[Smt-talk] I - II- IV as a progression

[Dmitri Tymoczko]

> Today I heard Donovan's 1968 "Atlantis," which prominently features  
> a I - II - IV chord progression, not unlike the verse of The  
> Beatles' 1965 "Eight Days A Week."  (I refer to the second chord  
> here as II rather than V/V because the dominant never figures into  
> the equation, except perhaps as a marked absence.)  Clearly, part of  
> the interest in such a progression is the chromatic descent from  
> scale degree five but I am curious as to whether there is a more  
> compelling and complete discussion of this harmonic move anywhere,  
> and also whether anyone knows of classical precedents for its use.

Very interesting question.  In classical-style diatonic contexts,  
direct progressions from ii (or II) to IV are extremely rare -- in  
fact, there's hardly a single example in the Mozart piano sonatas.  By  
contrast, IV->ii is extremely common.  This asymmetry between  
ascending thirds and descending thirds is in some sense even more  
striking than the classical asymmetry between V and IV.  (For example  
V->IV6 is reasonably common; whereas there's no inversion in which ii- 
 >IV is common.)  Furthermore, something similar can be said for  
almost all the ascending-third progressions, including viio->ii, IV- 
 >vi and I->iii.  (The one exception is vi->I6, which typically leads  
to IV/ii6.)  Making the ii chord major doesn't really change anything.

Off the top of my head, I suspect that I-II-IV is indigenous to rock,  
and wonder whether it might derive from the more bluesy bVII->I- 
 >bIII.  (If you reinterpret bVII as the tonic, you end up with I-II- 
IV.)  It would be interesting to trace its history in 20th-century  
popular music.

Another thought.  In classical music, you sometimes find progressions  
like D7->f6->E7.  (See, e.g. Beethoven Op. 54, second movement, mm.  
89ff or Mozart K. 576, movement 1, mm. 136-7.)  One way to interpret  
these is as an incomplete representation of a seventh-chord sequence  
D7->d7->E7, which involves only descending voice leading.  So you  
could see I-II-IV as being a mutated descendant of I->V7/V->ii7->V7,  
which though unusual isn't crazy.

Also, as you note, triadic ascending third progressions support  
descending stepwise voice leading in all three voices.  A cute fact,  
is that you can get descending voice leading for nearly-even chords  
whose roots progress by the interval -1 (mod 12/n) where n is the size  
of the chord.  So, for triads, that gives you -1, -5, and -9 semitones  
(= descending step, ascending fifth, ascending minor third) and for  
sevenths that gives you -1, -4, -7, -10 (descending step, descending  
major third, descending fifth, and ascending major second).

Nevertheless I would probably favor the idea that this is an import  
from the blues to more functionally tonal contexts.

DT

Dmitri Tymoczko
Associate Professor of Music
310 Woolworth Center
Princeton, NJ 08544-1007
(609) 258-4255 (ph), (609) 258-6793 (fax)
http://music.princeton.edu/~dmitri