Young Composers' Fugue Challenge

I know it's supposed to be an episode involving flying through as many keys as possible and sort of randomly playing around with your subject's head and tail, but other than that, I'm clueless as to what I should actually do.

No, the aim is not to traverse as many keys as possible. One of the aim of an episode is to quickly modulate from one key to another, among the set of closely related keys.

Any chance of some brief advice on what rules I should be following (bearing in mind I'm just about to have a cadence into the dominant). Do I need to get back to the tonic for the episode? What material do I base it on? What does it even include? What's with multiple episodes/developments?

After the exposition is done, there are no strict rules to follow. You can do whatever you want that pleases you and which remains "in the spirit" of the fugue: the subject must reappear from time to time, but episodes do not need to use the subject, they can use free material.

P.S I just listened to Brandon's treatment of the C-minor subject. I liked everything except the start of the exposition - I felt it got hugely complicated far to quickly. I think I'm more partial to expositions that get exponentially more ostentatious and dramatic as they progress.

Try to listen / look at the score of my C-minor fugue as well. You'll see that the exposition forms a true canon (each voice having the same material: subject, counter-subject#1, counter-subject#2, counter-subject #3 for the first voice, subject, counter-subject #1, counter-subject #2 for the second voice, etc...). But the first episode used to modulate to the relative major uses free material.

At the bottom of the first page, you will see an episode that is non-modulating: it's a purely diatonic sequence leading to a cadence in the dominant prior to a the subject statement.

Basically, episodes are there to provide a relief from the strict contrapuntal development of the subject. You can use whatever you want in episodes.

Good luck, looking forward to reading your score and listening to your fugue as well.

Raphael

Thank you - I think I shall wait until I've got something I'm pleased with until I post any sort of fugue here, because the exposition I was writing fell apart at the fourth entry (you'll hear exactly how it all went wrong in a second)

At the moment I'm just dabbling rather than composing seriously. I just need to improve my counterpoint and get a feel for what works and what doesn't (unfortunately my self-written subject had implied fifth-degrees in melody, so when I had the bass entry I ended up with unpassable 6-4s). . . . I was a bit of an idiot to expect that my own subject would be any good, but it's teaching me things all the same. Furthermore, the lack of a proper countersubject means that my attempt at a fugal exposition can't really be described as part of a fugue anyway.

I really enjoyed your fugues though - and having listened to Brandon's again it's really grown on me. My only complaint that remains was the lack of rhythmic diversity - but I suppose lots of this was to do with the midi file though. A decent registration would really make the voices properly audible. Also, there was a parallel when only two voices were sounding in bar 14, which was quite noticeable. It was fascinating to hear how both of you'd come up with totally different results from the same 'genetic material' as it were though. They are bizzare, fugues - a bit like kaliedoscopes that produces complex patterns by rearranging the same, few elements. Good fun though. I think I need to work on general counterpoint too, and have a go at using some decent subjects written by people who know what a 'decent' subject is. ;-)

Thanks.

Well - here's the failed exposition with my awful subject. My music teacher took one look at the first few bars and instantly knew that the piece would fail. Please be sympathetic to my first attempt using a subject of own devising, in future I'll take it from elsewhere. I should preface this form of audio-torture by saying that I've not actually done much contrapuntal stuff before; this is the first piece of baroque-style counterpoint I've written, besides stupid AS-level composition technique exercises. Hopefully there's small amount of redeeming material which indicates that in the future an exposition might not be beyond me:

fugue_edit.mid

Also, there was a parallel when only two voices were sounding in bar 14, which was quite noticeable.

Can you point exactly in which fugue and where in the score you see this parallel? I don't see any faulty parallel in bar 14 of the Cm fugue?

Well - here's the failed exposition with my awful subject. My music teacher took one look at the first few bars and instantly knew that the piece would fail.

I tried to reconstruct your subject from the MIDI. Without a score, I can't follow what you did (I'm a horrible listener, or a good one, depending on how you look at it...), can you confirm I got it right? I tried to reconstruct the meter and the tonality (D minor), maybe I got it wrong...

