In the domain of PVE as well as reintegration, the most interesting studies are arguably based on material collected first-hand from the individuals involved in the phenomenon of political violence or terrorism. As more individuals from the 2013-2016 wave of foreign terrorist fighters are exiting the criminal justice system, young individuals with no memory of that period are sympathizing with ISIS and others again are joining right-wing groups with violent agendas. Understanding the motives behind such engagement will always lead a portion of the scholars to pursue interview-based studies. This paper describes the research protocol used for a study which dealt with politico-ideological mobilization and violence in relation to causes and conflicts in the Arab World. More than one hundred interviews were conducted in Lebanon, Switzerland and Canada with individuals involved in politico-ideological mobilization or violence of different ideological orientations. Besides interviews, complementary material in the form of ethnographic fieldnotes and voice recordings via instant messaging were collected. The data was compiled into a MAXQDA database and coded according to the principles of Grounded Theory, using open, selective, axial and theoretical coding. The paper further discusses epistemological and ethical considerations.
- Introduction
The positivist dream of an epistemological state of perfect innocence papers over the fact that the crucial difference is not between a science that effects a construction and one that does not, but between a science that does this without knowing it and one that, being aware of work of construction, strives to discover and master as completely as possible the nature of its inevitable acts of construction and the equally inevitable effects those acts produce (Bourdieu, 1996, p. 18).
This paper describes the research protocol used in a study on politico-ideological mobilization and violence in relation to the Arab World (Ajil, 2023a). The objective of the study was to understand the motives of individuals from Switzerland, Canada and Lebanon who mobilized for causes and conflicts in the Arab World, such as the Israeli-Palestinian conflict, or the Syrian civil war. Research subjects were hence divided into two groups: Politico-ideological mobilization (PIM) including primarily non-violent activists, and politico-ideological violence (PIV) referring to individuals who espoused violent tactics and joined combating factions, some of which continue to be widely regarded as terrorist organizations. Expert interviews were also conducted but this paper will focus on the interviews with these two main groups of interest.
The main data collection instrument were interviews. The data collection and analysis strategy will be explained in detail in what follows. Further first-hand data was collected through instant messaging applications and through fieldnotes, alongside autoethnographic writing. Second-hand data was collected, where available, through trial files and open-source (e.g. social media profiles) information about interviewees.
Scholarship in the field of radicalization and deradicalization has evolved significantly. As the major wave of foreign terrorist fighters moves further and further into the past, many individuals are gradually exiting the criminal justice system, sometimes distancing themselves from the violent ideology they espoused and sometimes changing mainly in behavioral terms. This population will remain of interest to researchers in the field and is more likely to become available for research purposes, not least because both its engagement and the contact with the criminal justice system will have profoundly affected their life-course and psyche, and that the need to talk about these experiences becomes more apparent. On the other hand, terrorist organizations such as the group “Islamic State” are regaining force and attractiveness, with young sympathizers pledging allegiance to it and even committing violent attacks in its name. In that sense, the detailed research protocol presented in this paper should be of use to researchers and practitioners who engage with this population, by providing insights and lessons learned.
The paper will first present ontological reflections on the use of narratives in research on PIM/PIV (chapter 2). It will go on to elaborate on aspects related to the conduct of the interview (chapter 3). This includes the preparation, the setting and modalities of the interview. The choice in this study was to use semi-structured interviewing methods, and an epistemological positionality qualified as “understanding-based interviewing”. The process of transcription and analysis of the interview transcripts using Grounded Theory and the application MAXQDA will then be explained (chapter 4), before turning to the complementary primary data collection (fieldnotes, instant messaging, autoethnography) in chapter 5, additional collected material in chapter 6, and concluding with reflections on implications for practitioners and researchers (chapter 7).
- A TRUTH, TRUTHS OR NEITHER? ON THE ONTOLOGY OF NARRATIVES
Given that the main corpus of data in this study consists of original narratives, it is important to briefly lay out the ontological reflections guiding the research. Bonelli & Carrié (2018) argue that conducting interviews with former militants or individuals engaged in PIV implies risks of providing a platform that interviewees may use to publicize and politicize their struggle. Also, they suggest that interviewees will tend to rationalize and justify their past actions in light of present circumstances, restrictions, opportunities and aspirations. Horgan (2012) also notes an attribution bias, which he suggests may lead interviewees to overemphasize the role of structural grievances or coercion in an attempt to diminish their responsibility. Formerly engaged individuals may also reconsider past episodes of violent engagement in a less favorable, even condemning, light. This has certainly been the case for some of my interviewees. The following excerpt illustrates this well:
Actually, the injustice I was perceiving was clearly mistaken, but in my head at the time, it was an injustice that I couldn’t bear. […] I was telling myself, regarding all that was happening, when I see people that get totally…of course it was totally biased, because those were propaganda videos and not the reality, but I was influenced by that so when I saw that the people were being massacred and all these atrocious things… But then, how could you know who started it? (Max, Switzerland, emphasis added).[1]
The ‘reworked’ and self-censoring character of the narrative is evident. On two occasions, the interviewee does not finish the sentence that refers to the specific injustice he was perceiving and instead rushes to downplay it by invoking the bias of propaganda videos or the complexity of the situation. This ongoing dialogue – between the engaged individual in the past and the disengaged in the present – makes it sometimes difficult to clearly grasp why the individual radicalized or mobilized in the first place. Nevertheless, while at times methodically disadvantageous, it is also a marker of interviewees’ self-reflectiveness, which tends to provide for rich and dense narratives.
Horgan (2012) pointed out that a question frequently faced by researchers who conduct interviews with individuals engaged in deviant or antisocial behavior: ‘How do you know if they are telling the truth?’ (Horgan, 2012, p. 200). In line with Bourdieu’s (1996) arguments above and recalling constructivist perspectives of narrative criminology (Presser 2016), in this study and more generally, I acknowledge the constructed nature of all interviewees’ narratives, shaped not only by the timing of the interview (often after their engagement), but also the physical location of the interview, the presence of the gatekeeper, their perceptions of the interviewer’s intentions and trustworthiness to name only a few. Narrative is also closely tied to self-perception and identity, as Blount-Hill explains: “ The narration process—how interviewees choose to craft the story, selecting which parts to include, to omit, to emphasize, and then supplying a ‘lesson learned’ synopsis of their tale (or not)—reveals something about their interpretive tendencies and, at least, about the self they choose to present to the researcher” (Blount-Hill, 2021 , p.15). Interviewees tend to reassemble past experiences into more or less coherent narratives, in an aspiration to create a seemingly logical sequence of events. Whenever I noticed interviewees doing this, I would note it down and include these observations in my analysis of their narratives.
