Apr 11, 2015

Early Blues, Blue Notes, and Blues Scales

Here are a few thoughts regarding blues history and blues usage, that came up in the course of reading several interesting books on early published blues (see this review and this review). I don't claim to be an authority on the subject, beyond the fact that I'm a sax player who has played and listened to a lot of blues over the years.

Blue notes and Blues Changes

Composers of early published blues drew inspiration from folk musicians. In the absence of better documentation, examining these compositions can be one way of trying to understand what early (pre-1910) folk blues and pre-blues might have sounded like.

Early published blues, in turn, influenced the development of popular music in the years that followed, especially jazz and jazz-oriented blues, but published blues influenced subsequent folk blues as well.

In his autobiography Father of the Blues (1941), W. C. Handy describes his creative process in writing "St. Louis Blues" (1914):
[Black Southern folk musicians were] sure to bear down on the third and seventh tones of the scale, slurring between major and minor. Whether in the cotton fields of the Delta or on the levee up St. Louis way, it was always the same. Till then, however, I had never heard this slur used by a more sophisticated Negro, or by any white man. I had tried to convey this effect in Memphis Blues by introducing flat thirds and sevenths (now called "blue notes") into my song, although its prevailing key was the major; and I carried this device into my new melody as well. I also struck upon the idea of using the dominant seventh as the opening chord of the verse. This was a distinct departure, but as it turned out, it touched the spot.
It appears that Handy may have gotten inspiration not just from folk blues, but in part from Anthony Maggio's earlier published tune, I Got the Blues (1908). Maggio's melody consisted of a repeated riff, clearly meant to mimic the bent third that Maggio had heard from a folk musician. Maggio's b3-3-1 phrase was used by Handy in his "Memphis Blues" (1912), "Jogo Blues" (1913), and "St. Louis Blues" (1914).

Maggio's riff as it was used in "St. Louis Blues" (click to enlarge):

Handy's prior awareness of Maggio's tune is pretty clearly demonstrated in Peter Muir's book, Long Lost Blues. But be that as it may, there are two other noteworthy points in Handy's statement:

1) He describes b3 and b7 as blue notes, but not b5.

2) Handy describes the seventh as being bent, like the third, in folk usage. I don't doubt that, but in early published blues by Handy and others, when the b7 of the key is used in the melody or harmony, it is almost always in the context of setting up a I dominant (V of IV) sound, preparing a IV chord. That goes for the first bar of St. Louis Blues as well - perhaps a "departure," as Handy stated, but still acting as a V of IV, preparing the IV in bar 2.

Use of b7 notes in the context of a V of IV can be found in Sousa marches, arrangements of church hymns, and in popular classical themes - musical settings that were commonly heard in the late 1800s and early 1900s. Musicians like Handy, Maggio, and Artie Matthews (arranger of "Baby Seals Blues," pub. 1912) were relatively well-educated, and probably would have known quite well how a secondary dominant works. Usage of the b7 in that context is not at all unique to blues.

Vic Hobson, in his book Creating Jazz Counterpoint, has pointed out that adding a b7 to a triad was a common practice in barbershop singing - a popular pastime in the African American community in those years.  That may well have been another factor in b7 blues usage. But as far as a unique barbershop influence goes, I’d look more to the b7 that is added to the IV chord in blues, resulting in a b3 of the key (a non-classical usage), or to the use of a #IVdim7 in bar 6 of the progression, a feature that goes back to some of the earliest published blues.

In the first published blues tunes, and in late 1910s/early 1920s recordings, most “blue note” usage seems to occur on the third of the key. In sheet music intended for piano, a bent-note effect could be expressed by a b3 to 3 grace note, or by putting a b3 on the beat, resolving then to the 3 (as in the example above), or by simply sounding a b3 note over a major or dominant chord. Melodic b3 notes can also appear in measures harmonized with a IVdom chord (where the b7 in the chord is the b3 of the key), or in measures harmonized with a V7 (the b3 of the key is a #5 relative to the chord).

In early published songs and early recordings, the third seems to be scooped or bent far more than the seventh, which is relatively stable (still true now, IMO).

The b5 rarely occurs in these early blues. 

The M6 note is also an important part of blues vocabulary, then and now. It wasn’t flagged as a “blue note,” though, because it is a part of our familiar major scale. 

Gunther Schuller, in his book Early Jazz, traces b3 and b7 blue note usage to the b7 note that he states was common to the music of Central/West African societies, with the b3 resulting from singing parallel lines. I don't doubt that African music often used b7 notes, but the prevalence of the b3 and bent third in early blues, and the relatively “classical” usage of the b7 (V of IV context) when it does occur, do not seem to support this explanation.

