GRAID (Grammatical Relations and Animacy in Discourse) is a system of labels and conventions for annotating texts. GRAID targets syntactically relevant units, such as predicates and direct objects, and assigns to them labels for form and function. Typically, GRAID is used in conjunction with annotation tools such as Toolbox, or ELAN, to which an additional annotation tier for GRAID annotations can readily be added.
GRAID was developed by Geoffrey Haig and Stefan Schnell. It has been trialled on text corpora from five different languages. The data from these initiatives consists of the original texts themselves, the GRAID-glossed texts, the GRAID annotation line extracted from the annotations, and a spreadsheet containing the raw quantitative data, and in some cases some statististical analysis. These data are currently being made available, see the links to the right.
Additional work on the corpora of Gorani, German and Vera'a was also conducted as part of the research project "Pronominalreferenz im sprachübergreifenden Vergleich", undertaken in collaboration with Anna Margetts (Monash University) and funded by the German Academic Exchange Service / DAAD (Grant No. 54385124, 2012-2013).