Representing Spoken Political Discourse in the Digital Era: Can accurate and faithful semantic representations be obtained from Hansard transcriptions of Prime Minister’s question time in the House of Commons?

: The UK House of Commons is a political arena that is notorious for rowdy exchanges and political point scoring. This is particularly noticeably during the weekly session of Prime Minister’s Questions (PMQs) whereby members of the chambers are allowed to question the Prime Minister of the day who is expected to give direct and frank replies. Analysis of this session however suggests that it is anything but a question-and-answer session and that it often degenerates into a mud-slinging match between politicians and parties in an attempt to raise parity issues and set political agendas. Not only do members have the tradition of parliamentary privilege, which allows them to articulate any thought or idea however inflammatory without fear of legal redress, but much of the force of the exchanges comes through covert, and sometime overt, threats to positive and negative face.

The Hansard transcripts are produced as the official written records of the PMQs session and the media often makes reference to these records in interpreting policy and generating news stories. However, research has shown (Mollin 2007) that these transcripts are not always faithful to the original utterances and the force of the exchanges is often understated. In particular, the transcripts sometimes make significant lexical and grammatical changes to the original discourse and omit performance characteristics altogether (Mollin 2007: 187). This can limit the degree to which semantic representations can be faithfully made. It can also limit the degree to which accurate secondary reporting can be made and increases exponentially the potential for the misrepresentation of policy and ideas within the media.

In the run up to the 2015 UK general election, the accurate and faithful transcription of spoken political discourse in the media is a necessity if fairness and accountability are to be preserved in society. In this paper, we will discuss the implications of the Hansard transcriptions to the representation of discourse particularly in light of the availability of digital techniques to enhance representation. Two main questions will be asked: do the modifications to the original discourse and the lack of performance characteristics seriously affect the representation that is made by the reader and hence the media? Do politicians relying on Hansard transcripts to quote other members of parliament (MP) misrepresent the original utterances? If the answer to any of these two question is ‘yes’ then this suggests we require an augmented system of representing parliamentary discourse in the digital era.

Reference
Mollin, S. (2007) The Hansard hazard: gauging the accuracy of British parliamentary transcripts. Corpora, Vol. 2(2), pp. 187-210.