Sorry Ram! The parallels were in Brandon's fugue, not of yours! I should have made that clearer. I could hear no errors in yours, and although it was slightly less frenetic and exciting than Brandon's C minor fugue, yours had a better sense of rhythmic diversity.

You got the subject pretty much bang on. Here's a score for you:

http://img125.imageshack.us/img125/7016/fugueeditql2.png

OK, I see, you used a triple meter, and I used a duple one. BUT, with this slight change, I can harmonize your subject a little more easily.

Now that's very important. Whenever you invent a subject or work on another one's subject, the first thing you must do is sit down and see how you'd harmonize the subject if it were in the soprano, or if it were at the bass (i.e. which harmonies it would imply).

Can you try to do this exercice. Go back to your subject, and tell me how you'd do it. Given your meter, it's tough: you want to start with I but the E is going to be dissonant. And won't resolve as you keep the same note. Right from the start, you have a difficulty.

In my meter, things are simpler: I can use I over the first half beat and V on the second half, to finish with I on the second beat: the G will be dissonant, but an appogiatura of the following F, resolving naturally.

I did not know your meter, and I guessed "6/4" not really randomly but by looking quickly at how I would harmonize the thing and divide your subject without causing any harmony syncopation in my theoretical cheap harmonization.

Now when your subject is in the bass, things get a little tougher. But you can use an "exchange V 6-4" for the double E, if you consider the important notes in the bass to be (in capital) "D f E e g F f a". The soprano must then outline "F E D" to parallel the "D E F" in the bass.

If you don't like "V 6-4", you can think of E as the bass of a +6-3 chord, which is a "VII 6", and is more baroque in style: +6-3, like the exchange V 6-4 is connecting I and I6.

Anyway, can you finish this exercise first? Then you'll see that writing the counterpoint over your subject will be far easier and you should be able to redo your exposition in a convincing way in a matter of hours.

Thanks for your advice. I'll give it a try later today. The first step is really excellent advice - it seems obvious now, and I suppose for someone who writes without a computer it'd become obvious very quickly. Do you compose on paper, or by ear?

I use the computer as an electronic paper. But I have a piece of paper in front of my keyboard where I write down soprano-bass frameworks and harmonization sketches to see whether it will fit. However I fill the inner voices directly on the computer, unless treatment requires to be contrapuntal with one or several parts "obliged" in which case I write down most of it on paper first.

I've tried putting it into the time signature you suggested, and sadly it really, really doesn't work. I'm sure that theoretically it would be possible to harmonize the thing if you position the strong and weak beats in those locations, but it throws the subject too much - it sounds totally wrong and syncopated (I assume you meant to use dotted crotchet harmonies?)

I reckon I need to cast that last thing aside as an entertaining but ultimately failed experiment and use a friendlier subject. Any suggestions?

What about this harmonization? I never meant to use dotted crotchet harmonies!

[bTW, I think a slower tempo could be more appropriate -- I originally did the MIDI with 60 quarters per minute, but 40 could be nice as well -- what did you have in mind?]

[Further edit: I added a possible harmonization of the subject in the bass. Things are complicated by the fact that the A-D jump cannot remotely indicate a cadence in D, so you have to modulate before. I chose to modulate to the relative major for this passage. I included a 3-voice harmonization to make the overall harmony clearer than in 2 voices. I just noticed the bad parallel 5th between soprano and alto at the bar cross -- I did that too quickly, but you can simply move down the alto to F instead of Bb. Sorry for the wrong sample, but I don't know how I can easily edit the attachements...

Since the subject in the bass is problematic, you may want to start with the subject in the bass (the first entry is not harmonized by construction) and put the subject in the bass only once thereafter, or never at all afterwards -- it's not an obligation to have the subject re-appear in every voice]

harmonized_subject.mid

subject_lower.mid

Hi,

Most of the fugues on this thread are excellent, I've actually listened to them a few times over. I aim to get as proficient in the art as the examples here. Love fugues.

I tried one, using an untried subject, the Dmajor. It's not finished yet, but I'm itching to get feedback on it, as it's my third attempt ever at a fugue (first two were with the "puff the magic dragon" tune and another with a Shania Twain tune, neither of them got anywhere.)

