Who’s Driving This Engine?

This essay is an ABC New Media commissioned review of an interactive online comic by Simon Norton called ‘Testimony: A Story Machine’. Norton’s work won the Best Cultural Website 2004 AIMIA Multimedia Awards and is featured in ‘Strange Attractors: An Animation Showcase’, ABC New Media – which the following article is part of. It is still online at Strange Attractors and so you can read it there too.

WARNING: THIS ARTICLE MAY CONTAIN SPOILERS – it depends, however, on whether you are a “reader” or a “user”.

Simon Norton has created a “story machine”. Imagine a Ren and Stimpy-style animation of a monstrous mechanical device bulging, smoking and bleeding, churning out gold embossed books to naive curly haired children. Norton’s product is not this – it is aesthetically pleasing and not as prescriptive. Instead, Norton’s machine is intended to be a tripartite partnership between the author, computer and reader to produce an indefinite number of unique stories. After producing Testimony: A Story Machine, Norton described their respective functions as follows: the author creates the ingredients and sets limits, the computer determines sequences (randomly or following a program) and the reader interprets, provides meaning. Who is driving this story machine then? And how do we feel about sharing the wheel?

When an author produces a work they usually consider the elements of narrative: story and plot, character, setting, voice and so on. Fundamentally, story is what happens and plot is the order in which it is told. E.M Forster’s often cited illustration of this distinction is: “The king died and then the queen died” as story, whereas “The king died, and then the queen died of grief” as plot (Forster, E.M. (1941) Aspects of the Novel, London: Edward Arnold, originally published 1927). It is the author’s responsibility to ensure the events are linked causally, and the readers’ to interpret the links and associations and endow meaning. This engineering of content is controlled in a fixed media like literature, film or television where, once produced, the order of events is permanent. Digital media, however, permits interactivity, placing decisions about what occurs when in the hands of the user. It is in the new media context that narrative, specifically plot, is explored, experimented with and sometimes evolved; and where interactivity has become a much debated and pivotal element.

Norton did not begin Testimony with a linear story or a plot. Inspired by the collage writing technique developed by novelists William S. Burroughs and Brion Gysin in the 1960s, Testimony emerged from cutting-up newspaper articles, literature and his own writing. He then drew a few sketches, cut them up into squares, laid them out on a table and checked whether they could make sense in any order. They did, and since he had only precursive control over the readers’ associations he looked forward to their unpredictable inventive links.

Norton employs the principles of visual communication found in film and comics. Comic strips are traditionally associated with print media and have a particular mode of communicating a story. Essentially, each panel displays a complex set of ideas that are usually a leap in time from the previous panel (leaps can also be made within a panel). Comics are, therefore, known as sequential art. It is the reader’s task to interpret the iconography and fill-in the gaps. Scott McCloud, known as the “Aristotle of comics”, writes that the “natural world creates great beauty every day, yet the only rules of composition it follows are those of function and chance. Comics, at its best, should do no less” (McCloud, S. (1993) Understanding Comics: The Invisible Art, MA: Kitchen Sink Press).

Norton developed McCloud’s “function” and “chance” and created Testimony in Flash using Actionscripts to produce random panel sequences. Each sequence pulls from a different sized library, and thus the permutations vary and are hefty. Unlike a book where the amount of pages are easily viewed and numbered, and movies with mostly consistent running times, an online work’s boundaries are hard to ascertain. Norton cannot even predict what image will emerge where and does not know how many possible sequences can be displayed. Each session, therefore, is unique for the author and user. The calculations attest to the capricious production: the first sequence has three story panels that are selected from a library (the smallest) comprising forty images, this results in sixty-four thousand possible displays; further into the work the range varies between one and a half million to thirty-seven trillion (depending on how many panels with text balloons are seen). To achieve this story machine of function and chance Norton has employed an interdisciplinary array techniques – that of a writer, comic, animator and programmer.

Although the images, sounds, text and mode of interaction are pre-made and programmed by the author; and although the user has the ability to decide when they occur, they do not have control over the why. Why is this image followed by that? Why is this quote with this character? These decisions are made, or rather, these commands are executed, by the computer program. The program randomly selects images and text from a library – a simple Actionscript that is not exercising any attempt at intelligence. The presumption that some users have made though, Norton reports, is that the images are preselected by the author. The computer is therefore not perceived as a synthetic narrator but a medium between the author and user.

In Norton’s words, “Testimony is an interactive and animated comic strip.” “Interactive”” can refer to two elements – one is how the user interacts with the computer, and two, how or if the user interacts with the content (story, narrative) delivered by the computer. The human-computer interaction occurs in Testimony when the user inputs through a mouse-click to affect the transition from one screen sequence to another, to have images emerge under the mouse icon in the dream balloon, to direct the replacement of a panel and finally, to exit the work and email the creator at any time (the last two are of course outside the story world). However, the panels being affected by the user are images of Norton’s murder mystery. Settings, characters, events, quotes and props populate the panels. Therefore, when selecting a panel and choosing to progress to another screen sequence (presumably to satisfy a curiosity, fulfil a desire or confirm a hunch) the user could be seen to be interacting with the content, the plot. But is this possible?

As a user/reader, it was apparent to me from the beginning that the panels were randomly produced on screen but this did not affect my immersion in the story. I enthusiastically noted every panel, wrote notes on my assumptions about who the killer is and their motive, and kept a mental log of my decisions and expectations. I became disheartened once I realised the text quotes within the balloons were also randomly produced (meaning an accusation or admission could be said by many characters) and the panels depicted different characters doing the same action, for example various characters may be seen falling or jumping off the building. It is impossible therefore to assign a role – murderer, victim, accomplice, helper – to any of the characters because the narrator is now intentionally unreliable. This authorial anarchy has two effects: the author is not narrating – not storytelling – and the user is dissuaded from being a reader in the literary sense of making inferences.

Espen Aarseth observes that the difference between fixed and interactive media is found in the relative activity of the reader and user. For Aarseth, a reader makes decisions internally whereas the user actuates them through their interaction with the work (Aarseth, E. (1997) Cybertext: Perspectives on Ergodic Literature, Maryland: John Hopkins University Press). Although I am a user I am nevertheless unable to affect (through my actions) and solve (conceptually) any mystery. Instead the author wants me to consider the interface, the idea of a story engine and the unreliable nature of testimony. For Norton “the whole project is about process rather than result”. The process is that of the users. We are a large part of the story machine, more so than usual. We are being encouraged to play with our process of association, the construction of meaning and to generate our own denouement – if at all.

“I don’t like to be led by the hand by a filmmaker or a writer … It’s seems sort of natural to have that kind of hands-off approach.” If you are the type of reader Norton is then you will revel in taking the wheel and clinging on around the absurdist corners Testimony takes you, if you are a user of interactive stories, who still wants to be guided, then you will find yourself caught in the cogs. McCloud observes that “viewer participation is on the verge of becoming an enormous issue in other media. How a comic addresses this issue – or fails to – could play a crucial part in defining the role of comics in the new century.” Testimony is playing a crucial role for more than comics – it enacts the much theorised non-linear narrative structure and forces the user to participate in ways that have not been attempted in many new media products, whilst also playing with murder, cartoons and frypans along the way.

Check out Simon Norton’s work and the other great works and reviews at the Strange Attractors website.

Dena, C. (2003) ‘ Who’s Driving This Engine? ‘ Strange Attractors: An Animation Showcase , ABC New Media [Online] Available at: http://www.abc.net.au/arts/strange/articles/engine.htm