Thinking Asymmetrically in Times of Terror


© 2002 Colin S. Gray

From Parameters, Spring 2002, pp. 5-14.

In American common usage today, asymmetric threats are those that our political, strategic, and military cultures regard as unusual. Such threats differ significantly in character both from those that we anticipate facing from putative enemies and from the methods with which we plan to menace them. Much as international lawyers thus far have failed to define terrorism to the general satisfaction, so US national security specialists have found that the endeavor to define asymmetric threats has proved generally unproductive. Borrowing from the terrorism case, the most fruitful approach to the better understanding of asymmetric threats is not via a forlorn quest for the perfect definition, but rather by the identification of the principal characteristics of, and corollaries to, asymmetry.

Characteristics of Asymmetry

A problem with efforts to define an asymmetric threat is that they imply strongly that the universe of threats divides neatly into the symmetric and the asymmetric. Indeed, by definition we can make it so. Of course, this is at best misleading, if not downright nonsensical. Notwithstanding the apparent clarity of some cases, there is no more definitive a universal test for what is an asymmetric threat than there is for who is a terrorist. If one person's terrorist is another's freedom fighter, so one culture's asymmetric threat is another's standard modus operandi. Let us proceed by listing the characteristics of, and usual corollaries to, threats we generally deem to be asymmetric.

Asymmetric threats tend to be:

. Unusual in our eyes.

. Irregular in that they are posed by instruments unrecognized by the long-standing laws of war (which are keyed to control the conduct of regular military machines engaged in open combat).

. Unmatched in our arsenal of capabilities and plans. Such threats may or may not appear truly dangerous, but they will certainly look different from war as we have known it.

. Highly leveraged against our particular assets--military and, probably more often, civil.

. Designed not only to secure leverage against our assets, but also intended to work around, offset, and negate what in other contexts are our strengths.

. Difficult to respond to in kind. This is less true than we usually allow. For example, special forces can be unleashed to operate as "terrorists in uniform." Unconventional warfare of all kinds, including terrorism (and guerrilla operations), is a politically neutral technique.

. Difficult to respond to in a discriminate and proportionate manner. It is of the nature of asymmetric threats that they are apt to pose a level-of-response dilemma to the victim. The military response readily available tends to be unduly heavy-handed, if not plainly irrelevant, while the policy hunt for the carefully measured and precisely targeted reply all too easily can be ensnared in a lengthy political process which inhibits any real action.

. Friendly to the frightening prospect of the "unknown unknown." By analogy, if we do not scan the skies, including those of the southern hemisphere, comprehensively and routinely, we will probably not spot the asteroid (or other "Near Earth Object") that poses the ultimate asymmetric menace to our security. But even a superior defense community is going to miss some "unknown unknowns." We do not look for what we do not know to look for.

Undoubtedly some works of frontier social scientific scholarship one day will dissect the concept of the asymmetric threat and argue that it has N categories, Y subcategories, and who knows how many intriguing, and not wholly implausible, variations. To be useful to US policy, however, an understanding of asymmetric threats should focus only upon the core of the matter. In addition to the eight broad and overlapping characteristics itemized above, it is only a special class of asymmetric menace that need attract official US concern today. Specifically, the United States is interested not simply in threats that are unusual, different, or designed to evade American strengths. Instead, the United States has to focus on threats, which in this case happen to warrant description as asymmetrical, that if executed could wreak great damage upon American interests. In other words, it is not sufficient just for a threat to be different, also it would need to be prospectively effective. Many candidate asymmetric threats are not threats to achieve a measure of physical control, but rather work ju-jitsu fashion with the inadvertent cooperation of the victim. It follows that the effectiveness of those threats is not some absolute quality and quantity, but is very much ours to determine. This points to a general truth about the strategic utility of terrorism in particular, with the same rule applying to perpetrator and victim.

Typically, terrorists win when their outrages, though generally very minor as compared to the extraordinary events of 11 September or the kind of costs inflicted by regular warfare, induce the state-victim to overreact. The regular belligerent takes action which fatally imperils its own political legitimacy. Similarly, terrorists lose when their outrages delegitimize their political cause; this is what can be termed the mainstream strategic explanation of why terrorism succeeds or fails in particular cases. It is all but impossible for terrorist organizations themselves to inflict truly major physical damage upon the capabilities of states. For the parallel point, it is close to impossible for the forces of counterterrorism to root out all of the would-be warriors-by-terror. Each side usually has to be encouraged to defeat itself politically. The historical record on these points is quite clear--indeed, is overwhelming--though it is less well understood than it should be.

