America and our allies today face an array of enemies and threats daunting in their number and, taken together, their scale. For the moment, all of our enemies prefer to fight us through unconventional or irregular warfare rather than through traditional force-on-force engagements. Many of the threats we and our allies face are also focusing on their unconventional capabilities--either in the form of terrorism and other types of irregular warfare or through the development of nuclear weapons capabilities or both. Only China appears now to be actively preparing to face us in a conventional conflict sometime in the future.

But if our enemies and challengers prefer to fight us or deter us through unconventional means, it does not follow that we can or should prefer to use such means against them. On the high end of the unconventional spectrum, our nuclear force is intended entirely to deter the use of weapons of mass destruction. It is almost impossible to imagine a scenario in which our president might use nuclear weapons other than in response to a WMD attack or to preempt an imminent WMD attack that could in no other way be prevented. On the low end of the unconventional spectrum, the United States cannot use terrorism. Since our aim is to support rather than undermine global security, we cannot use insurgency. Because we are bound by international law and morality, we cannot encourage religious, sectarian, ethnic, or tribal violence within or between states--as our opponents do and as many empires of the past did to direct resentment away from themselves. What remains? We can use the traditional tools of statecraft, which include diplomacy both public and private, military operations conducted in strict accord with laws of war crafted for conventional warfare, economic aid and sanctions, international criminal law, and so on. In other words, we are constrained not by our systems and procedures, but by our nature as a state that aspires to be upright, ethical, law-abiding, and contributing to global stability, to fight our unconventional enemies with largely conventional means.

We have been using all of these and other tools against our current enemies and vital threats since 9/11. Much has been made of the current administration's efforts to engage our adversaries and challengers diplomatically with the argument that the previous administration disdained diplomacy. Certainly the Bush administration did not engage extensively in negotiations with Iran, for which it has been castigated. But it did engage vigorously in diplomatic efforts within Iraq through multiple channels engaging an extraordinary variety of Iraq actors--and to very good effect. This political engagement, led by Ambassador Ryan Crocker but purposefully supported by Generals Petraeus and Odierno, has been as essential to the transformation in the Iraqi political scene as the surge of forces in 2007. By all accounts the current administration is dramatically less engaged politically with Iraq, a circumstance resulting not only from the absence of an ambassador in Baghdad for the past several months but also from the failure of this administration to maintain the high-level and direct engagements with key Iraqi leaders that its predecessor had. Diplomacy should not be a zero-sum game, where increasing activity in Tehran results in distraction in Baghdad.

We have also been using economic levers of all varieties in many combinations, ranging from traditional foreign assistance in large amounts to countries like Egypt and Pakistan to sanctions of various forms against Iran, North Korea, and, until recently, Cuba. We have made skillful use of international law as well, seizing the assets of designated terrorists and targeting narco-traffickers and the terrorist beneficiaries of their profits. The Obama administration has continued in this tradition by bringing the surviving Somali pirate to New York for trial--precisely as provided for under international law. The use of these traditional instruments of statecraft is not controversial--the only arguments we have are about how to apply specific instruments in specific cases rather than whether or not to use them.

The more fundamental debate centers on the use of the military instrument. In particular, although most Americans agree that the U.S. must be prepared to use military force against its enemies, there is considerable disagreement about what kind of force to use and how to use it. It would be charitable to say that eight years of war have made us weary and therefore eager to follow the lures of those who claim to have found a silver-bullet solution to our problems either through technology or through our own use of irregular warfare. It would not be true, however. Advocates both of high-tech and low-tech alternatives to conventional military power dominated in the 1990s and many changed their dogmas only marginally after 9/11. It is distressing to hear today some of the same arguments we heard during the "strategic pause" of the Clinton years about the desirability of relying on technology to reduce the economic burden of defending America--despite the fact that it was an attempt to rely on precisely such theories in 2001 and 2003 that led us into near-disaster in Iraq and Afghanistan. It is natural to want to find an alternative to the unpleasant requirement to use large numbers of ground forces in far-off lands, whether that be smart bombs, Special Forces, local troops, or sweet reason. In particular, the desire to pull out of Iraq and Afghanistan (and, to be sure, the resistance by some to invading those countries in the first place) has led to the search for some kind of "counter-terrorism" strategy that would allow us to defeat our most dangerous foes without using conventional military force. Alas, there is no reason to believe that such a strategy could work, and much reason to believe that it will not.

