ISIS vs. the U.S. government: A war of online video propaganda
First Monday

ISIS vs. the U.S. government: A war of online video propaganda by William H. Allendorfer and Susan C. Herring

The Islamic State of Iraq and Syria (ISIS) relies heavily on propaganda in the form of videos distributed over social media to recruit supporters and new members to its cause, including from the U.S. The U.S. government has countered with anti-ISIS propaganda videos; however, sources claim that the U.S. is losing the propaganda war. We evaluate that claim through a comparative multimodal content analysis of the ISIS video Flames of War and the videos posted in response on the U.S. Department of State’s (USDS) Think Again Turn Away YouTube channel. Our findings shed light on some of the reasons why the USDS anti-propaganda videos are less rhetorically effective than the ISIS videos, including a one-dimensional narrative, a stance that could appear inauthentic, and a lack of sensitivity to Islamic culture. In concluding, we advance recommendations that the USDS could follow to strengthen its online propaganda defense against ISIS, and extend the implications of our findings to other social media fronts where the ISIS vs. USDS propaganda war is being waged.


Conclusion and recommendations




The self-proclaimed Islamic State of Iraq and Syria (ISIS) [1] is a militant group that claims to have established an Islamic caliphate in areas it has occupied since June 2014 in Iraq and Syria. ISIS relies heavily on propaganda in the form of videos and images distributed over social media to generate support and recruit new members to its cause, including from the United States. Paul Neuman, director of the International Center for the Study of Radicalisation and Political Violence (ICSR), reports that “[a]ccording to ICSR’s latest estimate, the total [number of foreign militants in Iraq/Syria] now exceeds 20,000 — of which nearly a fifth were residents or nationals of Western European countries” (Neuman, 2015).

To counter this trend, the U.S. Department of State (USDS) posts anti-ISIS propaganda online as part of a project called Think Again Turn Away that aims to advance “some truths about terrorism” [2] and discourage young people who are drawn to the beliefs and actions of extremist organizations. The project has a dedicated YouTube site as well as accounts on Twitter and other social media, including accounts in Arabic, Urdu, Somali, and English, where videos and messages countering jihadist claims and arguments are actively posted.

However, the effectiveness of the USDS’s efforts at discouraging people from joining ISIS has been called into question. Some critics have claimed that the State Department’s campaign “provides jihadists with a stage to voice their arguments” and thus strengthens, rather than weakens, ISIS’s recruitment efforts (McLaughlin, 2014). Moreover, although ISIS propaganda videos are regularly banned from major social media sites such as YouTube, Facebook, and Twitter for their violent content, which includes beheadings, ISIS often uploads its videos to sites like Liveleaks, Pastebin, and the Internet Archive, where users are allowed to upload information anonymously (Collier, 2015). These videos are then shared through social media networks by ISIS supporters, so that even if they are deleted from major sites, they are still accessible to interested audience members. On Twitter alone, a recent study estimated that there were at least 46,000 ISIS-supporting accounts, including 500 to 2,000 accounts of highly active users who sent more than 50 tweets on average per day between September and December 2014 (Berger and Morgan, 2015). This situation has led Michael Steinbach, the head of the FBI’s counterterrorism division, to announce recently that the U.S. government is “losing the [propaganda] battle” against ISIS. In Steinbach’s words, “the sheer volume” of ISIS messaging online, particularly as it is dispersed through social media networks, “eclipses [U.S. government] effort” (Levine, 2015).

“Media is more than half the battle” is the motto of the USDS Center for Strategic Counterterrorism Communications (CSCC), the department that produces the Think Again Turn Away campaign. It is a paraphrase of a statement purportedly written by an al-Qaeda leader to Abu Musab al-Zarqawi, the leader of al-Qaeda in Iraq in 2005 (Cottee, 2015), and the USDS has adopted this perspective in its online war with ISIS through social media. This social media war represents a historical shift in the control of media discourse. Through the Internet, ISIS is able to actively recruit with its video propaganda anywhere in the world. Meanwhile, whereas the USDS controls traditional, centralized media outlets, its efforts have been unsuccessful in containing the ability of online sympathizers to self-select and disseminate information.

The use of terms such as “losing” and “battle” frames the social media activities of ISIS and the USDS as a war. Rather than a military conflict [3], however, it is a rhetorical war; a “war of ideas” (Johnson, 2015), in which the weapons are words and images wielded in the service of conflicting moral ideologies, and in which the essential goal is to persuade. This article critically examines the rhetorical strategies used in ISIS and USDS propaganda videos to evaluate how these groups present their messages to a shared target audience of American Muslims, as well as how the values reflected through the structure and content of the videos may be affecting the videos’ reception by their target audience. The article concludes by suggesting that the success of ISIS videos is due in part to their ability to properly identify and appeal to their virtual target community, and that the USDS videos will be unable to appeal to that community until they understand and incorporate its values.




Motion pictures have been used for war propaganda purposes since the early 1900s (Véray, 2010). Initially, a projector was required to screen films, so it was not easy to distribute this propaganda across enemy lines unless someone physically brought the films. Nowadays, the distribution of war propaganda videos is benefitting from ease of access to the Internet, and through it, to populations that would have been difficult to contact previously. However, aside from the fact that such video has been converted to a digital format, it remains very similar to its predecessor: the war propaganda film. In Herring’s (2013) tripartite Web content classification scheme, the genre of online propaganda videos can be considered “familiar [in that it has been] carried over into Web 2.0 environments with minimal differences” from its off-line antecedents [4].

