ABSTRACT

A review of the group of studies included under the label frequently used by their authors-"the eHects of music"-has led us to the observation that the paradigms of the experiments and the kinds of questions they pose influence the results obtained. That review did not show that the subjects in those experiments, left to their own spontaneity, would have formulated judgements of signification in their minds, although the typological classifications they made would lead us to think so. What came out of our review was that the great majority of subjects lent themselves to such a formulation, and that, incidentally, the judgements were consistent. "Abstract" significations were, from all the evidence, associated with processes in which the musical form was the source of the thing signified; those judgments did not appear, in the way they were stated, to depend on individual personalities. They were not, however, just "responses to stimuli" in the sense of deriving from automatic perceptual reactions. The signifying musical pattern could have been the point of departure for normative or empirical judgments if the semantic attitude had been for some reason excluded by the subject. Musical signification is thus not identical to signification with words or images, which derives directly from the univocacy of attitude that those engender. We can doubtless take some interest in the typographic presentation of a text, or in the composition of an image, and thereby become distracted from their significations. But signification is still immediately given, before all other considerations. We might similarly suppose that semantic judgments apply to music only as the last links in a process that, in order to remain in some cases nonverbal, is nevertheless no less real and effective at the prereflexive level. To the extent that the musical signification is potential, that potentiality necessarily resides in a psychological reality that remains latent, in the form of

a vague impression, as long as the subject's attitude does not require an explicit verbal response of it.