Well any reasonable expectations for an end game has to have those who popularly have power with actual power, with checks and balances. Iraq isn't going to be like the Dubai or Turkey any time soon, and the only reasonable victory we can hope to achieve is developing a point of stalemate between the various forces which have sustainable power and influence in Iraq which ensures the more important of our interests to an acceptable degree.
Machiavelli said something along the lines of the only way to impose your will upon an unwilling populace is to absolutely break it, which will involve killing enough people to compell submission. This is possible theoretically, but it's clearly not something we're willing to do as a people (which is something our enemies also know, and gain a great deal of their power from this knowledge). And thus we cannot expect to ever successfully impose our unabridged will upon the Iraqi people.
So thus what we need to develop is what is referred to strategically as a Nash equilibrium of cooperation based on the indefinate repeating status of the games (strategy talk for competitions over the advancement of self interest) and knowledge of when the other is willing to defect.
The first picture is that of a classic prisoner's dillemma. A prisoner's dillemma is a situation in which the Nash equilibrium is Pareto inferior to another strategic point. The second picture is a demonstration of how pure strategy Nash equilibria are found. If Player 1 chooses to cooperate, then it is in Player 2's best interest to defect, as that carries a benefit of 4, rather than 3, and the same applies for Player 1 if Player 2 chooses to cooperate, and clearly results of either (4,1) or (1,4) are not optimal for at least one of the players. But if Player 1 chooses to defect (which s/he always will, as s/he always does better from defection, thus it strictly dominates cooperation), then it is in Player 2's best interest to defect also, and thus the only result in which both players have behaved strategically is (2,2), or that they both defect.
But this result is Pareto inferior to the result (3,3), in which both cooperate (Pareto inferior means that the result is either worse or the same for all players involved than another potential solution, and worse for at least one player involved), problem being that if any player chooses to cooperate, the other still has incentive to defect, and thus cooperation is strategically impossible.
Except if you take into account that there is more than one game. Now if there are a finite number of games, then cooperation is still strategically impossible, because if I know you have incentive to defect on the last time, then I have incentive to defect on the next to last time to offset that, and thus the other has incentive to defect on the 3rd to last time, and so on and so forth all the way back to the beginning, and considering that both have incentive to defect on the last game, both have incentive to defect every time, and thus cooperation is once again strategically impossible.
But then there are indefinately or infinitely repeating games, which is more relevant to Iraq (because the significant leaders know that, in the end, they're all stuck together geographically, and thus indefinately). Cooperation is possible in indefinately repeating games, depending on the conditions under which either player will either cooperate or defect. For example, if Player 1 is always willing to give Player 2 another chance, regardless of how many times Player 2 defects, then Player 2 will always defect, as there is benefit in defecting, and there is no cost. There is the tit for tat approach, which is that I will start out cooperating, and if you defect, then I will defect until you cooperate while I defect, at which point I will start cooperating again. There is also the Grim Trigger approach, which is I will cooperate until you defect, at which point I will never cooperate again.
And thus the difficulty in finding said Nash equilibrium, because the potential effectiveness of any approach grows as the consequences for defection increase, but then it is also reasonable for the other players to defect, in an attempt to call a bluff, at which point it cannot be a bluff, and thus both sides will have to deal with the opportunity cost of any such testing.
But it's possible, and considering we don't intend on doing what's neccesary to break their will as a nation, we have to find a solution in which the popular will is recognized and adhered to, and thus we need to make it so that those who make the strategic decisions with the power of collective will behind them will be left strategically coming to the courses of action that we prefer, which is finding said Nash equilibrium in which cooperation is strategically possible.