Narrative voice in popular science in the british pressa corpus analysis on the construal of attributed meanings

  1. Pérez Veneros, Miriam
Supervised by:
  1. Izaskun Elorza Director

Defence university: Universidad de Salamanca

Fecha de defensa: 26 June 2017

  1. Jorge Arús Hita Chair
  2. Blanca García Riaza Secretary
  3. Lise Fontaine Committee member

Type: Thesis

Teseo: 485980 DIALNET


This dissertation approaches the construal of attributed meanings in the dissemination of science in the British press by analysing the resources used by journalists to integrate in their narrations of scientific findings what other sources have said. Attribution in scientific, academic and media discourse has been previously described from an interpersonal viewpoint through the analysis of evaluation and appraisal. In addition, research has also addressed how experiential elements such as the source, the process (verbal and/or mental), and the structure (direct speech or indirect report) contribute independently to the construal of attribution. However, the approach followed in this dissertation attempts to provide a more comprehensive description of how attribution is construed experientially. The assumption made is that in polyphonic texts it is possible to analyse attribution comprehensively, by identifying to whom we can attribute each idea which is quoted, or reported, or narrated in the text. For this purpose, an analytical unit (‘unit of voice’) has been defined and studied, which distinguishes between external sources of attribution (‘attribution’) and the journalist’s voice (‘averral’). The aim in this dissertation is to explore how the experiential elements construing attribution co-occur in each unit of voice and contribute to the journalist’s interaction with his/her readers as well as to his/her epistemological positioning with respect to the attributed information. In order to obtain both quantitative and qualitative data, a corpus consisting of 180 popularizations has been compiled and 1,625 units of voice have been identified from the perspective of systemic functional linguistics. In addition, an annotation scheme for the comprehensive analysis of the three crucial components (processes, participants and logico-dependent relations) of the units of voice has been proposed. Results point to the fact that the texts in the corpus show a high degree of polyphony, due to the fact that in the 1,625 units of voice, the cases of attributed information almost double the cases of information averred. When taking the analysed features of attribution in isolation, results suggest that attribution is construed in these texts mainly through a balance between reporting and quoting, through neutral projecting processes, and through Human participants. These results correspond to traditional expectations pointing to the objectivity of the journalists in science dissemination, and seem to suggest that the journalist represents his/her mediation role from an invisible or almost invisible position. However, the analysis has also revealed that, within the unit of voice, the often complex intertwining of attribution and averral shows sometimes an ambiguous blurring between the voice of the journalist and the voice of the external source of attribution, which seems to suggest that the journalist also positions him/herself as literally aligned with the external source, by making both voices literally undistinguishable. In addition, the processes used by the journalist for projecting what others have said are varied, also including stance processes which the journalist uses to construe his mediating role in a more visible way, not really showing his/her personal views or opinions on the narrated information, but rather contextualising and interpreting its significance for readers, which is consistent with the pedagogic function expected from these texts. Results of the projection clusters considered show that journalist tend to construe the sources of attribution by labelling them either by their proper name or by their professional role when quoting them, whereas when reporting what they have said journalists show a much higher preference (up to one third of the total) to refer to material sources (e.g. the report, the study, etc.) instead. Preference is also shown to use projecting processes for quoting which are neutral together with participants construed as Human Named, versus a higher tendency to rely on stance processes when the journalist is reporting, for which they rely more often on the construal of participants as Human Semi-named. The comparison of these shows a clear difference on how the journalist represents his/her mediating role in each case, by not showing any kind of mediating presence in the case of quotes, to presenting a sounder presence as mediator in the case of reports. Finally, the journalist’s mediating role is also construed through embedding, particularly through the use of nouns of projection, which construe the journalist’s mediation as packaged and, therefore, not open to question, and which can be linked to a more prominent role on the part of the journalist in the control of the information narrated. This experiential account of the construal of attribution in science popularizations shows, in sum, that the intertwining of attribution and averral in the text is used by the journalist to construe a representation of the scientific findings narrated which relies on a mediating role of the journalist in his/her aim to guide lay readers along the narration which is essentially much more dynamic than previous accounts have shown.