Recidivism and Detention in Afghanistan

Since the War on Terror began in 2001, the U.S. has detained thousands of people in connection with terrorism. The targets of the war include “persons who were part of, or substantially supported, Taliban or al-Qaida forces or associated forces that are engaged in hostilities against the United States or its coalition partners, including any person who has committed a belligerent act, or has directly supported hostilities, in aid of such enemy armed forces.”[1] This criterion is used to assess whether continued detention is necessary to mitigate the security threat any particular detainee poses to U.S. or coalition forces.[2] Of the people who have been released from detention, many return to their homes and resume a peaceful life. However, some of the released detainees return to the fight against U.S. or coalition forces, and this phenomenon is called recidivism. This post will explain why examining the recidivism rate in Afghanistan is helpful, how the U.S. identifies recidivists, what the recidivism rate is according to reports from the U.S. military, how the extremely low reported recidivism rate may or may not be attributable to new review procedures and the detainee reintegration program, and what the reasons are to doubt the accuracy of the low rates reported by the military.

An examination of the recidivism rates in Afghanistan is helpful for assessing the progress of the U.S. mission in Afghanistan. Fear of high recidivism is one of the U.S.’s principal justifications for the prolonged detention of some detainees. By taking a closer look at recidivism rates, we can determine whether those fears are justified. An examination of recidivism gives us greater understanding of the insurgents’ commitment to their cause. Greater understanding of recidivism can also help in determining whether the U.S. should continue to detain the 66 third country nationals currently being held at Bagram.[3] Furthermore, looking at the recidivism rates can help verify the credibility of recidivism reports from Afghanistan, and in light of the scheduled U.S. withdrawal from Afghanistan in 2014, the credibility of recidivism reports from Afghanistan can help us assess the overall results of U.S. efforts; if the official reports provide an overly optimistic view of the situation in Afghanistan, then there may be reasons the U.S. should not withdraw from Afghanistan.

Before we can examine the recidivism rates, we should examine the process by which recidivism rates are calculated and recidivists are identified. U.S. forces use biometrics to identify and classify possible recidivist-detainees. While the exact procedures and criteria used to identify a recidivist-detainee remain classified, the Defense Intelligence Agency has provided some definitions from 2007 that shed light onto those procedures. Recaptured detainees are classified as “confirmed” or “suspected.”[4] The Department of Defense uses “fingerprints, DNA, conclusive photographic match, or reliable, verified or well-corroborated intelligence reporting” to identify a detainee, and to “confirm” him as “directly involved in terrorist activities.”[5] The “suspected” classification is limited to a detainee who “most likely associated with a specific former detainee” or as to whom there have been “unverified or single-source, but plausible” reports of involvement in terrorist activities.[6]

U.S. forces have given optimistic reports on recidivism in recent years. Vice Admiral Robert Harward, the detention operations chief, said in 2010 that “very, very few” detainees released from Bagram return to the battlefield.[7] As of January 2010, according to the Vice Admiral, “3000 Afghans have been detained by U.S. forces in eight years, and the U.S. could document a total of 17 ex-detainees who had returned to the fight, less than half of one percent.”[8] The Vice Admiral later stated that “of this year [2010], 550 have been released, there have been four individuals recaptured on the battlefield.”[9] Other reports from the U.S. government have given similar outlooks. In the Pentagon’s 2010 annual report to Congress on Progress Toward Security and Stability in Afghanistan, the Pentagon reported “the current recidivism rate for released detainees in Afghanistan is 1.2 percent.”[10]

One reason the recidivism rate is so low may be due to the new procedures for determining each detainee’s threat. In 2009, a new set of DRB procedures were introduced, with the intent of speeding up the processing of each detainee’s case and also empowering each detainee to better plead his case with the help of a personal representative.[11] Along with these procedural changes, U.S. detention centers also began segregating “the irreconcilables, those constituting an ongoing threat, from individuals who could be rehabilitated.”[12] Given the low reported recidivism rate, it appears that the new procedure is “successful in separating non-threats and the reconcilable from the irreconcilable.”[13] Reconcilable detainees are then enrolled into the reintegration program, the efficacy of which will be examined shortly.

Another possible reason the recidivism rates are low is due in part to the U.S.-backed and Afghan-operated reintegration program to educate the detainees, teaching these detainees valuable skills so that they can reintegrate into society after release. Combined Joint Interagency Task Force 435 (CJITF-435), the U.S. military task force responsible for U.S. detention operations, and for aiding the Afghans to operate their own detention program, assesses that the reintegration program is working to prevent previously detained individuals from rejoining the insurgency.[14] The reintegration program includes three phases. It begins with provincial officials reaching out to interested insurgents and resolving any local conflicts that may have pushed the insurgents to violence in the first place.[15] The next phase involves reviewing the detainee’s past actions and providing the detainee with “protection from targeting by government [Afghan] or international forces,” along with $120 dollars per month for the first three months for his basic needs.[16] It is in this “demobilization” phase that a detainee is expected to renounce violence and terrorism, and “becomes eligible for political amnesty.”[17] Finally, the detainee is given literacy training, vocational training in fields such as “tailoring, agriculture or baking,” and civics classes.[18] However, the reintegration program is plagued with numerous problems, such as poor internal coordination, the lack of a “systemic solution for their [the detainees’] personal or family’s security,” and concerns over corruption, all of which undermine the program’s success.[19] The problems with the reintegration program also give rise to doubts about the accuracy of the recidivism rates in Afghanistan.

