The Islamic State: constructing identity and articulating propagandistic discourse through intertextuality and recontextualisation

This paper examines the discourse of the Islamic State (IS) as a terrorist group claiming Islamic authenticity and the right to establish khilafa . The analysis of multimodal texts produced and posted online by this group shows how their discourse uses intertextuality and recontextualisation to authenticate their political practice in religious terms, and how they employ strategies of horizontal and vertical propaganda to maintain control and hegemony, and instil fear and terror in the lives of civilians and belligerents alike. The analysis shows that in order to survive, their discourse shifts positions in response to needs, desires and antagonistic social realities. The discussion of these shifting positions reveals the political foundation of their alleged religious discourse and shows how this will change in the course of history.


Introduction
This paper aims to study the discourse of one of the terrorist groups active in the Middle East region, namely the Islamic State (IS). The analysis of a large amount of data produced and posted online by this group reveals the mechanisms at play in the articulatory practice of their discourse. As this discourse claims to be religious and monopolises Islamic authenticity, the analysis focuses on the political dimension often denied or claimed by IS to be political theology. It reveals through the two categories of intertextuality and recontextualisation how positions and frames shift in response to historical realities and the antagonistic political relations that this group creates through strict boundaries and the practice of terror.
As IS multiplies frontlines, targets and enemies as per the contingency, its discourse modifies positions, concepts, frames and strategies; for instance, rapidly changing historical conditions compelled IS to change positions on certain issues like hijrah, the apocalypse, slavery and terror tactics. Therefore, the analysis tries first to understand the political identity of these groups from their perspective (their self-representation) and then proceeds to the critical assessment of their self-presentation (identity), their relation to others (difference) and their claim to authenticity and universality through the logic of what Hegel called 'oppositional determination' (contradiction) (Zizek, 1998, p. 131). It can be argued that the discourse of IS founds its claims of legitimacy and authenticity on religious terms to serve political ends. The selective use of religious intertext and its recontextualisation in terms of emerging contingencies and antagonisms is an inherent mechanism of the discourse of IS that is used to build up coherence and regularities between its dispersed discursive positions. The political identity of IS changes in terms of their discursive positions that constantly shift in response to new historical realities. The contradictions that emerge from these shifts are often concealed or reconciled through the processes of intertextual recontextualisation and propagandistic manipulation, or deferral in mythic terms in the face of impossibility.
In order to understand the contradictions in the political identity of IS, emphasis is put on the terrorists' contested claim of "authenticity" as constitutive of their religious and political legitimacy, thus imposing boundaries and engaging themselves in perpetual antagonism with the different Other, including other different Muslims, as a mode of identification ('Us' constituting itself as a totality versus 'Them' preventing the ambition of 'Us' to 'constitute full presence', Laclau and Moufe, 1985, p. 127). IS's claim of representing Islamic authenticity should be comprehended as an attempt at political hegemony by appealing to Muslims' sense of imagined community, and mobilising them to join the fight against the Other as a perpetual enemy of this community.
IS can, thus, be essentially classified as an extreme form of pan-Islamism and global jihad; however, other features can be added as secondary. Hegghammer (2010) has established a useful typology to distinguish the different forms of Islamist activism. He classifies Islamists into five types: 'socio-revolutionaries' fighting 'for state power against a Muslim regime perceived as illegitimate', 'violent irredentists' struggling 'for a specific territory against a local non-Muslim occupier', 'violent pan-Islamists', who are ummah-oriented, fighting 'to defend the entire Muslim nation and its territories from non-Muslim aggression' (he distinguishes between 'classical jihadists' who fight 'conventionally in local conflict zones' and 'global jihadists' who fight 'the West with all means in all places'), 'vigilantists' striving 'to correct the moral behaviour of fellow Muslims' and, finally, 'violent sectarians' who 'kill to intimidate and marginalise the competing sect (Sunni or Shiite)' (p. 6).
This typology may be used to classify IS as a global jihadist organisation alongside Al-Qaeda; however, some basic differences exist between the two organisations. Atouane (2015) presents some of these differences in terms of three variables: inception, ideology and priorities. He argues that IS was created as a state with (shifting) frontiers and sovereignty (part of Iraq and Syria), while Al-Qaeda is an organisation hosted by a sovereign country (Afghanistan). IS took advantage of the collapse of the state in Iraq and Syria and the weakening of the central authorities, the Western military and strategic intervention in Iraq and Syria, sectarian polarisation (Sunnis and Shiites), and the outbreak of the Arab Spring and its aftermath, whereas Al-Qaeda was the heir of the Islamic struggle against the Soviet occupation of Afghanistan that has now turned its hostility towards the West (pp. 9-10).
In terms of ideology, both organisations share certain concepts and frames like the sovereignty of God, loyalty and disavowal, considering Muslim rulers as illegitimate and the world order as an alliance of disbelievers (taghut), and proselytising as a religious practice necessarily leading to jihad as a factor of change. Nevertheless, IS known more for their belief in radical social change, their excessive use of cruelty as a strategy to achieve said change, and their undifferentiated hostile stance towards Muslim rulers and Western alliances (Atouane, 2015, p. 10).
Another scholar, Hafez (2020), highlights other differences between Al-Qaeda and IS opposing IS's puritanism to Al-Qaeda's populism. Hafez argues that '[t]he Islamic State has embraced puritanical extremism as its defining character. It insists that it constitutes the "Victorious Sect" that uncompromisingly adheres to Salafi orthodoxy in doctrine and practice ' (p. 40). He compares IS puritanism to the opportunistic populism of Islamist movements which, according to him, 'supposedly tolerate public blasphemy to avoid alienating supporters; delay establishing sharia-based states and instead choose to work within the confines of civil democratic states; and make alliances with secular factions or apostate governments in the name of realpolitik' (p. 40).
Finally, Hafez (2020) points out that IS accuses Al-Qaeda of populist Islamism as the latter 'seeks to win the hearts and minds of Muslims rather than mold them into believers through the strict application of "Islamic" law' (p. 40). Other scholars addressing the same issue argue that IS came out of the womb of Al-Qaeda. Baroudi (2020) contends that IS 'a hybrid seed that emerged from the cross-fertilization of elements of Salafi-Wahhabism with elements of the thought of the Muslim Brotherhood' (p. 230). He traces the roots of Salafi jihadism, quoting Lynch (2010), and pointing out that the origin of this new orientation were the ideas of Qutb who 'politicized Wahhabism' (p. 230).
IS another phase in the development of pan-Islamism into global jihad, and their claim of monopolising Islamic authenticity is a political attempt at gaining religious legitimacy and mobilising support and adherence to a cause that is basically political in nature. The study of their discourse shows contradictions highlighted by changing historical contingencies and emergent cases of political antagonism. The contradictions appear once the shifting discursive positions produce cracks in IS's system of justification and legitimation, making it impossible for their absolutist discursive claims to stand the test of those changing historical contingencies and antagonisms.

