Piano World Home Page Forums Our Most Popular Forums Pianist Corner Help approaching Bach's fugue BWV 849 and fugues in general

I am looking for experienced advice on a good way to approach a Bach fugue, or specifically the fugue in C# minor from WTC 1. I'm about to start this to pair it with the prelude and fugue in C# major, and I have to play both in four weeks. The C# minor would be my first 5-voice fugue, and it's an intimidating work. I still lack a proper system in learning a fugue and feel that there is always something random in the process, a lot of time wasted at the piano, especially for memory, and I wanted to ask others if they had a particular method. With the last two I learned (E major and C sharp major from book 1, both 3 voices), I've gone in sections (stopping e.g. at the end of the exposition, or at a cadence), learning each voice, then putting them in pairs, and finally together (also singing one and playing another, singing/ playing one and ghosting another, etc). But it was not entirely systematic, and something was still always physical when all the voices where put together. With BWV 849 I'm hesitant to try to play through first or to start at the piano too soon, but it's much more difficult to visualize without playing. It's also harder to mark sections or to know where to stop.

(I've played the E, Cm, Fm and Gm pairs from WTC 1 in the past, but most of it I don't consider to have learned properly. Last year I played more inventions and other baroque music in graded volumes, in preparation for approaching polyphonic music with more confidence, but I still feel very unqualified- aware that this particular fugue might be too ambitious at the moment).

4 weeks is pretty short...but you probably are a music student who spend a lot of time so I think it's doable. It took me about 4 weeks using several hours a day.

I have never really formally studied the historical norms of Baroque music but a harpsichord professor explained this fugue to me as such:

You don't have "a theme" (1: C#, B#, E, D#), or even a "secondary theme" (2: G# C# C# C# B# A# B#), the main body of the fugue is actually those two plus the (3: G# F# G# A G# F# E G# F# E F# G# F# E D# F#), and you get 5 choose 3 = 10 variations of it in different voices. He told me that in baroque fugues the "theme" is really this multi-part combination, not any particular one of them. This only starts on the 3rd line of the 2nd page. Having that pointed out to me made it much easier to remember what's going on.

Honestly it's a fairly long fugue, you have to drill it over and over before you can have it memorized. I find away from keyboard studying of the score helps memorization too.

It is a good thing to make a (secret) photocopy of the fugue and annotate it with points of analysis, e.g. appearances of the subject, countersubject, episodes, inversions, retrogrades, augmentations, diminutions, tonic or dominant pedals and so on. This helps to make it stick in the mind and gives a clearer picture of Bach's thinking when he wrote it.

An interesting technical idea is, whenever the subject occurs with other material in the same hand, practise the subject forte, legato with arm weight and a strong finger action, and the other material quietly, staccato with no weight and a weak finger action, hardly lifting the fingers at all. Playing staccato of course ensures that no arm weight is rested on the keys for these particular notes, so that all the weight is resting on the keys which are playing the subject.

Ah, making the subject stand out. I can hear the harpsichord professor getting angry about that.

That said, when you practice it is useful to try and pay attention to different voices each run through, and if playing them louder does it for you that is a good idea. Generally speaking for technique it is probably good to learn how to make certain notes stand out in a thick texture. If you can, try to "hear" 2 or 3 simultaneously, and switch up which 2 or 3 you are trying to pay attention to.

On a single manual harpsichord one uses articulation and phrasing to bring out the subject. Just because it is polyphony doesn't mean that some voices are not momentarily more important than others. On one of his organs, Bach actually wrote down the registrations he used for his preludes and fugues. From those stop lists we are able to see how he thought about polyphony and how he brought out individual lines in complex fugues.

Bach's favourite instrument was the clavichord, through which he could more easily and more natually bring out important polyphonic lines than he was able to on the harpsichord.

As I said, I don't pretend to be well-learned on this topic, but just pointing out that a famous harpsichordist/Baroque music professor told me this. Indeed, you can bring out voices through subtle differences in touch, articulation and such, and what he told me was that you should actually bring out the "glue" voices that are not the subject because they are more interesting as they aren't repeated. Indeed, why are there so many voices if we only want to hear a single boring 4 note subject over and over?

The reason I bring this up is that if you fixate on the "subject" you see in the first few bars as many piano teachers teach, you end up ignoring the rest of the structure which makes it harder to memorize. Intuitively, one's fingers would recall this structure once the piece is learned thoroughly but consciously noting the pattern is helpful. I find that after learning to look at BWV 849 this way, I felt much more secure about my memory of it, while before that I would just practice it over and over until I felt like my fingers remembered all the notes. There is much more structure to this piece than most pianists notice, IMO.

