Secure and Insecure Bases in the Performance of Western Classical Music

In The Modern Invention of Medieval Music (Leech-Wilkinson 2002), I suggested that Kuhn’s 1962 model of change in scientific thought applied effectively to changes in scholarship on the performance of medieval music. In Western classical performance as a whole, however, there has been only one Kuhnian revolution so far as we know, the development of historically informed performance (HIP) from the 1960s onwards. Nowadays that is generally seen, at any rate among scholars, as another modern invention, albeit artistically highly successful (Taruskin 1995, 2009). HIP performers still prefer to see it as historically well-grounded, for that provides the kind of justification for its practice that is absolutely required by the ideology underpinning performance of Western classical music (WCM). That ideology rests on the belief that the role of the performer is to reproduce the intentions of the composer. In fact, as has been comprehensively shown over the past 30 years (since Philip 1992), performance style changes constantly, with musicians insisting all the while that they have correctly translated the composer. According to the ideology, performance should always remain the same, and it is performers’ and WCM gatekeepers’ duty to ensure that it does; in practice it changes slowly, but over time very greatly. In a study drawing on Darwinian and post-Darwinian evolutionary theory (Leech-Wilkinson 2009a), I argued that this process of change is correctly understood as effected by mutations that are too small to be noticed but that spread rapidly through unconscious copying by others, so that over large spans of time performance changes so much that what is considered “naturally musical” alters beyond recognition. This explains the horror with which modern performers greet recordings made a century ago, where key modern values of musical performance are contradicted. These recordings include many (now unacceptable) performances by composers of their own scores (Slåttebrekk and Harrison 2010). Thus early recordings undermine the ideology itself (Leech-Wilkinson 2010, 2020a).