1980 (Lipman and Mullineaux 1981). Trillions of metric tons of volcanic ash were deposited thousands of kilometers downwind, and volcanic
aerosols that persisted in the stratosphere drastically lowered atmospheric
VACLAV SMIL
215
temperatures and plant productivity. This catastrophe is perhaps the best

explanation for a genetically well-documented late Pleistocene population
bottleneck when small and scattered groups of humans were reduced to a
global total of fewer than 10,000 individuals and when our species came
very close to ending its evolution (Rampino and Self 1992; Ambrose 1998).14
Judging by the timing of two previous Toba events, the probability of
a similar eruption is about 0.01 percent during the next 50 years, but for
North America a more direct threat is presented by recurrent eruptions of
the Yellowstone hotspot (Smith and Braile 1994). Eruptions of this
supervolcano left behind nine massive calderas during the last 15 million
years; the last one, about 630,000 years ago, produced about 1,000 km3 of
volcanic ash. The frequency of the Yellowstone hotspot eruptions actually
increased during the past 2 million years with the last two intervals being
about 800,000 and 650,000 years, and so another event during the next 50
years has a probabilityof about 0.007 percent. If it were to produce as much
ash as the last eruption, then it could bury about 2 million km2of America's
heartland (roughly a quarter of the country's territory, including all of the
leading wheat-producing states) under half a meter of ash.'5
Even a much smaller volcanic event could have enormous consequences if it were to cause a massive collapse of volcanic flanks into a nearby
ocean and hence generate an extraordinarilylarge tsunami. Such huge landslides, recurring roughly once every 100,000 years and creating waves in
excess of 100 m, were first documented in the early 1960s in the form of
massive hummocks of debris on the seafloor surrounding the Hawaiian Islands (Moore, Normark, and Holcomb 1994). Perhaps the greatest risk of
this kind is posed by a future eruption of the Cumbre Vieja volcano at La
Palma in the Canary Islands that could cause a catastrophic failure of the
mountain's western flank (Ward and Day 2001). The resulting landslide of
up to 500 km3 of volcanic rock would generate a rapidly moving (up to 350
km/h) mega-tsunami that would, even after crossing the Atlantic, hit the
eastern coast of North America with repeated walls of water up to 25 m
high (Figure 3).
The best evidence regardingthe frequency of CanaryIslands landslides
would indicate intervals of about 70,000 years and hence a probability no
higher than 0.07 percent during the next 50 years. But a 500 km3landslide
is the worst-case scenario that also assumes that the slip would take place
instantaneously during a single event and enter the sea at high velocity.
Halving the failed mass would reduce the maximum waves to 5 to 10 m;
and if, as argued by Wynn and Masson (2004) on the basis of their studies
of offshore deposits, each landslide were to be composed of multiple stages
of gradual failure, then the average collapsed mass could be as low as 10 to
25 km3 and it would be much less likely to generate tsunamis that could
inflict severe damage to the east coast of North America.
216
THE NEXT 50 YEARS:
FATAL DISCONTINUITIES
FIGURE3 An unlikely, but potentially devastating natural

catastrophe: a massive collapse of Cumbre Vieja in the Canary Islands
would generate a tsunami that would hit the eastern coast of North
America with a sequence of waves mostly between 10 and 25 m high.
Tsunami progress across the Atlantic would allow for ample
warning, and a staged collapse of a volcanic flank would produce
much smaller trans-Atlantic waves.
SOURCE:Based on results of a simulation exercise by Ward and Day (2001).
Viral pandemics
Modern hygiene, nationwide and worldwide inoculation, constant monitoring of outbreaks, and emergency vaccinations have either completely
eliminated or drastically reduced a number of previously lethal, deeply injurious, or widely discomforting epidemic diseases: alphabetically they include cholera, diphtheria, pertussis, polio, smallpox, tuberculosis, and typhoid.16 At the same time, new infectious diseases keep emerging: recent
decades have seen variant Creutzfeld-Jacob disease (the human form of bo-
VACLAV SMIL
217
vine spongiform encephalopathy commonly known as mad cow disease),

