The American Perception of Substate Threats

Taking a very long view, it quickly becomes apparent that the various troubles that today are labeled as substate threats are not at all new. The ancient world exhibited them all. The Roman Empire—the superpower of its day—contended with insurrections and rebellions. It dealt with lawlessness and absence of authority on its borders. The Romans certainly knew civil war, which led to their own transition from republic to empire. There was terrorism, too, especially religiously inspired terrorism, as indicated by some of the vocabulary we associate with it today. The original Zealots were members of a Jewish millenarian sect that fought against Roman rule in Palestine.²

If we take a not quite so long view, there is no clear trend of increasing substate threats. There probably is no such trend at all. The historian Niall Ferguson offers several explanations, which are grounded in changing economic and political structures, for why in most of the world (but perhaps not the Middle East) the 21st century is unlikely to be as bloody as the 20th century was.³ Interstate wars are a big part of that picture, of course, but so is substate conflict. Much of Ferguson’s analysis, drawing on things such as the degree of ethnic heterogeneity within nation‐​states, is about the prospect for less substate violence.

The monsters that the United States has searched for and confronted have taken a variety of forms, from single hostile states to whole categories of trouble that constitute much of what today comes under the heading of nonstate threats. There was the Axis of World War II, of course, which was replaced quickly by the Soviet empire. During the four decades of the Cold War, the Soviet Union served so reliably as the arch‐​threat in the American worldview that monster‐​searching did not go much beyond looking for Moscow’s hand in mischief and mayhem around the world. The sudden end of the Cold War, which some call World War III, brought a decade of fumbling by pundits for a new threat‐​defined way of describing the role of the United States. The terrorist event of September 2001 led many of those pundits to believe that way had finally been found, as the administration of George W. Bush declared a global “war on terror.“⁴

Since then, much discourse about other substate threats has been linked to the more specific issue of terrorism. For some, the United States has been in “World War IV,” a grand struggle against multiple manifestations of Islamic extremism.⁵ Others, who do not reduce U.S. national security challenges to such simplistic terms, nonetheless tie sundry nonstate difficulties to terrorism in other ways. Insurgents are categorized according to whether or not they are associated with al Qaeda. Internal disorder is looked at worryingly as providing a potential new haven for terrorists. A frequent question asked about criminal organizations is whether they have ties to terrorists. The distinction between countering terrorism and countering the proliferation of certain categories of unconventional weapons gets blurred—or obliterated altogether as Bush did in endeavoring to tie his case for invading Iraq to the “war on terror.“⁶

The conceptions of nonstate threats embodied in all of that thinking are reflections more of a long‐​standing American worldview and American habits of conceiving the U.S. role in global affairs than they are of characteristics of the threats themselves. Accordingly, a disconnect exists between how the threats are usually treated in American debate and how much of a danger they actually pose to U.S. interests. The disconnect is most often in the direction of threats being overrated. Understanding that disconnect requires a closer examination of each of the fundamental ways in which substate conflict can endanger U.S. interests. One way is through the overthrow of an incumbent regime, thus leading to the advent of an unfriendly government in its place. A second concerns threats that substate actors can pose more directly, without a change in regime. A third possibility is trouble stemming not from a particular regime or group but instead from internal disorder itself.

The fear that an insurgency or rebellion might overthrow a government that is relatively friendly toward the United States and replace it with an unfriendly one dominated U.S. thinking about substate conflict during the Cold War. It drove U.S. policies toward civil wars in Africa, Southeast Asia, and elsewhere. The prime U.S. objective, simply stated, was to prevent the establishment of new pro‐​Soviet regimes. The objective was not to tamp down the conflicts and save lives. The resulting policies entailed provision of aid to incumbent regimes, sponsorship of alternative resistance movements such as in Angola, and—in the extreme case of South Vietnam—large-scale U.S. military intervention.

Within the context of a global competition with the Soviet Union, that approach at least had a superficial logic to it. If one perceives a largely zero‐​sum contest in worldwide power and influence, and if the contestant on the other side is active across the game board in exploiting internal conflicts and aiding its own clients, pushing back at all points on the board seems to be an apt response. The coming to power of new regimes through insurrection and civil war and the alignment of those regimes with either the United States or the Soviet Union were entries on the scorecard of the global Cold War game.

