My interest in keyboard arrangements, and more generally in multiple versions of musical works, can be traced to an early encounter with Wittgenstein, long before I had become a harpsichordist or joined the ranks of musicologists preoccupied with historical performance practices. At Stanford in the late 1960s, pursuing graduate work in sociology, I audited a seminar on his Philosophical Investigations. As a budding sociologist, I could readily understand languages as games governed by sets of rules and conventions, often strongly entrenched, but far from immutable. It was likewise appealing to understand concepts and classification schemes as more dependent on family resemblances than on exhaustive sets of essential characteristics. Moreover, it seemed entirely accurate to expect that new speakers and writers were likely to alter both the rules of the games and the items on the family-characteristics menu, and furthermore that miscommunication might often arise should a speaker fail to follow the rules.

As one now preoccupied with research and performance in music, I find that many of the concepts from that seminar transfer quite well to a study of musical works and their transformations. The inevitable variety among performances and scores of “standard” works is not surprising, no matter how detailed the written directions, which we read with a variety of accents. Performances share merely a family resemblance, as new readers of received texts come away with new understandings of the works encoded therein. Sometimes those understandings are quite far removed from any that the composer may have imagined and may affect the work's very identity. It is from this perspective that I approach the study of keyboard arrangements. Variety is the norm and the boundaries of a work are permeable.

I have come to believe that the idea of a transcription or arrangement, a recasting of the “original” version to make it suitable for some other medium or performance context, or simply to update it so that its place in the repertoire can remain secure, is something to celebrate boldly rather than to bemoan as a departure from the composer's intentions.