A Counterfactual Look at the Afghan War:  the “SOF-only” COA and its Implications for the Future

A Counterfactual Look at the Afghan War:  the “SOF-only” COA and its Implications for the Future

by Michael Perry

Introduction

The U.S.-led war in Afghanistan began as a war to combat transnational terrorism but quickly evolved into something deeper and more profound.  To combat terror emanating from a foreign country the U.S. sought a cooperative Afghan government, and thus the war became an exercise in first toppling an uncooperative regime in the Taliban, and second establishing an effective government with a monopoly on force.  The first step proved easy, while the second led to a revival of counterinsurgent theory and doctrine in the U.S. military, as the deposed Taliban fought to undermine the newly established government.  With President Biden’s announcement all U.S. troops will be withdrawn after 20 years of engagement, it’s natural to take stock of what’s been achieved.  Most now recognize the error in the strategy of deploying large numbers of U.S. and Coalition troops to augment the Afghan defense forces.  Economically, through 2017 the combined efforts of the Afghan War had cost $877 billion, a price tag few would argue is justified by the realized returns.[i]  Many have even argued the large deployments of U.S. troops to Afghanistan have been counterproductive.  Micro-level studies of popular sentiment in Afghanistan have shown the Taliban is more popular in many regions than the Coalition,[ii] and macro-level studies have shown an approximate threefold increase in global terrorist manpower.[iii]  Throughout the war an alternative course of action (COA) that was often floated, but never materialized, was a light-footprint approach made primarily of special operations forces (SOF), who would continue to train the Afghan National Security Force (ANSF) that was created after the 2001 disposal of the Taliban regime, coordinate air support, and provide direct assistance on the ground when necessary.  It’s unclear whether such a COA is presently feasible given the terms of U.S.-Taliban negotiations and uncertainty surrounding what level of support the global community will provide Afghanistan moving forward, but this essay asks not whether this SOF-only COA is feasible now, but whether it was feasible from the outset.  In particular, I'll describe the initial stage of the Afghan War which was, in fact, prosecuted using only U.S. SOF and local Afghan militias, describe the state of Afghanistan immediately following the Taliban ouster, and argue that a continuation of the SOF-only COA could have achieved the U.S.' strategic objective in Afghanistan.  Comments will be made on the relevance of this analysis to the future of Afghanistan as well as to future U.S. engagements in counterinsurgencies.

Methodology

The case for the SOF-only COA will be made almost exclusively using historical evidence predating the arrival of conventional U.S. forces in Afghanistan.  The operations involving U.S. SOF teamed with Afghan militias in late 2001 will be used to illustrate the combat effectiveness of this combined force.  A comprehensive review of insurgencies since World War II will be used to assess the risks of regime collapse during an insurgency.  As the U.S.’ strategic objective in Afghanistan was to combat terrorism, irrespective of an ongoing insurgency, historical evidence of successful counterterror (CT) missions will be presented to assess whether CT can be performed in the midst of a chaotic insurgency.  Regarding this last point, some evidence will be drawn from sources that postdate the arrival of conventional U.S. forces in Afghanistan, as CT missions conducted by SOF in this environment are a valuable source of information.

Findings and Analysis

Part of the appeal of a SOF-only COA for Afghanistan stems from the incredible success U.S. SOF had in the first three months following 11 September 2001.  Due to a combination of senior officials’ reluctance to employ large amounts of U.S. troops and the logistical challenges of deploying a large force quickly,[iv] the first US soldiers on the ground in Afghanistan were a small contingent of U.S. Army SOF supported by conventional air power.  Operating under the name Task Force Dagger, on 19 October 2001, two 12-man Operational Detachment Alphas (ODAs) linked up with militias from the Afghan Northern Alliance (the primary adversary of the Taliban in Afghanistan) and within a month a total of 10 ODAs were on the ground working with militias, including one led by eventual Afghan president Hamid Karzai.  Collectively, this combination of ODAs, air power, and Afghan militias cut through the Taliban with incredible speed, culminating in the capture of Kandahar on 7 December 2001, and the disposal of the Taliban from power.[v]  In total, the Taliban suffered approximately 3,500 casualties compared to 52 for the U.S./Afghan alliance.[vi]

