Climate in the News: How Differences in Media Discourse Between the US and UK Reflect National Priorities

Abstract

Studies dealing with media coverage of climate change have increased steadily over the last decade or so, alongside the media coverage of climate change itself. This article aims to contribute to this growing literature on two levels: to deepen understanding of distinctive patterns of language use across nations speaking a common language and to demonstrate the usefulness of a new approach for finding such patterns. Articles in The (London) Times and the New York Times, published between 2000 and 2009, were analyzed using methods related to computational linguistics. Results show that the US seemingly still constructs climate change as a problem, whereas the UK focuses on finding solutions for the (established) problem of climate change. This linguistic and conceptual gap may hamper mutual understanding and the crafting of global climate change mitigation policies.

Keywords:

• Climate Change
• Media
• Computational Linguistics
• United States
• United Kingdom
• Carbon

Acknowledgements

We would like to thank the Economic and Social Research Council (UK) for its financial support, Alon Lischinsky for his advice on climate phrases and the Ngram viewer, and the anonymous reviewers and Jaspal Rusi for their helpful comments on matters of style and substance.

Notes

1. For continuous updates on articles appearing in the field see http://www.climatechangecommunication.org/ and http://agwobserver.wordpress.com/2010/10/04/papers-on-media-and-climate-change/ (for a site that constantly monitors media coverage, see Boykoff & Mansfield, online ; for one of the most recent contributions on the issue, see Boykoff & Smith, 2010)

2. On the many uses of the word carbon in so-called lexical (carbon) compounds, see Nerlich and Koteyko (2009a, b), Koteyko, Thelwall, and Nerlich (2010), and Koteyko (2010).

3. Previous research by Nerlich and colleagues (see Koteyko et al., 2010) has shown that the word carbon only began to be used extensively in debates about climate change around 2006. We therefore initially used two-year intervals before going for yearly extraction of data after 2006.

4. Notice that the n constituent tokens do not have to be found consecutively for the n/m-gram to match. (Specimen matches from fiction by Iris Murdoch.)

5. For a more detailed analysis of carbon footprint and other lexical “carbon compounds”, see Koteyko et al. (2010) and Nerlich and Koteyko (2009a, 2009b).

6. It should be noted that because the MDS procedure requires symmetric or near-symmetric distance matrices the underlying inter-term intervals are averaged before submission to the scaling procedure. Thus, intervals I–J and J–I become equal. Consequently any asymmetries (e.g., between carbon … trading and trading … carbon) are in effect smoothed out by this analysis.

7. For studies of this type of lexical creativity see Koteyko et al. (2010).

Additional information

Notes on contributors

Brigitte Nerlich

Brigitte Nerlich is a Professor of Science, Language and Society at the University of Nottingham

Richard Forsyth

Richard Forsyth is a Research Fellow in Computational Linguistics at the Centre for Translation Studies of the University of Leeds

David Clarke

David Clarke is a Professor of Psychology at the University of Nottingham