On the development of the Romance demonstrative systems

The paper provides a structured account of the evolution of the Romance pronominal/adnominal demonstrative systems, using both the analysis of medieval corpora and recent fieldwork. It identifies the mechanisms involved, building on a detailed analysis of the textual and extra-textual usage of demonstratives, which gives new insight into the nature of deixis and anaphora. The evidence shows that the genetic relation between deictics and anaphorics is not unidirectional: not only can anaphors originate from deictics, the reverse can also be shown to hold true, with pure anaphorics appearing to be a viable source for second-person deictics. The main driving force in the process is subjectification, which conditions the development at every stage. The evolution follows predictable pathways and turns out to be cyclic: three main recurrent patterns are clearly identifiable for all the varieties investigated.