In Search of a European Identity: a Multimodal Discourse Analysis of the 2014 European Parliament Election Campaign

Following Baldry and Thibault's micro-analytical approach to MDA (Baldry and Thibault, 2006), Table 3.1 is based on seven vertical columns: (1) Time; (2) Verbal Text; (3) Visual Frame; (4) Visual Image; (5) Kinesic Action; (6) Soundtrack; and (7) Metafunctional Interpretation: phases and subphases. Each vertical column corresponds to different modes – "the outcome of the cultural shaping of a material" (Jewitt, 2009: 24) referring to the functional choices people make across different semiotic resources simultaneously. Moreover, each column is related to a horizontal row which has "important integrating and cross-referencing functions" (Baldry and Thibault, 2006: 174), as explained below.
Column 1 refers to the time, in seconds, of the video recording. Choosing not to adopt the timing used by Baldry and Thibault, that is frame/second (f/s), we have selected one salient frame for each shot, leaving aside those characterized by the same multimodal combination in order to avoid redundancies. The time indicated in this column, therefore, refers to the duration of the shot selected. The corresponding horizontal row is "temporally correlated" (ibidem) to the extent that each row provides a description of "the semiotic modalities [co-patterning] at a particular moment in the text" (ibidem).
Column 2 refers to the Visual Frame selected for each shot. As explained above, we chose to reproduce one visual frame for each shot in contrast to Baldry and Thibault's second-to-second approach, mainly for economic and practical necessities, providing nonetheless a high level of accuracy. In the Act. React. Impact. video, instead, every frame lasts on average 3 seconds; shots are slow in the first phase, slightly accelerate in the second and slow down in the third. We decided, in point of fact, to select frames according to their "perceptual salience" (ibidem: 183), choosing one frame for each shot, thus providing a representation of resource integration in each dynamic sequence. Perceptual salience, here, refers to the indissoluble bond which ties the meaning of a text to the observers or participants. Text is, therefore, a meaning-based interactive event, in Halliday and Hasan's terms (Halliday and Hasan, 1985: 26), where participants' subjective interpretation actually activates functional meanings in discourse. Assuming that multimodal analysts are also participants in the communicative event, according to Baldry and Thibault multimodal analysis is never theory-neutral. Indeed, multimodal interpretation presupposes a process of making assumptions about the meaning of a text which may, nevertheless, facilitate the identification of the most salient "foregrounded copatterning of selections" (Baldry and Thibault, 2006: 183). As a result, although methodological criteria are needed in order to determine the consistency and the accuracy of the analysis, subjective criteria of selection lead us to focus only on the most salient features, according to their relevance to the main purposes of this study. [...]