That way of looking at substate conflict had deficiencies that were not widely recognized during the Cold War. Chief among those deficiencies were that worldwide communism was far from monolithic and that—even within the Cold War context—competition between Washington and Moscow in violence‐​wracked regions was not zero‐​sum. The dynamics of civil war and insurgency helped keep that competition from being zero‐​sum. Resistance movements that are in the pocket of an outside power are less likely than others to obtain the legitimacy necessary to establish a stable new regime, barring direct military domination by its patron.

Conversely, movements that have such legitimacy are more likely to have independent foreign policies once they gain power. Tito and his partisans in Yugoslavia and Ho Chi Minh and his Viet Minh both demonstrated that approach. Tito’s break with Moscow was sharp enough that even myopic Cold Warriors recognized that fissure in the communist world, but some U.S. policies seemed oblivious to other fissures, including those between North Vietnam and both China and the Soviet Union. Today, we hear echoes of monolith‐​perceiving Cold War thinking in the “World War IV” notion that the United States faces a radical Islamist adversary that would use the establishment of any new regimes as steps in the creation of a new caliphate.

Other deficiencies were not unique to the Cold War and should serve as qualifications to any assertions made today about dangers of substate conflicts bringing hostile regimes to power. It is difficult to look at an insurgency or any other kind of resistance movement and to make accurate predictions about the orientation and foreign policy of any regime into which the movement may convert itself in the future. The difficulty is partly a problem of making accurate assessments amid the fog of civil war. It is partly a matter of the inherent difficulty of foretelling how someone will behave when making future decisions.

It is also partly a matter of ends and means being two different dimensions. It is possible for a movement to use extreme methods to gain power and then pursue relatively moderate policies once in power—or vice versa. Some regimes that were established partly through the use of terrorism (e.g., Israel or post‐​apartheid South Africa) became friends or partners of the United States. Some that came to power through more peaceful means have come to be viewed as nasty adversaries; think of the Iranian presidency of Mahmoud Ahmadinejad. And, lest we forget, Adolf Hitler became chancellor of Germany after his Nazi Party won a large plurality in a free election.

Claims that the behavior of future regimes is more predictable than that are mostly exercises of hindsight. Looking at internal conflict in real time, we see it is often hard to distinguish agrarian reformers from totalitarian communists—or strong leadership from vile fascism. How, for example, should one have regarded Meles Zenawi and his Tigrayan People’s Liberation Front while still fighting a civil war, before a victory in that war led to Zenawi’s 17 years in power as ruler of Ethiopia? It was not clear. On one hand, almost anything might have been better than the military junta that Zenawi’s movement displaced. On the other hand, that movement had an ethnically narrow base and was led by a man with a Marxist‐​Leninist background. The final result was mixed: a regime that had human rights problems but that the United States came to see as a useful partner in dealing with security problems in the Horn of Africa.

Assertions that the horrible experience with the Nazis could easily have been foreseen and was “all in Mein Kampf” reflect hindsight; they also reflect how that one traumatic piece of history has done much to stimulate the ringing of many later alarm bells, including ones about how particular movements would pose a threat if they were to gain power. The Nazi rise has repeatedly been invoked, probably more than any other historical episode, as an analogy in sounding alarms about purported threats, including threats from nonstate movements. The analogy figured prominently in deliberations in Lyndon Johnson’s administration before the U.S. intervention against the communist insurgency in Vietnam.⁷ The tendency to overestimate one’s ability to forecast threatening behavior of a group not yet in power is sometimes a matter less of historical comparisons than of confidence in one’s gut sense. Alabama governor and presidential candidate George Wallace said in the late 1960s, looking back a decade, that “any cab‐​driver in Alabama could have told you, just by looking at him, that Castro was a communist.” But the Cuban leader’s foreign policies were less predictable while he was putting together an insurgency in the Sierra Maestra; his later association with communism was more a function of the choices and situations he faced once in power.⁸

Besides such unpredictability, U.S. policies toward many substate conflicts will understandably be diffident because, even if the behavior of the contenders once in power could be foreseen, U.S. interests do not all run in the same direction. What outcome of a domestic conflict the United States should want is not always evident. The upheavals in the Middle East known collectively as the Arab Spring have presented such situations. If U.S. policy toward Egypt, for example, has appeared vacillating, that is partly because different possible outcomes of the struggle for power in Egypt would serve in contradictory ways the different U.S. interests involved, including privileged access to the Suez Canal and the expansion of democracy as a way of discouraging radicalization. The civil war in Syria has also presented no clear answer to the question of what outcome would be most consistent with, or least damaging to, U.S. interests, given that there is much to dislike on both sides of that struggle.

