Abstract: The Islamic State is mobilizing children and youth at an increasing and unprecedented rate. The authors present preliminary findings from a new database in which they recorded and analyzed child and youth “martyrs” eulogized by the Islamic State between January 2015 and January 2016. The data suggests that the number of child and youth militants far exceeds current estimates. The article presents data on the children and youth’s country of origin, age, role, location of death, and under what circumstances they were killed. The authors also describe several trends in the propaganda before discussing the varied and complex implications of the Islamic State’s long-term vision for its children and youth.
Violent extremist organizations (VEOs) are mobilizing children at an ever-accelerating rate. The Pakistani Taliban run several so-called schools dedicated to graduating prepubescent bombers, Houthi rebels in Yemen have routinized the inclusion of children in their ranks, while the Lebanese Hezbollah has begun accepting adolescents into its ranks to boost its presence in Syria. The mobilization of children into VEOs (and featuring them in propaganda) is not a new phenomenon and has many historical antecedents. This is perhaps most common in the context of child soldiers. However, the Islamic State has so heavily championed the mobilization of children—on a scale rarely associated even with VEOs—that it suggests organizational concerns that far outweigh short-term propaganda benefits. That is not to minimize the importance of the latter for the Islamic State. Indeed, the publicity-hungry organization vividly depicts the wide-ranging and routine participation of children in its jihadist media projects.
Unsurprisingly, the majority of media commentary has highlighted the most public and dramatic roles played by children, namely as executioners in firing squads or beheadings. However, the presence and participation of children in the comprehensive corpus of Islamic State propaganda extends beyond ultra-violence. Indeed, on an almost daily basis, children are featured in multiple contexts, from highly publicized executions and training camps to Qur’an memorization fairs and dawa caravans.
So far, analysis of this phenomenon has been scant, with a few exceptions. While it is a well-known fact that the Islamic State uses child soldiers—defined here using a modification of the United Nations definition of “any person [of 18 years of age or under] who is, or who has been, recruited or used by an armed force or armed group in any capacity”—the nature and range of circumstances of their recruitment and engagement are rarely explored.[a] We seek to redress this gap by drawing on an original database of photographic martyrdom propaganda featuring children and youth. A cursory exploration of this data suggests that the Islamic State’s systematic use of children is more widespread than previously imagined. Indeed, the rate of youth deaths in the Islamic State’s name between January 2015 and January 2016 is more than twice the most regularly cited estimate.[b]
Below, after describing our methodology, we present findings from the archive, outlining key features and identifying, among other things, details about the children’s origins, places of death, and roles they played for the Islamic State. We then explore trends in the organization’s youth deployment before offering preliminary conclusions regarding the routinization and acceleration of children’s engagement on the Islamic State battlefield.
We recorded instances of young people (both younger children and youth) who were featured in official Islamic State reports as “martyrs” from January 1, 2015, to January 31, 2016, downloading the photographs and coding them into a database. Other data were recorded prior to this period but not systematically (nor was the earlier data subject to coder reliability checks), and thus were excluded. The majority of the 89 images collected, which were invariably of male children and youth, were sourced from Islamic State propagandists on Twitter. Three of the photographs were unofficially disseminated by known Islamic State fighters also via Twitter. Over the course of the year, as targeted account suspensions accelerated and Twitter became less hospitable to jihadis, propaganda dissemination shifted to the hybrid messaging app Telegram. Reflecting this, some of the data are also drawn directly from the Islamic State’s official Telegram channel. The constantly shifting nature of Islamic State propaganda dissemination means that we cannot claim (nor do we seek to) that this database is exhaustive. Rather, it represents a snapshot in time of how the Islamic State featured children and youth in its martyrdom propaganda.
Once the 89 images were downloaded, each was tagged and coded according to 24 variables, including the subject’s kunya (nom de guerre), nationality, location of death, facial expression, and backdrop/setting.[c] Each image was logged with as much detail as possible, such that any anomalies or outliers could be highlighted in the descriptive analysis.
