1970's Quebec, CA and other western countries.
Type of story
Theatre, mostly comedy.
Improvisation is defined as any performance in which no story is predefined before the performance. The story events are decided “on the fly”, by the actors. However, improvisation is not a totally free experience: constraints are usually defined beforehand, both by the general genre of improvisation and the specific performance that is prepared. Regarding the latter, it can be decided, for example, that the performance will last 4 minutes, involving two actors, and that one actor will talk, while the other not. For a long time, Improvisation theatre has been widely used in theatre as a training component for actors. Leaving aside the text, actors perform on a stage under various constraints in order to practice gestures, voice, staging, emotions, etc. Another common practice also consists in composing a dramatic piece from an initial improvisation that is later refined to obtain the final piece, to be displayed, without improvisation, to the public. Although improvisation forms can be found in history, it is only recently that improvisation theatre has become a more widely disseminated theatrical genre. General public now go to the theatre to see actors improvising on the stage. The audience, as it will be detailed below, plays a role in the improvised performance. Back to the 1970s, a specific improvisation genre has been created in Quebec area, Canada, which was a mixture of improvisation theatre and hockey sport game (Dramaction, 2009). This genre is now widely spread, and we will shortly review the main principles of this kind of improvisation that has a great influence on all recent forms of improvisation. Two teams, each composed of 7 actors (including one coach) compete for victory of a match. The stage was originally composed of an “ice ring”, a circular space delimited by a one-meter-high wall, similar to the hockey ring but much smaller. A match is divided into small improvisation performances. For each performance, a topic would be chosen randomly within a set of topics written by the public before the representation. Along with the topic is given:
- a duration
- the indication of the number of players allowed to play
- a category
- and the information whether each “team” shall improvise with the same constraints one after the other (“compared” mode), or together (mixed mode).
Then the two teams have 30 second to prepare the play. The match is governed by a referee, who plays a central role during the performance. Whenever a fault is detected, she/he informs players (and audience) about it. When a team gets three faults, it gives one point to the other team. There are many faults that the referee can detect. Some of them are simply related to the respect of the competition, such as the fact that the preparation should not be extended beyond the allowed 30 seconds. Other rules are designed for the sake of performance quality. For example, to impose a character to another player (“Hi father! How was your day at the factory?”) is forbidden and is punished. Other faults occurrs when actors do not listen to each other carefully enough, create inconsistency, for failed to make any progress in the story, etc.
When time is up (the referee uses a gestural code to inform players about the remaining time), the audience votes for one team to the other. The winning team gets one point. (See (LNI, 2009) for the complete set of rules.)
Improvisation theatre mostly contains humour and laugh, though serious and dramatic performance are not forbidden.
Match-based improvisation, becoming more and more popular, is now completed with less competitive genres and winning a match is not the goal. Each theatre company can creates its own genre, or “concept”, and test it on the public. A popular British TV show named “Whose Line Is It Anyway?“ has proposed its own form, with four actors improvising on TV based on various games. Note that a specific genre is called the “long form”, in which a full theatrical piece is improvised to the audience (actors might have prepared a coarse structure a few hours before). Except for their differences, these types of improvisation borrow from the original Quebecois genre (categories, time limits, rules, audience-chosen topics, etc.).
Relation with Interactive Storytelling
Improvisation theatre is of interest for IDS because it gathers two relevant features:
- The story is created dynamically during performance, rather than carefully scripted before the performance.
- It is interactive. Interaction occurs at two levels. First, the audience chooses the topic and reacts to it during performance. Second, the actors interact with each other to shape the story.
There are several references in the field of IDS to Improvisation (Hayes-Roth, 1994; Aylett and Louchart, 2003; Szilas, 2005; Swartjes et al., 2008; Swartjes and Vromen, 2007); Tanenbaum and Tanenbaum, 2008; Magerko, et al. 2009). However, there is no obvious mapping between Improvisation and IDS. Is Improvisation a model for emergent narrative? Should we take the coach as a model of a drama manager? should the IDS active user be matched with the audience or the improvisation actor? The difficulties to answer these questions explains why Improvisation Theatre has not served as a detailed model for IDS so far. Nevertheless, several strategies used by improvisational actors, as well as some of the constitutional rules of the improv games may serve as a guideline to design components of IDS.
Here follows a list of such rules:
- Listening. Actors must listen to the other actors to incorporate their propositions into their own story. This suggest an adaptive approach of storytelling.
- Yes rule. Actors must never refuse what is proposed by another character. This is a stronger and more specific formulation of the listening rule, as it ensures that the story is build collectively.
- Change location. When a story is stagnating, it is a good idea to change the location. Provided that a story in IDS contains several location and that the location enables new actions, this strategy could be used effectively in IDS.
- Calls for help. There is a code in improvisation to call for other actors in the stage. Typically, by mentioning, with a slightly insisting tone, a character who is not on stage, an actor is called to enter the stage and immediately play this character. While the code for calling for help is not necessary in IDS, adding a character in a stage when the story is not moving forward is a implementable strategy.