Literature review
Scholars of Islamic tradition and political Islam have debated this topic from different perspectives, highlighting the mechanisms used by different Islamic groups to gain legitimacy and authenticity. Lahoud (2005) argues that Islamism 'is a selective and literal approach to the foundation texts, Quran and Hadith, that is, selecting Quranic verses and Hadith reports without due sensitivity to the context or alternative traditional interpretations, but whose literal sense is conducive to their political objectives' (p. 2). The Islamists, according to Lahoud (2005), argue that the early Muslim community 'represents a fusion between the religious and political sphere ' (p. 15). She also claims that while different views and interpretations of the foundation texts exist, the Islamist current 'enjoys a greater "authenticity" among the general public' (p. 13), and that 'this seemingly faithful adherence to the teachings of the foundation texts has enabled it virtually to monopolize Islamic "authenticity" in the minds of many Muslims and non-Muslims' (p. 13). This monopolisation of "authenticity" is put into question by other currents (apologists and intellectuals), as each uses the same process of selective emphasis on certain Quranic verses and the Hadith, or other books of tradition, to support their political goals. The notion of "authenticity" becomes, thus, a negative void that can be filled with positive content from the Islamic tradition.
Other Muslim scholars deny that the political ideology of Islamists represents Islamic religious authenticity. Fatah (2008) argues that Muslims should not seek to establish an 'Islamic State' but a 'State of Islam', as the first 'requires a theocracy, the other a state of spirituality' (p. xii). Accordingly, he tries to question the notion of jihad on which the dream of an Islamic State is founded and which the gurus of the Islamist doctrine, like Maudoudi and Qutb, consider as obligatory. He contends that Islamists 'would like to see both Muslim and non-Muslim collectively submit to their fascist ideology of hate and supremacy where instead of life, death is to be celebrated' (p. 272). He also argues that 'like slavery, the time of jihad has gone' (p. 279).
From a different perspective, the conventional Western approach to terrorism focuses on shortterm assessments of terrorist threats in terms of a problem-solving dynamic (Gunning, 2007, p. 366), which has proven to be inefficient and has a Western state bias. This conventional knowledge can be 'weaponized to suit state interests' (Stump and Dixit, 2013, p. 55). Conversely, the postcolonial approach contends that terrorism is a social construct whose threats should be placed within a wider social and historical context of political violence, which 'originated in forms of control of colonized peoples' (Stump and Dixit, 2013, p. 55).
These two approaches view the social and political organisation of terrorist groups in terms of two reductive analytical models. On the one hand, the postcolonial approach does not account for the identity construction of the terrorists, which is not limited by national territory and specific national culture, as opposed to the coloniser's culture. On the other hand, the Western approach presupposes that these groups are socially and politically constructed in the same way Western societies are, where values of individuality, freedom, equality and deliberative democracy prevail. It can be argued, however, that the absence of these values is not conceived as a drawback by these groups and is actually what makes them pursue terrorism in the first place. The worldview of these groups and the political identity of their organisations draw on religious principles where authoritarianism, hierarchy and obedience through allegiance are viewed as values of political/religious organisation and conduct. Even anarchism (what jihadist scholars call 'the state of savagery') is viewed as a favourable condition for the establishment of the Islamic State.
Taking note of the previous views, which highlight the political dimension of terrorist organisations, the critical assessment of the phenomenon of terrorism is conducted in light of the power structures and relations within which they are actually operational. Given that terrorism is viewed 'primarily as a combination of violence and communication' (Schmid, 2011, p. 2), and given the antagonistic tension and boundaries created between terrorists and their enemies, the study of the discourse of terrorists in an overdetermined field of discursivity aims at highlighting the regularities in their dispersed subject positions.
These are the ways in which antagonism prevents the terrorist discourse from tying a final suture between itself as a totality and the workings of articulation and hegemony at play within their discourse in response to other discourses and in terms of emergent contingencies (Laclau and Moufe, 1985). Laclau and Moufe (1985) define articulation as 'any practice establishing a relation among elements such that their identity is modified as a result of the articulatory practice' (p. 105). The result of this articulatory practice is discourse which is actually constructed by modifying the elements (floating signifiers) into moments that are combined to form nodal points that fix partially the meaning of the social. The practice of articulation is solely felt when contingency and antagonism are taken into account, and this means, in the presence of these two factors, that the terrorist discourse merely succeeds at partial fixations of meaning, with totality effects only.
Describing the communication practices of terrorism as 'propagandistic' (Schmid, 2011, p. 2) means that the terrorist discourse uses strategies that deliberately and systematically attempt 'to shape perceptions, manipulate cognitions and direct behavior to achieve a response that furthers the desired intent of the propagandist' (Jowett and O'Donnell, 2015, p. 7). It is our purpose in this paper to study the ways in which the terrorist discourse tries to set intertextual links with the 'broader literary and cultural context in which it is imbedded' (Kristeva, 1980, p. 66) and to propagate itself vertically (leaders/public) and horizontally (within a social network) through intertextuality and recontextualisation (Bauman and Briggs, 1990;Wodak, 2012;Oddo, 2018). Like propaganda, intertextuality can also be both horizontal and vertical. Hodge (2015) elaborates on this distinction, borrowing from Silverstein (2005), by calling the former token-sourced intertextuality, 'where a token of speech (i.e. text) from a previous setting is placed into a new setting' (Hodge, 2015, p. 44), and the latter type-sourced intertextuality, where a type or genre connects 'the language used across different discursive settings ' (p. 45).
What is of interest to us in the study of intertextuality is its relation to discourse and power, that is, the way intertextuality makes the discourse of terrorists authoritative and hegemonic in terms of religion and even politics. Ritualised intertextuality constitutes a 'chain of authentication' (Hodge, 2015, p. 55) that elevates claims, knowledge and beliefs to the status of truth, making them gain traction with the public. What is more relevant here is the use of intertextuality as one mechanism, among many, of recontextualisation and propaganda. Fragments of discourse can be entextualised into texts and decontextualised (detached from situational context), and recontextualised in another context (Bauman and Briggs, 1990). These texts are made memorable for propagandistic ends and serve to consolidate power and its underlying claims (Bauman and Briggs, 1990;Oddo, 2018). Fairclough (2012) views recontextualisation as the appropriation of a discourse 'into the strategies pursued by particular groups of social agents within the recontextualizing field' (p. 11).
A final point is to consider the contexts of the terrorists' actions in terms of both the theory of political opportunity structure and what is called the theory of 'the management of savagery'. Terrorist groups are political opportunists who tend to take advantage of any weakening of states and the disorder resulting from political instability (civil wars, weak central authority, sectarian struggles etc.). Naji (2004) defines the management of savagery as the management of a state of savage disorder and divides the implementation of this theory into three stages: vexation and exhaustion (guerrilla warfare against an enemy until disorder is achieved), administration of the resulting savage disorder, and empowerment (the establishment of the Islamic State) (p. 15).
The practices of articulation, intertextuality and recontextualisation are at the heart of the discourse of IS and are realised through a process of reappropriation of floating signifiers (elements that are not articulated, like the apocalypse, slavery and hijrah to Sham -Syria) that flow intertextually in the discursive field and are recontextualised and operationalised to suit and serve political, social and psychological goals. Such modifications in IS positions have manifest impacts on their identity, such that the resulting contradictions make them search for an escape with a repetitive affirmation (their motto, "remaining till the end of times", is an obsessive idea that was entextualised and made memorable) into the future through a deferral of any stable achievement. The end that IS internalises is not the end of times, but its own end that takes the form, to use Zizek's terminology, of a 'point of impossibility ' (2008, p. xv), whereby IS endlessly struggling to postpone its own collapse.
It is within this theoretical and contextual framework that we ask the research questions that this paper will try to answer through the analysis of the collected multimodal texts posted by IS media.