We might expect the “familiar” genre of online war propaganda videos to utilize traditional rhetorical methods of the genre. Christie and Clark (2008) performed a content analysis of World War II-era Hollywood war propaganda films to see how closely they followed a 1942 list of enemy combatant characteristics identified by the U.S. government. They found that selected government-defined themes depicting Germans and Japanese during the war were systematically found in popular movies. As part of the present analysis, we apply the same checklist of enemy characteristics to a corpus of online propaganda videos from ISIS and the USDS to see how each group characterizes the other, and whether there are differences in their strategies of characterization.

Taking audience into account can help provide insight into the views of various groups involved in a conflict, as well as how those views could be interpreted. For example, Dimitrova, et al.’s (2005) content analysis of international news stories on the 2003 Iraq war found that different news frames were more prominent for audiences in different countries. In the present study, we analyze language use — English versus Arabic, translated in subtitles or not — as a clue to the identity of the intended audience. As Fairclough (1992) points out, “it is important to take account of how interpreters interpret texts if one is to properly assess (and not, for example, exaggerate) their political and ideological effectivity” [5]. In other words, it is possible to overestimate the effectiveness of online propaganda videos if one does not consider their relevance for the intended target audience, e.g., American Muslims.

While Dimitrova, et al. (2005) show the importance of differences in rhetorical framing in reaching different audiences, other studies reveal similarities between apparently disparate groups. For example, Hodsdon-Champeon’s (2010) analysis of discourse in online racially antagonistic debates found that people with opposing viewpoints employed similar rhetorical strategies, even when advancing different ideological positions. Thus, it is important to consider both differences and similarities when comparing the discourse of ideologically opposing groups.

Hodsdon-Champeon (2010) found that the participants in the debates she studied tended to use indirect intertextual references for ideas they felt were valid or true and direct quotations (which are easier to discredit) for ideas they considered invalid or false, independent of their ideologies about race. This observation is in keeping with van Dijk’s (2001) assertion that “a typical feature of manipulation is to communicate beliefs implicitly, that is, without actually asserting them, and with less chance that they will be challenged” [6]. Intertextuality is also manifested in social media through implicit cultural references, as well as through incorporating outside context by paraphrasing, quoting, retweeting, or linking to other texts elsewhere on the Web. The pro- and anti-ISIS videos analyzed in this study abound with references to Arabic and Islamic culture, and they reuse (and recontextualize) footage from other sources, including from each other. If this co-opted footage were simply analyzed for it surface characteristics, the broader implications of its recontextualization would be missed.




For the purpose of this study, we sought out comparable samples of video from ISIS and the USDS. The ISIS video clips are taken from an hour-long film titled Flames of War, a high production-value propaganda video that ISIS released in mid-September 2014 on various online sites such as Liveleaks. We chose to analyze this video for several reasons: 1) as the first full-length video released by ISIS, it lays out the group’s overall ideology; 2) it includes explicit references to, and characterization of, the United States; and, 3) the USDS posted a number of (shorter) videos over the following months that responded to many of the video’s claims. These response videos, of which eight were posted to the Think Again Turn Away Web site between September 2014 and February 2015 [7], constitute our USDS sample.

The USDS videos, considered collectively, are shorter at just over 18 minutes than the hour-long Flames of War, and the ISIS video includes content for which there is no analog in the USDS videos — notably, narration of battle sequences in Iraq and Syria in which ISIS was victorious. Therefore, we sampled a subset of video clips from Flames of War, excluding all but one of the battle segments. A total of six clips were sampled from the beginning, middle, and end of the film and assigned thematic titles, as shown in Table 1. The USDS videos with their original titles are also listed in Table 1. As the table shows, the total times for these two samples are approximately the same.


Table 1: Video samples.
Sample Time (minutes:seconds)Number of scenes
USDS videos 
Baghdadi’s ISIS schools 1:342
ISIS kills Muslims 2:504
ISIS leaves only tears and rubble 1:362
ISIS kills journalists 2:283
Young people fooled by ISIS/ISIL 2:552
ISIS and women: The real story 0:573
Baghdadi and the state of denial 1:572
ISIS inside the tent 4:106
ISIS video clips from “Flames of War” 
Flames of War Introduction 3:513
Fighting begins in Sham 4:434
The role of martyrs 2:372
Media war against the Islamic State 2:543
Benefits for believers & punishment for disbelievers 1:232
Flames of War closing 4:104


The clips were further broken down into ‘scenes,’ operationalized as continuous segments of video that address a coherent topic and/or that take place in a single setting. (That is, if there was a shift in topic and/or setting, we counted it as a new scene.) The two authors identified the video clips and scenes together. The ISIS sample has fewer scenes, as shown in Table 1, because scenes in the ISIS video tended to be of longer duration.

Finally, the language used in each scene was broken down into utterances. An utterance was operationalized as a sentence-like unit characterized by final intonation in speech; and in writing, by initial capitalization and final punctuation in English. Written Arabic lacks these indicators, but only one utterance written in Arabic was found in the data, so the issue of identifying others did not arise. The USDS videos contain 193 utterances, and the ISIS video samples contain 160 utterances.




