Reading the Counter-Jihad Discourse

Team Members

Steven Boegborn, Matthew Elworthy, Joseph Jackson, Charly Landman, Kieran Lewis

Introduction

In general, it is fair to assume that scholars on ‘counter-jihadism’ are familiar with the claims and arguments presented in the canon of popular books on this topic (what this study shall hitherto refer to as ‘counter-jihad’ literature). Authors of these texts are increasingly using “anti-Muslim rhetoric” to garner support for their beliefs and as Nick Lowles explains, with reference to his ‘Hope not Hate’ campaign; the ‘counter-jihad’ movement now represent “the new face of the political right” (2012). The movement is characterized by its broad and diverse alliance of people and ideas, making the ‘counter-jihad’ discourse a particularly difficult topic to study. By producing detailed reports on the movement’s dominant actors, Lowles’ ‘Hope not Hate’ tries to monitor the conversation and fight the rhetoric that in many instances leads to general ignorance and a violent hatred of Islam as a faith and Muslims as a people (Lowles 2012).

What is arguably just as important as understanding the rhetoric itself is understanding which of these rhetorics resonate most with the ‘silent supporters' of the movement, i.e. the readers of ‘counter-jihad’ texts. In order to investigate this further, this study aims to explore how the ‘popular highlights’ from Amazon Kindle users might be appropriated to understand the ‘counter-jihad’ discourse beyond the contents of the canon. Amazon’s Kindle website allows visitors to query books and authors for popular highlights made by readers using Amazon’s popular e-book reader. These highlighted text fragments are (anonymously) made available by Amazon as part of a backup service that allows Kindle users to save their personal notes and highlights in the ‘cloud’, i.e. Amazon’s server farms (Amazon 2013b).

Central to a study into the frequently marked text fragments in ‘counter-jihad’ literature is the assumption that highlights signal something meaningful, at the very least to the individual reader in which this study takes a specific interest. This study argues that Amazon’s ‘popular highlights’ are markers of importance because they indicate what readers of ‘counter-jihad’ literature personally discern as important, or at least memorable. Or as Amazon puts it, “every month, Kindle customers highlight millions of book passages that are meaningful to them” (“FAQ” 2013). Indeed, following the website’s tagline “read, review, remember,” what is arguably the main purpose of marking digital books – which already enable full-text search – is to create a personal collection of the ‘best bits’ from all of your favourite books (both academic and fictional) for later retrieval or reference. This includes the arguments that resonated with the reader - those they found to be most convincing. In aggregate, these highlights might tell us more about how readers of these texts ‘read’ the ‘counter-jihad’ discourse.

Such reader insights may add to the general understanding of what Peter Knoope, director of the International Centre for Counter-Terrorism in The Hague, calls the motivational drivers of radicalization (2012). Citing 375 pages worth of quotes in his 1,518-page manifesto, Norwegian terrorist Anders Breivik was clearly motivated by the ideas of fellow extremists writing about counter-jihadism (Lowles 2012). According to Lowles, “their writings had an important impact on the creation of [Breivik’s] political mindset,” possibly even inspiring his fatal killing spree (ibid.). Searching, finding and adopting extreme beliefs are at the base of the radicalization process, say NYPD analysts Mitchell Silber and Arvin Bhatt (16).

Following this definition of the radicalization process, it is important to not only understand which beliefs can be found in ‘counter-jihad’ literature but also which of these beliefs readers find particularly compelling. Put differently, what readers ‘think of beliefs’. For Allard Feddes, finding methods for studying the interactions between the different factors in the radicalization process, which includes the dynamic between ‘counter-jihad’ texts and its readers, represents one of the key challenges in the study of extremism (2012). The Internet may provide new courses for studying the widespread “anti-Muslim rhetoric” which, according to Lowles, “poisons the political discourse, sometimes with deadly effect” (2012). Through an analysis of popular text fragments highlighted by the readers of ‘counter-jihad’ texts, this study aims to add to the analysis of the lesser-known side of the ‘counter-jihad’ discourse. Additionally, through the Amazon Kindle system, we are exploring a totally new research method for gaining insights into the types of reasoning that resonate with the readers of a particular book, oeuvre or genre.