Lest there be any misunderstanding, I am not suggesting that the "regular" state party to an asymmetrical conflict should tolerate terrorist outrages. Underreaction, let alone no reaction, most likely would be interpreted as weakness, perhaps as evidence of successful intimidation. Nonetheless, when dealing with terrorists, a low-key response usually is preferable to heavy-handed action which both dignifies the enemy by signalling a large significance to his misdeeds and risks alienating political opinion. Though the attitudes and policies of most irregularly asymmetric foes will be accessible by carefully crafted threats and bribes, usually there is a hard core among the adversary whose mode of rationality literally will brook no compromise. In those instances, the rejectionists (of any compromise whatsoever) either have to be killed or placed in permanent detention. Ideally, the messy task of dealing with the hardest of hard-core among irregular enemies can be left to their former comrades-in-arms (as happened in Ireland, for example, in the early 1920s).

It will be abundantly clear by this juncture that for our current policy and operational purposes the asymmetrical threats of most interest are understood to be militarily, or even quite non-militarily, irregular in character. Nonetheless, we can conceive of asymmetrical threats very different indeed from menaces posed by irregular forces. Most obviously, the United States has to be ready to cope tactically, operationally, and strategically with the smart and unusual employment of regular armed forces by an enemy. Not all of America's foes, current and future, lack a regular military machine. Tactically, operationally, and strategically adroit belligerents use their regular forces in unexpected ways.

It is not obvious that smart tactics, refined operational artistry, and adroit military strategy warrant the ascription "asymmetric," even though they can manifest themselves in "different" behaviors. Indeed, careful reconsideration of the whole subject area of asymmetric threats, and responses to the same, leads the theorist and the practitioner at least to the working conclusion that good strategy on both sides is what this is all about. Because choices for asymmetric activity merge with common-sense approaches to strategy (e.g., doing what the enemy does not expect, generally practicing the precepts advanced by Sun Tzu), there is virtue in fencing off for distinctive discussion the phenomenon of hugely irregular asymmetric threats. This is not to suggest that asymmetry in conflict is synonymous with belligerency between regular and irregular foes, or between regular and irregular forces (regular security communities can license and employ irregular forces and methods), but that is the core of our current concern. How do asymmetric threats work? To repeat, although we must not equate such threats strictly with terrorism, by and large it is terroristic behavior that is the focus here.

. Asymmetric threats work by defeating our strategic imagination.

Every security community is the prisoner of its own strategic expectations. Recall that efforts at strategic deception tend to work when they show enemies what they expect to see. Our historical experience, culture, and geopolitical context, as well as the practical constraints of government (limited information, time, money, flexibility), direct us to prepare for some contingencies, but not others. We prepare against threats that our community agrees consensually merit contingent responses. It may be unjust, certainly strictly inaccurate, to identify failure of imagination as the strategic culprit, when really the problem reduces to knowing how to act in face of the full array of imaginative possibilities. More often than not, the difficulty lies not so much in a failure of imagination--someone will have thought of the threat at issue--but rather in an understandable failure of confidence in imaginative threat identification.

Experience suggests powerfully that the US defense community, with its hundreds of planning staffs, study groups, and respected theorists, has little difficulty imagining dire asymmetric threats (e.g., to the twin towers in New York City, a target previously assaulted unsuccessfully). The problem lies in locating decision rules to filter threats worthy of serious attention from the rest. Even the wealthiest country on Earth cannot afford to invest in protection against all conceivable threats.

. Asymmetric threats work by posing possible menaces so awful and awesome that countries dare not respond, at least not until actual experience provides incontrovertible evidence of the threat.

It is well worth remembering that the asymmetric danger leveled by terrorism can work strategically only with the unwitting cooperation of the victims. If we permit acts of terror to spread fear, despondency, and drive us into a variant of a garrison state, then at best we accept a very high price as the cost of living with this asymmetric foe. At worst, and this is the strategic logic of the terrorist, we find the responses we have initiated to counter terrorism so burdensome that we become discouraged and amenable to effecting a political deal (always assuming, of course, that our asymmetric foe is "dealable," which he may not be). Note that the (Provisional) IRA has bombed and shot its political wing, Sinn Fein, into government in Northern Ireland. While, for the least ambiguous, if in the long term historically ironic, example, Jewish terrorists bombed and shot the state of Israel into existence, as they rendered Britain's mandate over Palestine unsustainably costly.

