NATO's victory in Kosovo represents not a new paradigm but simply a
new turn in an old historical pattern that we would do well to study.
On the one hand, over the past five centuries Western technological
superiority has repeatedly brought military victories and ensured
Western domination of the globe. On the other, in every age there
have been numerous occasions when Western powers suffered local
defeats, sometimes so severe as to drive them from a region
altogether or bring down regimes in Europe. The reasons for these
defeats were generally to be found in a combination of local culture,
geography, and Western arrogance and ignorance.

The difference today is that Western tolerance of casualties is
lower, and the stakes--at least when it comes to the potential use of
weapons of mass destruction by non-state actors--are a great deal
higher. In 1879 the British could lose more than 1,500 men in an hour
at Isandlwana, shrug it off, and go on to conquer Zululand. In
1993-94 the United States was forced out of Somalia after 18 men were
killed.

What the examples of Mogadishu, Beirut and Grozny demonstrate is that
today the most dangerous terrain for the West is not the mountains or
jungles of the past, but the city. There, the automatic rifle and the
rocket-propelled grenade still give local irregulars technological
advantages of which their nineteenth-century equivalents could only
dream, and which Western forces still cannot neutralize. The most
dangerous enemy, meanwhile, is not the regular general, but some
combination of terrorist commander, ward politician, traditional clan
leader transposed to an urban setting, and criminal warlord or gang
boss.

Most Western military hierarchies desperately want to believe that
future wars involving their countries will be regular affairs against
regular enemy forces, preferably fighting at sea or in the middle of
an open desert. But to fight the United States on such terms would
indeed be suicidal--which is why our enemies are unlikely to attempt
it. Western technological superiority is only one reason why this is
not the kind of war we have to fear. Equally important is that the
ruling elites of most of the important states around the
world--Russians, Chinese and Indians included--are already to a
considerable extent included in the international economic order, and
as such have much more to lose than to gain from war with the United
States.

The real danger comes not from them, but from the excluded: all those
numerous social and ethnic groups who, for reasons of culture,
history or geography, are unable to partake of the world banquet.
Many of these groups and individuals are far too weak and miserable
to threaten any major power. But some have proud cultural traditions
that make it very difficult to accept peripheral, second-class
status; others have strong fighting traditions that give them a
distinct edge in certain kinds of warfare, organized crime, and the
areas where the two intersect. To fight successfully against such
people on their own ground requires a level of local knowledge that
is exceptionally difficult for outsiders to acquire. It may also
require an urban gang leader's mixture of flexibility and utter
ruthlessness--neither of which is encouraged by the militaries of
Western democracies. As Rick said to the German officer in
Casablanca, "Well, Major, there are parts of New York I wouldn't
advise you to try and invade."

Imperial Precedents

During the 1999 Kosovo war, the amount of damage inflicted by NATO
planes and missiles, compared to the complete inability of the
Yugoslav air defenses to inflict damage in return, was indeed highly
reminiscent of certain nineteenth-century campaigns. On November 3,
1839, two small, elderly British frigates shattered the Chinese
southern fleet off Chuenpi, in the first major engagement of the
First Opium War. Hundreds of Chinese were killed; one British seaman
was wounded. This set the scene for other such nineteenth-century
battles, like Ulundi in 1879, when the British killed thousands of
Zulus for the loss of ten men oftheir own, and Omdurman in 1896. The
First Opium War also echoed even more comprehensive victories three
centuries earlier, when, thanks to superior technology (and
infectious disease), a handful of Spanish adventurers destroyed the
great empires of the Aztecs and the Incas.

Unlike nineteenth-century Western victories, the most important
aspect of the NATO air campaign was not its effect on the enemy's
armed forces. Yugoslav forces on the ground in Kosovo suffered only
around five hundred dead. (Moreover, there is strong evidence to
suggest that most of those deaths took place when the Kosovo
Liberation Army, or KLA, launched an offensive over the mountains
from Albania. This forced the Yugoslavs to come out of concealment
and concentrate against the KLA, and thereby exposed them to NATO
fire.) What was indubitable, and much more important, was the effect
of NATO's bombardment on Yugoslav infrastructure. NATO air forces
(or, rather, those of the United States, with limited European
assistance) demonstrated with complete conviction that any large,
fixed target such as a bridge, power station, oil refinery or factory
could be destroyed from a safe distance. As long as U.S. cruise
missiles and "smart bombs" are around, the United States will be able
to inflict severe damage on the economies of much more powerful
states than Yugoslavia at a very small cost in U.S. lives.

