Pirooz Emad wrote: ↑
Thu May 18, 2017 6:19 am
In m. 23 Sor makes a strong approach to the dominant by half steps from both directions at once.
I certainly agree with this bit.
I would suggest the following:
At a cadence like this, it's often worth considering whether the first chord of the cadence could be a dominant 7th. Here the cadence goes to A, so we would be looking for E7 (E G# B D). That is exactly what we find, except that in place of the B we have a Bb. It's easiest to come back to that later and treat the chord as E7 for now, to see where that takes us.
The Neapolitan sixth chord is often / always found before a dominant seventh. It looks like the major triad a flat fifth above the root of the dominant seventh, so here we would be looking for the same notes we find in Bb major (Bb, D, F). That is what we find, except that there is also an A.
One theory holds that the Neapolitan sixth is an altered version of the IV chord, meaning that in this case Bb would be standing in for A and the A itself could (re)appear while the chord was sounding. Interestingly, that is exactly what happens on beat 2.
The question that leaves us with is why Sor does not go to the B on beat 3, instead giving us an altered version of the E7. It often happens - especially in basslines - that a note that would be a whole step away from where it is going is replaced by the note a half-step away. Here, rather than have the bassline Bb B A, Sor has held on to the Bb giving us a chromatic move down to the root of the tonicized A. It's not really so different from a progression like E7 E7/Bb / A, which you get all the time.
There are loads of video lessons and resources out there that help you assign names to chords, but in my opinion it is more useful to consider what the underlying progression is and what alterations have been made. This is really just a IV / V / I progression in A, but Sor has put in 3 or 4 alterations that make it much more interesting without altering the basic function. One mark of a good composer!