A simple question yet I cant seem to find an explanation for this in either of my tonal harmony books. They all parrot each other and talk about cadences where the melody note ends on ^1 ^3 and ^5 over the tonic chord or cadences where the melody ends on ^2 ^5 and ^7 over the dominant chord. But what about the other scale degrees ^4 and ^6? Melodic phrases can end on these scale degrees too right and they will need harmonic support yet these are not in any of the cadences discussed.

  • I am a bit confused by this question. By "end on ^1 ^3 and ^5 over the tonic chord" are you saying that they discuss cadences where the final melodic note is ^1, ^3, or ^5? Or does "end on ^1 ^3 and ^5" mean that the last chord contains those three pitch classes? If the latter, why do you say "over the tonic chord"? Because ^1 ^3 and ^5 is the tonic chord; it doesn't make sense to say that they are "over" the tonic chord.
    – phoog
    Commented May 4, 2023 at 9:53
  • Phoog, I edited the question. I was refering to the melody tone.
    – user35708
    Commented May 4, 2023 at 9:55
  • 1
    Can you cite an example of a piece from the common practice era where a melodic phrase ends on the 4th or 6th scale degree? If you can, the cadence in that example is a cadence one might use for that melody. If you can’t, that explains why the theory books don’t mention it. Commented May 4, 2023 at 11:20
  • Todd: Are you saying that musical phrases that end on scale degrees ^4 or ^6 are rare or strange?
    – user35708
    Commented May 4, 2023 at 11:35
  • See this discussion of a broader definition for "cadence" Commented May 4, 2023 at 15:32

6 Answers 6


In common practice style cadences are formulaic endings meant to affirm the importance of the tonic and dominant triads.

The standard formulas:

  • V I authentic
  • ...V half
  • V vi deceptive
  • IV I plagal (if you want to include that)

All of those formula involve either the tonic or dominant chord. The conclusiveness of each is determined by whether the final chord is the tonic.

In a deceptive cadence the typical voice leading is the deception part ^5 ^6 in the bass, with the other voices moving to ^1 or ^3, and where ^1 would be the usual tone to double. From the melodic point of view a deceptive cadence still moves to tonic chord degrees ^1 or ^3.

The hierarchy of common practice functional harmony, and especially the structural phrasing purpose of cadences is not egalitarian. The point is not to accommodate cadences to all scale degrees and diatonic chords. The point is to affirm the importance of the tonic and dominant chords.

This is a fundamental concept in tonality. Not all tones are of equal importance. The common practice style emphasized the relationship between a major or minor triad tonic and a major triad dominant. You could choose to emphasize different diatonic chords, like ♭VII I, and then maybe you could melodically cadence on ^4, but the tonality would not be common practice, it would be more like a modal style.

  • Thank you. I don't know Michael the subdominant triads are pretty important too.
    – user35708
    Commented May 5, 2023 at 8:19
  • 1
    Depends what you mean. In common practice cadences, actual structural cadences, the subdominant is unimportant. music.mcgill.ca/~caplin/caplin-classical-cadence.pdf. If you mean the subdominant in a progression like IV V I, yes the subdominant is important, but it isn't part of the cadence, only the V I is the cadence. Commented May 5, 2023 at 13:04

As you're well aware, there are four basic cadences (not including those more specific which seem to be used more in US).

They are:

Perfect and Plagal - ending ^1^3^5.

Imperfect - ending ^2^5^7.

Interrupted - ending ^1^3^6.

Those are using triads - 3 note chords. Add a fourth third, which is the oft-used next chordal note, and we get:

Perfect and Plagal - ^1^3^5^7.

Imperfect - ^2^5^7^4

Interrupted - ^1^3^6^5.

So, it appears there is ^4 coming out to play, and ^6 was there all the time!

I'm using the British labels - US use different names.

  • Hi Tim. I mean using triads only. I thought this would be obvious since all tonal harmony books teach cadences using triads (except for the dominant 7th chord resolving to I)
    – user35708
    Commented May 4, 2023 at 9:25

Doubtless your books discuss the four popular cadences, Perfect (authentic), Plagal, Imperfect (half) and Interrupted (deceptive). That's V-!, IV-I, something-V and V-vi.

Are we interested in just the final chord of each type? Then we've got ^6 in the Interrupted variety. You're right that none of the basic cadences end on a triad containing ^4.

I have no idea why, in my example below, B, C and D are classed as elementary types of cadence, A is not.

enter image description here


If you want to end a melodic phrase with ^4 or ^6 you cannot use a perfect, plagal, or imperfect cadence. The most common form of the interrupted cadence is V - VI, which would allow you to end on ^6, but not ^4. However an interrupted cadence is not restricted to dominant - submediant (V - VI).

From Wikipedia,

"A cadence is called 'interrupted', 'deceptive' or 'false' where the penultimate, dominant chord is not followed by the expected tonic, but by another one, often the submediant." (my emphasis)

If you play "shave and a haircut - two bits" (which ends on the tonic) with various final chords -

  1. I gives you the usual perfect cadence
  2. VI gives you the most common interrupted cadence
  3. IV gives you a different interrupted cadence

The third cadence would allow you to change the melody to end on ^4 or ^6 instead of ^1, noting that the preceding melody note might also need to be changed to make melodic sense.

If you are using ^4 as your final note you can also use the interrupted cadence V - II.

