This is a great performance, one that simply has to be heard. In the first movement, the question of tempo is crucial: an Allegro qualified by ma non troppo but an Allegro none the less (my Eulenberg score carries no metronome marking). Much as I admire the more ‘traditional’ (or should I say ‘comfortable’) approaches that hark back, in discographical terms, to Fritz Kreisler’s period, I cannot help wondering whether most earlier recorded interpreters got it wrong. Here, the hammering first tutti has tremendous force, yet Zehetmair’s free-wristed phrasing liberates a hitherto unsuspected wistfulness in Beethoven’s solo line. The first movement is built on a tug of war between dynamic extremes and, for once, it actually sounds like a concerto, and a brilliant one at that. The more familiar average playing time of around 25 minutes (Zehetmair’s lasts a mere 22) tends, for all its beautiful effect, to compromise on forward momentum. Only Bronislaw Huberman – either on APR or, more recently, on a newly-discovered live recording from Arbiter – is as arresting. Here, the use of period instruments adds extra fibre to the aural mix, and Bruggen’s conducting has a pressing urgency about it that, again, intensifies the drama. The Larghetto is full of subtle nuances and telling inflexions, and the Rondo has great rhythmic verve.
Two recording venues are used, the excellent Muziekcentrum at Enschede for the two Romances and the rather cavernous Vredenburg, Utrecht for the concerto. The former yields the more sympathetic acoustic (Zehetmair and Bruggen offer limpidly flowing performances of both pieces), but I cannot imagine that the slightly rougher-edged concerto recording will give much cause for complaint.
The most relevant comparison is with Gidon Kremer’s excellent COE recording under Nikolaus Harnoncourt. Harnoncourt opts for broader tempos in the outer movements and the modern-instrument orchestra produces an altogether warmer sonority. The main textual difference between the two recordings is in the first movement cadenza. Both use a version of the extensive cadenza that Beethoven composed for his piano reworking of concerto, but while Zehetmair stays with Wolfgang Schneiderhan’s fairly straightforward (though brilliantly achieved) violin transcription, Kremer brings back the piano for a sort of mini trio consisting of violin, piano and timpani. Innovative though it is, I am not sure that it wears terribly well. Schneiderhan’s is for violin and timpani alone.
This, for me, is one of the recordings of Beethoven’s Violin Concerto. Recent rivalry is headed by Hilary Hahn’s serenely beautiful rendition under David Zinman, a performance that, for all its deeply satisfying qualities, does not really extend our knowledge or experience of the piece. Thomas Zehetmair’s recording does, and is to be strongly recommended.'