Often, it should be recalled, the reasons for violent engagement are not always clear to the individuals themselves. In fact, as Akram (2014) argues, using Bourdieu’s concept of habitus, violent engagement may be the result of grievances, memories and experiences, which have been stored over a long period of time and are in some way present during the engagement, without necessarily being clearly articulated to the individuals themselves. Thus, one should be wary of believing that through interviews, a particular “truth” could be revealed. If anything, they reveal a truth – the truth of these individuals at a particular moment in time and in the specific context of the interview – that is captured and analyzed. An overall advantage in the sample in the said study is that none of the interviewees was, at the time of the interview, incarcerated nor immediately under the authority of a criminal justice institution, which avoids issues transparently laid out by Crettiez & Ainine (2017) or Bonelli & Carrié (2018), yet glossed over by Khosrokhavar (2006), such as pressures on the interviewee to participate, in hope of a favorable detention regime or sentence. Another advantage of the research design is that different views on the same phenomenon could be triangulated, such as the views of a Salafi-jihadist with that of social worker from Tripoli, Lebanon, and a researcher from Beirut. The diversity of the profiles allowed for the nuancing and balancing of inevitably positioned narratives.
Being an external bystander and not an institutional actor that individuals may be able to situate contextually was also helpful. By institutional, I do not only refer to criminal justice officials but also to actors involved in prevention work (P/CVE) who, depending on their affiliation, can also be seen as representatives of state power (for contrary evidence, notably on the perceived credibility of police officers, see Koehler et al. (2023), Thompson & Leroux (2023)). Given some interviewees’ aversion to the way “radicalization” has been demonized and criminalized, several interviewees explained that they would have refused to engage in this study had I as a researcher had any such connections. Several individuals also noted that they would not talk to a journalist, but that they did not mind talking to a researcher, provided anonymity of the account was ensured. Some of them had felt exposed by journalists in the past who had revealed their names and pictures.
- CONDUCTING THE INTERVIEW
Interview preparation, setting and modalities
Before the interview with individuals associated with PIM or PIV, I informed interviewees about my research intentions. I briefly explained the contours of the research ambition, the reasons for my being interested in them as potential interviewees and stressed the confidentiality of the interview in a message sent via E-Mail, WhatsApp (sometimes a voice message) or Facebook. These first approaches were personalized as much as possible, taking into account what I knew about the person’s background, ideology or political positioning, and making a clear link to the gatekeeper when there was one. It was sometimes helpful to enquire about the individual via the gatekeeper or by consulting their Facebook page or other information that was publicly available.
When interviewing individuals who are politically involved, sometimes even aligned with groups widely considered as terrorist, the importance of carefully preparing one’s wording and terminology cannot be overstated. When interviewing somebody aligned with Hezbollah, for instance, using the label “terrorist” or “extremist” organization can alienate the interviewee. In the view of Hezbollah sympathizers, the group is a “resistance” organization with legitimate aims. Interviewees will also likely be suspicious as to the motivation and aim of the researcher. What is their background, how are they going to use the information they provide in their research and what consequences will that entail for the perception of the organization and its adherents? Pursuing a doctorate in criminology, it was quite complicated for me to avoid conveying the perception, simply by stating my affiliation, that I was going to look at the interviewee’s story through a criminalizing lens. In one case, this problem became literally an obstacle to the mental aperture of the interviewee. The interviewee, a former member of the Iraqi popular mobilization forces (PMF), reacted strongly after he heard that I was pursuing a doctorate in criminology: “So, to you, the Hashd (Hashd al-Sha’bi, PMF) is a criminal organization?”, was his immediate conclusion, which I tried hard to deconstruct for the remainder of the interview. Hence, preparing one’s academic presentation with care will be beneficial for the researcher.
Furthermore, it is crucial to consider the aspect of temporality. There might be a discrepancy between the interviewee’s current life situation and mindset and the objective of the researcher. When interviewing someone in 2024 about their motivation to join a jihadist group in Syria in 2013, adequate weight should be given to the time that has elapsed since then, lest one risks fixating the interviewee in a past version of themselves they may not be in accordance with any longer. Disregard for this aspect can result in considerable epistemic violence and hamper the interview process.”
At the beginning of each interview, I would again present myself and the research in a personalized way. Sometimes, it even felt like pitching the research project to a potential funding body, explaining the need for the study, my motivation, the objectives, and the methods. I emphasized the confidentiality of the interview. Interviewees were asked to sign the form of consent[2], using a pseudonym, usually before, and in a few cases, where a basic level of rapport had first to be established, after the interview. In case the interviewee had difficulty reading, I or the gatekeeper would read and explain the form to them, recording their consent. They were informed that they were allowed to withdraw from the interview at any given moment, and that they were allowed to contact me using my full contact details within a year to nullify their participation. They were also guaranteed that their recordings would be deleted, upon demand within a year after the interview. They received their own copy of the form signed by me.
In the case of peripheral actors, I proceeded in the same manner. Especially since among the non-professional peripheral actors, there were many who had no experience participating in research. With professional peripheral actors, however, generally speaking, less time was invested in preparing the interviewee, explaining the contours of the research and the implications. Many of them were familiar with the process of conducting research and participating in interviews. They were sometimes more interested in the theoretical foundations of the research, the objectives and any intermediate findings. I took care to respond to their questions in the best way I could.
Interviews were conducted in the language preferred by the interviewee. Overall, the majority of interviewees were done in Arabic. In Lebanon, interviews were conducted in Arabic primarily, and in some cases in a mix between English, Arabic and French. In Switzerland, interviews were conducted in French, German and Arabic. In Canada, interviews were conducted in English, French and Arabic. Thanks to my fluency in these languages, no interpretation was required at any stage of the study.
Interviews often took place in cafés or restaurants. In some cases, they took place at the premises of the professionals I was interviewing. In some cases, where professionals were also gatekeepers, the interview with the individual (PIM/PIV) was also conducted in the premises, if possible (for example, a social worker’s office that the individual is familiar with). For most of the interviews in Tripoli, Lebanon, I was allowed to use a recording studio inside the premises of an NGO they were associated with, which provided an excellent setting for the interview and the recording. When the studio was occupied, I conducted the interview in my rental car which was parked outside the NGO.
