Mahler: Symphony No. 4

Our rating 
4.0 out of 5 star rating 4.0

COMPOSERS: Mahler
LABELS: Decca
WORKS: Symphony No. 4
PERFORMER: Dawn Upshaw (soprano) Cleveland Orchestra/Christoph von Dohnányi
CATALOGUE NO: 440 315-2 DDD
Following clinical earlier instalments in their respective Mahler odysseys, two master technicians play their trump cards. Sinopoli conducted Mahler’s Seventh in concert a week after Tennstedt, and to everyone’s amazement the Philharmonia performance was both more finely prepared and more expressively fluent.

Advertisement

So, too, it proves in the recording studio. Mahler’s strangest meeting of past rituals and futuristic nightmares needs a master-balancer to keep weird orchestral mixes in harness, but Sinopoli – for once – does much more than that. Every sound is graphically to the point, from screaming high frequencies – the Philharmonia woodwind and first trumpet (John Wallace, I presume) are superlative – to extreme rhythmic tension in the bass lines; and Sinopoli beats all competition, Rattle included, in his pacing of the outer movements.

The second Nachtmusik is controversially slow, but only to the purpose of lengthening the night shadows in this bitter-sweet serenade and emphasising its pivotal role between darkness and light.

Advertisement

The light certainly gets a better deal than the darkness in Dohnányi’s account of the Fourth Symphony, which is classically pure at the start, with no dynamic hair out of place, yet reluctant to admit to undercurrents. The Cleveland violins fall expressively short of their fellow strings in the slow movement, but heaven’s gates are convincingly stormed and Dawn Upshaw crowns Dohnányi’s vision with beguiling lightness of touch – infinitely preferable to the anodyne McNair on Philips. Bryn Terfel looked, on paper, to be the star attraction of the Sinopoli set; but his score-faithful reading of Kindertotenlieder, though phrased with supreme artistry, only springs off the printed page in the last, hallowed bars. So special a case among song cycles really does need the maturity of a Fischer-Dieskau. David Nice