Running a government, as well as a country, is starkly different from running a resistance movement. Incumbency involves demands and equities that strongly shape the motivations of leaders and constrain what they can do, at least without sustaining major costs and losses. That difference is the biggest reason for disconnects between the ideologies and behavior observed in a resistance movement and the policies followed once the same movement is in power. Happily, the disconnects are mostly in the direction of greater moderation when in power. The fundamental reason is that a regime has more to lose than an opposition group does. It has national interests, including strategic and economic ones, to protect, and it has to behave as a normal and responsible member of the international system to do a good job of protecting them. It also has national assets, with a fixed address, that are more vulnerable to reprisal should it misbehave.

The history of Vietnam‑U.S. relations reflects those realities. Bitter enemies in the most costly struggle the United States has waged in an effort to keep an insurgency from gaining power in a contested state, the two countries now have cordial relations. It is in the Hanoi regime’s interest to have such a relationship, to benefit from economic interdependence, and to play power politics in a way that helps protect its interests from any Chinese encroachment. The containment of Chinese power in the South China Sea is the foremost of several shared interests that have formed the basis for a deepening relationship between Washington and Hanoi.⁹

Globalization has increased the costs to incumbent regimes of misbehavior. It is more costly to be a pariah now than it was in a less globalized world. The costs include the opportunity costs of not benefiting from full participation in an integrated global economy as well as any direct penalties, such as specially imposed sanctions. Muammar el‐​Qaddafi’s sharp turnaround in the late 1990s—after heading one of the more troublesome regimes in the world over the previous quarter century—demonstrates the motivations involved. Qaddafi felt the pain, politically as well as economically, of the sanctioning and ostracism of Libya that the international community imposed largely in response to Libyan involvement in terrorism. The later unfortunate undoing by Western powers—when they intervened in an uprising in Libya to overthrow Qaddafi even after he had given up terrorism and his unconventional weapons programs—of that salutary lesson for other troublesome dictators does not negate the logic involved.¹⁰

That same logic provides most of the explanation for the significant decrease in state‐​sponsored international terrorism in the past 25 years. (A further and related explanation is the collapse of the Soviet Union and end of the Cold War, which made it more difficult for those ostracized by one set of great powers to find succor elsewhere.) State‐​sponsored (including state‐​conducted) terrorism is today only a shadow of what it was in the 1970s and 1980s.¹¹ The U.S. list of state sponsors of terrorism has dwindled, while some that remain on it, such as Cuba, do so for reasons other than terrorism. All of those facts should put into perspective one of the most frequently expressed worries about substate violence that might lead to regime change: the fear

that a new “terrorist state” will emerge. Old Cold War habits of thinking about regime changes in relation to a single international scorecard die hard. Even those observers who do not impose a “World War IV” template often apply a modified notion of good guys and bad guys aligning against each other. A model of the Middle East that is often applied sees Iran and its allies lining up against everyone else. That view gives rise to unrealistic concepts, such as a coalition of “moderates” that includes the United States, Israel, and Arab regimes with which the United States happens to have good relations, as if those players did not have at least as many differences among themselves as each of them does with the Iranians.¹²

Much American discourse about the conflict in Syria has used, explicitly or implicitly, such templates in assessing possible outcomes of that war. Outcomes get rated by the effect on a broader competition with the Syrian regime’s major allies, Russia and Iran. That way of looking at the Syrian civil war is misguided. The war can affect some U.S. interests at least indirectly—although, as noted, it is unclear which overall outcome would affect those interests relatively favorably or unfavorably. But the effects will not be measured primarily by a reenergized Russia dreaming old imperial dreams of obtaining warm‐​water ports or of making inroads throughout the Middle East. Neither will they be measured mostly by the strength or weakness of Iran, for which Syria has been a rather troublesome ally—all the more so since the Assad regime’s suspected use of chemical weapons against anti‐​regime forces in 2013 attracted the attention, as well as the opprobrium, of the international community.