cryptosporidiosis, cyclosporiasis, hantavirus pulmonary syndrome, SARS,
and, of course, HIV/AIDS(Morens, Folkers, and Fauci 2004). None of these
new threats, with the obvious exception of HIV/AIDS, appears capable of
changing the course of world history, and AIDS could do so only if new,
much more virulent strains were to afflict significant shares of populations
outside sub-Saharan Africa.'7
As far as the unpredictable discontinuities are concerned, only one somatic threat trumps all of this: we remain highly vulnerable to another episode of viral pandemic. High-frequency natural catastrophes have their somatic counterpart in recurrent epidemics of influenza, an acute infection of
the respiratorytract caused by type A and B viruses belonging to the family
Influenza epidemics sweep the world annually, mostly
Orthomyxoviridae.
during the winter months, but with different intensities. In the United States
between 250,000 and 500,000 new cases are reported every year, about
200,000 people are hospitalized, and 20,000 people die (less than 0.01 percent of the US population). Infection rates are by far the highest among
young children (10 to 30 percent annually) and in people over 65 years of
age (Centers for Disease Control 2000). Influenza pandemics have been
caused only by type A viruses when one of their subtypes, different from
strains that are already present in humans, suddenly emerges, rapidly diffuses around the world (usually within six months), and afflicts between
30 and 50 percent of people.
The illness, with its characteristicsymptoms of fever, headache, cough,
debilitation, and pain, is often complicated by bacterial or viral pneumonia.
The former can be now treated by antibiotics, but because there is no treatment for the latter it becomes a common cause of death during influenza
epidemics. The firstfairlywell-documented influenza pandemic was in 1580,
and six known episodes occurred during the last two centuries (Gust,
Hampson, and Lavanchy 2001): in 1830-33 (subtype unknown, originating in Russia), in 1836-37 (subtype unknown, originating possibly in Russia), in 1889-90 (subtypes H2 and H3, originating possibly in Russia), in
1918-19 (subtype HI, Spanish flu, originating simultaneously in Europe
and in the United States), in 1957-58 (subtype H2N2, Asian flu, originating
in Southeast Asia, total excess mortality of more than 2 million), and in
1968-69 (subtype H3N2, Hong Kong flu, originating in Hong Kong, with
excess mortality on the order of one million). The low death rate during
the last pandemic is attributable to protection that was conferred on many
people by the 1957 infection.
All of the nineteenth-century pandemics, as well as the 1957 and 1968
events, were relatively mild and hence did not make any noticeable upticks
in the secular trend of declining mortality. In contrast, the 1918-19 pandemic was by far the largest sudden infectious burden in modern times. Its
218
first wave began in February 1918; by May it had spread throughout most
of the United States, Western Europe, north Africa,Japan, and the eastern
coast of China; by August it was in Australia, Latin America, and India
(Patterson and Pyle 1991). Recent studies of this pandemic include books
aimed at a general readership (Davies 1999; Kolata 1999; Barry 2004) as
well as at experts (Phillips and Killingray2003). Scientific advances of the
1980s (polymerase chain reaction, permitting replication of genetic material) made it possible to identify the virus that was initially retrieved from
formalin-fixed,paraffin-embeddedlung tissue samplesand used to sequence
fragments of viral RNA (Taubenbergeret al. 1997). Perhapsthe most interesting finding from these studies is the remarkableuniformity of the pathogen, with strains showing 98.9 to 99.8 percent nucleotide sequence identity (Reid et al. 2003).
Statisticalanalyses of the best available data confirm a peculiar mortality pattern: in contrast to annual epidemics characterizedby a typical Ushapedmortalitypattern,the 1918-19 pandemickilledpredominantlypeople
between ages 15 and 35 years, and 99 percent of all deaths were among
people younger than 65 years (WHO 2005). Many of these deaths were
due to viral pneumonia that caused extensive hemorrhaging of the lungs
with death taking place within 48 hours. But there is little certainty about
the most relevant figure of all, the total global death toll: perhaps the most
commonly cited worldwide aggregate has been between 20 million and 40
million, but the latest World Health Organizationdocument refers to "upwards of 40 million people" (WHO2005) and the highest estimates are between 50 and 100 million. Even the lowest estimate is higher than all military and civilian casualties of World War I (about 15 million); the highest
figure would about equal the uncertain grand total of fatalities among the
populations of the world's two largest Communist regimes of the twentieth
century: StalinistRussia and Maoist China (White 2003).
During the late 1990s, two decades after the last and relatively mild
pandemic, concerns arose over the emergence of new avian viruses transmissible to people. In December 1995 a meeting in Bethesda, Marylandon
pandemic influenza heard from one of the world's leading experts that "at
this time, there is no evidence for or against the direct spread of avian influenza viruses to humans" (Webster 1997: S18). By the time this presentation was published, the subtype H5N1 had mutated in Hong Kong's poultry
marketsto a highly pathogenicform (firstidentifiedin April 1997) that could
kill virtually all infected chickens within two days, and in May 1997 came
the first human death (Sims et al. 2002).
The virus was eventually transferredto at least 18 people, causing six
deaths and the slaughterof 1.6 million birds (Snacken et al. 1999). This episode showed for the first time that avian influenza viruses could infect humans directly,without passingthrough pigs or other intermediatehosts. Two
VACLAV SMIL
219
years later Hong Kong had two poultry-to-people transfersof subtype H9N2,
and startinglate in 2003 and during2004 the highly pathogenic subtype H5N1
began to appear again in poultry in at least eight Asian countries. Eventually
more than 200 people were infected with avian viruses (besides H5N1 also
H9N2, H7N7, and H7N3) in Hong Kong, Netherlands, Canada,Vietnam, and
Thailandwith nearly 40 deaths in Vietnam and Thailandcombined. The Thai
outbreakwas particularlywidespread, and it requiredthe slaughterof 40 million chickensin 41 provinces (Chotpitayasunondhet al. 2004). Studiesshowed
that domestic ducks in China's southern provinces are the key reservoir of
H5N1 viruses (Chen et al. 2004). Because these viruses are highly pathogenic
and have become ineradicablethroughout parts of Asia, they have a clearly
pandemic potential (Liet al. 2004).
Expert consensus may be wrong but it is certainly disconcertingto see
that epidemiologists and virologists are in general agreement about a very
high probabilityof pandemic influenza in the not too distant future-if not
in a matter of months or years, then likely within the next one or two decades. Optimisticforecastssee 20 percent of the world's population falling ill,
one in every 100 ill people requiring hospitalization (provided the beds will
be available), and 7 million deaths in a few months (St6hr and Esveld 2004).
But the morbidityrate may actuallybe 25 to 30 percent, and the WorldHealth
Organizationbelieves that a new pandemic may affect 20 to 50 percent of
the world's population. Its toll, however, cannot be responsiblypredictedbecause we have no way of knowing the virulence of new infectious strains.
What is certain is that the appearanceof subtype H5N1 has brought us closer
to the next pandemic, and that whatever its actual magnitude we are utterly
unpreparedfor it and for its consequences (WHO2005).
The following possibilitiesindicate the imminence and magnitude of the
risk. The recurrence interval, calculated simply as the mean time elapsed between the six known influenza pandemics, is about 28 years, with the extremes of six and 53 years. Adding the mean and the highest interval to 1968
gives us the span between 1996 and 2021: we are, probabilisticallyspeaking,
very much inside a high-risk zone. Consequently, the likelihood of another
influenza pandemic during the next 50 years is virtually 100 percent, but
quantifying probabilitiesof mild, moderate, or severe events remains largely
a matter of speculation because we do not known how pathogenic a new
virus will be and what age categoriesit will preferentiallyattack.
If the eventual death toll were to resemble those of the last two
pandemics, with a few million dead, there would be virtually no consequences for the long-term course of world history. If it were a highly similar repeat of the 1918-19 event, with mortality of no less than 20 to 25
million people, the global toll would be relatively only about one-fourth as
large as it was four generations ago. But the overall mortality could also be
a proportionally potentiated replica of 1918-19: we now have a 3.4 times
220
larger global population, an at least an eight times larger (nearly 20 billion

compared to less than 3 billion) inventory of poultry (the main reservoir of
lethal viruses, and much of it now in large feeding facilities), and we use an
order of magnitude faster means of mass transportation to spread the infections around the world. In 1918 it took six days to cross the Atlantic on an
ocean liner; now it takes six hours on a jetliner, and there is no doubt that
air travel plays an important role in the diffusion of annual epidemics (Grais
et al. 2004).
Even when assuming "only" 25 million deaths in 1918-19, we could
see a proportionally increased global mortality surpassing 80 to 100 million
people; with 50 million deaths in 1918-19 the proportional total would rise
to 150 to 200 million. With a slightly more than 5 percent mortality rate
(the well-documented US mean in 1918-19) there could be at least 1.5 to 2
billion very sick people, and it would be clearly beyond the capacity of health
services to cope effectively with such a sudden burden. On the other hand,
there are positive factors of generally better nutrition, much better hospital
care, and incomparably greater virological understanding. The overall enormity of this ubiquitous morbidity and vastly multiplied mortality would pose
challenges that have been unseen in most countries of the world for generations: in China since 1961 (the end of massive famine), in Europe (except for some Bosnian villages decimated by mass executions during the
1990s) since 1945, in North America since 1918.
In addition, many specific impacts would complicate our ability to deal
with immediate challenges and would have long-term consequences. Detecting a smaller herald wave of infections several months ahead of the main
event (as happened in 1918 in the United States) may not be helpful: rather
than give us more time to prepare, it may actually cause more helplessness
and fear because any development of a new vaccine that would begin only
once the pandemic virus started its diffusion could not be completed before
the virus would cover the world (Stohr and Esveld 2004). But the event
may not be over in six months, as substantial mortality could continue during the second season, and many cities and countries may find it particularly difficult to cope with such a second wave.
That was certainly the case with Toronto's second wave of SARS in May
2003, minuscule in terms of total numbers but extremely taxing due to mental burdens and logistic problems caused by quarantined hospitals (none but
emergency operations, no visits, not even for terminally ill patients). The mortality burden may shift, as it did so stunningly during the 1918-19 pandemic,
to younger people: age-specific mortalities in New York showed little or no
impact among people aged 45 years and older but an unprecedented spike
for younger cohorts (Figure 4). Repetition of this pattern could strain the
availability and effectiveness of caregivers (health professionals ranging from
physicians to staff at retirement homes) and substantially worsen dependency
221
SMIL
VACLAV
FIGURE4 Age-specific mortality in New York City in October 1918

