Comparing verbal and visual rhetoric: – and the danger of “empirical shorthands”

This paper will evaluate the quantitative content analysis of visual rhetoric
which has been put forward by George Rossolatos in the context of audiovisuality and marketing research. Rossolatos bases his taxonomy of visual rhetorical figures on Groupe μ’s adaptation of classical rhetoric in Rhétorique génerale (1970).
With the aim of creating a systematic model for marketing analysis, he and his
collaborators have coded and classified an extensive number of audiovisual commercials from the top international strata of economically valuable brands. For comparison, I have during a period of four years given undergraduates the task of collecting and categorizing a minimum of ten audio-visual commercials, each of which should exemplify one of the figures referred to by Rossolatos. The outcomes of the task indicate that a random explorative study with a minimum of operationalization and training of coders is enough for obtaining results akin to those reported by Rossolatos. This may be regarded as a positive outcome from a pedagogical point of view, but it may also raise the suspicion that Rossolatos’s criteria are merely intuitive.
If the incidence of visual rhetorical figures is to be studied as an independent
content variable of audio-visual enunciations, the values/figures should be described with greater caution.