Creating a Story-Telling Universe

1 Introduction

One of the most int~resting forr~s _of sto_ry telling is extended story generation, the contmumg senal. Th1s can orcur as short fiction, movies such as Star Wars and The Empire Strikes Back, novel series, role-playing games and, in one of its most popular forms, the television soap opera. The writer or writers of such serials face the prodigious task of creating and keeping track of literally dozens of intricately interrelated characters and numerous past and present plot.s. Extended stories of this type may we'll be forerunners of the kinds of complex, interactive stories that will ultimately be created as computers are usC'd as more than simple word processors in toe creation of fic-tion.

In this paper · we will describe the first __phase of developnwnt. 01 a computer program, UNIVERSE. The goals of this program are twofold. Initially, we expect to dPvE'lop a program that can serve as a "writer's atd" for ex((·nded ~tory production. Ultimately, we hope to have l !0:1VERSE tell "soap opera" type stories. The as_pect of UNIVERSE to be di~cussed here is the creation of story­ telling universes IS'l~Us) comprised of charact~rs, their h1stones, famtly rclattons, and mterpersonal relattonships.

2 Motivation in Character Creation

Go<?d e_xtended ~tory telling is constrained by the need to mamtam cons1slency and coherence. A story is co11sislent if proP.ert1es and events of the story. world (mcludmg properties of characters) do not contam any Internal contradictions. Bitter enemies should not become close friends without explanation.

Coherence involves the idea that events should be logically derivable, at least in retrospect, from the information available to the reader. While it is neither neress:uy nor desirable to provide the reader with enough !nformation to actively predict every event that will occur Ill a story 1 the events should make sense when they do occur. fvwst notably, actions should be based on the personalities and backgrounds of the characters involved.

l TNf\l~HSE rreatrs a universe of characters before lw~in nin g to tell a story in order to maintain consistency ann coherence. It would be possible (and is sometimes

doctor warden klutz socialite movie-fiend professor flake swinger hari-krishna sports-fan store-owner nasty-person family-man bowler junkie

Figure 1: Some stereotypes used in UNIVERSE

Past event s are used primarily to provide variety and flavor and exQlain aspects of a charac ter that do not fit stereotypes. Hy including h_istorica! events. in an ST tJ, it is poss!l)le to have mterestmg vanat10ns m the vanous charac ters, along with adequate justification to assure coherence. (To achieve the coherence, it is, of course, necessary to refer to the various past events so that the reader is aware of them.) In the ultimate version of UNIVERSE, as the program tells stories it will naturally affect the history of the characters involved.

Fio-ure 2 a few of the_plots that can be run historically by U~TVEHSE during STC creation.

Figure 2: Sample historical plots

The fin al important class of information needed about the characters in an STU involves not individual characters, but interpersonal relationships between characters. Crucial to the selection of consistent plots is the need to monitor the way charac ters relate to each other. To achieve consistency of this sort , we must maintain information about relationships between, potentially , every pair of characters in our universe.

Figure 3 summari zes the information stored in a person frame. Trait s, goals and the role of marriages will be discussed further m later sections of this paper.

Name, Stereotypes, Trait modifications,

Individual goals, Interpersonal relations, Marriages, History (list of events)

Figure 3: Person frame informat ion

3.1 Stereotypical frames

UNIVEHSE stereotypes provide two basic ty_pes of information about the charac ters they describe -- descriptors of phys ical and J ersonality traits (which we will concPnt rate on here), an typical goals. Personality traits arc particularl v useful in selecting characters for a given plot (or plots tor a character) . In a blackmail plot , the bl a(" kmailer should not be a nice person, and the person blackmailed, or his family, should have something worth extorting.

Goa ls tend to be important in a somewhat different context. UNJVEilSE will make use of certain plot schemata, known as meta-P,lots, that are abstract in terms of the ev<'nts they descnbe. "Revenge" would be an example. To "execute" such plots, it is often necessary to have information about various characters' goals -- part icularly long-term goals -- in order to understand wha.t is important. to them.

The selection of tra its to be used in stereotypes is certa in lv not self-evident . We neerl. a selection of traits \ 1 ide enough to indicate which characters shou ld part icipate m which plots and how t h~y should fare in such plots. The traits chosen for UNNERSE are shown in Figure 4. No claim is made that this listing is exhaustive. It allows for a wide range of behaviors to be explained and can be easily extended.