What Dagger in essence showed was the new Afghan regime that was to be formed had a powerful army at its disposal:  U.S. SOF working with local militias, a fraction of which could have been reconstituted into an army dedicated to defending the new regime.  The militiamen involved in Dagger all had their own loyalties, but a good approximation of how many of these were dedicated to the defense of the new Afghan regime is 22,500.[vii]  In contrast, the Taliban had approximately 40,000 personnel, composed of native Afghans, al-Qaeda fighters, and Pakistanis, but with the casualty rates cited above manpower comparisons are not germane.  The combat effectiveness of Dagger made clear that if the Taliban were to fight back, they would either need to receive vastly enhanced conventional military support or fight as guerrillas.  Even the Taliban’s main backer at the outset of the war, Pakistan, surely would not have provided such conventional support; this would have been easily traceable and invited international backlash given the circumstances following 9/11.  Thus, the Taliban’s only option would be to launch a guerrilla campaign, which of course they did in the years following their ouster.

In evaluating the SOF-only COA, the question then begs as to how a regime with conventional military superiority can be defeated in an insurgency.  A RAND Corporation study titled “Paths to Victory” analyzed all major insurgencies post-WWII and attempted to identify causal factors for victory and defeat.[viii]^,[ix]  One factor included in their analysis was a binary indicator for the following statement:  “COIN force of sufficient strength to force insurgents to fight as guerrillas (or to prevail in the preponderance of conventional engagements, should overmatched insurgents choose to give battle).”  By analyzing the insurgencies RAND coded as a COIN loss, I looked for losses where this indicator was true and assessed whether the cause of defeat translates to a legitimate risk in the hypothetical Afghan War, where the new regime is supported only by U.S. SOF and air power.  I also filtered COIN losses down to those that did not end due to the withdrawal of a foreign occupying force, on the conceit that U.S. presence is small in the SOF-only COA, sustainable, and that forces could even be redeployed quickly were the Afghan regime to experience trouble following a withdrawal of U.S. SOF.  This left only eight COIN losses, for which I identified three general causes of defeat.

First, counterinsurgencies were observed to fail despite vast military superiority when the regime in power became so unpopular the military refused to follow the orders of their leader.  For example, in Colombia (1948 – 1958) opposing liberal and conservative factions organized a political settlement that would unseat the current president, Rojas Pinilla, who then ordered the arrest of the organizers.  Top military officials defected and deposed Pinilla.[x]  Similar examples include Nicaragua (1981 – 1990), Nepal (1996 – 2006), and Papua New Guinea (1988 – 1998).  In the context of the Afghan War, a similar situation could occur if powerful individuals capable of influencing the militiamen dedicated to defending the Karzai presidency (or any of his successors) called for his disposal.  If this occurred because Karzai was widely unpopular it wouldn’t be an inherently bad thing, but the U.S. nonetheless wouldn’t want anyone to be capable of utilizing the Afghan defense force to advance their own self-interest.  Mitigations for this risk will be discussed after reviewing other causes of COIN defeat.

A second cause of defeat occurred when a foreign power intervened in an ongoing insurgency.  For example, in the brief counterinsurgent campaign Pakistan launched to try to prevent Bangladesh from becoming independent (1971), India intervened to stop the genocidal nature of the counterinsurgency.[xi]  This scenario is unlikely in Afghanistan:  the U.S. can use her influence to prevent any blatantly genocidal behavior from the Afghan regime, and no foreign power is likely to militarily challenge U.S. interests in any case.

The final cause of defeat found in the historical case studies occurred when a counterinsurgent with de jure governmental authority in a region tried to regain control from an insurgent with de facto control, such as Russia’s war with Chechnya from 1994 – 1996.  Despite generally sound COIN practices, Russia failed to gain control of Chechnya, illustrating that counterinsurgencies are hard to win when the population you’re trying to gain control over doesn’t desire a change in de facto governance.  Afghanistan is historically tribal in nature and resistant to any form of central governance,[xii] and post-9/11 history has shown that while neither the Taliban nor the U.S.-backed regime is welcome in many districts, the Taliban is often preferred.[xiii]  In our counterfactual scenario (and in fact in any future counterinsurgencies, the U.S. must remain cognizant that no matter how well COIN tactics are applied they may fail in generating support in many parts of Afghanistan, and careful thought should be given to how much control is truly necessary to achieve the strategic objective.

