Each of these new recordings starts with an irritating disadvantage: Turovsky's has no fill-up, and Rozhdestvensky's does not provide the texts of the poems that are the basis of Shostakovich's symphonic song cycle. From the former point of view this gives a distinct edge to Haitink (on Decca), who adds the restrained but movingly eloquent cycle of Tsvetaeva songs, one of Shostakovich's finest vocal works, in an excellent performance with Ortrun Wenkel as the soloist. The 'musical fragments' from King Lear, offered by Olympia/Conifer as a make-weight to Rozhdestvensky's account, are agreeable enough but rather scrappy (tuckets and martial alarums, mostly: the sixth movement consists of five brief fanfares laid end to end) and the most entertaining 'fragment' of the lot, a sequence of ''Songs of the Fool'' (it includes a setting of Shakespeare to the tune of Jingle Bells), is omitted.
The omission of the texts is a graver flaw, though Russian speakers will find it easier to pick out the words in the Rozhdestvensky version of the symphony because his soprano has far better diction than Turovsky's. Haitink uses a version of the score (sanctioned by the composer) in which the poems are sung in their original languages (although Shostakovich set them all in Russian, only one of the 11 is by a Russian poet). I am beginning to have second thoughts about this: not only does it occasionally alter rhythmic values but there are several places where the climax of a phrase no longer corresponds with the word that inspired it, or where the original poem contains words or imagery omitted in the translation that Shostakovich actually responded to.
So far as recorded sound is concerned, Chandos enable one to stand very close to Turovsky's virtuoso ensemble of chamber players, and very close indeed to his soloists. This is no bad thing, perhaps, in the case of Storojev, who has just the sort of black, Slavonic solidity of sound that this music requires, and he is splendidly vehement in the witheringly contemptuous ''Zaporozhye Cossacks' Reply to the Sultan'', if a little more stolid elsewhere. Holleque, however, is quite miscast in the soprano songs: she has a pure, charming lyric voice (she would make a lovely Manon, perhaps) but she sounds uncomfortable with the Russian language, swallows her consonants and acts the words hardly at all. Turovsky makes up for this to a considerable extent by the drama and incisiveness that he draws from his players, and the Chandos recording is darkly, massively rich as well as very close. There is more space around the leanersounding orchestra in Rozhdestvensky's account, and although this makes for rather less clarity of perspective both soloists project without difficulty, and tie edged, very Russian-sounding Kasrashvili is vastly more expressive than Holleque (and she is happier with the mezzo-ish range that the part often moves into). Safiulin (another fine Russian bass) is graver than Storojev but also more intense. Rozhdestvensky's direction is no less urgent than Turovsky's (and a touch less histrionic, perhaps), but I don't think it is just the remarkable directness of Decca's recording (you can almost count the individual players at times, though the sense of their being in a real space is not sacrificed) that makes Haitink seem more eloquent still, yet without a hint of grease-paint. His soloists are more vividly communicative much more concerned with verbal colouring (well, yes, a hint of over-acting does peep through once or twice) than Turovsky's or Rozhdestvensky's and both are in splendid voice. Hearing the 'wrong' language is beginning to grate a little, but Haitink would still be my first choice for an account of this symphony that distils its full range of emotion and the inclusion of the Tsvetaeva songs clinches the matter.'