As mentioned, the majority of interviews with the core categories, PIM and PIV, were recorded (22 out of 32 PIV interviews, and all PIM interviews). In some cases, I relied on notes, either because interviewees refused to be recorded or because of my personal appraisal of the appropriateness of recording in a particular setting. My insistence on recording shifted throughout my research stays in Lebanon. While, in the beginning, I tried to record as much as possible, I relied more frequently on notes towards the end. On the one hand, this was reasonable because I was getting closer to empirical saturation and interviews were becoming more confirmatory, not bringing up much new information. On the other hand, it was sometimes methodologically wiser because interviewees (especially in Tripoli) would speak more freely and openly when they knew they were not being recorded. For me personally, it was also a relief, because I was often uncomfortable having such sensitive material on my recorder, which could potentially get the interviewees into trouble if security forces (which were omnipresent) got hold of it. Recordings were stored in an encrypted folder to which only I had access. Within 24 hours after the interview, I wrote up a short report based on my notes, fresh memory and impressions. The quality of the summary/report decreases drastically with each day elapsing since the interview, at least for me. Therefore, it should be written as soon as possible after the interview. When interviews weren’t recorded, I wrote down the most important parts of the interview as accurately as possible based on my notes immediately after the interview.
With peripheral actors, recording was omitted more often: 18 out of 51 interviews were recorded. This is either because some of them preferred not to be recorded; or because recording did not seem necessary (since the interview was about the phenomenon more generally, and no verbatim quoting would be necessary). In case interviews were not recorded, I relied on my notes and wrote up a report of the interview within 24 hours.
Interview-based data collection is the most commonly used research technique in the social sciences and semi-structured interviewing the most frequent interviewing technique in qualitative research (Kallio et al., 2016). It remains, however, a rare source of data collection when it comes to politico-ideological violence (Schuurman, 2020). Semi-structured interviewing can also be considered as an ethnographic research method in the field of criminology and criminal justice (Shover, 2012). It is particularly popular because it allows for versatile and flexible applications across various research domains and objects of study. It is sufficiently directed to maintain a focus on the research object and flexible enough to allow for the emergence of new and unexpected themes and patterns outside the preconceived research questions (Dearnley, 2005). Semi-structured interviews are based on themes and questions that guide the interviewee towards the research object. They are usually informed by a review of the literature, or through an exploratory study or interviews with experts (Kallio et al., 2016). The questions should be formulated as open-ended questions, in order to solicit spontaneous and in-depth answers as well as personal stories that reveal something about the object of inquiry (Dearnley, 2005). Ideally, answers should also reflect personal feelings (Whiting, 2008).
For said study, as mentioned, semi-structured interviews were considered as the most appropriate type of interview-based data collection. Interviews were therefore conducted using a semi-directed interviewing technique, in which I directed questions around an interview guide map that was established beforehand. At the beginning of each interview, I asked people to tell me about themselves, where they came from and what episodes shaped them early on in their lives, using open questions such as “Could you tell me a little bit about yourself?” , “How did you become [function, member, sympathizer, etc.]?” , “How did you become politically engaged?”. The idea was not only to capture early experiences and influences, but also to understand how interviewees defined themselves, built their narrative identities, and described their engagement. What interviewees chose to say in that rather spontaneous and unrestricted opening was in itself telling, for it was an indication of aspects of their lives and upbringing that were salient (or at least seemed so) to them at the moment of the interview. From questions revolving around the How, it was possible to transition to questions more directly preoccupied with the Why: “Why did a certain feeling arise?”, “Why did certain options of engagement seem more reasonable than others?”, “Why are you still engaged?”, etc.
The interview was structured around several themes: the process of engaging in activism/violent action; collective memory and collective identity; the types of grievances experienced and their sources; moral shocks; emotions related to grievances; violent and non-violent action repertoires; justifications for violent action; personal relationship with violence; reasons for and process of disengaging from activism or violent action; vision of future engagement and personal projects. While I made sure to cover the different dimensions, the interviewee was given a lot of freedom to expand and develop his narrative, as recommended by Luna Reyes & Andersen (2003). Truthful to the iterative nature of the Grounded Theory approach, the dimensions of the interview guide were continuously refined based on emerging insights regarding core categories (see chapter 4).
Specific context-related questions were also asked because they were pertinent for the different dimensions. For example, the bombing of two mosques in Tripoli in 2013 was a hugely incisive and traumatizing event for all residents of Tripoli, especially the Sunni population which already felt targeted by the pro-Syrian Alawite forces on Jabal Mohsen. I would therefore refer specifically to this incident (unless the interviewee brought it up himself, which was often the case), to understand how they felt and reacted in the aftermath of this event. In the Canadian case, I would ask about events such as 9/11 or the protest movements against the Iraq Invasion. Some of these contextual specificities also crystallized throughout initial interviews and were therefore included in the following ones where it seemed pertinent.
Interviews with peripheral actors were structured differently, depending on the profile of the interviewee. In case the interviewee was someone who had an in-depth knowledge of a specific case or trajectory, the dimensions in the interview guide were activated as well, i.e. “were there particularly shocking events that X was affected by?”. After exhausting their knowledge of a person’s trajectory, and given their expertise built from working with many individuals in relation to this phenomenon, they were also asked general questions about trajectories of mobilisation and engagement, the role of grievances, the role of the social and political context, etc. In the cases where peripheral actors did not have in-depth knowledge of specific cases, the questions asked were only of the more general kind, regarding the phenomenon and its manifestations in the context they were most familiar with.
The professionals among the peripheral actors were also asked additional questions that were intended to reveal more personal and emotional aspects as well as ideological positionings in relation to the phenomenon. These included “What made you work in this field?”, “How has working in this field affected you emotionally?”, “How do you feel about the often superficial ways in which public opinion deals with the phenomenon, given that you have an in-depth human understanding of it”. Such questions allowed for the surfacing of important insights into the motives as well as the complexities of working with individuals involved in this phenomenon and contributed to the findings, for example in relation to factors promoting or hampering disengagement.
Understanding-based interviewing
The posture of the interviewer in research on PIV has been the object of much debate and the epistemological positions on this aspect diverge strongly. I proceeded in a fashion that arguably presents a break from Horgan’s (2012) “nonreciprocal approach” of interviewing, which suggests that the researcher asks questions, and the interviewee answers them. My approach also stands in contrast to the one adopted by Nilsson (2018) who seems to agree with Horgan in principle but argues that non-involvement is practically impossible because the interviewee will attempt to “draw [the researcher] into a discussion of politics” (p. 423). I consider the interview as an instance of sharing stories and building rapport. While of course the interviewee is the center of attention, I proactively share insights and stories from my own life that may resonate with what the interviewee is talking about. This includes showing emotions such as astonishment, sadness, or even indignation, or not withholding a laugh or a smile when the interviewee tells a funny story. My intention is to transform a formal interrogation-like situation into one that resembles more of a dialogue or an informal discussion. It is also a form of doing justice to the vulnerability the interviewees accept to expose themselves to, while getting barely anything in return. I acknowledge that the interview situation is one that benefits me primarily, since I am able to extract the raw narrative material the interviewee provides, put it through a process that generates intellectual but also long-term economic value for myself through the positive impact on my career as a researcher.
