I'd like to put a word in here for the Études for piano, startling works. The (very different) recordings of these by Rosen, Gieseking, Jacobs and Uchida are all very well-worth hearing.

With respect to the Preludes, I'm not fussed on Osborne's approach personally, but would like to draw people's attention to Alain Planès's recording on Harmonia Mundi, on a turn of the century Bechstein (a piano Debussy particularly loved). Also to both of Gieseking's recordings of the works. And Michelangeli playing the Images and Children's Corner (less so his Preludes).

Did Ginette Neveu ever record the Debussy Violin Sonata?

What are anyone's favourite recordings of Pelleas? I only have one at the moment, the Abbado - surprisingly so as it's one of my favourite operas of all time! Never really investigated all the early recordings of it yet - any thoughts?

Probably not the type of playing that would make such an impact with British listeners, I suppose, but I like it very much. Schubert is OK, Janácek very fine. Debussy Études (on a modern instrument) are OK, though perhaps not amongst the best, Preludes are really wonderful (one of my favourite recordings of them), though.

A (relatively) rare opportunity to hear the Clarinet Rhapsody in the flesh at the Barbican tomorrow evening: Andrew Marriner/LSO/Gergiev - for those for whom the name Gergiev isn't totally off-putting. (Combined with Les Noces and an (even rarer) chance to hear Prokofiev's October Cantata in the flesh).

On Pelleas, I wouldn't presume to offer a comparative recommendation, not least because I haven't heard that many , but the old Ansermet LPs were my benchmark for years and, rightly or wrongly, I still tend to think of it as 'the' recording. Have also got a great deal (though I know it divides people) from the Boulez/Stein/WNO DVD. One of the very few exceptions to my rule of thumb that I don't really get on with opera DVDs.

No, not to do with it being 'impressionistic' (so is Gieseking), just being more about surface detail than emotional heart - doesn't move me in the way that Planès and many others do. Planès doesn't wear his interpretative strategies on his sleeve, in general, which probably wouldn't go down so well with audiences wanting to be dazzled in that respect. But overall I find British performers on all instruments tend to make Debussy into essentially a colouristic kaleidoscope, as well as sounding very calculated. Have you heard Debussy's own (very few) recordings/rolls? Very very different approach. Also very fond of the Debussy of Samson François or Zoltan Kocsis.

The question of 'intentions' in the context of the Stockhausen thread was to do with a possible discrepancy between what a composer intends something to achieve, and what it actually does achieve. A very different issue from how a composer plays their own music (which certainly shouldn't necessarily be seen as 'definitive', though).

Try a passage like the following from Et la lune descend sur le temple qui fut from the second book of piano Images.



Much of the piece consists of varying forms of 'processional' series of chords, sometimes parallel dissonant chords combining the root, fourth and fifth of a scale, sometimes triads - the latter generating a greater sense of distance and objectification to my ears. But from within this auratic world, Debussy extracts or superimposes moments of highly intimate (though understated) lyricism. Never using the more conventional resolution of a fourth onto the third of a scale, Debussy from bar 35 (and earlier in bar 16) instead uses the more askew resolution of the augmented fourth onto the third, more of a contrived resolution through the omission of the intervening fourth degree. The sense of distance is continued by the resolution of dominant seventh chords not onto the tonic but onto a distorted chord on the fourth, at that time a radical progression (and which in any time causes a blurring of any tonal sense, then through the use of chromatic half-steps within a whole-tone harmony (bars 37-38) before leading back to the 'processional'. Anyhow, all of this creates a context in which bar 41 achieves a high level of prominence (approached via a pedal B-flat and a G# major triad, so that, enharmonically, the B-flat becomes a ninth of an A-flat triad (but without the seventh), then can seem to resolve onto the implied fifth degree of D major, somewhat softening what would otherwise be a tritone shift. Bars 41-42, in part through their hushed and brief quality as well as the tonal resolution (with the extravagance of the extra harmonic colouration supplied by the melody), to me communicate a huge sense of yearning, an apparition if one likes, by these means, a moment of warmth and relief which would not be the same without the context (to some extent this is anticipated in bars 27-28). Other triadic harmonies are mostly approached via parallel chords; when one is not its impact is all the more striking.