Consider these two statements:

• “When you see a G7 chord you can use the G blues scale.”
• “When you see a G7 chord you can play a G altered lick.”

Are they true? The simple answer to each is an unsatisfying “sometimes.”
For a more satisfying answer, we need to understand an important distinction. A dominant chord can be used in two distinct ways we can call dynamic and static.

Dynamic Dominant Chords

We can define a dominant chord as a chord that contains a major third and a minor seventh. Those two notes create a certain dissonance. A dynamic use of a dominant chord is one where that tension is part of a cadence—where the tension gives rise to resolution to a different chord. The most common resolution is to a chord down a fifth (e.g. G7→C). Some other common resolutions are: down a half-step (as in the tritone substitution), up a whole step (see my post about the Back Door ii-V) or up a half-step (see my post about the common but nameless cadence used in the song “Whispering”).

Take any dominant chord that’s part of a cadence—in fact, almost any dominant chord in any Tin Pan Alley standard—and you’ll find that if you extend its duration it sounds stuck, like an falling object that somehow stops midair, violating the laws of gravity. Such chords are dynamic dominant chords. They are a means to an end rather than an end in themselves. They make a promise to resolve in a timely manner.

Static Dominant Chords

In contrast to dynamic chords, static chords can last indefinitely; when we hear them we don’t particularly expect them to change or resolve anywhere. With rare exceptions, major chords are always static. Dominant chords, it turns out, can be static too.

Just listen to almost any James Brown song—for instance, “Get On Up.” We hear the E7 chord vamping—and feel the tension of the tritone created by the major third and minor seventh… but we feel it as a color or a texture rather than a call for resolution. We call the E7 chord here static because its tension is stationary, part of the status quo, rather part of motion that yields to a resolution. (The words static, stationary and status all come from the latin verb stare, which means to stand.)

The blues have used static dominant chords steadily over the last century. Rock ‘n’ Roll used them when it emerged mid-20th-century, and funk not long after that. The Tin Pan Alley composers didn’t use them much in the jazz standards of the Great American Songbook, but jazz has used them wherever it has intersected with blues and funk. Some examples: “Blues in the Night” (1941, Arlen/Mercer), “Born to be Blue” (1946, Mel Tormé), “Night Train” (1951, Jimmy Forrest), “Watermelon Man” (Herbie Hancock, 1962), “Song for my Father” (1965, Horace Silver).

Dynamic… Static… Who Cares?

We mentioned that major chords are always static. In fact, we can think of a static dominant chord as a major chord that’s made more bluesy… through the darkening of the sound that results from flattening the major seventh… and through the dissonance created by the tritone between its third and that flattened seventh. No surprise, then, that the static dominant chords are the ones over which a blues scale sounds appropriate. If we view the Mixolydian scale (the most basic dominant scale, a major scale modified by flattening the seventh) as a bluesier version of the major scale, then the blues scale, with its minor third and flat five, is a version that’s bluesier still.

On the other hand, in dynamic cases, where we use Mixolydian for its need to resolve rather than its bluesiness, substituting a blues scale is not just unnecessary; it’s counterproductive. The blues scale lacks a major third, and that eliminates the tritone (formed by the major third and minor seventh) that is so central to a dominant chord’s desire to resolve.

If anything, we may want to increase that desire to resolve. We can achieve that by replacing the Mixolydian scale with other possibilities such as the altered dominant scale. If you compare the two sequences below—identical ii-V-I sequences in C, except that the second one uses an altered V7 chord—you find that the second one produces a greater feeling of urgency:

Increasing a dominant chord’s desire to resolve in this way serves well in dynamic contexts; not so for static contexts. Jimi Hendrix’s “ Foxey Lady ” vamps on an F#7(#9) chord in which both the darkness and the dissonance of a normal F#7 chord are intensified by the added sharp 9. But add a D natural to the F#7(#9) voicing, making it darker still, and you ruin the song. All of a sudden, the chord wants to resolve too much to just be for effect. It’s like the difference between a teenager who wears a spiked leather bracelet and one who carries an automatic weapon. The spikes symbolize danger but they won’t be used for anything; the gun actually means business. (Sorry, that metaphor was a little dark! But I hope you get my drift.) Adding the D natural to the F#7(#9) chord turns it into an F#7alt, and that chord can’t just be static.

