The majority of surviving fifteenth-century polyphonic music is vocal, and a great many upper-voice parts are provided with text in the oldest manuscripts; however, it is not always clear from these sources which notes are meant to be sung to which syllables. Much research has been devoted to helping editors and performers to cope with this situation, principally by providing editorial guidelines for the application of verbal texts to musical lines. This article does not aim to resolve such ambiguities as remain. Rather, it presents new evidence for an informed view of the situation in the sources. This may have implications for the development of editorial guidelines. However, its greater importance is as a contribution to the more general understanding of text-music relations in the period.

The recent publication of Heinrich Schenker's Das Meisterwerk in der Musik (1925-30) in a translation by Ian Bent, William Drabkin and others reminds us that in the highest modern traditions textual criticism and exegesis — editing and analysis — are inseparable. Much of Schenker's commentary on Mozart's Symphony in G minor, K.550, is devoted to isolating discrepancies between manuscript and edition, and some of his most fascinating analytical suggestions grow out of just this. Conversely, analytical strategies may have editorial consequences, and decisions taken by editors of the Neue Mozart Ausgabe have been challenged justly on stylistic grounds. If a passage occurs twice in a composition, for instance, does it have to be articulated in the same way on both occasions? And if there are differences in the articulations, what exactly do they tell us? Given the intricacy of the modern analytical debate over Mozart, such questions may not be answered by editors alone. Nor can editions be undertaken without reference to changing social circumstances: for composers in general have shown themselves more than ready to provide their own variants for different performing situations (again, Mozart's re-orchestration of the Symphony in G minor is a case in point). Indeed, in another field, scholars have asked that cultural issues should also influence editorial practice; these issues similarly require learning and sensitivity, and again deserve attention.

Mozart's partial autograph of the duet ‘Nun, liebes Weibchen’, K.625/592a, written for Emanuel Schikaneder's heroic-comic opera Der Stein der Weisen oder die Zauberinsel (Theater auf der Wieden, 11 September 1790) has not yet received the attention it deserves in the secondary literature. Perhaps this is due to the reluctance of scholars to accept Mozart's authorship. Because the initial string parts and all of the vocal lines are in another hand, and because Mozart did not enter the duet in his own handwritten catalogue, Alfred Einstein suggested that Mozart merely orchestrated original material possibly composed by Benedikt Schack, a suggestion that has been repeated as if it were a fact in several prominent studies.

Of what practical use are historical field recordings? This article is a partial response to that question, one which struck me on many occasions while working with ethnographic wax cylinders at the British Library National Sound Archive (NSA). There are, I believe, at least three possible answers to the question. One is that historical recordings can (and indeed should) be returned to the societies from which they originated, for which they serve as valuable cultural documents — documents both of indigenous cultural practices and of the engagement of scholars and other collectors with those practices (often within a colonial context). Another is that they can be used in studies of musical continuity and change, a possibility exploited most extensively by the ethnomusicologist Nazir Jairazbhoy, who retraced the path of his predecessor and mentor Arnold Bake's Indian field trips, comparing his own findings in 1984 with Bake's in 1938. My third answer is that historical recordings, together with the publications based on them, can contribute to studies of the historical development of ethnomusicological method — of the ways in which music has been recorded, transcribed and analysed. This article pursues this last possibility, taking as an example one of the classical works of early twentieth-century comparative musicology, A. H. Fox Strangways's The Music of Hindostan. I consider the author's aims, methods and conclusions, their status in the light of subsequent research, and their significance for music studies today.