DIY TO NONLINEAR CHOREOGRAPHY

What follows is an informal assessment of a five days research project made in collaboration with the composer Ged Barry and the musician Chieko Mori.
Its purpose is to give a simple example of possible non-linear choreographic modelling.

I have been working on concepts related to non-linear choreography for several years, particularly with the company Barriedale Operahouse (which I co-direct along with Michael Klien and Nick Mortimore) but also in collaborations with various artists or specific venues (such as the I.C.A.).
The work has raised a widespread interest, recently bringing the company to obtain a commission for the lyric opera house in Vienna (Volksoper, December. 2001) and one for Ballett Frankfurt (March 2002). In the wave of this excitement I felt encouraged to push forward the research about non-linear composition on a personal level, trying to approach its issues from different angles.

Non-linearity has been explored quite extensively in music composition and visual arts but very little (if anything at all) in choreography. Certainly there are examples of works presenting aspects that could be described as such; yet, in a certain way, the loss of linearity has mostly appeared as a by-product of other interests rather than an intentional purpose.
Moreover, when talking about 'non-linear' in choreography - but particularly in music composition and visual arts - it comes to the attention that we usually refer to works relaying on one common denominator: the use of sophisticated technologies: digital interfaces, feedback systems, cameras, monitors, sensors... these have always represented the filters for information processing and codes generation, in other words the means for breaking down linearity.

The focus of this project was centred on exploring the same concepts, but stripping them down of their "high-tech" connotations. A self-contained non-linearity in which all permutations generated within the choreography itself or the minds of the performers. Even the interactions with other mediums (such as music) was kept to a basic sensorial level, and the general regulations were based on a more subjective exchange of physical information rather than through the auxiliary of digital algorithms.
I have to admit that such work has proven itself to be extremely challenging: working with basic concepts is in a way tricky due to the thin line between simplicity and obviousness, moreover it is not possible to clearly refer to a previous choreographic tradition. For this reason most of the terminology and conceptualisation had to be subjectively drawn from different sources - normally ranging from complexity science and information theories to artificial intelligence, media studies and sociology... plus a few others.

In the following script, I decided to avoid what could be a too technical, and therefore confusing, terminology. Instead I tried to use the widely known terms relative to the HyperText Mark-up Language, in a way to the basic structure of the World Wide Web. An unusual language for my work, but I hoped that it would help to create a simple associative line for equally non-linear concepts. Other terms, such as "identities", are purely subjective and I will be glad to give further clarification at your request. Some key words have been emphasised in italic.

To conclude I would like to mention that the results of the work done until now have been extremely encouraging, raising a genuine interest in the people that have assisted to demonstrations and shows or heard about it through lectures and word of mouth.
I have received many enthusiastic email and requests of more in-depth clarification about the basic concepts of non-linear choreography. This has encouraged me to write this assessment (which I am using as a very informal introduction to the work) and to further develop my research.


PHASE 1 - DEFINE IDENTITIES

With the term 'identity' I define all that influences AND is influenced by the non-linear process.
Moreover, another fundamental requirement for the definition of an identity is the need of conceiving it as a recognisable entity, which means with a characteristic way of behaving and interacting with the changes of the performing context.
With this I stress that an identity is more than a simple interactive element; it is an individual bringing with it its own personality, "spirit" and also some specific behavioural codes. The identity has to carry a "soul" within, even when it is not represented by a human being but some sort of other medium - they can, in fact, be identified by all sort of elements: lights, pre-recorded or live music, video, even space and time themselves (NOTE 1).
It is better to clarify a possible misunderstanding: identities are not necessarily physical bodies, in other words the number of identities present in a given 'scenic space' it is not uniquely linked to the number of human beings involved in the 'performing context'.
On a practical level this is far less mysterious than it might seem: it's simply a matter of deciding how many performers and which other elements (music, lights, etc.) we want to be taking part in the dynamic exchange of information, that is to say the non-linear structure.
Non-linear composition is always a dynamic process dealing with variables and exchange of information between identities. It is not possible to create a non-linear process without some sort of interactive or variable elements. Another important aspect is that the process has to be regulated by a'meta-structure' limiting the possible outcomes. Metaphorically it is the same as creating a common language for each identity. How could we understand each other if we were not to speak the same language?
Identities are like the "users" on the WWW, interactive users in this case, that can change the nature of the site they are browsing.
Finally, I do not call identities elements that are part of the performing context (and even mutate throughout the representation) but they do so with no reference to other elements/identities, in other words they do not take part in the construction of the non-linear process. I prefer to call these elements 'presences'.

