I hate the term “ludonarrative dissonance.” It sounds needlessly florid, and it's the sort of thing that gives aid and comfort to the people who think that games writin' has gotten too fancy. Though it exactly describes the phenomenon in seeks to explain, it's the kind of phrase that stands in the path of common human understanding, and this is never a good thing. I wish I had a good substitute neologism at hand, but the plainer “story-game conflict” lacks zazz. Maybe the readers have a pithy and non-latinate alternative.
In a recent speech at the Montreal International Games Summit, Blow described ludonarrative dissonance as one of the three ways modern games are “conflicted.” (He calls it a conflict between story and “dynamic meaning”, but it's basically the same idea.) Here's the conflict: the game's mechanics lead you to have certain attitudes towards a character or situation, and the game's story tells you (or your character) to feel a different way. Because these two elements of the design pull in different directions, the game fails to achieve what it wants to communicate in either case. Hence the dissonance.
We might say: mechanics don't tell you anything. Mechanics are just sets of rules that define what you can do in the game: jump, run, shoot, whale on ninjas. This is right; textbook cases of dissonance revolve around a smaller subset of rules that define “progress” in the game: how it rewards you for doing certain actions by giving you new abilities and allowing you to progress through new terrain, and how it punishes you by forcing you to replay certain segments and preventing you from getting further. Blow advances the idea that these elements of the game's mechanics have a quasi-moral significance that is ultimately significant w/r/t the narrative. By giving and taking progress from the player, a game communicates a message about what is right and wrong to the player. Blow critiques World of Warcraft, for example, because he thinks that it showers unstinting rewards on the player's unpraiseworthy drudgery.
In the original case of ludonarrative dissonance, Clint Hocking's critique of Bioshock, the point of conflict was the player's decision to save or harvest the “little sisters”, a well-defended posse of bioluminescent poppets with a monopoly on a crucial resource which the player needs to upgrade their abilities. From the standpoint of the story, this decision is crucial; it's cast as a harrowing moral choice between virtue and self-interest, a decision that resonates with the story's larger thematic concern with the costs of untrammeled self-interest. But from the standpoint of the mechanics, either decision is more-or-less equal over the long term when it comes to progressing through the game. And this tells the player that the decision isn't as important as story says it is.
Almost all of Blow's examples focus on conflicts between narrative elements and the reward-structure of the gamplay. But there's a different kind of dissonance in games as well, which is as important to thematic coherence but is but harder to articulate. It's the way that game mechanics, the controls and movement and gameplay alltogether, invest your actions with a specific texture. Game mechanics (the controls, but also the structure of the levels and the enemy design) can create a certain tactile impresions to them that's hard to put into words. The way a game controls, the way it feels in you hands, can speak volumes.
An example might help: Mitch Krpata recently wrote a post on Gears of War 2, and said that he felt that there was a disconnect between the gameplay and the story. As a cover-based shooter, the gameplay centers around holding down a defensive position and keeping distance between yourself and the hordes of locust. But “The storyline... puts the COGs on the offensive for the duration. It's all rah-rah, take-it-to-'em stuff. You get all geared up to fight, pardon the pun, and then spend all your time with your head down. That doesn't make sense.” Krpata found the defense-based gameplay was satisfying in the context of the first Gears, where humanity was up against the ropes. But now that the story licenses the player to go out and get some, it seemed unsatisfying to spend your time cowering behind sandbags.
It's also worth noting where this marriage of narrative and gameplay goes right. I've been playing the most recent Prince of Persia game this week, and one of the things I like about it is distinctive feel of the platforming. Both the platforming and the combat have been likened to a rhythm game, and this is a strange-but-appropriate comparison. Because you don't control your momentum in midair, traversing the world is mostly a matter of timing your button presses correctly. The jumping and climbing has this nice tactile rhythm, and because the game is so forgiving, it allows you to focus on charting a fluid path through environment without worrying overmuch about death frustrated. With practice, you get the point where you can read the terrain like a row of glowing gems in Guitar Hero. The feeling of lightness you get from conforming to the tempo of the terrain complements the narrative, since the hero is depicted as a of carefree gadabout in a consequence-free magical kingdon. The banter, the visuals, the pace of the platforming, the supernatural aura, all these things work together to create this feeling of freedom and careless heroism. These gameplay systems wouldn't work in every game, but they function wonderfully in the context of the world and narrative that Prince of Persia attempts to create. Call it ludonarrative harmony.
I don't agree with all of Blow's arguments (and I take some issues with his examples, which I'll spare you out of an unusual dispassion for pedantry) but I do think he does an unusually good job of articulating the the problems and tensions in so many modern games. The challenge is finding ways to meld form and content, gameplay to story. These elements in modern game design have had one shotgun wedding too many, they deserve some happy nuptials.