From a tactical point of view, opening up personally and being in the interview setting as a human being has proven helpful to establish rapport and create a safe space for interviewees (this also applies to practitioners, for another example, see Haugstvedt, 2020). I also encouraged interviewees to ask me anything about the research project or about me as a person. Frequently, they gladly accepted this by asking about the ultimate goal of the research project, my background, experiences, or even political views. I always responded frankly, while being careful to remain diplomatic in my answers. This openness contributed to trust-building: not only did most interviewees open up themselves, but it also allowed me to ask more personal and uncomfortable questions. On other occasions, the outcome was less positive. In one instance, I was sitting with a group of men in a room on Jabal Mohsen in Tripoli, northern Lebanon, whose residents tend to be supportive of the Syrian regime, when Miz’il asked me what I thought about the Syrian civil war and Bashar Al-Assad. Although I think my answer was nuanced, pointing out the complexity of the war situation, and the difficulty of understanding it from afar, the absence of explicit support for Assad seemed to be unsatisfactory. After that meeting, Miz’il evaded my few subsequent attempts at meeting up.
All in all, my approach aligns with Bourdieu’s (1996) philosophy laid out in “Understanding” and the larger constructivist approach of this research (see opening quote). Considering the research instance as an inherently social and human one, I try to register and analyze in situ the ways in which the positions and positionalities of the interviewees but also, crucially, mine affect and transform the interview itself.
This stance includes a proactive engagement to reduce the symbolic violence inherent to an interview situation and the constant pursuit of non-violent communication. It also requires actively and methodically listening and interacting, which corresponds neither to nondirective laissez-faire interviewing nor to the rigidly interventionist character of the questionnaire. The approach resembles what Spradley (1979) describes as “ethnographic interviewing”. Dornschneider (2021) resorts to ethnographic interviewing in her research on violent and non-violent activism. Ethnographic interviews are based on creating rapport between the interviewer and the interviewee. They are largely non-intrusive and strive to maintain a situation that seems as natural as possible to the interviewee. Ethnographic questions allow interviewees to describe their behavior or their views in a language and using a lexicon of their choice, not one imposed by the researcher.
Throughout my fieldwork, I came to realize that the interview is, though initiated and directed by me, a co-production of knowledge. An important aspect of this was notetaking. I was very surprised the first time when an interviewee referred and pointed to something I had noted down (when I myself struggle sometimes to read my handwriting). It was then that I realized that my notes can be part of the interview process, in the sense of a platform to interact and discuss. Sometimes, when I felt it was appropriate, I also explained certain theoretical concepts to interviewees and asked them whether they resonated with them, whether they had anything to say about them. In some cases, this was an astonishingly fruitful approach.
Importantly, the understanding-based approach requires a commitment to going beyond the spontaneous and obvious both during interviewing and analysis. Bourdieu (1996) provides the example of expressions of hostility against foreigners and immigrants among farmers or small shopkeepers, who often lack any direct knowledge of immigrants. Through understanding-based interviewing, the interviewer may get closer to underlying feelings and experiences that shape such strong views:
The real bases of the discontent and dissatisfaction expressed, in inappropriate forms, in this hostility can only be brought to consciousness – that is to explicit discourse – where an effort is made to bring to light these things buried deep within the people who experience them – people who are both unaware of these things and, in another sense, know them better than anyone. (Bourdieu, 1996, p. 30)
In the research on PIV/PIM, this means digging deeper when individuals express hatred for a particular group and justify violence against them. It also means actively digging deeper when the reasons for engagement seem to be reverberations of narratives that are circulating more widely (so-called “master-narratives” (McLean et al., 2018)), or when narratives draw on representations borrowed from mediatic or political discourse. The following excerpt with Thulfiqar, a supporter of Hezbollah, may be useful to illustrate this:
Thulfiqar: Look, the whole thing is about religion. Everything is related to religion. You know how Karbala is something holy. Why did the Imam Hussein die? Because he was defending Karbala, Religion. We are doing the same thing. He fought till his last breath, so we want to do the same. This fire…this heat…is still here, that we want to defend our land, as Imam Hussein told us. […]
AA: But there are certainly other things, like politics?
Thulfiqar: No, none. It’s all about religion. Of course, your country, your land. But it’s mainly religion, you’re defending your religion. Why do you want to let someone enter your land, we can’t allow that.
A superficial reading of this excerpt may give the impression that sympathies for Hezbollah build mainly on religion or religiosity. A closer look reveals that what the story of Karbala and the sacrifice of the Imam Hussein illustrate is not primarily a theological story but a political one of sacrificing oneself to defend one’s land and people. In fact, defending one’s land seems to be the primary ideology, which becomes discursively embedded in a story with a specific religious character, and overshadowed by absolutist expressions such as “the whole thing is about religion.”
- TRANSCRIPTION & ANALYSIS
For the analytical process, the study drew on a non-rigoristic form of Grounded Theory. As Emerson, Fretz & Shaw (2011) point out, conservative grounded theorists insist on an exclusively inductive logic. Contemporary grounded theory scholars, however, they suggest, no longer emphasize the pure “discovery” of theory (see also Charmaz, 2014), but recognize that
“analysis pervades all phases of the research enterprise—as the researcher makes observations, writes fieldnotes, codes these notes in analytic categories, and finally develops explicit theoretical propositions. In this sense, then, analysis is more accurately described as both inductive and deductive, what some have termed ‘retroductive’ [e.g. Katz, 1988]. The process is like someone who is simultaneously creating and solving a puzzle or like a carpenter alternately changing the shape of a door and then the shape of the door frame to obtain a better fit” (Emerson, Fretz & Shaw, p. 301).
Also, in contrast to traditionally positivist Grounded theory, contemporary theorists insist on rooting the method in an epistemology of constructivism (Thornberg & Charmaz, 2014), meaning that data generation is understood as a co-constructed process between the researcher and the researched (Birks, Hoare & Mills, 2015).
The combination of inductive and deductive analysis can be considered as abductive, while “abduction means selecting or constructing a hypothesis that explains a particular empirical case or set of data better than any other candidate hypotheses, as a provisional hypothesis and a worthy candidate for further investigation” (Thornberg & Charmaz, 2014, p. 153). The abductive approach is also chosen as a way of providing heuristics, i.e. ways of analyzing complexity in the social world without the pretention of infallible explanatory promises, but nevertheless allowing us to understand in the way suggested by Raatikainen (2017): “Understanding […] refers most commonly to understanding the behavior and action of human agents on the basis of internal motives and the reasons that guide them” (p. 8-9). Importantly, Raatikainen goes on to argue that “understanding” is not anathema to causal “explaining”, but rather “a kind of causal explanation” (p. 9).