For this fugue in particular, I have noticed that the 3 part repeating polyphonic subject is usually voiced so that the C# B# E D# motif is either in the lowest or highest voice most of the time, making it naturally stand out for example. The eighth note motif is easy to hear as there are many stepwise moving notes, so the G# C# C# C# B# A# B# motif can use articulation to make it not lost.

I agree with you and him. Bach's fugues, in particular, have an intellectual and musical depth that needs careful study, far beyond just the subject. But, we need to occasionally hear the subject, just as we need to be brought back to the base melody in jazz improv, to ground us and make sense of the whole.

I feel that most of the interest of a fugue lies in the composer's ingenuity in presenting the subject (and often the countersubject also) in many different metamorphoses and polyphonic contexts. The only way to make this structural element clear to an audience is to emphasize the subject to an extent, although not so much as to obscure the surrounding polyphony. If we fail to do this, the fugue is likely to take on a somewhat amorphous and directionless character. Bach certainly did have the means to do this, especially in terms of the Clavichord, and as we now have the advantage of a modern piano, with its great tonal variety, I see no harm in using it to best effect ! It is certainly very standard practice on modern performances of fugues.

The difference in dynamic between the subject and the other voices can be quite subtle in performance. But in practising, I agree with Neuhaus's idea that dynamic differences in a texture should be greatly exaggerated. It is then very easy to scale down from this position to a more subtle approach, whereas to do the opposite would be very difficult.

I think the "rule" that the performer MUST emphasize each entry of a fugue's subject (and, furthermore, playing it with exactly the same inflection each time it appears) is a questionable one, even if it is pretty common. To me, there's something unnecessarily didactic about it, and the results for the listener can sound didactic, as well. I've heard far too many performances of fugues where the effect of underlining the subject ends up feeling like the performer is trying to teach me the basics of how fugues are made, which is something I already know and don't need to relearn every single time a fugue is played.

And too, sometimes the effect is weirdly self-congratulatory, as if the performer is showing off their skill at foregrounding one voice in a complex texture. That is a fine, advanced skill to have, but it is rarely the point of the music. Of course, if the audience is an academic jury, this kind of performance may be exactly what they want to hear, so doing this kind of thing might be appropriate in that situation.

To me, the important thing to remember is that fugues, while they are music of a certain special sort, are still music. I think it is useful not to unnecessarily restrict the imagination by imposition of an over-riding formula when working out an interpretation of them. Their musical variety is vast, and the range of musical personalities playing them is also huge. An emphasis on how they are put together may be one valid way of playing them, but it's hardly the only way, nor is it necessarily the best way. Said differently, the method by which the music was composed doesn't always need to be the one of the main points of the performance, not for the performer nor for the listener.

Wow, that just about summarizes my feelings toward this topic. Nicely put!

Thank you all! So much to think about.

trigalg693- thank you so much for your insights and specific tips. Very true that there's more structure than is immediately obvious- this is exactly what I'm trying to find.

Paulwhite- was that Cortot? He gives a similar practice advice in his notes to Chopin's Op. 10 no.3, and he says it applies above all to the music of J.S. Bach.
Prout- this is really interesting about Bach's stop lists. Would love to look further into this!



