Results 1 to 3 of 3

Thread: How to figure out if I have a good subject or not, is my strategy good?

  1. #1
    Senior Member caters's Avatar
    Join Date
    Aug 2018
    Post Thanks / Like

    Default How to figure out if I have a good subject or not, is my strategy good?

    So, here is my proposal for how to figure out if a melody that I have written down will fare well as a fugue subject. It all has to do with stretto. Not that stretto is important for a fugue, because even Bach rarely uses stretto in his own fugues. But the more strettos I can do and still follow the rules of Baroque counterpoint, the more I can use the subject to develop the fugue. And the more I can use the subject to develop the fugue, the longer of a fugue I can get that stays interesting.

    So here are the rules that I know of in Baroque counterpoint:

    1. No parallel 8ves or 5ths

      Pretty obvious why. This gets rid of all independence and makes it sound more like chords than melodies.

    2. No parallel 4ths because they will invert to parallel 5ths(well, unless you aren't inverting at the octave)

      Again, a pretty obvious rule.

    3. Parallel 3rds and 6ths are fine, but don't use them too much. About 4-5 parallel 3rds and they won't sound independent anymore.

      This one isn't so obvious because parallel 3rds and 6ths sound nice, especially if the 6ths are major 6ths. Not as much if they are minor 6ths because in some ways the minor 6th is a dissonance, namely an enharmonic equivalent to an augmented fifth and anything augmented sounds dissonant except maybe the augmented second or augmented third.

    4. Rarely use an accented dissonance unless the dissonance would naturally be accented

      So an appogiatura would naturally be accented so no problems there. A passing tone would generally not be accented. If you have to use an accented passing tone, don't use it on beat 1.

    5. Resolve 2nds via oblique motion to 3rds

      Pretty obvious one, because the other potential resolution to a unison is well for 1 thing, impractical outside of vocals or instruments with only 1 staff, and for another thing, it gets rid of independence just like parallel 8ves do.

    6. Resolve 4ths to 3rds

      You can't really resolve them to fifths because scale degree 4 has a downward gravity, not upward.

    7. Resolve the tritone via contrary motion to a 3rd

      The only exception to this that I know of from Bach is from towards the beginning of the fugue of Toccata and Fugue in D minor where Bach suspends the tritone by having a rest after the harmonic tritone(or in some editions, just a half note tritone), and during that rest having that same tritone but melodic before finally landing on an octave in the left hand which goes to a fifth in D minor harmony.

      Right here:


      Measure 73 has the tritone suspension and Measure 74 has the resolution of that tritone.

    8. Resolve 7ths to 6ths if they dont involve the leading tone

      The 7th outside of the leading tone, tends to have a downward gravity.

    9. If the 7th does involve the leading tone, resolve to an 8ve

      No choice here really, unless you want to leap in the lower voice. But if you want to keep it stepwise, the only choice here is going to an octave.

    10. If you reach an 8ve or a 5th, use either oblique or contrary motion to resolve it depending on the harmonic context.

      Pretty obvious, have to avoid parallels here.

    11. If you have any dissonant interval that involves the leading tone, resolving the leading tone takes priority over resolving the dissonance.

      As an example F# to B is a 4th, B is the leading tone in C major and fourths are to be treated as a dissonance so resolve to F# and C. Now you have a tritone that you have to resolve. In the context of C major, resolving it inward to G and B, a 3rd, makes harmonic sense.

    I'm sure there are other rules but having just these 11 rules helps make things easier.

    Here is my proposed strategy to see if I have a good fugue subject from an improvised melody:

    1) Simplify things by not including stretto at the second or the 7th. Realistically, these would be harder for a pianist to get across and my fugues are composed with the pianist in mind. On top of that, the stretto at the second risks voice crossing which while not your enemy in a piano fugue, I would rather not risk for my subject entries since in that case, the melodies might sound fragmented instead of sounding in full. This might be great for the episodes though, occasionally having voice crossing to make the melodies sound fragmented.

    This leaves me with these interval strettos:

    • Octave
    • Sixth
    • Fifth
    • Fourth
    • Third

    2) Since the fastest notes in my fugue subjects tend to be 16th notes and I have quarter note beats, check every half beat time interval up to and including 1 measure of overlap. Shorter than a measure of overlap and it might as well just be a canonic moment rather than a stretto. The ones left are possible 2 voice strettos where both voices are playing the original subject.

    3) Check this again with every 2 voice combination of original, inversion, retrograde, and retrograde inversion. Now you have all the possible 2 voice strettos in the fugue.

    4) Extend this process to 3 voices and if I plan on having 4 or more voices, continue extending it for as many voices as I have planned.

    5) Once I do this, I should check to see if anything changes as far as possible strettos with augmentation and diminution.

    6) Stretto table finished for subject, if I want, do this same process for 1 or more countersubjects.

    Will this work or is there something that I am glossing over? Will this be too labor intensive, especially considering that I have to make sure that the subject is invertible(in other words, it can be in any voice while still following the rules of counterpoint)?
    Last edited by caters; May-29-2019 at 05:21.

  2. #2
    Senior Member EdwardBast's Avatar
    Join Date
    Nov 2013
    Post Thanks / Like
    Blog Entries


    Quote Originally Posted by caters View Post
    Will this work or is there something that I am glossing over? Will this be too labor intensive, especially considering that I have to make sure that the subject is invertible(in other words, it can be in any voice while still following the rules of counterpoint)?
    No, this will not work. If you are still rehearsing "rules" in your head, you aren't ready to write a fugue. In any case, most of the eleven rules you have listed are either wrong, misunderstood, or trivial. Creating stretto tables would be an insane waste of effort.

    What you need to do if you hope to some day write a fugue is to start with simple two voice counterpoint. Can you write simple two-voice counterpoint? I've seen no evidence of it in the pieces you've posted. Given your tendency to think systematically, you should probably use a species counterpoint approach. After you demonstrate the ability to write coherent phases in two voices, you might be ready to take the step of writing imitative counterpoint — something like a two part invention. With a year's steady work you might reach this goal. But you should be writing exercises every day and having someone correct them.

    All of the elaborate planning and prose writing you are doing is time and effort that could be spent actually learning to compose counterpoint.
    Last edited by EdwardBast; May-29-2019 at 14:46.

    What greater comfort does time afford than the objects of terror re-encountered and their fraudulence exposed in the flash of reason?
    — William Gaddis, The Recognitions

    Originality is a device untalented people use to impress other untalented people and to protect themselves from talented people.
    Basil Valentine

  3. Likes mikeh375, Woodduck liked this post
  4. #3
    Senior Member mikeh375's Avatar
    Join Date
    Sep 2017
    Post Thanks / Like
    Blog Entries



    Edward is right on the money and as one who studied species counterpoint right through to fugue, I can honestly say that it will do wonders for you as a composer and for your music. You will gain a deep understanding and a firm foundation on which to build.

    If you do decide to do this, you must as Edward says, practice everyday (as you would an instrument) and make sure a competent teacher reviews and discusses your work with you. You do seem to have the right attitude and seem willing to learn, but in order to compose concert music well, you will have to master basics by studying and practicing in an organised, goal driven way. This foundational work will take several years but is eminently achievable and during that time, you will find out more about your own tastes and what style of writing best suits you. More importantly, you wont need to ask any questions of others when you start a piece of your own....
    Last edited by mikeh375; May-29-2019 at 20:52.
    New website and some new

  5. Likes Woodduck liked this post

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts