Cycles of Violence: Analysing media discourse in the newspaper reporting of bicycle users and road fatalities

New Voices Awards 2021

Thesis title : Cycles of Violence: Analysing media discourse in the newspaper reporting of bicycle users and road fatalities
Country : UK
University : University of Westminster
Year : 2020
Research supervisor : Rachel Aldred

1. What is your research topic? What thesis are you defending?

My dissertation investigated how cyclist and pedestrian deaths are depicted in newspaper reports, and how these reports help shape understandings of cycling, walking, driving and road safety. This topic forms part of an expanding research area looking at how media reporting shapes the public understanding and political priorities of road safety. We know from this research that cyclists and pedestrians are often described using details irrelevant to the collision, and that collisions themselves are commonly presented as isolated incidents (often ‘accidents’) rather than as part of a wider road safety issue ¹. These differences can have measurable effects on how readers assign blame and responsibility ². Consequently, media reporting may play a role in shaping how people perceive walking and cycling as forms of transport and how they interpret calls for improvements to walking and cycling conditions. In this way, reporting may influence the public acceptability of policies aimed at transitioning towards low-carbon transport.

But how are cyclists and pedestrians depicted differently as individuals and as part of a group in newspaper reports, and how does this compare to the depiction of driving cars? What do these differences tell us about how ideas of road safety, danger, and responsibility are communicated to readers?

2. If your thesis/dissertation involves empirical research, what does this consist of?

I examined a sample of newspaper reports of road fatalities in the London Evening Standard using a form of Critical Discourse Analysis called the ‘Social Actor Model’ ³. The articles were selected by a multi-stage process. First, I carried out a pilot search of articles in order to identify the different types of article that report on road fatalities, such as single event reports, multi-event reports, follow-up coverage and articles featuring editorial comment. From this typology, I chose to examine articles that represent the day-to-day reporting of crash events, as these articles follow a similar format to each other and the discourses presented in them are likely to be regularly repeated.

The final article selection consisted of identifying articles of the above type that reported on one of three different scenarios – cyclists killed in collision with car or van drivers, pedestrians killed in collision with car or van drivers, and pedestrians killed in collision with cyclists – and which were published between 1st January 2012 and 31stDecember 2019. Where the number of articles was large – as in the case of the first two scenarios – six specific articles were chosen at random across this time span. In the case of the third scenario, only five articles were found in total, and so all of these were used for analysis.

The Social Actor Model used for this analysis has not to my knowledge been previously applied to research on this topic. Previous research on newspaper reporting of road collisions has largely utilised content analysis, and where critical discourse analysis methods have been used they have been based on different models. What is particularly original about applying the Social Actor Model to this topic is how this approach focuses on the participants of the stories – the ‘social actors’ – which in this context are the bicycle riders, car drivers, and pedestrians. It is the differential representation of these social actors according to their travel mode that is suggested as an area for study by the existing literature.

The method itself consisted of close readings of the newspaper reports to identify patterns in the language used to describe and construct these social actors, their relative prominence in the reports and whether or not readers were encouraged to empathise with them. I examined whether and how characteristics of road users and the wider road environment were characterised as ‘relevant’ to the collision taking place – for example, whether the article drew attention to what a pedestrian was wearing, even if this had no bearing on the collision. I also examined whether the newspaper reports presented the collisions as part of a wider road safety problem or as isolated incidents.

Details of selected articles

3. What are the main findings?

There were four main findings. Firstly, newspaper reports about people killed whilst cycling include references to other previous collisions – such as ‘it is the third cyclist death in London this year’ – whilst those about people killed whilst walking did not include such references. In other words, pedestrian deaths were depicted as isolated incidents rather than as part of a wider road safety issue, even though there are more pedestrian than cyclist fatalities in London.

Secondly, references to previous collisions in reports about cycling fatalities were focused upon the involvement of a bicycle. References such as ‘the cyclist's death takes to seven the total number of cyclists killed on London's roads this year’ or ‘the cyclist's death comes just over a week after another cyclist (was killed)’ only make links to a growing tally of cyclist deaths. They do not make connections using other potentially important similarities, such as the type of road, presence of a junction, provision or absence of adequate infrastructure. Whilst the newspaper reports did therefore highlight a wider road safety issue around cycling, this issue was primarily depicted as being about the bicycle and its rider, rather than the road dangers imposed by the environment in which they were riding, or the behaviour of other road users.

