Séminaire de Recherche en Linguistique

Generative Grammar has undergone a paradigm-shift from its early emphasis on Universal Grammar, e.g. Chomsky (1965), to a focus on factors not specific to the Faculty of Language, e.g. Chomsky (2007; 2015). The latter factors are known as third factor principles and “have the flavor of the constraints that enter into all facets of growth and evolution.... Among these are principles of efficient computation” (Chomsky 2007: 3). Third factor principles include Minimal Search, Labeling, Determinacy, and Economy and can be seen at work in specific syntactic structures and restrictions on them. In this talk, I argue that they can also be detected in how language changes. For instance, labeling requirements pressure against {XP, YP} merges and the need for determinate derivations eliminates superfluous movement.

Language change involves a cyclical dynamic between economy and elaboration or, as von der Gabelentz (1901: 256) put it, language history moves between comfort and clarity. Innate, third factor principles bias the acquisition of a specific grammar in economical ways but external factors, such as pragmatic strengthening, can complicate the grammar of a language. I mainly examine the comfort side of the linguistic cycle by considering the role of third factor principles in language change, in particular determinacy and labeling. Earlier structural Economy Principles, such as the Head Preference Principle, Late Merge, and Featural Economy (van Gelderen 2004, 2011), influence change but still depend on linguistic information and, if less emphasis is placed on these, they should be reformulated as non-linguistic, i.e. genuine third factor principles. 

                  The changes to be discussed as resolutions to the labeling difficulties are the well-known pronoun to agreement and demonstrative to copula, complementizer, and article ones. I also show that phi- and Q-feature sharing is very different. Changes that can be argued to be due to Determinacy are topic to subject and verb to auxiliary changes. Because determinacy can be resolved in a number of equally optimal ways, there is variation in how languages resolve possible violations, e.g. the CP-TP `bottle-neck’. In this bottle-neck area, the phi-features of C are inherited by T and a number of issues can arise, e.g. indeterminacy may occur because there are too many positions in close proximity to each other through which wh-elements must move. In Modern English, a strategy is followed where the complementizer that and the CP can delete; in Old English and Dutch, the TP is not present so indeterminacy is avoided that way. Two other differences follow from this.