during the peak of the influenza pandemic shows a W-shaped rather
than the common U-shaped (epidemic) profile.
10,000
8,000 -
6,000 -
4,000 -
2,000 -
0
0
10
20
30
40
50
60
70
80
Age (years)
Plottedfromdatain MonthlyBulletinof theDepartment
SOURCE:
ofHealth,CityofNewYork(November1918).
ratios, particularly in Europe's aging populations where the dependency ratios are already rising to unprecedented levels.
Massive mortality of people in their prime ages would also cause enormous strain on the life insurance industry and it would depress real estate
values. And what would 24-hour nonstop news media, so adept at flogging
a few accidental deaths to all-day marathons of despair and end-of-the-world
comments, do with so many deaths that would just keep coming, day after
day, week after week? Can we foresee how the financial markets would
react to such massive and indiscriminate dying? More importantly, what
would be the long-term economic cost in fear and depression that would
come on top of the immediate social and economic insults to the previously
insulated Western way of life? To what extent would Europe's ravaged countries become open to unchecked Muslim immigration, speeding up the demise of the continent's centuries-old global influence? What would the resulting cessation of global trade do to the lives of hundreds of millions of
factory workers in Asia?
Transformationalwars
While trying to assess the probabilities and consequences of recurrent natural
catastrophes and catastrophic illness, we must remember that the historical
222
record is unequivocal: these events, even when combined, did not claim as
many lives and did not change world history as much as the deliberatekind
of fateful discontinuities that RichardRhodes (1988) calls manmade death,
the single largest cause of non-natural mortality in the twentieth century.
Violent collective death has been such an omnipresent part of the human
condition that its recurrence, on wide-ranging scales of time (conflicts lasting days to decades)and casualties(fromindividualhomicidesto democides),
is guaranteed. Long lists of past violent events can be inspected in print
(Richardson1960; Singer and Small 1972; Wilkinson 1980) or in electronic
data bases (White 2003; International Institute of Strategic Studies 2003;
InternationalPeace ResearchInstitute 2004).
Even a cursory examination of this record shows yet another aspect of
that terrible toll, the fact that so many violent deaths had no, or perhaps
only a vaguely marginal,effect on the course of world history. Others,however, contributed to outcomes that truly changed the world. Large-scale
death tolls of the twentieth century that fit the first category include the
Belgian genocide in the Congo (that began before 1900), Turkishmassacres
of Armenians (mainly in 1915), Hutu killings of Tutsis (1994), wars involving Ethiopia (Ogaden, Eritrea, 1962-92), Nigeria and Biafra (1967-70), India and Pakistan (1971), and civil wars and genocides in Angola (startedin
1974), Congo (since 1998), Mozambique (1975-93), Sudan (1956 and perhaps finally finished in 2005), and Cambodia(1975-78). Even in our greatly
interconnected world, such conflicts can cause deaths of more than one
million people (as did all of the just-listed conflicts, with the highest estimated death tolls for a few of them exceeding 2 million), and some of them
can go on for decades, without having any noticeable effect on the cares
and concerns of the remaining 98 to 99.9 percent of humanity.
In contrast, the modern era has seen two world wars and a number of
interstate conflicts that resulted in long-lasting redistributionof power on a
global scale. And there were also intrastate wars that led to the collapse or
emergence of powerful states. There is no canonic list of such transformational wars of the nineteenth and twentieth centuries. While the key events
are obvious, I also readily concede that a more liberal definition of (indirect) worldwide impacts could extend their list. A long-term transformational effect is the key criterion for inclusion in this category, but all but
one of the conflicts I have selected also share another commonality: they
could be labeled as megawars, claiming the lives of more than 1 million
combatants and civilians. Or, according to Richardson's (1960) definition
based on the decadic logarithm of total fatalities,all but one would be magnitude 6 and 7 wars. Their fairly restrictive enumeration starts with the
Napoleonic wars (begun in 1796 with the conquest of Italy) that ended in
1815 in a refashioned-and for the next 100 years also a remarkablystableEurope.18
223
SMIL
VACLAV
The next entry on my list (see Figure 5) is the protracted Taiping war
(1851-64), a massive millennial uprising led by Hong Xiuquan (Spence
1996). This may seem a puzzling entry to readers not familiar with China's
modern history, but the Taiping uprising exemplifies a grand transformational conflict because it was its resolute challenge (rather than the British
victory in a short war of 1842 and the acquisition of Hong Kong) that fatally undermined the ruling powers of the Qing dynasty (at that time the
ruler of the world's largest economy), enmeshed foreign actors in China's
politics for the next 100 years, and brought, in less than two generations,
the end of the old imperial order (and cost more lives, about 20 million,
than the aggregate combatant and civilian losses of World War I).
Conversely, the American Civil War (1861-65) should be included because it opened the way to the country's rapid ascent to global economic
primacy. America's gross domestic product surpassed that of Great Britain
by 1870; by the 1880s the country had become the technological leader
and the world's most innovative economy, firmly set on its rise toward the
superpower status. World War I (1914-18) traumatized all European powers, utterly destroyed the post-Napoleonic pattern, ushered in Communism
in Russia, and brought the United States into global politics for the first
time. And, a fact that is often forgotten, it also began the continuing destabilization of the Middle East by dismembering the Ottoman Empire and by
leading eventually to the formation of such artificial, and precariously ex-
FIGURE5 Wars of magnitude 6 and 7 occurring between 1850 and