Thirdly, the newspaper reports tended to describe the collisions themselves as if all parties involved possessed equal physical power – for example that ‘a cyclist has been killed after a crash with a van’, rather than that the larger and faster party collided with the slower and more vulnerable one. This was particularly apparent when the victim was riding a bicycle and had the effect of downplaying the greater power of – and danger posed by – motor vehicles involved.

Finally, car drivers involved were mostly referred to indirectly, for example ‘a man in his 30s died last night after being hit by a car’. This had the effect of making the driver of the car appear as a passive third-party whose involvement is distanced from the collision, and further focused questions of safety on the bicycle rider or pedestrian. The exceptions were where the drivers were associated with some other offence such as failing to stop, for example ‘Police were today hunting a suspected hit-and-run driver after a cyclist was killed’. In these cases, the drivers were depicted as part of an exceptional group of ‘rogue’ drivers.

Examples of sentences found for different socio-semantic categories

4. What is your contribution to theoretical and policy debates?

Building upon previous research, my dissertation confirmed the tendency for newspaper reporting to downplay both power differences between different road users and the actions of human drivers. However, whereas previous research found that both bicycle rider and pedestrian deaths tended to be reported as isolated incidents, my study found that these newspaper reports did include references to previous collisions when describing bicycle rider deaths. This was unexpected and may suggest that the newspaper concerned is trying to improve its reporting. It may be influenced by the involvement of media specifically in debates around cycling safety in London ⁴ .

However, pedestrian deaths were still reported as isolated incidents, while the references to previous collisions involving people on bicycles did not highlight the sources of danger involved, risking reinforcing beliefs about cycling being in itself inherently risky (whereas most cycling fatalities involve motor vehicles). I would suggest that unless newspaper reports communicate the dangers posed by the unequal physical power of motor vehicles compared to people walking or cycling, public understanding of these dangers and political will to address them will be limited. My findings have already contributed to a consultation on the development of media reporting guidelines for road collisions in Britain ( https://www.rc-rg.com), with international interest in this.

10 Media guidelines for reporting collisions ⁵

1. At all times be accurate, say what you know and, importantly, what you don’t know.
2. Avoid use of the word ‘accident’ until the facts of a collision are known.
3. If you’re talking about a driver, say a driver, not their vehicle.
4. Consider the impact on friends and relatives of publishing collision details.
5. Treat publication of photos with caution, including user generated footage or imagery.
6. Be mindful if reporting on traffic delays not to overshadow the greater harm, of loss of life or serious injury, which could trivialise road death.
7. Journalists should consider whether language used negatively generalises a person or their behaviour as part of a ‘group’.
8. Coverage of perceived risks on the roads should be based in fact and in context.
9. Avoid portraying law-breaking or highway code contravention as acceptable, or perpetrators as victims.
10. Road safety professionals can help provide context, expertise, and advice on broader issues around road safety.
    
    

5. What questions have arisen from your research that could be addressed in the future?

In using the ‘Social Actor Model’, my approach was different to most previous research in this area. Since I looked at one newspaper published in London, questions arise as to whether newspaper reports published in other newspapers – and outside of major cities and in other countries – display similar or different characteristics when examined through the model I used. For example, to what extent might newspaper depictions of bicycle riders’ safety be different in countries such as the Netherlands, where more journalists and readers are likely to be regular users of bicycles? My research identified differences between how bicycle users are depicted compared to pedestrians in newspaper reports of fatal collisions with people driving cars, and this raises questions as to how collisions between pedestrians and bicycle riders are depicted. Whilst I included such collisions, the number of fatalities and hence newspaper reports were very low with no consistent patterns found. An analysis of other types of media coverage of such cases – such as opinion pieces and reports on court cases – might help to answer this question.

Download

Download the full thesis