I think it was Bryan that wrote "Whichever you like as long as you specify all the notes unambiguosly."
It has just occurred to me that the notion that the notes are in some way more fundamental than the mode is actually wrong. It assumes that the tune is always played/sung the same way. That does seem to be more or less true for songs (the singers I've heard don't seem to change the tunes much) but is *not* true for instrumental tunes. The better (folk) musicians that I know are *always* taking liberties with the tunes and if they wrote the tune themself then it can change every time. But there is something that doesn't change. I've heard it called "the bones of the tune". So the real tune is the sum total of all the variations that the composer would regard as not straying so far as to be a different tune. (A cloud of points in a high dimensionsal, abstract, tune space if you like sesquipedalion language). So what's actually written down is one sample version (one droplet from the cloud) which is somehow supposed to capture the essence of all those variations and ambiguities. The chord structure of the tune gives another stab at capturing it. That can have variations too. (Does John Brown's Body have the interesting B Em | A7 D7 | G ending, or does it just go G G | D7 D7 | G) so there's another set of dimensions to the cloud. The tonic and mode are yet another way of describing the tune. These too can be played about with (our band plays a minor key version of Black Jack), but to my mind they are usually a little more stable (we call it "White Jill" - so we sort of acknowledge that it's become a new tune). Laurie To subscribe/unsubscribe, point your browser to: http://www.tullochgorm.com/lists.html