Having heard M3 in Milwaukee a couple weeks ago, I have surveyed most of the recordings I own, along with another from the library. I will do this with a Mahler work every few years, triggered usually either by a concert or by hearing a previously unknown recording. Each time, I often come to somewhat different views of the performances.
Typically, I will start with the more recent or unfamiliar recordings, then move toward my previous favorites, which come last. I find that I almost try to avoid arriving at that previous "winner," perhaps suspecting that its charm will have faded. (I'm not alone; one Fanfare reviewer [though possibly not for the same reason] mentioned that he rarely listens to his favorite recordings.)
I claim nothing rational or orderly about this process, agreeing with Emerson that "a foolish consistency is the hobgoblin of little minds." In each round of listening, I'm noticing somewhat different characteristics. If I demote a recording, it is with regret, but I am always delighted to find virtues that previously eluded me. My conclusions (not really conclusions, because they are always tentative) rarely change radically.
In this round of M3 listening, I am surprised that I found Bertini to be a very strong performance, in respectable sound. I didn't care for this performance at all when I first heard it almost three decades ago. Bychkov also rose in the rankings; this recording seemed uninvolving last time I listened to it, but now find it well thought out and nicely recorded, one of the more appealing choices. Levine/CSO is a fine performance, sounding much better on CD than in its vinyl incarnation.
To help cement my reputation as an unreliable reviewer, I was favorable impressed with Boulez when I bought it a couple years ago, less so in the early rounds of my current survey, and back to liking it a lot when I tried it again. (The SQ alone, unusually lifelike, makes this worth a listen.)
Lopez-Cobos/Cincinnati and Litton/Dallas have gone to the out stack. Neither captures an idiomatic Mahler sound. I cast a jaundiced eye at Zander; this recording has its virtues, but suffers from a softness and lack of detail in the bass. I couldn't get past the first movement.
Haitink/RCO, though in dated sound, still holds its own as a fine performance. Chailly also sound very good to me this time around, far better than before.
The previous favorites? Ozawa and Inbal/Frankfurt have always struck me as among the strongest entries in their respective sets, and I still think each is outstanding.
A very honorable mention goes to Leinsdorf/BSO. While the lightweight sonority often strikes me, at least initially, as unidiomatic, the clarity and detail in this recording is breathtaking, and for me, Leinsdorf's interpretation commands attention. It's compelling, if unusual, from start to finish, but note particularly the third movement. He really speeds up the more intense episodes of the finale, fully justified by the score. RCA had long captured radiant string tone in Boston—you'd have to pry those Munch recordings from my cold, dead fingers—and this recording is an outstanding example.
A final note on my peculiarities as a listener: I often have a negative reaction when I first hear a recording, for reasons I cannot explain, and I've learned not to take it too seriously. I bought most of the Jansons/RCO Live CDs a year ago and found them somewhere between underwhelming and annoying at first. A few months ago, I played his M4 and couldn't get past the first few minutes. Yet I've just listened to 1, 3, 4, 5 and 8, and all strike me as fine performances in very good sound, a few trivial quibbles aside.
And as I stagger to the finish line of this absurdly long post, I think it is a measure of the greatness of Mahler's music that it comes alive in such a variety of interpretations. Surely we are living in a golden age to have so many choices arrayed before us.
Topic: M3 along with some comments on listening (Read 4775 times)