Figure 4: Person stereotype traits

3.2 Interpersonal relationships

While it is important to have individual characters adequately represented 1 it is also necessary to represent interpersonal relationsnips (IPRs) between characters. We liave chosen to do this with a set of numerical scalest based on those from the psychological ex~eriments o [Wish et al. 76], extended for Al use in Schank and Abelson 77]. This work posited t ree scales -- positive-negative, intimatl~dis tant and dominant­ submissive -- to describe the way people relate to each other. To these we have added a fourth scale, att. rac t. t>dness, ap_p ropriate for soap opera universes. More eomplete sets of scales are being considered for later versions of UNNERSE. Ou r scales can include asymmet ric perceptions of each participant in an TPR.

As wi th individual charac ter information, values of scales can come either from stereotypes or past events. Past events seem to be more impor ant in shaping in terpersonal dynamics. While there are a reasonable quantity of stereotypical IPRs -- among family members and people in occupations, for example -- events such as shared crises, conflicts at work , marital conflicts, and the like, tend to be more significant in explaining how one person relates to another.

4 Building Up an STU

!laving specified the kind of information we would like to maintain about the charac ters in a story-telling universe. we are still left wi th the problem of how to generate t hat information. We could specify properties of charac ters and interpersonal relations randomly or only when needed for a plot (a necessary mechanism in any case). Howeverf to see how we might add coherence to our set o charac ters, consider the universe of characters in a typical television soap opera, Days of Our Lives. Figure 5 lists some of the 40 or so current ac tive characters in this NBC soap opera.

Tom and Alice -- Horton family P-atriarchs Mickey and Maggie -- son and wife Sarah -- their daughter (by artificial insemination) Stephano -- the baa guy Tony -- his son Renee -- Tony's half-sister Liz -- Ton(s estranged wife Don-- Liz sex-husband (sort of, she was actually still married to Tony then) Marlena -- another of Don's ex-wives Roman -- Marlena's fiancee

Anna-- Roman's first wife, presumed dead The· first thinf!: we see from Figure 5 is that the relations :1mong the enaracters are gutte complex (an informal count showed at least 20.J identifiable mterpersonal relationships), and that many of the characters are related to each other by past and present marriages of one sort or another. Past and present husband-wife and parent-child relationships add coherence to the character set in most soap operas.

Figure 5: A sample soap opera universe

We would like to have this same sort of coherent intricacy among the characters created by UNIVERSE. The most feasible way to achieve this is to, in effect, do a sim_Ql~ simulation of the past lives of the characters in the STU, creating spouses, children and other assorted characters as necessary. We view marriages as a motivation for characters rather than as arbitrary relationships that should be established after we have created a set of characters.

The . simulation used by UNIVERSE involves cyclical character creation. A queue is maintained of characters who have been created during the simulation, but have not had the details of their lives filled in. Each character is, in turn , removed from the queue, and UNIVERSE steps through his or her life, creating spouses and children. When the present is reached 1 further details of the character, such as occupation ana other descriptive stereotypes are filled in.

The selection of the specific events that occur can be done either by a user (in "writer's aid" mode) or somewhat arbitrarily . The latter is not as unreasonable as it may sound, as the idea is often to have manv interconnections among characters, regardless of what they might be.

5 Adding More Background

Although the cyclic character creation process described in the previous section provides the framework for the STUs created in UNIVERSE, it is not quite enough by itself. There are three main problems with the universes created: 1) Almost all the characters are related in one way or another; 2) Almost all the IPRs in the universe have to do with marriages and childrenj 3) There is no history, other than marriages and o1fspring, to the charactl'rs, which is detrimental to coherence.

Fortunately, we can solve all three of these problems with a sing!_~ mechanism, one related to the main thrust of the UNfVERSE program. The answer is to simply add history to the characters' backgrounds by_ "executif!g" in the past simplified versions of the plots UNIVERSE will create for the present.

It is crucial to note that the past plots run by UNIVERSE need not be executed in the same detail as during actual story telling. Instead, it is enough to merely determine how each plot affects the characters involved and the IPRs among them and modify their histories appropriately. UNiVERSE can add coherence to its current stones by referring to events in the past. In addition, the mechanism for selecting the plots to be run need not be the same as for real story telling. Insteadi'Pwe can select. a given plot in order to obtain a desired R, without requtring the same level of motivation as is need'ed in real story telling.

6 Conclusion

We have shown here a method for generating extended story-telling universes, consisting of characters, interpersonal relationships and historical events. The method of creating characters focuses on maintaining rons-i8lency and coherence. The methods described are fl exible enough that they can be extended to address additional problems, such as specific author intent, by adding constraints in the character generation process. The work described here provides the framework for the extension of UNIVERSE to actual story telling.

Acknowledgements

Special thanks go to Susan Rachel Burstein who helped develop many of the ideas described here. Comments by Kathy McKeown on an earlier draft of the paper also were extremely helpful. A more complete description of the work described here is available from the author.