Multimodal analyses of content themes and language use were performed on the scenes and utterances in each video, following the Web content analysis paradigm developed by Herring (2010). The unit of analysis for the thematic analysis was the scene. We coded thematically by 1) applying the checklist of enemy characteristics used by the U.S. government to portray enemies in films since 1942 (Christie and Clark, 2008); and, 2) using a grounded theory approach (Glaser and Strauss, 1967) to identify other emergent content characteristics, such as participant status and demographics. For the grounded theory analysis, we ended up coding the following categories: gender, age, occupation, and health status for each person portrayed (or group of persons, if no single individual was salient) in a scene.

We coded for language use in the videos at the level of the utterance. The language analysis focused on the use of Arabic and/or English in speech and subtitles. Most of the video clips include both speech and subtitles, although some USDS videos have no subtitles. We also coded for the source of each utterance, for example, whether it was ISIS, the USDS, or another source such as a news broadcast or television talk show. Also as part of the language analysis, we identify expressions in Arabic that were left untranslated in the videos and discuss their significance.

The coding was done manually by the first author, who speaks and reads Arabic, and checked by the second author; disagreements were resolved through discussion. The findings are presented using descriptive statistics, except for the analysis of untranslated Arabic terms, which is qualitative.




Enemy portrayals

Christie and Clark (2008) derived a list of 10 suggested ways to portray the enemy from the Government Information Manual for the Motion Picture Industry, section II, which was produced by the U.S. Office for War Information in 1942. This list is reproduced as Table 2.


Table 2: U.S. government (1942) recommendations for portrayal of enemies in film.
Enemy characteristics
1. Having no legitimate government order or alliance
2. Cruel
3. Cynical (insulting)
4. Deceitful, not trustworthy
5. Disregarding basic people’s rights
6. Dividing the U.S. (causing fear and distrust among Americans or allies)
7. Dominating by force, power
8. Lying, spreading rumor, false optimism and defeatism
9. Not invincible (can be defeated)
10. Sabotaging (e.g., spies who damage American industry/war machinery)


The USDS is still using the same enemy frames in video propaganda about ISIS, as Table 3 shows. The USDS videos especially characterize ISIS as cruel (#2), deceitful/untrustworthy (#4), disregarding people’s basic rights (#5), dominating by force (#7), lying/false optimism (#8), and not invincible (#9), with an increasing focus on conquerability and lying toward the end of our sampling period. In fact, the only items that are missing are dividing the U.S. (#6) and sabotaging (#10). Homeland terrorism by foreign groups is a sensitive issue in the U.S. post-9/11, and the USDS may not wish to emphasize that possibility lest it empower a group like ISIS to take that initiative.


1942 checklist of enemy characteristics, USDS videos
Table 3: 1942 checklist of enemy characteristics: USDS videos.


The following are (textual) examples of how ISIS is characterized in the USDS videos:


Excerpt 1: Deceitful; cruel
ISIS claims to be the defender of Muslims ...
But the only thing ISIS knows how to do is kill
(‘ISIS kills Muslims’)
Excerpt 2: False optimism; not invincible
ISIS’ claims can’t stand up to the truth.
Inside a tent, ISIS boasts of its plans to conquer the world ...
Openly inviting military action and making claims of invincibility ...
Meanwhile, outside the tent ...
Coalition airstrikes destroy ISIS armored personnel carrier
Coalition airstrikes destroy ISIS armed military truck [etc.]
Back inside the tent, ISIS continues its façade of invulnerability ...
(‘ISIS inside the tent’)

The ISIS videos demonstrate a different approach to presenting their war propaganda, however. As Table 4 shows, the main characteristics from the U.S. list that ISIS attributes to its enemies are that they lie (#8) and are untrustworthy or deceitful (#4).


1942 checklist of enemy characteristics, ISIS videos
Table 4: 1942 checklist of enemy characteristics: ISIS videos.


This attribution is repeated explicitly in Flames of War, as in the following spoken (transcribed) excerpt about the U.S. leadership:


Excerpt 3: Lying
[Narrator] You are with us or against us, Bush had said.
[Bush] “Either you are with us or you are with the terrorists”
[Narrator] They thought they had won.
[Bush] “Major combat operations in Iraq have ended. In the Battle of Iraq, the United States and our allies have prevailed.” *Applause*
[Narrator] They lied! The flames were only beginning to intensify. Obama then claimed it was over.
[Obama] “Iraq’s future will be in the hands of its people. America’s war in Iraq will be over.”
[Narrator] They lied!
(‘Flames of War Introduction’)

In addition, the ISIS video clips characterize enemies in religious terms, as ‘Christian,’ ‘Shia,’ or kufar (‘nonbeliever’), all of which are terms of disapprobation. All are infidels in the eyes of ISIS, which sees itself as waging jihad, or holy war, against the enemies of God. The USDS clips also characterize ISIS in religious terms, as ‘irreligious,’ ‘heretical,’ and ‘hostile to Muslims,’ to show that ISIS is hypocritical and to undermine the legitimacy of its claims to establishing an Islamic government that is favored by God. As these observations suggest, religion is a heavily contested site of discourse (cf., Foucault, 1991) in this war of propaganda.


