No CrossRef data available.
Published online by Cambridge University Press: 06 May 2021
This article questions the prevalent account of North Germanic tonogenesis, which proposes that at the outset, Accent 2 was characterized by a double-peaked melody close to the one found in central Swedish today (Riad 1998, Kingston 2011). The spreading patterns observed in the data analyzed here are difficult to reconcile with this hypothesis. My analysis instead offers support in favor of the alternative hypothesis that the phonetic roots of the accentual contrast are to be found in a difference in timing between single peaks, specifically, peak delay in plurisyllabic domains, but not in mono-syllables due to lack of space. The variation observed in the single peak Dalarna varieties today, from robust timing differences in the south to absence or only partial implementation of the tonal contrast in the north, strongly suggests that the accentual contrast has been spreading northwards through incremental peak delay in Accent 2 words. I argue that this situation mirrors the initial stages in the development that, through additional peak delay, eventually resulted in a double-peaked Accent 2 melody in central Scandinavia. At the same time, the older single peak patterns are still retained in Dalarna and scattered around the geographical margins of Norway and Sweden.*
My warmest thanks go to the speakers from East Mora, Malung, Skattungbyn, Sollerön, Vinäs, and Våmhus, who generously let me record them in their homes. Gunnar Nyström put me into contact with them, and without his deep and extensive knowledge of the Dalarna dialects and his generous help and advice along the way this article could not have been written. I also owe Eva Olander special thanks for helping me with the recordings in 2008, Jan K. Hognestad for valuable comments on an earlier version, and Kari Elida Eriksen for drawing the map in figure 3. Finally, I thank the two anonymous reviewers for valuable comments, which have greatly improved the article.