Bach on modern piano
This text was originally prefacing an edition of selected pieces out of the "Klavierbüchlein für Anna Magdalena Bach", as a theoric introduction to the didactic purpose of the publication. The original text is presented here in a corrected, slightly revised version. Original source: Klavierbüchlein für Anna Magdalena Bach - Minuets and polonaises - ROLAND AM 91795 - © 1993 EMI Music Publishing Belgium/Roland Benelux Music Publishing.
Interpreting piano works by Bach and his contemporaries on modern pianos
I will not even try to count the numerous versions, theories, musical studies, and other analysis that the bewildered pianist of this century feels compelled to plough through in order to come up with a logical interpretation of the works of J. S. Bach.
From the short-sighted absolutist versions demanding a simple transfer, without any changes, from the harpsichord used by Bach to the Steinway grand, to the widest instrumental amplifications, overblown Lisztian paraphrases, all possible formulas have been tried and heeded to, only to be rejected when they no longer met the taste or fashion of the times.
There is no such thing as an absolute truth in this matter. The most faithful approach would actually be to play Bach only on those instruments that were around in his days - but that woud mean that one had to be absolutely sure about the way they were played in the 18th century. The few authentic indications that have stood the test of time are far to elusive to be of any help. Conversely, it would be a pity to withhold this essential part of the repertoire, which baroque and pre-classical music are.
To my opinion, the most important thing is to follow as many concrete elements as possible, inasmuch as they are related to the instruments for which these pieces were written, so as to adapt, in the most faithful way possible, your technique to these works.
Such an approach should permeate all parameters related to the interpretation, including a fresh reading of the original text, and trigger a reflection on what some people call "the composer's gesture". Let's briefly look at those parameters.
Bach gives no hints concerning the tempo of his pieces. His indications usually refer to formal or stylistic aspects, such as "Prelude", "Polonaise", "Gigue" etc. The pianist thus has to decide for himself and rely his aesthetic understanding. You should, however, shy away from extremes. An exceptionally slow tempo will make it almost impossible to detect the continuity of the musical phrases, whilst too much brio tends to reduce the music to a mere technical showcase.
With the exception of the rare "Fantasias", the tempo should remain constant from start to finish of a work or movement.
The possibilities of contrast, virtually non-existent at the time of baroque instruments do not come anywhere near the expressiveness of even the blandest present-day piano. The harpsichord, for instance, will always respond in the same way, whether one plays pianissimo or fortissimo. The clavichord, on the other hand, allows for rathersubtle nuances, yet it is worth keeping in mind that dynamics effects meant so little to Bach that he didn't event bother to write them down. To him, expression is closely related to register, intervals, as well as harmonic and contrapunctal structures.
A 21th century pianist thus has to keep in mind the original point of view and should therefore use the dynamic possibilities as a last resort, and always in a subtle, passive way, as they are of less importance than the other means of expression, such as phrasing and rubato. You should, however, shy away from overtly mechanical interpretation of the works.
The only acceptable effect in the baroque context seems to be the technique of echoing reprises. Indeed, quite a few instruments were equipped with two keyboards, or registers (similar to those of organs) that were operated with levers and allowed to switch from single string mode to multi string mode (which you could call the ancestor of the chorus effect). These switches can be effectively simulated on the piano, but beware of overusing this technique. Every instance should be preceded by the question of the Wise Musician: "Is it really necessary?"
Phrasing (legato, portato, staccato)
Now we come to the most important aspect for the interpretation of baroque works on a piano. In keeping with our principles, we rely on the few indications we have, more specifically the resonance capacity of the original instruments.
The adsr profile of a baroque instrument and a piano are very different, especially regarding the values of decay and sustain, much shorter for the baroque keyboard.
It ensues that the difference between legato and staccato is less obvious on a baroque instrument. Hence, if the following excerpt was played very well legato by both hands on a harpsichord, the left hand notes would be perceived as a reverberated portato because of their distribution in time:
If the same legato was performed on a piano, the musical phrase would lack definition.