It is also instructive to consider self-descriptive terms used in the videos. Using a grounded theory approach, we identified five main characteristics that ISIS attributes to itself, as shown in Table 5. ‘Favored by God’ is mentioned in the majority of scenes; other self-portrayals include ‘persistent,’ ‘honest,’ ‘believers,’ and ‘fearless.’


ISIS portrayals of itself
Table 5: ISIS portrayals of itself.


In contrast, the USDS videos say nothing about the USDS or the U.S. This is one of the most striking differences between the two video samples: While ISIS portrays its enemies in a negative light, the main purpose of Flames of War is to recruit supporters to the ISIS cause by portraying the group in a positive light — its glorious mission, military successes (evidence of the favor of God), and the (purported) good it has done for the people of Iraq and Syria. In contrast, the exclusive focus of the USDS videos is to portray ISIS negatively. This rhetorical asymmetry favors ISIS, in the sense that a positive or balanced message tends to be more persuasive than a purely negative one.

People depicted in the videos

The fact that the focus of both groups (whether positive or negative) is on ISIS means that the same kinds of people tend to be depicted in both video samples. Tables 68 summarize the results of the content analyses of people depicted in the videos. A total of 174 individuals or groups of people were coded in the USDS videos, and 215 individuals or groups of people were coded in the ISIS video clips.

The majority of people in both video samples are adult males, as shown in Table 6; however, the USDS videos showed somewhat more females and children than the ISIS videos [8].


Table 6: Gender and age of people depicted.
USDS157 (90%)17 (10%)24 (14%)139 (80%)11 (6%)174 (100%)
ISIS213 (99%)2 (1%)6 (3%)196 (91%)13 (6%)215 (100%)


In terms of occupation (Table 7), a majority of people depicted in both samples are soldiers, consistent with the status of ISIS as a military organization. Again, the USDS videos showed somewhat more diversity, in that they included more civilians and journalists. Conversely, the ISIS videos included more people in political and/or military roles [9].


Table 7: Occupations of people depicted.
USDS107 (61%)16 (9%)4 (2%)8 (5%)39 (22%)174 (99%)
ISIS159 (74%)2 (1%)24 (11%)6 (3%)24 (11%)215 (100%)


As Table 8 shows, although most people depicted in the videos were alive, a not insignificant number were dead. More dead persons, including non-combatants killed by ISIS, were depicted in the USDS videos, but the ISIS videos were more likely to show actual dead bodies. They also showed more wounded soldiers, many of whom were ISIS soldiers (mujahidin, or ‘holy warriors’) represented as returning to God in heaven, whereas the death of enemy (e.g., Iraqi or Syrian) soldiers was represented as righteous justice befalling kufar [10].


Table 8: Health status of people depicted.
USDS148 (85%)2 (1%)24 (14%)174 (100%)
ISIS189 (88%)9 (4%)17 (8%)215 (100%)


One reason for the general similarities in the depiction of people in the two video samples is that the USDS videos often incorporate footage from ISIS, as well as from religious leaders and news media in the Middle East. The USDS recontextualizes this footage to support its position that ISIS is a terrorist organization that is killing innocent people and cruelly abusing human rights. Thus it is important to take into account not only who is depicted, but what message their depiction is intended to convey.

Authorial source

In the USDS videos it is mostly the USDS perspective that is presented, in the sense that the text that appears on-screen appears to have the Think Again Turn Away campaign people as its author, although they never speak, and no actual author is identified. The only identifiable people who speak are in embedded video segments taken from other, typically Middle Eastern, sources. Thus the USDS videos recontextualize segments in which ISIS members speak, creating the effect of a dialogue or debate between ISIS and the USDS intended to convey the impression that the USDS is winning the debate. The USDS videos also incorporate testimony by Saudis who were formerly ISIS soldiers speaking against ISIS on a cleric talk show, and Middle East news agencies reporting on ISIS’s victims and military defeats. This breakdown is summarized in Table 9.


Table 9: Authorial source of USDS utterances.
 USDSISISMuslim clericsSaudi soldiersNews agenciesNon-profitUnknownAll
USDS76 (41%)36 (19%)9 (5%)29 (16%)16 (9%)8 (4%)11 (6%)185 (100%)


Conversely, the source of most utterances in the ISIS clips is ISIS itself, but unlike in the USDS videos, it is actual, identifiable ISIS members who speak. Also represented are Muslim clerics (some speaking for and some speaking against ISIS) and the text of the Quran, which is quoted to support ISIS’s methods and mission. The video additionally includes an extended scene that purportedly shows a captured Syrian soldier speaking submissively as he digs his own grave. U.S. voices are heard in only four percent of the utterances, e.g., when snippets of speeches by Presidents Bush and Obama are incorporated to accuse the U.S. of lying, as in excerpt 3 above. Thus there is some dialogic exchange between pro- and anti-ISIS perspectives, but not as much as in the USDS videos. Most of the perspectives presented are pro-ISIS. The distribution of the various authorial sources in the ISIS sample is summarized in Table 10.