Of the three causes of defeat identified above only two appear pertinent to the U.S.' situation following the fall of the Taliban regime:  a disloyal military turning against the central government, and attempting to overextend the control of the central government.  Given the SOF-only COA relies initially on militiamen hastily constituted into an army to defend the newly found regime, disloyalty is a major concern.  Militias are loyal to individual leaders, not a conceptual elected office, and the initial militiamen who were tasked with the defense of Afghanistan in 2002 were largely loyal to the former Northern Alliance leaders rather than the elected President Karzai.[xiv]  The long-term implication of this is that Afghanistan would need to transition towards a defense force designed to be unambiguously loyal to elected officials rather than specific individuals.  Such a force would be composed of tribally diverse, highly-vetted individuals, roughly equal in size to the initial militiamen employed in 2002 (22,500).  In addition, so that U.S. SOF wouldn’t need to support the Afghans in perpetuity, an indigenous group of elite special forces personnel must be trained and equipped with the advanced weapons systems that allowed Task Force Dagger to be so effective.  Task Force Dagger was, to a good approximation, composed of one U.S. Army Special Forces Group (SFG) and accompanying personnel to provide additional air power;[xv] this force could be conservatively replaced with 3,000 additional highly trained personnel.  While training both the regular and special Afghan forces would take time, the goal of training some 25,500 forces is very reasonable.  The actual COA employed by the U.S. of training a large ANSF on the scale of 300,000 personnel has been fraught with issues, ranging from a lack of motivation to a lack of ability on the part of recruits,[xvi] but the U.S. has successfully trained an elite group of soldiers with proven combat effectiveness called the Afghan National Army Command Corps (ANACC) which was 21,000 strong in 2017, with plans of expansion.[xvii] The ANACC was only founded in 2007, and presumably the training timeline could have been sped up had the training of a small, elite force been the COA from day one.  A force structure of 25,500 also would have been sustainable given the state of the Afghan economy in 2002, which could support an estimated 30,000 personnel.[xviii]

It’s worth saying a few words about why, in reality, the U.S. sought to develop a large ANSF that currently stands about 300,000 strong, even though this was never sustainable without the U.S.’ financial support.[xix]^,[xx],[xxi]  This figure, coupled with U.S. and other foreign troop support, is most likely informed by COIN doctrine that recommends large amounts of troops for every 1,000 inhabitants of a region, and a hesitancy to rely on local, autonomous militias in the defense of Afghanistan.  Most of the literature on the use of local militias in the Afghan War is negative or at best mildly supportive, but these critiques are almost always written within the framework of the large ANSF COA and view militias are a stop-gap solution while the ANSF is in development.[xxii]^{,[xxiii],[xxiv]}  They cite legitimate issues such as the shifting loyalties of militias, human rights abuses, and fighting amongst militias, but they fail to acknowledge most of the benefits militias provide.  Not only are militias a financially feasible solution for national defense, but Afghanistan has historically been resistant to centralized governance and thus a more realistic national defense strategy may involve influencing militias to act in the national interest.[xxv]  In fact, before the group-think set in regarding the primacy of a large ANSF, a model for the ANSF was proposed by Anja Manuel and Peter Singer that called for a small, centralized force of 30,000 troops (similar to the SOF-only COA), while utilizing local militias in what was described as “a ‘national guard’ that incorporates tribal and warlord militias into formal units responsible to provincial governments.” [xxvi]  While these militias would remain relatively autonomous, the central Afghan government could influence their behavior via economic agreements and by giving them a seat at the political decision-making table.  What Manuel and Singer don’t mention, which creates further incentive for the militias to act favorably towards the government, is the extraordinary combat capability of a small ANSF modeled on Task Force Dagger; not only does this provide a “stick” with which to threaten militias hostile to the central government, but it’s also a “carrot” that can be offered to aid local militias under threat from a resurgent Taliban.

The second of the two concerns found through examining historic cases of COIN failure was that expanding territorial control of the central government is often difficult.  Part of the logic behind the use of local militias in a national guard framework of defense is to create regions that are autonomous yet cooperative with the central government, thus avoiding the risk of defeat in insurgent warfare all together.  Yet, the U.S. still would need to plan for the eventuality that many of these autonomous regions will be poorly governed and policed, and become subject to attack from a resurgent Taliban, a safe haven for terrorists, or both.  Given the U.S.’ objective in Afghanistan was to combat terrorism, to evaluate the SOF-only COA it must be assessed whether effective CT can be performed within such an environment.  Under the SOF-only COA, while the U.S. doesn’t seek to lock-down the entire country of Afghanistan with a conventional COIN campaign, it does gain a base of support from which it can launch CT missions; this is precisely what was lacking under the Taliban regime and necessitated Task Force Dagger.  It should be noted that Dagger was launched out of the Karshi-Khanabad (K2) airbase in Uzbekistan, but diplomatic U.S.-Uzbek relations have since soured, highlighting the need to establish a solid base in Afghanistan.