Research questions
This paper examines two main questions. The first is related to the articulation constitutive of identity and discourse in terms of contingency and antagonism. The second scrutinises the probable impact of the practice of excessive savagery on the rhetoric of propaganda and its effects: -How does IS construct its discourse and identity as a comprehensive totality and around which concepts and frames? To what extent do antagonism and contingency produce cracks in such a totality through contradictions, deferral and escapism?
-How does propaganda address audiences and produce effects, and to what extent?
Method and materials A significant amount of data was collected from IS official magazines: Dabiq (in English, 13 issues), Rumiyah (in English, 15 issues) and Annabaa (in Arabic, 207 issues), covering a seven-year period (from 2014 to 2020). A non-probability sampling of the materials collected was adopted, based on convenience for the goals of our research, namely locating basic concepts and frames, and propagandistic strategies. In order to accumulate more evidence and to increase credibility, other materials were collected from the IS online Hima library and from supporters' publications, like Takwa and Albattar. The total number of multimodal texts analysed was 1,500 and consists of still pictures with text, infographics, video cover images, documents, leaflets and postcards.
This triangulation was also applied in terms of approach, as the data collected was analysed first in search of indicators, coded categories and concepts, followed by theories of discourse analysis: discourse theory (Laclau and Moufe, 1985), critical discourse analysis (Wodak, 2012;Dijk, 2012;Fairclough, 1992Fairclough, , 2012, and multimodal discourse analysis as conceived and developed by scholars within the framework of social semiotics (Kress and Leeuwen, 2006). These approaches are complementary, as they share a common concern with how power is constructed and maintained through the use of language and other semiotic systems. In this paper, we define discourse as a social practice that constitutes social reality and is constituted by it, as a way of representing social practices and as a form of knowledge. The multimodal approach is more concerned with all the semiotic modes of a text, 'framed as one field, as one domain. Jointly they are treated as one connected cultural resource for (representation as) meaning-making by members of a social group at a particular moment' (Kress, 2012, p. 38).
As different modes have different affordances, they give the same meaning different realisations, which all partially contribute to the meaning of the multimodal text. The interest of this paper lies in comparing and contrasting these realisations of meaning in different modes and highlighting the ways intertextuality and recontextualisations of competing discourses and concepts flowing in the discursive field are conducted. The practical analytical model used in this paper draws largely on the semantic-functional categories that Kress and Leeuwen (2006) mapped from Halliday's theory of transitivity (participants, process and circumstances) and its meta-functions (ideational, interactional and textual).
Other researchers (Bolt, 2012;Brachman, Boudali and Ostovar, 2006;Ostovar, 2017) have studied the visual motifs of jihadists using a content analysis approach. Ostovar organised the images of jihadists in two types of networks: casual and deliberate, in an attempt 'to explain intentionality behind the adoption of certain symbols and imagery' (p. 85). He contends that '[i]n a casual visual network, visual imagery can be shared by both affiliated and unaffiliated groups. Casual networks are unsuccessful at articulating a unified movement because they are not intended to do so. Imagery used within a deliberate network, however, is intended to highlight affiliation between groups or to express solidarity within a movement' (p. 85).
This classification is in line with our classification of propagandistic imagery as being either horizontal or vertical. Brachman, Boudali and Ostovar (2006) stress the importance of studying the visual motifs of jihadis claiming that they 'accomplish several objectives for jihadi propagandists. First, they create a mental conception of reality for their audiences. The use of carefully edited images evokes existing emotional or historical memories, eliciting an emotional response that may be conscious or subconscious' (p. 6). Their analysis of visual motifs is often repetitive and does not reveal explicit contextual or cultural motivations for the meaning-making process of their production. Bolt (2012) focuses on the process, stressing the importance of using pictures in what he calls the 'propaganda of the deed', and the emotive effect of the images jihadists use to highlight the underlying grievances of communities.

Multimodal analysis
The analysis of a multimodal text presupposes that the text is 'the material site of emergence of immaterial discourse(s)' (Kress, 2012, p. 36), that signs are motivated and that the text realises the interests of its maker (the designer or the rhetor). Examination begins with the representation structure of the visual describing its processes (narrative and conceptual), then moves on to the interaction among the participants studying the gaze, the frame, the distance, the perspective and the modality. Finally, the textual is analysed in comparison to the visual describing the process of meaning-making in the composition through an examination of three elements: information value, salience and framing.
A large number of multimodal texts was analysed fully and the resulting semiotic features were matched to recurrent discursive nodes. The semiotic meaning of the different components in this analytical model, however, depends on the context, such that the same element can be interpreted differently in different contexts. Certain regularity is maintained to some extent in terms of gaze and contact, perspective, colour and illumination, framing and salience. For instance, the absence of gaze and contact between the represented participant(s) and the interactive participant(s) (the viewer) is striking in most pictures, offering images for contemplation rather than demanding direct calls for involvement. This has meaning in terms of rhetoric, as it implies that this kind of images calls for their identification with religious obligation more than an attempt at persuasion. The underlying principle and motivation are closely connected to the discursive position of Islamic supremacy, excessive pride and self-exaltation. Pratt elaborates on this by arguing that 'in the process of negating the other, the self is asserted as inherently superior. My God is greater than your god. My Truth reigns over your ignorance. The authenticity of my faith contrasts with the feeble delusion you entertain ' (2006, p. 16).
The analysis combines semiotic meanings articulated in the textual and the visual, according to the affordances of each. The meaning of icons relies mostly on shared visual memory, and a relation of visual intertextuality (intericonicity; intericonic dialogue) features as constitutive of their network of recontextualised images. The visual repertoire draws mostly on the icons mentioned in the foundation texts, like the fruitful tree, light as representing God, the rope (of God) and clasped hands (both as emblematic of unity), the spiderweb (metaphor for weakness), fire and flood (metaphor for punishment), flock of sheep and the shepherd (political relationship between ruler and ruled in Islam), the prophet's war standard (monopoly of authentic Islam) etc. The textual puts emphasis on the process of anchorage and justification, recontextualising the foundation texts selectively to serve their goals. The multimodal text as a whole combines selectively visual elements and text in an attempt to establish a coherent process of identity construction, articulate a corresponding discourse, and propagate their model of social construction.
The analysis shows that the discourse of IS composed of six nodal points: sovereignty of Allah, hijrah (emigration), jamaa (congregation), loyalty and disavowal, jihad and khilafa (caliphate). Each nodal point leads to the next so that they are interdependent and interconnected. In other words, they are presented as a package to be taken as a whole. These nodal points fix the meaning partially and make predication in terms of identity possible. They are also constituted within an intertextuality that overflows them, and in the presence of the contingent and the antagonistic that are constantly attempting to subvert them.
In order to take a closer look at how these nodal points are formed, the analysis was conducted according to the critical model outlined below, with special focus on relevant semiotic features.