Table 10: Authorial source of ISIS utterances.
 ISISU.S.Muslim clericsSaudi soldiersQuran/HadeethAll
ISIS92 (58%)6 (4%)10 (6%)36 (23%)15 (9%)159 (100%)


The fact that ISIS members speak for themselves directly in Flames of War lends the video an authenticity and sincerity that the USDS videos, with their disembodied, unidentified narrators, lack.

Intended audience

The language used to convey the message of the videos provides insight into their intended audiences. Perhaps unsurprisingly, given their countries of origin, the ISIS videos use more spoken Arabic than English, and the USDS videos use more English than Arabic, as shown in Table 11.


Table 11: Language of utterances (as a percentage of number of utterances).
 All ArabicMostly ArabicHalf/HalfMostly EnglishAll EnglishAll
USDS75 (39%)0 (0%)11 (6%)2 (1%)105 (54%)193 (100%)
ISIS79 (49%)0 (0%)2 (1%)31 (19%)48 (30%)160 (100%)


Moreover, ISIS provides subtitles with translations of English into Arabic and Arabic into English, such that a monolingual speaker of either language could understand everything in the videos (except possibly certain untranslated Arabic words, see below). However, the USDS only translates Arabic into English, and does not translate English into Arabic (see Table 12). This suggests that the target audience for the USDS videos is English speakers or bilingual Arabic speakers, whereas the target audience for the ISIS video is broader, including monolingual Arabic speakers. That is, even though the title of the video is in English and it was posted to English-language online forums, the ISIS video does double duty as a recruitment tool for both Middle Eastern Arabic audiences, which generate many ISIS recruits, and Muslims living in the U.S. and other English-speaking countries.


Table 12: Language of subtitles (as a ratio to number of utterances).
 All ArabicMostly ArabicHalf/HalfMostly EnglishAll EnglishAll
USDS0 (0%)0 (0%)1 (1%)14 (7%)56 (29%)71 (37%)
ISIS80 (50%)0 (0%)1 (1%)16 (10%)60 (38%)157 (99%)


These U.S. Muslims may have been born and raised in the U.S. by immigrant parents who passed on cultural concepts and values to their offspring, or they could be converts who are still developing an Islamic religious identity. Indeed, both the USDS and the ISIS videos assume that the viewer has knowledge of Arabic cultural and religious terminology. The Appendix lists Arabic terms from the video samples that were not translated into English, along with their English translations. We found 67 such tokens in the ISIS videos and 20 tokens in the USDS videos, representing 45 different words and expressions.

The untranslated Arabic terms are almost all religious. Some terms were left untranslated by both ISIS and the USDS; of these, some are common expressions that have entered English usage, such as jihad (‘holy war to maintain Islam’), mujahidin (‘people engaged in jihad’), and insha Allah (‘God willing’), while others express more advanced Islamic religious concepts, such as kufar (‘infidel, unbeliever’) and tauhid (lit. ‘oneness’; monotheism, belief in God’s singularity). The terms left untranslated by the USDS only are of both types; they include Allah akbar (‘God is the greatest’), fitna (e.g., ‘strife’ or ‘trial’), qibla (the direction Muslims face when praying), sahabah (the first Muslims), shahada (the declaration of belief in Islamic faith), shirk (‘attributing a partner to God’, the opposite of tauhid), and takfiri (‘accusing someone of being a kufar’). These words speak directly to Muslims because they are foundational to Islamic creed.

ISIS also leaves untranslated many words in the same religious semantic domain, and the ISIS video clips use many more Arabic terms without translation, despite translating more of the video’s content overall, as noted above. These terms are more likely to have political connotations (e.g., sharia ‘Islamic law’; khilafa ‘an Islamic form of government’) or be highly specialized terms for religious concepts such as tuaghit (‘anything worshipped other than God’) or names of specific religious and military groups, such as al-Muhajirin wal-Ansar (a brigade of pro-Islam foreign and native fighters in Syria). Thus the vocabulary in the ISIS video requires specialized knowledge. This could be a way of appealing to Muslims who are keeping themselves very informed regardless of their geographical location, or it could be because the target audience of the video includes Muslims living in the Middle East, who are assumed to be familiar with these terms.




According to the U.S. State Department, to date no extremist has publicly declared that he or she has reconsidered joining ISIS or any other extremist group because of the department’s messages (McLaughlin, 2014). Our findings suggest several rhetorical reasons for this lack of persuasive success.

First, there are differences in the content and presentation of videos by ISIS and the USDS, despite the fact that they address many common topics and entities. Scenes in the USDS videos depict ISIS using a wide array of enemy characteristics from the 1942 Office for War Information checklist. However, ISIS only uses a few of these characteristics to describe its Western enemies, especially, that the West lies. ISIS is more likely to attribute lack of religious faith to its enemies in the Middle East and elsewhere, whereas the USDS videos cast ISIS members as religious hypocrites and not true Muslims. These differences point to a difference in the underlying narrative of each group. The USDS narrative portrays ISIS as an enemy group, but portraying the U.S. as an enemy is only part of ISIS’s narrative. Its overarching message is that its establishment of an Islamic caliphate is favored by God, and that ISIS soldiers are holy warriors. ISIS’s message is affirmative and initiating, whereas the message in the USDS videos is negative and reactive, and this defensive position is inherently weaker. The coordinator of the USDS Think Again Turn Away program at the time these videos were posted, Alberto Fernandez, has acknowledged the difficulty faced by his department in countering the narrative produced by ISIS: “We don’t have a counter-narrative that speaks to that. What we have is half a message: ‘Don’t do this.’ But we lack the ‘do this instead’” (Cottee, 2015).