---------

NOTE 1: In this case, when mentioning space and time, I am obviously referring to what is normally intended as the relative space and time of a performance, where a certain scenic event takes place. This is different from the absolute time and space (representing the physical reality) which would be pretty useless to define as an identity, considering it would be far beyond our means to cause any kind of modification to them. The relative time and space are subjective visions of their absolute nature: for example the total surface available for the execution of a choreography, or the total length of each sequence. Here it becomes possible to reduce or expand the scenic space by re-establishing its parameters, and to shorten or extend the duration of a sequence by changing its "tempo".

PHASE 2 - DEFINE PRIMITIEVES (POSSIBILITIES)

Once defined the identities we need to define the 'primitives' for each of them. Primitives are elementary building blocks. They represent the arbitrarily chosen basic level of informational content, the bricks used to generate a choreographic system. This is what makes up the fixed information content used to build a non-linear performance. It can be represented from a single action to a long dance sequence (if for instance referring to dance) or from a single note to a complete music piece (if referring to music) and so on for any of the chosen mediums. Primitives are subjectively intended as linear elements, in other words they refer to the bricks only, not to the ways of arranging them.
In the WWW metaphor they would be the pages of a web site. Web pages can contain an endless range of different contents yet, when seen solely as locations in the 'cyber space', they are also independent from their content.
I sometimes refer to primitives also using the term 'possibilities'. In my opinion this is an appropriate way of representing them, in fact they are nothing more than the ensemble of the possible "states of being" for each identity, the possible movements and words, the possible positions in the space, etc. In other words, the " represented material" generated through the given restricted access (the created building block) to the endless number of possible permutations (all possible building blocks).

PHASE 3 - DEFINE A BASELINE FOR EACH IDENTITY

This is somehow an optional stage of the process, and a pretty straight forward too, as it is just a matter of choosing (or creating) a primitive as baseline for each identity (NOTE 2).
What is a baseline and why there is the need to identify one?
A baseline is a point of reference that remains present throughout the show, it is like the 'home page' of a web site. Ideally if no codes were activated or no links created, the baseline would be the only primitive to be performed. In any case the baseline is always the primitive with the lowest priority, this means that it is always overcome by the appearance of others primitives, but it will invariably re-assert itself as soon as the primitive with higher priority ends. Generally speaking it is always possible to create a hierarchy of priorities, in a way that primitives with higher priority will always superimpose the one of relative lower priority.
Let's go back to the original question: why to create a baseline in the first place? As I said, this is an optional stage in the construction of a non-linear choreography, anyway to define baselines and to establish priorities helps greatly in the managing of the overall flow and coherence of the non-linear process. I find especially important to maintain a sense that each identity is a coherent and well-defined entity. This is a basic requirement in order to avoid the sliding of non-linearity into chaos.

-----------------

NOTE 2: it is also possible to select more than one baseline even if this adds a slightly higher degree of complexity to the codification of transitions. I would anyway recommend not to go beyond the two baselines in order not to loose the very concept of baseline.

PHASE 4 - DEFINE HYPERFIELDS

An 'hyperfield' (HF) is the entire group of possibilities that each identity will be able to access. Each identity can only move within its own HF, even though this may contain others identities primitives (for a practical example refer to the "DIALOGUES" layout below). It is the all web site, the collection of all web pages - but this site could contain also the pages belonging to other sites.
This is a very important phase of the process on all levels: theoretical and practical. Here is where the major parameters of freedom/limits are created, here we define what the "reality" of each identity will contain and here we establish the primary connections between them. Moreover here is where we establish the reasons, the meaning and poetry of such exchanges.

------------------

NOTE 3: presences do not posses HFs. The collection of possibilities accessible by them are not defined by the term HF but by closed field, in this way we make a distinction between the different nature of the two.

