Signs that this may be a day of things not working.

I went back to the original piece that the last eleven bars come from and tried to follow through the serial logic. It was very clever � but it was one of the last pieces that Webern published � I read somewhere that there were many different accounts of the way that serial patterns had led to the piece being as it is.

Perhaps in consequence I did have a �Fifths� thought though � a kind of symmetry thought. There is a harmony made of two fifths, say C and G plus A and E � the notes of an Am7 chord. The C/G fifth also occurs in a Cm7 chord, the A/E in a F#m7 which includes as its other fifth F#/C# and this appears in the Ebm7 chord. This set of overlapping fifths could be used to make another piece like the existing Fifths

But then it occurred to me that this set of notes also contained the possibility of less static structures. Take all the notes from this set and make a scale: C C# D# E F# G A Bb � 8 notes � an octatonic scale . The four notes outside the scale make a diminished chord D F G# B.

The octatonic can be used over a set of dominant seventh chords � on the same four roots as the minor seventh chords that the scale was derived from . This is a bebop trick which is used to make ordinary chord sequences sound more outlandish

Suppose the notes making up the dominant seventh (eg C7) are taken out then the remaining notes from the octatonic scale are C# D# F# and A � this is a D# half diminished � the top half of a B9. So if you use the octatonic scale over the C7 then you are throwing in some �weirder� notes which give a more modern sound. In effect you are playing the required chord and the chord that it might move to. I used to use this ploy in the Steve Pheasant band a fair amount.

But I was surprised to see this scale emerging as part of the Fifths analysis � not where I expected the analysis to end up.

The notes outside the octatonic scale make a diminished chord - B D F Ab. As there are only three diminished chords � the other two must be in the octatonic scale. Somewhere in these structural relations I have the sense of another piece lurking � apiece which exploits static elements of fourth or fifth chords but also the dynamism of dominants � while avoiding conventional harmonic relationships.

Another idea which I wish I had time to follow up involves using quite bluesy guitar timbres to play conventional Broadway standards. I tried a little today first thing. I felt it was leading somewhere but I had to interrupt the flow of ideas � something I always resent.