Congregation (jamaa)
Loyalty and disavowal hijrah jihad Sovereignty of Allah khilafa Figure 1: The main nodal points of IS discourse

Sovereignty of Allah vs secular laws
This paper is more concerned with how IS presents itself in different semiotic modes. Figure 1 is an infographic that places IS in juxtaposition with the rest of the world. This relation is framed as an opposition between the sovereignty of Allah on earth and the rule of secular laws, which they call jahiliyyah (ignorance), drawing on the works of Maududi and Qutb, and recontextualising the original meaning as it figures in the Hadith and the Quran where the term refers both to a The composition matches the foregrounded tree of flames with its background as interconnected, and partially realises the meaning of the textual in the visual. While the visual creates a distance with jahiliyyah, the textual explains it by creating a chain of authentication with a verse from the Quran that encourages peoples and tribes to recognise each other and acknowledge differences, and sets piety as a divine principle of selection that only God can know and judge. Then, a definition of jahiliyyah by a religious authority (Ibn Taymia) is given, stating that 'Anything that deviates from the call of Islam and Quran, based on lineage, country, ethnicity, madhhab or order, then is an affiliation of Jahiliyyah'. The kinds of jahiliyyah are represented in circles and a warning is mediated by a hadith that is explicit about the punishment of partisans of jahiliyyah in the afterlife. Then, jahiliyyah is represented as a disease or flaw that impacts character (arrogance, conceit and boasting) and the Muslim community (division and parting from jamaa). A final warning is launched against fighting for Jahili partisanship citing a hadith reinforcing the idea of punishment in the afterlife.
While the principle of selection acknowledges the recognition of difference among peoples, the definition draws a line between affiliation to Islam and affiliation to other laws. This demarcation rejects the Quranic recognition of difference (the universal) and establishes an antagonistic relation between Muslims (Us) and others (Them) (the actual). Political order is built on one absolute affiliation, lack of which leads to punishment in the afterlife. This discursive position establishes the cornerstone of the identity of IS: IS sovereignty equals the sovereignty of Allah; the rest of the world is its enemy (partisan jahiliyyah).
The discursive nodal point of sovereignty is constructed through a chain of authentication that uses intertextuality and the recontextualisation of four main concepts: sovereignty, jahiliyyah, partisanship and blights. While the use of sovereignty draws on different texts and events (the Quran, a principle used by the Khawarij during the civil war in early Islam revived by Qutb and Maudoudi), the religious concept of jahiliyyah has changed meaning (with Maudoudi and Qutb) to suit the goals of the modern political Islamic movements. IS uses both concepts as a benchmark of demarcation, authenticity and difference.
The third concept of partisanship establishes a link with early Islam and the life of the Prophet, and the Quran itself. Partisanship is derived from the Arabic word "parties" (ahzab), who were allied Arab tribes who fought the Prophet in a battle called the "battle of the Trench" (Khandak). This intertextual allusion is recontextualised by IS to be used as a denigrating name for the enemies, and identifies IS as representing authenticity.
The fourth word, "blight" (afa in Arabic), rhymes intertextually with a saying of the Prophet that classifies the blights of religion in three categories: a corrupt scholar, an unjust ruler and an ignorant interpreter of religion. The term is recontextualised to classify the blights of jahiliyyah into five categories: the first three concern character (arrogance, conceit and boasting), while the other two are related to community (dividing the Muslim rank and parting with congregation, and breaking the tightest of all bonds, that among faith, loyalty and disavowal). The blights of character are only a camouflage for the blights in relation to the community (IS), as the latter are more important for social control. IS discourse tries to maintain regularities through intertextual links among its nodal points in a systematic way.
In terms of propagandistic rhetoric, IS discourse addresses both the general public (vertical communication) and the followers of IS (horizontally in a social network) with different strategies and effects (Oddo, 2018). The aim of the selection of citations from the Quran, the Hadith and Ibn Taymia's work is to make their discourse recontextualised favourably, especially among followers, as the citations belong to authoritative voices that have a huge capacity for semiotic mobility. Besides, this effect is reinforced by what I would call "a logic of exchange", according to which religion sets two types of exchange between people and God: promise of heaven in exchange for faith, and punishment in hell in exchange for disbelief. This logic is partially exploited to warn disbelievers and Muslims alike against siding with jahiliyyah, hoping to coerce audiences into fear and guilt. In semiotic terms, the burning flames remind them of the punishment that awaits them.
2. Hijrah to Sham: a mythic frame for a political struggle Hijrah is a fundamental principle in political Islam as a strategy of demarcation from the rule of jahiliyyah in Muslim society. Its psychological meaning as an emotional separation (Qutb) from Muslim society is stressed by IS as a prerequisite for joining the Muslim community as they imagine it. While, however, the psychological (and also social) meaning of hijrah is maintained, IS puts more emphasis on the physical meaning of hijrah as movement from societies of jahiliyyah to the land of Islam where sovereignty belongs to God. This recontextualisation of hijrah was done at a time when IS needed to mobilise people in divided, weak and unstable Syria to join the IS soldiers, taking advantage of the civil war and the state of savagery and disorder there. The intertext links such hijrah with the Abrahamic tradition, highly respected by Muslims as an emblem of monotheism (tawhid).
The picture at the top ( Figure 3) represents hijrah in a narrative process where the actors are people riding camels across the desert heading towards a destination (Sham). Though this means of transport is seldom used for such purpose today, it is invested with cultural meaning linking the concept of hijrah to the Islamic (hijrah to Medina) and Abrahamic traditions. Recontextualising the concept of hijrah as a religious belief and merging horizons by abolishing historical differences make the identification of Muslims with such tradition easier. The photo consists of an extreme long shot frame, and though it may seem impersonal, the image of camels in the desert is culturally embedded as emblematic of a historical religious event. The combination of light and shade creates the religious aura of a tradition and sets an epic tone and mood.
In terms of composition, the image represents hijrah as an ideal rooted in the Muslim imaginary. It is a movement seeking radical change that is much revered and has an emotive appeal to Muslims, as it is reminiscent of a time of transition in the life of the Prophet, and of the tradition of Abraham. The text is the reality to which this ideal is to be applied. The expansion of IS to Sham is supported with ahadith stating that hijrah will never end ('Hijrah after Hijrah') and with the quotation of a famous Salafi scholar (Ibn Taymia, saying that 'Islam in the end of times will be more manifest in Sham'), represented as legitimate. It is even required, according to tradition, to join the major battle of Dabiq. The portents of the hour are manifest and Muslims have to get ready for the malahims (great battles) in Sham. The recontextualisation of hijrah by establishing a link between the Abrahamic tradition and the expansion of IS to Syria is political, as the call for the revival of the Abrahamic tradition is meant to mobilise volunteers for jihad to whom the movement to Syria is framed within a revered religious tradition. The token-sourced intertextuality serving such recontextualisation links IS discourse to Abraham, who is the emblem of monotheism for Muslims. Another frame used for such mobilisation is the frame of the apocalyptic. IS's attempt to dramatise history and give a mythic dimension to their battles is foregrounded in their representations investing their struggle with religious meaning. IS appropriates the idea of the apocalyptic as it is described in the tradition, especially in the Hadith, and makes use of it by recontextualising it as applicable to modern times. The chain of authentication is reinforced by the fact that the area of Dabiq in Syria is mentioned in the Hadith, and is specified as the location of a great battle at the end of times (malhamah). The intertext in the discourse of IS referring to the hadith does not guarantee the link all the time, as IS changes its interpretation of the historical events actually taking place -especially after their defeat became imminent -claiming that the battles they fought there were only minor signs of the Hour and that the definitive battle is yet to come. This deferral translates into a denial of defeat and an escapism into an unknown future, in order to save face. IS internalises its defeat as impossible because they believe that the struggle will continue and they will be the "victorious sect" in the end. Their perseverance cannot be maintained (baqiya: "remaining till the end of times") in realistic terms but they resort to the mythical transhistorical apocalypse and the continuous deferral of this end for strategic ends.
The two frames of hijrah and the apocalyptic provided a religious and mythic cover for IS expansion into Syria and reinforced the polarisation between right and wrong (true believers vs Dajjal, the anti-Christ) dramatising the conflict in transhistorical terms. Their defeat in Dabiq was articulated as temporary and modified this discursive nodal point, as well as their propagandist strategies by mobilising would-be jihadists to stay home and cause terror locally with whatever means available to them.

