This is Petri’s second time recording these works. The first time, in 1991, reviewed two years later by Kevin Bazzana, was for RCA Victor, and paired Petri with harpsichordist Keith Jarrett, in one of his occasional stints as a classical musician. Bazzana liked it a lot; I have that disc as well, and it’s certainly a top choice, if you have no objection to hearing works composed for the traverso played on a recorder. Last time, Petri had to transpose four of the sonatas (BWV 1031, 1032, 1034, and 1035) in order to play them on either a descant or alto recorder. This time, BWV 1031 has been left in its original key, presumably because she has acquired additional instruments in the interim; several tenor and alto recorders are listed in OUR’s booklet. She still transposes BWV 1032, 1034, and 1045, however.
Another change is the addition of a viola da gamba in all six of the sonatas. This is not discussed in the booklet note, but it is not unprecedented. Several recordings in which the sonatas are played on a flute add a cello or a viola da gamba, and to good effect, as it helps to fill out the lower end of the tonal spectrum. Given Perl’s popularity among early and Baroque music enthusiasts, including her in this project makes sense on multiple levels.
Bazzana noted Petri’s and Jarrett’s preference for “very quick tempos.” That remains the case here, although there has been a slight moderation since 1991. In timings, there is only one large deviation, and that is in the Andante of BWV 1034. That is not because of tempo, however, but because Perl is given a chance, at the start, to pluck out the movement's melodic outline—a lovely, still moment in a program characterized by many lovely moments.
Petri remains a masterful player. I didn't expect otherwise, and she does not disappoint. This disc is a lesson in what the recorder is capable of, when it is played by a musician who possesses the ultimate in technique and discernment. I also like what Perl brings to this disc. She and Petri are on the same wavelength. Esfahani is no Jarrett, however. Next to the powerful personalities of Petri and Perl, he sounds pale and lacking imagination. I think the sound of his harpsichord is partly to blame; a new instrument by Jukka Ollikka, based on a German model by Michael Mietke, it is thin and tinkly. Jarrett's instrument, a 1982 Carl Fudge modeled on Taskin, has more depth. Perl fills in some of what Esfahani lacks, but it would have been better, of course, if both of them had been equal partners with Petri. For what it's worth, I will note that I liked Petri's and Esfahani's collaboration (sans Perl) on a disc of Corelli sonatas reviewed in 2015. Esfahani played a different instrument there. As I have written several times, my reaction to harpsichord recordings depends at least as much on the instrument as it does on the instrumentalist.
Reservations aside, this disc is a very good argument for performing these works on a recorder instead of a flute. If another recorder player surpasses what Petri has done here, I will be surprised. That said, this SACD does not clearly supplant her earlier release with Jarrett, which stimulates me just a little bit more.