One source of evidence for the U.S.’ ability to conduct CT in a hostile environment is the U.S.’ own experiences in Afghanistan thus far.  While this is somewhat at odds with a counterfactual analysis of the SOF-only COA, the reality is that even under the current COA, the Afghan government only has uncontested control of approximately 53.8% of the country,[xxvii] and U.S. SOF have performed many CT missions within this environment.  Navy SEALs famously raided Osama bin Laden’s compound near the Pakistani border and killed the al-Qaeda leader, but this is only one of many successful direct action (DA) missions by SOF.  In 2010, for example, SOF conducted an average of five raids per day with about half capturing the intended target, and the remaining missions often generating valuable intelligence.[xxviii]  In addition to DA, SOF have led hunter-killer teams through ungoverned territory in Afghanistan to capture al-Qaeda and Taliban forces in hiding and seize weapons caches.  A successful example of this occurred in 2003 when teams discovered Taliban forces were coalescing in the mountains near Kandahar; SOF then led a team of indigenous forces and air power to attack the Taliban, destroying most and forcing the remainder to flee.[xxix]

The U.S.’ experiences in Afghanistan provide confidence CT missions can be equally successful under the SOF-only COA, but no discussion of CT in a hostile environment is complete without talking about Israel’s historical record on this issue.  Israel has conducted successful CT missions throughout the Middle East and North Africa, all originating out of its home territory.  Israel has conducted elaborate, long-range DA missions in areas as distant as Iraq and Tunisia;[xxx] emulating this capability would allow U.S. SOF to perform DA anywhere in Afghanistan, utilizing a solid base of support in Kabul.  Israel has also performed larger scale CT operations in hostile territory.  For example, during the Lebanese Civil War (1975 – 1990), Israel successfully expelled the Palestine Liberation Organization (PLO) from Lebanon in 1982 by way of Operation Peace for Galilee, in which the Israeli Defense Force (IDF) maneuvered from their occupied territory in southern Lebanon to Beirut to attack the PLO, causing them to relocate to Tunisia.[xxxi]  The expulsion of the PLO by no means ended the Lebanese Civil War, which was a messy, multi-player war not unlike the current war in Afghanistan, but Israel’s CT objective was achieved.  The lesson for the U.S. in Afghanistan is that under the SOF-only COA, al-Qaeda strongholds could have been pursued in attempts to eliminate Afghanistan as a terrorist safe haven.  The U.S. needn’t have lost sight of the objective of combating terrorism by becoming too engaged in the civil war against the Taliban.  Larger CT missions such as Operation Peace for Galilee would require Afghan troops to work with U.S. SOF.  Assuming an ongoing insurgency, the Afghan regime’s focus would be on fighting the Taliban rather than al-Qaeda and other terrorists; a natural quid pro quo is that in exchange for U.S. assistance in defending the central regime around Kabul, the regime in turn assists the U.S. in combating terrorism.

To conclude the discussion on the risk of defeat in a counterinsurgency brought on by trying to overextend the control of the central government, I’ve argued that exerting influence on local militias lessons the need to extend government control and that effective CT can be performed as long as the U.S. has a solid base of support in country.  Under the SOF-only COA, in addition to teaming with local Afghan forces to ensure the regime is protected, U.S. SOF would have performed both DA and more sustained CT missions to combat al-Qaeda.  As the ANSF develops and U.S. forces departs, turning over the mission of protecting the regime to the ANSF wouldn’t cause an agency problem, but the CT mission may lag in quality.  Maintaining basing rights to in Afghanistan is therefore important to deploy small numbers of SOF, such as a SEAL Team or an ODA, to execute high-priority CT missions based on actionable intelligence.  In fact, being able to execute such small deployments anywhere in the world should be considered a core competency if the U.S. truly intends to fight terror on a “global” level.