2000. Plotted fatalities are minimal to average (and heavily rounded)
estimates from a variety of sources listed in the text.
(8)100 _
WWII
Taiping
::
-=
M?~ ^/^in=~~~~
E r(7)10-
m^
n_i~~~~~~
~~WWI
Civil War
Vietnam
Civil War
_Spanish
X-
War
~Sino-Japanese
~~Chinese
Russian Civil War
War
Afghanistan Wars
Wars in Sudan, Biafra,
Mexican War
,(6)1
-
Algerian War
US Civil War
I
1850 1850
1900
Year
I
1950
Congo, Mozambique,
Iran-Iraq War
I
I
2000
224
isting, states as Jordan (in 1923), Saudi Arabia and Iraq (1932), Lebanon
(1941), and Syria (1946).
World War II (1939-45) is the most obvious entry on the list of transformational wars not only because of the sweeping changes it brought to
the global order but also because of the decades-long shadows it cast over
the rest of the twentieth, and even into the present century. Virtuallyall of
the key post-1945 conflicts that involved the protagonists of World War II
(Russia,the United States, and Chinain Korea;Franceand the United States
in Vietnam; Russia in Afghanistan; superpower proxy wars in Africa) can
be seen as actions designed primarilyto maintain or to challenge the outcome of that war. Arguably, other conflicts might seem to qualify, but a
closer examination shows that they did not fundamentally alter the past
but rather reinforced (often at a high cost) the changes set in motion by
transformationalwars.
The two most obvious cases are the undeclared but no less fatal wars
that were waged, by a variety of means ranging from outright mass killings
to manmade famines, against populations of the Soviet Union by Stalin between 1929 and 1953, and against China's people by Mao between 1949
and 1976: the actual toll of these brutalitieswill be never known with any
accuracy, but even the most conservative estimates put the combined toll
at above 70 million (White 2003). Another arguable change would be to
redefinethe durationsof the listed transformationalwars:for example, World
War I beginning with the wars in the Balkans in 1912 and ending with the
conclusion of the civil war that establishedthe Soviet Union in 1921; World
WarII startingwith Japan'sinvasion of Manchuriain 1933 and ending with
the Communist victory in China in 1949.
Even a restrictedlist of globally transformationalwars adds up to 42
years of conflicts in two centuries, conservatively estimated total casualties
(combatant and civilian) of about 95 million, a mean recurrence rate of
about 35 years, 17 million deaths per conflict, and an implied probabilityof
a new conflict of that category at roughly 20 percent during the next 50
years. I readily admit that all of these numbers could be reduced by extending the retrospective for another century. The eighteenth century had a
remarkablylower intensity of all violent conflicts than the two preceding
and the two subsequent centuries (Brecke 1999); but as most of it, and
most of the then major powers, belonged distinctly to the premodern era,
its exclusion makes sense.
Three primaryconclusions emerge from the examination of all armed
conflicts of the past two to three centuries. First, an upward trend in the
total number of conflictsstartingin each decade;second, an increasingshare
of wars of short duration (less than one year) (Kaye, Grant, and Emond
1985). Implications of these findings for future transformationalconflicts
are unclear. But the most importantfinding comes from Richardson's(1960)
VACLAV SMIL
225
search for causative factors of war and his conclusion that wars are largely
random catastrophes whose specific time and location we cannot predict
but whose recurrence we must expect. That would mean that wars are like
earthquakes or hurricanes, leading Hayes (2002: 15) to speak of warring
nations that "bangagainst one another with no more plan or principle than
molecules in an overheated gas."
At the beginning of the twenty-first century one could argue that new
realities have greatly diminished the likelihood of recurrence of many possible conflicts. The European Union is widely seen as a near-absolute barrier to any armed conflict involving its members; America and Russia may
not be strategic partners but they are not the same adversaries they were
for two generations preceding the fall of the Berlin Wall in 1989; the Soviet
Union and China came very close to a massive conflict in 1969 (a close call
that soon prompted Mao to turn to the United States), but today China
buys the top Russian weapons; and Japan is not about to attack anybody in
East Asia. This reasoning would negate, or at least severely undercut,
Richardson's argument, but it would be a mistake to use it when thinking
about long spans of history. Neither short-term complacency nor the understandable reluctance or inability to imagine the locale or the cause of
the next grand transformationalconflict is a good argument against its rather
high probability.
In 1790 no Prussianhigh officer or Czaristgeneral could suspect that a
diminutive Corsicanfrom Ajaccio, who soon became known to his troops as
le petitcaporal,would set out to redraw the map of Germanybefore embarking on a mad foray into the heart of Muscovy. In 1840 the EmperorDaoguang
could not have dreamed that the dynastic rule that lasted for millennia would
come close to its end because of a failed candidate of the state Confucian
examination who came to think of himself as a new Christ.And in 1918 the
victorious powers, dictating a new European peace in Versailles,would not
have crediteda destitute, neurotic would-be artistand veteran of trench warfare from Munich with powers large enough to undo, in a mere two decades,
their new order and plunge the world into its greatest war.19
New realities may have lowered the overall probability of grand, globally transformational conflicts but they have not eliminated their recurrence. Richardson's reasoning and the record of the past two centuries imply that during the next half century the likelihood of another armed conflict
whose outcome would have the potential to change the course of world
history is no less than about 15 percent and most likely around 20 percent.
As in all cases of such probabilistic assessments, the focus is not on a particularfigure but ratheron the proper order of magnitude:no matter whether
the probability of a new transformational war is only 10 percent or 40 percent, it is one to two orders of magnitude higher than that of globally destructive natural catastrophes that I described earlier in this essay.
226
Relative fears
Determined and concerted actions can modify, even reverse, many seemingly deeply entrenched trends, but as of now there is nothing we can do to
avert any low-probability natural catastrophe whose consequences can
change the course of history: in this sense our civilization is no different
from the cuneiform or hieroglyphicrealms of the Middle East of 5,000 years
ago, or from the early modern world of the seventeenth century. We are
getting better at anticipating some volcanic activity-clear warnings given
ahead of Mount Pinatubo's 1991 eruption were perhaps the greatest success (Newhall, Hendley, and Stauffer 1997)-but discerningthe likely magnitude of an eruption remains beyond our ken, as does the ability to predict
a major earthquake.
Surprisingly,we may soon be able to do something in space ratherthan
on Earth. As already noted, by the end of 2008 we will have surveyed all
near-Earthobjectslargerthan one km in diameterand will be able to calculate the risk of a very large-scaleterrestrialimpactwith unprecedentedaccuracy. Moreover, at the currentrate of discovery some 90 percent of all nearEarth objects large enough to pose a global risk should be detected by the
year 2020 (Rabinowitzet al. 2000). So far, observations of known objects
and computations of their trajectorieshave not discovered an object on a
collision course with the Earth,but theoreticalproposalshave alreadybeen
made for possible deflection of asteroids (Milani 2003). Such an operation
(contraryto first impressions) is not a matter of wishful science fiction: it is
likely that it could be accomplishedwith a relativelylimited dischargeof energy,just enough to deflect the objectfrom its collision-boundtrajectory.
Appraisalsof naturalcatastrophesthat can have a nearly instantaneous
effect as well as generations-long worldwide consequences show low probabilities during the next half century-but, at the same time, such quantifications enter a realm that is alien even to those experts who routinely analyze various risks. Leading hazards encountered in modern society have a
fairly high frequency of fatalities but they kill in discretely small numbers,
and most of the losses (be it mortality, injuries, or economic damage) are
sustained through voluntary actions and exposures whose risks people almost uniformly underestimate. Annual mortality aggregatesof such exposures may be relatively high but they come to public attention only if a
particularevent of that kind is unusually large.
Car accidents are perhaps the best example of the peculiar attitude
with which humans treat voluntary risks that have a high-frequency rate
but a low-fatality rate per event. They now cause worldwide nearly 1.2 million deaths a year (WHO 2003), but more than 90 percent of individual
events involve the killing of just one or two people; these events are widely
reported only when the per-event mortality rate rises (albeit in absolute
VACLAV SMIL
227
terms it still remains fairly small): fog- or ice-induced pile-ups of scores of

cars causing a dozen or more casualties are the most common instance. In
sharp contrastis the attitude toward terroristattacks:their fatalitiesare large,
their potential dangers are real, but it is our self-inflicted terrorizing that
wildly exaggeratesboth their likely frequency (the notion of ubiquitous complex terrorism) and their impact (suitcase nuclear bombs devastating cities,
new bioterror plagues killing on massive scales).
Failures of machines (airplanes, trains, ferries) or their control systems
more
(or,
likely, of their operators) that result in hundreds of deaths are
rare. Naturalcatastrophesthat kill almost instantly tens or hundreds of thousands are very rare and, as has been clearly illustrated with the December
2004 Indian Ocean tsunami, they elicit incredulity at the scope of their destructive impact. But even mortality on this scale does not matter in terms
of world history. Even if the latest tsunami toll were to surpass 300,000, it
would be nowhere near the record level for a modern natural catastropheand none of those left any long-lasting mark on global affairs. Most notably, every one of China's recurrent (and now almost totally forgotten)
droughts during the first half of the twentieth century (1907, 1928, 1936,
1941) claimed at least one million people, as did the Huanghe flood in 1931.
And in an entirely different world, with weather satellites and electronic
telecommunication, the Bhola cyclone still killed about half a million people
in East Pakistan in 1970.
We must remain agnostic about the eventual impact of what Rapoport
(2001) called the fourth wave of terror.20Historical lessons are clear: ending terror everywhere is not a likely outcome of even the most dedicated
and most taxing effort. Terrorismis deeply rooted in modern culture and
even a virtual eradication of one of its forms or leading groups settles little
in the long run as new forms and new groups may emerge unexpectedly
(as did plane hijackings in the 1960s and al-Qaidain the 1990s). The real
question is about the intensity and frequency with which terrorist actions
can be sustained over a long time period, and the unknown course of those
two variables will determine their eventual impact on modern history.
Good arguments can be made for seeing terroristactions as a shocking
(also painful and costly) but manageable risk among other risks and to point
out the general tendency to exaggerate the likelihood of new, infrequent,
but spectacular, threats. On the other hand, it is understandable why a responsible political leadership would tend to see terrorism as an unprecedented, and intolerably deadly, challenge to the perpetuation of modern
open societies. In any case, nobody can assign any meaningful probabilities
to these different outcomes, whereas a clear judgment is possible regarding
any future natural catastrophes: in order to leave a mark on world history
they would have to be on scales not experienced during the historic era
and would have to claim, almost instantly or within a few months, many
228
millions of lives. Events of sufficient magnitude to produce such a toll took

place within the past million years, but none of them has probabilitieshigher
than 0.1 percent during the next 50 years (Figure 6).
In contrast, there is a high probabilityof an influenza pandemic that
would rival or surpassthe greatestsuch event on record,and a simple probabilistic assessment shows that the risk of a transformationalmegawar is of
the same orderof magnitude (Figure6). But I emphasizethat the past record,
while highly indicative and inevitably suggestive, is basically one of
singularities, events that are too few and mostly too far apart to allow for
any meaningful statistical evaluation beyond the simplest calculations of
highly uncertain return frequencies and approximaterecurrenceprobabilities. Still, we now know enough about near-Earthobjects to rank the danger from colliding asteroids as by far the least likely discontinuity with a
potential to change near-term history. Moreover, the overall risk may be
revised substantiallydownward during the coming years.
We can take many steps to lessen the risk of a future megawar and
even to reduce the likelihood of the worst imaginable terroristattacks, but
we are entirely at the mercy of unpredictable viral mutations. With luck
the next pandemic may resemble much more the last bout than the 191819 event, or it could be the greatest viral cataclysm in history. These are,
indeed, the most disturbing marks of our ignorance: we have no way to
know whether we are exaggeratingor underestimatingwhat is to come, be
it from the depth of fundamentalist hatred or from random mutations of
viral genomes. We can take many steps to be better prepared for another
terrorist attack on the scale of 9/11 and for a pandemic mortality on the
scale of the two last episodes. But if we are grossly underestimating these
risks, then there is much less we are likely to do to make any fundamental
difference: there is simply no way to prepare for a terrorist attack with a
hijacked nuclear-tipped missile or for dealing with more than 2 billion infected people and more than 100 million deaths.
But many steps we can take will make a marginaldifference,and those
margins may translateinto a large number of saved lives. The probabilityof
a terrorist campaign that would be sufficiently intense and prolonged to
change the course of history is unknown and unknowable. Even so, a combination of rational steps (in contrast to the exaggerated, often plainly cosmetic, and patently wasteful post-9/11 responses) rangingfrom better evaluation of available intelligence (still marred by shortages of translatorsand
coordinated data bases) and more flexible armed response to gradualsocial
and political transformationscan clearly reduce the likelihood of many terroristattacks and moderate their consequences.
But if we are to act as rational risk minimizers, then the current preoccupation with this risk should not blind us to what are historically two
much more likely threats: another megawar and another pandemic during
VACLAV
229
SMIL
FIGURE 6 Probabilities of massively fatal discontinuities during the

next 50 years, ranging from low chances of asteroid impacts to a high
probability of a virulent influenza pandemic. Starting points at the
right vertical axis are calculated on the basis of the historical record of
the past two centuries; all curves are approximations, aiming
primarily to convey correct orders of magnitude
influenza
megawars
volcanoes
tsunamis
asteroids
1
0.01
0.05 0.1
0.5
Probability (%)
10
50 100
230
THENEXT50 YEARS:FATALDISCONTINUITIES
the next 50 years. Early interventions to defuse any emerging causes of

potentiallymassive armedconfrontationsand betterpreparednessfor a major
pandemic would be the most rewarding risk-reducing steps. Forget about
near-Earthasteroids,supervolcanoes, and monster tsunamis;with foresight
and some luck we can even live with terrorism,but we must not underestimate the chances of another megawar and must remember that unpredictably mutating viruses will be always with us.
Notes
1 On Fukuyama:he rightlycomplainsthat
he has been misunderstoodsince he did not
suggestthat events will come to an end. What
he maintainsis that no matterhow large and
grave future events will be, history itself ("as
a single, coherent, evolutionary process")is
over because nothing else awaitsbut an eventual triumph of liberaldemocracy.This claim
irritatesbecause of its combinationof wishful
thinking and poverty of imagination: if we
were to believeit, then the attacksof 9/1 1, fundamentalist Islam, terrorism,nuclear blackmail, globalizationof the labor force, and the
resurgenceof Chinaare nothing but inconsequential events because "all of the really big
questionshad been settled."
On Nineveh: I am farfrom convinced, despite the enormous challenges we face, that
our civilizationwill be soon transmutedinto a
defunctheap. And even if it were the case, we
would still not be one with Nineveh: myriads
of our artifactsmade of billions of tonnes of
steel, other metals, glass, and plasticsthat we
leave behindwill be betterpreserved,especially
once buried,than the AssyrianEmpire'sshortlived capitalof claythat was so thoroughlydestroyedby invadingBabylonians.
2 Some of these scenariosare availableat
a staggeringcost. For example, GlobalBusiness Network'sGBNScenario
Book2003:History
in Motion (127 pages of spiral-bound text,
graphs,and collages) is availableto organizations (but not individuals)only as part of an
annual US$40,000 subscription.See owww.
gbn.com,>.
3 Several remarkablediscontinuities are
embedded in the fossil record, none more so
than the great Cambrianexplosion of highly
organizedand highly diversifiedterrestriallife.
Thisgreatevolutionaryjump began about 533

million years ago and it had producedwithin a geologically short spell of just 5 to
10 million years, equal to less than 0.3 percent of the entire evolutionary span-virtually all of the animal lineages that are known
today (McMenaminand McMenamin 1990).
4 For comparison,Bostrom(2002), in his
study of existentialrisks (those that could annihilateintelligentlife or permanentlyor drasticallycurtailits potential)classifiedthem into
bangs (extinctions due to sudden disasters),
crunches (that thwart future developments),
shrieks (resulting in very limited advances),
and whimpers (changesthat lead to the eventual demise of humanity).
5 Some readersmight have noticed that
this typology is a differentlyworded version
of an undeservedlymalignedenumerationofferedby DonaldRumsfeldduringhis pressconference on 12 February2002 (US Department
of Defense 2002):
Reportsthat say that somethinghasn'thappened arealwaysinterestingto me, because
as we know,thereareknownknowns;there
arethingswe knowwe know.We alsoknow
thereareknownunknowns;thatis to saywe
know therearesome thingswe don'tknow.
Buttherearealsounknownunknowns-the
ones we don'tknow we don'tknow.
6 A biosphereprone to rapidtakeoverby
a single microorganismcould not become differentiatedinto millionsof species,with thousands of them closely interdependentwithin
complex food webs of rich ecosystemsand all
of them connected through global biogeochemicalcycles:symbiosishas been no less
criticalthan interspecificcompetitionfor life's
survival(Smil2000; Sapp 1994).
VACLAV SMIL
There are even weirderostensiblysciencebased suggestions regarding civilization's demise, including the idea that we are living in a
simulation of a past human society that is run
by a superintelligententity that can choose to
shut it down at any time (Bostrom 2002). In
responseI would only note that the mind running the exercise has been a very patient one
as the simulation has been going on for nearly
4 billion years; or does Bostrom dismiss evidence of the Earth'sevolution and our emergence as one of its results?
7 In November 1951 the Soviets deployed
their first deliverablefission bomb but a more
appropriateyear might be 1955 when both superpowers acquired their first nuclear-tipped
missiles (Johnston 2005): unlike bombers,
launched ballisticmissiles cannot be recalled.
8 On a much more mundane level, these
situations are akin to avoided fatal car crashes
when a few centimeters of clearance makes a
difference between death and survival:these
events happen worldwide thousands of times
every hour, but an individual has only one or
two such experiences in a lifetime and these
singularities make it impossible to calculate
probabilitiesof any future clean escape.
9 As the fanatics of Aum Shinrikyo discovered, it is not easy to disperse a nerve gas
and kill a large number of people even in such
a densely populated setting as Tokyo'ssubway
(12 people died as the result of their March
1995 attack), and the US anthrax scare of October and November 2001 was primarily a
matterof irresponsiblyexaggeratedfearsrather
than of any history-alteringmedical catastrophe. Indeed, many experts argue that the entire threatof bioterrorismhas been vastlyoverblown (Enserinkand Kaiser2005). Given the
enormous amount of money now pouringinto
US researchon potentiallydeadly anthraxand
smallpox (more than half a billion dollars in
FY 2004), the work on common pathogens
that annually claimmillions of lives worldwide
gets shortchanged;and (a counterintuitivebut
credible possibility) there is a much greater
dangerthat a bioterroragent could be released
by a disgruntled employee of one of 14 new
research super-labs built to handle the most
dangerous pathogens, rather than being carried to the US by a fundamentalistzealot in a
suitcase.
231
10 Who now remembersthe greatAmerican blackouts of 1965 and 1977? And the latest repeat, the US-Canadian blackout of August 2003, provided a remarkableillustration
of both technical and social resilience. Caused
by a series of preventabletechnical failures,it
affected some 50 million people in the Northeastern United States and in Ontario, and (a
perfect example of creatingmass havoc) it extinguished all trafficlights and stoppedall subway trains in New York.But trading on Wall
Street continued, the blackout did not lead to
any catastrophicdisruptionof the region'sbusiness or its economic growth, hospitals were
able to maintain adequate care, and, despite
hot weather, the crime rate actually dropped.
11 The most accessible interdisciplinary
source of accurate information on the 2004
Indian Ocean tsunami is Wikipedia'slong entries for "Tsunami"and "2004 Indian Ocean
Earthquake"at http://en.wikipedia.org/wiki/
2004_Indian_Ocean_earthquakeand at http:/
/en.wikipedia.org/wiki/Tsunami
12 The explosive power of one metricton
of TNTis equal to 4.18 billionjoules (GJ).There
is no definite number for the explosive yield
of the Hiroshima bomb, but the most authoritative source (Malik 1985) puts it at 15 (+3)
kt of TNT.
13 Ice-freeland accountsfor about 26 percent of the Earth'ssurface and densely populated regionsadd up to about 4 percent of such
land, or almost exactly to 1 percent of the
planet's surface. Hiroshima,with a bowl-like
setting that acted as a natural concentratorof
the blast, had about 40 percent more fatalities
and more destruction with a 15 kt blast than
did Nagasakifrom a 21 kt explosion. Another
complicatingfactoris that a Tunguska-likeblast
may not be a point-source event (similarto a
nuclear bomb) but rather a plume-forming
event (similarto a line of explosive charges)
and hence could be caused by much less powerful objects (NEOSDT2003).
14 This theory relies on studies by
Harpendinget al. (1993) of mitochondrialDNA
that indicate a severe population bottleneck
between 80,000 and 70,000 years before
present. For its critiqueand reply see Ambrose
(2003).
15 This assumes, for illustrative reasons
only, an even downwind distribution;the ac-
232
tualash coverwould rangefromseveralmeters

