John Sinclair lecture: “Primed for Violence? A corpus analysis of jihadist discourse”

It was a great honour to give this year’s Sinclair Lecture at Birmingham University. I have long been an admirer of John’s work – there are many ideas he developed that are well worth critically engaging with. So to be asked to give a talk in his memory and honour was a challenge I happily took on. The topic of the talk I chose carefully – John liked ground breaking work and was a producer of daring and new ideas. So I thought an off-the-shelf piece of work was not right for this talk – it was more in keeping with the event to give a talk on a piece of work in progress. Having made that decision I then knew I should talk on the work I am developing on language and violence.

Language and violence is, in my view, a terribly under-researched topic. It is also an area which, sadly, has on-going relevance to human society. More positively, it is a topic on which linguists can – and to some extent do – provide insights. This talk was given in that spirit. I aim to show, as one would rightly expect of the ESRC Centre for Corpus Approaches to Social Science, the insights that a corpus approach may provide to an issue which reaches across the social sciences and beyond. In doing so I had to work with some quite challenging data. I think, however, that the results are at least indicative both of how this area of research may be opened up and how  linguists may contribute to its exploration. I think John would have liked this bold new venture and so I feel very comfortable in dedicating this talk to his memory.


‘How are people persuaded to be violent? How might a small group of people influence members of a larger group of people to behave in ways that they may normally find abhorrent? This talk looks at these questions, which are typically summarised as ‘radicalization’, using the example of jihadist language.I will explore how language may be manipulated in order to legitimate violent acts against certain groups or individuals in jihadist materials. However, I will also be exploring the important claim that there is a direct link between what the jihadists write and what other Muslims write, an assumption held by policy makers, academics and the media.

This talk examines how we look for linguistic evidence of this process, with an emphasis upon incitement to violence. If there is evidence that the manipulation of language in jihadist writing leads to a corresponding adaptation in either the Muslim mainstream media or the writing of ordinary Muslims over time, then we may begin to accept and understand with some linguistic sophistication what is at the moment assumed by many. We may also, however, be able to see how such radicalization is resisted, and hence better understand the process of resistance to radicalization also.

Central to my account of incitement to violence are the linked ideas of collocation and lexical priming. Together these begin to explain, I will argue, both the rhetorical process around incitement to violence and the broader dynamics in discourse that alienate and leave open to persuasion sections of society that may be persuaded to undertake violent acts.

My exploration is based on tens of thousands of words of corpus material, including i.) transcripts of so-called ‘martydom’ videos; ii.) texts by those who exhort jihadists to acts of violence; iii.) muslim news media and iv.) comment data from the Muslim news media. By drawing upon a range of sources like this, I will be better able to characterise the competing forces being brought to bear as different groups try to influence mainstream Muslim discourse.’