wr, I agree. Sometimes, though, I do think it is important to finally arrive at how a fugue (or any piece?) is put together to interpret it thoughtfully. Isn't performance a reconstitution? You start with the whole to arrive at the skeleton. Is it important to arrive at the skeleton? I think it is, if one wants to get the most out of the music, and (especially?) if one is not only playing for one's own enjoyment but wants to responsibly translate notation into sound and deliver it to other people. I've been trying to liken this to analytical reading, specifically to Mortimer J. Adler's approach to reading in his masterpiece "how to read a book" (it would be fascinating to systematically apply this to reading (/playing?) music. But it's precisely this difference between "reading", "playing", "listening to" and "interpreting" music that I need to understand and define better). With Bach's WTC at least, I always feel like I'm missing something by just playing it. Composers taught themselves by studying it. Beethoven could play it when he was 11 ("to put it in a nutshell, he plays chiefly The Well-Tempered Clavier of Sebastian Bach, which Herr Neefe put into his hands. Whoever knows this collection of preludes and fugues in all the keys- which might almost be called the non plus ultra of our art- will know what this means"). Nadia Boulanger knew it by heart by the time she was 12, and she made all her students memorize a pair every week- to be able to write out the separate parts by heart, then to reconstitute the whole. Schumann advised to make it one's daily bread. "Everything is there," Brahms said. It was, after all, written "for the use and profit of the musical youth desirous of learning, as well as for the pastime of those already skilled in this study" (i.e. the study of fugue). It is, first and foremost, beautiful music, as you say. But should "studying" it by separate from "playing" it? Should that be left to the composition classroom? I want to enjoy it and to play it as music, but also to get the most out of it, but I often feel extremely unqualified. To me, the issue is not so much "what to do" with a fugue, but how to start learning it. I feel like I'm always improvising a way, and there's always a lot of time lost. I really believe in the importance of learning away from the piano first, but it's something I'm new to, and I'm still teaching myself the discipline for it and don't have a system, especially with a fugue. Should I learn each voice all the way through? Then pair them together? Going all the way through voice by voice doesn't feel right, because I feel like the voices are not independent strands that are later superimposed; they affect and are affected by one another, interact and intertwine flexibly, respond to one another almost spontaneously- at least it should, or can, sound that way. Do I learn all pairs then the whole section by section? It's incredibly difficult to hear more than 2 horizontal voices simultaneously without the aid of memory or physical sound. I guess it will take at lot of trial and error over time with a lot of fugues, but I wanted to know if anybody had a way they found successful or efficient.

I wasn't at all advocating that the performer shouldn't study and analyze fugues, but just saying I didn't agree that there was one specific way of performing them that was the only right way.

As far as your issue of how to learn them is concerned, you probably will just have to try various approaches until you find what works best for you. Although for I've been systematically playing through the entire WTC during the course of a year for a number of years now, I've never tried to memorize any of it, because I have no need to. And besides, I don't think memorization was a common practice in Bach's day, anyway. He might well have been astonished that anyone would attempt to memorize his fugues.

The most recent Bach I have tried memorizing is the F major invention, and I did it specifically as a memorization exercise. I have never been good at memorization and this was a challenge I set for myself to try out the technique of writing out the music away from the piano, as a way to get it solidly into memory. I cheated a bit by doing it mostly in my head, rather than on paper, but I think the principle is pretty much the same. And although it was very difficult, especially at first, it really worked. The difficulty at the beginning was about understanding just what was required to really and truly get those notes set in my memory so I could later reproduce them. It was a strange and humbling experience, but worthwhile, and rewarding in unexpected ways.

Oddly enough, a couple of days before you started this thread, I was thinking about Bach and why his music is so hard to memorize, for many people. One major thing that always comes to mind when I think about this stuff is that we simply aren't used to thinking/hearing contrapuntally, at least not in the sense that Bach was used to thinking. I feel like I'm finally, after over 50 years of playing, getting a bit of a feel for it sometimes, but Baroque fugues definitely are not a "natural" mode for me. Which just makes sense - I may love the WTC, but I'm also quite aware that it comes from fairly remote reaches of the history of keyboard music, before anything like the modern piano even existed. The more comfortable I am playing it, the more it feels like there is a subtle gear shift in my brain that happens when playing Bach fugues. It's clearly a special kind of musical thought that did not originate with a single person at a keyboard, and is one of the most "artificial" ways I can think of for an individual at a keyboard to make music. Sometimes I think it is so far from our own era that the only musicians who really should be playing this stuff in public are ones who are immersed in it full time, and who don't really play more recent music.

Thanks Rania ! I wasn't aware that it was Cortot, but he may well have said that ! It would certainly apply to Chopin Op. 10 no. 3 (the right hand melody and accompaniment), and also to any music with a melody and accompaniment in the same hand. Your observations about analysis are very interesting and valid. As a composer who plays my own compositions a lot, I find it unthinkable that anyone could perform music without understanding its structure and musical language. That would be like someone reciting a story in a foreign language which they don't undersand - just a lot of meaningless syllables ! It would also be very hard to memorize a story without understanding its meaning, and I think the same is true of music. Whether the audience understands it or not is up to them, and it will vary with each listener, according to their musical background, but I feel the performer should certainly have analysed at least the bare bones of it. This is not a dry academic exercise. It is a way of trying to get inside the musical imagination and thinking processes of the composer.