Let's start by defining the enemy more closely. An enemy is a group, state, or individual that is working actively to attack America, its citizens, or their property. Enemies are distinct from threats--groups, states, or individuals that may become enemies at some point in the future but are not now attacking us. China is a threat; al Qaeda is an enemy. Iran, interestingly, is both--it is an enemy in the sense that Iranian agents are working actively in Iraq and Afghanistan to help Iraqis and Afghans kill Americans and defeat our aims. But these efforts are less significant strategically to us than Iran's attempts to develop nuclear weapons and other activities around the region that are not being used to attack us or our allies now, but may be used for very significant attacks in the future.

Leaving Iran aside, the list of our enemies is lengthy. Al Qaeda and its affiliates in Iraq, Algeria, Somalia, Yemen, and elsewhere aim at our complete destruction and work toward that goal every day. The Taliban group based in Quetta, Pakistan--the Mullah Omar or Quetta Shura Taliban--is one of our principal enemies in Afghanistan; the Haqqani network based in Miramshah, Pakistan is the other. Smaller and less significant enemy groups in Afghanistan and Pakistan are also fighting us, including the Hezb-e Islami Gulbuddin (HIG), the Tora Bora Front, and numerous drug lords. A number of Pakistan-based groups are also fighting us in Afghanistan, although their main target and focus is Pakistan itself.

They include the Tehrik-e Taliban-e Pakistan (TTP), led by Beitullah Mehsud and based in South Waziristan; the Tehrik-e Nafaz-e Shariat-e Mohammadi (TNSM), led by Sufi Mohammad and based in the area from Bajaur through Swat; and the Lashkar-e Toiba (LeT), which is based in Muridke, a small town near Lahore in Punjab, and operates throughout Pakistan as well as in Kashmir and India. In addition to these groups fighting us in Afghanistan, our Pakistani allies in their own country, and our strategic partner, India, we still face dangerous enemies in Iraq as well: former Baathists within Iraq and in Syria and Jordan; the Naqshbandi network; Ansar al Sunnah; various fragments of the Sadrist movement; and the Iranian Qods Force and its networks. Apart from the threat it poses to our Israeli allies, Lebanese Hezbollah has also been active in Iraq killing Americans.

One could lump all of these enemies together and speak of a global insurgency within the Muslim world, and there is considerable validity to that viewpoint. Taken together, these groups, both Sunni and Shi'i, form a small minority of the worlds 1.5 billion Muslims who wish to impose upon their co-religionists first (and then all of the world's non-Muslims) a peculiar and heretical interpretation of Islam that holds little popular appeal. For that reason, these groups have had to use force even against the Muslim populations where they hold sway to compel those people to adhere to an extremist ideology mostly alien to them, but cloaked in religious language. Thus Taliban rule in Afghanistan in the 1990s was brutal and inhumane. When al Qaeda in Iraq ran Anbar it committed unspeakable atrocities to keep the Anbaris in line--and ultimately facilitated its own destruction when the Anbaris "awoke" with the aid of the surge. TNSM installed itself in Swat through violence and has maintained itself there through violence, just as it had earlier established itself by force in Bajaur and Dir. The good news is that our problem is not convincing Muslims to reject this hideous ideology--the overwhelming majority of Muslims already do reject it. The bad news is that the enemy groups know how to take and hold power through force if they are not opposed, and the opposition of the local people is rarely enough to throw off these new tyrants.

