A.C. Douglas links to this interview with Paul McCreesh. I would suggest that Mr. Douglas listen to McCreesh's 2000 recording of Bach's towering Matthäus-Passion.
He falls into all the standard HIP tricks: super-fast tempi and one-singer-to-a-part. Gardiner's recording, pretty much standard for a HIP Matthäus-Passion, though not as glorious and grim as Klemperer's echt-Romantic interpretation, is marginally slower and infinitely more dramatic. McCreesh revels in the "scholarship" of Parrott and Rifkin, which work attempts to undo two hundred years of received performance wisdom.
I quote from an interview in the liner notes,
However, if the early music movement is serious - and I'm not always sure that it is - how can we ignore 20 years of astoundingly erudite scholarship without trying to make it work?
Hmmm. Apparently, about as easily as you can ignore the performance history of the work since Mendelssohn's revival in 1829. The notes pay lip-service to the first critical editions of Bach, undertaken by the German Bach Society, while criticizing their concept of Bach. This performance is nothing more than HIP standard operating procedures applied to smaller-than-average forces. McCreesh might not claim a normative interpretation, but he certainly keeps step with his colleagues.
Only Sir John Eliot Gardiner, of the HIPsters of the first rank, understands that you can adopt period-performance style with a large ensemble. Had Bach foreseen the size of a modern orchestra, would have done things differently? You bet. These people, except Gardiner, don't understand that. Not for one second.
McCreesh ruined the soprano part, O Lamm Gottes, in the opening chorus. He'll never be forgiven by me for that gaffe.