In 1900 Macaulay suggested that some of the earliest surviving Gower manuscripts were copied in ‘the poet’s “scriptorium”’, and argued that some of these (most notably the Fairfax manuscript of the Confessio amantis) were revised under the author’s meticulous supervision. Macaulay’s suggestion was based on the quality of the texts, the nature of the spellings, and above all on the frequent appearance of certain scribes who either copy the text or enter revisions in the manuscripts. 1 Behind this suggestion lies a principle frequently invoked in the study of manuscripts copied in the early Middle Ages: that where one finds two or more scribes collaborating in two or more manuscripts, and that when their handwriting and other scribal practices exhibit sufficient points of similarity to suggest the existence of a ‘house-style’, one may infer the existence of a scriptorium.