Of course I knew the work of Roland Kirk and Harry Carney and the specific uses they would make of circular breathing, so I knew it was physically possible.

I think the solo playing, the decision to start playing solo, came out of having discovered what lay behind the doors that that technique opened for me.

In a certain sense, aspects of my solo playing were developed in order to test the theory about how long particular elements could be, as parts of so-called free improvisations.

Those early steps are very important in understanding the evolution. But in themselves, maybe now you need the later records to understand the significance of the earlier records!

To speak about notation as the only way that you can guarantee structure of course is already very suspect.

The argument we always used to use was that keeping records in the catalog was good for people that were coming new to the music, but I think that was talking over a ten year or fifteen year time span.

I'd met Roscoe in Europe quite a few times over the years, and we'd say hi and so on, but this was the first time we'd actually played together.