For example, if you have a progression in D major that ends on G, would adding an A to the G chord and losing the 3rd make the chord dominant? You would have the notes G A and D. It certainly sounds like it is functioning like a dominant chord but note sure how or why.


Chords are not defined simply by the notes in them; rather, they are defined by their musical role. So adding a 9th to a subdominant chord ... it's still a subdominant chord – subdominant is a role.

However, ignoring the original name of the chord, a chord (in the key of D) containing G, A, and D could easily sound like a Dsus (tonic) or an A7sus4 (dominant) depending on the context and harmonic role it plays.

  • thanks, your posts are always helpful since I don't always express myself as well as I could and you seem to have a genuine ability to interprete my questions. If I wanted the progression to go from G > A > D (IV V I cadence) but the A doesn't work well with my melody and G does how can I fix this? Well the pull from G to D is not good enough so I kind of see the G with the A note as a hybrid in a sense since you essentially get that root movement which is familiar in the IV V I. This is what I am trying to do and I like this G chord as it harmonizes better with my melody
    – armani
    Apr 6 at 7:42
  • So in a sense I am trying to harmonize a melody and make a subdominant chord sound more dominant. THe C# in the A chord is the one that is a problem specifically because my melody used the D note which clashes with the C# of the A chord... I want a chord that will sound dominant that will avoid this m2 dissonance and even though the C# is the leading tone, I still think that the notes G > A to D is kind of familiar and creates a more satisfying cadence than just IV I.
    – armani
    Apr 6 at 7:44

Basically, you are hinting at a A7sus4 chord, but the method you describe to get there is the issue.

I think this description...

...add a 9th to a subdominant IV chord...

and this one

...in D major...adding an A to the G chord and losing the 3rd...You would have the notes G A and D.

...are presenting some contradictory ideas that are part of the confusion.

When we start with a IV the fundamental idea is a triad, it's a symbol to represent something in tertian harmony.

But, when thirds of chords are omitted and when then have stuff like G A D as chords we drift into a vague quartal harmony.

G A D could sound like a voicing of A7sus4.

G A D could sound like a voicing of Dsus4.

G A D could sound like a voicing of Gsus2.

Another way to look at it is in D major all three tonal degrees (the tonic, subdominant, and dominant) are G A D. Each of the three tonal chords is represented by two tones from the set of three: D A are root and fifth of I, G D are root and fifth of IV, and A G are root and seventh of V. Notice how all three have their third omitted. You cannot answer the question of which tone is the root through the perspective of a tertian stack of thirds. That is the essential reason for the sort of vague sound of quartal chords.

If you want to have tonal identities like I, IV, V, you will want to treat G A D as containing some kind of non-chord tone, like A7sus4, and look to the voice leading and bass movement for sensible root progressions. The sus4 doesn't necessarily need to resolve, but somewhere in the music should be chords with thirds and voice leading to reinforce the triadic foundation. The point here is that A7sus4 isn't really the bona fide chord. A7 is and the D is a suspension, a displacement of C#.

If the music is filled with quartal stacks, if the voice leading avoids treating certain tones as resolution of dissonance, it might make sense to not use analysis symbols like I, IV, V. For example, in D major, the G stepping down to F# is a strong tonal voice leading move. A7sus4 with the G moving down to F# make a kind of V to I progression which will reinforce a sense of tertian harmony. Avoid such things deliberately to create a quartal sound. If you do something like alternate A D G (a supposed A7sus4) with G D A (a supposed G9 with omitted third), move the bass around while avoiding F# and C#, and you will get a vague quartal/pentatonic sound. At that point tertian chords like I, IV, V don't really make sense. You might just identify a D tonic.

  • Hi Michael. Aren't the 3 tonal degrees the 1, 3 and 5 tones? In other words the "stable" tones?
    – armani
    Apr 8 at 9:23
  • @armani, using ^ to show scale degrees: ^1 ^3 ^5 are the tonic chord tones, ^1 ^4 ^5 are the tonal degrees, ^3 ^6 and sort of ^7 are the modal degrees. I say "sort of ^7" because it really gets the special name "leading tone" when a half step below tonic or "subtonic" when a full step below tonic. It really different categorically. Apr 8 at 12:46
  • In terms of stability I think the thing to consider is tonality is about the stability/instability of the tonic & dominant chords, so "tonal" and "stable" are not synonymous. Also, ^3 of course determines whether the scale/key is major or minor, which is a modal consideration, so ^3 is the definitive modal degree. Apr 8 at 13:00

Adding a 9th does nothing but add a new note to an existing chord and that does not make it dominant. You have more than one alteration to your chord not indicated in your title. Namely dropping the 3rd.