In line with this approach, analysis of the interview started the moment I met the interviewee. This included how they presented themselves, how they reacted to me and the interview setting, and the kind of questions they asked. I made a brief note or sign whenever I had a personal observation and elaborated on this aspect in the interview report.
I continued the analysis during the transcription process. Listening to the recordings allowed me to relive some of the experiences I had during the interview and remember subtleties, impressions and emotions. Consistent with the understanding-based approach, I took care to include both the “personal voice” and the “voice of science” in the transcription process. I wrote down memos about conceptual or theoretical ideas (Birks & Mills, 2015; Walker & Myrick, 2006) but also about personal feelings and impressions within brackets in the transcript itself. I also made links to other interviews, emerging codes and categories, as well as theoretical concepts. The following examples illustrate this:
It is normal that the citizens take matters into their own hands, like the Hashd did in Iraq for example. To defend the land and liberate it. [violence as a necessity when ‘nobody protects us’]. And until now, the army is not able to protect this country. (Jaafar, Lebanon)
You can be in solidarity with the whole world, but you need to understand that there are struggles that belong to some and other struggles to others. The Swiss have a hard time understanding that [This is something that Souhail also mentions, but Richard doesn’t. I wonder whether having foreign parents (which is the case for Daniel and Souhail) might help develop a sensitivity in this respect]. (Daniel, Switzerland)
When you see a wrong being committed, you imagine what if it happened to you. To your family, to your neighbours, to your friends [vicarious victimization]. (Aziz, Lebanon)
I transcribed interviews into the language they were conducted in except for interviews in Arabic (which I translated directly into English), since they were conducted in vernacular Arabic and given that I personally am more efficient typing using the Latin alphabet. Important notions or expressions which cannot be literally translated were inserted in Arabic nevertheless (e.g. (transliterated): al-gheera, al-li’b ‘alal a’sab, tasleeh ‘atifi). Interviews conducted in Swiss German were transcribed into standard German. French and English interviews were transcribed literally. The transcription for interviews in the categories PIV and PIM was integral, unless there were longer parts that consisted of repetitions or side talk where no verbatim transcription was deemed necessary. For those parts, I either noted down bullet points or wrote a brief summary. For the category of experts, I summarized the content of the interview. Literal transcription was only used for parts of the interview that discussed specific aspects in detail.
The interview reports and transcripts were compiled into a common MAXQDA database. MAXQDA is a software package for qualitative and quantitative data analysis.[3] Coding was performed on MAXQDA. In qualitative research and, more specifically, grounded theory, coding can be seen as “iterative, inductive, yet reductive process” which consists of breaking data down and organizing it in order to “construct themes, essences, descriptions, and theories. Coding […] is what transports researchers and their data from transcript to theory” (Walker & Myrick, 2006, p. 549). Overall, transcripts were analyzed individually and collectively (using the coded segments) to identify patterns and themes, i.e., “statement[s] of meaning that [run] through all or most of the pertinent data” (Ely et al., 1991, p. 150). Throughout the in-depth examination of each interview, I also tried to maintain a theoretical sensitivity (Birks & Mills, 2015; Strauss & Corbin, 1998) by contrasting emerging codes with theoretical concepts, in order to develop the latter and get a more nuanced understanding of how they might be interacting and acting on individuals’ trajectories. The ensuing conceptual and theoretical reflections were therefore grounded in the data or, at least, repeatedly brought back to it (Emerson, Fretz & Shaw, 2011).
In line with suggestions made by Strauss & Corbin (1998) and Thornberg & Charmaz (2014), I proceeded to coding in four phases: open, focused, axial and theoretical coding. Given the large number of interviews collected over a period of two years, phases of interviewing, transcribing and coding would overlap. Consequently, the different phases of coding were not neatly subsequent but largely iterative. In every case, interview transcripts were first analyzed for themselves (vertical analysis), before proceeding to a transversal (horizontal) analysis across interview transcripts (on vertical and horizontal analysis in qualitative research, see Deslauriers, 1987).
I started out with the open coding of each interview, i.e., freely reading through the transcript and labelling the fragments of narrative, whether sentences or paragraphs, using codes that succinctly express their meaning. Coding was conducted mainly in sentence-by-sentence fashion. Where an entire paragraph contained a coherent piece of information, coding was done paragraph-by-paragraph. While advancing throughout the coding of transcripts, the open coding gradually transforms into focused coding, where certain recurring codes are used more intentionally as analytical categories to sweep through the data. As a part of open and focused coding, I also used auto-coding, i.e. coding the results of lexical searches (e.g. rage, violence, terrorism, revenge, etc.) in order to provide an overview of the frequency of specific terms, identified based on recurring themes. Each code was added to the code list and therefore available for the coding of subsequent interviews. Where codes present similarities, they were subsumed into an overarching code or category. In a second step, after transcribing and coding all interviews, I proceeded to axial coding, i.e., comparing codes with each other and categories with sub-categories, in order to organize the codes and make sense of how they relate to each other. For axial coding, as well as for the analytical process in general, MAXQDA’s code matrix was used. The matrix provides a user-friendly and clear overview of the most frequent intersections between codes. This means that when a segment (sentence or paragraph) is coded with two different codes (e.g., “grievances”, “violence”), these will be considered as intersecting. Across the data, there are pairs of codes that intersect or overlap more frequently. Frequently overlapping codes indicate some form of relationship that can then be investigated more in-depth. Another useful function of MAXQDA is the code map, which provides a visual network of the relationships between codes. Codes that appear more frequently together in a text segment which is defined by the researcher (e.g. 3 paragraphs or 10 lines) are located more closely to each other on the code map. The code map therefore allows for the identification of clusters, i.e., groups of codes that seem to have a stronger relationship between each other. Again, the code map cannot be used as a result per se, but it is very effective in guiding the researcher throughout the analytical process and indicate where analysis should be pushed further.
Finally, I used theoretical coding, which means analyzing the codes that emerge from the data in light of pre-existing theoretical and analytical concepts. As Thornberg & Charmaz (2014) suggest, “theoretical codes consist of ideas and perspectives that researchers import to the research process as analytic tools and lenses from outside, from a range of theories. Theoretical codes refer to underlying logics that could be found in pre-existing theories.” (p. 159). As mentioned, the way the interview analysis was approached in this study can be qualified as abductive. In principle, codes were built inductively from the data. However, given the effort put into building the analytical framework before starting the interview phase, I had incorporated many theoretical concepts into my thinking about the phenomenon. The literature was also reflected in my interview questions and therefore also influenced the narratives I collected. It was therefore hardly possible (nor completely necessary) to dissociate myself completely from the existing literature while coding. Theoretical coding was therefore a crucial step to organize categories conceptually to gain a level of abstraction from the empirical data and build a theoretical framework around the phenomena of PIM and PIV.