Nonstate actors can pose threats even if they never gain control of a state. It is with regard to that possibility that fears of international terrorism especially dominate American ideas about substate conflict. Relevant fears of terrorism, even of the religiously defined variety, have roots that extend back before the end of the Cold War; the bombings and kidnappings by Lebanese Hezbollah during the 1980s elicited significant policy attention during the Reagan administration.

There is no doubt, however, that the terrorist attack of September 2001 is the overriding influence today on American thinking about terrorism and thus on American thinking about substate threats in general. Public and political concern about terrorism usually spikes after a terrorist incident and then gradually subsides as time passes without another attack. We see today reflections of some such subsidence with acrimony over the collection activities of the National Security Agency that would not have caused a comparable stir in the first few years after 9/11. But 9/11 was such an enormous national trauma and the subsequent reactive spike of fear and concern was so high that even after more than a decade of breath catching, it is still the defining event in how Americans perceive threats to themselves from nonstate actors.¹³

September 11 further filled the national need for a known, prominent adversary by equating serious terrorist threats to the United States with a single group going by the name al Qaeda. A name and an organization are much better suited to play the role of the nation's enemy than is a mere concept—much less a tactic, which is what terrorism is.

To understand how Americans' conception of terrorism today is a function at least as much of their own political and psychological needs and circumstances as it is of the attributes and abilities of any foreign group, consider how the salience of terrorism as a public issue has, or has not, correlated with the amount of terrorism taking place in the United States. In the 1970s, there was a lot of it, which was perpetrated by groups ranging from Croatian nationalists to domestic American radicals. So much occurred during the middle part of that decade— airplane hijackings, bombings in the Capitol and Grand Central Station, and much else—that comparable acts today would be seen as a horrifying wave of terrorism.¹⁴ But the Cold War competition with the Soviet Union was still going strong, and coming away from the Vietnam War and the Watergate scandal, Americans were in no mood to expand, rather than contract, the powers of their national security agencies or to declare any "war on terror." The mood over the past decade has been far different.

Given that current American attitudes about terrorism are based more on moods and milieu in the United States than on an objective reality about foreign terrorist groups, American perceptions about terrorism are disconnected from that reality in at least two important respects. One concerns the overall magnitude and intensity of the threat. As John Mueller and Mark G. Stewart have argued in detail, the attention and resources devoted to counterterrorism greatly outstrip the damage that is being countered, even when including the off-the-charts 9/11 attacks.¹⁵ The impact of terrorism is, of course, at least as much political and psychological as it is physical; a hazard is that exaggerated emphasis on countering terrorism may accentuate the political and psychological impact of the terrorism itself.¹⁶

The other major disconnect concerns the habitual references to "al Qaeda" as an enemy that embraces most of what is feared in terrorism, at least of the radical Sunni variety. The references reify an organization that does not exist. Core al Qaeda, the group that perpetrated 9/11, is a shadow of its former self and is almost certainly incapable of mounting another operation comparable to 9/11. Applying indiscriminately the label "al Qaeda," as if it denoted a single cohesive organization, to the sundry violent Sunni groups in Africa and Asia—even to those who have seen advantages in adapting the "al Qaeda" brand name— distorts the nature of the problems that such groups pose to the United States. Most of them have a mixture of motives that are focused primarily on politics and power within their own regions. The United States figures into their thinking mostly to the extent that U.S. actions have made them and the United States direct adversaries. The indiscriminate labeling also loses sight of how, even in al Qaeda's heyday, Osama bin Laden never gained broad support across the community of Sunni jihadists for his strategic concept of hitting "the far enemy" as a way of toppling "the near enemy."¹⁷

The various radical groups that compose what is too loosely referred to as al Qaeda do warrant attention from the United States, but not as part of some global challenge from an archrival terrorist organization. Instead, those groups are factors in the politics and stability of countries from Mali to the Philippines. They are a complication for U.S. foreign policy as it relates to the countries directly affected by violence and to allies of the United States that may be supporting those countries.

