Rules tend to break at the extreme high and extreme low end of the system. Its okay to me that it might wonk a bit in the case of characters with a +0 attack bonus fighting a Dex 1 enemy who then passes out. (If only because very few combats continue on after the enemy has passed out, and when they do, its usually just the players saying "We beat him to death.") The situation is rare enough that I'd be fine with however the GM adjudicated it, either with a circumstance bonus, or just by following the RAW and making an unconscious guy slightly harder to hit than the conscious, crippled and flanked guy.
To me it comes down to what you consider "flanking" to be. Does the bonus derive from two allies coordinating their strikes to minimize the opponent's defenses (say, with one striking high while one strikes low), or is it something that happens to the defender when he has to split his attention in two directions? If you rule its the former (which the PHB seems to), then only allies can flank. If you rule its the latter (which is understandable), then its something two unallied attackers could do to a third.
Even if you rule attackers must cooperate in order to provide flanking bonuses, there's no reason why two enemies beset by a third couldn't agree to work together. I'd rule if two characters are seeking flanking on a third, they'd get it, even if they weren't allies. (I also could see two characters in a slow 5' step chase around a larger creature, taking their flanking bonuses each round but waiting to get back to the business of killing one another when the third opponent drops.) On the other hand, say in the case of a fighter trying to kill a rogue who are attacked by an ogre, I could see the fighter denying flanking just to make the rogue's life harder (or shorter). To me, this would be funnier if the fighter had to 5' step out of flanking every time the rogue moved into it, which is why I'm leaning towards the second interpretation of flanking.