PHASE 5 - DEFINE HYPERLINKS

The simplest way to explain the concept of 'hyperlink' (HL) probably is to refer directly to the World Wide Web (after all that's where the word itself comes from).
Each web site is constituted by a certain number of pages (as each HF is constituted by primitives) and each page is accessible through a series of hyperlinks (as it is each primitive) but, even though they are all part of the same web site, from any given page we will be able to access only a restricted number of alternative pages, depending on the hyperlinks branching out from it.
Same can be said for the non-linear choreographic process: some possibilities will be accessible only if departing from an hyperlinked starting point. There is no way to move directly between two primitives that are not hyperlinked. Ever.
This process is important for the general rhythm of the changes, this will guarantee to avoid, for example, "dead spots" or too long moments of the same dynamic, or similar problems relative the overall balance of a performance.

PHASE 6 - DEFINE PROCEDURES

With the HL we establish the paths, with the 'procedures' we define the rules on how and when to move along them; they regulates the access to HLs and subsequently to primitives.
For each identity procedures can be passive or active. They are active when activated by a conscious choice of the identity in question, passive if triggered by an external event or by another identity (for whom it will be active).
Procedures are arbitrarily chosen and are normally cued by different elements: from elements belonging to the scenic space such as a sound source, a position of a dancer in the space, a specific movement, a change of light, etc., to elements present in the 'public space' such as a train passing by, the variation of people in the audience, changes in the stock marked, even commands sent from a remote control (i.e. from the PC of a remote user connected to the Internet).
There are only two real requirements for the codes: they have to be clearly identifiable by the identities and they have to be dynamic, as an ever present code is obviously of scarce use. A third recommendable requirement is their practicality. For instance, procedures triggered by elements of the public space often require the use of expensive technology (sensors, computers, monitors, etc.) and a complex interfacing of the information through a specific filtering software. Moreover simple codes (but not necessarily obvious) have the advantage of requiring a minor amount of continuous attention to the possible appearance of each of them. In a complex environment it is in fact very tiring for the performer to keep focusing on a large number of intricate possible codes. This requires a certain amount of energy that often is lost at expences of the qualitative execution of the choreography itself.
This is a simple and practical observation learned by experimenting with non-linear choreographic systems but, to be honest, I admit that I find very intriguing to test the implications of working with different degrees of complexity.

PHASE 7- SIT BACK AND ENJOY THE RIDE


The following is a schematic rapresentation of the elements composing the global structure of "DIALOGUES".
It is divided in two parts:
The first part is a list of the various elements. At the appearance it might seem complex, but a pen and a white sheet of paper will soon prove the impression wrong. All it is needed is to draw a wide circle for each identity and, within them, some smaller circles representing their respective primitives. Choose the baselines, and number the other primitives. Next step is to connect the primitives according to the specifications in the "Define Hyperlinks" list. Finally the "Define Codes" list will tell you when something happens and how many times.
These bring us to the second part. Following the instructions, the result should be a connectionist model of the entire show similar to the one closing the list. The graphic solutions I have chosen are not necessarily the best one, different solutions are obviously possible. It's worth to give it a try.


PHASE 1 - DEFINE IDENTITIES

Identity 1: dancer - static
Identity 2: dancer - dynamic
Identity 3: musician (koto player)
Presence 1: electro-acoustic effects
Presence 2: lighting

PHASE 2 - DEFINE PRIMITIVES(P)

Identity 1:
· P 1: (slow circular walk on spot)
· P 2: (dance sequence in enclosed space)

Identity 2:
· P 1: (dance sequence with circular movements)
   * P 1a: (P 1 performed increasing travel in space) - OPEN SPACE
   * P 1b: (P 1 performed on spot) - CLOSED SPACE
· P 2: (dance sequence with balletic steps and jumps)
· P 3: (series of steps with sudden explosions)
· P 4: (dance sequence with sharp "staccato" movements)
· P 5: (slow circular walk on spot) - as in identity 1

Identity 3:
· P 1: (melodic music piece in 8/8)
· P 2: (rhythmical music piece in 7/8)
· P 3: (suspended notes)
· P 4: ("andante" music piece in 8/8)
· P 5: (dance sequence inspired by traditional Japanese dance)