It is therefore necessary to adapt your style in order to maintain the transparency of the original instruments as much as possible. Here are the theoretical solutions I can provide for your interpretations.
The majority of Bach's works are based on an underlying principle which is related to the almost constant use of two basic values. The first, which we shall call speed, is the shortest basic duration. Speeds should always be played legato, unless that is technically impossible, in which case you should still try to play them portato. Even faster values (usually transitional notes or grace notes) should also be played legato.
The other basic value, which we shall call pulse, is twice as long as the speed in binary rhythms (e.g. speed=16th note, pulse=eighth note), and three times as long in ternary combinations (e.g. speed=eighth note, pulse=dotted quarter note). Pulses are habitually played portato or staccato, according to the tempo or the type of the work. Playing a pulse legato has usually expressive reasons.
The remaining note values are to be played legato. They are, by definition, values with a special expressive connotation that stick out from the motoric context (refer to the chapter about ornaments). The duration of notes followed by rests has to be strictly respected.
Of course, every rule, no matter how reassuring and logical it may seem, has a lot of exceptions. The only principle that suffers no exception is the clearcut distinction between speeds and pulses.
Occasionally, and for reasons of expressiveness, I suggest exaggerating the legato style, by releasing a note only after playing the next one.
You won't find many phrase marks in Bach's manuscripts nor in those of his relatives and friends, and if there are they tend to be more of a problem than of any help. Certain slurs seems so strange that I simply cannot believe they serve as indications to play legato.
A lot has been written about ornaments in baroque pieces. Most editions of Bach's piano pieces contain the short explanatory table of the Klavierbüchlein für Wilhelm Friedemann Bach. Quite a few of them add some complementary examples, and almost all of them state that these comments are a crucial element for the interpretation of baroque works.
I think it is a good idea to remind the reader of the fact that, in those days, ornaments (as to their frequency and diversity) were usually left to the performer's discretion. It would have been rather incongruous to abide by the composer's indications as they were mere suggestions, and as such no more than the fingering in present-days editions.
Ornaments are inflections, the process of emphasizing certain notes, in short: the equivalent of an expressive accent. They often serve as technical accents, or as elongating tools used to fill the gaps between notes that are supposed to be held for a long time, in order to preserve the continuity of musical phrases. In the light of all that has been said so far about the difference in resonance between instruments, you will have little trouble to accept that it would be awkward to play these ornaments, that were originally intended for a harpsichord, on a piano. Playing them would actually constitue a serious modification of the musical phrase, that tends to become blurred and sometimes even completely obscure. I therefore suggest to simplify or even omit these ornaments, while remaining faithful to the original idea. Hence, rather than embellishing a pulse, you should slur it to the next note. This simple modification suffices to introduce the inflexion symbolised by the ornament the composer had in mind. That comes down to slightly oversluring a note in a legato phrase or to using incidental rubato.
With the exception of the rallentando in the coda, which is perfectly natural, I believe in a strict tempo because it is in keeping with the motoric principle I mentionned before. Let us not forget that extremely useful tool referred to by some as "horizontal accent", i. e. hardly perceptible fermata that gives a note the inflection it needs, provided that the phrase is played with the utmost stability.
This incidental rubato successfully manages to take the place of an ornament on a speed.
Finally, I should like to point out that rhythmic stylisations, especially for characteristic dances, should never be mistaken for ad lib rubatos. Thus, the rhythm of a Polonaise works better when you play the 16th notes slightly tightened to the next note.
I never use the sustain pedal, nor the sostenuto pedal, typical features of the romantic and more recent periods, out of place in a kind of music where everything that needs to be sustained, is held by means of the fingers. The soft pedal, on the other hand, allows you to contrast two timbres and is thus a performance tool similar to the registers, perfectly natural in the baroque context.