The Hansard hazard: gauging the accuracy of British parliamentary transcripts

Abstract

Large databases of transcribed speech, downloadable from the Internet, are a corpus linguist's dream. They turn into a corpus linguist's nightmare, however, when the transcriptions are not linguistically accurate. In this paper I assess the suitability of the Hansard parliamentary transcripts (200 million words, downloadable) as a corpus linguistic resource, comparing a sample of the official transcript to a transcript made from a recording of a House of Commons session. The findings are that, as could be expected from earlier research, the transcripts omit performance characteristics of spoken language, such as incomplete utterances or hesitations, as well as any type of extrafactual, contextual talk (e.g., about turn-taking). Moreover, however, the transcribers and editors also alter speakers' lexical and grammatical choices towards more conservative and formal variants. Linguists ought, therefore, to be cautious in their use of the Hansard transcripts and, generally, in the use of transcriptions that have not been made for linguistic purposes.

The 385+ million word Corpus of Contemporary American English (1990–2008+)

“…First, are they accurate -do they accurately reflect what is found on the original audio or video recording? (This is a serious problem with transcripts of the British parliament, as discussed in Mollin 2007). Second, are they really spontaneous (as would be hoped), or is there too much scripted material?…”

The Transcription and Representation of Spoken Political Discourse in the UK House of Commons

“…This could indicate that the editors are including more of the original utterances in Hansard compared with 10 years ago which supports the suggestion in the previous section that the report has become less formal in its style and more faithful to the original spoken discourse. This is further supported when we consider that Mollin (2007) analysed a variety of sessions in the House of Commons most of which were more "sedate" affairs than the "rowdy" PMQs studied here. The nature of PMQs means that the discourse is likely to contain performance features which will be removed eventually from Hansard.…”

“…108-109) according to Slembrouck which enables institutionalised ideology and power to be manipulated through the use of language. Mollin (2007) found that the Hansard report tended to omit "performance characteristics of spoken language" (2007, p. 193) such as false starts, repetitions and reformulations as well as "parliamentary speech" that is characteristic of the specific setting (such as the cry to "order!" from The Speaker).…”

“…The site makes clear that Members' words are "edited to remove repetitions and obvious mistakes, albeit without taking away from the meaning" and is an indication that Hansard does not pretend to be what it is not-i.e., a full Conversation Analysis transcription of speech. The transcripts are made live by a number of reporters working in shifts and these transcripts are then handed on to editing staff who compile the day's report to be published the next morning (Mollin, 2007). The transcripts are thus "thrice removed from the original" (Mollin, 2007, p. 189).…”

“I entirely understand” is a Blairism

“…A further problem that needs considering is that of the transcribers. As Slembrouck (1992) and Mollin (2007) have shown, the Hansard staff do not only transfer speech with all its performance characteristics into a correctly written text, but also alter MPs' lexical and grammatical choices in line with a certain house style. However, Mollin's (2007) study also shows that maximisers and maximiser collocations are not generally subject to changes by the Hansard transcribers and editors -only absolutely is sometimes omitted (but not substituted with any other maximiser), probably in order to reduce hyperbole.…”