|
m. |
Event |
Cadence |
m. |
|
24 |
a1 |
59 |
PAC (i) |
|
59 |
tr |
91 |
HC (iii) |
|
91 |
b1 |
149 |
PAC (III) |
|
149 |
b2 (cl) |
177 |
IAC (III) |
|
m. |
Event |
Cadence |
m. |
|
273 |
a1 (mostly in i) |
HC (#iii) |
312 |
|
312 |
b1 (mostly in #iii) |
PAC (#III) |
370 |
|
370 |
b2 (cl) (mostly in #III) |
DC (i) |
412 |
|
412 |
Coda |
PAC (i) |
438 |
In the recap, cadences are hard to find! I've shown the simplest analysis above for the recap, omitting the transition and lumping some development-like passages into the second group. You may feel that some of these passages deserve separate mention.
Clearly A-flat major in the recap = G# major = #III, NOT flat IV.
Take note of the A minor passage in mm. 396-403. This patch of subdominant is as close to a statement of b1 in a closely related key as we ever get. It compensates somewhat for the #iii / #III statements of the second group in the recap, which are perplexing to those seeking to understand this movement in terms of the sonata principle.
If we consider the overall harmonic progression of the movement, we get
(Exposition:)
|
m. 24 |
91 |
149 |
|
a1 |
b1 |
b2 (cl) |
|
i |
iii |
III |
(Recap:)
|
m. 273 |
312 |
370 |
396 |
408 (or so) |
432 |
|
a1 |
b1 |
b2 (cl) |
Coda |
||
|
i |
#iii |
#III |
iv |
V |
i |
So even if we can't explain the movement in terms of the sonata principle, we can hear the exposition as unfolding an incomplete version of the i - iii - V - i arpeggiation of the tonic triad that is presented in its complete form in the recapitulation (with a picardy third!). And since Dvorak had already used natural iii and III (i.e. G minor and G major) in the exposition, he had to use the raised, major-mode forms of those keys in the recap.