The nature and degree of African influence on the development of blues styles is an open question. Some blues scholars see extensive African influence, others see relatively little.

Willie Ruff has suggested the possible influence of “lining out” worship singing, introduced in America by English and Scottish settlers perhaps hundreds of years ago, on popular music practices in the South. "Lining out" singing still exists in a few congregations in Scotland (Presbyterian) and in the South (Baptist) - see this very interesting video. Check out the similarities and differences in the melodic embellishments and bent notes used by congregations from Scotland, Kentucky, and Alabama. There's room for some speculation on the influence of this type of singing in the origins of blues and gospel style.

In my non-expert opinion, recordings by male "down home" blues artists in the late 1920s seem to show more use of the b7 as a melodic blue note than we see in early published blues, or in recordings of female blues singers in the early 1920s. This may reflect the rural folk style of earlier, undocumented times (pre-1910). On the other hand, styles can change over time. Jazz recordings of blues tunes seem to show freer use of the b7 as a "blue note" (not just a secondary dominant note) as time goes on, from the 1920s through the 1940s. 

The b5 seems to have been used increasingly beginning around perhaps the early 1930s, and became an integral part of blues/jazz/pop melodic vocabulary (e.g., Ellington's 1931 tune "It Don't Mean a Thing," or Matt Dennis' 1941 "Angel Eyes"). By the 1940s, blue-note b5 licks were a part of the bebop language.

Of course, since musicians have been taught that there is a "blues scale" for perhaps 70 or so years, it is now universally accepted that blues-scale-derived licks are basic blues melodic vocabulary.

Blues scales

As I see it, it's a bit misleading to teach students that playing a “blues scale” is the way to create a good blues solo. Historically, and to this day, much of the blues musical vocabulary does not conform to any sort of “blues scale.” In the early days of jazz and blues, the concept of "blues scales" did not exist. The idea of a “blues scale” seems to have come about in the late 1930s, when academically-inclined musicians looked for some sort of underlying principle that would explain the use of “blue notes” in a major-key musical context.

A number of different “blues scales” have been proposed over the years. The one that most of us have settled on (1 b3 4 b5 5 b7 1) has some utility: We can give it to beginning improvisers, and they will usually sound good immediately, which inspires self-confidence. I do teach this way, but I always follow up by saying that lots of great blues licks don’t use the scale, and that any note could sound good, depending on how it is used.


  1. As usual Peter, you have tweaked my interest in late nineteenth- and early twentieth-century American popular music.

    One of the earliest published 12-bar blues songs is "Negro Blues," copyright 1912 (published as "N....r Blues" in 1913 by Le Roy White. There are several versions on YouTube. It follows the pattern of early blues that you noted above in that the b7 is used on the I chord as a dominant preparation for the IV chord and there is no use of the b5 as a melody note. Of interest is the ambivalence about the use of the b3 with the IV chord: in measure 5 of the form, the natural 7th of the IV chord is used as a melody note but in measure 6, the 7th is flatted, as it would be today. In all other instances, the third of the tonic scale is always natural. Thus, there is little use of "blue" notes in this early example of a published 12-bar blues.

  2. Hello Peter,

    I'm a long-time blues researcher and it's refreshing to see someone so on the right track, imo.

    If I understand right, the modern nonsense we see about b5 being a blues note comes from theorists wanting to see the use of b5 in blues by modern _jazz_ musicians as significantly related to early blues, but it wasn't, it was inspired (in early bebop, and among jazz musicians who influenced bebop, such as Roy Eldridge) by classical.

    "[S]imply sounding a b3 note over a major or dominant chord" was routinely done vocally by early-born black and white folk artists, in non-blues and blues. Scales similar to 12356 and 1b345b7 were both well-known to black folk musicians. Those scales (sometimes both of those mixed with each other) were superimposed by blacks onto the chords taught in the guitar instruction books that came free with guitars in the mail in the 1890s on. And whites did much the same.

    Of course for pre-1910 sounds we can research the ages of recorded blues musicians and when they said they began singing and playing. Mance Lipscomb learned "All Out And Down" in roughly 1909. Gus Cannon learned "Poor Boy Long Ways From Home" in roughly 1902. W.C. Handy recalled hearing "Got No More Home Than A Dog" before 1900, and he was recorded singing it and accompanying himself on guitar in 1938.

    Here are some blues guitarists who were born before 1893:
    Daddy Stovepipe (e.g. "Sundown Blues") (implied in an interview that he began playing guitar before 1900)
    Henry Thomas (e.g. "Lovin' Babe") (reportedly born in 1874)
    Crying Sam Collins (e.g. "Yellow Dog Blues")
    Frank Stokes (e.g. "Bedtime Blues")
    Peg Leg Howell (e.g. "New Prison Blues") (recalled that he began playing guitar in 1909)
    Charley Jordan (e.g. "Dollar Bill Blues")
    Allen Shaw (e.g. "Moanin' The Blues")
    Charlie Patton (e.g. "Green River Blues") (his friend Booker Miller recalled that Charlie said he began playing guitar at about 19, i.e. in about 1910)

    The influence of lining out on black folk music has been discussed frequently over many decades, long before Ruff got press for doing it again. The Edison magazine in 1919 described lining out heard in Southern blues in about 1909, e.g.

    There is no "the blues scale." Even individual artists, such as Lemon Jefferson, would emphasize closer to "minor pentatonic" in one blues piece they knew and closer to "major pentatonic" in another.

    As you note, the purpose of "the blues scale" as typically taught is to make the beginner sound (superficially) "folky"/"funky" quickly.

    1. Joseph,

      A few comments:

      Thanks for the list of blues musicians who learned their stuff pre-1910. All of the recordings you mentioned except Handy are on YouTube; I checked them out. Overall, I hear a lot of what today we would call major pentatonic with bent thirds. Considering that these were were recordings by ten different players, from various geographical areas, it seems reasonable to think that this was generally how blues (or pre-blues) sounded pre-1910: lots of bent thirds and major 6ths, a few b7’s.

      Early folk musicians surely did not think in terms of “major pentatonic” or “minor pentatonic” scales. I’m sure that you did not mean to suggest this, but were only using these scale terms as a convenient way to describe sets of notes. I think that to early musicians, it was all about licks and melodic vocabulary, not scales.

      About the b5: Although it apparently wasn’t part of the blues language pre-1920, by the 1930s licks with b5’s were solidly in the jazz/blues vocabulary. You stated that you believe the b5 came from classical usage. I’d be very interested to know about any pre-1930 classical usage that resembles the b5 blues licks in Louis Armstrong’s Hot Five recording of “St. James Infirmary” (1929), Ellington’s “It Don’t Mean a Thing” (1931), Matt Dennis’ “Angel Eyes” (1946), or Charlie Parker’s “K. C. Blues” (1951). All of these use a similar b5-4-b3-1 shape, and all of the pieces except the Parker track are in minor.

      The 1940’s bop players also used the b5 in a harmonic context, and as a melodic color note, and those usages may owe to classical, but b5 blues usage is something else again. I really don’t know how it entered the vocabulary, but I’d have to be convinced that b5 blues usage was borrowed from classical.

      I absolutely agree that for early players, there was no “blues scale,” and they didn’t much use the b5. But usage changes over time, and the b5 has become integral to the blues language. Bent notes have become more extreme, too (think of Hendrix-derived rock guitar).

      I use the blues scale in teaching not just to make the beginner sound “funky” quickly. The process is more like this: I give the student a limited set of notes (i.e., a blues scale) to work with, which makes it easier for them to get started improvising. The student tries a few notes, and quickly goes into a comfortable blues zone, as the scale reminds the student of licks that he or she has heard before. Students usually don’t just run the scale - rather, the scale helps the student fall into “ear” playing (a good thing). Students usually sound pretty good right away, which gives them confidence in their creative efforts (a good thing). And they have fun (a good thing).

  3. "Overall, I hear a lot of what today we would call major pentatonic with bent thirds. Considering that these were were recordings by ten different players, from various geographical areas, it seems reasonable to think that this was generally how blues (or pre-blues) sounded pre-1910: lots of bent thirds and major 6ths, a few b7’s." Agreed.

    "... I’m sure that you did not mean to suggest this..." Right.

    "I think that to early musicians, it was all about licks and melodic vocabulary, not scales." I think they thought in terms of, had familiarity with and liking for, rough scales. And (roughly speaking) would often combine two of those rough scales, just as Lennon and McCartney for that matter would.

    "I absolutely agree that for early players, there was no 'blues scale,'" My understanding is that during the '10s and '20s, people who encountered Handy's publications and _were ignorant of Southern folk music_ thought of normal black folk scales as "blues scales," because of their interest in the blues fad as such, when they are better described as normal black _non-blues or blues_ scales of the 1890s-1910s.