There has been a long-standing controversy over the extent to which metaphor serves an emotional or a cognitive function. Mac Cormac (1985) recently proposed a model predicting, among other things, that semantic anomaly is a major determinant of the emotional impact of a metaphor. Thirty subjects were asked to rate semantically normal and semantically anomalous metaphors in terms of the extent to which they felt a subjective emotional impact. Although the experimental prediction inspired by Mac Cormac's model was supported, three interpretations of this obtained effect were judged plausible. First, as stated by Mac Cormac, increasing semantic "distance" between the tenor and vehicle of these metaphors may have had an increasing emotional impact. Second, the emotional impact of the metaphors may have been due to cognitive properties that are intrinsic to the literal sense of the metaphors. Third, the emotional impact of the metaphors may have resulted from cognitive and linguistic properties intrinsic to the vehicle. It was concluded that further study of the emotional impact of semantic anomaly should involve a procedure whereby the literal and figural senses of the metaphors are equalized (or manipulated) in terms of the intrinsic anomaly (or strangeness) of each sense, taken separately, and whereby the tenor and vehicle are also equalized (or manipulated) along several dimensions, such as familiarity, length, variability, imageability, and proportional presence of adjectives and nouns. (Claude Braun)