Congregation and the appropriation of authenticity and legitimacy
While hijrah mobilises volunteers to join the community of Muslims as it is perceived by IS, the concept of congregation, as defined and exalted by the Sunni tradition based on the foundation texts, is an utter appropriation of authenticity and legitimacy. The Sunni ideology of congregation is constructed as a means of identification and exclusion. All Muslims should belong to jamaa, while dissidents should be viewed as alien to it and even as disbelievers. IS adopts this concept of congregation and founds it exclusively on monotheism, which should be accepted as a 'package-deal phenomenon' (Pratt, 2006, p. 17), including the six points outlined earlier (sovereignty, hijrah, congregation, loyalty and disavowal, khilafa and jihad).
The authoritarian character of monotheism is manifest in their insistence on the fact that power should not be shared and that it is the leader (imam) who wields it. Authoritarian power is strictly hierarchical and requires complete obedience (ta'aa). The representation in this design ( Figure 5) is a conceptual, analytical process. The participants (flock of sheep and a dog) are carriers of certain attributes; the flock of sheep is united, obedient and peaceful, and the dog is watching over them from a distance. The interaction between the represented participants is that of control and obedience. The distance between them informs the existing power relation. The image shows no gaze and is offered for contemplation and identification. The eye-level perspective reinforces the created desire for identification (for the viewer), as it connotes unity and equality. The image is framed as a wide shot showing a naturalistic view of a green field in saturated colour and highly illuminated by sunlight. The abundance of light reinforces the tranquillity and peace that such unity (horizontal and vertical) conveys.
In terms of composition, the structure of information shows a top to bottom organisation. The ideal that is portrayed visually is anchored textually through a chain of intertextual authentication and recontextualisation. At the core of this chain of authentication, we find two ahadith and an allusion to a Quranic verse. The first hadith explains the metaphor represented by the picture that 'every man is a shepherd and every shepherd is responsible for his flock', and concludes that 'every imam is a shepherd and is responsible for his flock'. The second hadith stresses the importance of sticking to the jamaa (congregation) of the Muslims and their imam, warning Muslims of the danger of joining other groups. The purpose of this attempt at authentication is to prove that the noble rank of imam that God has given to Abraham (in a Quranic verse) is not only religious but political as well, as the imam should be a man of justice and virtue and should have the force to fulfil both.
The analysis of various designs of the same orientation shows that many symbols are invested to consolidate this notion of jamaa. Clasped hands is a reiterant symbol, as they portray unity and ba'yaa (allegiance), which is a Muslim social contract between the imam and the Muslim community. A second symbol is the rope, used as a metaphor in the Quran for unity (the rope of God) around monotheism and the worship of God. In order to consolidate the meaning of unity, IS also uses the symbol of the strong thread that is mentioned in a hadith about a foolish woman known for always undoing the thread she had spun. In the same vein, they use the metaphor of the worn-out garment from the Hadith to denote political mobilisation, as it is representative of faith that has become weak, and is used to call Muslims to renew their faith and join jamaa.
Another attempt at manipulation is the overlap they construct between obedience to God, which is religious, and obedience to the ruler (imam), which is political. IS uses this overlap as a means of control, investing the submission of followers with a surplus of meaning that is illusionary, namely that God is pleased with them and promises to reward them in the afterlife if they obey the imam. They also protect the imam of the Muslim jamaa from challenge and accountability by insisting on his noble lineage (the name of the imam should mention the name of the tribe to which the Prophet belonged, Qurashi or Hachimi).
The intertextuality sets ties with tradition and the recontextualisation does the rest.
Recontextualising the religious imamah as political is in line with IS's attempt to consolidate the power relation that their discourse constructs -and is constructed by -to give the political organisation of their group a religious anchorage and authentication. It is true that these types of recontextualisation are based on selective intertextuality; however, this does not mean that recontextualisation is exclusively political. It is more the result of an attempt at 'compliance with the teachings of the pious predecessors (al-Salaf as-Salih)' (Lahoud, 2005, p. 4) for whom religion was also politics. Recontextualisation often fails not because of selection (as it is a common practice in all religious groups and sects) but because the interpretation of the selected text remains literalist and is done in terms of 'perspectival absolutism and immediate inerrancy' (Pratt, 2006, p. 12). Pratt argues that 'the assertion of the immediate inerrancy of the text -namely reading the text as being immediately applicable and providing a nonmediated access to ultimate or divine truth -in fact involves an implicit assertion that there is only one normative interpretive reading allowed ' (2006, p. 11). This tendency to reject all other interpretations is the expression of an ideological extremism that affects the degree of correlation between their narrative and the real world. Religion, as applicable by IS in a pluralist global reality, 'would reduce to a simple and obvious fairy-tale superimposed on the "real world", or set aside from it' (Pratt, 2006, p. 15).

The panopticon of loyalty and disavowal
Belonging to jamaa in the name of tawhid (monotheism) entails obedience, which is put under surveillance using the principle of loyalty to jamaa and disavowal of disbelievers. The two elements of this principle are like the two sides of a coin and they should not be separated. It is not enough to be loyal but one has to show no affection even towards one's close relatives if they are not loyal to jamaa (IS). Loyalty and disavowal have to be manifest in one's conduct and followers should be able to exhibit such signs. In Figure 6, the concept is represented by an analytical process, where the carriers are passports of different nationalities, perforated by bullet shots, jumbled up with empty bullet shells, and having the attributes of useless documents. It is an image offered for contemplation and identification.
The multimodal composition is organised in a top to bottom structure, where the text foregrounds the principle as an ideal that should be followed by all Muslims. It is defined by a hadith: 'Love for the sake of Allah, hate for the sake of Allah, make allegiance for the sake of Allah, and make enemies for the sake of Allah, for the wilāyah (loving guardianship) of Allah is not attained except by this. One will not find the taste of īmān (faith) even if his prayer and fasting is great until he is like this'. The text places the principle at the core of religion and even subordinates prayer and fasting to it. Love and hate for the sake of Allah are recontextualised from the religious context of brotherhood to the political context of an Islamic state. It becomes a principle of political organisation, where jamaa is a political construct based on the inclusive principle of loyalty and the exclusive principle of disavowal, and framed within strict boundaries.
The verbal anchorage from the sacred text and tradition justifies the decision to abolish the idea of national states (Sykes-Picot agreement) and establishes the idea of khilafa, one nation that should expand globally to transform the whole world into the land governed by the rule of Allah. The picture at the bottom confirms this ideal of one nation in reality by abolishing division among Muslims; the perforated passports show that this notion of one nation would not be possible without fighting and jihad (the bullet shots). The text warns against having affection for those who oppose Allah and his messenger and by extension IS. It also warns Muslims against becoming Jews or Christians without knowing it by sympathising or having affection for them. Self-censorship is required. Surveillance is categorical and conformity should be total.  Figure 7 reinforces the meaning of the principle in positive terms. This is represented in the picture by a conceptual process whereby two participants (both men) of different racial and national origins, an Arab and an Asian, are laying hands on each other's shoulders and smiling. The gaze is directed at each other, making contact through the double gaze (the viewer) and involvement obvious. The picture is framed in a medium shot with close social distance connoting intimacy and brotherhood. The perspective on the horizontal angle involves the viewer; on the vertical angle, there are no power differences, reinforcing the impression of harmony. In naturalistic saturated colour, the background points to a probable mountain frontline, while the two men are carrying guns and are on a mission of jihad. The text is placed in the middle just between the two men: 'walaa and baraa versus American racism'. Religion abolishes racial differences and makes all people equal. In democracy, racism is still common practice.
While Figure 6 addresses followers urging them to maintain total obedience to jamaa and distance from enemies (of IS), Figure 7 compares the equality among Muslims to racism in the USA. The objectives are to demonstrate the superiority of Islam to other Western societies and create a feeling of pride among Muslims. Figure 7 is propaganda-oriented and works on both horizontal and vertical levels. Presenting Islam and IS as promoters of equality serves IS's ideological claims of being a global universal movement open to all people of different nationalities and races. This claim paves the way for their attempts at mobilising people across the world. Figure 6 is more horizontally oriented, as it seeks to control followers and rally them in the name of love and hate, at the cost of Allah alienating them even from their own families. The ideological dimension of this principle lies in the fact that followers internalise it as constitutive of their subjectivities, becoming the principle of their own subjugation, willing to follow and kill in the name of Allah (IS). Total conformity and enmity are framed in religious terms and have an excessive dimension, in line with IS cruel savagery. The "Big Brother" is watching not only over visible acts but also over inner feelings, as no affection should be given to enemies. Figure 8 describes explicitly how followers should act towards enemies: 'Spilling the blood of the mushrikin (disbelievers) is the greatest form of disavowal'. The bloody hand shows what disavowal means for IS followers. The utmost cruelty is framed as hate for the sake of God and further alienates IS followers from the universal religious common sense.

