U.S. Strategy For Countering Violent Extremism: An Assessment


By Will McCants and Clint Watts

The United States and its allies devote considerable financial and human resources to countering violent extremism (CVE). CVE is a central pillar of the United States’ domestic and international counterterrorism effort, following the lead of the United Kingdom’s Prevent initiative begun several years earlier. Like the United Kingdom, the United States launched its CVE enterprise in response to a perceived increase in radicalization among its Muslim citizens. The U.S. enterprise, however, lacks a clear definition, is based on flawed assumptions about what works, and its proponents have yet to question whether CVE is worth doing in the first place. The United Kingdom’s approach suffered from similar shortcomings when it was first introduced, many of which were corrected in a later program update. It is time for the United States to do the same.


Although U.S. government documents frequently employ the term CVE, there is not a shared view of what CVE is or how it should be done. Definitions range from stopping people from embracing extreme beliefs that might lead to terrorism to reducing active support for terrorist groups. The lack of a clear definition for CVE not only leads to conflicting and counterproductive programs but also makes it hard to evaluate the CVE agenda as a whole and determine whether it is worthwhile to continue.

In the interest of clarifying the activities covered by CVE and encouraging debate on their relative merits, we propose the following definition: reducing the number of terrorist group supporters through non-coercive means. This definition is useful for several reasons:

  1. It is broad enough to cover most of what people describe as CVE without being biased towards one type of activity. For example, we favor programs that persuade active but law-abiding terrorist supporters to abandon their support. Others favor an approach that prevents people from becoming supporters in the first place.[1] Our definition captures both of those approaches and makes it possible to have a debate about their efficacy because there is a clearly stated goal.
  2. The definition is broad enough to include support for any kind of terrorist organization but narrow enough to exclude support for other kinds of violent organizations like insurgent groups and gangs, which don’t primarily attack civilians for larger political objectives (the definition of terrorism).
  3. The definition suggests a metric: reducing the number of a terrorist group’s supporters. The focus is not on reducing support for ideas, which is difficult to judge, but rather support for specific organizations that embody those ideas and seek their realization, which is easier to document and more closely related to criminal behavior.
  4. The definition excludes coercive kinds of “countering” activity that are better left to law enforcement and militaries (e.g. arresting, killing).
  5. The definition tacitly acknowledges that there is a spectrum of organized support, ranging from lone wolves to bona fide members of a terrorist group. Moreover, support does not necessarily imply criminal behavior. The “supporter” label applies to anyone who actively promotes a terrorist group’s program and actions.

As the last point suggests, there are a range of supporters. We categorize them into two broad types: law-abiding supporters and law-breaking supporters. Obviously, what constitutes “law abiding” and “law breaking” depends on the laws of the country in which the supporters live. The law-breaking supporters can be further divided into mobilized (those who have not gotten caught) and demobilized (those who are in custody).

One final distinction is worth keeping in mind before we proceed: there is a difference between sympathy for and support for terrorist organizations. Sympathizers agree with the words and deeds of a terrorist organization but do not actively promote those words and deeds. They are the people that show up on Pew and Gallup polls expressing sympathy for al-Qaeda. Sympathizers can certainly become supporters but they have not done so yet.


People sympathize with a terrorist group because they believe the cause is justified, there are no other means to achieve it, and the terrorist group does more good than harm. Change one of these perceptions and sympathy decreases.

There are several ways to change these perceptions, ranging from easy to difficult: let the terrorist group undermine itself; actively point out inconsistencies between a group’s message and actions; make sure no one hears the terrorist group’s messages (e.g. by shutting down radical Internet websites); provide positive information about the terrorist group’s targets that contradicts the terrorist group’s messages about the targets (e.g. spreading good news about America); and, most difficult, solve the underlying problems that push people to sympathize with terrorist organizations. Polls, focus groups, and sentiment tracking are ways to assess how effective your efforts are.

CVE programs that seek to reduce sympathy for terrorist organizations are premised on the assumption that reducing sympathy leads to a reduced support. This is not necessarily true, at least in the short term. For example, sympathy for al-Qaeda has fallen steadily over the past decade but it still attracts a steady stream of supporters.

The task of reducing the number of terrorist group supporters is often left to law enforcement and the military because people view non-incarcerated supporters of all stripes as irredeemable. It is true that supporters are harder to reach but there are ways to turn them around, particularly if they have not engaged in criminal activity. Exposing them to criticisms of ex-terrorists and non-violent extremists can be effective, as is derogatory information about terrorist group members. Interventions by law enforcement can also work, especially if done with one’s peers, family, and respected figures in the community. Most difficult, the government or NGOs can provide law-abiding supporters with positive, alternative outlets for their activism.  If supporters have engaged in criminal activity, amnesty and exit programs can work. If they have been arrested, it helps to separate them from more committed radicals and facilitate their transition back to society.