Second, the USDS videos suffer from a lack of authority and credibility relative to ISIS. Since at the time these videos were made the U.S. was not officially at war with ISIS, its narrative stance of ISIS-as-enemy-of-the-state could be questioned; ISIS adopts no such stance. Moreover, representatives of the U.S. government do not speak directly in the USDS videos, but rather condemnation of ISIS is made through embedded clips, taken from Middle Eastern media sources, of Muslim clerics and former soldiers speaking out against the group. In contrast, the ISIS videos feature narrators, including high-ranking members of ISIS, who speak directly to the audience. In these videos, ISIS appears more open and forthright than the USDS, whose disembodied voice from behind the scenes could appear to be hiding. Given ISIS’s repeated assertion that the West is deceptive, this is an unfortunate impression for the USDS to give off.

The use of English, spoken and in subtitles, as well as Arabic religious terms without translation in both the ISIS and USDS videos suggests that the common target audience of both is English speakers with some knowledge of Arabic culture and religious terminology, hence, probably, U.S. Muslims, although the ISIS videos appear to target monolingual Muslims in the Middle East, as well. It is problematic, therefore, that the USDS videos display a lack of sensitivity to Islamic cultural values. Although not specifically addressed in our analysis, one USDS video includes an extended scene of supposed ISIS members discussing the market for slaves among conquered people. The USDS video claims that ISIS members are barbaric slaveholders, a portrayal that is designed to make ISIS look bad in terms of American values. But it may also cause problems for Muslims who see this portrayal as characterizing Muslims in general, since the issue of slave ownership is discussed in the Quran, and some interpretations allow for such slave ownership. In another USDS video, the USDS accuses ISIS of violating human rights by requiring women to cover themselves and by segregating them from men, but many Muslims believe that these practices are appropriate and morally correct, and they could reasonably take offense at the USDS’s characterization, even if they do not support ISIS.

Another example of cultural insensitivity involves the use of background music in the videos. Instrumental music is a feature of every USDS video, but there is no instrumentation in any ISIS video. Instead, the ISIS videos use anasheed (sing. nasheed) or lyrical poems, sung a capella, which appeal to religious values (al-Awlaki, 2009). In contrast, the USDS videos all feature instrumentation, which is a problem for adherents to Islamic schools of thought that advise that musical instruments are forbidden. Thus, while the U.S. videos aim to engage the emotions of their viewers through the use of background music, which superficially resembles that in the ISIS videos, this difference of religious acceptability could undermine the appeal of the USDS videos to some Muslim viewers.

In short, unlike the ideologically opposed groups studied by Hodsdon-Champeon (2010), ISIS and the USDS employ divergent rhetorical tactics in their online propaganda videos. Our findings shed light on some of the reasons why the USDS anti-propaganda videos are less rhetorically effective than the ISIS videos, including a one-dimensional narrative, a stance that could appear inauthentic, and a lack of sensitivity to Islamic culture. These qualities could cause Muslim viewers, the ostensible target audience, to find the USDS videos unpersuasive, and possibly even offensive.



Conclusion and recommendations

The U.S. government believes that it is losing the online propaganda war to ISIS for reasons of quantity: “[T]he sheer volume” of ISIS messaging online, particularly through social media, “eclipses [U.S. government] effort” (Levine, 2015). The findings of our study suggest that in addition to quantity considerations, the rhetorical quality of USDS online counter-propaganda contributes to this outcome. The findings further suggest that the success of ISIS is due in part to its ability to properly identify and appeal to its virtual target community, and that the USDS will be unable to appeal to that community as effectively unless it understands and engages with its values, while speaking from a position of authenticity and credibility.

It would be most effective for the U.S. to establish its own positive narrative with regard to ISIS rather than simply focusing on its counter-narrative. This positive narrative does not need to address the same issues as the ISIS narrative; instead, it could focus on the reasons why American Muslims do not need to join a militant group in the Middle East. In order to promote this narrative, videos could focus on positive aspects of modern Muslim life in the United States and other countries of the world. For instance, these videos could engage with such issues as religious tolerance and acceptance, with an emphasis on the importance of these values in Islam. In this way, the videos could appeal to values, like the viewers’ responsibility to take care of their own families and communities here in America, in a positive way rather than simply condemning the actions of ISIS. In addition, in order to establish authenticity, the U.S. narrative could benefit from allowing American Muslim voices to speak directly to the target audience rather than using foreign clerics or impersonal on-screen text to deliver the USDS narrative.

At the same time, the USDS counter-narrative could be enhanced through a focus on criticism of ISIS without co-opting the materials of this group. As the language of both the USDS and ISIS videos demonstrates, the intended audience of these videos is Muslims with at least a basic understanding of Islamic creed. In order to appeal to the values of this audience, the counter-narrative could incorporate authentic Islamic materials (e.g., the Quran and Hadith [11], both of which are present in ISIS videos and absent from USDS videos) to enhance the legitimacy of its claims for viewers who are versed in this Islamic discourse. In addition, the question of the legitimacy of ISIS’ claims of establishing a caliphate could be addressed by providing historical education on the many disagreements over rightful leadership that have existed since the creation of the first Islamic caliphate. Thus, it is our belief that the establishment of both an authentic, positive narrative that promotes constructive portrayals of American Muslims and an informed counter-narrative grounded in Islamic texts and history could better position the USDS to offer a persuasive alternative to ISIS recruitment propaganda.