Beyond groups that are appropriately labeled as terrorist groups because terrorist attacks are the dominant mode of their operations, the security challenges to U.S. interests posed by nonstate actors mostly involve actions that take terrorist forms. For example, the leftist insurgent organizations in Colombia, even if driven these days by motives that are at least as much economic or criminal in nature as political, are legitimate concerns insofar as their violence affects foreign interests directly or continues to impede the achievement of stability in Colombia. Other organized criminal organizations affect U.S. interests, but as a law enforcement problem rather than as a truly national security concern. The one substate threat in recent years that cannot simply be labeled as terrorism but that has been big enough and bloody enough to be properly considered a U.S. security issue is the violence associated with Mexican drug cartels, some of which has bled across the U.S. border.¹⁸

People often see internal conflicts in foreign countries, especially open civil war, as a security threat to the United States for reasons that go beyond the possibility of a hostile regime’s taking power or any immediate danger that particular nonstate actors may pose. The disorder itself is seen as breeding longer‐​term dangers, in at least two respects. One centers yet again on terrorism, and on the idea of territorial havens used by terrorist groups. The other concerns the potential for civil wars to spill over international boundaries.

The concept of a safe haven as contributing to the threat any terrorist group might pose to U.S. interests has played a prominent role in discussions in the United States about counterterrorism. U.S. policymakers justify the continuation of the long American military involvement in the civil war in Afghanistan chiefly in relation to preventing the establishment of a new terrorist haven there. One form of the worry about terrorist havens is that internal conflict might give rise to a regime that would willingly host a group that performs terrorist operations in other countries. That concern is just a variation of the aforementioned worry about new hostile regimes and state sponsorship of terrorism, and it is subject to the same limitations noted earlier. The Taliban in Afghanistan, which already experienced the disadvantages of its earlier hosting of al Qaeda—including that group’s 9/11 operation provoking a military intervention that inflicted the biggest setback the Taliban ever suffered— would have little incentive, if it regained power in most of Afghanistan, to reestablish anything like that earlier relationship.

The other customary worry about terrorist havens concerns not what any new regime would tolerate, but instead how terrorists would exploit disorder and lawlessness to establish such a haven. The group does what it wants to do, in other words, because the chaos of domestic conflict places it beyond the reach of any regime, domestic or foreign. That scenario presents somewhat more basis for concern because it does not postulate regimes getting into the terrorist business against their own interests, but it suffers from two other limitations. One is that although a little bit of disorder may help keep a terrorist group beyond the reach of law and government, a lot of disorder does not often help it. Terrorist organizations find it hard to operate in truly chaotic situations for the same general reasons legitimate businesses and other organizations find it hard to operate in such situations. That is why al Qaeda did not make as many inroads in Somalia as many predicted during the two decades that country was the archetypical chaotic failed state.

A second limitation is that terrorist safe havens are, quite simply, overrated. They may seize our attention as a spatially satisfying way of keeping score of how we are doing against any adversary that, like the United States, operates internationally. Among all the variables that help determine how much of a threat any one group represents, however, having a small patch of real estate is not one of the more important ones. That is all the more the case in an era of globalization and globe‐​spanning information technology in which planning, recruitment, and the direction of operations take place at least as much in virtual space as they do in physical space.¹⁹ When physical space is involved in a terrorist threat to U.S. interests, it is at least as likely to be in an apartment or mosque in a Western city (or a flight school in the United States) as on a piece of ground in some strife‐​riven land outside the West. Preparation of the most famous terrorist operation of all—9/11—is a prime example.

Spatial thinking is also involved in the other frequently expressed concern about civil wars: that they will “spread” across international boundaries. That imagery is a natural way of thinking; we can understand how something messy, like a gooey sauce, can spread over a physical boundary, especially when it is boiling or otherwise increasing in heat, size, or intensity. The true spreading of a civil war across international boundaries is a limited phenomenon, however, that involves the direct movement of people or munitions into adjacent states. Some of that has occurred, for example, with the Syrian civil war. Artillery shells fired from Syria have struck in Turkey. The movement of fighters across the boundary with Lebanon carries the risk of the Syrian battlefield effectively being extended into Lebanon. Most of the secondary regional effects of civil war, however, are not actually a spread of the war itself—and how could it still be described as a civil war once it crosses boundaries?—but instead a stimulation of other troubles.

Some of those other troubles can be complications for U.S. foreign policy, even if they do not directly threaten U.S. security. Even if a spilling of men and materiel over international boundaries is limited to a few pieces of geography, such as some of Syria’s borders, ideas and emotions can transgress boundaries more widely and more easily. Dramatic disorder in one country can stimulate disorder in other countries because citizens of the latter see new possibilities, or they feel kinship with some of the protagonists in the first country’s conflict, or they have preexisting anger that is catalyzed into an eruption by the news of occurrences elsewhere.

