Answers are grounded in this paper and cite their sources.
This article challenges the principle of strict parallelism in Optimality Theory by providing evidence that Duke-of-York derivations, deemed to be impossible by McCarthy (2002), exist in phonology. An analysis of two independent fragments of Polish phonology, chain shifts in velar palatalization and labial fission, shows multiple Duke-of-York effects because segment inventory constraints posit conflicting requirements for the well-formedness of outputs at different depths of derivation. It is concluded that Optimality Theory must permit constraint reranking and admit three derivational levels: two lexical levels and one postlexical level.
No takes yet. Share an insight, caveat, or question.
Jerzy Rubach (2003) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: