Gustav MAHLER (1860-1911)
Symphony No. 9 in D major (1909)
Vienna Philharmonic Orchestra/Bruno Walter
rec. live, 15-16 January 1938, Musikvereinsaal, Vienna
I detest the current almost universal use - one might say abuse - of the word ‘iconic’. However, I might stretch a point in the case of this famous recording of Mahler’s Ninth. It was taken down live, in an act of extraordinary vision, by Fred Gaisberg and his HMV team in January 1938. That was just a matter of weeks before the Anschluss which made continued life in Austria an impossibility for Bruno Walter and a good number of the players who had taken part in this performance.
It was only a few months ago that I welcomed a fine Pristine re-mastering of Bruno Walter’s 1961 stereo recording of Mahler’s Ninth and now we have a new transfer from them of the celebrated 1938 recording, which was the first-ever recording of the symphony. Incidentally, if Pristine gave the correct date for the 1961 recording - and in my review I suggested they may not have been quite correct - then the sessions took place forty-three years to the day after the 1938 Vienna performance. While on the subject of dates, I’m intrigued to see that Pristine indicate that the 1938 recording took place over two days. I’ve always believed that HMV recorded a single performance - on 16 January - and that seems to be confirmed by a 1944 article by Fred Gaisberg in which he specifically states that the performance was recorded that day though he does also say that there were “five rehearsals during which our engineers could make their tests and experiments in ‘mike’ positions.” That article is reprinted in the booklet for the transfer of the performance that I’ve owned for many years. It’s a transfer by Michael J Dutton (Dutton Essential Archive CDEA 5005).
So much has been written about this extraordinary reading that it seems almost superfluous to say more. It’s often a scalding interpretation and the urgency of the music-making is as remarkable as it is palpable. In my review of the 1961 disc I drew attention to the fact that the 1938 reading is some twelve minutes shorter. In the absence of any other recordings of Walter around this time it’s impossible to know how representative of his thoughts on the symphony at the time this 1938 event was. One would normally assume that a performance presents the artist’s considered view at the time it was given but it’s possible that the feverish political atmosphere in Austria in early 1938 - and the trepidation this must have caused people like Walter - may have added an extra febrile quality to his music-making at the time. Having said that, a photograph of him, which was taken in the green room immediately before the 1938 concert, shows him looking serious but calm.
The performance was discussed in some detail by Tony Duggan in his survey of some of the recordings of the Ninth and I largely concur with his judgements. I share his relative disappointment over the quite swift pacing of the finale but Tony seems happier with the very sturdy pace for the second movement Ländler than I am: it seems almost stolid to me. I could also wish that Walter had been more expansive in the trumpet-led nostalgic episode in the third movement (from 5:38). Tony was absolutely right to comment on the strain under which the orchestra is audibly working and one of the ironies is that the clearer a modern transfer is the more those frailties show up. One wonders whether the frequent fallibilities in the playing owe more to the unfamiliarity with this complex score or to those volatile times. It’s salutary to remember that this was then a very new piece: Walter had led the premiere only in 1912 and one wonders how many members of the VPO had played it before.
A comparison of this new Pristine transfer and the one by Michael Dutton has been very interesting. The first bit of comparative listening I did - the opening of the first movement - was to the Dutton disc. Leaving the controls unaltered I then played the same passage on Pristine. The Pristine transfer is at a higher level and, to be honest, a comfortable level for listening to Dutton was a little uncomfortable for Pristine. Overall, my impression is that the Pristine sound is a little more defined but it’s also rather fierce at times. I think perhaps Andrew Rose of Pristine may have retrieved a bit more detail from his source material - but there’s a good deal of detail on the Dutton transfer too. Thus, for example, we can hear it quite clearly when a member of the orchestra drops something during the first movement (7:24); it’s just that bit more obvious on the Pristine version, as is audience noise generally. Perhaps there’s a little more space round the orchestral sound in the Pristine transfer but it’s marginal and you get a good sense of the hall’s ambience in both transfers.
The Dutton transfer makes for more comfortable listening when the orchestra is playing loudly. Thus, for example, the first big climax in the first movement (around 3:00) sounds a bit strident with Pristine and later on in the same movement (around 19:50) the brass do blare rather more than they do with Dutton. On the other hand, lightly scored passages, such as the end of that movement, come off well with Pristine. I’ve referred earlier to Walter’s sturdy way with the Ländler music in the second movement. Near the start of the movement he gets the strings really to dig in and the sense of that is almost tangible with Pristine.
Come the great concluding Adagio and I think it’s the Dutton transfer that does more justice to the tone of the VPO strings. Whichever version you hear allowances have to be made for the age of the recording but there’s more edge to the sound of the strings with Pristine. Both versions report a good, solid string bass sound. The big climax, starting around 12:00, is another example of unpleasantly blaring brass in the Pristine transfer. This is another instance where the Dutton transfer tames the brass a bit more - to beneficial effect. In either version it’s fascinating to listen from about 15:00 and to hear the succession of downward portamenti in the strings (around 15:20). There’s aching nostalgia in the playing and a sense of a world that was very soon to vanish for ever. As I listened to the last two or three minutes of the movement in each transfer I forgot about making comparisons.
After I’d done my listening and as I sat down to type up this review I decided to see what Tony Duggan had had to say about Walter’s 1938 recording. I was interested to see that he’d compared an old EMI Références transfer, which I used to own years ago, with the Dutton. While praising the latter for, among other things, a gain in detail, he felt that the EMI transfer offered a more comfortable listening experience. As it happens that sums up my feelings after this present comparison. The Pristine transfer is very clear and present but perhaps it shines too bright and unforgiving a spotlight. Some may feel that the Dutton tames the sound too much but I feel that it offers a more comfortable experience for domestic listening - though, arguably, listening to Mahler’s Ninth should never be ‘comfortable’.
If you already have the Dutton I see no reason to part with that and acquire the new Pristine version. However, whichever transfer you opt for this famous performance demands to be heard. It’s amazing that this remarkable, truly historic reading speaks to us, and does so vividly. seventy-five years after it was given. For that we must give thanks for the vision and technical skills of Fred Gaisberg and his team and also for the skill and dedication of transfer engineers such as Michael Dutton and Andrew Rose.
John Quinn
A new transfer of Bruno Walter’s remarkable 1938 Mahler Ninth brings mixed results. 

Masterwork Index: Mahler symphony 9