  • Thank you Peter. By far the best answer so far IMO. I often find IV as a final chord at a cadence in music where you might not want the typical V I cadence which sounds too "classical" This chord is especially useful in that it allows the tonic note in the melody but I also wondered if it could support those other 2 melodic tones. I still wonder however why in tonal harmony those tones are not even in any of the cadences as if they are just plainly outcast.
    – user35708
    Commented May 7, 2023 at 9:42
  • Perfect and plagal cadences end on I, which is ^1 ^3 ^5. An imperfect cadence ends on V, which is ^5 ^7 ^2. That leaves only interrupted cadences, which can end on ^4 or ^6. As to why - I suspect it is because when harmony is so important it constrains the melody. We have become accustomed through centuries of music to looking for a tonic chord at the end of a piece of music, and perhaps a dominant chord at a subsidiary ending. This affects the melody lines too.
    – Peter
    Commented May 8, 2023 at 4:28

On the harmonic possibilities for a cadence on melodic 4 or 6

The idea of a cadence, or course, is that it represents a point of (relative) stability in a piece of music, which is often desirable at the end of a phrase or the end of a piece. Harmonically, the tonic chord is the most stable, with the dominant chord also providing stability, though less than the tonic. The submediant chord also provides a sense of stability, since it can serve a tonic function.

When speaking in terms of triads, scale degrees 4 and 6 are harmonically problematic for cadences. 4 could be harmonized by VII, II, or IV, and 6 could be harmonized by II, IV, or VI (the last being covered by the deceptive/interrupted cadence). Here is why, except for 6 against VI, those are rare, at best.

The VII chord

In nutshell, the VII chord is simply too unstable to be a cadence-point in Common-Practice Tonality. There is no sense of rest, between the presence of both the leading tone and diminished fifth. (The V7 chord also contains both the leading tone and diminished fifth, but the presence of 5 stabilizes [in relative terms] the chord. In the V chord, of course, the diminished fifth isn't present, also helping make that chord more stable that VII.)

The II chord

The II chord doesn't have the stability problem of the VII chord, but it doesn't set up well to create a satisfying cadence. Notice that the penultimate chord in the three standard cadences is (usually) either the dominant chord (authentic, deceptive/interrupted) or predominant chord (half, plagal). Since the II chord can function as either a dominant or predominant chord, it's presence at a cadence point would tend to take on the characteristic of the chord preceding it. Thus, it would be perceived either as an extension of a half cadence or as a pre-dominant harmony, usually not stable enough for a cadence.

The exception here might be a phrase that ended I-II; a sort of half cadence with II standing in for V. However, this is certainly uncommon, at best, in the Common Practice.

The IV chord

This has essentially the same problem as the II chord, but without the II's exception of I-II. A I-IV cadence will tend to sound like a modulation, since I is also V/IV.

  • this is good thank you. I do want to say that I dont think that V supports stable tones. The leading tone is the most unstable of all tones in the scale supported by V at a half cadence.
    – user35708
    Commented May 7, 2023 at 11:16
  • @armani That's not what I intended to communicate. I've updated the post to try to clarify my meaning.
    – Aaron
    Commented May 7, 2023 at 12:07
  • Thanks Aaron. You added some good stuff there.
    – user35708
    Commented May 8, 2023 at 8:51
  • Aaron, Why are all your numerals bigger? I get confused when seeing VII as it kind of makes me think of a major chord not a dimished chord
    – user35708
    Commented Jun 16, 2023 at 8:47

I've found a category of examples of cadences in the classical period where the melody has ^6 on a strong beat at the cadence. The category is sonata form, secondary theme in the dominant key area and the harmony is a half cadence - like a V/V.

  • Example: Maria Hester Park, Piano Sonata in C major, first movement, measure 33, half cadence in G on a D major chord with A in the melody.

Obviously this won't do for ^4 because V/V has a raised fourth scale degree as its third.

I found a cadence with ^4 in the melody in the recapitulation of the same piece.

  • Example: Maria Hester Park, Piano Sonata in C major, first movement, measure 101, half cadence in C major on a G7 chord.

I see your comment on Tim's answer about "triads only". I don't understand how or why you would impose that limitation. If you're asking "how to harmonize ^4 at a cadence", the answer is "with a V7 chord as part of a half cadence".

If every line besides the melody is playing ^5, ^7, and ^2 and the melody is playing ^4, then we will hear a V7 chord even though the ^4 is in the melody. That harmony works and is at least somewhat popular.

So your answer is half cadences in both cases.

  • Toddd thank you. My textbook "Harmony and voice leading 5th edition" says that although V7 does support 4 in the melody it usually does not do so at a "point of repose" like at a half cadence... As a chord leading into the cadence then yes but not as the end chord. This is why I said triads. Because triads form the consonances necessary for the feeling of "repose" Look at all the chords used in all tonal harmony textbooks. When they teach you about PAC, IAC, HC etc the chords that support the melodies are triads not 7th chords.
    – user35708
    Commented May 4, 2023 at 17:33
  • By the way, why can´t IV support those tones? I would imagine that that would be the first choice.
    – user35708
    Commented May 4, 2023 at 17:34
  • In your first example, a half cadence after modulating to the key of the dominant, it really does not make sense to analyze the scale degree as ^6. After the modulation the scale degree would be ^2. Commented May 4, 2023 at 19:51
  • 1
    @ToddWilcox My book doesnt say never, it says usually.. more precisely that the dissonant nature of a 7th chord wants to keep moving and as such often makes it innapropriate as the last chord in a half cadence.
    – user35708
    Commented May 5, 2023 at 7:23
  • 1
    @armani I should explain my goal post comment. It's not clear if you believe there are cadences that are used with ^4 and ^6 in the melody, or if you believe there are not cadences that are used with ^4 and ^6 in the melody. If you believe there are not ^4 and ^6 cadences, then that's the answer to your question: they are not discussed because they do not exist. If you believe there are ^4 and ^6 cadences, then it's confusing that you're rejecting other people's examples of ^4 and ^6 cadences without supplying examples of what kind of ^4 or ^6 cadence would satisfy you. Commented May 5, 2023 at 16:51

Your Answer

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