There are three reasons to doubt the accuracy of those previously mentioned reports. First, the recidivism rate of detainees transferred to Afghan control may be higher. Vice Admiral Harward has said the recidivism rate for detainees is less than half of one percent, but he did not have any information regarding the recidivism rates of “insurgents turned over to the Afghan authorities.”[20] Afghan detention procedures are “vaguely worded, resulting in troubling uncertainty over who can be detained, on what grounds, and for how long.”[21] Furthermore, the release of the detainees is “widely seen as a potentially powerful bargaining chip in negotiations with the Taliban, and there could be multiple opportunities for political interference.”[22] The political interference means that some of the detainees under Afghan control could be released without any assessment of the threat that they pose or the likelihood of their return to the insurgency. “The lack of any rules or transparency governing the [Afghan] Commission’s role in making a decision to release a detainee leaves the system open to arbitrary decision making and political horse-trading.”[23] Given that the U.S. transferred custody of over 3000 detainees to Afghanistan early this year, and that the likelihood of recidivism is much higher for detainees under Afghan control, there are serious questions regarding the timing of U.S. withdrawal from Afghanistan and the overall anti-terrorism mission’s progress.

Another reason to doubt the optimistic official reports is that the reported recidivism rates for former Guantanamo detainees have been much higher.[24] In a 2012 summary, according to the journalist Andy Worthington, the Director of National Intelligence reported “of the 599 prisoners released from Guantanamo, 95 (15.9%) are described as ‘Confirmed of Reengaging,’ and 72 others (12%) are described as ‘Suspected of Reengaging.’”[25] These numbers are much higher than the previously mentioned statistics from Afghanistan. This discrepancy can mean that the program in Afghanistan is extremely effective at reducing the rate of recidivism in comparison to the program in Guantanamo; or the Guantanamo detainees are more likely to be hardliners who are irreconcilable; or alternatively, the rate of recidivism from Afghanistan is actually higher than the rate in the official reports.

Furthermore, the way that U.S. forces calculate recidivism may be inherently inaccurate. The percentage that the Vice Admiral was referring to only accounted for previous detainees who have been recaptured by U.S. forces,[26] meaning that the calculated recidivism rate excludes possible recidivist-detainees who were later killed after their release but whose bodies could not be identified for various reasons. [27]  If a former detainee has returned to the fight but has avoided recapture, he too would not be factored into the recidivism rate estimate.

Examining the recidivism rate critically is a vital aspect in assessing U.S. progress in Afghanistan, particularly because the war has gone on for over a decade and will possibly come to an end in the near future. If the recidivism rate is truly as low as one percent, then we might infer that the detention and reintegration policies have been extremely successful in converting former insurgents into peaceful civilians; and those policies should continue to be promoted to reduce terrorist threats around the globe. However, there are many reasons to doubt the reported rates, leading to two questions: how effective is the detention policy in Afghanistan, and how can it be improved?


[1] Memorandum from Deputy Secretary of Defense, Policy Guidance on Review Procedures and Transfer and Release Authority at Bagram Theater Internment Facility Afghanistan 2 (July 2, 2009), available at

[2] Id.

[3] Barack Obama, Press Release, Letter from the President — Regarding the War Powers Resolution, The White House (June 14, 2013), available at

[4] The Constitution Project, The Report of the Constitution Project’s Task Force on Detainee Treatment 298 (2013), available at

[5] Id.

[6] Id.

[7] Spencer Ackerman, U.S. Detentions Chief in Afghanistan Says Recidivism Is Very Low, Wash. Indep., (Jan. 27, 2010, 11:03 am), available at

[8] Id. (reporting Harwards figures).

[9], DOD News Briefing with Vice Adm. Harward from Afghanistan, U.S. Dep’t of Defense (Nov. 30, 2010),

[10] U.S. Dep’t of Defense, Progress Toward Security and Stability in Afghanistan 61 (Nov. 2010), available at .

[11] The Constitution Project, supra note 4, at 79.

[12] Id, at 80.

[13] Robert Chesney, Detainee Recidivism in Afghanistan, Lawfare (Nov. 30, 2010, 10:23 am),

[14] The Constitution Project, supra note 4, at 98.

[15] Deedee Derksen, Peace from the Bottom-Up? The Afghanistan Peace and Reintegration Program 6 (May 2011), available at

[16] Id.

[17] Id.

[18] Combined Joint Interagency Task Force 435 Public Affairs, Officials OK Afghan Detainee Vocational Program, U.S. Department of Defense (Apr. 5, 2011), .

[19] Derksen, supra note 15, at 11-17.

[20] Ackerman, supra note 7.

[21] Open Society Foundations, Remaking Bagram: The Creation of an Afghan Internment Regime and the Divide over U.S. Detention Power 3, 22 (Sept. 6, 2012), available at .

[22] Id. at 26.

[23] Id.

[24] Andy Worthington, Guantanamo and Recidivism: The Media’s Ongoing Failure to Question Official Statistics (March 14, 2012),

[25] Id.

[26] U.S. Dep’t of Defense, supra note 9.

[27] Id.