I'll try to answer to the various points that arose.
First:
Your Summitship. I know you are teasing Bat (and he deserves it). But Beecham had in mind the horrible Pleyels, Gaveaus and Ehrads of Wanda Landowska, which are actually horrible. A good harpsichord is bright, but never dry (if played correctly: no recorder thing!

). That said, using a harpsichord is no guarantee that you'll perform well. On the contrary, it needs a lot of control on micro-agogics which is quite difficult to master. But you know all that
The cello suites.
Dear Bat:
No, a one fingered harpsichordist could not play the suites: you would need two hands. But that is not the point. The point is, as many have mentioned before, the music is subtle, very moving and is, in many cases, polyphonic (there actually is a fugue, in, if I am not mistaken, the 5th prelude). So get yourself a good version and listen until it has grabbed you: at one point you will understand the pieces.
As a matter of fact, I think that this kind of music, exactly like d'Anglebert's, is the best way to learn to really like music, because it is all about singing and declamation. Music, higher music at least, is a combination of vectors, curves, tension, release, balance and unbalance. You will find this almost in its pure form in the cello suites. I would go as far as to say that if you do not like the cello suites, you won't like much of what Bach has written.
And this leads me to:
Dear Pe-Zulu
You mention the importance of the pulse. Many musicians defend just that. Peter Hurford and, especially, Walcha come to mind. But even Walcha, who stressed the importance of the pulse, very often transcends it. In fact, while maintaining the pulse, he actually structures the works in terms of very long stretches. This is perhaps more evident in the organ fugues. He is metrical, but the pulse is never stressed. For instance, the b minor fugue (second reading) almost floats in the air.
I would say Bach's music is, in many cases, just what you said: a long and boring succession of semiquavers; what makes it interesting is precisely the overall structure. In the F major toccata the interest is precisely in the confrontation of two very different sections: the semiquavers and the chords. What is miraculous is the way he can keep the interest in such a monotonous way. Of course one cannot ignore the fact that the semiquavers are close to being boring, and therefore one must avoid it, either by not stressing the beat or by differentiated articulation, as you suggest. But this is, I think, only the beginning of the interpreter's difficulties: what is really difficult is to make the semiquavers as such unimportant: just a melodic and rhythmic pattern that will be combined with the chord pattern.
As this combination is not easy, all the interest of the music is how the two patterns will be integrated. Well, at least this is why I like the toccata.
If you favour a structural approach, I think this is the way. But you may play it like Koopman's first rendition (in the Garrell organ), beat by beat. And this is actually boring, I agree.
Is this to say that I favour really long phrases, in a romantic way? Yes and no. Yes, I favour long phrases; but not in a romantic way. A long phrase is rhythmically and agogically determined (for instance, in the doric toccata, ta.ra.da.ta.ta.ta.ta.ta.ta.ra.da is a possibility) and Bach very often uses a kind of rhythmic and melodic unit (the Orgelbüchlein is all about that) as a building block to the music. So what is important is actually this unit, and one must be able to change it according to the moment in which it occurs within the work.
A very good example here is the C major fugue (of the toccata, adagio and fugue): the theme is actually a kind of spring, which you can flex or loosen by and by.
However, Bach actually uses longer phrases: many fugue themes are a good example (for instance, the doric, but almost any of them is a good example). As an example, I recall the g minor fugue (from the fantasia and fugue): there are parts which completely flow (semiquavers in both manuals and pedal), and when the theme appears, it structures the flow.
This is becoming boring because I cannot actually provide the music. But what I want to say is that the themes structure the otherwise flowing (longish phrases, although rhythmically characterized) parts.
So, I would not agree that the beat is important in Bach: only inasmuch as it allows one to understand the basic rhythmic-melodic unit.
On the registration I suggested for the a minor prelude. This is, of course, a rather extreme example. But it is not completely far fetched: there are actual examples of very low bass registration with somewhat mild upper work. Anyway, what did I mean by it: the prelude may be viewed as a disorganized flow of one voice that is progressively structured by rhythm. So, if playing the beginning rather freely, slowly and then introducing a 'hound of hell' in the bass, you create the despair and horror of the beginning; then you must progress to some sort of plenum. Anyway, regarding registration, I think I will post my feelings in another thread. (Just to bring LordSummiT to complete despair) :RdS: