So what is Transmedia? It is not watching a TV show on an iPad. That’s “watching TV.”
It is not telling many movies in the same universe on one screen. That’s “franchising,” like the Marvel Universe movies.
Ideally, transmedia is the telling of a story on multiple platforms. By this definition, the earliest example of transmedia storytelling is with the Bible: a book, music, paintings, stained-glass, sculpture, miracle plays, and meetups (designed to trigger every one of our five senses, including incense) all helping tell one grand narrative.
Examples:
- Falling Skies produced a prequel graphic novel which provides insights into the motives of characters in the “core text” – the TV series – while explaining backstory.
- ABC produced tons of world-building information all over the internet about the Dharma Initiative and the Hanso Foundation for Lost.
- I just wrote about the collection of websites about Pawnee, Indiana for Parks and Recreation.
I love the possibilities here. Not just for callbacks, tie-ins, or secondary narratives, but to actually enrich the telling of the primary story. To further the plot, to deepen the experience of the story.
Why should TV evolve in this direction? First, audiences want it. Fans are getting there before TV is willing to go there. A sign your show does will is when Reddit fills with fan theories, or AO3 fills with fanfiction, or DeviantArt fills with fanmade graphics. This is good behavior. Hollywood should embrace that people are curious about the worlds of their favorite shows.
Good Future TV Business
But mainly I think ancillary transmedia content is good for business. It’s good for business because it cuts down on a pirate’s ability to destroy your business model by cutting and pasting a single flat file. It’s good for business because it offers advertisers extra ways to reach wandering eyeballs in a second-touch and third-touch offering. For many shows, their online extensions can serve as access points to watch or listen to something live, which is great for business. Also, as evidenced by the booming video game sector, this kind of immersive experience is obviously addictive for an audience, it is going mainstream (especially among younger people) and all of that is good for future growth.
USC’s Richard Jenkins teaches this subject. He works with the film school, but approaches transmedia from an academic perspective. Jenkins breaks down most transmedia content into serving one or more of the following functions:
- Offers backstory
- Maps the World
- Offers us other character’s perspectives on the action
- Deepens audience engagement.
Jenkins is free to be hooked on his ideal of balanced transmedia, with no single medium or text serving a primary role over others – he’s an academic.
But the reality is that any content producer in a Hollywood guild needs to think about how many media are deployed and who works on what, and who is paid for what – the writing, directing, producing. Therefore, the model of unbalanced transmedia, with one clearly identifiable core text (eg. on TV) and a number of peripheral transmedia paratext extensions like podcasts, features, comic books, soundtracks, live theater, or social media narrative breadcrumbs (which are all integrated into the narrative whole) is better for television.
The reality is that the TV industry has incredible power right now – power to leverage – for the purpose of evolving TV storytelling into our multi-platform future. If they don’t, I fear they will lose primacy to video games.
We’ve seen American television already evolve from highly episodic structures (mostly self-contained) to much more heavily serialized structures. Most shows, though, combine elements of the episodic (a procedural plot which can be wrapped up in a single episode) plus the serial (an evolving character relationship, an unfolding mythology). With modern TV, the individual episodes work like chapters within a larger plot. Our shift towards seriality on American television plays a large role in preparing audiences for transmedia TV storytelling. We’re getting primed for it.
We’ve already seen some of this story expansion, exploration, and these early transmedia forays happen on fan-generated websites, while other experiments are pre-planned by the production companies or networks. They’re working. This is partly due to the nature of serialized television. It’s also partly due to the nature of our brains.
Time Seriality
The unfolding of a story over time is seriality. Time is an essential element of all storytelling, but it’s even more crucial for television. On TV, typically this happens through a process of “chunking” (creating bits of the story) and “dispersal” (delivering the story in installments). Central to that process is the creation of a story hook or cliffhanger which motivates the consumer to come back for more of the same story after the break. It can’t be overstated how important “set-up” and “pay-off” is to storytelling. The curiosity that comes between the two is a trigger for the prefrontal cortex. Our dopamine response to good payoffs is for TV what nicotine is for Philip Morris.
Chemistry aside, time is of the essence. There’s three different temporal streams within all narratives:
1. There’s narration time, the temporal framework involved in telling and receiving the story. For literature, this is variable because everyone reads at a different pace. The business of television was most profitable when narration time and schedule were strictly controlled.
The business side of TV believes it needs to focus on narrative time in serial storytelling because revenue is dependent on the structural gaps between episodic entries – those commercial break gaps between individual Acts. But as narration time becomes far more mutable and variable for viewers, the industry loses control of the original broadcast form (and Nielson loses the ability to count viewers) we lose the control over the gaps.
It is these temporal gaps that define the serial experience. Additionally and importantly, these gaps allow viewers to continue their engagement with a series in-between episodes, participating in fan communities, reading criticism, consuming paratexts, and theorizing about future installments – all vibrant practices that increase engagement, prevent piracy, and provide more advertisement opportunities.
2. The creative side of the business is already enamored with exploring the second temporal narrative stream: Discourse time. Discourse time is the time structure as told within a given narrative, or mimesis. There’s always been a great deal of experimentation with this aspect – influenced by technology, video games, and more. Such narratives often reorder events through flashbacks, retelling past events, repeating story events from multiple perspectives, and jumbling chronologies—these tricks are all manipulations of discourse time. Classically, mysteries fiddle with discourse time in order to engineer suspense, but increasingly, many complex narratives or mind-game movies, play with layering and chronology to engage viewers. Mulholland Drive, Memento, Pulp Fiction and, the video game culture-influenced Inception, or Source Code are examples. The building of tension, the raising of stakes, and achieving payoff, increasingly hinge on writers fiddling with discourse time.
3. Story time is the timeframe of the diegesis, or how time passes within the storyworld. This typically follows straightforward chronology and linear progression from moment to moment, with exceptions for movies which are about time-travelers.
Throughout its history, the TV industry depended financially on controlling narration time. They regimented story “dispersal,” exclusively on a TV screen, at specific days and times. But now that power is waning.
I believe Hollywood can make up for the loss of control and power by taking the reigns of mastering discourse time and multiple screen experiences. People will always depend on payoffs to get the rush of a story well told. As discourse time experimentation, and increasingly transmedia forays become more dependable for story payoffs, eventually the two will converge.
Then people like me will say I told you so.
2 thoughts on “Transmedia, TV Time and the Brain”