Because most imagined threats do not occur, it follows that most of them can be safely ignored. Of course, it can be difficult to know with high assurance which threats can be ignored with impunity, and which cannot. Such "acts of God" as giant tsunamis (e.g., triggered by the collapse of a mountainside in the Canary Islands) or collision with a Near Earth Object, tend to be classified in official and popular minds along with mass bio-terrorism and even nuclear missile strikes, as events so awful as to be all but beyond prudent policy response. Apart from the obvious danger of public panic, which may be gratuitous (since nothing can be truly certain until it happens), the difficulty and cost of suitable anticipatory responses are self-deterring. Even when the asymmetric threat approaches both high plausibility and amenability to a fairly reliable solution--as, for example, with the menace of rogue missile attacks--government and public are likely to opt for the non-response of psychological denial. After all, it may never happen.

. Asymmetric threats work by challenging successfully our ability to respond effectively.

By its nature the executive agency for asymmetric threats, and possibly the political force behind that instrument, will be dissimilar to us. Ideally, from his point of view, the purveyor of asymmetric threats does not leave a business card with an address at the scene of the crime. The highly irregular warriors of asymmetry can succeed tactically only in the mercifully rare cases when they are indifferent to personal survival, or when they can merge anonymously into the urban human mass or into forbidding physical terrain. Since strategy is not solitaire, even a country as powerful as the United States requires that its enemies have map coordinates as a necessary condition for chastisement. Although irregular foes generally can function only with the willing or coerced acquiescence of host polities, it is by no means an elementary matter for the United States to drain those particular swamps, as the popular pejorative expression has it.

Among other difficulties: the state-swamps at issue are inhabited by many people deemed to be innocent; they will have civilizational affiliates elsewhere, some of whose official and popular opinion we will need to take seriously; and operational problems most likely would make a mockery of robust intentions and muscular language (e.g., draining the swamp) on our part. It is not sufficient for American responses to asymmetric threats to be effective; in addition, they must be politically and morally tolerable in our culture. The Roman Republic and Empire devised and practiced exceedingly brutal standard operating procedures against irregular foes, domestic and foreign, that were extremely effective. Those procedures could not be followed today by our society in the contexts of the laws of war (as revised, to accommodate internal strife) and the CNN factor.

Americans will need to decide whether asymmetric foes are criminals or enemy soldiers. If we redefine what the concept and legal idea of "war" encompasses, then so also will we have to redefine who can wage it legitimately, which is to say who, and what kinds of behavior, enjoy some legal recognition and protection. In addition, there will have to be reconsideration of the precise meaning of a distinction that has been fundamental to the development of the laws of war, that between combatants and noncombatants ("innocents"). At present the civilized world is trapped somewhat in a timewarp of arguably obsolescent political, ethical, and strategic assumptions and practices. Had three thousand Americans been killed on 11 September 2001 in a regular attack by the conventional forces of a state-enemy, the US response would have been swift and bloody indeed. Given the terrorist nature of the attack, the US defense community had to adjust to an unfamiliar strategic context. There is a considerable danger that today's new (sometimes asymmetric) menaces will be addressed by thoughtways and operating procedures of unduly conventional character.

. Asymmetric threats work by acting against what appear to be our strengths.

Bearing in mind the restricted domain allowed asymmetric threats in this discussion--confined largely to the terroristic outrages committed by the physically relatively very weak--it is the symbols, the apparent exemplars, of our strength that must attract the hostile strategist of asymmetry. In comparatively minor key, the attack on USS Cole in November 2000, and in truly major key, the assaults on the World Trade Center and the Pentagon, both illustrate our argument all too clearly. Notwithstanding even horrific scenarios, the Osama Bin Ladens of this era cannot wreck US global military or financial hegemony, or the political context which lends a widespread legitimacy to that preeminence--only ill-judged US policy itself can do that. But the asymmetric threat posed by expressive acts of terror can occasionally succeed in inflicting damage on a scale and of a kind that could be truly damaging to US political prestige in the world.

While US policy and operations must seek to prevent and, if need be, thwart, acts of terror, the impracticality of achieving permanent 100-percent protection (of what?) suggests the wisdom in a policy which scores well at political damage limitation. Given the very restricted physical damage that most asymmetric threats could pose (weapons of mass destruction generally are another matter), we have to think innovatively about ways to minimize loss of prestige when such outrages succeed tactically (as they will, from time to time).

Fortunately, our problem is noticeably strategic in form. The Osama Bin Ladens are not literally madmen. They are highly intelligent, resourceful, and bent upon acting in ways that, in their reasoning, will have beneficial effects. If we are to perform competently in deterrence we need to address empathetically the issue of how, by our policies, we can negate the political effects of tactically successful terrorism.

