(Part 1 of 2) Back when I was a teenager, I used to play Dungeons & Dragons with a group of friends. D&D, for those who have never played it, is essentially a pen-and-paper version of World of Warcraft. Instead of a computer running the game, that role in D&D was served by a person—the “Dungeon Master,” or DM—whose job it was to map out a location in advance and fill it with monsters, traps, and other characters and events, and keep that material hidden from the players until they reached certain points in the game. Whereas computer games display information visually to the players as they wander through the game world, in D&D the DM simply describes what happens, according to his or her pre-established plan and a set of fixed rules governing things like combat.

One of my friends was a particularly good DM. He would create settings full of portentous omens, intrigue, and mysterious characters. Seemingly random encounters would lead to unexpected coincidences. The world he created seemed richly populated with deeper meaning, and my friends and I loved exploring it.

But we ultimately discovered that there was no deeper meaning; there was no there there. He was making it all up on the fly, and when we started scratching beyond the surface, improbable barriers such as invulnerable gods and invincible monsters began to block our path. The narrative structure of the world collapsed under its own weight.

I’m reminded of all that by the final season of Lost. Lost, I think, illustrates a trap that the creators of D&D dungeons and network television series alike are apt to fall prey to: it is much, much more tempting to build suspense than to resolve it.

I should make clear that I’m a fan of the show. And it’s final season is a far sight better than the final seasons of other series that I’ve loved, such as X-Files or Star Trek: The Next Generation or heck, even the Simpsons (which I no longer watch). I may even (I hope!) wind up eating these words after Sunday’s Lost finale. (For other pre-finale commentary, see the New York Times, the Washington Post, NPR, and my friend Jon Kay in the National Post.)

But at this point, it appears that the creators of Lost, early on in the series, fell into the trap of building suspense by piling on mysteries. The attraction of mystery is simple enough to understand: a mystery is an extremely economical means of telling a story. That’s because a mystery arises in the absence of detail. It’s a kind of end-to-end storytelling: much of the work of constructing the plot is actually done by the viewer. Given a few intriguing clues, viewers start to construct a set of consistent explanations, and try to figure out which explanation best fits what’s going on. Just like a jigsaw puzzle, as more pieces of information are added, the overall structure of the narrative should become clear.

But there are three major downsides to this form of storytelling. The first is that, whatever the ultimate resolution is, it can’t possibly compete with the gestalt sense of wonder created by the simultaneous possibility of all of the potential resolutions combined. That is, if mystery 1 has possible resolutions A, B, C, and D, a piece of information that eliminates A and B as possibilities lessens the mystery by 50%. An obvious way around that problem is simply to avoid divulging such information. But you can’t just have a plot that goes nowhere; something has to happen. If you can’t divulge much information, and you can’t do nothing, then one way to maintain the suspense is to add more mystery.

That leads to a second downside, which is that it is close to impossible to time the suspense/resolution balance when the plot is of uncertain duration. Single-episode formats don’t face this problem. Suspense created at the beginning of a half-hour Twilight Zone episode has to be resolved in 25 minutes; that certainty imposes certain requirements on the plot. The mystery can’t be too involved, or there won’t be time to wrap it up. And any pieces of critical information have to be divulged within 25 minutes, with the kicker (the bit that the Twilight Zone was famous for) coming right at the 25-minute mark.

But a multiple-episode series with no fixed end date doesn’t have the luxury of knowing when the important bits of information need to be released. Do it too soon, and the suspense will evaporate before the series is over. Piling on additional mysteries is a good short-term fix for this. But it leads to a long-term problem, which is that the many layers of mysteries take on a byzantine complexity that becomes impossible to resolve. That’s what happened to the X-Files; it’s no accident that the most memorable X-Files episodes are the self-contained stand-alone monster stories: The Pusher, Fluke Man, Victor Toombs.

There is a third downside, the one my friend ran into, and the one I believe the creators of Lost have encountered. As I noted above, it’s very difficult to resolve a mystery in a way that lives up to the many potential resolutions it has before it’s explained. A mystery with several potential resolutions is a bit like an electron “cloud” from particle physics—it’s almost its own entity, with different properties and different behaviors than the actual measured electron, which after being observed behaves like a boring old (if infinitesimally tiny) particle. Even stand-alone stories encounter this difficulty: the explanation sometimes does not fully justify the suspense. And there’s nothing special about television, either: in the Sherlock Holmes story (one of my favorites), why in the world does Julia Stoner describe what she saw as a “speckled band“?

A series of uncertain duration and uncertain success presents an opportunity—or perhaps a temptation—to short-circuit this problem. Where the mystery does not have to be resolved by the end of the hour, or the book, or the film, the writer can introduce mysteries for which he or she simply has no ending in mind. That is, she can leave all or most possibilities open long after the mystery is introduced. But there is a danger with that technique. It risks destroying the illusion of the separateness of the fictional universe and the real universe, the one the writer is in. If the most plausible explanation for most mysteries is, “Well, the writers didn’t really know how that would turn out,” that collapses the two worlds together. And when the mysteries become so convoluted that there is no hope of resolving even very significant plot points, universe collapse is almost inevitable.

From their answers to interview questions, it seems apparent that for most of the mysteries of Lost, series creators Carlton Cuse and Damon Lindelof had little if anything specific in mind for how the resolutions would play out. I’m very interested to see what they come up with the series finale; but I expect that some of the larger mysteries of the show will simply evaporate. I’ll go through my list of suspects in Part 2 of this post.

[Cross-posted at Madisonian.net.]