Research Questions

Q1: As explained in the outset, this study is not so much interested in the actual content of ‘what’ is being read but rather in ‘how’ the texts are being read, as indicated by the popular text fragments highlighted by Kindle users. More specifically, what this study is interested in are the types of reasoning that resonate with the readers of ‘counter-jihad’ literature. Presenting itself as “more mainstream and respectable [than the] old racial nationalist politics of neo-Nazi and traditional far right parties,” books on this topic are widely available, even in e-book form (Lowles 2012). Our study sets to explore what insights can be gained from the readers of these texts.

Additional Questions

Q2: With retrospective reference to our overall sample can we conclude that the texts we gathered were both sufficiently representative of, and related to other works from within the canon of ‘counter jihadist’ literature.

Q3: Working in accordance with the Digital Methods Initiative, how can our research assist in the development of a tool for further exploration of the Amazon Kindle popular highlights data? As an outcome of our study, what would we require this tool to do and how could it be best incorporated into future research?

Methodology

Before we could begin our research process we first needed to decide upon an initial collection of texts suited to our research demands. Said list was obtained from the website Jihad Watch – a popular entry point online for the counter-jihad movement. Notably, the website is affiliated with both the David Horowitz Freedom Center, a prominent organisation within the US Islamophobia network (Center for American Progress). It is run by the author Robert Spencer, perhaps the most well-known author within the field of ‘anti-jihadism’, with two of his publications making it onto the New York Times Bestseller list. Using this convenience sample of 44 recommended books on jihad and dhimmitude (appeasement to Islamic demands), we set out to investigate which rhetoric readers of these books find particularly compelling. We approached this question by querying each book title on Amazon’s Kindle website for ‘popular highlights’, as the search feature is called. For each search query, Amazon returns a selection of the most popular highlights for books and authors matching that query, typically about a dozen or so. We compiled a research corpus of all these highlights, additionally taking note of the book’s title, author and the number of readers that have marked that particular text fragment. After collecting the available highlights for all 44 books, we classified our research corpus according to an emergent coding scheme.

After we collected and analysed all the “popular highlights” from our sample of 44 books, we conducted a small-scale case study in order to determine the significance of this sample in relation to the canon of “counter-jihadist” literature. As a starting point for this research we used the most highlighted book from our corpus, Ayaan Hirsi Ali’s Infidel (2008). This book is ranked number 17 in the Kindle Store category “Leaders & Notable People”, ranked number 28 in the “Political Leaders Biography” category and has spent thirty-one weeks in the New York best sellers list (Amazon, 2013).# The book is described as a biography and a criticism on many aspects of Islamic religion (Ibid.). Taking the popularity and the sales rankings on Amazon into account, it can be argued that Infidel is part of the ‘counter-jihadist’ literature canon. We wanted to know to what extent other authors and books from our corpus would relate to Hirsi Ali and the book Infidel. Therefore, we set out to examine whether the following four most highlighted books are related (i.e. related according to Amazon when visiting the product page for Infidel) to Hirsi Ali’s Infidel. The aforementioned books being America Alone (Steyn, 2008), The Politically Incorrect Guide to Islam (Spencer, 2005), The Truth About Muhammad (Spencer, 2007) and While Europe Slept (Bawer, 2007). The tool that would allow us to assess the ‘degree of relatedness’ was developed and provided by the Digital Method Initiative (DMI). This tool aggregates the books that are related according to Amazon, when visiting the product page of a specific book. The tool was set to capture the related books to the third degree. Gephi was used for visualizing the network. We choose the “Author” as partition parameter and the “Degree” as ranking. The results are discussed in the findings.