Jihad for victory or martyrdom
Jihad is a focal node in the discourse of IS. What is so special about jihad for IS that it is perpetual and existing as long as the struggle continues between jihadi Salafism and the rest of the world. IS articulates hijrah with jihad. Figure 9 is a composite design, where both conceptual and narrative processes are at play. It features carriers, who are soldiers attacking enemies at night, and are watching over foes ready to launch rocket-propelled grenades (RPGs) from behind a trench. The main attributes assigned to them are determination and engagement. The scenes also narrate the process of attacks targeting invisible goals. The interaction is maintained between the rhetor and the viewer, and the absence of contact with the represented participants offers the image up for contemplation. The shots, taken from behind, make the soldiers look like iconic, archetypal figures and stand-ins for the viewer. In terms of power dynamics, the represented participants, seen from below, look superior to the viewer, inspiring awe and fear.

From hijrah to jihad
The composite design, with four images featuring soldiers attacking enemy camps, presents determined and engaged jihadis. The merging of top, bottom (present) and centre (text from the past) is a fusion of horizons, of an ideal past and a historical reality. The text is a hadith that links hijrah to jihad: 'Hijrah will not cease as long as there is jihad'. The text gives an anchorage to the visual making these attacks seem legitimate and justified. The verbal anchor controls meaning-making and recontextualises the military attacks in the religious context of jihad.

5.2.
Jihad and tourism (siyaha) Figure 10: The siyahah (tourism) of jihad (Rumiyah, 2016, 5, p. 20) IS makes the link between hijrah and jihad clear and warns against seeking other ends. It is not tourism, as contended in Figure 10. In order to make jihad appealing to would-be jihadists looking for the glamour of adventure and fighting, however, IS resorts to conceptual blending, subordinating "tourism" to jihad and framing it as a religious experience.
In Figure 10, the process is conceptual and analytical; the carriers are two soldiers standing on a hill against a misty background, wearing uniforms and holding rifles. They are not fighting but looking at each other. Contact is maintained between the represented participants. It is an image offering contemplation of the concept of jihad as tourism. The picture is framed as a long shot, with an eye-level angle. As focus is on the concept, the soldiers are presented as archetypal. The colour is less saturated and the background is misty. With no further details and low illumination, the scene looks as if it were a sort of dream.
While the title anchors the image and orients its interpretation, the text frames that interpretation by redefining the concept of tourism as not traveling from one country to another for pleasure but doing so while practising jihad. Though they quote a hadith ('the siyaha of my Ummah is jihad for Allah's cause'), they reinterpret the word siyaha to serve their dual goal, that of articulating it with jihad and of making jihad appealing to adventure seekers.
At the same time as discarding the monasticism practised by deviant monks, the reinterpretation of siyahah as jihad for the cause of Allah restricts its meaning to include what is found in the monastic siyaha of zuhd, namely reclusion, prayer and worship; in other words, siyaha is preparing oneself for jihad while enjoying the landscapes and the adventure. The discourse of IS blends concepts and creates new meanings, recontextualising them in terms of their ideology of jihad to which all social and economic activities are subordinated.
The propaganda at work in such designs tries to appeal to Western adventure seekers and to what is called 'jihadi cool' (Cottee, 2019;Picart, 2015). Picart (2015) examines the process of transition of lone wolf terrorists from radicalisation of thought to radicalisation of action, and argues that the appeal of the counter-cultural 'jihadi cool' propagated through the Internet is one of the factors of self-radicalisation. According to Picart, terrorist groups like IS use social media to appeal to youth circles by posting 'videos created specifically for western audiences […] to depict jihad as a Hollywood-like video game ' (2015, p. 365). The targets of this jihadi-cool propaganda are what Picart calls the 'badass', who are people who 'embrace symbols of deviance' (2015, p. 362) and replace the gangster face with a jihadi face. Cottee (2019) quoted an IS fan called Bint Emergent (probably the Twitter handle of a girl): 'Jihadis […] look cool -like ninjas or video games warriors' (p. 51); in another quote, he reported her saying that 'the Islamic state is the classic sci-fi underdog battling a seemingly all-powerful Evil Empire America against impossible odds' (p. 51).