Some Working Propositions

Let us now turn to how we can best prepare for and shape our responses to these threats.

. We cannot predict specific asymmetric threats (unless we have excellent intelligence) and therefore we cannot protect everything at risk.

What this means is that, as in any war, the friendly side will take losses. While the United States should do all that it can, consistent with maintenance of decent standards of behavior, to make life difficult for would-be terrorists, this character of asymmetric conflict is peculiarly unrewarding to careful defense. The reason should be obvious. Simply stated, we and our friends and allies offer too many targets around the world for preclusive protection to be anything other than a worthy policy goal. Deterrence will be especially important, despite the likely fact that it will be unusually (culturally) difficult to achieve. A confident assumption that Americans are very good at deterrence--witness the course and outcome of the Cold War--needs to be jettisoned forthwith. We are urgently in need of culturally sophisticated profiles of asymmetric foes, so that we may stand some chance of understanding what might best discourage them from proceeding.

. We tend to lock onto yesterday's event and project it forward as the menace of the era.

It is of the essence of the irregular, asymmetric threat that it will not comprise a replay of yesterday's outrage (though the World Trade Center was attacked more than once). We must not give the impression that we believe that our asymmetric enemies always will be successfully cunning and proficient. From time to time they will succumb to unduly routinized, "regular," and conventional thinking, they will behave incompetently, and "friction" of several kinds can thwart them. All of which is both true and somewhat comforting to recognize, but alas it cannot serve as the basis of our policy. Bureaucracies--military and civilian--and indeed any hierarchical organization which rewards rule-following, are inherently ill-suited to think innovatively about asymmetric threats. The US armed forces have handfuls (no more) of people amongst their substantial special operations forces who truly can think "outside the box," and who can reason and, if need be, behave like "terrorists in uniform." It is not likely that even an elite group of US officials blessed with relatively unconventional mindsets would offer much of value with respect to specific asymmetric threat-spotting; there are just too many possibilities out there. But at the least such a group should be able to frame an intelligent generic strategy for response, and therefore deterrence.

. Although we are not likely to perform well at the identification of very specific dangers, we should be able to identify, and therefore plan how to protect against, the kind of threats that would do us major harm.

This thought really is complementary to the merit in the idea of trying to access the enemy's culture. Since his exact operational choices are likely to remain a mystery to us, we at least should know what we value most, and take measures to afford such protection as is feasible. The lore on sound principles for the guidance of defense planning includes the injunction not to avoid being surprised, but rather to avoid serious damage from the effects of surprise. The more one thinks about the problems of coping with asymmetric threats, the more relevant do traditional, historically founded approaches to defense planning appear to be.

. We need to be especially alert to the possibility that asymmetric threats can wreak their greatest damage through ill-judged measures of response that we ourselves choose to undertake.

While we do need to worry about, and plan to prevent, the damage that asymmetric threats might cause, we have to be particularly alert to the danger that relatively minor physical damage inflicted by terrorists may be translated--by us--into truly major societal and economic costs as we dignify the asymmetric belligerent by overreacting. If decisive action against asymmetric threats is possible, ideally after the fashion of defeating piracy by burning out the pirates' lairs, all to the good. However, the challenge to US policy lies not so much in those cases where there is a military option, but rather when there is not. The temptation to do something, for the sake of being seen to be doing something--even something strategically stupid--can be politically irresistible.

One should not forget a basic rule outlined above: the terrorist (as an asymmetric opponent) can succeed only with our assistance. He lacks the resources himself to inflict significant direct damage upon us. Even if armed with weapons of mass destruction (WMD), the scale of the terroristic asymmetric menace is but a pale shadow of the damage that the superpowers might have inflicted in a World War III. This is not to suggest that terrorists with WMD are insignificant--far from it. But it is to say that we need to keep a sense of proportion. There may be a great Sino-American struggle in our future, a possible conflict which, in its potential for harm, would demote today's roguish terroristic perils to the second-order problems that they are, historically. Less than 20 years ago we faced some danger of a war wherein casualties easily could have numbered in the many millions. Without diminishing the tragic loss of three thousand lives, it cannot be strategically sound for America to allow itself to be permanently traumatized by such an outrage, or consequently to recast its national security policy on a grand scale.

. We need to identify and think hard about threats to which we lack obvious responses.

