On languages for describing suitable motions

Darren Kelly (kelly@x4u2.desy.de)
Tue, 14 Oct 1997 14:04:17 +0200 (MST)

Dear "Nik" (is that your first name ? ) and dance-tech,

Thankyou Nik for rightly picking me up on my logical sloppiness, which
crime is certainly greater than connecting a camera to an amplifier.
Yesterday seems to have been my day, another technical correction follows
this mail. I'd like to restrict my dance-tech mail to practical matters,
so I'd be glad to dicuss these tangerines one-to-one. Brief remarks
follow, relevant to dance, and some questions and a request for
advice to others.

On posing -> dancing
[A tangerine: skip to next underlined section for dance-tech pure]

On Mon, 13 Oct 1997, nik wrote: (in reply to Darren)
> >
> >There is, quite simply, no way to define the difference between
> >standing and dancing.

> Unless of course one views dancing as a proper subset of motion and standing
> as another proper subset; and therefore motion and dancing might overlap or
> be mutually exlcusive but not necessarily equal.

So you can tell me when I'm dancing and when I'm not ? This might explain
why obtaining funds for the neglected dance forms such as "sitting-on
a-bar-stool-dancing" or "while-driving-along-a-country-road-dancing" is
so difficult ! sm:)e

The problem is that you can't define your subsets easily. It's like the
old problem of telling the difference between blue, grey, and green eyes
(spectra). In everyday language we use these subsets, the boundaries are
however fuzzy. One of the interesting things one can emphasise with
performance is just how fuzzy those borders are. So a performer might
move smoothly from standing to dancing, the audience never being able
to tell when the performer is "dancing", when the performer is
"standing on one leg", or when the performer is "swaying". In a similar
manner a PLAY might emphasise the situation of a person who has lived
for long periods in two different countries/cultures and doesn't quite
know whether he, for example, is Australian or German. Nur ein Beispiel
natuerlich. Jawohl.

The (x,t)=(p,E) bit, by the way, is an allusion to the Heisenberg
principle, which emphasises fuzziness of measured quantities in
quantum mechanics.

[end of tangerine: Back to dance and tech...]

On notation for dance motions (with a request for advice)

> >However, physics and maths notation describe the most general
> >rules, and hence are incredibly compact without missing ANYTHING (as far
> >as one can measure).
> This is simply not true. All math that is based in formal logic is incomplete.
> That is, there are true things which aren't logical and logical things
> which aren't true.
I just new as I typed that sentence I'd get burnt. The mystery is why I
pushed the "send" key. You are right, my proviso "as far as one can
measure" is not enough, one must also question the logic system (as
examination of FOPA and Goedel's theorem tell us), and there are more
caveats, many which occur to me yet didn't mention. I see I'll have to be
much sharper. Don't worry, I'm not a fundamentalist physicist convinced
that maths is the tool to describe "emotion". I was just lazy.

Will you grant that the mathematical and physical description of motion
signals is the most useful available language for discussing what I call
"Motion-to-sense" algorithms ? This question is of immediate importance
to my work.

Maths is clearly NOT a good language for interacting with dancers, a
problem I'm trying to address. I am developing an ever increasing set of
"useful" performance motions which exploit the properties of acceleration,
but others trying out Drancing-A (without insight into the operation)
don't exploit these properties, and it's difficult to explain. The person
trying it out doesn't gain much from a demonstration of a single motion,
since one needs to understand a single "property" exploited by a class of
motions, i.e. a continuum of motions within a single "mode" not single
moves. Otherwise one is destined to repeat the same boring single motion
forever. Rather, one must understand the class of motions exploiting one
property of acceleration.

Sometimes when I use Drancing-A I have a visual picture of the motion I
wish to perform. I see the motion as a space-curve (like a 3d plot) of
exactly the type I'd use to explain the properties of acceleration to
physics students. I picture the Frenet frame motion vectors moving along
the curve. And I have a set of formulae in my head telling me in advance
what the signals will look like during the move, and with sufficient
dexterity they do. In the CMJ article I'm preparing I'll have example
plots of these motions (both plots from formulae and signals from

Sometimes I try to switch off these expectations and just go for it. The
results are very different and just as valuable (or musically pleasing).
The aim I guess is to train specific motions exploiting physics so that
they can then be exploited subconciously.

I just can't see how I can teach dancers these techniques without
standing at a blackboard with chalk drawing graphs. I enjoy doing
exactly that with physics students, and I've always received good
feedback about my teaching style at uni, but is this reasonable
with dancers not familiar with physics and maths ?

I welcome comments, especially from dancers and choreographers.