In the first movement, this means that the triplet crotchets underlying the second theme will approximately equal the normal crotchets of the Majestoso. This imposes certain limits on your choice of tempo for the Majestoso. If you start as slowly as Klemperer, this tempo relationship, if observed for the second subject, would result in such a slow motion rendering that not even he could attempt it, and in fact he just ploughs on without changing the tempo at all.
Quite a lot of conductors get this crotchet-equals-triplet-crotchet relationship right, but not so many of them hold this slower tempo until actually instructed to return to the original Majestoso.
Keilberth and Horenstein were not the only two to play the first movement with the tempo relationships as instructed, but they were the only two (to my knowledge) who also played the finale as written, i.e. no tempo change for the second subject (because none is written), but a marked change where one is actually requested, and then holding the slower tempo until actually instructed to return to the original one.
You might say, well, so what? Isn't the important thing that it sounds convincing? Maybe, but the differences between what is written and what is often played amount to the symphony having a different structure. Can both be right?
To the best of my knowledge, no other Bruckner symphony has such clear cut markings, though the finale of no. 4 raises the issue of what we should do about the "half-tempo" marking present in Nowak but not in Haas - and again, when we should go back to the first tempo. I discussed this in some detail in my review of the Klemperer box.
Klemperer seems to have liked the idea of the half-tempo. In no. 5 he practically doubles and halves his tempi at will, though nothing of the sort is marked here.
Chris, you raise a very interesting point - and one that probably applies to other Bruckner symphonies as well. My immediate reaction is that the tempo indication at the start of the Sixth - 'Majestoso' - leaves a lot of room for discretion on the part of the conductor.
Patrick and I will consider this quetion and post a more considered response in the next few days. In the meantime, it will be interesting to see what other readers may have to say on the subject - I note there have already been interesting messages from Ralph Moore and Lee Denham.
Thank you for taking part in the MusicWeb International Forum.
Len Mullenger - Founder of MusicWeb