You can add the 9th to the IV chord and have a G add 9 = (G B D A). If you drop the 3rd you have a suspended 2 chord (G A D). This would most likely occur when resolving to G minor and using a suspended resolution. The fact is that G is the b7 of A so you may have also created a very slim version of an A11 chord and you could be hearing that along with a tendency to resolve to D.


No, changing the chord quality via suspending the second of the chord is not going to make it dominant. But with that said, placing that A in the bass would make a strong case for reinterpreting the chord as a different type of dominant chord: [A G D] is nearly [A G B D], which would be G/A or A9sus, a suspended dominant chord. Gsus2 is a fairly ambiguous chord itself, so it could be used in a dominant context in many cases.

This is a specific result that only works because the resulting chord happens to have another interpretation, normally modifying the quality of a chord with extensions, additions, or suspensions does not have an effect on the function of the chord. Usually, those types of modifications can be understood as extra colourful variations on the basic chord quality rather than defining the harmony and its function.

  • But even having the notes G > A > D in the inner voice leading are kind of indicating a cadence arent they? Yes the bass is strongest but if you hear this root motion in one of the voices then doesn't the ear hear a cadence IV > V > I albeit more subtle?
    – armani
    Apr 6 at 7:59
  • @armani kind of, yes... but I don't follow where you're going with that. Are you stating that the addition of a 9th or other note could carry that "slight indication" of a cadence and thus affect the function?
    – user45266
    Apr 6 at 19:50
  • I just noticed your comment mentions root motion. But adding/replacing notes in a chord wouldn't have any effect on the root, no? I'm a bit confused on what you're saying here...
    – user45266
    Apr 6 at 20:01

would adding an A to the G chord and losing the 3rd make the chord dominant? You would have the notes G A and D.

Not really. The main issue is the note D, which clashes with the leading tone, C#. Presence of notes A and D in the same chord suggests the root isn't A.

Some other people suggested it might be interpreted as A7sus4 – yes, maybe, but this doesn't produce a strong dominant feeling.

However, if you picked ii chord Em: E-G-B, which is typically interpreted as a subdominant, and add note A, so that we have Emadd4 (E-G-B-A), in some contexts it could function as a dominant, especially if it resolves to tonic and you resolve G downwards to F#.

It could also work with a chord comprising of G, B and A notes.

Of course lack of note C# makes it weaker dominant than a regular A chord.

  • Good points, thank you... unfortunately, the C# is precisely the note I want to avoid in this "dominant" chord. My melody features very prominant use of the D note as I am using a pentatonic scale so this C# is clashing and it needs to go. I know it is the leading tone but essentially I was looking for a chord that could sound dominant via other methods. I know a cadence is strongest when using the leading tone but there are other voice leading techniques that make a cadence sound like a cadence right?
    – armani
    Apr 6 at 7:57
  • @armani Yes, relations between the chords are defined by the context, So yes, A7sus4 is a possibility. You can emphasize motion between A7sus4 and D (assuming you want to resolve the dominant) by resolving the 7th (G) downwards to 3rd of tonic (F#). Also it may help to move root of A (A) to root of D (D) – it's tempting to leave it as a common tone, but it would produce less motion. Apr 6 at 14:43
  • @armani Another possibility is subdominant. It still produces some tension with respect to tonic. G, Em7 maybe Bm7 (very weak). You can add dissonance to make the tension stronger, e.g. G9 (containing 7th, F), borrow chords like Bb7, Ebmaj7... Do7? Apr 6 at 14:54

A dominant chord is generally made from the root of the dominant note in a key. In key D, the dominant note is 5 - A. Just adding A to a chord can't make that chord dominant. The dominant chord there is A C♯ E, with G added to make it a dominant seventh.There are also secondary dominants, but this is unlikely to be one here. And dominant chords do not have to be followed by the tonic.

The G chord (no 3) with an added A note gives the three notes G A and D. It will either be called Gsus2 (or properly Gret2), or Dsus4. With A as its root (possible), it could be perceived as A7sus, and that's probably where you get the 'dominant' feeling from.

  • Yes I think so that the sus7 chord might be what makes it sound like that... what about if you have a G/A... would that sound like some kind of altered dominant A chord?
    – armani
    Apr 6 at 7:48

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.