Intonation, the modulation of voice pitch, is essential for communication as it conveys information that helps listeners make inferences about the pragmatic intent of the speaker. Despite increased understanding of intonation’s importance, there is little agreement even about essential aspects of its structure and meaning. This is in large part because research has focused either on the form of intonation, often taking a reductive approach to meaning, or has concentrated on meaning but without full scrutiny of form. Crucially, most research has eschewed the study of intonational variability, seeing it as a problem, rather than a natural facet of speech production that needs to be understood and accounted for. Examining all three aspects in tandem is critical for understanding how intonation is structured and functions in communication: considering meaning in the study of intonational form (i.e. phonetics and phonology) can help delimit intonational categories and uncover the limits of within-category variability; in turn, a robust understanding of form will lead to insights into intonational pragmatics. The present proposal will take exactly this integrative approach, based on the PI’s recent research, to examine intonational phenomena attested in English and Greek that have vexed researchers for some time (uptalk, high accents, question tunes). Two varieties per language will be studied, Standard Southern British, Bristol English, Standard Athenian, and Corfiot Greek. Their systematic differences with respect to the phenomena under investigation will allow me to examine cross-linguistic differences, and dialectal variation and its role in communication. The investigation will involve phonetic and pragmatic analysis and modelling, followed by series of behavioural and neurophysiological experiments. Together, these methods will shed light onto the realization, structure and function of intonation, and lead to a robust model of intonational phonology and pragmatics.