For another example, look at the beginning of Horace Silver’s “Song For My Father.” Here’s what the chord changes normally sound like:

The second chord is an Eb7. It is always a static chord in this song, both here in the A section as well as in the bridge. If we replace it with an altered dominant chord, here is the unfortunate result:

A Better Understanding

We can now revisit the question we posed at the beginning, offering these improved guidelines:

• When you see a G7 chord used statically, consider using the G blues scale; a G altered lick won’t sound good.
• When you see a G7 chord used dynamically, consider using the G altered scale; a G blues scale lick won’t sound good.

Even though the blues scale always gives a static sound, it does not work in every static situation; sometimes it is just darker than what is called for. The Eb7 chords of “Song for my Father” are a good example; it’s fine to play a little bluesy on them here and there—a passing reference to the minor third or flat five—but all-out use of the blues scale would sound too minor and too dissonant.

Likewise, altered dominant is virtually always a dynamic sound, meant to resolve, but it doesn’t work in every dynamic situation. Often it is just too dark, or its force of resolution is too strong.

An Example

Look at the first five measures of the Harold Arlen standard, “My Shining Hour”:

and the first five measures of a simple blues in G:

I’ve labeled three of the measures (1), (2) and (3).
All three of these measures specify G7 chords, and it works fine to play this mixolydian phrase (or any other) in any of the three:

Here is a G altered phrase—that is, a phrase based on the G altered scale that works over the G7alt chord:

Bar (1) is a dynamic dominant chord resolving down a fifth. No surprise that the altered phrase works great there.

Since the first four bars of the blues example amount to a bluesy vamp in G, the G7 chords there are static… so we can use a blues scale phrase in any of them, including bars (2) and (3). Here’s a sample blues scale phrase:

And here is the phrase used in those two measures of the blues:

The blues phrase would not work well in bar (1) of “My Shining Hour” because the role of that G7 is dynamic.

Likewise, the altered phrase above wouldn’t work well over any of the static chords of the blues. Or would it?

Sure enough, in does not work in the midst of a G7 vamp. But on the last chord of the vamp it actually works fine, because that chord is followed by a C7, which is a perfect fifth down from G7. Listen:

So the fourth bar of the blues is unusual in that the dominant chord there can be thought of as either static or dynamic. The two ways of thinking of it give rise to two different ways of playing over it that we just heard.

Ambiguity

Deciding whether a chord in a song is static or dynamic is not always as clear as, say, whether it is major or minor. Many factors may come to bear—even tempo. (An example: “Sweet Georgia Brown” has a sequence of dominant chords, each lasting four bars. At a brisk tempo, a player could treat them all as dynamic and use altered chords, for an edgy, modern sound. At a slow tempo, they must be treated more as static, lest the unresolved tension last too long.) The dynamic/static distinction is not so much about objective qualities of chords as about how we choose to treat chords. Perhaps a good analogy is the friend/acquaintance distinction. In lots of cases it is clear whether someone is a friend or just an acquaintance, but there are many cases where context is a factor (a person we know from home can be an instant friend when we encounter them wandering in a far off country) or simply personal choice. We may view someone as a friend or not, and our actions will vary accordingly.

Some Additional Thoughts

Major Chords

I mentioned earlier that major chords are always static. There is perhaps one exception: It’s common for a II chord to be Lydian Dominant, so as to resolve down to the I. But if the melody note it must harmonize is the root of the I chord, it must instead be Lydian. An example is the second to last chord of “Samba de Uma Nota Só” (“One Note Samba”), in which a dynamic Bmaj7 with B in the melody leads to final B major chord:

Minor Chords

Minor-major chords (chords with a minor third and major seventh, written as C(ma7)) are always static. For examples, listen to Strayhorn’s “Chelsea Bridge” or Horace Silver’s “Nica’s Dream.” The first two chords of both songs are B(ma7) followed by A(ma7). They contain tension, but not the sort that wants to resolve to any chord in particular.

Half-diminished chords are generally dynamic. Their main use—and virtually their only use in the Great American Songbook—is as the two chord of a minor ii-V progression, which is to say, a darkened ii-V progression in which the ii chord is half-diminished, the V7 chord is altered and the destination chord is, more often than not, minor.

In modern jazz they have a static use that calls for a Locrian scale. It’s a very dark sound, akin to a version of the (already dark) Phrygian scale that whose fifth is flattened, darkening it still more. The first chord of Joe Henderson’s “Inner Urge” is a static half-diminished chord.

Minor seventh chords can be dynamic or static. They are extremely common as the ii chord of a ii-V progression, which makes them dynamic. But they are also used very often as minor tonic chords in static situations (e.g. “So What,” “Cantaloupe Island,” “Mr. P.C.”). The distinction between static and dynamic is even more blurred for minor chords than for dominant chords for a couple reasons. First, minor seventh chords have much less force of resolution than dominant chords. Their chord tones contain no tritone because the third and seventh are a perfect fifth away. That reduces their tension and, accordingly, their need to resolve. So the dynamic ones don’t feel especially dynamic. Secondly, compare a Cm7 to Cm, the canonical, static, minor tonic chord. Whereas there is a big difference in sound between C and C7, the difference between the Cm and Cm7 is small. They normally call for the very same Dorian scale, and the decision to write one or the other can be somewhat arbitrary. Thus, any minor seventh chord is already very close to its most static counterpart. There are plenty of songs (e.g. “Invitation,” “Recorda Me”) where what seems like a static minor seventh chord silently morphs into a dynamic one that is part of a ii-V.

Diminished Chords

Diminished chords are mainly dynamic. They tend to resolve either up a half-step (e.g. second chord of “Ain’t Misbehavin'”), down a half-step (e.g. second chord of the bridge to “Ja-Da”), or to the major chord of the same root (e.g. first chord of “Spring is Here”).
As for static diminished chords… it’s a bit hard to argue that any given diminished chord is static because no matter what chord follows it, it can always be seen as making one of the of the above three resolutions. For instance, if a diminished is followed by a G chord, whatever the chord is we can think of it as either a G, G or A diminished chord, possibly over a new root. Adim = Gdim/A, Edim = Gdim/E, Fdim = Adim/F, etc.

In theory there’s an argument to be made that a diminished chord can serve as a (static) minor-major chord with a flattened fifth. But I can’t think of an example that can’t be thought of equally well as dynamically leading to the chord that follows it. Can you?

Deceptive Chords

A chord can be dynamic even though it doesn’t resolve. Consider the end of the standard bebop treatment of “All the Things You Are,” where the melody is hijacked two bars before the end of the form by the intro/ending figure. (The figure was originally from Charlie Parker’s song “Bird of Paradise,” a contrafact of “All the Things You Are,” but players have used it for “All the Things You Are” ever since):

The E7 chord is dynamic. It is part of a ii-V progression that prepares us to resolve to A. In the original song it makes that resolution but here, when the ending is used, it never does.

We see a similar thing in the last four bars of Stanley Turrentine’s “Sugar,” where the first G7alt does not reach its natural destination of C, but is still dynamic:

In these examples and many others, the case for calling these chords dynamic is not undermined by the fact that they do not reach their natural destination. Regardless of whether or not the dominant chord resolves, or how it resolves, these examples share the fact that the dominant chord effectively makes a promise to resolve, and that promise must be fulfilled or reneged within a short amount of time—most often two beats or one measure.