Chomsky's latest linguistic work has problematized projection and labeling. With the earlier Phrase Structure Grammar and X’‐bar theory, it is taken for granted that a phrase contains a specifier, head, and complement; the current work only assumes a labeling algorithm to meet requirements of the conceptual‐intentional interface. Such an algorithm automatically rules out particular configurations, for instance, ones with a specifier unless they share certain features. In this brief paper, I point out how regular patterns of language change can be seen as resolutions to the labeling paradox and how they themselves can possibly shed light on the precise nature of this labeling algorithm. I focus on the change where specifiers are reanalyzed as heads.
Stored only when both papers are in the corpus — 40 citations exist upstream.