Unless supporters are incarcerated, tracking success in reducing their numbers is more difficult than tracking success in reducing sympathy. The number of people who support terrorist groups is much smaller than the number of sympathizers and they are indistinguishable in polls. It is not impossible, however, to come up with a rough estimate, particularly when many terrorist group supporters are active online. Combining this kind of publicly-available information with various kinds of intelligence can give a sense of the overall number.

Although there are various ways to reduce sympathy and support, not all of them are worth doing. In most cases we favor using methods that are most amenable to assessment in the near term and have the fewest downsides.


The U.S. CVE initiative aims to reduce the number of sympathizers and demobilized supporters. Law-abiding supporters and mobilized supporters are left to law enforcement and, overseas, the military. In addition, the United States seeks to reduce the number of people vulnerable to becoming sympathizers. The government’s primary approaches are “challenging justifications for violence,” “promoting the unifying and inclusive vision of…American ideals,” and providing support to vulnerable individuals in communities targeted for terrorist recruitment. For each type of activity, the United States encourages local communities targeted by terrorist organizations to play a leading role in carrying out these approaches.[2]

There are at least two major problems with this strategy. Anyone can potentially sympathize with a terrorist organization if the conditions are right. By framing the problem this way, you risk alienating a broad swathe of people, as the United Kingdom found several years ago in its treatment of British Muslims. Similarly, the current administration exclusively focuses its domestic CVE efforts on Muslims, endeavoring to inoculate them against al-Qaeda propaganda—or “building resilience” as it is called. This approach assumes that American Muslims are actually susceptible to this propaganda in large numbers when in fact that has not proven the case.

Another downside to focusing on “potentials” or “vulnerables” is that you cannot prove how many “potentials” or “vulnerables” did not become terrorists as a result of your program. They might have been just fine without it. Teaching women English, a common CVE program overseas, is a worthy thing to do for a host of reasons but no one can demonstrate how it ends support for terrorism.

The second major problem with the U.S. CVE agenda is that the U.S. government has no interest in turning around law-abiding supporters of terrorist organizations. This is particularly a problem domestically because U.S. law enforcement has no incentives to stop young kids from ruining their lives and every incentive to build cases against them or to recruit them to build cases against their friends. Contrast this to the approach to gang violence in Boston where the police work with families and community leaders to confront individual gang members in an attempt to turn them and their peers around.

What if the United States treated the problem of AQ support domestically like it treats supporters of white hate groups? What would its CVE program look like? It would aim to decrease the public’s sympathy for the group and seek to turn around the group’s law-abiding and incarcerated supporters rather than reaching out to the broader white communities of which they are a part. This approach is commensurate with the threat; its success can be measured; it carries less risk of alienating communities from which terrorists arise; it undermines the narrative that these communities are potential threats; and it is far less threatening to civil liberties than the current approach.

All that said, we are not sanguine about the possibility of turning around law-abiding or incarcerated supporters of terrorist groups in the United States. It is incredibly difficult, particularly given the intractable nature of larger political problems that drive some forms of terrorism. Moreover, recommending that the U.S. government reorient its domestic CVE policy toward dissuading law-abiding and incarcerated terrorist group supporters does not mean that the government is the best suited to do it. Neither does it imply a certain way to go about it nor does it mean that this approach will work in every country. Every context will require a unique blend of policy prescriptions.


Given its focus on vulnerable populations, it is not surprising that the U.S. CVE strategy combines three population-centric approaches to fighting violent extremism: U.S. counterinsurgency doctrine, the United Kingdom’s Prevent model, and U.S. law enforcement’s Community Oriented Policing model.  None of these three popular strategies appropriately assesses and counters the U.S. homegrown extremist threat because each model rests on assumptions that prove faulty in the U.S. context of CVE.

The “whole-of-government” approach advocated in the White House CVE implementation plan repeats counterinsurgency mantras popular during the 2009-2010 timeframe, advocating a “surge” to “win hearts and minds” of vulnerable populations—namely the U.S. Muslim population. The United States is not fighting a subversive Muslim insurgency in the homeland, and the call to “win hearts and minds” is unnecessary as American Muslim hearts and minds, while likely irritated by the incessant scrutiny of their communities, have not been lost. Equally frustrating is that the model contains the faulty assumptions of past failed counterinsurgency strategies, namely 1) that more actions done poorly across many bureaucracies are better than fewer actions done well by a handful of experienced agencies; and 2) coordination of many actions across many agencies to counter a complex threat will occur easily and sufficiently to keep pace with an evolving threat.

