Are there any circumstances (at least in the common period use) where the scale degree 2 in V7 goes up to 3 as opposed going down to 1?

Why does this seem to be disallowed in textbooks, even more than 4 going up to 5, or 7 going down to 5, despite these two tones being arguably more important as tendency tones?

3 Answers 3


It is not very common, but possible

(for clarity's sake I'll assume C-Major and use roman numerals for chords, and scale degree numbers and names for notes)

Doubling a chord's third

Generally. in polyphonic music and especially in four-part writing, we avoid doubling the mediant in a tonic chord. That is because doing so would alter the chord's character, and make it sound more like the iii. We usually only double the third of the I when resolving the vii6, but that's for another question.

On the other hand, doubling the third of the ii (namely, the subdominant), makes it more like the IV, which is often desirable, and also makes for a smoother voice-leading.

A problem with your requested voice leading: I tends to iii

In a typical voice leading of your example, the dominant chord's seventh would to resolve to the third of the I, as well as its fifth. So F->E, and D->E, or 4->3 and 2->3. This means a doubling of the mediant. Here, the dominant's seventh also functions as an upper leading-tone, which in conjunction with the former would lead to further altering the character of the I towards that of the iii.

A possible Solution: An irregular resolution of the dominant's seventh (but not really)

This troubling situation can be remedied by stepwise motion of the subdominant note. A E->F->G procession of the voice is preferrable for two reasons.

  • It's contrapuntally agreeable as a stepwise occurring dissonance.
  • It avoids the doubling of the mediant
  • The movement by thirds between the bass and the voice containing the 4 means it's now technically not a seventh. It only dissonates in relation to the leading tone, (a tritone relation).

For all of these reasons, it's okay not to resolve the 4 to the 3 in this example, since its downwards resolution is usually required in the context of a 7-6 dissonance.

Disclaimer: The V43 chord is quite common in the Classical era but - in my experience - the vii6 is more commonly used instead in the Baroque era.

enter image description here


It's very common, at least in the baroque period. Normally we're thinking of final cadences where the melody usually has ^2 ^1, but in intermediate cadences you will often find the melody or an internal part having ^2 ^3 (perhaps relative to a secondary tonality). The first Bach chorale I checked, BWV 253, has this in its first cadence, in the alto voice (E to F♯ against A to D in the bass).

In a final cadence in a four-voice texture it will be rare because the melody will normally be ^2 ^1, the bass will be ^5 ^1 and the doubled tone of the dominant chord (if it is not a dominant seventh) will be the root, ^5, not the fifth, ^2. In a five-voice texture it will be more likely.

Another move you see fairly often in Bach is that one voice has subdivisions on the dominant chord in the last cadence, moving from ^4 to ^2 and then ^3 for the final tonic chord. An example of this is found at the close of the Gloria from Bach's B-minor mass.

Finally, if the soprano doesn't have ^2 ^1 but instead ^7 ^1 then it's quite likely for an inner voice to move in parallel sixths below, as at the end of Handel's Messiah, where the tenors have E F♯ in D major,


In Music in the Galant Style Gjerdingen gives some examples like these for the DO-RE-MI schema...

enter image description here

enter image description here

Gjerdingen also has a modern edition of the rule of the octave from Fenaroli where ^2 ^3 is in the bass and the harmony involves an inverted V7...

enter image description here

So, the basic idea is ^1 ^2 ^3 is basically harmonized as I V I and it's an opening phrase, an incomplete ending.

From there I think the question needs to be specific as to which voice is ^2 ^3 in, is implied broken chord type motion acceptable, is the seventh on V absolutely required?

Otherwise, I think the answer is: generally, yes ^2 can ascend to ^3 with some kind of V7 I harmonization.

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

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