Meanwhile, online videos are one of the main ways ISIS currently recruits new members and supporters. In addition to longer, Hollywood-style videos like Flames of War, short videos and animated GIFs are posted to sites like Twitter and Tumblr, where the USDS also maintains active antiterrorist accounts. It could be fruitful to analyze the interactions there using the kinds of methods employed in this study to see whether the present findings extend to other social media contexts, especially those contexts where the more rapid exchange of short postings would likely reveal additional rhetorical dynamics. The data from these sites, as well, could be mined for insights to inform the creation of more effective antiterrorist propaganda. End of article


About the authors

William H. Allendorfer is a Ph.D. student in Second Language Studies at Indiana University Bloomington. His research focuses on second language acquisition and use in online discourse. His other interests include computer-mediated communication, Arabic, language pedagogy, and pragmatics.
E-mail: wallendo [at] indiana [dot] edu

Susan C. Herring is Professor of Information Science and Linguistics and Director of the Center for Computer-Mediated Communication at Indiana University Bloomington. Her research focuses on structural, pragmatic, interactional, and social language use in digital communication. Her recent interests include multimodal CMC and telepresence robotics. She is a past editor of the Journal of Computer-Mediated Communication and currently edits the online journal Language@Internet.
E-mail: herring [at] indiana [dot] edu



1. The acronym ISIL, where the L refers to the Levant, is also used. The group is also known in Arabic as ad-Dawlah al-Islāmiyah fī ’l-ʿIrāq wa-sh-Shām, leading to the acronym Da’ish or Daesh;, retrieved 24 November 2015.

2., retrieved 28 July 2015.

3. There is also a military component to the conflict between ISIS and the U.S., but this paper is concerned only with its online, discursive manifestations.

4. Herring, 2013, p. 1.

5. Fairclough, 1992, p. 292.

6. Van Dijk, 2001, p. 358.

7. After February 2015, posting of USDS videos to the site dropped off.

8. Persons between the ages of 0 and 15 were coded as children; seniors were considered to be older than 55; and all others were coded as adults.

9. Persons wearing uniforms, who were armed, and/or who were engaged in military acts were coded as soldiers. Abu Bakr al-Baghdadi, the leader of ISIS who has been proclaimed by his supporters to be the Caliph, or the political and religious leader of the Muslim community, was coded as political/military rather than religious because he is portrayed in this role in both the USDS and ISIS videos. Religious figures include clerics and people on talk shows discussing religious/moral issues. Journalists were identified on the basis of their activities and/or because they were specifically named as news agency employees. Unarmed people in civilian clothing were coded as civilians.

10. We coded persons as dead if they were deceased but living pictures of them were shown, as was often the case in the USDS videos. We coded persons as wounded if they had visible injuries or bloody clothes.

11. Hadith (lit. ‘report, narrative’) is a collection of reported sayings and actions of the prophet Muhammad (S.A.W).



Anwar al-Awlaki, 2009. “44 ways to support jihad,” Victorious Media, pp. 1–21, at, accessed 31 July 2015.

J.M. Berger and Jonathon Morgan, 2015. “The ISIS Twitter census: Defining and describing the population of ISIS supporters on Twitter,” Brookings Project on U.S. Relations with the Islamic World, Analysis Paper, number 20, at, accessed 7 December 2015.

Thomas Christie and Andrew Clark, 2008. “Framing two enemies in mass media: A content analysis of U.S. government influence in American film during World War II,” American Journalism, volume 25, number 1, pp. 55–72.

Kevin Collier, 2015. “Fearing ISIS, India censors dozens of popular websites,” Daily Dot (1 January), at, accessed 30 January 2015.

Simon Cottee, 2015. “Why it’s so hard to stop ISIS propaganda,” Atlantic (2 March), at, accessed 31 July 2015.

Daniela Dimitrova, Lynda Kaid, Andrew Williams, and Kaye Trammell, 2005. “War on the Web: The immediate news framing of Gulf War II,” International Journal of Press/Politics, volume 10, number 1, pp. 22–44.
doi:, accessed 7 December 2015.

Norman Fairclough, 1992. “Intertextuality in critical discourse analysis,” Linguistics and Education, volume 4, numbers 3–4, pp. 269–293.

Michel Foucault, 1991. “Politics and the study of discourse,” In: Graham Burchell, Colin Gordon, and Peter Miller (editors). The Foucault effect: Studies in governmentality. Chicago: University of Chicago Press, pp. 53–72.

Barney Glaser and Anselm Strauss, 1967. The discovery of grounded theory: Strategies for qualitative research. Chicago: Aldine.

Susan C. Herring, 2013. “Discourse in Web 2.0: Familiar, reconfigured, and emergent,” In: Deborah Tannen and Anna Marie Trester (editors). Discourse 2.0: Language and new media. Washington, D.C.: Georgetown University Press, pp. 1–25.

