Is there an answer to the question ‘what is a refrain?’ Traditionally, scholars of thirteenth-century French song begin their reply by noting that a refrain is the characteristic formal feature of a rondeau, that is, an element which is repeated, usually with a separate half-repetition, in alternation with a strophic ‘respond’:

Moreover, most modern attempts to define refrains have explicitly begun by collecting rondeaux. But since only a small proportion of refrains occurs in rondeaux (in van den Boogaard's bibliography approximately one tenth), it is clearly not a comprehensive definition.

‘Such is your divine Disposation that both you excellently understand, and royally entertaine the Exercise of Musicke.‘ So wrote John Dowland of Robert Cecil, first earl of Salisbury, in the dedicatory epistle of Andreas Ornithoparcus his Micrologus, published in 1609. Beneath the hyperbole natural to a dedication lie the essential characteristics of Cecil's musical patronage. The first part of this paper examines his royal entertainment of music in terms of the form and scale of his patronage and the ways in which music could be used within the patron-client relationship. The second part explores Cecil's excellent understanding of music and comments on his personal taste.

Since it has been observed that artistic singers and also instrumentalists have somewhat digressed from the notes here and there, and thus have given occasion to invent an agreeable kind of figure; for what can be sung with reasonable euphony might well indeed also be written down.

Henry Playford's The Second Part of Musick's Hand-maid, the famous collection of keyboard pieces published in London in 1689, contains an anonymous and untitled ground in C minor which has today become well known both for its beauty and for its fine exemplary qualities as a post-Restoration English ground. It also presents us with an interesting study both in authenticity and in compositional evolution, inasmuch as its early history is as engaging as its more recent scrutiny has been controversial. The first half of the piece is reproduced in Figure 1.

It so happens that almost all the music which Schenker edited had previously appeared in editions by Hans von Bülow. On each occasion, Bülow's edition provoked from Schenker what he himself described as his ‘frequent and energetic criticism’. Examining the basis of these criticisms may lead us to a better understanding of Schenker's own editorial work – work which had an importance within Schenker's total output that theorists, if not musicologists, have tended to underestimate. For Schenker, editing meant restoring the masterworks of the past to their authentic state (the title-page of his edition of the Beethoven sonatas describes it as a ‘reconstruction’) and thereby ensuring their survival in a world that, in Schenker's eyes, had lost the capacity to appreciate their significance. And such editing called for respect for the masterworks and a profound understanding of them; hence the extraordinary virulence with which Schenker attacked editors such as Bülow, who in his view lacked these qualities. In his study of the G minor Symphony, Schenker wrote: ‘Editions of Mozart's works no longer have anything in common with the original manuscripts.’ And he added: ‘Then is Mozart dead?’ This extreme sense of the editor's responsibility – a responsibility that extended to the life and death of the masterworks – informed everything that Schenker did. I would go so far as to say that Schenker's theory of levels was conceived as, more than anything else, a decisive contribution to editorial method. And to say this entails a considerable revision of the image of Schenker prevalent today.

The assumptions that lie behind what have variously been called ‘realizations’, ‘completions’, ‘restorations’ or ‘reconstructions’ of unfinished works are generally unarticulated. Over the last few years such ‘realizations’ have become media events that threaten to blur the line between scholarship and public relations. Since for most of the general public these projects represent the sole contact with the discipline of musicology, the issue of ‘realizations’ is one that affects all members of our scholarly community. If you look to The New Grove Dictionary of Music and Musicians or any other standard music reference work for guidance on the subject of ‘realizations’ or ‘completions’, ‘restorations’ or ‘reconstructions’, you will find no such article. Indeed, I am not aware of any study that examines the conceptual basis of ‘realizations’ in the arts.