But to defeat or deter a state, as the British did in the case of the
Manchus or NATO did in the case of Yugoslavia, you have to be
fighting against a state. Yugoslavia surrendered as it did because
the Yugoslav Army and police are the disciplined forces of a modern
state, and were under the effective control of a semi-autocratic
president, Slobodan Milosevic. When Milosevic decided to give in--not
because Yugoslavia could not have gone on resisting for a
considerable time, but for his own political reasons--the Yugoslav
forces obeyed and withdrew from Kosovo.

If the Serbian population in Kosovo had been much larger, and had
supported its own autonomous militia under local warlords, would such
forces have surrendered? If not, how would NATO have coerced them? If
the new regime in Serbia or a future one should support indirect
attacks on NATO forces, how will NATO respond if economic and
diplomatic pressure should fail to work? Will it be by defeating and
occupying the whole of Serbia and suppressing any armed revolt from
among the population? If not, how?

The trouble is that bombing bridges and factories has no effect on
irregular forces, and bombing villages and concentrations of people
is supposedly outlawed today by the rules of humanitarian
warfare--though this prohibition is likely to prove a brief one when
we next get involved in a really important war. The only recourses
for NATO in Kosovo against an autonomous Serbian militia would have
been either a compromise involving the partition of the province
between Serbs and Albanians, or giving massive help to the KLA to
crush the Serbian militias on the ground. In the latter case--as with
U.S. help to the Croatian army in 1995, or the Israeli alliance with
the Christian militias in Lebanon in 1982--NATO would have borne
moral responsibility for the massacres of Serbian civilians that
undoubtedly would have followed.

While NATO's victory over Kosovo may therefore be considered a more
or less impressive event in the history of warfare between states, it
provides no instruction whatsoever for other kinds of war to which it
has been linked: wars against peoples in arms; and wars against
decentralized, anarchical or "tribal" societies. It also provides no
answers for the dilemmas involved in urban warfare and in occupying,
administering and policing recalcitrant areas.

To make the point, between the two Western victories against Iraq and
Yugoslavia came the debacle of U.S. involvement in Somalia. During
the urban fighting in Mogadishu, many aspects of U.S.
tactics--weaponry, intelligence and, above all, political analysis--
proved wanting to some degree. Somalia also proved the limitations of
Westernhumanitarian concern in real battles on the ground. According
to Captain Kevin W. Brown of the U.S. Marine Corps, during the
October 3, 1993 attempt to capture General Mohammed Farrah Aideed,

"U.S. restraints [on the use of force] unraveled once Aideed's rebels
began to exact a toll in American lives. In the ensuing desperate
rescue attempt, our forces fired at rebel gunmen who were
interspersed among crowds of Somali civilians. . . . When facing
fierce opposition, both Russian and U.S. rules of engagement have
either been disregarded or relaxed once infantry forces began taking
heavy casualties. This pattern has held true over the past 50 years
of urban conflict. . . . What is the proper balance between avoiding
collateral destruction and properly protecting our infantry forces?
Unfortunately, history reveals that neither the U.S. nor Russian
militaries have found a way to simultaneously accomplish both
objectives in the face of a determined and capable urban foe."

This is not intended as a criticism of the way in which firepower was
used or threatened in Somalia, for in the last resort and when faced
with death, soldiers like most other human beings will fight with all
the means available; and the second duty of every commander--after
the achievement of victory--is to his own men. The Rangers' fight in
Mogadishu should, however, be a most searing reminder that the sort
of orderly, "sanitized", limited war fought by NATO over Kosovo
remains very much the exception, both historically and in today's
world.

