When talking about the MDA framework, there seems to be an assumption that the mechanics of a game (sometimes combined with assets) create the dynamics while the game is played, and that the dynamics of the game cause the aesthetics to appear in the mind of the player. But I feel this diminishes the part of the player in the game.
A game's mechanics can do nothing to determine the behavior of the game at run-time; the mechanics only determine the possibility-space of the game, many parts of which will never be reached by anyone but an overworked beta-tester. The dynamics are a bit like roads. By that I mean, while a road isn't the only way to traverse terrain, it's the one most people will use. Or as Marc LeBlanc said, "in hockey there's no rule that says you have to skate backwards, but if you're playing defense, it's a good idea."
He then went on to say, "it's an emergent behavior that arises from the rules of hockey. Dynamics emerge from mechanics." However, he also said that "the player is part of the system too, so some of our understanding of game dynamics has to be an understanding of human dynamics." If we look at the hockey example, we can see the influences of both the mechanics and human dynamics. Skating backwards is a good idea, "if you're playing defense." Thus, the value of skating backwards is determined by two things: (1) the end which the dynamic of skating backwards seeks (viz. defense) and (2) the mechanics of the game, which encapsulate the outcome of this behavior at this point in this game, either causing or preventing it from achieving its end. (From the point-of-view of any one player, the others are functionally identical to game mechanics, hence the existence of AI NPCs.) The presence of skating backwards as a dynamic in actual hockey matches can only be because of this value.
Defense is also a dynamic. (One may note, a dynamic of dynamics.) There are no rules in hockey which mandate that players go on the defense, but if they're setting up for offensive play, it's a good idea. That in turn is a dynamic, and if we follow the chain up we find 'setup for offense' -> 'score' -> 'win' -> 'experience challenge'. Challenge, of course, is one of the MDA paper's enumerated aesthetics, and of course it, unlike its predecessors, is outside of the game and inside of the mind. And so, it is correct that challenge be an aesthetic, and not a dynamic. But I can't avoid the conclusion that it, or rather its consequences, are what is meant by human dynamics. Winning would have no meaning if there were no challenge to it, and so players would not play to win, and the dynamics would be different; hockey would devolve into combination figure skating and boxing.
It is important that the chain reaching from skate backwards to challenge is not a chain of cause and effect. It is a hierarchical relationship from the most specific action to the most universal. There is no single burst of challenge when one's team wins a hockey match; challenge arises in all specific actions because and only insofar as they work toward winning.
It is also important that this hierarchy is not of the same sort by which bricks make up a house, and if the house were never built, the bricks would survive. Game dynamics could not exist without both (1) their being for an aesthetic and (2) their being possible given the mechanics.
The reason I'm posting this here is because I need critique. This is just an afternoon musing, and probably makes very little sense -- at least so far. So what do you think?
TL;DR: Aesthetics create dynamics, not vice versa.