When the Islamic State releases images of martyred children and youth, it does not provide biographical details. This created a significant obstacle to analysis. To resolve this challenge and determine country of origin, the individual’s kunya was used as a proxy for nationality—for example, al-‘Iraqi denotes that they are/were Iraqi, whereas al-Lubnani would correspond to being Lebanese. However, in 24 percent of the cases, where the kunya was al-Ansari, it was not possible to determine with a high degree of certainty whether the nationality was Syrian or Iraqi (as it could feasibly be either). For that reason, these cases of children and youth who died in Syria and Iraq were coded as “Syrian/Iraqi.”[d] The remaining eight children and youth had kunyas that did not designate their origins—for example al-Muhajir (literally, ‘the emigrant’) or al-‘Ilami (‘the media man’)—and were coded as “undesignated.” Age is another key detail not reported by the Islamic State. As such, the sample was classified according to categories provided by Dr. Heidi Ellis and Ms. Elizabeth Nimmons of Boston Children’s Hospital’s Refugee Trauma and Resilience Center and follow from age categories for child development—“Pre-Adolescent” (8 to 12), “Adolescent” (12 to 16), or “Older Adolescent” (16 to 21).[e] We used inter-coder reliability checks for this classification. Each image was independently coded by three people and, where possible, cross-compared with media reports.[f] In all but 10 cases, the age categories were coded consistently. The remaining 10 were discussed and reconciled by the entire team.
A similar database of adults featured in martyrdom propaganda from November 19, 2015, to January 31, 2016, was assembled as a control sample to contrast trends in adult eulogies with those of the children and youth, thereby allowing us to determine whether significant variation was present (location, type of target, role, etc.). Photographs of each adult featured in official Islamic State “martyrdom” propaganda were downloaded and coded to the same variables. It is worth noting, at this juncture, that there are more adult eulogies available—114 adults in the space of 73 days, compared to 89 youth in 395 days—because more adults than children and youth are involved in the Islamic State’s military operations.
From January 1, 2015, to January 31, 2016, 89 children and youth were eulogized in Islamic State propaganda. Fifty-one percent were alleged to have died in Iraq, while 36 percent died in Syria. The remainder were killed during operations in Yemen, Libya, and Nigeria. Sixty percent of the sample was categorized as “Adolescent” based on Islamic State photographs, 34 percent were classified as “Older Adolescent,” and 6 percent were “Pre-Adolescent.” Thirty-one percent were Syrian, 25 percent Syrian/Iraqi, and 11 percent Iraqi. The remaining 33 percent were from Yemen, Saudi Arabia, Tunisia, Libya, the United Kingdom, France, Australia, and Nigeria.
Figure 1: Deaths and suicide operations
Of the 89 cases, 39 percent died upon detonating a vehicle-borne improvised explosive device (VBIED) against their target. Thirty-three percent were killed as foot soldiers in unspecified battlefield operations, 6 percent died while working as propagandists embedded within units/brigades, and 4 percent committed suicide in mass casualty attacks against civilians.[g] The final 18 percent were inghimasis (derived from the Arabic “to plunge”), meaning they died in what we more commonly term marauding operations in which a group of mostly adult fighters infiltrates and attacks an enemy position using light automatic weapons before killing themselves by detonating suicide belts. Forty percent of the time, the children and youth died in operations targeting state security forces (including military and police targets). Twenty-one percent were killed fighting against paramilitary forces (militias and non-state opposition), and only 3 percent carried out suicide attacks against civilians. No target was specified for the remaining 36 percent.
Figure 2: Location of deaths
A number of stylistic elements of the data are worth mentioning. In 6 percent of the photographs, the children and youth are masked. Of the remaining cases, 46 percent are depicted with smiles on their faces, 27 percent with neutral expressions, and 26 percent with angry expressions.[h] Their mode of dress is evenly distributed between casual civilian and formal military attire. Fifty-three percent are wearing battle fatigues and are often holding weapons; the rest are dressed in casual clothes, from caps and t-shirts to Afghan-style pakol or salwar kameez. The location of the photoshoots ranged broadly: 27 percent were taken in a military context (e.g. in a trench), and 22 percent were taken in an unremarkable setting (e.g. a beige room). A further 28 percent of the children and youth were standing in orchards and meadows, scenery presumably chosen to echo the paradise to which they thought they were destined. The theme of happiness at the prospect of martyrdom occurred regularly. In one case, a pre-adolescent child smiled widely at the camera, wearing his baseball cap back-to-front, sitting atop what appears to be a fairground ride. While there was an Islamic State flag featured in 19 percent of the images, none featured a Qur’an, which is an oft-used motif in jihadist martyrdom propaganda.
Figure 3: Role
The overwhelming majority of the deaths (63 percent) were celebrated in photos corresponding with Islamic State “Breaking News” updates, the blue and red infographics that appear within minutes or hours of an operation. The rest appeared sporadically in Islamic State photographic propaganda campaigns that occurred across the 13-month period under review. The most prominent of these campaigns was “Caravans of the Martyrs,” which featured in 25 percent of the data. The remaining cases were part of other, more irregular campaigns, like “Among Them Is He Who Fulfilled His Vow,” “Ink of Jihad,” and “Media Man, You Are a Mujahid.”
