The other day, Alex Ross had a few choice words on the subject of Roger Norrington's recreation of Mendelssohn's revival of Bach's St. Matthew Passion. (Got that? Oh, good.) He's taken the posting down now on the grounds that it didn't make much sense. He's wrong about that - Norrington went over some kind of common-sense limit regarding the value of historical recreation - but I have a few choice words about Roger Norrington myself.
Norrington's had a long and distinguished career in early music, as you can read in his biography at the EMI Web site. His major fame in the United States certainly rests on his orchestral work with the London Classical Players, and more specifically on his Beethoven symphony recordings. I gather they were considered somewhat revolutionary at the time they were first published, in the late 1980s and early 1990s. The Beethoven symphonies were already available in one period-orchestra recording, that of Christopher Hogwood with the Academy of Ancient Music. But Norrington's aggressively fast tempos were something new and exciting.
I wonder if the excitment has worn off by now. I've heard a couple of those recordings, and they don't have very much going for them other than the unusual tempos and the period instruments. Those fast tempos are rigid and lacking in rubato, and I can't imagine that lack of rubato is a correct period practice. Sure, it's nice to hear Beethoven on period instruments - but there are excellent sets by John Eliot Gardiner and Franz Brueggen, who are both more interesting conductors.
There's one Norrington set that is still a revelation to hear, though, and that's the Wagner disc. I suspect his very fast tempos aren't any more correct than the Beethoven tempos, but the transparency of the orchestra and the bright, individual sounds of the mid-19th century instruments are a far cry from the homogenous, dense sound that so many consider the right sound for Wagner. That's a sound that is only produced by a modern orchestra.
Between that orchestral sound and the size and design of the Bayreuth Festspielhaus, his ideal theater, you have to wonder what kind of voice Wagner really wanted. We know from the trills and decorations to be found in the Ring operas that he wanted flexible voices. Yes, Wagner singers have to have stamina, but less-than-immense voices wouldn't present any problems of audibility at Bayreuth with a period orchestra.
The most interesting moral of the Norrington story is very likely that you don't have to be a trumpeter like Flagstad or Nilsson to sing Wagner.
Updated January 31, 6:13 a.m.
ACD mostly agrees with with what I say above, for which thanks, but takes issue with a couple of my points. Herein some responses.
I think I will just have to disagree that with Wagner, the genius lies in the massing (not the details) and therefore it's wrong to play him transparently. There is much genius in the massing, and yet there's also genius in the counterpoint. I've frequently gasped or been brought to tears by the details of the inner line in Wagner. I love Furtwangler and Knappertsbusch, masters of massing, but also Bernstein and Boulez, masters of transparency. And perhaps it's not a coincidence that two great composers are the ones to most bring out that inner detail in Wagner. It's certainly revelatory to me to hear, as well see, those details.
I have not yet been to Bayreuth, and it sounds as though ACD has been, so I am interested in hearing more about the "circle-squaring" aspects of achieving transparency in that theater. My experience is strictly with performances recorded there, in which, of course, transparency may be achieved by the clever placement of the microphones. (Boulez, though, has stated in writing that he wants to achieve transparency.)
I'm not saying Wagnerian voices shouldn't be big, beautiful, and expressive. I'd distinguish between "big" and "immense," especially when the hugest voices can't execute the ornaments or when they move with such difficulty that the music doesn't flow well. I love Varnay as an interpreter, but all that glue in her tone...!
Lastly, I'd have to take issue with "beautiful" as applied to Nilsson's voice. Again, I never heard her in the house - my loss - and everyone I know who really loves her heard her live. On record, even at her freshest, in her debut recordings from the late 1940s and on recordings up to at least her first Tristan, there's an awful lot of steel under not very much velvet. From the mid-60s forward, on record, her sound thinned out over time and records as if it's often just missing the core of the pitch, sometimes by different amounts during a phrase. The last time I put on Boehm's Tristan, this bugged me enough that I had to take it off.
Not everybody hears this in her, and I've had some rather sharp discussions with friends about her. But I think most will agree that she doesn't trill where Wagner wanted those trills, and she executes written-out ornaments (that one in the dawn duet, for example; yes, I can produce a measure number if you want one) without much grace. That also does Wagner a disservice.