Stars, bars and charting rhythms

Igor Stravinsky once said that the only recording angel who interested him was the one with the big book and not CBS, but it was CBS that made Stravinsky the first classical superstar. Whether in the process it "saved'' classical music or merely accelerated an "inevitable'' change from concert-going to record-buying, participation to consumption, as the fundamental musical nexus is a lot less clear. Unfortunately, historian Andre Millard's textbook is less concerned with "recorded sound'', as his subtitle claims, than with the advancing technologies of sound recording, which presents a very different set of issues.

Like television, sound recording was not so much a pure invention as a long-predicted and carefully mediated branch of existing technologies. Millard offers useful technical information on early systems, reinforcing a perception still shared by cylinder and shellac enthusiasts that even early commercial recordings were far from "primitive''. The anthropological analogy has an important part to play in the story, and has a clear sociological adjunct. Though early domestic replay machines were the preserve of the well-to-do middle class, cheaper systems filtered down the social scale and, in American terms, across ethnic divides, with astonishing speed. Anyone who indulges unsubstantiated diatribes about the suffocating number of new recordings available on CD today would do well to look at Millard's pre-first world war figures. As with so many cultural generalisations, those about the record industry are often made relative to the special circumstances of the Depression and second world war years. The phonograph and its cognates were almost instantly successful and were diffused faster (arguably) than any technology before television. Their appearance on mid-western dirt farms and in southern black shacks is vividly attested in Millard's photographic evidence, which is not sufficient, but is well backed up elsewhere.

One of the most public and obvious ways in which American society is still racially divisive is the slowly disappearing division between "popular'', or "rock'', charts and "black", or "dance'' or "r'n'b'', charts. It is a distinction that has survived stylistic levelling and the enfranchisement of black music on MTV. It has its origins in the early record companies' desire to split repertoire into mainstream and "race'' labels; Millard shows Edison labels for "dance'' (that is, foxtrot and waltz discs), "rube'' (or country) records and "banjo selections'', which are explicitly aimed at the black market. The generic and stylistic compartmentalisation of music is almost entirely a recording phenomenon and one of many ways in which Millard's argument suggests more purely musical concerns that seem to lie outside his areas of interest. Precisely how recording technologies and musical styles interact, how music is tailored to what would originally have been a second-order transmission, is the fascinating subtext to America on Record, but rarely put in the foreground.

Millard makes it clear that he is aware of these interactions. They are in the present-day scratch and hip-hop musics, which are definitely regressive in technological terms. And they made a huge impact on the development of jazz, a style of music coterminous in historical terms with sound recording, and very largely driven, not by "live'' performance as most music historians instinctively argue, but by the demands of recording. Millard shows how the cult of Caruso was largely generated by early recordings which then in turn generated his awesome bankability as a stage and recital singer. It was a voice that, unlike Melba's or Patti's, lent itself to the feasible spectrum of pre-electric recording, and as such it became the norm for subsequent performers, putting a premium on emotionalism and dramatic presence over the more abstract musical values of the previous generation.

Millard persistently conflates music and its dissemination, even where there ought to be some separation. This is most obviously the case with jazz, "the folk music of the machine age'', whose rise coincides with America's imperial project but whose role in that project is left unexamined here. A frustrating book, then, valuable for giving the vast contemporary recording industry a degree of context and a few cautionary relativities. Millard will be a useful adjunct to American studies courses and to those who need a rapid resume of the combined technical and economic genius of the early pioneers. It is less useful and, on occasions, disconcertingly unthoughtful, about the interpenetration of technics and art. That task remains to a more musically literate observer.

Brian Morton is senior arts producer, BBC Radio Scotland, and presents jazz and classical music programmes for BBC Radio 3.