4

Many continuo manuals (both historical sources and modern treatises) are careful to point out that a continuo realization should be contrapuntally correct, i.e., one ought to avoid parallel fifths and octaves in the extemporized parts, as well as other voice-leading mistakes. (With some amount of license allowed in very full-voiced textures.)

However, such prescriptions only seem to apply to the internal consistency of the continuo realization. Most continuo manuals that I've seen only show music examples for the realization itself, and not for the composition as a whole. For that reason, I'm wondering how the realization should interact contrapuntally with the other written parts of a composition.

Suppose, for example, that we have three string parts (2 violins, viola) above a continuo line. If I want to play a four-part continuo realization, should the resulting seven-part texture be correct contrapuntally? That seems almost impossibly difficult to me, especially if it has to be extemporized. Or am I allowed to double in the right hand one or more of the written parts, so that we effectively get a less than seven-part texture?

I find this a rather confusing aspect of Baroque musicianship.

Edited for clarity.

3

Yes and no.

More precisely, octave and unison doubling is probably permissible with inner voices. As always, this is subject to considerations of style and taste.

Suppose, for example, that we have three string parts (2 violins, viola) above a continuo line. If I want to play a four-part continuo realization, should the resulting seven-part texture be correct contrapuntally?

Another way of asking this is whether the resulting texture really ought to be a seven-part texture. It probably shouldn't.

If you really want to go into the question in depth, study some larger orchestral or choral/orchestral works of the period. A piece with four string parts and two wind parts, for example, will not employ a six-part texture throughout. There is a lot of doubling, sometimes at the octave. Bach's festive choral music, for example, frequently has the first trumpet doubling the alto at the octave.

One technique that Bach, at least, employs frequently in his orchestral writing is to use different "choirs" of instruments antiphonally. I don't know how readily applicable that is to a continuo realization, but it brings to mind a related point: using some sort of figuration in the continuo realization might help avoid unwanted doublings. I once heard Christopher Hogwood suggest, in a workshop rehearsal of Dido and Aeneas, that the continuo harpsichordist consider a realization inspired by the opening of this through-composed piece for solo harpsichord:

(To be sure, I'm talking about the first few measures where the right hand is in the alto range, here on the upper manual, not the part where the right hand jumps to the soprano range, here on the lower manual, and begins playing the ornamented melodic line.)

Parallel unisons and octaves are going to be impossible to avoid in a piece with a full choral or orchestral texture. They should generally be avoided in accompanying a soloist, and they should probably also be avoided with the soprano/first violin/treble wind instrument in a fuller texture.

Parallel fifths should be avoided more assiduously.

Disclaimer: I have limited experience actually playing continuo realizations, so there are probably many practical considerations that I'm not aware of. I wish Wheat Williams were still on the site.

4
  • Thanks a ton! Exactly what I wanted to know. Yes, I was mainly asking about very full textures, such as the tutti sections from Handel's Messiah. Aug 25 at 14:47
  • @KimFierens funny, I was just looking at Messiah as a sanity check against some of the assertions in this answer. It's fascinating, in fact, how the orchestra jumps into and out of doubling the voice parts. At the opening of Worthy is the Lamb, the first violins, doubled with the first trumpet, are independent in m1, doubling the soprano at the octave in m2, and doubling the alto at the octave in m3. By the cadence in mm6-7, the first trumpet and second violins are doubling the soprano at the unison while the first violin is doubling the alto at the octave.
    – phoog
    Aug 25 at 15:02
  • @KimFierens and I don't see a reasonable way for a continuo player to realize that cadence without doubling the 7-6 suspension in the soprano and the B-to-A♯ resolution in the alto.
    – phoog
    Aug 25 at 15:11
  • Thank you; interesting. Come to think of it, I believe we can generalize your observation: if there is a dissonance anywhere in the written parts that is also notated in the continuo part, then doubling is inevitable, because the resolution of dissonance is more or less unique (in the Baroque style at least). Aug 25 at 17:47
0

Yes. The bass part (especially in Figured Bass style) stands out. Parallel fifths or octaves with a melody line (or two melodies in trio sonatas or even more) gives the effect that a voice dropped out.

You can test this by playing some simple songs and play a bass line that imitates the melody. It sounds like a full melody but not two parts.

4
  • 2
    The question isn't just about the bass part — which is written in basso continuo parts. The question relates to the harmonic realization.
    – Aaron
    Aug 25 at 1:22
  • Yes, Aaron is correct. I was asking about the extemporized right-hand parts of the continuo. Should those also not form faulty consecutives with the other written parts? Aug 25 at 1:24
  • Yes, the other parts have the same problems with parallels. The bass just stands out more so that's what I used. In inner parts, one may get away with doublings but not parallels.
    – ttw
    Aug 25 at 1:39
  • The bass isn't discretionary; it's the given starting point from which the continuo realization is, well, realized. Parallel fifths and octaves between the bass and the melody (or any other part) are an error on the part of the composer, not the performer.
    – phoog
    Aug 25 at 13:29

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.