On June 23rd 2016, the British people will vote on whether to remain in or to withdraw from the European Union. The announcement to hold the referendum was announced by PM David Cameron in January 2013.
In this talk I want to compare and contrast the reactions to the referendum proposal in two English newspapers, the left-leaning Guardian and the right-leaning Daily Mail. In the first stage, two corpora were compiled, each containing all the articles in the two newspapers whose headline or leading paragraph contained the items eu OR european union OR brussels OR frankfurt for the years 2013, named respectively DM13_eu and GN13_eu. These were downloaded from the Lexis Nexis database. Another similar pair of focused corpora were compiled from 2005, named DM05_eu and GN05_eu in order to conduct a diachronic comparison to examine whether the newspapers’ stances have altered over this time period as a result of changing circumstances, especially the Eurozone crisis. In a second updating stage, two more corpora were compiled of articles from the two newspapers which contain the terms eu OR europe AND referendum in the first three months of 2016, named respectively DM16_eu and GN16_eu. These corpora were compared and contrasted with each other using the WordSmith key-item tool which produced lists of words and short phrases (or ‘clusters’) which were more frequent in one data-set than another. This affords a window into the relative particular preoccupations of each paper at these particular times.
Observations from the research include that British EU-scepticism - and EU-enthusiasm - come in various shades and varieties. Anxieties over the EU have changed over time (especially if we reference Teubert’s 2001 seminal work on EU-sceptic discourses in 2000). The data show clearly that it is not simply a right versus left issue, and to divide viewpoints into just two camps - a pro-EU and anti-EU one - is hugely simplistic. But the problem with Referendums is that they demand a ‘Yes’ or ‘No’ response, and shades of opinion cannot be envisaged.
The principal endeavour of Corpus-assisted discourse study (CADS) is the investigation, and often comparison, of features of particular discourse types, integrating into the analysis techniques and tools developed within corpus linguistics, shunting between statistical quantitative overviews of data and traditional close reading, often of segments of data identified as potentially significant by the overview. The aim of the CADS approach is the uncovering, in the discourse type under study, of ‘non-obvious’ meanings and patterns of meanings, that is, meanings which might not be readily available to naked-eye perusal (Partington, Duguid and Taylor 2013).