The year was 1978 and I had been assigned the G major French Suite of J. S. Bach by my piano professor at the RCM. I duly went off to the Kensington Music Shop (which is still there by South Kensington tube station) to buy the Henle Urtext edition, and then found a practice room to explore. Uncertain as to the exact meaning of the ornament signs that littered the pages, I decided I ought to listen to a few recordings from the College’s record library only to discover that each performer did them differently. So what was a poor undergraduate to do? In those days, I assumed that anyone good enough to issue a commercial recording of anything had to know what they were doing, so I was bemused and confused by what I had heard.

It struck me that perhaps all of these different versions of the ornaments were OK, and I could just do whatever took my fancy. Somehow this didn’t seem quite right, surely there had to be some sort of difference between the squiggle with the line through it and the one without it. Since it was going to be a whole week until my next lesson, and I wanted to take at least the first two or three movements along, I thought I had better ask around. Accosting members of staff in the hallways, one eminent professor of piano told me one thing, while another said he thought it should go like this (there ensued a whistling session) and I was left none the wiser.

This was beginning to really trouble me! My studies of Shakespeare and the bible at school had impressed upon me that quibbling over textual matters was important, because it could change the meaning of something if you misinterpreted it. In my quest for truth, I toddled off to the main library and expressed my concerns to Dr. Harold Watkins Shaw (Librarian-in-Chief at the time, and famous for his still-revered edition of Handel’s “Messiah”). I remember him as a very kindly, besuited man who was happy to help students, and he produced the only ornament table J.S. Bach left for posterity, the “Explication” from the preface of the Clavierbüchlein vor Wilhelm Friedemann Bach.

This didn’t help much, because it tells only half the story (but how was I to know that?). Did the trillo mean I had to play that exact number of notes in all cases? What if it was over a dotted minim rather than the given crotchet? There were so many notes to fit in and sometimes there just wasn’t room. Bach hadn’t addressed all these possibilities!

As I later discovered, an ornament table will give the basic shapes and designs of each type of ornament, and assumes you will tailor these to specific circumstances. You also need to know that, provided you stick to the basic outline, you can add and subtract repercussions, and bend and inflect the ornament rhythmically to make it fit its surroundings. This is very important! Ornaments should sound free, personal and improvised, and the military precision of the notation in ornament tables is misleading (how else could the composer have done it though?) given that the final result is not possible to notate precisely. And it shouldn’t be carved in stone either – if you are feeling particularly expressive you might want to lean longer on that first note of the trill, to get the maximum amount of juice from the dissonance. These things vary from performance to performance, depending on mood, whim, the instrument, the acoustical space and how many cups of coffee you had for breakfast. You can, of course exuberate in the repeat by adding even more notes to a trill, and adding more ornaments and embellishments (a different thing, but let’s not get into that here). Let only good taste be your guide.

Why do I get so bent out of shape when a pianist consistently plays a three-note Schneller (or “upper mordent” if you’ve been reading a certain familiar 1970s theory book, now properly updated) in Bach? Is it irritation that they haven’t bothered to do their homework and read the instructions on the packet? Do they think ornaments are like manners, which change from time and place? Are ornaments subject to evolution like everything else? Is it because the modern pianist is a nineteenth century animal, from an age when players happily did what they wanted (this freedom extending to adding or subtracting notes the composer had written)? Certainly “text as dogma” has clouded the issue. If you knew your teacher was brought up to accept Edward Dannreuther’s decree that it was simpler for all trills to start on the note, regardless of period, then you might be able to question this in light of modern scholarship (throw it out, more like – not questioning your elders or betters is the main problem here, of course). No, for me a Bach trill (a French trill) is a chain of two or more appoggiaturas. The dissonance comes on the beat, or on the strong parts of the beat, and the resolution on the weak parts of the beat. If the trill starts from the main note or does not coincide with its bass, you lose this. The trill may elongate the note, may accent it, may be expressively melodic or zingily rhythmic but it begins on the upper note and on the beat. All eighteenth century sources stress this with monotonous regularity.

I think there is still a lot of confusion around the tied trill (or tremblement lié) and C.P.E Bach’s Schneller, or “upper mordent” (it amuses me that people quote C.P.E chapter and verse and apply it to music of his father, but J.S. Bach came earlier of course, and C.P.E had embraced a very different style). The tied trill may sound like a three note ornament that starts on the beat (and might be indistinguishable from it when fast) but it actually feels different. A tied trill is usually notated by a slur that connects the trill to the previous note, when the first note of the trill is the same as the preceeding note. Rather than replay it, it is tied over. You may think this is splitting hairs, but actually you still get the sense of appoggiatura, of dissonance and resolution but in a more legato way without the rhythmic accent. The repetition of the upper note is not to be avoided in a rhymical or non-legato context – this reiteration adds even more emphasis to the dissonance. Whereas F. Couperin was meticulous in notating his tied trills, Bach was less so.

To wrap up this potted history of the trill, remember the year 1828. This was the year of the death of Schubert, and the publication date of Johann Nepomuk Hummel’s Anweisung zum Piano-Forte-Spiel where the instruction is to begin trills on the main note. Here is Dannreuther’s translation of Hummel’s text, and presumably the start of pianists’ misunderstandings:

In the next post, I would like to explore the “how” of achieving these ornaments beautifully and skillfully at the piano. It took me some years to figure it out, for it to become reliable and natural, even on so-called heavy-action pianos. I will share it with some video links I plan to record over the next few days.