to a few centimeters. Based on the effect of
previouseruptions(Fisher,Heiken,and Hulen
1997) and assumingprevailingnorthwesterly
winds, this ash-coveredareawould encompass
Wyoming, Colorado,Nebraska,Kansas, and
Oklahomaand parts of South Dakota, Texas,
New Mexico, and Utah. Thinnerlayersof volcanicash couldbe easilyincorporatedinto soils
by plowing, but even the most powerful tractorscouldnot handledepositsof 0.5-1.0 m and
an inevitable consequence would be (at least
temporary)abandonmentof croppingon large
areas of the GreatPlains. Unstableash layers
would be easily dislodgedby heavy rains and
springsnow melt, creatingenormous flooding
and streamsiltinghazards.The economic costs
of such an event could be assessed only generationsafterit took place.
16 But these are battles with no assured
outcomes. The endgame has not been easy in
the case of polio owing to an irrationalresistance to vaccinationin northernNigeria(Roberts2004). Pertussis(whoopingcough) is coming back among children too young to be
vaccinated (Tozziet al. 2005). And there is a
long list of more sinisterreemerginginfections
(Morens,Folkers,and Fauci2004) with a new
Diseases,published
journal-EmergingInfectious
by the Centers for Disease Control and Prevention-devoted to this globalchallenge.
17 HIV/AIDShas had widespreaddebilitating impacts (social, mental, economic) in
sub-SaharanAfrica(where the highest country ratesof infectionnow surpass20 percent),
but its annual globaldeath toll (about2.8 million in the early 2000s) is less than mortality
due to diarrheaand tuberculosis,two diseases
that we know perfectlywell how to eradicate