- Avoid too many people on stage. It is better to limit the number of significant characters on stage to around 4. By “significant”, it means that actors playing a tree or a immobile guard do not count as “significant”.
- Ellipsis. To move forward the story, a technique of entering the stage consists in announcing, as a narrator, a time ellipsis. Typically, an actor enters the stage and says “three years later”. Actors on stage adapt immediately and create a new scene following this constraint.
- Contrast (Bernardi, 1992). Showing contrast on the stage is advisable. If the first actor entering the stage plays a happy character, it is a good idea for the second character to enter crying, because he/she creates immediately a conflicting situation and raises questions in the audience's mind: why is this character so sad while the other so happy? Interestingly, the second character does not need to know why he is crying, or at least, not at the moment he enters the stage.
There are many other such rules, and they depends on improvisational players approach and training.
Rather than relying of this kind of prescriptive rules, Brian Magerko and his colleagues have investigated experimentally the cognitive processes that occur when actors improvise (Magerko, et al. 2009). They found some general mechanisms such as inferring from what is said, building a game within the scene, or ensuring the reality of the scene. Some of them naturally overlap with the above mentioned prescriptive rules. This research opens the way to a computational simulation of improvisational actors for generating stories according to the user input. However, how this simulation can interact meaningfully with a player is still an open question.
Systems/Tools using this theory
- Research from Sandy Louchart and colleagues on emergent narrative has looked at improvisation theatre as a model for emergent narrative. The concept of distributed drama manager is inspired by how the story unfold in improvisation theatre, although specific rules of improvisation theatre are not implemented in Fearnot.
- The concept of late commitment (Swartjes, et al., 2008), which consists of deciding character's attribute or props according to the current situation has been implemented in the IDS system called Virtual Storyteller.
- Szilas (2005) has proposed a general architecture for IDS in which distributed agents and a drama manager are organized in a non hierarchical manner. This architecture was directly inspired by Improvisation Theatre.
This idea that agent's parameters and beliefs should be changed according to a narrative constraint is implemented in the fitting algorithms used in Thespian (Si, et al., 2009).
- Brian Magerko and colleagues have implemented Improvisation Theatre rules into a micro-agent (Baumer and Magerko, 2010).
- Aylett, R., and Louchart, S. (2003). Towards A Narrative Theory of Virtual Reality. Virtual Reality 7: 2-9.
- Baumer, A. and Magerko, B. An Analysis of Narrative Moves in Improvisational Theatre. In the Proceedings of the Third International Conference on Interactive Digital Storytelling, Edinburgh, UK. (pdf)
- Bernardi, P. (1992). Improvisation starters – A collection of 900 Improvisation Situations for the Theater. Betterway Books, Cincinnati, Ohio, 1992.
- Dramaction (2009). Dramaction. http://www.dramaction.qc.ca/fr/improvisation/histoire-de-la-lni/
- Hayes-Roth, B., Sincoff, E., Brownston, L., Huard, R., and Lent, B. (1994). Directed improvisation. Stanford University Report KSL-94-61.
- Johnstone, K. (1992). Impro: Improvisation and the Theatre. New York, New York: Routledge / Theatre Arts Books.
- LNI, 2009. http://www.lni.ca/fr/reglements-officiels.php3
- Magerko, B., Manzoul, W., Riedl, M., Baumer, A., Fuller, D., Luther, K., and Pearce, C. (2009). An Empirical Study of Cognition and Theatrical Improvisation. In the Proceedings of ACM Conference on Creativity and Cognition, Berkely, CA.
- Si, M., Marsella, S. C., Pynadath, D. V. (2009). Directorial Control in a Decision-Theoretic Framework for Interactive Narrative, Proceedings of the 2nd Joint International Conference on Interactive Digital Storytelling: Interactive Storytelling, December 09-11, 2009, Guimarães, Portugal
- Swartjes, I., Kruizinga, E., & Theune, M. (2008). Let's Pretend I Had a Sword: Late Commitment in Emergent Narrative . In U. Spierling & N. Szilas (Eds.), Interactive Storytelling (Vol. 5334, pp. 264–267). Berlin / Heidelberg: Springer.
- Swartjes, I. and Vromen, J. (2007). Emergent Story Generation: Lessons from Improvisational Theater. In: AAAI Fall Symposium on Intelligent Narrative Technologies, 9-11 Nov 2007, Arlington, Virginia.
- Szilas, N. (2005). The Future of Interactive Drama. Australasian Conference on Interactive Entertainment (IE'05), Sydney, Nov. 2005.
- Tanenbaum, J., Tanenbaum, K. (2008). Improvisation and Performance as Models for Interacting with Stories. In: Proceedings of the 1st Joint International Conference on Interactive Digital Storytelling: Interactive Storytelling, Erfurt, Germany, pp. 250-263, 2008.