The Gulf War and Kosovo were both preceded and followed by other
instructive defeats. The stationing of U.S. troops in Lebanon in
1982, as in Somalia, led to dangerous mission creep. This, in turn,
alienated much of the local population, drew U.S. troops into ethnic
conflict, and, finally, exposed them to devastating terrorist attacks
from elements of the local population. In Kosovo, the NATO occupation
since the end of the war there has failed to protect the local
Serbian minority, and has also failed to create a democratic, let
alone multi-ethnic, Kosovo. As a result, NATO will sooner or later be
faced with a set of choices, every one of which may involve a measure
of moral disgrace and political humiliation.

By not invading and occupying Iraq in 1991, the United States avoided
terrorist and guerrilla attacks by local militias. This decision,
however, also meant that Saddam Hussein would stay in power and
remain a threat and an irritant to the United States (as well as to
his own people) ever since. In other words, Western victory in the
Gulf left George Bush with a dilemma for which military superiority
gave no solution. For as Major General Robert Scales has observed,
"without physical occupation warfare is nothing more than punishment
from a distance, something that any nation with a will to resist can
endure indefinitely." And even if the enemy government does give in,
the effect may prove short-lived.

Fighting Tribes

Given sufficient technological superiority, organized states under
"rational" leaderships can be defeated with relative ease. Once their
leaders accept defeat, the war is over. Perhaps even more important,
once conquered they can be administered using the bureaucratic
institutions and habits of mass obedience inculcated under the former
order.

It is vastly otherwise with tribal and semi-tribal societies, whose
social disciplines are based not on submission to state authority or
laws given from above, but on forms of "ordered anarchy": where no
permanent, centralized authority exists, and where there is no "army"
that can surrender--only local volunteers without uniforms or regular
units, fighting on the basis of blood, personal ties or religious
allegiance. Too many Western observers and staff officers in every
age have dismissed such adversaries as "savages", their sacrificial
courage as mere "fanaticism"--and have subsequently paid the price.
As Kipling put it concerning the Sudanese followers of the Mahdi,
"You're a poor benighted heathen but a first class fighting man." In
covering wars in Chechnya, Afghanistan and elsewhere, I formed a deep
respect for such locals as warriors--the same people seen as alien
and often contemptible from a Western cultural perspective.

Not merely is it far more difficult to defeat such societies (at
least without murdering much of the civilian population in the
process), but officers and officials of modern states often have the
greatest difficulty understanding how such societies work and
devising even theoretical means of dealing with them. Russian
officials who planned the intervention in Chechnya in December 1994
have good cause to remember this, as do U.S. officials who served in
Somalia. A key aspect of these societies is summed up by the
anthropologist I.M. Lewis in a passage about the Somali tradition,
aspects of which have carried over into the urban clan and gang
warfare of today:

The Somali have no indigenous centralized government. And this lack
of formal government and of instituted authority is strongly
reflected in their extreme independence and individualism. Few
writers have failed to notice the formidable pride of the Somali
nomad, his extraordinary sense of superiority as an individual, and
his firm conviction that he is sole master of his actions and subject
to no authority save that of God. If they have noticed it, however,
they have for the most part been baffled by the shifting character of
the nomad's political allegiance and puzzled by the fact that the
political and jural unit with which he acts on one occasion, he
opposes on another.

The complex clan, family, religious, personal and opportunistic
allegiances of ordinary Somali fighters are not amenable to
examination and analysis by U.S. satellites or unmanned spy planes.
The same goes for the Chechens, who, though not nomadic, are a tribal
people, with no tradition either of generating their own state or of
accepting one from outside--except, to a limited extent, when this
comes in the name of Islam. "In peacetime, they recognize no
sovereign authority and may be fragmented into a hundred rival clans.
However, in time of danger, when faced with aggression, the rival
clans unite and elect a military leader."

In the late eighteenth and early nineteenth centuries, the Russian
imperial army won a whole string of victories over the Turks and
various European rivals. It even fought Napoleon to a standstill. But
it took decades to subdue comparatively tiny numbers of Chechens and
other tribesmen in the mountains of the Caucasus. Nor was this a new
pattern: the Spanish conquistadores subjugated the great Aztec and
Inca empires in a few years--and then spent several centuries
attempting to subdue much smaller numbers of primitive Yaquis,
Apaches and Araucanians in the deserts of northern Mexico and the
freezing wastes of southern Chile.