The large-scale approach assigns many tasks to many agencies to counter a recruiting problem that produces only dozens of wannabe terrorists annually. In doing so, the U.S. CVE enterprise dilutes the collective capacity of government as a whole.  Each agency assigned an interagency CVE task cannot focus on performing their primary task to the best of their ability and instead must engage in many tasks of limited consequence and low impact. For example:

  • The strategy designates U.S. Attorneys as the lead for CVE in their jurisdictions, which distracts them from their primary duty to prosecute criminals and terrorists in U.S. court.  The approach turns U.S. Attorneys into diplomats, glad-handing community leaders when they can more effectively serve their communities by using their precious time and limited resources to tackle the overwhelming case load of corrupt bankers, Internet predators, and drug traffickers – all of which have a more detrimental impact on American society than an occasional violent extremist.
  • The strategy tasks agencies like the Departments of Education and Health and Human Services, which have limited resources and incredible challenges to tackle far outside the sphere of counterterrorism. The Department of Education should focus on improving inner-city education and restoring American student competitiveness in math and science rather than trying to find a dozen troubled youth googling “al Qaeda.”  Health and Human Services can do more to counter radicalization by addressing the rising rates of teenage depression and childhood obesity – two likely yet unexplored factors that contribute indirectly to social isolation and lone actor violence. But, this should not be framed as counterterrorism.

The second model inappropriately inserted into the White House CVE strategy is the U.K. Prevent Model.  The United Kingdom recognized after the 2005 London subway bombings that it needed to prevent local Muslims from joining al-Qaeda’s cause.  The United Kingdom, a nation with large Muslim Diaspora populations not fully integrated in British society, recognized a critical need to directly engage with Muslim communities to discern their grievances in hopes of ameliorating them.

The U.S. duplication of the U.K. Prevent strategy is ill-advised for three reasons.  First, Muslim community composition and integration in the United States varies wildly from what one finds in the United Kingdom. American Muslims are not in the same position as British Muslims.  Second, critics in the U.K. criticized their governments Prevent strategy for “securitizing” the Muslim population—making normal citizens feel like a security threat.  Finally, the initial Prevent strategy was too expansive in scope and funded programs that could not be proven effective. In recent years, the United Kingdom has scaled back its CVE efforts and published lessons learned from its initial Prevent strategy, outlining several objectives for refining future British CVE efforts. The revised strategy calls for:

  • a clear definition of the terms “extremism” and “extremist;”
  • an assessment of where CVE efforts are most effective and efficient in preempting terrorist recruitment;
  • integrating CVE efforts in the virtual and physical environments where al-Qaeda recruitment is taking place; and
  • designing CVE projects to address terrorist recruitment in specific populations with an emphasis on pre-determined measures of project performance and effectiveness.[3]

The U.S. CVE strategy is several years behind the United Kingdom because it has adopted the early Prevent strategies while ignoring the program’s lessons learned and failing to adjust for the marked difference between the U.S. and U.K. Muslim populations.

The third concept found in the White House CVE strategy is the Community Oriented Policing (COP) model instituted across the U.S. during law enforcement’s era of tackling gangs and narcotics distribution networks.  At a time of economic austerity and police budget cuts, local law enforcement has transformed these programs into Community Oriented Policing for Muslims. This modified COP approach also has several drawbacks when morphed for CVE purposes.  Violent Muslim extremists in the United States are usually not a part of the broader Muslim community—they more often interact virtually than physically with terrorist propaganda and recruiters, they rarely operate in groups (at least in recent years), and they often evade the notice of their parents and neighbors. COP programs routinely rely on families (parents) to police their children. Whether in Columbine or in 9 Somali youth in Minneapolis, parents are notoriously poor at detecting emerging violence from their kids, especially violence emanating from young, socially-isolated perpetrators.

The COP approach may inadvertently create a “D.A.R.E. effect” in the communities it encounters.  Some research suggests that during the 1980s and 1990s law enforcement education programs about the evils of drugs inadvertently piqued the curiosity of students about illegal narcotics. Most young people in the United States have little or no exposure to al-Qaeda’s propaganda. Aggressively informing U.S. communities about al-Qaeda propaganda might result in a similar outcome as the War on Drug’s D.A.R.E. program (obviously, on a much smaller scale) as a result of greater exposure to al-Qaeda’s message and the repeated warnings to stay away from it, which only encourages curiosity.

Finally, the COP approach sends the wrong signal to Muslim communities. The White House is sensitive to the securitization problem and says so in its strategy. But it proceeds nonetheless to focus almost exclusively on Muslims.  The subtext remains: You Muslims are a potential threat and we, the government, have to co-opt you. As seen in the first iteration of the U.K. Prevent model, this approach isolates already marginal communities even further. The approach risks perpetuating the “us-versus-them” dichotomy the White House is trying to overcome.


As the Obama Administration prepares for a second term, it needs to reconsider its domestic CVE enterprise. The primary question to answer is whether a CVE program is necessary. If the recruitment problem is small enough for law enforcement to manage through traditional means, it should be left to do so. If not, then the U.S. might consider the following changes to its CVE program:

  1. Clearly define the meaning of CVE and delineate a spectrum of support for terrorism, explaining where on the spectrum the USG can usefully focus its efforts.
  2. Stop focusing on vulnerable populations.  Continue to focus on reducing sympathy for terrorist organizations and deradicalizing incarcerated supporters.
  3. Consider expanding CVE to include non-coercive programs that address law-abiding supporters of terrorist groups. Even if a law-abiding supporter breaks anti-terrorism laws by doing something relatively harmless, like translating a document, consider using it as leverage to get him or her to turn their lives around.  The aims of this strategy are measurable and self-reinforcing as each law-abiding supporter turned will provide increased influence on his or her associates.
  4. Seek to reduce sympathy and support by means that are measurable, low risk, small scale, and narrowly targeted on a specific, well-defined audience.
  5. Stop reaching out to or empowering communities targeted by terrorist recruiters, which is usually just code for talking to Muslim community leaders. Instead, offer to inform all citizens in a locale about the range of terrorist organizations that are recruiting among them.
  6. Enlist the support of non-religious NGOs that work to change the minds of youth who admire violent groups. There is ample opportunity to do this online. A lot of sympathizers and law-abiding supporters in the United States openly express their views on social media. Start with them.
  7. Use undercover officers and informants to find the most serious suspected militants. Do not require agencies that conduct outreach to also conduct intelligence gathering.  Pursuing both strategies from the same agency undermines the effectiveness of both programs.
  8. Train law enforcement officers to detect domestic terrorist group supporters of all ideological stripes.  Do not try to make officers into religious experts or narrow their sensory field by focusing them on only one of dozens of strains of terrorism they might encounter in their jurisdictions. We know the USG accepts a broad approach in principle, but we have seen little of it in practice.
  9. Place CVE responsibilities squarely in the hands of one lead agency.  Spreading CVE responsibilities to a dozen entities results in many agencies lightly pursuing CVE efforts that are a low priority when compared to their primary missions.  Measurable results in CVE will only be achieved if agencies are measured by their CVE results – “extremists deterred,” not “tasks completed.”
  10. Use the federal government to support the CVE and intelligence priorities of local jurisdictions. Pursuing a top-down approach for countering extreme ideologies that most jurisdictions will never encounter is a waste of resources. The FBI’s Behavioral Analysis Unit support for dealing with serial killers and the FBI Crime Lab are two such examples of how consolidated federal support can assist local jurisdictions.
  11. Just as foreign models of CVE should not be imported whole cloth into the United States, the United States’ CVE enterprise should not be exported overseas. Other countries will require very different approaches from those adopted in the United States. Get in the habit of speaking of “good practices” rather than “best practices.” Most CVE “best practices” are not replicable.

Again, we ask the U.S. government to take a hard look at the number of people who support terrorist groups. If their numbers are small and if their violent acts are few, a traditional law enforcement approach might be all that is required. We know that CVE is viewed by some as a more holistic approach to terrorism. But the downside of holistic approaches is that they can do a whole lot of harm and consume a disproportionate amount of resources. If CVE is required, identify sympathizers and supporters, select a limited set of actions, execute them through a few agencies and measure their effectiveness against defined objectives. Such measures might prevent a few people from pursuing a bad course in life, and if the measures fail, law enforcement knows what to do.


[1] The United Nations defines counter-radicalization broadly as “deter(ing) disaffected (and possibly already radicalized) individuals from crossing the line and becoming terrorists.” See United Nations, First Report of the Working Group on Radicalisation and Extremism that Lead to Terrorism: Inventory of State Programmes, available at: http://www.un.org/terrorism/pdfs/radicalization.pdf

[2] See the National Strategy for Empowering Local Partners to Prevent Violent Extremism in the United States, August 2011 (http://www.whitehouse.gov/sites/default/files/empowering_local_partners.pdf), and the Strategic Implementation Plan for Empowering Local Partners to Prevent Violent Extremism in the United States, December 2011 (http://www.whitehouse.gov/sites/default/files/sip-final.pdf).

[3] See the United Kingdom, Prevent Strategy – Presented to Parliament by the Secretary of State for the Home Department, June 2011. Available at: http://www.homeoffice.gov.uk/publications/counter-terrorism/prevent/prevent-strategy/prevent-strategy-review?view=Binary

Published by the Foreign Policy Research Institute

Founded in 1955, FPRI (http://www.fpri.org/) is a 501(c)(3) non-profit organization devoted to bringing the insights of scholarship to bear on the development of policies that advance U.S. national interests and seeks to add perspective to events by fitting them into the larger historical and cultural context of international politics.

Leave a Reply

Your email address will not be published. Required fields are marked *