Jihad and the logic of exchange
Jihad is the culmination of the effects of all the previous discursive articulatory practices in an antagonistic contingent historical context, and it is the force that empowers IS and its project of khilafa. Jihad is viewed by jihadi Salafism as another complementary pillar of Islam (including five other pillars) in constant need for activation. It is highly exalted in their literature in that all the other pillars are reduced to nothing if they do not lead to jihad. The concept of jihad is a complicated one and, as a nodal point, it includes many moments that are enunciated through an articulatory practice using a chain of intertextual authentication and recontextualisation. Jihad is a negation of the self in that the jihadis renounce everything with a view to getting something in exchange. This logic of exchange is what informs their will to sacrifice themselves. The analysis of hundreds of designs and multimodal texts shows that this logic could possibly explain their motivated negation of the self and the killing of others.
First, the jihadists perceive their military actions as a sacred mission serving the cause of Allah in terms of a covenant structured by a clear exchange between them and God. Figure  11 shows soldiers sitting in a circle, joining hands and showing a great sense of solidarity, unity and brotherhood. This conceptual process offers no interaction with the represented participants and urges viewers to contemplate the scene. The picture is framed as a wide shot from a high-angle perspective. The image is taken from above highlighting a power relation where the represented participants are depicted as inferior, submissive and obedient, and where the perspective is created in such a way as to give the power over those people to God. The viewer is invited to look at them using that perspective but s/he is not given any power over them. Those obedient represented participants are proud of being inferior and submissive, but only to God.
It is not surprising to see Muslims, especially extremists, accept obedience and submission not only to the divine power but also to the power of the imam. This is part and parcel of a strong belief in the jamaa and the obligation of maintaining allegiance, obedience, heeding and submission. From a cultural point of view, these extremists accept an unequal distribution of power and do not feel any embarrassment in showing obedience and submission to their leaders. It is even considered a virtue.
The colour of the image is a shaded unmodulated brown, creating an effect of homogeneity. The soldiers are not differentiated; they constitute a whole made from undistinguished parts. The image creates an ideal typical representation of unity in the Islamic culture, namely clasped hands. Below the image is its verbal anchorage of the legitimation of the concept and practice of 'the pledge to fight to the death'. The text is structured in such a way as to highlight the obligation of keeping the pledge made to the Prophet and Allah by praising believers who are patient and steadfast during wars, and warning those who retreat before the enemy. While constructing intertextual links with the Quran and the Hadith, however, the interpretation articulates the pledge within the historical contingency, recontextualising it as a pledge to fight with jamaa represented by IS. Fulfilling the covenant with God is an obligation if the deed is permissible. Otherwise, it is forbidden. The text reinterprets the verses, ahadith and quotes from tradition by specifying what is permissible today and can be the object of a covenant made with God. The only permissible deed is to fight in obedience to God. If the combat is against the Muslim jamaa (IS), it is forbidden and the covenant must be broken. Fulfilling such a pledge is rewarded by eternal pleasure in paradise. Figure 12: Taking the wealth of disbelievers is permissible (Rumiyah, 2017, 8, p. 12) Second, the negation and the alienation of the self in exchange for a life in paradise does not mean that jihadists do not get worldly rewards. The revival of historical tribal practices in wars, like booty (ghanima) and slavery (of women and sexual abuse in the name of religion), take place in terms of contingency. The justification offered by IS discourse frames the notion of booty in a fikh (law) of jihad (Figure 12), rejecting the approach of some scholars who are called scholars of taghut and promote the fikh of pacifism that is founded on the peaceful coexistence of all religions, pacifism and the abolition of jihad. Alterity is negated by the revival of the fiqh of jihad, placed at the centre of religion and social life. The world is divided into two parts: dar al Islam (land of Islam) and dar al Harb (land of war). Accordingly, those who live outside dar al Islam (namely the Islamic State / khilafa) are permissible targets, including harbi kuffar (belligerent disbelievers), who consist not only of soldiers but of all those to whom Muslims have not granted security, either with a dimmah contract (with Jews and Christians) or a ceasefire treaty. The blood and wealth of disbelievers, including apostates, are permitted to be violated by Muslims. Dar Al Kufr is the land of ibaha (permission) and the disbelievers have no ismah (protection).
The reinterpretation proceeds by selecting texts, opinions and exegeses that support their political ideology, and meets the needs of IS to wage a global war where everything is permissible by religion and, thus, legitimate. Legitimising the killing of people who are not at war with one, and stealing their money by force, or fraud and deception are other forms of legitimating and normalising terror. Another reward that jihadists get is the permissibility of enslaving captured women and having sex with them as slaves (ima'). Figure 13 shows a convoy of cars celebrating the conquest of Nainawa for propagandistic ends. The long-shot frame builds an epic tone focusing on the convoy as a whole from a narrative high angle to tell the story of a conquest and victory, and give it context. IS discourse links the End Times (the Hour) with the revival of slavery. The reappearance of the practice of slavery is one of the portents of the Hour according to a hadith: 'the slave girl gives birth to her master'. The conquest of Nainawa, home to a population of Yazidis, a pagan group, raised the question of how to treat the captured women. Before any decision could be made, IS had to give them a status, choosing between two options. The first was apostates, which does not give IS men the right to marry Yazidi women, as female apostates cannot be enslaved and married to IS men. They can only be given an ultimatum to repent or face the sword. The second option was mushrikin. According to IS, the Yazidis worshipped Ibliss (Satan) and should be treated as mushrikin (disbelievers). IS has chosen to give Yazidis this status so that they can enslave their women and "marry" them. Captured women are distributed according to Sharia (law) among the fighters.
This interpretation is an attempt to justify in religious terms the sexual abuse of Yazidi women. Recontextualising a tribal practice as a religious one in a world where even wars have laws and ethics shows how IS interprets the foundation texts to serve their interests, desires and ideology in an ever-changing historical reality. Third, although worldly rewards are framed as the religious rights of jihadists, a surplus of meaning is always needed to maintain the religious aura of jihad in their minds. Jihad is essentially carried out in the name of the sacred covenant with God, and that is the reason why people believing in it agree to surrender everything and make a total negation of their selves in exchange for a reward deferred to the afterlife. Figure 14 features five soldiers holding rifles, smiling, cheerful and happy. They are called martyrs, as they have carried out an attack on a place identified as Gulshan. The picture was taken prior to the attack.
The direct gaze at the viewer from the smiling faces of would-be martyrs gives the impression that these five soldiers are happy with what they are going to do. The gaze reinforces the idea that martyrs are not dead people but continue to live on in heaven. One cannot be happy to face death unless one has this conviction. This is a demand image; it is an invitation to join the shuhada. The image is a superimposition of the photo of soldiers on the IS flag, and together they form an ideal, namely fighting for IS/Islam.

