There was a thread on Lee Konitz's 10-step method for improvisation. It seems to have been removed. Shame, as I'd just written quite a long response.
The only source I know of for the basic idea is an interview from Down Beat, 1985, posted on the web by Mel Martin, at http://www.melmartin.com/html_pages/...ws/konitz.html You will have to open this page to see what on earth I'm talking about in the discussion below.
Someone asked if a specific description of each step in this supposed "10-step" process could be provided.
Lee Konitz may well have a "10 step method" for moving incrementally from 1) original melody to 10) pure inspiration, but if he does, there is not enough information in the Down Beat article cited to tell us exactly what it is.
If you want a more detailed description of what the various "gradients" involve, you may have to take a lesson with Mr Konitz, unless there is some other more detailed article or interview out there.
Having said that, I do think the article implies rather more content to the 10-step method than just "to start with a simple melodic variation on a tune and build it up in complexity over ten or so choruses" as one contributor suggests.
Konitz: "I suggest the kinds of compositional devices that are available: a trill, a passing tone, an appoggiatura that can bridge one melody note to another. The point is, you're still playing the melody, but you're doing something to it now. And there are many levels of this process before you get anywhere near creating new melody material."
and later: "I started backtracking, and it was in my own backtracking that it occurred to me that there might be a way of possibly taking some of the mystery out of the process with more knowingness."
I think [the original thread-starter] is impatient with the "mystery" and would like to get some of the "knowingness". Fair enough, as Konitz certainly implies that's the point of the exercise ....
So here's my first stab at articulating what seems to be implied by Konitz's scribbled examples.
Gradient 1) The original melody.
Actually this is not exactly as Kern wrote it. LK has already changed some rhythmic details, probably because he doesn't remember precisely what Kern wrote. However to all but the most finicky listener, this is the original line.
2) No comment from LK, but his example has identified "target" notes in the melody. These may be simply the longest and most noticeable notes, or (more likely) those that fall on the first beat of the bar (or wherever the underlying harmony changes) and that therefore act as signposts to the basic underlying (harmonic/melodic) skeleton. In this version he keeps those target notes prominent (and generally in the same location as the original) but adds additional shorter notes connecting them. So the initial f of bar one is preserved, the high bflat in bar 2 is still there, but interrupted by a neighbour note a, the initial a of bar 4 is there and moved back a beat, the gsharp in bar 6 similarly, the csharp of bar 7 as in the original (and similarly preceded by d as classical voice-leading prefers).
Still very recognisably the first 8 bars of ATTYA; though in my view Konitz has in fact gone rather further from the melody than "one step" would indicate to me. For example, between bar1 and 2 the simplest way of connecting f and bflat, if you want to introduce two new intervening quavers, is to have g and a, a simple ascending line. LK has already introduced a sort of contrary motion, one more level of unnecessary complexity.
NB ATTYA is a rather untypical candidate for this kind of treatment ... in fact it's rather too obviously qualified for it, as each main melody note is simply the third of the underlying chord .... in fact "third of underlying chord, chords move in diatonic cycle of 5ths starting on VI minor, key change between chord 4 and chord 5 effected by using diminished 5th as the step instead of perfect 5th" is a description of the musical content of the first 8 bars of ATTYA.
So my summary of step 1 (slightly different from what LK's example implies) would be: -- identify "target tones"; preserve them in your new line but you may introduce new passing tones between them, shortening the target tones as desired
3) Every pitch of Kern's melody is preserved, but concealed in additional notes.
My summary (again, not exactly what LK's example demonstrates); - keep target tones and passing tones ... now you may also introduce neighbour tones, and you can experiment with changing melodic direction (i.e. where the original has an ascending jump, instead of introducing ascending passing tones, use a descending line and a wider skip, or a wide skip and then a descent to the target note).
4) Imagine that there is a line entirely composed of quavers (8th notes) in which the target notes are still to be heard. Play this line, introducing rests to create logical musical phrases. Use the target notes, passing tones, upper and lower neighbour tones.
5) You may introduce arpeggiation of the underlying chord, and chromatic neighbour tones. It now may be desirable to displace the target tones further from their original positions (i.e. anticipate them or delay them).
6) at last a comment from LK, though not a very transparent one "Still using melodic targets but displacing for new melodies" The target tones all appear in the bars they belong to, but are no longer necessarily regarded as the most important tone; for example, though bflat still appears in bar 2 it is simply there to characterise the underlying Gm7 chord, and the most prominent note in the melody is now a high a (the ninth). Perhaps generally the student should now be aiming at creating a melody where higher chord tones (sevenths, ninths etc) are more prominent in the melody than 3rds, roots or fifths.
Alternatively LK may simply mean, "Keep your target tones, but you can now transpose them into a new octave, and try and make them subsidiary to other tones if you can"
7) Comment from LK "More new melodies" -- process of introducing higher chord notes (and altered chord tones, e.g. the flat 9th and flat 10th of C7 in bar 3) and more chromatic neighbour notes; perhaps introduce different rhythmic values, i.e. no hint of the original note values, but hopefully your phrasing now also conceals the subterranean quaver melody I suggested in step 4.
8) LK: "Still a subtle reference to the original song" -- I guess this means the target tones or some characteristic intervals are still preserved, though completely subservient to the new fabric.
9) LK: "Totally new theme" -- well, this is arguable ... I think an experienced jazz musician who knew the tune would listen to someone playing this line and would at once say, "Well, it's All The Things You Are, ain't it?" But I suppose the original melody is no longer present.
However note that LK uses NONE of his original "target" tones in bars 1,2,3 and 5 ... though he keeps the a in bar 4 and the gsharp and csharp in bars 6 to 8; these notes are so crucial to the harmonic structure that avoiding them altogether would have been rather perverse (I mean it's quite possible to do it, but you'd be likely to end up with something that sounded rather colourless).
10) LK: "An act of pure inspiration" ... yeah well, who knows. I guess the challenge is to come up with a melody so strong that anyone who hears it says, "Yeah, it's kind of like All The Things You Are, but better..." Maybe like take 3 of Charlie Parker's "Bird of Paradise" if you like that kind of playing (though maybe Bird is actually at step 8, as described by Andre Hodeir, "revealing and concealing the original melody in the folds of his magician's cape").
Anyway, that's one attempt, I daresay Lee Konitz would have said different. I'm afraid it was beyond me to do this "in simple English with generally accepted musical terminology" as was asked; it was as simple as I could make it without spending hours on it but I don't know how to explain these things without using quite technical terms -- after all musical terminology was developed because everyday language didn't have the necessary concepts. If someone doesn't know what "passing tones", "neighbour tones" "skips", "roots" or "ninths" are, or what I mean by "harmonic structure' then they'll need to read up some music "theory". (It ain't theory at all, in reality, it's just nomenclature).
I may have overstated the concentration on pitches. A different and possibly quite enlightening direction could have been taken by concentrating on rhythmic figures -- e.g. preserve the original note values of the melody, but ignore the original pitches -- use, if possible, a single pitch, throughout, or if that is harmonically ugly, use as few pitches as you can. Now introduce rhythmic complexity and new figures to the line but don't change the pitches.
I may say if I wanted to present this kind of material to a student, I'd simplify it quite radically, so that the instructions were as unambiguous as possible -- Identify target tones; play ONLY target tones; introduce diatonic passing notes; introduce neighbour notes; arpeggiate chord in bar 3: etc etc. However that is only a way of getting them started, there has to come a point when the student can follow things for themselves, i.e. intuitively, to some extent.
And some people get the idea WITHOUT any technical description at all, just by listening and having a go. Depends how "fuzzy" you like your thinking and learning to be.