The Arab Spring is the most obvious demonstration of that phenomenon. The uprisings in several Middle Eastern countries beginning in late 2010 have exhibited at least as many differences from each other as similarities. Even in neighboring states such as Tunisia, Libya, and Egypt, the course of events and the political outcomes have so far been quite different. Such outcomes have reflected different preupheaval conditions in each country, particularly with respect to the role of the military.²⁰ Nonetheless, a panregional contagion effect is clearly involved. It would be far too unlikely a coincidence for that many uprisings in the Middle East to happen to break out for completely independent reasons within such a short time frame.

Substate conflict in one country thus can pose a challenge for U.S. foreign policy partly because of its potential to stimulate messy events in other countries. What happened in Tunisia in late 2010 can be seen in retrospect as important not only, or not even mainly, because of what it would lead to in Tunisia—one of the smaller and less significant states in the region—but also because of what it would lead to elsewhere in the region. If one were to view the totality of the unrest coming under the label of the Arab Spring as on balance a net negative for U.S. interests, an implication is that the United States should have tried to do something to prevent the abuse of a Tunisian fruit vendor that started it all.

Clear and practical implications for policy toward the Middle East, however, do not follow from such thoughts. First, the Arab Spring, messy though it may be, is not necessarily a net negative for U.S. interests. Although the unrest has presented many difficult choices for U.S. policymakers, the political changes taking place are in important respects desirable and supportive of U.S. interests. Second, calculations in hindsight are far different from decisions that must be made in real time. Even if the United States knew what it would like to see in the Middle East, it was impossible to foresee how any unrest in the region would play itself out. Third, the Arab Spring reflects a potential for upheaval that could have been triggered by any of numerous possible events. If it wasn’t the abuse of a Tunisian fruit vendor, then it would have been something else—although the timing of the something else would also be unpredictable.

Another sort of panregional contagion effect in the Middle East, woven into some of the confrontations of the Arab Spring, has been heightened sectarian consciousness, particularly a consciousness that pits Sunni against Shia. The underlying differences and disagreements have been around for centuries, but the past 10 years have seen an intensification of that divide, with many people in the region identifying their interests primarily in relation to their confessional group and with the divide assuming a violent form in some countries. The intensified sectarianism is definitely a negative for the United States, if for no other reason than it has forced Washington to make decisions that others perceive, correctly or incorrectly, as taking sides in religiously infused conflicts. How the United States might have acted differently to help avoid that surge in sectarianism is not a matter of what it failed to do but instead of what it did: it launched a war in Iraq in 2003. That war was the single‐​biggest stimulus to the heightened sectarian conflict in the Middle East over the past several years. Toppling the Baathist dictatorship in Iraq uncorked the fears and ambitions of each confessional group in a country in which a Shia majority had been dominated for decades by a largely Sunni regime. The mostly sectarian and still continuing civil war in Iraq that followed did not literally spread across international borders as much as some feared, but it stimulated parallel fears, ambitions, and distrust among confessional groups in other Middle Eastern states. We see the consequences most vividly today in the civil war in Syria, where the sectarian tables have been turned with a largely Sunni revolt seeking to topple an Alawite‐​dominated regime.

The foregoing should make apparent how limited the capacity of the United States is to curb many instances of substate conflict, as well as how easy it is for the United States to be counterproductive or otherwise damaging in its influence. The Iraq War is by far the biggest lesson of recent years in that regard, both because it was the costliest overseas expedition of the United States in the past half century and because it is a glaring example of U.S. action boosting or creating substate conflict. The U.S. invasion led directly to severe nationwide violence that blended the sectarian civil war with an anti‐​occupation insurgency. It also created a serious international terrorist problem in Iraq that had not existed before the invasion, having the directly opposite effect in that regard from what the selling of the war would have led one to believe.²¹