Such successful opposition could exist even in close proximity to the
most developed states of the day: the British state did not succeed
in "pacifying" the Gaelic clans of Scotland until after the Jacobite
revolt of 1745, and then only by repressive measures, the savagery of
which prefigured that of the anti-partisan wars of the twentieth
century. This long British failure was due not only to the rugged
terrain of the Highlands, but to the clans' mixture of highly
decentralized authority with fanatical loyalty to the clan chiefs. As
a result, the British authorities could rarely be entirely sure that
an apparent ally would not change sides at a crucial moment. Nor
without massive repression could the British force the highlanders,
or the Gaelic Irish clans before them, to accept the authority of the
new British state. In all these cases, the other side was
strengthened not just by ethnic and cultural loyalties but also by a
profound attachment to an ancient way of life menaced by modernizing
forces.

Afghanistan and Chechnya

Throughout its modern history, Afghanistan has displayed similar
features. As bewildered and infuriated Western journalists and
officials used to say of the Afghan mujaheddin: "You cannot buy an
Afghan--you can only rent him." The British-Afghan War of 1839-42
provides a perfect counterpoint to the contemporaneous First Opium
War. Whereas in China the defeat of the imperial army and navy led to
a settlement with the imperial government and an end to the war, in
Afghanistan in the 1840s, as in the 1980s, the multi-headed tribal
and religious resistance could not be defeated in this way because
there was no unified enemy authority with whom it was possible to
make a settlement; and because, unlike the unarmed and generally
passive Chinese peasantry, the Afghan tribesmen--like the Chechens
and Native Americans--were fierce fighters, honoring warfare, trained
to fight without mercy from an early age, and equipped with firearms
that (like the Kalashnikovs and rocket-propelled grenades of partisan
forces today) were extremely effective in certain circumstances.

In China, British officers at the time recognized the complete
impossibility of invading and occupying the inland provinces, given
the huge size of their territories and populations and the very small
size of the British Army and its Indian auxiliary troops.
Fortunately, there was no need to do so. When the British had
demonstrated their ability to move at will down China's rivers, and
had menaced Beijing, the resistance of the Chinese government
collapsed.

In Afghanistan, the British defeated the Afghan forces with relative
ease when the latter tried to defend the supposedly impregnable
fortress of Ghazni in a straight fight. But there, too, the British
did not have remotely enough troops to garrison the whole country. By
late 1841, the force based in Kabul consisted of only one British and
three British Indian infantry battalions and one cavalry regiment,
with other "armies" in Kandahar and Jalalabad. The British thus
relied on local Afghan allies, motivated by a mixture of political
allegiance to the British puppet ruler, Shah Sujah, and British
financial subsidies. Neither proved sufficient. The presence of
"infidel" troops in Muslim Afghanistan whipped the anger of religious
leaders and their followers into a frenzy. As so often occurs in such
situations (for example, as with U.S. support for the Shah of Iran),
infidel British support for Shah Sujah, far from strengthening him,
actually undermined his authority over his own people. For one thing,
the British could not check atrocities by Shah Sujah's followers
against their personal and political enemies. As well, the British
government of India--under pressure from the government and
Parliament in London to economize--finally refused the funds
necessary to strengthen the British camp at Kabul and cut the subsidy
to the tribes who controlled the British lines of communication.

The result of all this was the defection of the British allies, a
general rising, an appallingly bungled attempt to retreat from Kabul
across the mountains in midwinter, and the destruction of the entire
British force. In addition to all their other failings, the British
force had been encumbered by an immense train of camp followers,
servants and officers' luxuries--a syndrome not without its parallels
in the comforts the U.S. Army now expects when deployed in the field.

In 1842 a much stronger British Army returned to recapture Kabul and
wreak vengeance on its inhabitants: but the lesson had been learned.
Never again did the British attempt permanently to station troops in
Afghanistan, or reduce its rulers to the completely subservient
condition of the British Indian princes. Instead, the British relied
on a variety of "indirect" means to exert influence over Afghanistan.
This approach saved them from renewed disasters and the draining
financial and human costs of an endless war of occupation.