For decades we have hoped that we could do enough damage to such groups by targeted strikes against their leaders to render them harmless, if not to defeat them outright. Ronald Reagan used airstrikes to respond to the Marine Barracks Bombing in Beirut and to Libyan terrorism. Bill Clinton used cruise missile strikes to respond to al Qaeda attacks against US embassies in Africa. The attacks against Libya--a state sponsor of terrorism--were relatively effective at deterring further Libyan terrorist attacks. The airstrikes against Hezbollah and al Qaeda were ineffective and neither deterred nor prevented either group from operating against us. After the withdrawal of US ground forces from Somalia in 1993, the US (and the international community) has done little to prevent Somalia from slipping further into chaos and serving as a base and breeding ground for extremists with ideologies similar to al Qaeda. After 9/11, the US responded by building up Joint Task Force-Horn of Africa (JTF-HOA), which relied on Special Forces and indigenous troops to achieve its aims. JTF-HOA was not able to prevent the extremists from overrunning Somalia and, after Ethiopia invaded and occupied Somalia briefly, has not been able to prevent them from returning. In Iraq between 2003 and 2007 we relied extensively on targeted raids against enemy leaders, supported by 150,000 troops. We killed the enemy leaders at a terrific pace, and even succeeded in killing Abu Musaab al Zarqawi, the head of AQI, in June 2006. But the enemy replaced its lost cadres faster than we could kill them. To the argument that it was the very presence of US forces that facilitated that replacement one could point out that the surge of forces in 2007 and the change of strategy that accompanied it did what the targeted counter-terrorism approach could not do: it drove the terrorists out of their sanctuaries and rallied the support of the Iraqi people against them.

One could also point to our experiences in Pakistan, where the US has attempted to use a combination of targeted strikes and indigenous forces to combat al Qaeda and many of the other enemy groups named above--some of them direct threats to the Pakistani state. Where is the success to show for this strategy? One can hardly complain that it has not been sufficiently resourced—the US has given billions in aid to Pakistan since 9/11. Nor have we maintained a "large footprint"--on the contrary, there have been no conventional American forces in Pakistan. We are left only with the argument that we have not been applying the strategy correctly. But how credible is that argument? Four presidents have attempted to apply this strategy in various areas over the course of decades, and it has never succeeded against a terrorist group. Is it really likely that, although we have been trying this approach for more than a quarter of a century, no Democratic president, no Republican president, no military commanders, no Directors of Central Intelligence, have ever figured it out? And, if it really is that hard to figure out, why should we believe that we can do it now? The burden of proof is on those who claim that we should try again a strategy that has been tried repeatedly and failed to show some reason why it should succeed now.

So what is to be done? First, recognize that our challenge is not to find some magical way to beat the terrorists at their own game either in terms of propaganda or irregular warfare, but rather to find ways to use the tremendous capabilities at our disposal to maximum advantage in difficult circumstances. Our experience in Iraq suggests that this can be done, but different problems require different solutions. Our principal challenge in Afghanistan now is counter-insurgency and state-building. The cultural background, the economic situation, and the political climate all require significant modifications to the approaches that worked in Iraq. But the basic principles of counter-insurgency and state building apply, and we should not dismiss our experiences in applying those principles to Iraq simply because we rightly recognize that Afghanistan is different.

The key problem in Iraq in 2006 was the rising spiral of sectarian violence that threatened to engulf the country in full-scale sectarian civil war. Providing the population with security from that violence was the essential precondition for any sort of progress on any other front. Afghanistan faces different problems. There is virtually no ethno-sectarian violence in Afghanistan--almost all of the fighting occurs within Pashtun areas against international and Afghan forces. There is very little insurgent violence against civilians at all (although there is a great deal of criminality). We do need to provide security to the Afghan people, working together with Afghan Security Forces, but that is not enough. The key problem in Afghanistan today is that the government is widely seen to be illegitimate because of corruption, criminality, and its inability to provide justice, security, and rule of law to its people. US and international forces have not hitherto focused on the need to address the yawning legitimacy gap in Afghanistan. That must change. It must become the core objective of our strategy to which all others, including the establishment of security, are subordinated. This approach is as classic a counter-insurgency strategy as the one implemented in Iraq--but suitably modified for different conditions. As with all counter-insurgency approaches, it will be difficult, costly, and time-consuming, but it can work if done right, as history shows. It has the advantage, among other things, of allowing us to use the conventional tools of statecraft at which we excel to best advantage against unconventional enemies.