Conclusion

This essay has argued the U.S.’ strategic objective to combat terrorism in Afghanistan could have been achieved using only U.S. SOF, air power, and a modestly sized force of indigenous ground troops.  The justification of the actual COA employed of using a large deployment of conventional U.S. troops and a large ANSF was that by deploying the doctrinally recommended number of troops to achieve victory in a counterinsurgency, the U.S. would in turn be victorious in combatting terrorism.  The SOF-only COA, on the other hand, maintains that effective CT can be performed in the midst of an insurgency.  This is possible by establishing a strong base of support in Afghanistan from which long-range CT missions can be launched.  The historical record of Task Force Dagger provides confidence that neither the Taliban nor any other insurgent group would be capable of toppling a regime backed by U.S. SOF, and successful long-range CT missions into hostile territory by the U.S. and Israel provide confidence the U.S. could make Afghanistan inhospitable for al-Qaeda without attaining complete control of Afghanistan.  Further advantages of the SOF-only COA are that it’s economically feasible, and relies on building mutually beneficial relations with local militias.

While this essay was counterfactual and provides lessons for how the U.S. ought to think about future counterinsurgencies, the discussion can’t be complete without considering how the analysis presented here relates to the current situation in Afghanistan, where the government has collapsed with the removal of U.S. military support.  Recall the three crucial factors identified for the SOF-only COA to succeed.  Most fundamentally, the indigenous forces coupled with U.S. SOF and air power must be able to prevail in the preponderance of conventional battles.  Second, the indigenous military must be willing to follow the orders of the regime in power.  Third, it was observed that counterinsurgencies are hard to win where the counterinsurgent attempts to extend its influence into communities where lacks popularity relative to a local de facto government.  While it’s premature to draw conclusions regarding the collapse of the Afghan government upon the U.S.’ withdrawal, all three factors could have raised concerns.  The open-source reporting has not indicated U.S.-SOF were fighting alongside Afghan partners as they surrendered, and it’s been speculated air power wasn’t provided with the same regularity during the drawdown.[xxxii]  A lack of U.S. involvement during the Taliban’s offensive would have put enormous pressure on the ANACC to act as a vanguard, and perhaps overwhelmed their capacity.  The ANSF often surrendered without a fight; while this may indicate Afghan commanders believed they couldn’t win individual battles, it’s more likely an indication they weren’t willing to fight for the government.  Corruption and competency within the Afghan central government have long been concerns, while the Taliban, on the other hand, are at least perceived as an organized movement.  Commanders may have rationally been positioning themselves on the right side of history with their surrender.  Regarding the third factor, polls indicated very low popular support for the U.S.-backed central government in many regions of Afghanistan.[xxxiii]  While the central government wasn’t in direct conflict with local, de facto government bodies, they couldn’t marshal support from local militias to resist the Taliban’s offensive.[xxxiv]  The U.S.-backed regime’s engagement with local leaders as part of their strategy for governing Afghanistan has experience fits and starts over the 20-year war, but has not been prioritized in recent years.  A key program to empower local militias (the Afghan Local Police) was completely defunded during this critical time.  In short, none of the prerequisites for the SOF-only COA appear to have been in place during the U.S. withdrawal, and the Afghan government’s collapse is in fact aligned with the historical record of counterinsurgency failures.

References

Akmam, Wardatul. “Atrocities against Humanity during the Liberation War in Bangladesh: A Case of Genocide.” Journal of Genocide Research 4, no. 4 (December 2002): 543–59. https://doi.org/10.1080/146235022000000463.

Asia Foundation. “A Survey of the Afghan People,” n.d. http://asiafoundation.org/where-we-work/afghanistan/survey/resources/#archive.

Celeski, Joseph D. Hunter-Killer Teams: Attacking Enemy Safe Havens. Hurlburt Field, Fla.: Joint Special Operations University Press, 2010.

Chan, Samuel. “Sentinels of Afghan Democracy:  The Afghan National Army.” Military Review, February 2009, 25–40.

Cobban, Helena. The Palestinian Liberation Organisation: People, Power, and Politics. Cambridge Middle East Library. Cambridge [Cambridgeshire] ; New York: Cambridge University Press, 1984.

Cooper, Helene. “Afghan Forces Are Praised, Despite Still Relying Heavily on U.S. Help.” New York Times, August 20, 2017.

Crawford, Neta. “United States Budgetary Costs of Post-9/11 Wars through FY2018:  A Summary of the $5.6 Trillion in Costs for the US Wars in Iraq, Syria, Afghanistan and Pakistan, and Post-­‐‑9/11 Veterans Care and Homeland Security.” Brown University, Costs of War, 2017.

Feickert, Andrew. “U.S. Special Operations Forces (SOF):  Background and Issues for Congress.” Congressional Research Service, July 16, 2010.

Felbab-Brown, Vanda. “Hurray for Militias? Not so Fast: Lessons from the Afghan Local Police Experience.” Small Wars & Insurgencies 27, no. 2 (March 3, 2016): 258–81. https://doi.org/10.1080/09592318.2015.1129169.

Gibbons-Neff, Thomas, Helene Cooper, and Eric Schmitt. “Departure of U.S. Contractors Poses Myriad Problems for Afghan Military.” New York Times, June 20, 2021.

Grenier, Robert. 88 Days to Kandahar: A CIA Diary. First Simon&Schuster Paperbacks trade paperback edition. New York: Simon & Schuster Paperbacks, 2016.

Henriksen, Thomas H. The Israeli Approach to Irregular Warfare and Implications for the United States. Hurlburt Field, Fla.: JSOU Press, 2007.

Iqbal, Anwar. “Afghan Army to Collapse in Six Months without US Help:  Ghani.” Dawn.Com, January 18, 2018.

“Joint Special Operations Task Force - North (JSOTF-N) (Afghanistan) ‘Task Force Dagger.’” GlobalSecurity.Org, n.d. https://www.globalsecurity.org/military/agency/dod/jsotf-n-af.htm.

King, Marvin L., III. “Optimizing Counterinsurgency Operations.” Colorado School of Mines, 2014.

“Leave Them in No Peace:  America’s Afghan Exit.” The Economist Radio, July 5, 2021.

Lefevre, Mathieu. “Local Defence in Afghanistan:  A Review of Government-­backed Initiatives.” Afghanistan Analysts Network, May 2010.

Livingston, Ian S., and Michael O’Hanlon. “Brookings:  Afghanistan Index.” Brookings Institute, September 29, 2017.

Long, Austin G., ed. Locals Rule: Historical Lessons for Creating Local Defense Forces for Afghanistan and Beyond. Santa Monica, CA: RAND/National Defense Research Institute, 2012.

Lyall, Jason, Graeme Blair, and Kosuke Imai. “Explaining Support for Combatants during Wartime: A Survey Experiment in Afghanistan.” American Political Science Review 107, no. 04 (November 2013): 679–705. https://doi.org/10.1017/S0003055413000403.

Manuel, Anja, and P. W. Singer. “A New Model Afghan Army.” Foreign Affairs 81, no. 4 (2002): 44. https://doi.org/10.2307/20033239.

Paul, Christopher, Colin P. Clarke, Beth Grill, and Molly Dunigan. Paths to Victory: Detailed Insurgency Case Studies. Santa Monica, CA: RAND, 2013.

———. Paths to Victory: Lessons from Modern Insurgencies. Santa Monica, CA: RAND, 2013.

Peltier, Isaac J. “Surrogate Warfare: The Role of U.S. Army Special Forces.” Fort Leavenworth, KS: U.S. Army Command and General Staff College, May 26, 2005. https://doi.org/10.21236/ADA436109.

Rempe, Dennis M. The Past as Prologue?  A History of U.S. Counterinsurgency Policy in Colombia, 1958-66. Carlisle, Pa.: Strategic Studies Institute, U.S. Army War College, 2002. http://books.google.com/books?id=0ndsAAAAMAAJ.

Roggio, Bill. “Analysis: US Military Downplays District Control as Taliban Gains Ground in Afghanistan.” Long War Journal, January 31, 2019. https://www.longwarjournal.org/archives/2019/01/analysis-us-military-downplays-district-control-as-taliban-gains-ground-in-afghanistan.php.

“The Taliban Takeover of Afghanistan.” Middle East Focus. Middle East Institute, August 18, 2021.

Thrall, A. Trevor, and Erik Goepner. “Step Back:  Lessons for U.S. Foreign Policy from the Failed War on Terror.” Policy Analysis, the Cato Institute, no. 814 (June 26, 2017): 26.