Alignment and scaling contrasts in Majorcan Catalan falling accents Maria del Mar Vanrell Universitat Autònoma de Barcelona mariadelmar@mallorcaweb.net In their classic study about English tonal range, Liberman and Pierrehumbert (1984) found that gradual increasing in emphasis correlated with increasing tonal range of the pitch accent, without changing its phonological make up. Nowadays, a central assumption of the standard autosegmental- metrical model (and of most work on intonation) is that pitch range variation is paralinguistic (that is, it expresses exclusively differences in emphasis or prominence). This assumption relies on a version of the so-called Free Gradient Hypothesis (Ladd 1994, Ladd 1996). It is indeed clear that one of the most common effects of gradually expanding the pitch range of a given pitch accent is the pragmatic reinforcement of the utterance, that is, an increase in the degree of speaker’s involvement in the speech act. Recently, though, some work within the autosegmental model has revealed that pitch range can also display categorical effects and steps have been taken towards the phonologization of the pitch range category (Hirschberg and Ward 1992, Ward and Hirschberg 1985, Ladd 1994, Ladd 1996. This study presents a close examination of interrogative intonation of Majorcan Catalan, which has not yet been studied in detail. The main data are drawn from prior work on Majorcan Catalan interrogative intonation (Vanrell 2003), Map-Task recordings, and my own speech. The article will demonstrate that while in some cases tonal alignment is the main cue to distinguish between two types of interrogatives, in others, pitch scaling is the main distinctive cue. For example, the main difference between a neutral yes-no question (1) and a yes-no question with surprise/complaint (2) is relative alignment: while in (1) Te n’anaves? Did you leave?, the low tone (L) is aligned with the stressed syllable, in (2) Que te n’anaves? Did you leave?, it’s the high tone (H) that is aligned with the stressed syllable. (1) Te n’anaves? (2) Que te n’anaves? H+L* L-L% L+H*L-L% The following are two examples where scaling of tonal features show clear categorical effects. (3) is an example of a neutral wh-question and (4) is an example of a reiterative wh-question. Leading tones mark the difference between (3) Com ho duries? How would you carry that? and (4) Com ho duria How would I carry that?. In (4), the leading tone H is upstepped. (3) Com ho duries? (4) Com ho duria? H+L* L-L% H+L*L-L% The same basic difference in pitch height is found in another pair of pitch contours, namely, yes-no neutral questions and echo questions. In the case of echo questions, the leading tone H is upstepped. In the paper, I will present a phonetic study of the Majorcan contours found and the results of perception experiment tasks that will tease out the role of scaling in distinguishing the relevant type of interrogative. Within the ToBI framework, I will claim that in order to account for scaling differences in Majorcan Catalan interrogatives we need to resort to upstep features affecting leading tones in falling accents.