So what of Pakistan? A counter-insurgency strategy is required there as well, but the balance of forces must be entirely different since we have neither the desire nor the means to send large numbers of American troops to fight there. Here the prospects are less promising. One problem, perhaps the greatest problem, is the unwillingness of the Pakistani government to engage in a serious counter-insurgency campaign. Its fitful efforts against TNSM first in Bajaur and now in Swat are exceptions that prove the rule--they are totally inadequate to the task at hand, but cost the state's leadership so much pain that they seem daunting in themselves. American efforts to coerce or incentivize successive Pakistani governments to engage in the necessary counter-insurgency campaign have failed repeatedly to change the equation.

I will not attempt to evolve here the strategy for "fixing" Pakistan because I do not believe that there is one. We should abandon the chimerical chase for a grand unified field theory of Pakistan strategy and focus on the problems we can usefully address. Succeeding in Afghanistan--by which I mean establishing a stable, secure, and legitimate Afghan state--will provide us with enormous leverage on Pakistan. Instead of continually begging Islamabad to help us defeat enemy groups that elements of the Pakistani military are actively supporting, we will be able to stand with a functioning Afghan state without Pakistan's "help." Success in Afghanistan will also provide us with the best possible vantage point for seeing the sanctuaries of our most dangerous foes in Pakistan and influencing the people among whom they live. In most areas along the Afghan-Pakistan border tribes cross the Durand Line. The tribal leaders living in Pakistan are not oblivious to what their fellows across the Durand Line are doing. On the contrary. As we and the Afghan government have made Khost--long the heartland of the Haqqani network--a success, both the Haqqani network and its Pakistani sponsors have made clear their determination to reverse our success, which poses the danger of undermining their credibility and authority over a large area.

But, above all, we must recognize that the conventional wisdom about Pakistani involvement in Afghanistan has it backward. The problem is not that Pakistanis fear that the US will abandon Afghanistan and they must therefore hedge their bets by supporting the Taliban and Haqqani. The problem is that they fear we will succeed. Pakistanis often speak of the need for "strategic depth" in discussing Afghanistan--and their support for the Taliban--leaving many Americans puzzled. Surely they do not mean to retreat into Afghanistan if the Indians invade? No, indeed. They fear that the Indians will establish strong ties with an Afghan government that is at the moment strongly inclined toward New Delhi. They watch Indian companies building roads and infrastructure and Indian investment flowing into Afghanistan, and they fear that they will be surrounded. And the elements within Pakistan that support the Taliban and Haqqani see the US as the Trojan Horse that is inserting Indian influence into Afghanistan. Convincing the Pakistanis that we will stay in Afghanistan is not enough. We must convince them that the proxies they are now supporting will fail. The only way Pakistan can have influence in Afghanistan in the future is by working with the government in Kabul; if Islamabad persists in supporting insurgents, it will end up achieving all of its worst nightmares. Succeeding in Afghanistan is not tantamount to succeeding in Pakistan, but it is an essential precondition.

The bottom line is that there is no one-size-fits-all solution to the problems we face, even in the realm of counter-terrorism. Abstract discussions of the problem are fine within limits, but real answers can be found only in the context of real and specific problems. For that reason, among many others, maintaining strategic flexibility is absolutely essential. That flexibility requires not just flexibility of thought, but also the strong and broad mix of capabilities that our position as the preeminent state in the world brings. In particular, it requires large and capable armed forces that can face foes across the spectrum of conflict, as well as the ability to integrate those forces into a sound strategy using all other elements of statecraft to succeed. We do not need to become irregular warriors to defeat irregular warriors--and we could not do so in any case. We do need to continue creatively to apply our strengths against our enemies' weaknesses and to succeed by being ourselves, only better.

Frederick W. Kagan is a resident scholar at AEI.