- FURTHER PRIMARY DATA
Fieldnotes
Apart from ethnographic notes relating to the interview situation itself, focusing on the aspects pointed out above and captured using memos and interview reports, fieldnotes were taken more specifically in situations of immersion where no organized interviews were taking place. During my research stay in Lebanon, I was taking fieldnotes on an almost daily basis. To take notes in situ, I often used a small black notebook that was handy to carry around. As Emerson, Fretz & Shaw (2011) point out, notetaking may impede proper immersion in a particular setting of interest. Noticing that someone is taking notes could also add to individuals’ hesitation to open up and speak freely. On several occasions, notetaking did not feel appropriate. In those situations, I relied solely on immersion and wrote down the fieldnotes from memory. Also, since I was driving a lot in Lebanon, and because I did not always have the time to sit down and write down my thoughts and impressions, I used my voice recorder instead. On the one hand, it allowed me to register my fieldnotes orally while they were still fresh. On the other hand, I was also able to trace back particular emotional states that were underlying my thoughts and impressions, which might have been less easily “storable” in written form. As soon as possible, usually later the same day, I transferred my fieldnotes to my laptop, complementing them with as many details that I could gather from memory.
Lebanon is a politically charged context, and events, episodes and encounters that may turn out to be pertinent for my research are virtually on every corner. I would use notes for seemingly mundane episodes (e.g. a discussion at the gym), for moments related to my recruitment efforts (e.g. at the military headquarter in Saida trying to get a permit to enter the refugee camp Ein-el-Hilweh and ending up talking to officers about their perceptions of the camp and the terrorist threat in it), and during intentional field immersion, which included one day in the refugee camp Ein-el-Hilweh (including the difficulty of being allowed inside) and six days of intentional immersion in Bab-el-Tabbaneh and Jabal Mohsen. By intentional I mean that I was not conducting interviews, but simply spending time with people from the neighborhood, participating in and observing their daily routines, and getting to know them outside the semi-formal interview setting. I also visited the Hezbollah Museum in Mleeta in Southern Lebanon, with a group composed predominantly of young men sympathetic to Hezbollah that I met at the gym. With my gatekeeper for ex-detainees from Tripoli, I visited a precarious neighborhood in West Beirut, where members of Islamic State or Al-Qaida were reportedly hiding. With one of my interviewees, Pierre, I attended a lecture on Lebanese identity and collective memory in a café in Beirut, given by Charles Corm, a Lebanese writer considered as the founder of the Phoenicianism movement in Lebanon, whose intention it is to create a common Lebanese identity based on their supposedly Phoenician ancestry. Finally, I made three visits to a cultural café in Beirut, which runs projects aiming to bridge sectarian divides between young residents in the area.
In Switzerland, I relied on fieldnotes in situations where no formal interview was arranged, such as professional meetings on matters pertinent for my research (e.g. meeting with a prison director regarding their efforts in terms of “radicalization prevention inside prisons”) or during conferences and workshops on counterterrorism in Switzerland. Atthe level of policymaking, significant developments were taking place while I was conducting this study, and through my professional engagements, I was able to participate in the formal and informal discussions surrounding the phenomenon and the ways in which Swiss authorities chose to engage with it. Fieldnotes were also crucial during my visits to the Swiss Federal Criminal Court, when terrorism cases were being tried. These trials usually take several days and can be exhausting. Members of the public are allowed to assist in the rows behind the various actors implied in the trial. The questions asked by the judges, the interactions between the defense attorney and the federal prosecutor, or the responses and emotions of the defendant are examples of highly valuable aspects which can only be collected by being present during these trials. The final judgement which is published online contains mostly strictly legal matters.
In Canada, I also relied on fieldnotes whenever recording of the interview was not possible. Fieldnotes were also used during and after the interview to note down observations that went beyond the mere content of the interview, regarding the posture of the interviewee, or emotions that are expressed before, during and after the interview situation. Finally, fieldnotes were used more generally during fieldwork in Canada to document my ongoing analytical engagement with fresh insights gained from encounters with interviewees.
Fieldnotes were also added to the MAXQDA database in case they were not already included in the interview transcripts. The analysis of fieldnotes was performed analogously to that of interview transcripts, in line with suggestions by Emerson, Fretz & Shaw (2011).
Instant Messaging
A tool that came in handy were instant messaging applications such as WhatsApp, Telegram or Signal. Widely and globally used, they enable the exchange of messages, voice notes, pictures and place calls. Telecommunication fees in Lebanon are relatively expensive and people frequently resort to WhatsApp calling and messaging. Unsurprisingly, a proposed tax on WhatsApp calls was one of the reasons for the outbreak of protests in October 2019. In Switzerland, WhatsApp is also frequently used for communication, although less in professional settings. In Canada, the use of WhatsApp seems to be less frequent. People seem to prefer Facebook Messenger as a means of social instant messaging. With individuals who are particularly careful about their data and communication, the application Signal was the way to go.[4]
I used both WhatsApp and Facebook Messenger communicate with interviewees and gatekeepers. In Lebanon, I frequently exchanged voice messages with interviewees, particularly helpful with interviewees who may have difficulties writing or reading. Sometimes, interviewees sent me further documentation (e.g., newspaper article about the trial of one of my interviewees) or references (e.g., videos, pictures related to things mentioned in the interview) via WhatsApp. In one case, an interviewee sent me a live picture from protests that were taking place in the refugee camp Ein-el-Hilweh on 16 July 2019, which would otherwise be difficult to obtain.[5] Finally, individuals who refused to be interviewed or were unable to put me in touch with potential interviewees often sent me a brief voice message to apologize and sometimes explain why. This was useful for my reflections on recruitment and obstacles related thereto.
In Switzerland, WhatsApp has been very useful for one of my interviewees, who I met on two occasions. I asked him, however, to tell me about his trajectory using voice notes, which he agreed to do. Since May 2019 (throughout my stay in Lebanon) I was receiving regular updates about his intimate thoughts, worries, and moments of despair. By now, the voice notes amount to dozens of hours, providing very insightful material. I transcribed parts of those voice messages that seemed most relevant for the research.
WhatsApp written or vocal messages were integrated in the analysis not as a separate corpus of data, but as a means to enrich accounts of interviewees, on the one hand, and inform reflections about methodology, on the other. To analyze the transcripts of vocal messages, I added them to the database in MAXQDA and coded them analogously to interview transcripts and fieldnotes (see chapter 4).
