The creole anatomy of the Rich Agreement Hypothesis
The Rich Agreement Hypothesis concerns the generalization that V-to-I movement is conditioned by rich subject agreement on the finite verb. In order to rehabilitate this hypothesis (Rohrbacher 1999, Bobabljik and Thráinsson 1998), Koeneman and Zeijlstra (2014) resort to reanalyzing the subject clitic system of Colloquial French as Agreement, due to two facts: (i) the inflectional paradigm of Colloquial French falls short of counting as rich in their definition of richness; (ii) Colloquial French still exhibits the "V-Adv-Obj"-signature that is indicative of V-to-I movement. This paper looks at the status of the pronominal systems in subject position, the patterns of adverb placement, and how these properties are distributed among different creoles. With regard to French related creoles, we show that only a few of them exhibit the "V-Adv-Obj"-signature, whereas all of them have a designated set of subject pronouns, that in (at least) some cases are to be analysed as (subject) clitics in the sense of Cardinaletti & Starke. It becomes clear from the discussion that the distribution of the two properties does not show any effect of the purported correlation, and we argue that this constitutes a serious problem for the Rich Agreement Hypothesis (weak as well as strong version). We further discuss two other creoles, Papiamentu (Veenstra 2008, Kouwenberg 2006) and Saramaccan (Veenstra 1996, McWhorter & Good 2012), showing that they lack the "V-Adv-Obj"-signature but do exhibit syntactic subject clitics. These languages are, we argue, direct counterexamples to the Rich Agreement Hypothesis.