Figure 4: Country of origin
When comparing this data with the adult group, there are some noteworthy observations. While there was slightly more variation regarding adult roles and countries of origin, on the whole, the datasets were strikingly similar.[i] Like the youth sample, most adults were killed in Iraq. Similarly, most adults died as VBIED operators, with the remainder killed— in descending order—as foot soldiers, inghimasis, and propagandists. Target types are remarkably similar, with 46 percent of adults attacks aimed at state security forces, compared to 40 percent of the children’s.
The data unambiguously suggests that the Islamic State’s mobilization of children and youth for military purposes is accelerating. On a month-by-month basis, the rate of young people dying in suicide operations rose, from six in January 2015 to 11 in January 2016. The rate of operations involving one or more child or youth is likewise increasing; there were three times as many suicide operations involving children and youth in January 2016 as the previous January (2015). It seems plausible that, as military pressure against the Islamic State has increased in recent months, such operations—especially those of the inghimasi variety—are becoming more tactically attractive. They represent an effective form of psychological warfare—to project strength, pierce defenses, and strike fear into enemy soldiers’ hearts. We can expect that, as their implementation increases, so too will the reported rate of child and youth deaths.
It is equally striking that the Islamic State’s children and youth operate in ways similar to the adults. Children are fighting alongside, rather than in lieu of, adult males and their respective patterns of involvement closely reflect one another. In other conflicts, the use of child soldiers may represent a strategy of last resort, as a way to “rapidly replace battlefield losses,” or in specialized operations for which adults may be less effective However, in the context of the Islamic State, children are used in much the same ways as their elders.
Notwithstanding some limitations—for example, the data provides a partial snapshot of the reality (in January 2016, there were 85 suicide operations in Iraq and Syria, only 53 of which were reported in photographic propaganda[j])—this remains the first such database of its kind, and provides a baseline for future research on VEO utilization of children and youth.[k] Yet despite these limitations, we can assert with confidence that the use of children and youth has been normalized under the Islamic State’s rule. Instead of hailing them as young heroes, the Islamic State media team merely celebrates them as heroes. Indeed, the actual age of the martyr is never mentioned, even in passing. In exceptional cases, Islamic State supporters might celebrate children and youth, but, aside from one instance in which a pre-adolescent boy is photographed while bidding farewell to his father before a mission, there is no special consideration given to age or separation from family. If anything, to the Islamic State’s propagandists, the youth of the martyr is incidental.
When considered in the context of the child soldiers in other conflicts, this is somewhat counterintuitive. Historically, when militant organizations enlisted children, they did so surreptitiously, a pattern that emerged with the release of the Machel Report on children in armed conflict in 1996 and the UN resolutions against youth recruitment that followed. The Islamic State bucks this trend brazenly by boasting about its young recruits, something that is indicative of the fact that it is using them differently than the child soldier norm. The data suggests that the Islamic State is not recruiting them to replace lost manpower— children and youth only constitute a small proportion of its battlefield losses overall—and they are not engaging in roles in which they have a comparative advantage over the adults. On the contrary, in most cases, children and youth are dying in the same circumstances as adults. Additionally, existing research argues that children and youth will be used more to attack civilian targets among whom they can blend in better. However, the data shows that Islamic State’s children and youth have been used to attack civilians in only 3 percent of the cases.
It is clear that the Islamic State leadership has a long-term vision for youth in its jihadist efforts. While today’s child militants may well be tomorrow’s adult terrorists, in all likelihood, the moral and ethical issues raised by battlefield engagement with the Islamic State’s youth are likely to be at the forefront of the discourse on the international coalition’s war against the group in years to come. Furthermore, as small numbers of children either escape or defect from the Islamic State and as more accounts emerge of children’s experiences, there is an urgent need to plan and prepare for the rehabilitation and reintegration of former youth militants.
Mia Bloom is Professor of Communication at Georgia State University and author of Dying to Kill: The Allure of Suicide Terror and Bombshell: Women and Terror. Follow @miambloom
John Horgan is Professor in the Global Studies Institute and Department of Psychology at Georgia State University. His most recent book is The Psychology of Terrorism (2nd Edition). Follow @drjohnhorgan
Charlie Winter is a senior research associate at Georgia State University’s Transcultural Conflict and Violence Initiative. Follow @charliewinter
[a] The Paris Principles and Guidelines published by UNICEF in 2007 states that “a child soldiers associated with an armed force or armed group refers to any person below 18 years of age.” “The Paris Principles: principles and guidelines on children associated with armed forces or armed groups,” UNICEF, February 2007, p. 7.
