I think that JA COULD have had a point, as [MENTION=85555]Bedrockgames[/MENTION] explains it there's a fairly reasonable consistent dissociation criteria. I can even buy the definition JA uses for it here. It is all the other screed that goes around it in The Alexandrian that makes it ridiculous edition warring. Had he adopted a more balanced viewpoint and a neutral tone, looked at and analyzed the game in more depth and truly understood it then he might have provided some interesting and useful analysis. IMHO as it is that didn't happen... Luckily I don't have to read other people's blogs though, so overall the only way it bothers me is the sheer number of times I've had to endure bad arguments rehashed here or on other forums in even less coherent forms.
In his second run at the topic, JA does, indeed, remove
most of the overt ties to 4e, but the agenda is still quite clear and his argument seems to me to have two gaping holes.
What he does in that blog is first set up a supposedly clear "definition" of dissociated mechanics as "decisions by the player that do not map directly to decisions by the character in the game world", and then he insists that such decisions are the definition of "playing a role".
Okay, objection number one: that's not what "playing a role" has ever meant before. The most recognisable meaning in common parlance is when an actor "plays a role"; but in this case they (generally) follow a script, deciding to say things and take actions in order to serve the story being told, rather than deciding to say and do those things for the same reason the character does. Even in improv acting, there is (necessarily) a good deal of "meta" input to the actors' decision making processes. Those wishing to "get closer to the character" sometimes choose "method acting" - but even this is not an excuse to simply ignore the script and director, it is (as I understand it) the process of internalising the decisions mandated by the script to build the character so as to make those decisions his or her own. Quite similar to the way in which some 4e players describe internalising 4e's supposed "dissociated mechanics", in fact.
Objection number two is the idea that there can be a system without dissociative elements. For this I need to break down how character decision making might work, so I need to make a few explanations. First off, there is no such thing as "the character"; they are figments of our imaginations, as indeed is the world they live in. As such, when I say "the character does such-and-such", this is actually shorthand for "we might imagine with plausibility intact that the imaginary character will experience the imaginary world they inhabit in this way". That "imaginary world" could, in principle, work in any way whatsoever, limited only by our imaginations. I have sometimes played with the idea of doing 'Flatland' as an RPG, but the world as perceived by the characters would be almost completely dissimilar to our own experience, making the imaginative leap very hard indeed. Which brings me to the point that RPGs are
usually set in worlds superficially quite similar to our own and peopled by characters we might reasonably imagine to have perceptions of
their world that are as complex, as nuanced and as instinctive to them as our own perceptions of this world are.
This brings up a problem. The player in a roleplaying game has a proxy for the character's experience of their world conveyed to them via what is, compared to that complex, broad and nuanced perception we enjoy of our own world, a hopelessly restricted bandwidth. As a result, there is a rich palette of limitations and opportunities for the
character, the full scope of which the player cannot possibly be aware. This alone makes "taking decisions for reasons that map directly to the character's reasons" problematic. Either the mechanics limit the information upon which a character decision can be made to that tiny subset of information that is also available to the player, or some way must be found to represent the myriad limitations and opportunities that are apparent to the character in the decision making of the player. Neither option allows us to have simultaneously both a plausible character in relation to the world, and player decisions that are based on the same logic as that used by the character. Both are "dissociated"; one has a decision making logic "dissociated" from that which the character uses, the other has character logic that is "dissociated" from what we experience in the real world.
Let's look at JA's "one handed catch" ability as an example. The chances are the receiver will make such a move only once or twice in a game, at most. So, when will they happen? Well, as he runs, the receiver has a whole plethora of sensory input. In the moments before he either tries for, or doesn't try for, the catch these will give him a picture of just how likely such an attempt is to succeed - information the player of this "character" can base only on a few words from the GM or the rough position of some minis on the table. What happens next is key; the receiver either
decides not to try the catch, because he doesn't think it's feasible given the circumstances, without expending any effort or time on the "failed attempt", or he goes for the catch with, if he is suitably skilled, a pretty good chance of success. Do the mechanics for this model the "pretty good chance for success"? If so I see a character who catches an infeasibly large number of one handed catches. Or do they model the "mean" chance of success? If so, I see a character inexplicably failing to do anything because they spend their "action" to try a catch at a likelihood no sane receiver would try for - or they simply don't use the ability at all.
I would boil the problem down to one of modelling "skill" in an RPG. Skill is composed of two parts: perception and habitual action (or "knack"). The 'perception' part - instinctively 'seeing' a situation in a way that the unskilled really don't, like a sculptor "seeing" the finished figure inside the wood, or a mechanic sensing what is wrong with an engine based on the sounds and movements it makes, or a wide receiver judging whether a one handed catch is "on" based on sensory information about the field around him and his own body - is extremely hard if not impossible to model in an "associated" way. The habituated physical movement (or even mental activity, thinking of facility with arithmetic and maths) is relatively easy to model - and some seem to believe that, once they have modelled that, they have modelled "skill". But that leaves, at least to my perception, a gaping hole where "skilled perception" should be.
Oh, hey - I have
got to pick up an album recorded by those guys!