A few posts ago I noted that I do not believe in the concept of modulation which, briefly put, is transitioning from one stable tonal center to another, usually via what is called a secondary dominant.
The reason I don't go that route is because it leaves out the fundamental nature of accidentals. (An accidental is a sharp or flat added into the sounds of the piece, not accounted for in the key signature.)
Some accidentals do lead to a change of key, but they do this by effectively untuning one of the home key's principal tones.
The major scale is comprised of two identical tetrachords, or sets of four tones, whose internal relationships are identical, one a fifth higher than the other. C D E F is the lower tetrachord of the key of C: whole step, whole step, half step. G A B D, the upper tetrachord, has the same internal spacing.
If I introduce a C# into the piece (as Mozart frequently does at the very beginning of a sonata) I undo that symmetry by untuning the G a fifth above C. This is a much more interesting way of letting the accidental work on the sound; it is full of potential.