at an acceptablylow cost and that now claim
annually about 3.4 million lives (WHO2003).
Moreover,low and steady rates of HIVinfection in many countries and falling rates in
some previously badly affected nations (particularlyUgandaand Thailand)show that the
diseasecan be managed.
18 Thiscentury-longstabilitywas not fundamentallyalteredeither by briefconflictsbetween Germanyand Austriaand Germanyand
Franceor by repeatedacts of terrorthat killed
some of the continent'sleading public figures
(and some, including Kaiser Wilhelm I and
ChancellorBismarck,escapednot just one but
two assassinationattempts).
19 Failureof imaginationis our constant
companion as we face new political, economic, and strategic realities. Perhaps the
three most spectacularexamples of the past
generation were the rapid and nonviolent
demise of the Soviet Union, the sudden post1990 retreatof Japan from being the world's
leadingeconomy, and, of course,young, welleducated middle-class Saudis and Egyptians
steering Boeing 767s into the World Trade
Center.
20 Accordingto Rapoport(2001), "Since
the 1880s, four successive,overlappingmajor
waves of terrorhave washed over the world....
The firstthree lasted approximatelya generation each; and the fourth, which began in
1979, is stillin process"(p. 420). Theyear 1979
saw not only the rise of theocracyin Iranbut
also, symbolically,the beginningof a new century in the Islamiccalendar.
References
Albright,D. and K. O'Neill(eds.). 2000. SolvingtheNorthKoreanNuclearPuzzle.Washington,DC:
Institutefor Scienceand InternationalSecurity.
Allison, G. and P. Zelikow. 1999. EssenceofDecision:
ExplainingtheCubanMissileCrisis.New York:
Longman.
Ambrose, S. H. 1998. "LatePleistocenepopulationbottlenecks,volcanic winter, and differentiation of moder humans,"JournalofHumanEvolution34: 623-651.
.2003. "Didthe super-eruptionof Tobacause a human populationbottleneck?Reply to
JournalofHumanEvolution45: 231-237.
Gathore-Hardy and Harcourt-Smith,"
Andriolo,K. 2002. "Murderby suicide:Episodesfrom Muslimhistory,"American
Anthropologist
104: 736-742.
VACLAV SMIL
233
Barry, J. M. 2004. The Great Influenza: The Epic Story of the Deadliest Plague in History. New York:
Viking.
Bland, P. A. and N. A. Artemieva. 2003. "Efficient disruption of small asteroids by Earth's atmosphere," Nature 4214: 288-291.
Blight, J. G. and D. A. Welch. 1989. On the Brink: Americans and Soviets Reexamine the Cuban
Missile Crisis.New York: Hill and Wang.
Bostrom, N. 2002. "Existential risks," Journal of Evolution and Technology9(March), <http://www.
jetpress.org>>.
Brecke, P. 1999. "Violent conflicts 1400 AD.to the present in different regions of the world," paper
prepared for the Meeting of the Peace Science Society, 8-10 October, Ann Arbor, MI.
Britten, S. 1983. The Invisible Event. London: Menard Press.
Brown, P. et al. 2002. "The flux of small near-Earth objects colliding with the Earth," Nature
420: 294-296.
Bundy, M. 1988. Danger and Survival: ChoicesAbout the Bomb in the First Fifty Years. New York:
Random House.
Calder, N. 1979. Nuclear Nightmares. London: BBC.
Capek, K. 1921. R. U. R. Rossum's UniversalRobots.Praha: Aventinum.
Casti, John L. 2004. "Synthetic thought," Nature 427: 680.
Centers for Disease Control and Prevention. 2000. "Prevention and control of influenza from
the Advisory Committee on immunization practices," Morbidityand Mortality WeeklyReport
2000:1-38.
. 2005. Mortality Data from the National Vital StatisticsSystem. Atlanta, GA: CDC. <http://
www.cdc.gov/nchs/about/major/dvs/mortdata.htm)
Cha, V. and D. C. Kang. 2003. Nuclear North Korea:A Debate on Engagement Strategies.New York:
Columbia University Press.
Chapman, C. R. 2004. "The hazard of near-Earth asteroid impacts on earth," Earth and Planetary
ScienceLetters222: 1-15.
Chapman, C. R. and A. W. Harris. 2002. "A skeptical look at September lth: How we can
defeat terrorism by reacting to it more rationally," SkepticalInquirer, September-October.
<<http://www.csicop.org/si/2002-09/9- l.html>>
Chen, H. et al. 2004. "The evolution of H5N1 influenza viruses in ducks in southern China,"
Proceedingsof the National Academyof Sciences101: 10452-10457.
Chen, Y. et al. 1988. The Great Tangshan Earthquake of 1976: An Anatomy of Disaster. New York:
Pergamon Press.
Chotpitayasunondh, T. et al. 2004. "Human disease from influenza A (H5N1), Thailand, 2004,"
Emerging InfectiousDiseases 11: 201-209.
Coale, A. J. 1985 "Nuclear war and demographers' projections," Population and DevelopmentReview 11: 483-493.
Davies, P. 1999. The Devil's Flu: The World's Deadliest Influenza Epidemicand the ScientificHunt for
the Virus that CausedIt. New York: Henry Holt.
Dolgov, Y. A. (ed.). 1984. Meteoritnyeissledovaniiav Sibiri: 75 let tungusskomufenomenu. Novosibirsk:
Izdatel'stvo Nauka.
Ehrlich, P. and A. Ehrlich. 2004. One with Nineveh: Politics, Consumption,and the Human Future.
Washington, DC: Island Press.
Eldredge, N. and S. J. Gould. 1972. "Punctuated equilibria: An alternative to phyletic gradualism," in T. J. M. Schopf (ed.), Models in Paleobiology. San Francisco: Freeman, Cooper, pp.
82-115.
Enserink, M. and J. Kaiser. 2005. "Has biodefense gone overboard?," Science307: 1396-1398.
Fisher, R. V., G. Heiken, and J. B. Hulen. 1997. Volcanoes:Cruciblesof Change. Princeton, NJ:
Princeton University Press.
Forrow, L. et al. 1998. "Accidental nuclear war-A post-Cold War assessment," New England
Journal of Medicine 338(8): 506-511.
Fukuyama, F. 1992. The End of Historyand the LastMan. New York: Free Press.
234
Grais,R. F. et al. 2004. "Modelingthe spread of annual influenza epidemics in the U.S.: The
Science7:127-134.
potentialrole of air travel,"HealthCareManagement
Gust,I. D., A. W. Hampson,and D. Lavanchy.2001. "Planningfor the next pandemicof influenza,"Reviewin MedicalVirology11: 59-70.
Harpending,H. C. et al. 1993. "Thegenetic structureof ancient human populations,"Current
34: 483-496.
Anthropology
Man.London:K. P. Trench,Trubner
Hatfield,H. 1928. Automation;
Or,theFutureoftheMechanical
& Company.
90: 10-15.
Hayes,B. 2002. "Statisticsof deadly quarrels,"AmericanScientist
Hills,J. H. and M. P. Goda. 1993. "Fragmentationof small asteroidsin the atmosphere,"AstronomicalJournal105: 1114-1144.
Homer-Dixon,T. 2002. "Therise of complex terrorism,"ForeignPolicy128(1): 52-62.
Huixian, L. et al. 2002. TheGreatTangshanEarthquake1976.Pasadena,CA:CaliforniaInstitute
of Technology. <http://resolver.caltech.edu/CaltechEERL:EERL.2002.001)
Interational Instituteof StrategicStudies.2003. ArmedConflictDatabase.London:IISS.<http://
acd.iiss.org>>
InternationalPeace Research Institute. 2004. ThePRIO/Uppsala
ArmedConflictDatabase.Oslo:
PRIO.<<http://www.prio.no/cwp/ArmedConflict>>
Jakes, L. 2005. "Security report outlines terror scenarios," ABCNews, 16 March. <http://
abcnews.go.com/Politics/,
Jewitt, D. 2000. "Eyeswide shut,"Nature403: 145-148.
Johnston, W. R. 2005. Nuclear WeaponMilestones.<<www.johnstonsarchive.net/nuclear/
wrjp205.htmlb
Joy, B. 2000. "Why the future doesn't need us," Wired8.04 <<http://www.wired.com/wired/
archive/8/04/joy.htmln
A Compendium
Kaye, G. D., D. A. Grant,and E. J. Emond. 1985. MajorArmedConflict:
of Interstate
1720to 1985.Ottawa:Departmentof NationalDefense.
andIntrastate
Conflict,
Kolata, G. 1999. Flu: TheStoryof the GreatInfluenzaPandemicof 1918and the Searchfor the Virus
thatCausedIt.New York:Farrar,Strausand Giroux.
Kurzweil,R. 1999. TheAgeof SpiritualMachines.New York:Viking.
Li, K. S. et al. 2004. "Genesisof a highly pathogenicand potentiallypandemicH5N1 influenza
virus in easter Asia,"Nature209-212.
Lipman,P. W. and D. R. Mullineaux (eds.). 1981. The1980EruptionsofMountSt.Helens.Washington, DC:US GeologicalSurvey.
Los Alamos, NM:Los
Malik,J. 1985. TheYieldsof theHiroshimaandNagasakiNuclearExplosions.
AlamosNationalLaboratory.
McMenamim,M. A. S. and D. L. S. McMenamin.1990. TheEmergence
ofAnimals:TheCambrian
New York:ColumbiaUniversityPress.
Breakthrough.
material-virtual or realhazards?,"Science300: 1882-1883.
Milani,A. 2003. "Extraterrestrial
Moore, J. G., W. R. Normark,and R. T. Holcomb. 1994. "GiantHawaiianlandslides,"Annual
ReviewofEarthandPlanetarySciences
22: 119-144.
Mind.New York:OxfordUniversityPress.
Moravec,H. 1999. Robot:MereMachineto Transcendent
Morens, D. M., G. K. Folers,and A. S. Fauci.2004. "Thechallenge of emergingand re-emerging infectiousdiseases,"Nature430: 242-249.
NEOSDT(Near-EarthObjectScience DefinitionTeam). 2003. StudytoDetermine
theFeasibility
of
to SmallerLimitingDiameters.Washington, DC:
Extendingthe Searchfor Near-EarthObjects
NASA. <<http://neo.jpl.nasa.gov/neo/neoreportO30825.pdf>n
Newhall, C., J. W. Hendley, and P. J. Stauffer.1997. "Benefitsof volcano monitoringfar outweigh costs-The case of Mount Pinatubo,"U.S.Geological
SurveyFactSheet115-97.<<http://
wrgis.wr.usgs.gov/fact-sheet/fsl15-97/,
Norris, R. S. and H. M. Kristensen.2005a. "U.S. nuclear forces, 2005," Bulletinof the Atomic
Scientists
61(1): 73-75.
.2005b. "Russiannuclearforces,2005," Bulletinof theAtomicScientists
61 (2): 70-72.
VACLAV
SMIL
235
Patterson,K. D. and G. F. Pyle 1991. "Thegeography and mortalityof the 1918 influenza pandemic,"Bulletinof theHistoryofMedicine65: 4-21.
Phillips,A. F. 1998. "20mishapsthat might have startedaccidentalnuclearwar,"Nuclearfiles.org.
<<http://www.nuclearfiles.org/kinuclearweapons/anwindex.html>>
Phillips,H. and D. Killingray(eds.). 2003. TheSpanishInfluenzaPandemicof 1918-19:NewPerspectives.London:Routledge.