But it also meant that throughout the entire period of the British
Empire in India, Afghanistan was the source of a whole variety of
security threats: notably, the preaching of anti-British jihad by
local Muslim religious leaders (like the "Fakir of Ipi"), and the
continual raiding of British territory by tribal "bandits." During
the period of their presence on the Afghan frontier,the British tried
various responses to these dangers, from the use of locally recruited
elite troops (the famous Scouts and Guides) to seize individual
troublemakers, to aerial bombardment of villages and flocks, to
full-scale military expeditions.

By the 1920s, civilian casualties resulting from these tactics were
drawing strong criticism from the Labour Party in Britain, and from
socialist and pacifist groups abroad. More important, the effect of
all these responses tended to be both limited and temporary, because
it was largely pointless to try to improve things by putting pressure
on the government in Kabul. This government did not control its own
territory and its own subjects in the frontier areas.

This same set of problems confronted the Russians after their
withdrawal from Chechnya following Chechen victories in 1996. The
Russians pulled out completely (rather than trying, as a minimum, to
retain some sort of "security zone" in northern Chechnya) because
they believed that the Chechen commander, General Aslan Maskhadov--a
former Soviet artillery colonel with whom Russia had been negotiating
the previous year via the Organization for Security and Cooperation
in Europe--would prove an effective leader and would pursue a policy
of pragmatic cooperation with Russia. Unfortunately, while Maskhadov
did indeed win presidential election in January 1997 with almost
two-thirds of the vote, the destruction, militarization and
brutalization caused by the war, together with Chechen traditions of
resistance to authority and of raiding, meant that it proved
completely impossible to create any kind of effective Chechen state.
Instead, Chechnya became the base for a great wave of kidnapping and
raiding into surrounding Russian territory. Even more menacingly, a
strong group of international Islamic revolutionaries (led by a Saudi
Arabian who had fought in Afghanistan) established itself in Chechnya
and formed an alliance with Chechen warlords. The group's explicit
intention was to launch a jihad to drive Russia from the northern
Caucasus, and in August 1991 it did indeed invade the neighboring
Russian republic of Dagestan, leading to the second Russian attempt
to reconquer Chechnya.

By 1999 the great majority of Chechens were bitterly hostile to both
the Muslim radicals and the Chechen warlords--or, at least, that was
the view of the Chechen refugees with whom I spoke in December of
that year. The casualties inflicted by Russian bombardment have also
bitterly alienated most Chechens--just as the civilian casualties
inflicted by U.S. troops in Mogadishu infuriated even those Somalis
who loathed Farrah Aideed.

Unavoidable Conflicts

While the United States could withdraw from Somalia without any great
cost beyond damaged pride, there was clearly no way that in the long
term Russia or any other modern state could tolerate the existence on
its own frontier of the kind of threats that Chechnya represented
from 1996-99. In other words, although there has been much to condemn
in Russian policy and tactics, it is also true that Russia was faced
with a dilemma that we would do well to study.

Critics will say that the United States can always avoid such
conflicts by simply refusing to get involved in them, but to assume
that this will always be possible would be outrageously complacent. A
country that would spend tens of billions of dollars to create a
missile shield to help save North Korea from committing suicide,
while failing to prepare for urban and partisan warfare, would in all
likelihood be regarded by future generations as having established
new parameters for both paranoia and irresponsibility. Apart from
anything else, such an outlook would require the assumption that our
leaders will never make mistakes or engage in reckless military
deployments--a rather optimistic assumption, given the historical
record.

After all, U.S. administrations did not intend that U.S. troops
should be blown up in Beirut, shot to pieces in Mogadishu, or face a
possible ground war in Kosovo. And quite apart from the likelihood of
blunders, there may at some stage also be real and inescapable
reasons to intervene in areas of vital U.S. interest--for example, in
the Persian Gulf or Central America. Terrorism may also compel such
interventions, especially if at some stage terrorists gain access to
weapons of mass destruction. The existence of states that foster or
tolerate terrorism has led the United States in the last fifteen
years to bombard targets in Libya, Afghanistan and Sudan, and the
latest attack on the U.S.S. Cole has elicited vows of retaliation
from U.S. leaders. In the former Soviet Union, terrorist actions
helped lead Russia into its second invasion of Chechnya and are
becoming a source of Uzbek intervention in neighboring states. This
is the down side of the "global village": that it does indeed contain
universal threats and requires a measure of universal policing.

