Vol. 39:1 (2022) ► pp.39–87
Persistent innovations and historical conspiracies as reanalysis and extension
This article explores temporally extended innovations as a form of phonological reanalysis and extension. Polabian (West Slavic) exhibits multiple dissimilatory innovations that target the reflexes of Late Common Slavic (LCS) *v/w. In this paper, I propose that the outputs of syllable structure changes in early West Slavic were reinterpreted as restrictions on the distribution of [w], thus introducing dissimilation to the language. The new grammar was not able to stop other innovations from occurring (prophylaxis) and instead restructured intermediate outputs from subsequent innovations into an acceptable phonetic form (repair) thereby extending dissimilation to new items. I demonstrate that (a) outputs of an earlier onset epenthesis grammar conform to the surface-true generalizations enforced by the reanalyzed dissimilation grammar and (b) a single grammar can account for both the dissimilation which developed during the West Slavic period and subsequent extensions which occurred after Polabian was fully differentiated from its relatives.
Article outline
- 1.Introduction
- 2.The Polabian segment inventory
- 2.1The segments of Polabian
- 2.2Feature representation
- 3.The structure and timing of Polabian dissimilation
- 3.1Comparative and internal evidence of dissimilation
- 3.2The diachronic sequencing of Polabian dissimilation
- 3.3Summary of dissimilation
- 4.Labial dissimilation from a cross-linguistic perspective
- 4.1Types of dissimilation
- 4.2Does a process confine the analysis or does the analysis confine the process?
- 5.Towards a unified account of Polabian dissimilation
- 5.1Change in historical OT and phonological reanalysis and extension
- 5.2Modeling dissimilation in agreement by correspondence
- 5.3Polabian grammar models
- 5.4Summary: The structure of West Slavic dissimilation
- 6.Closing
- Acknowledgements
- Notes
-
References
https://doi.org/10.1075/dia.19034.bur