Grammaticalization is seen as a cross-linguistically homogeneous phenomenon. Our project starts out from the idea that there is a certain degree of areal and cross-phylogenetic variation in two domains as stated in the following hypotheses:
(i) Grammaticalization paths of the type [SOURCE → TARGET] show areal/genetic variation in terms of the sources that undergo grammaticalization as well as in therms of target concepts.
(ii) There are differences in the covariation of meaning and form, i.e., the coevolution of meaning and form shows differences between semantic change and different form-related parameters and it is likely that there also are some areal differences.
Moreover, we are currently preparing an online accesible Databank, which will consist of all the grammaticalization paths collected in the project, and aims not only at allowing access to the (revised) database of the publications, but also encourage more quantitative research on grammaticalizaton paths.
As part of the project also a conference on the said topic took place, which was held in 2015 in Mainz (s. Conference). A collection of the resulting papers has already been published and is accessible online.
For more information on the methods and the workflow of the MAGRAM project, Methodology.
To get in touch with the project, see Contact.