And I'm not at all clear on a few things here:

Progressions, how do you clear up chordal progressions? What I do is I lay out bass lines on the main beats, ONE two three, ONE two three, in this case, and then use them to fill out a line. But I don't know how progressions should go, so you'll hear a few times in this that it seems to not know where it's going and I sort of "squeeze" it to where it should go. It can be interesting, but I'd prefer to know what I'm doing. So I guess my question here is: Are there definite rules to sort out progressions?

Another thing, I can't get it to sound baroque. It must have something to do with the embellishments? Mine just ends up sounding West Wingish.

And after my intro, (exposition?) I move into a piece (which sounds familiar to me, it wrote itself so I suspect it's half a bit of greensleeves or something), into a piece where I've laid out the bass lines for the polyphonic line, but I decided that I would leave it that way, the bass line just bumping along. Is this okay in a fugue?

I'm not at all happy with the part exiting this, where it leads to a cadenze before inverting the subject, should I go somewhere else with it?

I also wrote this with a sequencer, which is a habit I've gotten into, that's using the mouse to "draw" on the notes and the midi playroll. Good or bad, but anyway, I run it through a sampler and reverb and it sounds nice, especially on woodwinds. When I take it out and play it MIDI however, it sucks a bit. Is this something to do with the tonal quality? With a fugue I would imagine you would want the clear defined aspect of the different voices as opposed to a reliance on the blended tone, soft woodwinds, reverb etc.?

But all that aside, I'd appreciate some critique, I'm learning as I go along here.

P.S I don't like Shania Twains music.

3rd D Orchestrated.mid

Shaniafugue.mid

If you want to write something Baroque sounding, unless you've played lots of baroque music and have an ear for what baroque stuff sounds like, the common consensus seems to be that you start with something very simple, like a two-part invention, before attempting fugues. My first fugue sounded baroque, but was a technical disaster simply because the things rely on so much theory - bear in mind that Bach wrote the first fugues of the Well-Tempered Clavier whilst in prison, and without access to a keyboard of any kind. Indeed, writing Baroque counterpoint originally relied on sticking to rules, singing melodies aloud and writing counterpoint in a manner totally unlike the way I did it. . . which was just to try to do it by ear.

Without a score, it's hard to tell if you're following the basic rules correctly, but if you want to make your pieces sound baroque then I suggest you listen to lots of baroque music, and start simply.

And Ram - thanks for your harmonization attempts, I particularly like your suggestion of starting in the bass. I do however think that this one is now a lost cause, certainly the subject does not lend itself well to either stretto or inversion, since it doesn't really have a good set of cells. I've spoken with the teacher who's starting counterpoint work with me next term, and he's explained that we're to do it with no instrument, just a sheet of manuscript paper, a set of rules that must be adhered to, and through singing the melodies first aloud, and then in the head. It all sounds rather exciting, but horrendously complicated! The idea of writing music without a keyboard seems kinda cool though.

The two part thing would be a good idea. Something I've read on this thread will combine well into that exercise: playing eighths (sp??) in one voice while the other plays halfs and then swapping, like you find in Bachs two parts. It keeps the overall texture consistant in eighthththts while the voices individually don't etc.

I listen to a lot of Bach. Goldberg variations over and over and over, can't get enough.

Do you know a free program that can convert MIDI to sheet music with? I doubt I'm sticking to the rules anyway, aside from staying away from parallel fifths and octaves, and modulating to the dominant for the second voice then back to the tonic for the third. That's about all I know.

Thanks for the input.

Here's an mp3 file of my D major fugue. *I'm having no luck uploading the mp3..*

Could you post the scores somewhere, please? I can't get back to the score with the MIDI only, and without seeing the score, it's hard to comment on fugal writing. My ear is not good enough to reconstruct it by simply playing the MIDI.

Well, this is about the best I could come up with. If it's too much of an eyebleeder then don't worry about it.

@@

Well, this is about the best I could come up with. If it's too much of an eyebleeder then don't worry about it.

Thanks. It's clear from that score that your writing is not fugal at all.

That does not mean your work sounds "bad" at all. But simply that its form cannot be said to be a fugue.