Findings

Querying 44 Jihad Watch-endorsed books for public highlights, we compiled a research corpus of 288 popular highlights. Fifteen books did not return any highlights, possibly due to their unavailability on Amazon’s Kindle reader. Another explanation is that these books did not have any passages that were marked by multiple Kindle users. In order to be considered in the ‘popular highlights’ results, text fragments need to be marked by three or more users, Amazon explains (“FAQ” 2013). The text fragments found for the other books from our sample, 29 in total, were highlighted by anywhere from three (Amazon’s baseline) to 390 readers. Aggregated, the 288 highlights in our corpus represent almost seven thousand independent text markings. Note that the number of unique readers (i.e. the contributors to our research corpus) is likely to be considerably lower than seven thousand, given that Kindle users may highlight multiple text fragments per book.


Following our emergent coding scheme, the results stand as follows;

Table of Results

It became evident that a large portion of the collected highlights could be appropriated to support the argument that Islam is a violent religion. One out of four highlights can be found to be (implicitly) supportive of this general claim. The ‘Islam is violent’ category was indicative of highlights that described Islam in such a manner, pertaining to the supposed ‘militance’ of Islam as a religion and the nature of its people. The ‘Islam is ridiculous’ category was indicative of highlights that seemingly ‘scorn’ Islam and its practices, commonly undermining the peculiarities of Islamic history, the religion in general, and its people. ‘Islam is oppressive’ was suggestive of highlights that refer to any apparent oppressiveness against human rights within the Islamic belief system and/or nations. This included highlights that made specific reference to the treatment of women and people alike, as well as the notion of a ‘submissive’ people. The ‘Inferior’ category was indicative of highlights that note or define examples in which Islam is portrayed as inferior to Western nations, examples including their progression with technologies, ideologies and overall societies. The ‘Intolerance’ category included highlights that include notations of suggestive ‘bigotry’ within Islamic beliefs and/or nations. The ‘Dangerous’ category included highlights that made note of possible ‘threats’ to Western nations on behalf of Islamic beliefs, with these taking form as mentions of terrorism on behalf of Islam, as well as quotes regarding the proximity of Islamic ideologies to Western nations. The final category encompasses all highlights that incorporate a ‘Nationalist Argument’, and encompasses examples that are directed towards Western nations with regards to the ‘acceptance’ or lack of action towards Islamic beliefs and/or nations, i.e. social criticism of social-democratic states.

Interrelatedness

Starting with Ayaan Hirsi Ali’s Infidel (2007) (the most highlighted book from our corpus and arguably a canonical work in ‘counter-jihadist’ literature), we set out to determine whether the following four most highlighted books were related according to Amazon. Effectively, we set out to determine the representativeness of our sample in terms of a ‘counter-jihadist’ canon, by examining the five most highlighted books. The number of highlights made by the Kindle-users (in Infidel) seems to resonate with the rankings of Hirsi Ali’s book on Amazon. The readers (i.e. the readers using the Kindle) of Hirsi Ali’s book produced an aggregated amount of 2556 highlights. Considering the total amount of highlights from our corpus, a total of 6854, the highlights from Infidel make up for 37,3% of the total amount of highlights. The five books combined make up for a total of 5108 (74,5%) highlights. Therefore, these five books in turn can be said to be representative of our own corpus. After we visualized the output derived from the Digital Method Initiative tool, we could analyse the author network of Hirsi Ali. Figure 1 shows the full network (to the third degree) of Hirsi Ali’s network, based on the book Infidel. All four books were identified within this network. The (arrow marked) central node in the middle of the network represents Infidel. More importantly, all the four nodes (representing the four mentioned books) were clustered around the centre node (see figure 2). One could say that the graph shows the ‘closeness’ or rather ‘degree of relatedness’ in Hirsi Ali’s network for the book Infidel. Therefore, we could establish that Hirshi Ali’s Infidel and the four other authors and their books are all related to one another within the third degree. Accordingly, the four books show a (relatively) strong degree of relatedness to one of the bestselling books (Infidel) on the subject matter, that being ‘counter-jihadist’ narratives. Therefore, when asking whether the sample is representative for the canon ‘counter-jihadist’ literature, one can argue that the authors and books are closely related in the overarching network of this particular canon of work.