Autoethnography denotes, in essence, the study and documentation of self. Autoethnographers study their own life stories to reveal sociological phenomena through “privileged” knowledge that comes by lived experiences (St. John et al., 2019). Self as subject offers advantages unavailable through other methodologies. Narrative theories of identity emphasize that we understand our lives through constructed narratives of a cohesive self (McAdams & McLean, 2013). Typical methodologies capture only pieces of individual life narratives. In autoethnography, the full scope of perceived experience is accessible, including unflattering or taboo aspects typically edited in others’ responses (Chang, 2016). Adding autoethnography as an instrument allows for a triangulation of data collection, a thorough engagement with the researcher’s subjectivity and therefore a more holistic engagement with the phenomenon under study. In addition to fieldnotes, I therefore engaged in a continuous process of autoethnographic writing. The purpose of these writings has been to document the reflective and autoethnographic confrontation with the research project and object, the methodological difficulties and the personal challenges. Autoethnographic and reflective writing allows the reader to contextualize the findings presented by the researcher. Importantly, autoethnographic analysis in relation to the research method and the research object provides research findings per se, which are of particular interest to other social science researchers.
The autoethnographic data consists of notes that I took during activities immediately related to the research, such as fieldwork, transcription and analysis, in the form of fieldnotes and memos as explained above. Aspects of interviews that were deemed of methodological pertinence were coded using the label “method” – the coded segments were added to the autoethnographic material. Beyond this material that is immediately related to the study, I also wrote reports whenever I communicated orally on my research, for instance during conferences, workshops and lectures. I also included reactions to my written communication about the research, in the form of anonymous peer reviews for instance, replies by editors to submitted opinion pieces, as well as reactions to published opinions.
Finally, the study was, to a large part, a personal journey. I therefore included ongoing reflections and spontaneous thoughts about the research during unrelated activities, which I took care to always note down in my personal digital notebook (I use Evernote[6] for that purpose). In another recent project parallel to the writing of the study, I engaged in a collaborative autoethnography with a colleague of mine and compiled a considerable body of material including reflections on the research process and challenges related thereto (Ajil & Blount-Hill, 2020).
The above-mentioned material was added to the MAXQDA project and coded analogously to the main material, i.e. using open, focused, axial and theoretical coding. However, since much of the material was scattered across different documents and sometimes part of one single document that was filled with notes, analysis did not start out with vertical analysis, but immediately with horizontal analysis across the entirety of the autoethnographic material. This autoethnographic work has been a continuous process. It has helped me review some of my analysis and findings more critically and, perhaps counterintuitively, create some emotional distance by engaging thoroughly with my emotions and thoughts. It has helped me identify many of my personal biases and evaluate the ways in which they impact my work (Ajil, 2023b).
- ADDITIONAL MATERIAL
Beyond the primary and complementary data collection, I considered some accessory material for this study, namely Facebook profiles and court judgments (the latter only for the Swiss context). This material did not undergo a separate analytical process. It was used to complement and illustrate specific insights emerging from the analysis of the main corpus of data.
Social media profiles and posts were also consulted, and relevant material added to the database. Using social media profiles has become a prominent feature of research on individuals engaged in PIM and PIV. Using a Facebook profile, where I use a pseudonym, I befriended and followed most interviewees either before, i.e. during recruitment, or after the interview, when interviewees were referred to me outside of social media. When messaging a potential interviewee, I would introduce myself formally, describing the research endeavor and my institutional affiliation. After the interview, I occasionally followed their activities and posts, which were sometimes revelatory of particular thoughts and stances, as well as the causes they were involved in. Interviewees were informed that I would be following their activities on social media. In fact, for several of my interviewees, their activity on Facebook (posting on particular issues, raising awareness, commenting, etc.) was considered to be part of their engagement (e.g. Nidhal or Sami when it comes to speaking out in favor of Hezbollah).
Publicly accessible judgments on individuals who engaged in foreign fighting or were convicted in relation to terrorism-related offences, were used for the Swiss sample. The judgments, sometimes over a hundred pages long for a single case, often provide valuable pieces of information regarding the activities and the motives of individuals accused of terrorism-related offences, although seen through the lenses of police, prosecutors and judges. In the case of interviewees who had been sentenced before, the judgments were used as complementary information pieces, especially valuable to understanding the judicial background of their trajectories and the institutional lens through which they were viewed.
- CONCLUDING REMARKS
This paper presented the multipronged research protocol used in a study of politico-ideological mobilization and violence in relation to the Arab World. While mainly based on interviews with individuals directly involved in the phenomenon, expert interviews, fieldnotes, voice notes, social media profiles and judgments were included to ensure triangulation of the findings.
Beyond the technicalities, I also argue for an in-depth engagement with one’s positionality, epistemic positioning and views on the ontology of the narratives collected. This reflective work, before and throughout the research process, is crucial to ensure that knowledge production is theoretically and methodologically sound, but also clearly situated in terms of the researcher’s positionality. Especially given that this kind of research, dealing with individuals who face criminalization and are confronted with the weight of state power, can never be completely neutral and will always be part of a larger sociopolitical process of negotiating power. Research on these phenomena is “messy” and therefore demands methodological consciousness (Woon, 2013).
At the same time, choosing an understanding-based approach and an “immersed” research rapport can cause traumatic experiences for researchers which should not be understated (see, for instance, Rustemi, 2022). It is recommended that researchers remain sensitive to any impact the research process may have on them, seek discussions with colleagues and professional advice when necessary. I recall several PVE practitioners, especially in Lebanon, recommending seeking counsel in parallel or subsequently to the research process. I also advise researchers to engage in the autoethnographic writing process throughout, to make sense of traumatic experiences which may sometimes remain stuck below the discursive surface.
As the pool of potential interviewees is increasing and researchers as well as practitioners continue to work on the phenomena of terrorism, violent extremism, radicalization, and P/CVE, it is crucial to employ research methods that are thoroughly reflected upon both technically and ethically, to produce, or rather, co-produce, fresh insights useful to scholarship, praxis, and society at large. In that sense, this paper intended to provide a detailed and accessible research protocol that can be adapted to different settings.
- References
Ajil, A. (2023a). Politico-ideological mobilization and violence in the Arab World: All in. London: Routledge.
Ajil, A. (2023b). Studying Terror Through my I’s: Autoethnographic Insider/Outsider Reflections of an Arab-Muslim Researcher, Perspectives on Terrorism, Vol. XVII, Issue 2, 74-91.
Ajil, A. & Blount-Hill, K.-L. (2020). “Writing the Other as Other”: Exploring the Othered Lens in Academia Using Collaborative Autoethnography. Decolonization of Criminology and Criminal Justice, 2(1), 83-108.
Akram, S. (2014). Recognizing the 2011 United Kingdom riots as political protest: A theoretical framework based on agency, habitus and the preconscious. British Journal of Criminology, 54(3), 375-392.
Birks, M., & Mills, J. (2015). Grounded theory: A practical guide. London: Sage
Birks, M., Hoare, K., & Mills, J. (2019). Grounded theory: the FAQs. International Journal of Qualitative Methods, 18, 1-7.
