MusicWeb International One of the most grown-up review sites around 2024
60,000 reviews
... and still writing ...

Search MusicWeb Here Acte Prealable Polish CDs

Presto Music CD retailer
Founder: Len Mullenger                                    Editor in Chief:John Quinn             



AmazonUK Amazon US

Johannes BRAHMS (1833-1897)
Symphony No. 3 in F major op.90 (1883) [39:45]
Serenade No. 2 in A major op.16 (1858/9) [29:17]
London Symphony Orchestra/Bernard Haitink
rec. live Barbican, London, 21-22 May 2003 (Serenade), 16-17 June 2004 (Symphony)
LSO SACD LSO0544 [69:12]


Error processing SSI file


There is a lot of the art that conceals art in Haitinkís performance of the Serenade. For one thing, I donít think Iíve ever heard it sounding so completely lovely. All too often in this work rasping oboes and booming horns prevail over the other wind instruments while Brahmsís use of only the lower strings creates muddiness. Here the textures are always light and warm.

And then, it speaks much of Haitinkís command of the orchestra, together with his long experience, that the piece actually sounds like a serenade for once. Thereís a gently flowing first movement, a bucolic scherzo, a gravely paced but not turgid slow movement, a graceful minuet and a finale whose high spirits never get out of hand. However, I did begin to lose interest about two-thirds through the slow movement and never quite regained it.

A suggestion that the music deals with larger issues came from a comparison of the version by Sir Adrian Boult. You will see at once from the timings that there is one big difference, but let me proceed movement by movement.

Serenade no.2† I†† II III† IV†† † V
Haitink††† 08:12† 02:44† 07:14† 05:14†† 05:38
Boult† 08:59†† 02:24†††† 04:43† 04:09 05:22

In the first movement Boultís nostalgic, romantic account hints at depths and symphonic dimensions which Haitink seems unwilling to recognize; there is a breadth and at times a grandeur which can only partly be accounted for by the slower tempo, since the chronometer reveals that there isnít actually all that much difference. It sounds completely different, though, and in different moods I could be very grateful for both.

If this is much as you might expect (this was a very late Boult recording), the remaining movements tell another tale. Boult takes Brahmsís ďvivaceĒ at its word in the scherzo and gives it a symphonic drive which Haitink seems to be deliberately avoiding.

From the first the Boult recording was controversial on account of the very swift tempo he chose for the ďAdagio non troppoĒ. He gets away with it with phrasing that still sounds relaxed and I canít actually say I got the impression that the expression is being squeezed out of the music, as usually happens when a tempo is too fast. On the other hand, can this tempo be called an adagio? Returning to Haitink, there remains the problem that the music begins to outstay its welcome somewhere after the five-minute mark; no such problem with Boult who has got it over and done with before that anyway. Also, under Haitink one notices that certain passages donít seem to have much activity in them, and there hardly seem enough notes to fill the texture, which is no doubt what led Boult to take a much faster tempo.

Might there be a happy medium? Well, fishing around among other performances I had on tape, I discovered that Wolfgang Sawallisch (Rome 1975) drew this movement out to 08:52, so perhaps Haitink is the happy medium. However, I feel that Renť Leibowitz, at 06:25 (Rome 1962) probably hit the nail on the head Ė sufficiently mobile not to hang fire but with just that little bit more space than Boult allows. The trouble is, Leibowitz made no commercial recording of the work (and the unattractive timbre and modest discipline of the Rome band means there would be little point in any historical label trying to exhume this particular one).

In the remaining movements the differences are minimal, but Boultís swifter tempi seem to flow a little more in the minuet and to provide more in the way of sheer zip in the finale. But you might prefer Haitinkís more relaxed touch. In view of the striking differences between these interpretations you could do worse than to have both of them and I for one will be glad to have the choice. If forced to plump for one or the other, in the end it would have to be Boult, since the music seems to have greater dimensions in his hands.

In the symphony, Haitink again prevents Brahmsís textures from ever clotting; the bass-line is never turgid and the brass are always a warm support rather than craggy in the Klemperer manner. This is a relaxed, serene performance which almost grudges Brahms such moments of power or energy as Haitink is compelled to acknowledge. The opening is more majestic than passionate and subsides gratefully into the lyrical foothills of the second subject where it almost becomes becalmed. Though the actual playing is incisive, the development section and the coda again avoid any suggestion of drama. The lead back to the recapitulation is very slow indeed.

The middle movements take on almost a dream-like quality, very beautiful in their gentle way though I felt the intermezzo too placid altogether, especially in the middle section. The strings offer, presumably under Haitinkís instructions, more portamento than we usually hear today. Has Haitink been studying the work of his great Dutch predecessor Willem Mengelberg? The trouble is that, in this context, the portamenti just add to the rather syrupy effect. The finale is again majestic, with no concession to headlong drama, and the serene close leaves you wondering what the point of the faster music actually was.

And here lies the problem. It is a very beautiful performance, indeed a perfect realization of a particular point of view. But it seems to me to leave too many elements out of the symphony. Itís not just a question of tempo, but note how Haitink spreads himself compared with two illustrious predecessors and one contemporary:

Symphony no.3 I II III IV
Haitink†† 14:24††† 08:53††† 06:45† 09:27
Boult† 13:08†† 08:32 06:03†† 09:08
Klemperer 13:04 08:17 06:12 09:14
Colin Davis† 13:27 10:00 06:38 09:33

It is salutary to find Klemperer the swiftest in two of the movements; the fact that Colin Davis presents a vastly long slow movement and exceeds Haitink in the finale too, points to the fact that interpreters in the latter part of the 20th Century took Brahms interpretation into realms undreamed of by those musicians like Boult and Klemperer whose roots were in Brahmsís own world.

But statistics only tell a part of the tale; neither of the two earlier conductors, nor Colin Davis either, present such a rigorously one-minded view of Brahms. The autumnal beauty is there, but so are drama and passion. With the various elements properly balanced the final descent into serenity has sense. This time I do not put Boult at the top of the list since, while his interpretation is admirable, it doesnít seem one of his most inspired efforts (for that you have to go to Symphony no.2, at least as far as his late cycle is concerned); I am not always a Klemperer fan but no.3 is the one symphony of the four where he really does seem to have all the answers, unfolding it in a single breath and giving full weight to each of its single elements. Haitink is very beautiful in his way, as I say, but I think you would have to know the work already very well to appreciate his view. I should hate to think of any newcomer to Brahms picking up this record.

The recording (which I heard as a normal CD) has a very natural bloom and it is somewhat surprising to learn that these are live performances; the audience must have been gagged and apparently didnít even applaud. Furthermore, there is no suggestion of the tension and communication of a live performance Ė for that you must go to Klemperer.

Christopher Howell





Return to Index

Error processing SSI file