What makes me say that a particular song is a blues song? Specifically in terms of form. I consider the three previous examples to be blues songs, but none of them follow the standard format. In fact, so far we've had quite a few examples of songs that I think of as blues songs, and none of them follow the standard format, either! So who and what is right?
I actually think that last question leads to a dead-end and is musically, culturally, and pedagogically uninteresting. A quick digression to illustrate a point: In classical music, we often learn about sonata form as a very formulaic set of phenomena or rules that are followed. But when you look at the many, many, many pieces that use this form, few of them (if any!) actually follow the supposed format, even though theses pieces still obviously qualify (and not just because they are titled as such). High up in academia, musicologists began thinking about sonata form in a more flexible, open way as a spectrum of practices/phenomena/ingredients. I don't believe in trickle-down economics, but I do believe in trickle-down academics. Eventually, the ideas from the so-called ivory tower are synthesized and flow down to become practical pedagogical tools. "Sonata theory" is now being taught at the undergraduate level, instead of the stricter, more formulaic ideas we used to teach about the form.
Usually when you learn about the blues as a form, you learn a few set things:
there are 12 measures, or bars
there are specific harmonies that happen in specific places within those twelve bars
the lyrics are often set up in an AAB form, in which B is a sort of response to the situation presented in A (like an antecedent-consequence, to get fancy about it)
there are specific pitches or pitch areas that do not happen in European folk or classical musics — we call these "blue notes"; otherwise we try to stuff the square peg of blues into the round hole of European music theory1
Unlike with sonata form, there are in fact many, many, many blues songs that do follow these exact specifications. Yet in the same way that something sounds definitely like a sonata, but doesn't have all the same ingredients as all the other sonatas, there are songs that sound like blues songs in form (as well as, say, blue notes in the melody), but don't follow the same road map. To see and hear how this is so, let's look at what is usually presented as the standard:
With scale degree chord function indications:
I7 IV7 I7
|/ / / / |/ / / / |/ / / / |/ / / / |
IV7 I7
|/ / / / |/ / / / |/ / / / |/ / / / |
V7 IV7 I7 (V7)
|/ / / / |/ / / / |/ / / / |/ / / / |<— This happens when you are repeating the form.The end of the song ends on the tonic.With Riemannian chord function indications:
T S T
|/ / / / |/ / / / |/ / / / |/ / / / |
S T
|/ / / / |/ / / / |/ / / / |/ / / / |
D S T (D)
|/ / / / |/ / / / |/ / / / |/ / / / |<— This happens when you are repeating the form. The end of the song ends on the tonic.In jazz (and probably before that?), people started combining the sounds from more harmonically complex Tin Pan Alley/Vaudeville/Broadway songs, and adding more chords. This is where Riemannian theory is kind of nice because it's based around the idea that other chords are simply extensions or substitutions for the basic tonic, subdominant, dominant functions. So that when you, say, put a vi chord where the I chord should be, that is explained by the idea that the relative minor chord (or major, if we are in minor) can act as a substitute for the tonic — an idea that is obscured a bit by using abstract Roman numerals. Here's a common souped-up jazz version (there are many):
With scale degree chord function indications:
I7 IV7 I7
|/ / / / |/ / / / |/ / / / |/ / / / |
IV7 I7 iii7 V7/ii
|/ / / / |/ / / / |/ / / / |/ / / / |
ii7 V7 I7 (ii7 V7)<— when repeating
|/ / / / |/ / / / |/ / / / |/ / / / |With Riemannian chord function indications:
T S T
|/ / / / |/ / / / |/ / / / |/ / / / |
S T Tg D/Sp
|/ / / / |/ / / / |/ / / / |/ / / / |
Sp D T (Sp D)<— when repeating
|/ / / / |/ / / / |/ / / / |/ / / / |Tg is the Tonikagegenklang, which takes some explaining. To Riemann, chords that had two common tones could substitute for one another. Gegenklang refers to the leading-tone — the 7th scale degree that "wants" to lead up to the tonic, because it's only a half-step away. The chord built on the mediant (iii) can substitute for the tonic because it shares two common tones and the non-common tones are only a half-step away.