Presence 1:
· P 1 (prerecorded music piece)
· P 2 (live electronic manipulation of sounds)

Presence 2:
· P 1 (general, fixed lighting)

PHASE 3 - DEFINE A BASELINE FOR EACH IDENTITY

Identity 1: P 2 chosen as baseline
Identity 2: P 1 chosen as baseline
Identity 3: P 1 chosen as baseline
Presence 1: no baseline
Presence 2: P 1 maintained as fixed baseline

PHASE 4 - DEFINE HYPERFIELDS (HF)

Identity 1 HF = P 1 + P 2 + identity 2's baseline.
Identity 2 HF = P 1 to P 5 + identity 1's baseline + identity 3's P 5
Identity 3 HF = P 1 to P 5
Presence 1 F = S 1+ S 2
Presence 2 F = S 1

PHASE 5 - DEFINE HYPERLINKS (HL: ===)
(Abbr.: identity 1's baseline = 1baseline, identity 1 P 1 = 1P 1, and so on.
===> indicates possible directions of interchange)

HL 1: 1baseline <===> 1P 2
HL 2: 1baseline <===> 2baseline
HL 3 & 4: 2baseline <===> 2P 1a & 2P 1b
HL 5 to 7: 2baseline <===> 2P 2 to 2P 4
HL 8: 2baseline ===> 2P 5
HL 9: 2P 5 ===> 2P 1a
HL 10: 3P 5 <=== 2baseline

PHASE 6 - DEFINE PROCEDURES (Active or passive)
(procedures are active or passive in relation to the different identities. They are active if arbitrarily triggered by a conscious decision of a specific identity, passive if triggered by an external source).

HL 1 to be activated three times (X3)
· 1baseline ===> 1P 1 // identity 1 active twice, passive once (triggered by identity 2's P 5)
· 1P 1 ===> 1baseline // identity 1 active

HL 2 to be activated twice (X2)
· 1baseline <===> 2baseline (1&2 swap places simultaneously all the time)
   * phase 1: identity 1 passive, identity 2 active (first change)
   * phase 2: identity 1 passive, identity 2 passive (return to original baselines, triggered by identity 1 ending her sequence twice)

HL 3:
· 2 baseline ===> 2P 1a // identity 2 active
· 2P1a ===> 2baseline // identity 2 passive (triggered by ending sequence once)

HL 4:
· 2 baseline ===> 2P 1b // identity 2 active (but space conditions had to be satisfied: possible only if passing on one of the five predetermined points)
· 2P 1a ===> 2baseline // identity 2 passive (triggered by ending sequence once)

HL 5 to be activated twice around beginning and ending of music (X2)
· 2baseline ===> 2P 2 // identity 2 passive (triggered by begging of identity 3's P 2)
· 2P 2 ===> 2baseline // identity 2 passive (triggered by ending of identity 3's P 2)

HL 6 to be activated once (X1)
· 2baseline ===> 2P 3 // identity 2 active
· 2P 3 ===> 2baseline // identity 2 passive (triggered by ending spatial pattern)

HL 7 to be activated three times, each time with different front (X3)
· 2baseline ===> 2P 4 // identity 2 active
· 2P 3 ===> 2baseline // identity 2 passive (triggered by ending choreographic sequence)

HL 8: (to be activated at least three time, each time of different length)
· 2baseline ===> 2P 5 // identity 2 active

HL 9:
· 2P 5 ===> 2P 1a // identity 2 passive (triggered by ending of 2P 5)

HL10 to be activated once (X1)
· 3P 5 <=== 2baseline // identity 2 passive, identity 3 passive (triggered by end of music)

Additional conditions for identity 2:
· baseline could be executed once or twice in a row (but no more than twice). Second time front had to be reversed.
· P 3 & HL 4 could be activated only at the end of baseline sequence.
· if P 2 & P 5 were activated in the middle of baseline when returning to baseline the sequence had to start where it had been left.
· baseline could never be performed in CLOSED SPACE / OPEN SPACE twice in a row.

PHASE 7- SIT BACK AND ENJOY THE RIDE

The above relations can be simplified in the following graph:




copyright © 2000 Davide Terlingo All Rights Reserved