To state the obvious, a hugely impressive achievement: a comprehensive, academically rigorous, analytical and thought-provoking history of Western music, which manages both to cover the canon and challenge preconceptions with new insights or at the very least unusual (albeit consistent and justified) angles.
To state the only slightly less obvious, it's also a cracking good read. Taruskin clearly set out to make it more than a mere reference book, and although I didn't whip through it with any speed (there is an awful lot to take in), I certainly read it from cover to cover.
It occasionally feels as though Taruskin overreaches himself in his aim to make this a good read, with little bouts of silliness or slightly patronising comparisons that suggest that he doesn't quite get that what he has written is still very, very niche. I have a degree in music and this was a heavy read, so I don't quite think this is going to be lifted off the shelf by your casual enthusiast. That being the case, I'm not sure that comparing Caccini's 1601 description of 'graces' to Frank Sinatra and Perry Como adds much - he can leave those comparisons to the likes of Alex Ross, who is welcome to them.
My other gripe is perhaps only specific to my paperback copy, but given the absolutely vital nature of the musical examples and illustrations I do feel that OUP might have taken a little more trouble over the reproductions. 'As you can see at a glance,' says Taruskin in the chapter on Music For An Intellectual and Political Age, 'the manuscript… uses a different kind of notation.' Well, no I can't -even following up my unfruitful glance with much screwing up of eyes and peering, even with a year's study of early notation up my sleeve, I couldn't identify any such difference. What price a few plates, OUP? Honestly, I'll fork out for them.
Mind you, Taruskin does sometimes make assumptions about what we mere mortals will be able to follow, and if there's one genuine flaw in the writing here it's that his analysis doesn't always guide the reader through the examples with enough clarity. I haven't found this a problem in later volumes, so it may simply be that the terminology of the middle ages is so unfamiliar, or that without the reassuring presence of bar numbers and instrumental names it's much harder to give precise navigational directions, but trying to find a particular section of plainchant on a poorly reproduced manuscript when the only clue given is that it's 'the triplum part' would have required more dedication and page flicking than I was quite prepared to give.
The failure to account for such human failings is, of course, the remit of great intellects, and Richard Taruskin is clearly a man of very great intellect. This is an unparalleled achievement and nothing has put me off diving into the next weighty volume.
After a pause for air, obviously.