In effect, this point advises asymmetric, even unconventionally irregular, approaches on our part. We have to learn to respond differently, but effectively, to threats which cannot be answered in kind. The United States has to ask imaginatively what it is that its asymmetric foes value highly, and devise ways and prepare means to hurt those values severely. If there are cultural barriers on our side to incorporating particularly murderous options into our policy, strategy, or operational intent, then we may need to reconsider some of our attitudes and rules of engagement. After all, war is war. Combat against terroristically asymmetric foes is likely to be about as far removed from the "clean" conditions of, say, war at sea or in the desert (where there are no civilians) as can be imagined.

A Skeptical End-Note

In the history of strategic ideas, the contemporary American fascination with asymmetry comprises rediscovery of the stunningly obvious. To behave in ways different from those expected by an enemy can be simply good tactics, operational art, and strategy. Since asymmetrical merely means different, it is a little hard to understand quite why the notion has been elevated as the latest fashionable Big Idea (following on from the concept of a revolution in military affairs). In this essay I have confined asymmetrical threats to those emanating from an irregular foe. However, with equal, if not greater, justification I might have set out to diminish this Big Idea by pointing out that all of America's wars have been asymmetrical contests. Even aside from the bloody, two and a half century-long experience of struggle against native American irregulars, when has the country waged a plausibly symmetrical conflict? Imperial Britain was radically different--grand strategically--from the revolting Colonies, as the Confederate States were from the Union, as Germany, Japan, and then the Soviet Union were from the United States in the 20th century.

Defense and war planning always have a significant asymmetrical dimension, which should find expression at every level--tactical, operational, and strategic. Competent tacticians, operational artists, and military strategists are obliged to be aware of salient actual and possible asymmetries. In fact, the quality of being or behaving differently--which is all that asymmetrical means--is so natural to effective defense professionals that they can be excused wondering why the US defense community today is so excited by the concept. Historically assessed, symmetrical warfare has been the rare exception, not the rule. Belligerents always differ from each other, usually in ways that are or could be strategically significant.

It is entirely admirable for the US defense community to recognize the potential importance of asymmetry. This recognition should help offset the peril of indiscriminate strategic autism to which very great powers are prone. A less happy consequence of the current fascination with asymmetry is the imputing of extraordinary efficacy and significance to it. To a greater or lesser degree, all tactical, operational, and strategic behavior is asymmetrical. There are no identical belligerents, with identical forces, who behave identically. But to listen in to the current American defense debate is to hear senior officials talking as if they had just discovered extraordinarily dangerous asymmetrical enemies who pose similarly awesome asymmetrical threats. To be different, or to behave differently, is not necessarily to be strategically effective. There is nothing inherently strategically magical about different--i.e., asymmetrical--behavior. There is some excuse for journalists who become overexcited when exposed for the first time to the apparently new Big Idea of asymmetry, but we defense professionals should know better. From the time of its founding, the United States repeatedly has waged war asymmetrically, as it was obliged to do against a series of "different" enemies.

A little reflection reveals that asymmetry essentially is a hollow concept. As a relational variable, that which is asymmetrical can be labeled as such only with reference to that which is symmetrical--and what is that? The concept may have some limited merit if it is corralled, as in this essay, with a carefully specified meaning (focused on an irregular foe favoring terroristic activity). As a contribution to the general lore of strategy, however, asymmetry is a complete non-starter. Given that competent American military planners have always plotted how to defeat particular enemies in the distinctive ways best suited to the individual cases at issue--albeit in ways preferred by American strategic and military culture--what exactly is novel or even especially interesting about the concept of asymmetry? Because all warfare is asymmetrical (there are no sets of identical belligerents), in effect no particular wars or warfare is distinctly so. In this respect, a course of instruction on "asymmetrical warfare" would be content-free.


An earlier version of this essay was prepared as a contribution to the study of asymmetric threats being led by Dr. Steven Lambakis for the National Institute for Public Policy, Fairfax, Va.

Dr. Colin S. Gray is Professor of International Politics and Strategic Studies, and is Director of the Centre for Strategic Studies, at The University of Reading, England. He is also European Director of the National Institute for Public Policy. He took his B.A. in government at Manchester University and his D.Phil. at Lincoln College, Oxford University. His public service record includes five years as a member of the President's General Advisory Committee on Arms Control and Disarmament (1982-87). Of recent years his work has included studies on technology and strategy and on strategy for space. His most recent book is Modern Strategy (Oxford: Oxford University Press, 1999). In 2002 he will publish Strategy for Chaos: RMA Theory and the Evidence of History (London: Frank Cass, forthcoming).

Reviewed 6 March 2002. Please send comments or corrections to