No army--least of all that of the world's hegemonic superpower--can
assume that it will not have to fight in cities. Since the beginning
of recorded time, cities have been the seat of government, wealth and
military power, and armies have had to besiege and capture them.
Today, as more and more of the world's population lives in cities and
the cities themselves continue to expand, it would be simply
ludicrous to draw up any military strategy that did not have urban
warfare as a major component.

But if anything, U.S. armed forces (with the exception of the
Marines) are adapting less, rather than more, for operations on the
ground involving relations with hostile or potentially hostile local
populations. Useful strides are being taken in producing equipment
for urban warfare, but there will always be a limited degree to which
weaponry can compensate for the risks, the brutality and the moral
ugliness of this kind of fighting.

Moreover, as the British have discovered in Northern Ireland, even in
a small territory, successful armed policing operations require very
large numbers of men. In Kosovo--also a very small territory--sheer
lack of manpower has been partly responsible for the failure of the
NATO forces adequately to protect the Serbian minority from their
Albanian neighbors. When it comes to high-technology weaponry, most
of the political talk of the U.S. military's unpreparedness is
nonsense; but when it comes to the inadequacy of present manpower
levels for repeated and prolonged emergency deployments, it is only
too accurate.

Perhaps most important of all are the intellectual and cultural
weaknesses of Western armed forces in general, and U.S. forces in
particular, when it comes to the armed policing of alien societies.
In part, this factor reflects worries about suffering heavy
casualties among ground troops, and (to a lesser extent) about using
ground troops to inflict heavy casualties on civilians--as may well
be unavoidable on occasion. As well, it reflects the sheer lack of
knowledge and insight that is inevitable when officers and officials
go into a place on six or even three-month contracts--the norm for
Western officials in Kosovo. Old-style Western colonial officials had
their faults, God knows, but they did at least spend their entire
working lives in the countries they were governing.

Misplaced Rationality

The passionate desire of so many U.S. military thinkers to believe in
the efficacy of air power alone reflects an equally passionate desire
to avoid thinking about this kind of issue. Perhaps almost as
important as the fear of casualties is the desire to impose order on
warfare, which is alien to the essentially chaotic and irrational
nature of war itself, but which appeals to old tendencies in the
military mind and military culture. These tendencies have only been
strengthened by the computerization of the U.S. military machine,
with its impression of control, rationality, cleanness and distance.

From this point of view, one is concerned to learn of the U.S. Army's
intention to rely on the RAND Corporation to help draw up some kind
of "general doctrine" of urban warfare. While a purely military
doctrine concerning methods of isolating and storming a citymay well
be useful, it should already be obvious to the U.S. military that the
great majority of future urban operations will not be straightforward
military battles like Stalingrad, but will involve strong elements of
politics and partisan warfare. Given the enormous differences between
societies, a universal doctrine would be of little use. The key
factors must be a willingness to learn from experience and to listen
to outside advice as needed. Today the best teachers of the U.S.
military concerning urban warfare would be former Vietcong, Somali
and Chechen urban fighters--and they are not impossible to find. Yet
it seems that too many members of the military hierarchy and its
advisers have learned nothing and forgotten everything from Vietnam.
The reason for this is not individual foolishness, but the military's
very nature as a modern, Western, "rational", bureaucratic
machine--together, of course, with deep Western cultural and social
trends.

According to Christopher Coker,

In the course of the twentieth century, even the vocational worlds
which included war became increasingly bureaucratized. Soldiers
became technicians. The nuclear age required the military to become
more bureaucratic in their thinking than ever before. The main
virtues of bureaucracy, predictability and consistency, were
especially valued by the military in the nuclear era. No wonder the
generals came to hate their own profession.