Potts,R. 2001. "Complexityand adaptabilityin human evolution,"paperpresented at the AAAS
conference "Developmentof the Human Species and its Adaptationto the Environment."
Rabinowitz, D. et al. 2000. "A reduced estimate of the number of kilometer-sized near-Earth
asteroids,"Nature403: 165-166.
Rampino, M. R. and S. Self. 1992. "Volcanicwinter and accelerated glaciation following the
Toba super-eruption,"Nature359: 50-52.
Rapoport, D. C. 2001. "The fourth wave: September 11 in the history of terrorism,"Current
History100: 419-424.
Reid, A. H. et al. 2003. "1918 influenza pandemic caused by highly conserved viruses with two
Diseases9: 1249-1253.
receptor-bindingvariants,"EmergingInfectious
Rhodes, R. 1988. "Man-madedeath: A neglected mortality,"JournalofAmericanMedicalAssociation260: 686-687.
Richardson,L. F. 1960. Statistics
ofDeadlyQuarrels.PacificGrove, CA:The Boxwood Press.
Roberts,L. 2004. "Polio:The final assault?,"Science303: 1960-1968.
Rose, I. W. and C. A. Chesner. 1990. "Worldwidedispersalof ash and gases from the earth's
largestknown eruption:TobaSumatra,75 ka,"Palaeography,
Palaeoclimatology,
Palaeogeography
(GlobalandPlanetaryChangeSection)89: 269-275.
Sagan, S. D. 1993. TheLimitsof Safety.Princeton,NJ:PrincetonUniversityPress.
Sakharov,A. 1983. "Thedanger of thermonuclearwar,"ForeignAffairs61: 1001-1016.
A Historyof Symbiosis.
New York:OxfordUniversityPress.
Sapp, J. 1994. EvolutionbyAssociation:
Simpson, G. G. 1983. FossilsandtheHistoryof Life.New York:ScientificAmericanLibrary.
Sims, L. D. et al. 2002. "Avianinfluenza in Hong Kong 1997-2002," AvianDiseases47: 832-838.
Handbook.New York:
Singer, D. J. and M. Small. 1972. TheWagesof War1810-1965:AStatistical
John Wiley.
Smil, V. 2000. Cyclesof Life.New York:ScientificAmericanLibrary.
.2003. Energyat the Crossroads:
GlobalPerspectives
and Uncertainties.
Cambridge,MA: MIT
Press.
Smith, R. B. and L. W. Braile. 1994. "TheYellowstone hotspot,"Journalof Volcanology
and GeothermalResearch61: 121-188.
Snacken,R. et al. 1999. "Thenext influenzapandemic:LessonsfromHongKong, 1997,"Emerging
Diseases5: 195-203.
Infectious
Spence, J. D. 1996. God'sChineseSon:TheTaipingHeavenlyKingdomof HongXiuquan.New York:
Norton.
Starr,C. 1969. "Socialbenefits versus technologicalrisk,"Science165: 1232-1238.
Stohr, K. and M. Esveld. 2004. "Willvaccines be available for the next influenza pandemic?,"
Science306: 2195-2196.
Stuart,J. S. 2001. "Anear-Earthasteroidpopulation estimate from the LINEARsurvey,"Science
294:1691-1693.
Taubenberger,J. K. et al. 1997. "Initialgenetic characterizationof the 1918 'Spanish'influenza
virus,"Science275: 1793-1796.
Tozzi, A. E. et al. 2005. "Diagnosisand management of pertussis,"CanadianMedicalAssociation
Journal172: 509-515.
United Nations. 1991. WorldPopulationProspects
1990.New York.
1999. WorldPopulationProspects:
The1998Revision.New York.
The2002 Revision.New York. <<http://www.un.org/esa/
1.PDF
population/publications/wpp2002/WPP2002-HIGHLIGHTSrev
The2004 Revision.New York. <<http://esa.un.org/unpp>>
236
U.S. Departmentof Defense. 2002. DoD News Briefing-SecretaryRumsfeldand Gen. Myers.

Tuesday,February12, 2002. "(http://www.defenselink.mil/transcripts/2002/t02122002_
t212sdv2.htmlb
Ward, S. N. and E. Asphaug. 2000. "Asteroidimpact tsunami: A probabilistichazard assessment," Icarus 145: 64-78.
Ward,S. and S. Day. 2001. "CumbreViejavolcano-potential collapseand tsunamiat LaPalma,

Canary Islands," GeophysicalResearchLetters28: 3397-3400.
Webster,R. G. 1997. "Predictionsfor future human influenza pandemics,"TheJournalof Infectious Diseases 176(Supp 1)S14-S19.
White, M. 2003. Historical Atlas of the Twentieth Century. http://users.erols.com/mwhite28/
20centry.htm
LosAngeles, CA:Universityof CaliforniaPress.
Wilkinson,D. 1980. DeadlyQuarrels.
WHO (World Health Organization). 2003. The WorldHealth Report2003. Geneva: WHO.
.2005. Avian Influenza:Assessingthe PandemicThreat.Geneva: WHO.
Wolfe, Tom. 1968. "Whatif he is right?,"in ThePumpHouseGang.New York:Farrar,Straus&

Giroux,pp. 119-154.
Wynn, R. B. and D. G. Masson. 2004. "CanaryIslandslandslidesand tsunami generation:Can
we use turbiditedepositsto interpretlandslideprocesses?,"Southampton:Southampton
Oceanography Centre. (<http://www.soc.soton.ac.uk/CHD/staff_stu/Russel-Wynn/
Wynn&Masson.pdf

Smil 2005

Uploaded by

Copyright:

Available Formats

Smil 2005

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Smil 2005

Uploaded by

Copyright:

Available Formats

The Next 50 Years: Fatal Discontinuities

Author(s): Vaclav Smil

The Next 50 Years:

THE NEXT 50 YEARS: FATAL DISCONTINUITIES

of these repeatedlyfailedanticipations.The unprecedenteddecline of Russia's

cial, or economic) whose impact is likely to be large enough to shape the

THE NEXT 50 YEARS: FATAL DISCONTINUITIES

THE NEXT 50 YEARS: FATAL DISCONTINUITIES

ingly devious leadership that is engaged in a risky game of serial extortion

THE NEXT 50 YEARS: FATAL DISCONTINUITIES

demonstratesthat similar,or even larger,failureswould not rise to the level

FIGURE1 US fatalities per person per hour of exposure plotted

* terrorism (US only)

* terrorism (US citizens)

motor vehicle accidents

THE NEXT 50 YEARS: FATAL DISCONTINUITIES

THE NEXT 50 YEARS: FATAL DISCONTINUITIES

rowed the total number of kilometer-sized near-Earthasteroidsto just over

THE NEXT 50 YEARS:

FIGURE2 Size and frequency distribution of near-Earth asteroids (all

temperatures and plant productivity. This catastrophe is perhaps the best

THE NEXT 50 YEARS:

FIGURE3 An unlikely, but potentially devastating natural

SOURCE:Based on results of a simulation exercise by Ward and Day (2001).

vine spongiform encephalopathy commonly known as mad cow disease),

THE NEXT 50 YEARS: FATAL DISCONTINUITIES

THE NEXT 50 YEARS: FATAL DISCONTINUITIES

larger global population, an at least an eight times larger (nearly 20 billion

FIGURE4 Age-specific mortality in New York City in October 1918

THE NEXT 50 YEARS: FATAL DISCONTINUITIES

FIGURE5 Wars of magnitude 6 and 7 occurring between 1850 and

THE NEXT 50 YEARS: FATAL DISCONTINUITIES

THE NEXT 50 YEARS: FATAL DISCONTINUITIES

terms it still remains fairly small): fog- or ice-induced pile-ups of scores of

THE NEXT 50 YEARS: FATAL DISCONTINUITIES

millions of lives. Events of sufficient magnitude to produce such a toll took

FIGURE 6 Probabilities of massively fatal discontinuities during the

the next 50 years. Early interventions to defuse any emerging causes of

Thisgreatevolutionaryjump began about 533

THE NEXT 50 YEARS: FATAL DISCONTINUITIES

tualash coverwould rangefromseveralmeters

that we know perfectlywell how to eradicate

THE NEXT 50 YEARS: FATAL DISCONTINUITIES

Phillips,H. and D. Killingray(eds.). 2003. TheSpanishInfluenzaPandemicof 1918-19:NewPerspectives.London:Routledge.

THE NEXT 50 YEARS: FATAL DISCONTINUITIES

U.S. Departmentof Defense. 2002. DoD News Briefing-SecretaryRumsfeldand Gen. Myers.

Ward,S. and S. Day. 2001. "CumbreViejavolcano-potential collapseand tsunamiat LaPalma,

Wolfe, Tom. 1968. "Whatif he is right?,"in ThePumpHouseGang.New York:Farrar,Straus&

You might also like