Lessons can also be drawn from the intervention in Libya in 2011, although the scale and direct cost of U.S. involvement were far less than in Iraq and the motivations and rationales for the intervention much different. Another major difference is that an armed insurrection was already well under way in Libya before the foreign intervention. As in Iraq, however, the removal of a long‐​standing dictatorship opened a floodgate for pervasive violence and disorder. There was not the same sectarian division as in Iraq, but there was the legacy of the four‐​decade‐​long regime of Muammar el‐​Qaddafi, which had destroyed much of civil society and had left state institutions that were little more than instruments of personal rule. With the dictatorship gone, Libya became one of the more disorderly and unruly countries on the planet. The Libyan government comes nowhere close to having a monopoly on the legitimate use of force. What passes for internal security in most of the country is provided by a hodgepodge of militias left over from the insurrection. The effect on U.S. interests is negative, including at its most painful a lethal attack on a U.S. diplomatic facility in Benghazi in 2012.²² One other source of lessons from recent U.S. experience is the military intervention in Afghanistan, which has become

America’s longest war. The first few weeks of the intervention succeeded in expelling al Qaeda from its Afghan home and in ousting from power in Kabul its Afghan hosts. But since then, the war has involved a failure to quell the violence, to establish stable and legitimate governmental institutions, to resolve the differences and divisions that underlie the violence, to prevent a resurgence of the Afghan Taliban, or to end deadly spillover into Pakistan.

Different though they are in other respects—and not to be equated on any scorecard of foreign policy decisionmaking—the U.S. military interventions in Iraq, Libya, and Afghanistan have shared the attributes of being successful in removing a targeted regime but conspicuously unsuccessful in resolving the substate violence that followed regime change. That disappointing record in using military force—which could be presumed to be the most decisive foreign policy instrument available—reflects certain inherent limitations to any U.S. attempt to deal with substate conflict.

One limitation concerns how the level of effort required to resolve an internal conflict is disproportionate to the stake the United States has in the outcome of the conflict. Richard Betts has observed that limited, impartial interventions do not end civil wars. A limited intervention may be able to end a war if it tips the balance of forces enough in favor of one contender for it to prevail, but such an intervention would not be impartial. An impartial intervention can end a war only if it is so large that it overwhelms all sides in the conflict, imposes a peace, and thus is in no way limited.²³ It is difficult to find substate conflicts in which the United States has either a strong enough interest in stopping the conflict to make a big enough effort to do so, or has enough interest in one side winning to tilt the battlefield in favor of that side.

Those realities, which have been exhibited by the conflicts in which the United States has intervened in recent years, are also present in the troubling Syrian civil war. The war has tugged at American heartstrings as a humanitarian issue, but American stomachs have not been prepared for large‐​scale intervention in another Middle Eastern war. Any thought of more limited intervention must confront the question of what outcome the United States ought to favor in that war, in which there have been atrocities on both sides and in which a brutal dictatorship is facing off against radicals with terrorist connections.

Any action that even appears to take sides incurs the resentment and anger of someone else. That is a second respect in which U.S. intervention, especially military intervention, can be counterproductive by making enemies of those who would not otherwise have been enemies. The intervention leads contenders who are insular and otherwise focused on the domestic conflict to broaden their focus to the foreign intervener. The substantial anti‐​occupation flavor of the insurgency in Iraq exhibited that point, in addition to being a violent response to occupation per se. A similar situation has prevailed in Afghanistan. The Afghan Taliban are highly insular in their orientation and objectives, and they are concerned about the United States only insofar as the United States interferes with the Taliban’s objectives for the social and political ordering of Afghanistan. The same cannot be said of the transnational terrorists of al Qaeda, who were once the Taliban’s guests. It can, however, be said of most of today’s Islamist terrorist scene, consisting largely of radicals—even among those who get described as “affiliates” of al Qaeda—who are concerned primarily with events in their respective local areas of operation, be they in Africa, South Asia, or elsewhere.

Referring back to the customary way in which Americans have come to view substate threats, we see unfortunate elements of self‐​fulfilling prophecy. American perceptions of those threats tend to overstate the danger they pose to U.S. interests. The overstatement leads to greater efforts overseas to neutralize or defeat the threats than would have been made with a more accurate appraisal of the danger. The efforts, especially ones involving the use of military force, elicit responses that turn some presumed threats into real ones. That cycle, rooted in natural American ways of looking at the place of the United States in the world and in natural foreign responses to the exertion of American power, will be difficult to break. Any chance for breaking the cycle will begin with more accurate assessments of substate threats and of our ability to mitigate them.