Presentation Script

Slide 1) Research Skills Project by Group 9. Conducted by Megan Lawton, Kathryn Myers, Hayley Yates, Vicki Haworth and Tabitha Dunn

Slide 2) After some complications with exactly where our focus was going, we came up with a research question that best defined what we were aiming to find out from our project: How do Creative Insults Vary in their Linguistic Features? Focussing Specifically on Twitter. The term ‘creative insults’ was one we felt appropriate in covering all of the phrases we would be looking at.

Slide 3) Our initial inspiration for the project developed from a keen interest in the cases of high profile internet abuse reported recently in the media. We felt that we could examine these cases from a linguistic point of view, focusing on how the language used in the ‘troll’ posts caused offense. However, due to difficulties surrounding data collection and applications of the literature, we later opted for the focus of ‘creative insults’, with the intention of comparing and contrasting the different linguistic features found in several cases of ‘trolling’ on the social networking site twitter.

SLIDE 4) In order to collect the data for the research project, each member of the group individually researched a high profile twitter incident, where an insult/threat of some sort occurred, in many cases the tweet/series of tweets were considered so severe/offensive they were taken to court. There was one exception amongst the case studies, in the form of the comedian Frankie Boyle. We analysed several of Frankie’s tweets purely for a point of comparison amongst the others, as we believed his insults are intended to be humorous and not hurtful. Linguistically we were intrigued to find out what deems a tweet offensive and humorous. 

After each member had chosen their case, we each read various articles, blogs and pieces of academic literature relevant to the research question, for example Labov’s  ‘rules for ritual insults’, the texts enabled us to create a linguistic analysis/comparison between the tweets.

Slide 5) Case study one looks at tweets sent from @Riley_69 during the Olympics the tweets were a mix of direct threats, insults and responses to the media attention.  The 8 separate tweets are just a small number out of the many that could have been analysed but these give a good example of the creativity and variation that insults possess.  The celebrity status of the victim and the severity of the language created a media hype that spanned tabloids and broad sheets alike. The police became involved due to the threatening nature of some of the tweets but no charges were bought. It is interesting to note that more media attention was put on the tweets that contained other forms of dysphemism other than threats.

SLIDE 6) Case study two unlike the other 4 case studies consists of one tweet only. The tweet, written by Paul Chambers in January 2010, is a threat aimed at ‘Robin Hood Airport’, the tweet declares Chamber’s intentions of “blowing the airport sky high”.  This is an example of an impoliteness output strategy (as defined by Anna Holmberg) that aims to ‘frighten’- in other words the tweet has the potential to make others believe an action will occur that is harmful. The impoliteness output strategy of ‘Frightening’ flouts the maxim of relevance. Chambers was taken to court under the Malicious Communications Act 2003, however was cleared after it was argued messages sent in jest, regardless of the subjective view of the appropriateness or quality of their humour should be taken as such. Holmberg provides an interesting perspective that insults are down to interpretation, and this seems heavily the case with this tweet. It’s also worth considering sarcasm, blatantly insincere in nature, but effective in causing havoc if interpreted as the author did not wish.

Slide 7) Unlike the other Case Studies analysed in the research, Case study 3 focuses on one main victim of abuse, the ex-England footballer and sports pundit Stan Collymore, and analyses four racially provoked tweets personally directed at him by three unrelated authors. The two most prolific of these tweets were sent by @JoshuaCryer1. Cryer addressed Collymore with stereotypically racist language, using the Twitter convention of ‘hashtagging’ in order to sadistically summarise and reiterate his abuse, changing his form of register to promote humour. Cryer’s was  charged under section 127 of the Communications Act for sending grossly offensive messages, and ordered a two year community order and fine. 

Slide 8) As Holmberg’s thesis demonstrates, looking at completely different types of online insults is a useful way of gaining an insight into just how many different ways people can creatively insult others; something that became a prime focus from early on in our research methods. Case Study four focussed on the tweets from Student Liam Stacey, who wrote various tweets about footballer Fabrice Muamba just moments after the footballer had suffered a near fatal heart attack and was fighting for his life. For the threatening tweets, Stacey was sentenced to 66 days in prison, with the conviction that the tweets were posted intentionally to cause harassment.