Figure 1: The red-marked node in the centre of the graph represents Hirsi Ali’s Infidel.

Figure 2: A close look on the cluster of nodes surrounding Hirsi Ali’s Infidel. Marked by red arrows are the five most highlighted books from our sample.

Discussion

Compiling a Narrative

When we consider all of our gathered highlights as a unified whole the narrative that appears reflects that which has been seen in much of the rhetoric surrounding the counter jihad discourse, both in terms of the counter jihad movement itself and the counter-terrorist discourse employed by government agencies. This is somewhat unsurprising, as Kundnani highlights that official security narratives, jihadist and far-right narratives mutually reinforce one another (10). The most compelling arguments, according to Kindle readers, constructs a narrative that paints Islam as at odds with (Islam is inferior), and a threat to (Islam is dangerous), Western liberal ideals. Islam is depicted as totalitarian in nature, and accordingly is seen as oppressive -- specifically to women, as well as to the western notion of the individual (Islam is oppressive). Through its reliance on strict interpretations of holy texts, such as the Quran and Sunnah, and their resulting structure of Sharia law, the religion is portrayed as both inherently violent (Islam is violent) and absurd, or worthy of mockery (Islam is ridiculous). In combination, such perceptions enable the argument that, not only is multiculturalism an unacceptable outcome, but is a passive cause of the Islamification of Europe (Nationalist arguments).

Resulting Tool

In essence, the highlights from our corpus represent a user-generated, or 'reader-authored', if you will, collection of quotes. This offers new opportunities for quantitative content analysis, which now can be filtered according to this qualitative selection. Rather than having a script tell us which words appear most frequently in counter-jihad literature, we can now assess which topics resonate with the readers of these texts, as indicated by the frequency of highlights. However, the application of a qualitative coding scheme would be extremely difficult, if not impossible, to do using code. Therefore, the resulting tool -- developed in continued coordination with the Digital Methods Initiative - takes a much more quantitative approach for organising Kindle popular highlights around potential themes. The tool scrapes all popular kindle highlights for the users desired input search terms, displaying as an output a tag cloud ordered according to each word’s highlight frequency, along with their number of highlights. Most interestingly and indeed most useful for potential research demands, each word is clickable, displaying all highlighted quotations that contain the word within its body. Highlights therefore become organised according to specific themes, enabling the analysis of narratives across publications. As previously mentioned, the categorisation of highlights by this tool can be considered a quantitative, rather than qualitative approach, relying on the frequency of words rather than subjective perception of the overall themes of the highlighted text.

Further Reflections

A rich irony of the net is that large sets of so-called ‘public data’ aren’t actually available to the public, or in this particular case; scholars of extremism. These sets of data are provided by the public, typically the users of a ‘free’ service, but are kept private after the data has been exchanged (for a service, e.g., saving notes about books in the ‘cloud’) (Andrejevic 2009, 418). Amazon’s public highlights are a textbook example of this asymmetric data exchange, sharing only a dozen-or-so highlights per queried book, even when it is obvious that further highlights do exist. This limits the potential of Amazon as an indicative source of what readers of ‘counter-jihad’ literature specifically discern as compelling, beyond the book as a whole.