Sp is the Subdominantparallele, the relative minor to the subdominant (Germans call this parallele, which is confusing to us English speakers, since we use that word for a different concept. Be patient with yourself and other languages.). The relative relationships between any two chords also share two common tones; so the chord on the supertonic (ii) can appear where you might usually have a subdominant (IV) chord. We often call these "pre-dominant" functions. D/Sp is the dominant of the relative of the subdominant. These double-dominants are the kind of thing you can stick before any chord to add harmonic interest,2 and create a sense of motion — the beat doesn't go faster, but the when the rate of harmonic change increases, that can create a different sort of motion and emotion.
Up to this point, we've relied on scale degrees and Riemann. What about Tagg? Tagg developed his terms to talk about shorter, cyclic progressions and says specifically that the 12-bar blues is too long to be considered a loop. However, I've found his terminology to be useful, as I've written elsewhere, to help clarify the idea of chord functions in plain English, where scale degrees and Riemann are more abstract.
In classical music, the tonic is "home." It's the place where the music seems to come to rest, at the phrase level, at the section level, at the entire piece level. Still thinking in a larger scope, the subdominant usually serves the function of what Tagg calls the "counterpoise" and what I like to call simply "away." Where do you go when you are not home? Away. The dominant usually serves the function to signal that we are returning home; Tagg calls this "incoming." Overly simplified, everything else to the classical music theorist is decoration. It extends or decorates one of these three functions. We can add the fourth Taggian function which is "outgoing," that's the chord that signals we are leaving home, that transitional space between home and away.
When we look at our blues progressions in this way, we see something interesting happening with the IV and V chords. Starting with the first progression:
H = home A = away I = incoming O = outgoing
H A H
|/ / / / |/ / / / |/ / / / |/ / / / |
A H
|/ / / / |/ / / / |/ / / / |/ / / / |
? ? H (?)<— for repeating
|/ / / / |/ / / / |/ / / / |/ / / / |As we've seen so far in this whole blog, when you only have two chords, you can only really have home and away. In that two-chords context, it's almost as if outgoing, away, and incoming are merged into one thing. Since Tagg conceived of his terms in the context of short, repeated phrases, it makes the most sense to apply them phrase by phrase. The difficulty arrives in the last phrase, which is usually dominant-subdominant-tonic-(dominant). If the dominant is incoming, then it should go directly to the tonic or a tonic substitute. Yet, it definitely does not go directly to the tonic. Can the subdominant act as a substitute for the tonic? After all, it does share two common tones. In Riemannian terms, you might be able to make that argument. Yet, we've already heard many two-chord songs in which the subdominant is the away chord — meaning it's also the outgoing and the incoming all rolled into one.3
Usually in a four-chord loop you have:
tonic/home-outgoing-medial- (in this case, the chord that's in the middle, but could also be considered as counterpoise/away) - incoming
This way of thinking about functions applies nicely outside of loops, too. Since we know what the tonic is, we can fill in the rest of the functions in the last phrase of the 12-bar blues using this loop order. That is, before home, we have to have incoming; and before incoming we have to have away/counterpoise/medial.
H A H
|/ / / / |/ / / / |/ / / / |/ / / / |
A H
|/ / / / |/ / / / |/ / / / |/ / / / |
A I H (?)<— for repeating
|/ / / / |/ / / / |/ / / / |/ / / / |In a loop, home would be followed by outgoing, but in this case we are going back to the top of the form, which starts with home. So we can call the single turnaround chord away:
H A H
|/ / / / |/ / / / |/ / / / |/ / / / |
A H
|/ / / / |/ / / / |/ / / / |/ / / / |
A I H (A)<— for repeating
|/ / / / |/ / / / |/ / / / |/ / / / |I mean, we could also call it a turnaround, but let's stick with away for now. Comparing Tagg with scale degrees and Riemannian names, it should be clear now that a single chord sonority can have different functions, depending on where it appears in the song. Sometimes the IV is away; sometimes the V is away. Using Taggian terms makes this really explicit, which is why I've become so enamored of them. The Riemannian terms can show how we can vary our harmonies with substitutions. Would I teach all of this? Probably only to high school and above, and who were choosing to take a theory course. Otherwise, keep it simple.
