Writing out an improvisation

[ronde_des_sylphes]

So, there's been some discussion on here about the crossover between improvisation and composition, specifically mentioning the notion of composition through improvisation and re-improvisation.


I've had a few improvisations on here which I am very partial to and I believe are sufficiently strong that they can transition to compositions with a bare minimum of editing. However, that means writing them out or learning them by ear.. with that in mind, I have been writing out the improvised piano sonata I uploaded last year. Half way there..

Here is the finale. It isn't absolutely 100% notationally perfect as a representation of what I played, but I'm trying to compromise between what is in the improvised audio and what will make most compositional sense on paper, trying to keep it as close as possible to the original while accepting that certain imprecisions and ambiguities are a part of improvisation and not going overboard trying to notate them pedantically.

This segment took three days to write out

[ranjit]

Nice work! I was surprised to realize that I still remember fragments of the melody here from when I heard it a few months back, which I think speaks to the quality of your writing. It's a very memorable melody at the start especially, and a composition taking that as its basis could work really well imo.

[ronde_des_sylphes]

Thanks! I'm happy to have it as the finale of the whole larger scale work-to-be, as I think it has a lot of presence and ends everything on a suitably dramatic note. The full improvised sonata will gradually get written out: I probably have 7 mins of the 18 total playing time left and at that point I will consider if there are any editorial and/or compositional changes I deem necessary.

[quantum]

Great stuff!

I appreciate the amount of work that must have gone into this.  Unlike a transcription of another performer whereby the object would be to be as reasonably accurate to the recording as possible, you have to balance the directive as creator of the recording and creator of the score.

[ted]

Firstly, the resulting composition is very enjoyable and to your credit regardless of its mode of creation. The functions of written notation and spontaneous playing, how they interact in the creative mind, are both diverse and highly personal among musicians and players. I have stated my own position on it several times on forums but it is just my view and holds no universal validity. In earlier years I wrote out a large number of compositions originating from improvisation and I know only too well the hard grind involved so your persistence is to be highly commended.

[ronde_des_sylphes]

I have a certain agenda in mind. I am *really* happy with the full (18m) improvisation, but it goes without saying that a home recording during circumstances so unavoidably unfavourable through global events that I can't even get the instrument tuned doesn't do the improvisation full justice as a musical entity. Swings and roundabouts. Would I have improvised something I am so partial to under "recording" or live performance situations? Maybe, but really I can't guarantee it.9

So.. at the end of the day, a full score will enable me to perform it as a finished composition should live recitals return, or even to record it under studio conditions. I don't know, philosophically, if it will necessarily have the same immediacy when it becomes "interpreted" as opposed to improvised, but I feel it would have a lot of power in a concert hall situation.

En passant, movement two, which has been written out in entirety, not absolutely perfectly but to the point where it should serve as a basis for future performance:

I'm taking my time and care over this because, as both of you clearly realise, this is a major undertaking. Addressing quantum's point, in a sense I am happier writing it out in a dual composer/transcriber role because in a way it is more comfortable to write it out, where ambiguity exists, as "this is clearly what was meant, compositionally" (and, to be clear, I do feel my role as improviser is one of real-time composition) rather than an attempt at a pedantic rendering. I dread those moments of notational and metrical irregularity though - I think editorially resolving them satisfactorily costs the most time.

One plus is I know what figurations my hands naturally gravitate towards!

[ronde_des_sylphes]

As my personal view is increasingly to improvise as a form of real time composition, I do try to maintain a form of structural cohesion even if it is simple as bringing motifs back albeit in different associated contexts. So there is definitely an element of discipline in there, though the process of music creation remains based more within intuition rather than analytical and purely calculated development.

[ted]

I dread those moments of notational and metrical irregularity though - I think editorially resolving them satisfactorily costs the most time.

One plus is I know what figurations my hands naturally gravitate towards!

You have put your finger on the core of the issue there, Andrew. The problem is about rhythm, not notes. Pitch detection has come so far now that, given a decent recording and the haptic memory you mention, even someone with a Woolworths ear like mine can recover all the pitches. Rhythm is a much deeper property of music than either notes or written notation. I could very easily sit down and play something, and no doubt you could too, which is at once coherent, meaningful and replicable but for which an easily constructed written representation is difficult or does not exist at all.