Blount-Hill, K.-L. (2020). Exploring a Social Identity Theory of Shared Narrative: Insights from Resident Stories of Police Contact in Newark, New Jersey, and Cleveland, Ohio. Criminal Justice and Behavior, 48(6), 810-827.
Bonelli, L., & Carrié, F. (2018). La fabrique de la radicalité – Une sociologie des jeunes djihadistes français. Paris: Le Seuil.
Bourdieu, P. (1996). Understanding. Theory, Culture & Society, 13(2), 17-37.
Chang, H. (2016). Autoethnography in Health Research: Growing Pains? Qualitative Health Research, 26(4), 443-451.
Charmaz, K. (2014). Constructing Grounded Theory. London: Sage
Crettiez, X. & Ainine, B. (2017). Soldats de Dieu, paroles de djihadistes incarcérés. Éditions de l’Aube, Fondation Jean-Jaurès.
Dearnley, C. (2005). A reflection on the use of semi-structured interviews. Nurse researcher, 13(1), 19-28.
Deslauriers, J. P. (1987). L’analyse en recherche qualitative. Cahiers de recherche sociologique, 5(2), 145-152.
Dornschneider, S. (2021). Analyzing ethnographic interviews: Three studies on terrorism and nonviolent resistance. International Political Science Review, 42(2), 1-15.
Ely, M., Anzul, M., Friedman, T., Garner, D., & Steinmetz, A. M. (1991). Doing Qualitative Research: Circles Within Circles. London: The Falmer Press.
Emerson, R. M., Fretz, R. I., & Shaw, L. L. (2011). Writing Ethnographic Fieldnotes. Chicago: University of Chicago Press.
Haugstvedt, H. (2020). Research Note: ‘How Can I Get Them Offstage?’– Critical Reflections on a Researcher’s Approach to Qualitative Data Collection with Social Workers Involved in Preventing Violent Extremism in Norway, Journal for Deradicalization, 23, 214-234.
Horgan, J. (2012). Interviewing the terrorists: Reflections on fieldwork and implications for psychological research. Behavioral Sciences of Terrorism and Political Aggression, 4(3), 195-211.
Kallio, H., Pietilä, A. M., Johnson, M., & Kangasniemi, M. (2016). Systematic methodological review: developing a framework for a qualitative semi‐structured interview guide. Journal of advanced nursing, 72(12), 2954-2965
Khosrokhavar, F. (2006). Quand Al-Qaïda parle. Témoignages derrière les barreaux. Paris: Grasset.
Koehler, D., Clubb, G., Bélanger, J. J., Becker, M. H., & Williams, M. J. (2023). Don’t kill the messenger: Perceived credibility of far-right former extremists and police officers in P/CVE communication. Studies in Conflict & Terrorism, 1-20.
Luna-Reyes, L.F. & Andersen, D.F. (2003). Collecting and analyzing qualitative data for system dynamics: methods and models. System Dynamics Review, 19(4), 271-296.
McAdams, D. P., & McLean, K. C. (2013). Narrative identity. Current Directions in Psychological Science, 22(3), 233-238.
McLean, K. C., Lilgendahl, J. P., Fordham, C., Alpert, E., Marsden, E., Szymanowski, K., & McAdams, D. P. (2018). Identity development in cultural context: The role of deviating from master narratives. Journal of Personality, 86(4), 631-651.
Nilsson, M. (2018). Interviewing Jihadists: on the importance of drinking tea and other methodological considerations. Studies in Conflict & Terrorism, 41(6), 419-432.
Presser, L. (2016). Criminology and the narrative turn. Crime, Media, Culture, 12(2), 137-151.
Raatikainen, Panu (2017). Explanation and Understanding Revisited. In Niiniluoto Ilkka & Wallgren Thomas (eds.), On the Human Condition: Philosophical Essays in Honour of the Centennial Anniversary of Georg Henrik von Wright. Acta Philosophica Fennica, vol 93 (pp. 339-353). Helsinki: The Philosophical Society of Finland.
Rustemi, A. (2022). Ensuring security while conducting research and fieldwork on countering and preventing religious and far-right violent extremism: The case of Southeastern Europe. Journal for Deradicalization, 31, 95-118.
Schuurman, B. (2020). Research on terrorism, 2007–2016: A review of data, methods, and authorship. Terrorism and Political Violence, 32(5), 1011-1026.
Shover, N. (2012). Ethnographic methods in criminological research: Rationale, reprise, and warning. American Journal of Criminal Justice, 37(2), 139-145.
Spradley, J. P. (1979). The ethnographic interview. Long Grove, IL: Waveland Press.
Strauss, A., & Corbin, J. (1998). Basics of qualitative research: Techniques and procedures for developing grounded theory (2nd ed.). Thousand Oaks, CA: Sage.
Thompson, S. K., & Leroux, E. (2023). Lessons learned from dual site formative evaluations of Countering violent extremism (CVE) programming co-led by Canadian police. Journal of Policing, Intelligence and Counter Terrorism, 18(1), 24-42.
Thornberg, R. & Charmaz, K. (2014). Grounded theory and theoretical coding. In U. Flick (ed.), The SAGE handbook of qualitative data analysis (pp. 153-169). London: SAGE Publications Ltd.
Walker, D., & Myrick, F. (2006). Grounded theory: An exploration of process and procedure. Qualitative health research, 16(4), 547-559.
Whiting, L. S. (2008). Semi-structured interviews: guidance for novice researchers. Nursing Standard (through 2013), 22(23), 35-40.
Woon, C. Y. (2013). For ‘emotional fieldwork’ in critical geopolitical research on violence and terrorism. Political Geography, 33, 31-41.
[1] Unless quotes are shorter than two lines, they will be placed into a separate paragraph and without quotation marks, as this example shows. While the author acknowledges that it is uncommon to introduce findings in the section on research methods, the quotes were selected specifically in order to illustrate methodological specificities.
[2] The consent form was translated into English, French, Arabic, and German by the author.
[3] For further information: www.maxqda.com
[4] Given the external storage of communication data by WhatsApp or Facebook, I would, however, recommend using encrypted instant messaging services such as Signal.
[5] L. Al-Arian, July 26, 2019, In Lebanon, Palestinians protest new employment restrictions, npr, https://www.npr.org/2019/07/26/745041157/in-lebanon-palestinians-protest-new-employment-restrictions (Last Access: 6 May 2024).
[6] Evernote is a tool that can be used for various purposes, among other things taking notes, which are then synchronized between all devices connected to one’s account: www.evernote.com.
This paper was published in Journal for Deradicalization, Summer issue, 28 June 2024. PDF below.