Duration of Onset Consonants in Gay Male Stereotyped Speech Sean Crist (University of Pennsylvania) To date, there has been only one phonetic study comparing the speech of gay and straight males. Gaudio (1994) found that listeners are largely accurate in identifying the sexual orientation of male speakers from recordings of a read passage of neutral content, although the specific phonetic cues involved in this identification are not known. Gaudio made several kinds of analyses of the F0 patterns of gay and straight men, but did not find significant differences between the two groups. The present study concerns the duration of certain consonantal segments. Six adult male speakers of American English, including both gay and straight men, were asked to read a short story of neutral content, once in an "ordinary" voice, and again in "the most flaming, exaggerated gay male stereotype you can do." (It should be emphasized that the cues which speakers employ in producing stereotyped speech are not necessarily the same cues which actually distinguish gay and straight male speakers in ordinary speech; however, if there is a connection between the two, then one might expect these cues to be more easily identifiable in exaggerated speech.) Durations were measured for selected consonants occurring in the onset of word-initial stressed syllables. The /s/-frication of /sp/ and /sk/ clusters was found to be significantly longer in the stereotyped speech for all speakers except one. /st/ clusters do not participate in this pattern. A similar significant lengthening in stereotyped speech was found for word-initial /l/; however, not all lexical items with word-initial /l/ participate in this pattern. In particular, the instances of /l/ in looks like are consistently very short and show no systematic differences between stereotyped and non-sterotyped speech. Implications for further research are discussed. For example, it is not yet known whether the artificial manipulation of the duration of /s/-frication in /sp/ and /sk/ clusters, or of /l/ in the appropriate lexical items, is sufficient to induce a change in listener judgment of the sexual orientation of the speaker. Also discussed is Gaudio's negative finding in the comparison of gay and straight mens' F0 patterns. Gaudio's finding was surprising in light of frequently cited intuition (e.g. Zwicky, in press). A possible, testable explanation for this difference between intutition and measurement is discussed, relating the present study of consonant duration with existing work relating F0 contour with perceived segment length (van Dommelen, 1993). Gaudio, R. 1994. Sounding Gay: Pitch Properties in the Speech of Gay and Straight Men. American Speech 69.1:30-57. Van Dommelen, W. 1993. Does dynamic F0 increase perceived duration? New light on an old issue. Journal of Phonetics 21:367-386 Zwicky, A. In press. Two Lavender Issues for Linguists, in Kira Hall and Anna Livia (eds.), Queerly Phrased. Oxford Univ. Press.