I am myself trying to learn how to write in a more "fugal" form, and I've been through that stage of yours already, so don't worry: things can improve.

May I give you a piece of advice, that my teacher in the conservatory gave me last year when I chose to write a fugue as my final assignment (we had to write a short free piece, and I chose to write a 3-page fugue instead of the requested 1-page piece...)? He asked me to look at the C-minor fugue from Bach's Well-Tempered Klavier #1 (that's the second one).

You should do the same. Print the score (I can send you a PDF by e-mail if you need it) and underline the score, marking instances of the subject, of counter-subjects, labelling entries and episodes. Pay particular attention to partial subject statements (the head usually) and to possible inversion, augmentation, etc...

This study will help you understand the "internals" of a fugue, and how the polyphonic voices are constructed. Don't loose sight that although not the primary concern, the vertical harmonic progression resulting from the combination of the voices MUST also make sense. That's the sheer difficulty of the exercice.

Is this thread still going? Having a go at the F major (great subject). Will post shortly.

Yes Gregorious, two part inventions use lots of double counterpoint at the octave (I think that's the right term for rearranging the left and right voices in the manner you described), and I think it would profit you to start with them first, rather than trying to write a fugue yet. I feel a bit stupid offering you advice when I struggle to write even a fugal exposition myself, but I can write mediocre two-part inventions without much trouble, and writing good fugues is such a challenge that the best university in the world claims you'll be able to write a fugal exposition after a year of constant musical study there. . . . the problem is that what you wrote doesn't really appear to follow the fugual skeleton at all. Try taking a look at the wikipedia entry.

Ram's advice is good, take a look at some fugues of Bach, analyse them and see how they work. Keep listening to his music too, and have a go at those inventions. Since I don't think you'll want to take the theoretical approach, trying them should help you develop an ear for how to write baroque-sounding stuff. I hope this hasn't sounded out-of-place coming from me, but I think this advice will be agreed upon by most others (except the purists who say we should all start with a textbook, sharp pencil and no piano!).

Good luck.

TEXTBOOKS!!!

*sorry*

Thanks. It's clear from that score that your writing is not fugal at all.

That does not mean your work sounds "bad" at all. But simply that its form cannot be said to be a fugue.

I am myself trying to learn how to write in a more "fugal" form, and I've been through that stage of yours already, so don't worry: things can improve.

May I give you a piece of advice, that my teacher in the conservatory gave me last year when I chose to write a fugue as my final assignment (we had to write a short free piece, and I chose to write a 3-page fugue instead of the requested 1-page piece...)? He asked me to look at the C-minor fugue from Bach's Well-Tempered Klavier #1 (that's the second one).

You should do the same. Print the score (I can send you a PDF by e-mail if you need it) and underline the score, marking instances of the subject, of counter-subjects, labelling entries and episodes. Pay particular attention to partial subject statements (the head usually) and to possible inversion, augmentation, etc...

This study will help you understand the "internals" of a fugue, and how the polyphonic voices are constructed. Don't loose sight that although not the primary concern, the vertical harmonic progression resulting from the combination of the voices MUST also make sense. That's the sheer difficulty of the exercice.

I've located a copy of the fugue at mutopia.com and will do that. What I've seen so far has me realising how much preproduction goes into making a fugue - I just jumped in.

Thanks for the advice, appreciate it. Waiting to hear more fugues..

F Major Fugue. There's a little problem in the d minor stretto entry which I will correct when I have time.

MP3

Why have you shouted Textbooks? I can't afford Textbooks! (Although I shall be getting some after exams. . . full of tasty fugue subjects and contrapuntal advice. Buying counterpoint textbooks also gives me an excuse to march into a bookshop, find the most attractive store-assistant and ask loudly "I'm looking for Fux!")

Get them off amazon my friend, that's where I get all of my books, cheap, and very fast delivery :(

You've got a score Praeludium and Fugue? I wanna take a look at it.

You've got a score Praeludium and Fugue? I wanna take a look at it.

PM me if you want a scan of the original paper and pencil score. I just copied it into Sibelius for the sound file but the voices and rests are not properly written out there.