[b] In July 2015, the Syrian Observatory for Human Rights reported that eight children had died in Islamic State operations between January 2015 and July 2015 by ‘detonat[ing] themselves using vehicle bombs.’ However, according to our data, in the same time period, the number of children who died on suicide operations was actually 21. “Over 50 ISIS child soldiers killed in 2015,” NOW, July 15, 2015.
[c] Images were coding according to the following variables: date of publication, country in which death was reported, province in which death was reported, subject nationality, kunya, age category, role, target, target type, propaganda type, presence of weapons, presence of suicide belt, flag, presence of a Qur’an, microphone used, backdrop, clothing/attire, headgear, subject alone or accompanied, staged/unstaged, raised finger (to symbolize tawhid), facial expression based on Matsumoto and Hwang’s criteria, facial hair, and accompanying narrative.
[d] It should be noted that al-Ansari—a kunya denoting localness that harks back to the time of the Prophet Muhammad— does not just refer to Syrians and Iraqis. However, in the territorial context of Syria and Iraq, it does.
[e] We used age categories based on child development theory with a maximum age of 21 years old. However, no one over the age of 18 was included in this sample. Any “martyrs” whose age was known to be over 18 years old were inserted into the control group database.
[f] In at least five cases, we were able to verify the actual age of the bombers because of additional media reporting in the Western press, and in each case, the range had been coded correctly. Such cross comparison was only possible in a minority of instances as most military Islamic State operations do not receive media attention.
[g] Here, “mass casualty attacks against civilians” refers to an operation in which there are “three or more killings in a single incident,” as per the Federal Bureau of Investigation (FBI) definition. Investigative Assistance for Violent Crimes Act of 2012, 28 USC 530C(b)(1)(M)(i).
[h] Expressions were coded according to research conducted by Matsumoto and Hwang. Like with age categorization, expressions were subjected to intercoder reliability checks. David Matsumoto and Hyi Sung Hwang, “Reading facial expressions,” Psychological Science Agenda, American Psychological Association, May 2011.
[i] To a certain extent, greater variation is to be expected, given that there are more adults from which to choose.
[j] This came to light upon comparing the dataset with a recent infographic published by the official Islamic State propaganda outlet, A’maq Agency. See “Istishhadi Operations in Iraq and Syria,” A’maq Agency, February 2, 2016.
[k] Further research will be conducted as part of a future Minerva Research Initiative Dataverse. According to its website, the Minerva Initiative is a “Department of Defense (DOD)-sponsored, university-based social science initiative launched by the Secretary of Defense in 2008 focusing on areas of strategic importance to U.S. national security policy.”
 Lara Logan, “60 Minutes: Child suicide bombers,” CBS News, May 17, 2015; Mohammed al-Qalisi, “Mother mourns for son recruited as one of the Houthis’ child soldiers,” National, October 11, 2015; “Hezbollah Yushayyi’ Tiflan Qutila fi Al-Qalamun,” [Hezbollah Holds Funeral for Child Killed in Al-Qalamun], Al-Araby Al-Jadid, April 28, 2015.
 Lin Jenkins, “Isis video shows killing of Syrian troops at Palmyra amphitheatre,” Guardian, July 2, 2015; Lizzie Dearden, “Isis video shows young boy beheading Syrian soldier near ancient city of Palmyra,” Independent, July 17, 2015; Katie Zavadski, “ISIS’ new child executioner speaks English,” Daily Beast, February 4, 2016.
 “Implementation of qisas against the killer of a Muslim,” Homs Province Media Office, January 6, 2016; “Cubs of the caliphate camp,” Khurasan Province Media Office, January 28, 2016; “One of the Qur’an memorization competitions,” Anbar Province Media Office, February 9, 2016; “The da’wa caravan for cubs of the caliphate,” Nineveh Province Media Office, April 20, 2015.
 See, for example, John Horgan and Mia Bloom, “This is how the Islamic State manufactures child militants,” VICE, July 8, 2015; and John Horgan, Mia Bloom, Max Taylor, and Charlie Winter (forthcoming), “From Cub to Lion: A Community of Practice Perspective on Child Socialization into the Islamic State,” Studies in Conflict and Terrorism.
 Peter Singer, Children at War (Oakland, CA: University of California Press, 2005), p. 98; Michael Wessells, Child Soldiers: From Violence to Protection, (Cambridge: Harvard University Press, 2007), p. 71.
 “Impact of armed conflict on children,” UN General Assembly, August 26, 1996, A/51/306. See also “UN Security Council votes to name and shame governments, groups who abduct children,” Deutsche Welle, June 19, 2016.
 Michael Gross, The Ethics of Insurgency: A Critical Guide to Just Guerilla Warfare (Cambridge, MA: Cambridge University Press, 2015), p. 127.