I am trying to harmonize a melody I wrote and am stuck at the end. I like the 3^ over the bass, but what chord should it be? I64? iii?

Also, what other improvements could be made?

Enter image description here

  • 1
    Where did the melody come from? Aug 27 '21 at 17:09
  • Must we end the phrase on the tonic chord? If the melody is the only "sacred" element of the composition, one of the simplest solutions is to change the last two chords from G C into E Am. This momentary shift into the parallel minor seems to fit your prevailing contrapuntal style well (maybe try G G# A in the bass). Of course, there's also altered dominant chords and non-chord tones as well as nondiatonic harmonies to explore.
    – user45266
    Aug 28 '21 at 4:40

I'd like to add an additional possibility to the answers that have been given so far.

A fair number of the chorale melodies that Bach harmonized feature ^3 ^1 at cadences. When Bach harmonizes these passages, I believe he usually uses the harmonies I IV. For example,

  • from Nunn komm, der Heiden Heiland, n. 28 in the Breitkopf collection. (NB I is turned into V/IV here by the C-natural.)

enter image description here

  • from Freu dich sehr, o meine Seele, n. 29 in the Breitkopf collection. enter image description here

(You'll notice these examples come from consecutive chorales from the Breitkopf collection. I didn't have to look very long to find examples.)

Note, however, that this harmonization only works in the middle of the piece, since it ends on a IV chord.

Another possibility (also for the middle of the piece) would be to harmonize ^2 ^3 ^1 with V V6/vi vi, providing a nice chromatic bassline G G# A.

Finally, I would like to suggest that, besides V13 or V(add6), it would also be possible to harmonize the E with iii6, which (in this context) is "really" a dominant with the D simply replaced by an E échappée. Although none of these possibilities are very likely in Bachian chorale style, the iii6 triad may be less unlikely than the 13th chord.

  • 1
    Nun komm der Heiden Heiland is in the minor mode, in this case B minor, so that's actually V/VI to VI, not V/IV to IV, and the scale degrees are ^5 ^3. But that doesn't really detract from the answer; it just illustrates the versatility of this move.
    – phoog
    Aug 27 '21 at 22:57
  • Agreed, the global tonic is B minor, but I hear it as moving into D major in the second phrase (and for the beginning of the third phrase, not shown). I should have been explicit about that.
    – msailor
    Aug 27 '21 at 23:39
  • It's worth noting that Bach could have harmonized F# D as ^5 ^3 in B minor, as a V I imperfect authentic cadence, but chose not to. Possibly because it would have been somewhat monotonous to have two consecutive B minor cadences?
    – msailor
    Aug 27 '21 at 23:40
  • Indeed, or it might have been because of the text, the absence of which is a huge shortcoming of the 371 chorales. The text is typically the explanation for dissonant and chromatic chords, at least.
    – phoog
    Aug 28 '21 at 0:42
  • brilliant thank you... you know I was thinking of iii6 because I am only really familiar with triads so far in my text book and those two tones are part of the iii chord so not sure why it wasn't considered a possibility yet since it is a simple diatonic chord chord.
    – armani
    Aug 30 '21 at 8:42

Dekkadeci is right; really, the only way to have this ^3 is with a V13 (or Vadd6) chord. This is something that happens fairly frequently in popular music and folk tunes, but it's relatively rare in the "classical" style. It's much more common to have ^2 above that V chord, but that ^2 briefly moves to ^3 as an escape tone before resolving down to ^1.

I have encountered one—precisely one!—instance of someone calling this a "Slovenian cadence," claiming that this ^3 turns the V chord into a iii6 (!!) right before the cadence. But I have to say that I find this explanation really troublesome. (See What is a "Slovenian cadence"?)

Otherwise, I'd invite you to look at the chords in your example and how they relate to the meter. Typically, the stronger beats 1 and 3 will have more tonic chords, and the weaker beats 2 and 4 will have more dominant chords. In a few spots, you reverse this relationship (see, for example, m. 2) so that dominant happens on beats 1 and 3 and tonic happens on beats 1 and 4. This is certainly possible to do, but it's relatively rare in this eighteenth-century style. An easy fix is to change an earlier quarter note to a half note, thereby moving everything else forward one beat.

