Comparison Recordings
BWV 1050:
London Baroque Ensemble/Haas Westminster LP WN 2211
Mandeal, Members of the George Enescu PO, AIX 1338 AX DVD-Audio
BWV 1061:
Richter Müller, DECCA ["London" mono LP] CM 9157
Rafael Puyana, Genoveva Galvéz, Jenkins, CCO [ADD] Mercury
BWV 1067:
Petri, recorder, Kussmaul, Berliner Barock Solisten
Galway, modern transverse flute, I Solisti di Zagreb [ADD] RCA 6517-2 RG

Many fine recordings over the years have taught me that they know Bach in Leipzig, so I expected a lot from this recording, and wasn’t disappointed. These are possibly the best, or at least equal to the best, performances of these frequently performed works I’ve ever heard.

They are very fast, but there is no sense of the music being rushed; it simply erupts at this tempo as if it couldn’t help itself, as if this were the only way it could possibly be played. Having just finished reading and reviewing a book on the origins of our ideas of original performance practice, this recording is a perfect example of what it was all about, Bach’s music pretty much the way he played it and heard it himself. This music could not have been played this way fifty years ago, even thirty years ago, because it was necessary not only to know, but to get used to knowing, how to play like this. And, these are chamber music performances, not "orchestra" versus "soloist" performances, again, close to the way we presume it was played in Bach’s time.

The "Brandenburg" Concerto No. 5 is generally credited with being the first solo keyboard concerto ever written, and it was at one time played on the piano by famous pianists before famous orchestras. These large forces required slower tempi, and those slower tempi tended to stick in the first recordings for smaller forces. Then, with the first explorations of original performance practice, musicians tried playing faster, uncomfortably so, because they felt they ought to. But this performance, the fastest I’ve ever heard, is also the most natural sounding. This is more an accompanied trio, like the Beethoven Op. 56 than a keyboard concerto. It would be interesting to hear a performance where the violin and flute acted like co-soloists and improvised their own cadenzas at appropriate places; this would be the final step in ultimately authenticating this music.

The Concerto in C may have been originally written as a duet for harpsichords, and the orchestral accompaniment may have been an afterthought.

The Second Orchestral Suite has been frequently performed as a solo concerto by flutists such as Galway, Petri and Rampal, and has been recorded by Simion Stanciu on the panpipes and Maurice André on the trumpet; it always works, of course. But like the other works on this disk it is chamber music and while the flute carries the bulk of the musical responsibility, everybody should be heard and the interactions evident. These "Suites" of Bach are actually "overtures" in the direct line of ancestry of the late Haydn symphonies which set the form for almost all modern symphonies. Indeed Mahler "modernised" the orchestration and programmed them as Bach’s "Symphonies" in Vienna; but this of all the suites is the least suitable for that treatment, and the style of this recording comes closest to the authentic nature of the work.

The surround sound perspective is front-centred with ambient information from the rear channels. It is a shame the orchestra seating wasn’t expanded so the extra channels could give us greater separation of the voices. Some would denounce that as a "gimmick", but the truth is chamber music is written to be played by musicians sitting in a group, close to each other. Our modern concert hall perspective, sitting 50 metres away in the middle of a crowd of 1000 or more people, a perspective all but unknown before the nineteenth century, is not the natural way to listen to chamber music or any eighteenth century music. Each player hears the other players from all around him. It is the audience-versus-stage perspective in chamber music that is false, that is the "gimmick." A surround sound recording of a string quartet should invariably have one instrument in each corner of the listening room, or at the very least, offer a sound stage of 120 degrees or more. That would be the most authentic recording perspective, the one that most closely expresses the composer’s intentions.

An obvious point here relates to the two harpsichords in the concerto. They are placed very close to each other, just to the right of centre, with harpsichord II just behind harpsichord I. The writing is a duet for soloists; they comment on each other’s lines now and then in question-and-answer style. Perhaps placing the instruments close together makes it easier for the players to hear each other and stay together, but we, the audience, are missing out on much of the fun of the interplay between them. It would be easy to seat the harpsichordists close to each other but have the sounding boards pointing out, and then accentuate this with microphone placement. The ideal is what we have from AIX records, a surround sound "stage mix" with the instruments in a circle around the listener; that way the listener can participate in the music as the composer intended.

Paul Shoemaker