4

In four-part harmony only the root tone can be doubled in dominant seventh chord ("Guide to harmony" - Tchaikovsky). Why, what is the reasoning behind it? I am interested only from classical harmony perspective (!)

For the vast majority of harmonic rules I always find some explanation/motivation and try not to study them as just a set of recipes. In this case I seem to fail.

Disclaimer: Yes, I understand that all these rules are constantly broken even during Common Practice Period.

  • 3
    The rules say that a dominant seventh chord has to resolve: the third rises and the seventh falls. But if you double the third or the seventh the result is parallel octaves, which are forbidden. The reason you can't double the fifth is just that it was thought to sound unpleasant or awkward. – Old Brixtonian Oct 5 at 19:30
  • @OldBrixtonian Well in this case I thought that a slight deviation from the rules just solves the problem. E.g. doubling two sevenths I can lead one of them a step down, and the second one a step up. In some cases we can lead the seventh a step up, and I thought here it may be tolerable as well. – NickQuant Oct 5 at 19:39
  • Why not, if its what you want to hear. You might try staggering them to avoid parallel movement; or inserting a passing-note. It's still against the rules, but it's what Bach did sometimes. By the way, I was as interested as you in learning WHY the rules existed, and I believe the reason parallel octaves and fifths weren't allowed was that people had been hearing them for a long time in organum, on certain organ stops, in folk-singing and simply as overtones and they were simply sick of them! – Old Brixtonian Oct 5 at 20:47
  • @OldBrixtonian it's extremely common for the third to resolve down by a major third if it is in an inner voice. – phoog Oct 6 at 19:54
  • @phoog - only in the Baroque period - from Haydn onwards, resolving the third downwards was generally avoided. Almost the only places it happens are in choral music intended to imitate an older style. – Alexander Woo Oct 8 at 9:16
3

The 3rd (ti) and 7th (fa) shouldn’t be doubled because of their tendency as leading tones. Ti is leading to do, fa is leading to mi. Whether these are harmonic/acoustic laws or just traditional use be left open.

If we double the 5th we shall have either no root tone or no 3rd, as the 7th is needed: otherwise it wouldn‘t be a 7th chord! And with no root tone this would actually be ti,re,fa (missing the sol) which is considered as functionally equivalent to the Dom7 chord.

  • 1
    And what is the problem with dropping the 3rd? It is said in the beginning of the chapter that we can drop the 3rd under some circumstances. – NickQuant Oct 5 at 22:11
  • 1
    I like you‘re asking, as I‘ve been always skeptical to rules and was not satisfied by answers: like it sounds bad. I‘m even doubting that leading tones want to go somewhere. If someone wants something it was the composer or the singers. The rule of the dropping 3rd I recognized already as a young boy in practice of singing songs: in tenor and alt. But I think we wouldn‘t do it in the soprano. Or does anyone know an example? This would be an interesting question! (I‘ve heard Tchaikovsky was stricter than others in his theory of harmony. – Albrecht Hügli Oct 5 at 22:29
  • 1
    I am not sure I understand what you are saying here: "The rule of the dropping 3rd I recognized already as a young boy in practice of singing songs: in tenor and alt. But I think we wouldn‘t do it in the soprano" – NickQuant Oct 5 at 22:41
3

Perhaps the voice leading becomes a problem. It's difficult (in four part harmony) resolve both the diminished fifth and augmented fourth at the same time. The augmented fourth is usually resolved outward and the diminished fifth inward. (Augmented intervals tend to expand, diminished intervals tend to contract or at least that's the way composers played it.)

A concrete example: take B-F-B' (in C major): resolving the two (inequivalent) tritones would lead to C-E-C' with parallel octaves. F-B-F' would give E-C-E again giving parallel octaves.

  • Yes, I had that in mind as well, but Tchaikovsky offers quite a handful of exceptions for voice leading, I therefore assumed I can introduce slight voice leading modifications as well. – NickQuant Oct 5 at 19:41
  • The question becomes whether the music sounds like a voice dropped out. In this case (especially with a dominant seventh which stands out on its own), I think it would be noticeable. Another problem is that doubled thirds in a major may sound like a Neapolitan harmony (perhaps starting a short tonicization or even a modulation.) – ttw Oct 5 at 21:16

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.