Speaking of keeping it simple, now I want to show how we can simplify this form and still have it feel like the same form. This will create a different sort of motion and emotion. Some other element will carry the song.
For example, we can take out that first switch to the subdominant and back. We don't need it. It's not structurally or even narratively necessary:4
With scale degree chord function indications:
I7
|/ / / / |/ / / / |/ / / / |/ / / / |
IV7 I7
|/ / / / |/ / / / |/ / / / |/ / / / |
V7 IV7 I7 (V7)
|/ / / / |/ / / / |/ / / / |/ / / / |With Riemannian chord function indications:
T
|/ / / / |/ / / / |/ / / / |/ / / / |
S T
|/ / / / |/ / / / |/ / / / |/ / / / |
D S T (D)
|/ / / / |/ / / / |/ / / / |/ / / / |We can ditch some of the activity in the last phrase, too. This is precisely what happens in Las Vegas Tango:
With scale degree chord function indications:
I7
|/ / / / |/ / / / |/ / / / |/ / / / |
IV7 I7
|/ / / / |/ / / / |/ / / / |/ / / / |
IV7 I7
|/ / / / |/ / / / |/ / / / |/ / / / |With Riemannian chord function indications:
T
|/ / / / |/ / / / |/ / / / |/ / / / |
S T
|/ / / / |/ / / / |/ / / / |/ / / / |
S T
|/ / / / |/ / / / |/ / / / |/ / / / |Gil Evans gets rid of the turnaround chord at the end, too. Let's look at this in Tagg functions:
H
|/ / / / |/ / / / |/ / / / |/ / / / |
A H
|/ / / / |/ / / / |/ / / / |/ / / / |
A H
|/ / / / |/ / / / |/ / / / |/ / / / |Real simple, right? But we can go further, and people do. Now that the second and third phrases are the same thing, we can just get rid of one:
H
|/ / / / |/ / / / |/ / / / |/ / / / |
A H
|/ / / / |/ / / / |/ / / / |/ / / / |Voilá: an 8-bar Blues. And here's the interesting thing: That away chord could be either a IV or a V. Rockin' Pneumonia, Tol' My Captain, and Keep A-Knockin' use the V as the away chord; the verses of The Name Game use IV as the away chord in an 8-bar form. Astute readers will notice this is also the harmonic rhythm that I like to call four-two-and-two. There are blues-based songs that use other harmonic rhythm patterns, but four-two-and-two feels like a 12-bar blues, even though it's not.
Chuck Berry's Memphis, Tennessee elongates the form (20-bar blues), while still keeping it harmonically simple. It maintains the home-away-home-away-home set-up of a 12-bar blues, but the time spent in each area is quite different. Yet again, somehow it feels like a 12-bar blues, even though it's not.
Then there are certainly blues-flavored songs that are unrelated to the form. We've covered feel like jumping, can i kick it, achy-breaky heart, but there are zillions of others that we may never look at.5 Usually when people talk about The Blues, they mean the flavor plus the form.
Our internet-friend here on substack, Ethan Hein has written extensively about this topic and you can read what he has to say, if you'd like to dig further:
https://www.ethanhein.com/wp/2011/blues-basics/
We are wise to also remember that European music theory is to a large extent a square peg itself, stuffed into the round hole of ancient Greek music theory. It's a giant tradition of trying to reconcile practice (or, really, practices — after all, Europe is hardly a monolithic culture. What place is?) with neat, tidy theoretical ideas from someone else's practice. Also: that's why it's "theory" and not "science."
They can also be overdone, but I definitely believe in overdoing something as a pedagogical tool, so boldly experiment with inserting more double-dominants in chord progressions!
When Tagg looks at these kinds of loops, sometimes the function is determined by the harmony, but sometimes the function is determined by its temporal place in the loop. But I don't think it's the case here…
I suppose we could say that about a lot of things that happen in music. That would be the crux of Romanticism (all those gossamer details!) vs. Minimalism (only what we need!). Necessity would depend on your world-view…
Such as the oeuvres of the Rolling Stones, the White Stripes to name but two out of said zillions.
Thanks for the post. I like the home/away approach and appreciate too how you offered a few different lenses of interpretation for each of your examples.