Just terror politics and the state of savagery
Excessive violence has been one of the hallmarks of IS terrorism, especially their videos of scenes from the execution of prisoners. Many scholars have debated the issue. While Waller (2002) argues that terrorists kill their victims because they consider them morally dead and dehumanised and, thus, not worthy of any compassion, Bandura (1998Bandura ( , 2004 argues that terrorists are able to commit such horrible acts because of what he calls 'moral disengagement', which Hafez defines as 'the process by which (these) cognitive codes (inhibiting people from killing) are deactivated by transforming immoral conduct into ethical imperatives ' (2006, p. 169).
IS terrorists have exceeded all limits with their shocking videos of cruel executions. Atouane has added another feature, namely the harsh and tribal nature of the people of Iraq. This may be untenable as it needs further scrutiny; however, the suffering of the Sunni people of Iraq after the US invasion has sharpened their feelings of injustice regarding what was inflicted upon them. In terms of religion, these feelings could be explained by a chain of intertextual authentication drawing on selected verses and ahadith that call Muslims to be harsh towards disbelievers by terrorising them. Though this call was issued at a time of war between the Prophet and disbelievers, IS discourse recontextualises terror as a mode of communication that does not only address the belligerent disbelievers but also the innocent, peaceful people living in the land of Kufr. IS's call for would-be jihadists to stay home and launch operations with whatever means they have (stabbing, arson, bombing etc.) is also meant to create terror among civilians and to give 'a lesson to those nations that wish to take heed' (Figures 15-16). Creating fear has propagandistic ends; IS uses both emotional coercion, by activating fear and worry among civilians, and proximisation strategies, as they make them expect attacks from anywhere, even from places they deem safe.
Excessive cruelty is also in line with the state of savagery in which IS operates, especially in Iraq and Syria. They communicate fear to their rivals to dissuade them from fighting or surrender (this reportedly happened in Syria). Excessive cruelty is more of a war tactic than a religious stance, though IS uses religious terms to normalise the killing of prisoners and civilians. IS discourse modifies tactics in terms of the ongoing situation and the emergent conditions of their changing political circumstances in the region.
6. Khilafa: the deferred dream The declaration of the Islamic State came at a time when the state of savagery both in Iraq and Syria presented a favourable political opportunity for IS. The creation of a state by terrorists was unprecedented and was actually enabled by the political disorder reigning in the region. The political identity of the Islamic State was explicitly presented in the discourse of their leaders. It is a revival of khilafa, the well-guided rule of Caliphs in early Islam. The khalifa (leader) of IS chosen thanks to his charisma and noble lineage, which according to tradition is a factor of unity. Figure 17 is a design in which distance, salience and framing are at play. The khalifa is presented in a medium shot but made more salient with personal distance. The coalition is presented in a long shot but pushed to the background and made less salient by the adjoining picture of the Alaksa mosque in Jerusalem surrounded by barbed wire. The collage creates a composite picture, where modality applies to composition. Original pictures are assembled with a clear hierarchy and framing: the foreground (the khalifa) is disconnected from the background and given prominence; the latter is composed of two pictures, the coalition and Alaksa, as a reminder of the hypocrisy of the Gulf countries and their submission to the West.
The text reaffirms the meaning of the title 'conspiracy against conspiracy', which has an intertextual relation with a verse in the Quran and is recontextualised by appropriating the meaning of conspiracy made by God in response to the conspiracy of disbelievers, and in the process identifies IS with the will and deed of God, as if God is conspiring on their behalf. In Figure 18, the declaration of the khilafa is framed as a response to the humiliation of Muslims by the West and is a call for 'the Ummah of Muhammad […] to wake up from its sleep, remove the garments of dishonor and shake off the dust of disgrace and humiliation'. Feelings of superiority and pride were at their peak after the invasion of Mosul.
The Islamic State was careful as to how others should see it. The headlines of the designs I collected to this end, show that IS an akida (creed) watchdog and this creed places it in antagonism with all other sects, religions and doctrines: Jews, Christians, apostates, Rafida (Shiites), the Muslim Brotherhood, the Taliban, wicked scholars, forces of jahiliyyah, and the rest of the world. IS presents itself as a new historical actor in a new era and as a state providing services to its citizens (collecting and distributing alms, moral policing (hisbah), education, dismantling cells, destroying temples, proselytising, medical care etc.).
IS discourse presents itself in an either/or logic, asking others to choose between IS or the Flood, IS or disorder ( fitna), IS or humiliation. These phrases are intertextual and have a high capacity for semiotic mobility as they are reminiscent of mythical (Noah's Ark), historical ( fitna during the early days of Islam after the death of the Prophet) and ongoing events and states (invasion of Muslim lands seen as humiliation). IS in these terms the saviour of Islam and Ummah (the nation of Islam).
After the defeat of IS in Syria and its expulsion from Mosul and other cities in Iraq, the IS khilafa is shrinking and looks more divided, seeking allegiances from other jihadists in Sinai, Mali, Nigeria, Mozambique and other places in the world. Ideologically speaking, IS clings to the principle of collective takfir and now targets other Islamic sects, including the Taliban, Al-Qaeda and the Shiites. The sectarian dimension has now become more overriding, while the khilafa is becoming more of a deferred dream than a historical reality.

Discussion
The analysis of multimodal texts shows that IS has a shifting identity in terms of tactics (hijrah, just terror politics, the revival of slavery, the apocalypse), though general strategies have remained partially the same so far (collective takfir, sectarianism, will to power and hegemony). This shifting nature leads, at times, to contradiction between what IS doing (the particular) and what it preaches, like "equality", "love for the sake of Allah" and the "recognition of difference" (the universal). These contradictions emerge when the discourse and the actions of IS are not compatible, which makes them, due to antagonistic new historical realities, change their discursive positions in order to survive and maintain consistency and credibility.
The articulatory practice that is at work relies mostly on intertextuality and recontextualisation. IS discourse depends on selective intertextuality to construct a chain of authentication that is used as anchorage and justification. The aim is to gain legitimacy in the eyes of Muslims. They select texts, detachable (memorable) fragments from foundation texts, popular icons from the collective visual memory (basically drawn from religious texts), and concepts and frames that are embedded in religion and culture. They are careful to set a link with the Salaf to gain authenticity and monopolise it, excluding others from representing Islam as tawhid.
Recontextualisation is at the heart of this articulation of the intertext and it is effected in different ways. First, IS selects a concept or frame from the tradition by entextualising it through intertext, and decontextualising it from its immediate past context, while creating a new meaning and inserting it in the present context of antagonism and contingency. The appropriation of these concepts, meanings and frames is done to serve political goals, like the use of the concept of jamaa to ensure social control, exclude other sects and gain monopoly over authenticity, and the revival of slavery to meet the needs of its fighters and appeal to their desires. In a similar vein, apocalyptic elements are used to frame their struggle in a religious mythic mode.
Their shifting discursive positions show that their discourse is in constant flux and will change in terms of future antagonistic relations and historical realities. The partial fixations observed show that the sectarian dimension will be stronger in the future and the tendency to collective takfir will isolate IS further. IS will keep shifting positions, concepts and frames to serve their political ends of survival.
The propagandistic orientation of their discourse is manifest. It works horizontally on followers to maintain loyalty by using the rhetoric of identification (absence of gaze, offer images, mythical tones etc.) and also authoritative voices that have high semiotic mobility. When addressing Western audiences, their discourse plays on the fear of civilians through emotional coercion and proximisation. Propaganda also uses detachable memorable fragments in both directions to spread favourable concepts and frames. It also takes advantage of the counter-cultural appeal of IS to young people in the West.

Conclusion
The study of IS discourse has shown that it is political in motivation, religious in justification, and intertextual in its recontextualisation. There is an inherent linking of religion and politics in their discourse, as they invoke the foundation texts in a highly selective manner that is conducive to their political goals. The analysis has shown that the discourse of IS changes positions, strategies and claims in alignment with shifting historical contingencies and the new antagonisms within which it has to justify its claims and actions. Reappropriating religious authenticity is meant to reinforce the legitimacy of IS in the eyes of Muslims, and its discursive positions and actions in the face of antagonistic claims.
IS discourse, as has been argued, uses religious intertext and recontextualises it to support its political claims. The revival of the notions of jihad, slavery, booty, emigration, the apocalypse, loyalty and disavowal serves to give a religious justification and authentication to IS's practices of terror, framed in mythic and religious terms, rape, theft and looting to appropriate resources, the mobilisation of would-be jihadists, and the surveillance of its members. IS discourse takes advantage of the affordances of the new media to propagate its claims and strategies of terror. On the one hand, it uses terror as a mode of communication that targets antagonistic audiences, intimidating them through strategies of emotional coercion and proximisation; on the other hand, it uses intertextual recontextualisation to justify its actions and persuade in-group members, would-be jihadists or the Muslim community through the use of authoritative voices with high semiotic mobility. Memorable fragments are used as slogans in a propagandistic discourse that is constantly changing tactics, targets and justification.
One of the main aims of research in terrorism studies is to understand this phenomenon, predict its future shifts in discursive positions and counter any future form of terrorist extremism. The discourse of IS will usually justify and authenticate its political claims in religious terms using fragments, arguments or authoritative voices from their vast religious textual heritage, and recontextualising them to serve and justify their shifting discursive positions and actions. Changes in IS discourse can, thus, be predicted to some extent and the dangers of its ideology can be less effective in a world that is united in tackling the problems of political disorder, the state of savagery, and guerrilla wars in politically unstable countries.