A big part of success in writing improvisation therefore depends whether the player has become accustomed to actually thinking in notated rhythms to start with in the creative process. I recall my teacher spending a whole hour's lesson getting me to feel a rhythm in one of his written compositions. I protested I was following the music and he replied that the rhythm he wanted to hear could be written in music only as a crude approximation. I did not realise the far reaching significance of this episode until many years later.   

[robertus]

Improvising and 'written' composing are different disciplines.

Improvisations, by their very nature, of ephemeral and 'of the moment'. This doesn't mean they don't have artistic value- but there a just once-off. They have a unique capacity to respond to the moment, but seldom stand up when notated.

And notating them down doesn't make any sense. For one thing- it's a huge amount of effort, and then going to be a huge amount of effort for someone else to learn. For a typical improvisation of a few minutes (such as

), it would take weeks to notate, and then take up so many pages, and then be almost impossible for anyone else to play.  Why bother doing this, when it would be infinitely easier just to toss off another improvisation?

The closest I come to notating improvisations is if I do an improvisation and record it, one or two good compositional ideas may pop up. But that is the exception rather than the rule.

[ronde_des_sylphes]

*ugh, duplicate post, deleted*

[ronde_des_sylphes]

Improvisations, by their very nature, of ephemeral and 'of the moment'. This doesn't mean they don't have artistic value- but there a just once-off. They have a unique capacity to respond to the moment, but seldom stand up when notated.

And notating them down doesn't make any sense. For one thing- it's a huge amount of effort...

it would take weeks to notate, and then take up so many pages, and then be almost impossible for anyone else to play.  Why bother doing this, when it would be infinitely easier just to toss off another improvisation?

I don't really agree tbh, other than the first part of this. Maybe I'm kidding myself, but I think the best of my improvisations stand on their own as compositions (I've actually performed in concert the full piece this particular extract is from, and I don't think that (other than entitling it 'Sonata improvvisata') anyone was any the wiser.) Writing them out is certainly onerous, but if I think the improvisation is either promising or close to a finished work as it stands, then I'm quite prepared to write them out and if necessary do some editorial work to give the final bit of polish. Putting it another way, if I feel a lot of ideas have coalesced in a way I find intrinsically artistically satisfactory I'd rather write it out and consider whether or not I want to make cosmetic changes rather than improvise afresh and hope for something better. I'd also state that I'm quite certain a lot of Liszt's operatic fantasies evolved from improvisations, and indeed that he wasn't averse to improvising round the framework of finished compositions.

My (imo) best improvisation
https://www.pianostreet.com/smf/index.php?topic=69430.0
seems a fair compromise between fully free and the structural tautness of a thought-through over time composition; motifs and progressions come and go within the texture and are presented in varying forms.

[ted]

I suppose it all depends on what we as individuals consider ephemeral. I have difficulty in describing recordings of improvisations by Jelly Roll Morton, Keith Jarrett, Cecil Taylor, Marilyn Crispell, Chick Corea and many others as musically or artistically transient, as even in the purely historical, social or statistical sense this is contrary to the observable facts. The recordings, the sounds, are themselves perfectly viable artistic end products without need of visual representation. So in that sense I disagree but I do agree that the functional processes of creating written music and improvised music, while overlapping in some areas, are essentially disparate. The pragmatic difficulty of converting one to the other is likely not germane to the much deeper issue of the lasting value of either.

[robertus]

I suppose it all depends on what we as individuals consider ephemeral. I have difficulty in describing recordings of improvisations by Jelly Roll Morton, Keith Jarrett, Cecil Taylor, Marilyn Crispell, Chick Corea and many others as musically or artistically transient, as even in the purely historical, social or statistical sense this is contrary to the observable facts. The recordings, the sounds, are themselves perfectly viable artistic end products without need of visual representation. So in that sense I disagree but I do agree that the functional processes of creating written music and improvised music, while overlapping in some areas, are essentially disparate. The pragmatic difficulty of converting one to the other is likely not germane to the much deeper issue of the lasting value of either.

Yes, I agree with this. Maybe my choice of word "ephemeral" was wrong. But the whole point of an improvisation is that it is spontaneous.

I'm pretty sure if we heard someone like Beethoven or Liszt improvise a completely free fantasia live, it would be absolutely amazing- probably more moving and impressive than any of their notated works. But this doesn't mean that if their improvisation was perfectly notated, and then played perfectly, it would be a masterpiece. This is because improvisations respond to the audience, the situation, the atmosphere.

I almost always include an improvised treatment of something in my recitals- and I feel that I am responding to the audience.