Slide 9) Case Study 5 was chosen with the intention of being a comparison against the four other studies. Whilst all the other  ‘creative insults’ we looked at were from members of the public who were originally unknown, and only came to be known as a result of their insults on Twitter, this study was of Frankie Boyle – a well-known comedian.  Labov suggests that ritual insults should be responded to using the special forms appropriate to the insult exchange, and not as if they were intended literally, with the knowledge that the content of the insult cannot in fact be true. This suggestion of Labov provided an interesting stance to look at the insults of Frankie Boyles. The main intention was to see how they differed because they were written with the intention of being humorous – is it clear that ‘the content of the insult is not in fact true’, and if so how does the language differ in order to portray this?  Boyles tweets, are generally not aimed at anyone in particular, but tend to be a humorous remark about something current in the news. For example: ‘Just found out I’m the only non-child molester in Britain and it’s all some mad paedo show.’ (around the time of the Jimmy Saville case).

Slide 10) Sentence Types: We analysed the individual case studies by sentence structure to try and see if one kind of sentence type was more common than the others. The easiest way to do this was to look at whether the 4 main sentence types – imperatives, interrogatives, declaratives and exclamatory – were present. Imperatives were found in 3 of the studies, interrogatives in only 1, declaratives in all of them and exclamatory sentences in 2.

Slide 11) As our results from the table show, four out of five case studies report extensive use of individual lexis, with the only exception being Frankie Boyle. Reasons for this may include the fact that as a comedian, his purpose is slightly different in that offending people is not his primary focus, whereas the other case studies demonstrate the intent to offend.

The results of the focus on the subject aspect demonstrate how tweets more often than not were directed explicitly to the subject in question. The focus on the subject in the case of Frankie Boyle, however, provided a stark contrast with that of the others. The general nature of his twitter page shown in the examples on the hand-out demonstrates how having a specific subject is less likely to occur.

Our analysis of typical insult employment helps to evaluate the levels of common cultural reiteration in the tweets. Three out of the five case studies featured typical insults ranging from stereotypical racist remarks to common offense turns of phrase. The employment of ‘your mum’ in case study four parodies on the common playground joke; such use of language certainly reduces the formality of the register in the tweet, however in context does not reduce the seriousness of the message related.

All of the case studies in this research demonstrated various aspects of negativity. Case studies 2 and 3 employed threatening language to convey their messages, whilst case study 4 was filled with extremely offensive lexis, all of which reflected negativity. In case study 5, negativity is implicated at, however, appears less offensive than the other cases. This is reflective of Frankie Boyle’s stylistic dark humour which is able to shock, but not particularly offend the audience.

Slide 12) Summary of Results: With regards to sentence type, it was declaratives that appeared the most throughout all of our examples. As a group, we concluded that this might be due to the nature of social networking – twitter in particular. As the concept of Twitter is to state your thoughts, with a limit of 140 characters, people are most likely to be declaring a thought they are having, be it insulting or not. Again, this might explain the lack of interrogatives we found. When people are creating an insult they are generally aiming a horrible statement at someone, without the need to invite response. A particularly interesting result was how explicit offensive lexis was used throughout all the cases except Frankie Boyle; who tends to create an insulting or explicit semantic implication of his tweets, as opposed to using explicit language within his creations. A final thought was that typical insults seemed to be used  often as an attempt to create humour through an insult. Phrases such as ‘your mum’ and ‘you look like a monkey’ are often used, in these context it seems, as a way of the author or the tweet seeming funny; perhaps a method of being more successful concerning the response of the insult – how others perceive it. 

SLIDE 13) Overall the project ran smoothly and we are happy with the final outcome, however we did encounter some minor problems throughout the term. We struggled to define our research question, particularly in relation to linguists. Initially the original study was centred around internet trolls, however we found this topic too broad, and we were carried away with our personal interests. At one point we formed a questionnaire but found the answers wouldn’t benefit our study in any way, at this point we refined our research question, focusing on creative insults on the social network of twitter, with particular emphasis on their linguistic features. After finding a criteria for features commonly used in creative insults, we could then easily analyse the case studies and find correlations, similarities and differences amongst them, we now feel satisfied that our project has reached a conclusion and we have gained knowledge from the study.

If we were to extend the project we would focus on the origin and culture of typical insults, paying interest to where they emerged from, for example rap battles.  We also feel a comparison could be made between various social networks where insults are concerned, does the nature of twitter invite a certain form of insult in comparison to Facebook for example?

Slide 15) As a group, we feel the credits of this project should be spread equally between us. We were efficient in organising weekly meetings, to which all members attended without fail.  At the meetings, all of us made equal contribution to the task, and, particularly when we came to having problems with the focus of our research question, everyone put an equal amount of effort into helping get back on track. As well as the blog, our group created a Facebook group, which worked well in ensuring all members knew about any changes to meetings, as well as an easy method of posting files and passing thoughts onto each other, before they were ready to posted onto the blog.

 Slide 16) Thank you for taking the time to listen and watch our presentation. We are happy to answer any questions you may have.