Quite frankly, a career spent climbing up the ranks of an ordered,
hierarchical, obedient military bureaucracy in peacetime may equip a
man for the technical and organizational aspects of warfare, but it
is often an extremely poor psychological and intellectual preparation
for any sort of command in the field. It naturally tends to diminish
a capacity for flexibility and improvisation; to encourage a belief
in and liking for what in these pages Andrew Bacevich has called
"stylized warfare"; and to produce what Ralph Peters, a retired
officer and leading military thinker, has unkindly referred to as
"Stepford Officers." That is why every revolution or national upheaval has thrown up men like Leon Trotsky or Shamil Basayev who, with no formal military education at all, prove much better commanders than most "professionals."

This tendency toward military theorizing and planning is, of course, linked to very real and immensely useful advances in technology. But it can also be seen as linked to wider developments in the intellectual and academic sphere, one aspect of which is "rational choice theory." Indeed, this discipline has had a direct link to the military, through its role in the formulation of the theory and practice of strategic games and weapons development for major war. In this limited respect, its role has been a useful one. There are, however, two major dangers that rational choice theory poses to the military. These are in essence the same dangers that have been noted in the application of rational choice theory to international relations.

The first is the illusion of complete control and understanding, encouraged both by the supposedly mathematical and "scientific" basis of the approach, and by the supposedly limited number of possible developments and outcomes. This has obvious analogies to the belief in military circles in complete understanding through technological surveillance and analysis. As has been pointed out by Stephen M. Walt and others, rational choice theory is seriously out of sync with the reality of international relations, especially in times of rapid change, insecurity and crisis. It is really only suited to periods when the international situation is settled and orderly, and when the number of actors is limited, is made up of known states and state leaderships, and does not include new and unpredictable non-state forces acting "from below."

This, of course, makes it even less appropriate to warfare and, above all, to internal civil wars involving numerous players whose ultimate allegiance and intentions are sometimes unknown even to themselves. Thus, observation technology would not have been especially useful at the Battle of Bosworth Field in 1485, where the key issue was not the strength or weaponry of the opposing forces of King Richard III and Henry Tudor, but the intentions of a third, uncommitted force--that of Lord Stanley, who did not make up his mind until after the battle had actually been joined. Bosworth has almost exact parallels in the dilemmas of Russians trying to deal with Chechen warlords of uncertain loyalties who are capable of shifting from one side to another without warning.

The second weakness of rational choice theory is its rootedness in a particularly narrow version of late twentieth-century American culture--limited even in American terms because, after all, there are streets not ten blocks from Columbia, Yale or the U.S. Congress whose inhabitants have a better natural understanding of how things work in Mogadishu or Grozny than do most international affairs experts. As Chalmers Johnson and E.B. Keehn have noted, "this ideological orthodoxy is almost surely one of the less pleasant, unintended consequences of the end of the Cold War and Americans' perceptions that they 'won' it."

Inadequate even to explain the behavior of many modern states (the United States included), rational choice as defined by its U.S. practitioners is comically, grotesquely irrelevant to the behavior of people from radically different cultures. In terms of American educated middle-class behavior, there was nothing at all "rational" about the actions of the Chechen fighters I met in 1994-96, or of the Somalis who immolated themselves during the U.S. intervention.

The influence of rational choice theory tends to make Western analysts and planners poorly equipped to understand such societies. This is not just because of intellectual narrowness and prejudice, but because, by encouraging concentration on supposedly universal, "scientific" bases of analysis, it discourages students and career-makers from developing knowledge of local cultures and languages. This weakness is of special importance among armed forces whose strategy involves not the seeking of decisive battle, but less direct ways of bringing the enemy to surrender--for such a strategy is critically dependent on good political intelligence and analysis.

In armies that have had to fight ground wars at regular intervals, the influence of experienced fighting soldiers counteracts to some extent that of the military bureaucrats and technicians. It provides a continual reminder that the qualities which go to make up a good fighter are not necessarily those taught in formal military courses; and that in war there is always the possibility of a clash escaping the control of commanders and becoming a "soldiers' battle", in which what really counts is a mixture of tactical skill, discipline, flexibility, quickness of decision, coolness under fire, endurance and courage. The U.S. military has always been able to relearn these virtues when it had to, but it is a pity that this often happened so late, and that it was so often our enemies who were the teachers.

Anatol Lieven is a senior associate at the Carnegie Endowment for International Peace. His book, Chedmya: Tombstone of Russian Power, was published in paperback by Yale University Press in 1999.