Notwithstanding the Amazon-imposed limitations, studies into the highlights made by Kindle users can still add to the study of extremism online. Literature studies may provide insights into the contents and prototypical reasoning found in ‘counter-jihad’ literature. The research method proposed here may add to such studies by highlighting, what rhetoric resonates with the readers of these texts. This understanding of the ‘counter-jihad’ discourse, the reader’s view on things, may prove valuable in countering extremist claims, exposing them for the ignorant foundations on which they are so often based. The reader insights may, for example, indicate where claim debunking can be most effective (e.g. with claims concerning the alleged violent nature of Muslims).

Conclusion

To conclude, it is important to stress the non-obstructive and unobtrusive qualities of this study’s approach to understanding the counter-jihad discourse. Contrary to insights gained from the analysis of forum messages, tweets and users comments found on social media, Amazon’s public highlights do not represent new opinions but rather what might be called ‘meta-opinions’, that is, ‘opinions’ underlined by readers. Moreover, as readers highlight books for personal reasons, these insights are not skewed by expectations of peers or other observers. Put differently, these highlights do not contribute to an online personality based on ones own media consumption, in the manner most commonly seen on social networking sites, and one which has been discussed within media theory since the internet’s early days (Turkle 1995). Highlights made on Kindle devices are public by default; it is an opt-out backup feature (Amazon 2013b). Accordingly, many readers may not be aware of the latent audience of their highlights. Anonymity is still provided, however, as Amazon only discloses the number of highlights per quote. Because of the statistical improbability of multiple readers highlighting the same piece text without discerning that fragment as important, analysing popular highlights might act as an accomplice to traditional techniques for obtaining reader insights, such as conducting interviews. In doing so, Amazon may add to the understanding of the counter-jihad discourse.

References

Ali, Wajahat et al. Fear Inc.: The Roots of the Islamophobia Network in America. Center for American Progress. 2011. 17 January 2013. <http://www.americanprogress.org/wp-content/uploads/issues/2011/08/pdf/islamophobia.pdf>.

Amazon. “Frequently Asked Questions.” Amazon. 2013. Amazon.com Inc. 22 January 2013. <http://kindle.amazon.com/faq#PopularHighlights0>.

Amazon. “Organizing Your Kindle Keyboard Content”. Amazon. 2013[b]. Amazon.com Inc. 17 January 2013. <http://www.amazon.co.uk/gp/help/customer/display.html?nodeId=200504380>.

Amazon Kindle. 2013. Amazon.com Inc. 15 January 2013. <http://www.kindle.amazon.com>.

Andrejevic, Mark. “Exploitation on YouTube: Contradictions of User-Generated Labor.” The YouTube Reader. Eds. Pelle Snickars and Patrick Vonderau. Stockholm: National Library of Sweden, 2009. 406-23.

Feddes, Allard. “Studying the Radicalization Process in Social Media.” Studying Extremism and Counter-Jihadism Online. Digital Method Initiative. Amsterdam, 14 January 2013.

Jihad Watch. Robert Spencer, 2003. 15 January 2013.
<http://www.jihadwatch.org/recommended-books-to-understand-jihad-dhimmitude.html>.

Knoope, Peter. “Radicalization and the Internet.” What does the Internet add? Studying Extremism and Counter-Jihadism Online. Digital Method Initiative. Amsterdam, 14 January 2013.

Kundnani, Arun. “Blind-Spot? Security Narratives and Far-Right Violence in Europe”. International Centre for Counter-Terrorism - The Hague. 2012.

Lowles. Nick. “Counter-Jihad Report: Introduction.” Hope not Hate. 24 August 2012. Searchlight Information Services Ltd. 21 December 2012. <http://www.hopenothate.org.uk/counter-jihad/>.

Silber, Mitchell, and Arvin Bhatt. “Radicalization in the West: The Homegrown Threat.” NYPD Intelligence Division. 2007. The New York City Police Department. 18 January 2013. <http://sethgodin.typepad.com/seths_blog/files/NYPD_Report-Radicalization_in_the_West.pdf>.

Turkle S. (1995), Life on the Screen: Identity in the Age of the Internet, New York: Simon & Schuster.