Susan C. Herring, 2010. “Web content analysis: Expanding the paradigm,” In: Jeremy Hunsinger, Lisbeth Klastrup, and Matthew Allen (editors). International handbook of Internet research. Berlin: Springer, pp. 233–249.
doi:, accessed 7 December 2015.

Connie Hodsdon-Champeon, 2010. “Conversations within conversations: Intertextuality in racially antagonistic online discourse,” Language@Internet, volume 7, article 10, at, accessed 28 July 2015.

Natalie Johnson, 2015. “How ISIS is waging a ‘war of ideas’ through social media,” Daily Signal (8 June), at, accessed 28 July 2015.

Mike Levine, 2015. “FBI: ‘We are losing the battle’ to stop ISIS radicalization online,” ABC News (26 February), at online/story?id=29241652, accessed 28 July 2015.

Jenna McLaughlin, 2014. “Why the U.S. government’s counter-terrorism tweeters are finding it tough to fight ISIS online,” Mother Jones (19 September), at, accessed 28 July 2015.

Peter Neuman, 2015. “Foreign fighter total in Syria/Iraq now exceeds 20,000; surpasses Afghanistan conflict in the 1980s,” International Centre for the Study of Radicalisation and Political Violence (26 January), at, accessed 1 August 2015.

Teun van Dijk, 2001. “Critical discourse analysis,” In: Deborah Schiffrin, Deborah Tannen, and Heidi Hamilton (editors). Handbook of discourse analysis. Malden, Mass.: Blackwell, pp. 352–371.

Laurent Véray, 2010. “1914–1918, the first media war of the twentieth century: The example of French newsreels,” Film History, volume 22, number 4, pp. 408–425.
doi:, accessed 7 December 2015.


Appendix: Untranslated Arabic expressions in the videos


Arabic expressionEnglish translationUSDSISIS
al-ḥamdulilah‘All praise to God’02
Allah akbar‘God is the greatest’40
amir al-muminin‘Prince of the Believers’01
al-Muhajirin wal-AnsarMuhajirin Brigade of foreign and native fighters in Syria01
AqidaLit. ‘to tie/knot’; any religious belief system or creed01
DawaLit. ‘making an invitation’; proselytizing Islam02
FitnaMeanings include ‘temptation’ and ‘trial’30
ghurbaLit. ‘foreigners’; foreign ISIS fighters01
ḥadudLit. ‘limits’; crimes forbidden in the Quran01
ḥijriYear-numbering in the Islamic calendar01
ḥukumLit. ‘judgment’; Quranic punishments for crimes01
insha Allah‘God willing’11
istishahadin‘The ones seeking the death of a martyr’02
jihadLit. ‘to strive/struggle’; Islamic term for maintaining religion23
khawarijLit. ‘outsiders’; unorthodox Islamic extremists03
khilafaIslamic form of government02
kufar‘Disbeliever, Infidel’16
muḥajirinLit. ‘Emigrants’; refers to the Muhajirin Brigade01
mujahidinPlural of mujahid13
mujahid‘A person engaged in jihad [a struggle to maintain Islam]’01
munafiqinLit. ‘hypocrites’; a disbeliever who outwardly practices01
nifaqLit. ‘hypocrisy’; feigning love for Islam to conceal disbelief01
NusayriyyaThe beliefs of a Shi’ite sect called Nusayri02
NusayriA Shi’ite sect that maintains unorthodox Islamic beliefs03
qiblaLit. ‘direction’; the direction Muslims face when praying10
rab al-alamin‘Lord of the worlds’01
rabi al-awwalThe third month in the Islamic calendar02
rasul Allah‘Messenger of God’02
rawafidLit. ‘refusers’; Shi’ites, who refused to accept Sunni rule01
SafawiLit. ‘of the Safavid Dynasty’; derogatory for Shi’a01
ṣaḥabahLit. ‘companions’; the first Muslims20
ṣala Allah alayhi wa salam‘May God honor him and grant him peace’03
shahadahLit. ‘Testimony’; the declaration of belief in Islamic faith10
ShamTerm used for the geographical area known as the Levant15
shariaIslamic legal system derived from the Quran and Hadeeth01
shirk‘Attributing a partner to God’; the opposite of Tauheed10
shuhadaPlural of shahid [‘martyr’]01
takfiriLit. ‘accusing of being a kufar [‘disbeliever’]’10
ṭauḥidLit. ‘oneness’; monotheism, belief in God’s singularity11
ṭuaghitPlural of taghut [‘anything worshipped other than God’]01
ummaLit. ‘nation’; the collective community of Muslims04
wallahi‘I swear to God’02
zakatLit. ‘that which purifies’; an obligatory tax in Islam01
N=45 2067



Editorial history

Received 2 December 2015; accepted 10 December 2015.

Creative Commons License
“ISIS vs. the U.S. government: A war of online video propaganda” by William H. Allendorfer and Susan C. Herring is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

ISIS vs. the U.S. government: A war of online video propaganda
by William H. Allendorfer and Susan C. Herring.
First Monday, Volume 20, Number 12 - 7 December 2015

A Great Cities Initiative of the University of Illinois at Chicago University Library.

© First Monday, 1995-2017. ISSN 1396-0466.