  • thanks for the tip about meter... Very helpful
    – armani
    Aug 27 '21 at 15:50
  • Aren't there some Lutheran chorales that cadence ^3 ^1? I agree that it's rare, but it certainly isn't unknown. It was fairly common in early-20th-century popular song, though.
    – phoog
    Aug 27 '21 at 16:32
  • @phoog I don't know of any off the top of my head, but I wouldn't be surprised if some Lutheran chorales do that.
    – Richard
    Aug 27 '21 at 16:39
  • @Armani, if you do the one beat shift forward, then when you get to the final dominant chord and the ^2 ^3 ^1 could then take rhythm eighth, eighth, whole, and the ^3 on a weak upstroke will feel more clearly as an escape tone. Aug 27 '21 at 17:15
  • Thanks Michael but I am not allowed to use in between non chord tones yet
    – armani
    Aug 27 '21 at 17:23

Scale degrees 3 and 1 are unusual choices for the melody of a cadence because that shuts you out of plagal cadences and forces you to use extended V chords for authentic cadences, but if you insist on those 2 scale degrees, you can still write an authentic cadence by using V13 - I.

  • what about a cadential 64 chord, doesnt that have the 3^?
    – armani
    Aug 27 '21 at 14:53
  • but that G bass with the E does sound quite good there is there no triad or normal 7 chord that can go there with those two notes?
    – armani
    Aug 27 '21 at 15:02
  • @armani for a popular song of the early 20th century, I like V7 if the voicing is right. The major ninth and major seventh with ^2 and ^4 can be lovely. In the chorale style I might omit the seventh and maybe also the fifth, which would make it iii6/5 or iii6, strictly speaking. I have a vague recollection that there's a Bach harmonization of a chorale with a ^3 ^1 cadence; if I think of what it is I'll come back and let you know.
    – phoog
    Aug 27 '21 at 16:38
  • @armani - The cadential 6/4 chord needs to resolve to a V chord (or maybe a V7 chord); I've never heard of an imperfect cadence or half cadence that ends with V11 (which you need to use scale degree 1 as the 2nd soprano note in that cadence).
    – Dekkadeci
    Aug 28 '21 at 11:31

If you're working without any particular imposed rules like species counterpoint, I'd encourage you to consider that not every melodic note has to change the bass (or therefore the chord. If the cadential "D, E" in the melody were over a half-note G chord, I'm not sure I'd bother calling it anything other than a V that just happens to have an escape tone. (Of course, if you're free to change your melody, changing the E to a B is even better.) You might also consider moving a bit slower in the bass in the first three measures in order to "go somewhere" other than the tonic—imagine, say, in half notes: C B A E F G C. And then with that in place, if one wanted a busier bass, you could fill in the gaps with various passing or neighbor notes.

  • this is just an exercise... I need each tone of the melody to have a harmony, its just to get an idea of which tones go with which harmonies... but I hear you
    – armani
    Aug 27 '21 at 14:54

From comments...

this is just an exercise... I need each tone of the melody to have a harmony, its just to get an idea of which tones go with which harmonies...

I agree with @Richard's answer that as is the ^2 ^3 ^1 cadence part should treat the ^3 as an escape tone.

But, if you really mean to make each soprano tone a chord tone, there is a bit of a problem, because an escape tone is a non-chord tone.

I think the mediant ^3 in the melody is most strongly associated with the tonic chord.

After the tonic I chord the iii and vi chords are the diatonic triads that can harmonize it. That categorically rules out an authentic cadence as a harmonization of ^3 as a strict chord tone. iii and vi are secondary, modal chords. Functionally they are pre-dominant harmony. I think it is common to see them in sequential passages or at least with a secondary dominant preceding them.

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.