Purcell's surviving autographs have long been a source of fascination, and the extensive evidence they preserve of his working methods has been examined in several academic studies. However, methodology has focused on analyzing notational changes within the sources, and neither the appropriateness of the standard labels "working draft" and "fair copy" nor the directness of the relationship between the creation of a composition and its encoding in notation has been questioned. This article challenges current assumptions about Purcell's compositional processes, arguing that because the autographs were created for specific, often practical, purposes not necessarily directly connected to the compositional process itself, it is misleading to interpret them divorced from their function and the cultural and social contexts in which they were produced. There are three main factors that help us to determine the different purposes for which the manuscripts were copied: First, Purcell maintained a clear separation between different genres of music in both his private and court scorebooks, so it is possible that his compositional techniques varied between repertories; second, Shay and Thompson have established that the methods Purcell used to record his compositions changed as his employment conditions altered; and third, it is clear from contemporary descriptions that the modern tendency to translate the Restoration terms "fowle originall" and "fayre writeing" as "rough draft" and "fair copy" is misleading, and that Purcell's "fowle originalls" were not essentially private documents of his work in progress. Instead, the sources seem to have been notated for at least five different purposes, the categories including performance materials, file copies, and what we might term "transmission" copies, which, as Robert Ford has highlighted, were sent by the composer to provincial colleagues. Analysis of the autographs surviving for two contrasting genres - liturgical sacred music, written for choir and organ alone, and court odes - serves to illustrate the ways in which a contextual perspective on the sources can transform our understanding of Purcell's compositional methods, particularly when considered alongside significant nonautograph sources. There are distinct differences between the functions for which the manuscripts for these repertories were copied, with a notable emphasis in the liturgical sacred music on transmission around the provinces. More significantly, however, the extant "fowle originalls" for the two genres demonstrate variations of approach in the earliest notated stages, while comparisons between the two main court scorebooks and related non-autograph sources indicate that file copies were also produced for different reasons and do not relate to the cre-ation of performance materials in the same way. The manuscripts also suggest that more of Purcell's compositional amendments in the liturgical sacred music may have been made without a specific performance context in mind than has previously been thought, an observation that draws into question modern assumptions about the ontological centrality of musical performance in this period. © 2006 by the Regents of the University of California. All rights reserved.