Just a small, hasty thought today. 
 
As I look at the battle moves in Apocalypse World, and specifically the subterfuge moves, I’ve been thinking a lot about the snowball, and the discussion on the Barf Forth forum that I mentioned back in post no. 149 about the changes that the Bakers made to seize by force for the second edition in order to make the move a battle move rather than a basic move.
 
I link to and quote a lot of the forum thread in my earlier post, so I’ll just quote one small relevant passage here:

The thing that I’m interested in is the way moves are built to “put [their] consequences off into the snowball” rather than resolving the conflict or the situation entirely.  What this reminds me of is the once-hot topic of task resolution vs. conflict resolution.   (You can read Vincent’s early thoughts on this distinction on his anyway blog here: http://lumpley.com/hardcore.html (it’s in the fourth post, titled “Conflict Resolution vs. Task Resolution”)).  Task resolution, at its most basic understanding, is that the thing being resolved is the success or failure of a specific action without a larger set of actions, such as a single sword swing in a battle, or an attempt to move silently over one defined area as part of a larger silent approach.  The scale can be vast or narrow.  In conflict resolution, the things being resolved is the entirety of the conflict at issue, the whole of that specific battle or of that specific infiltration.
 
Moves in Apocalypse World resolve one situation into a different situation where the fiction has significantly shifted in one way or another.  But the set of battle moves are designed to be interconnected as the fiction requires so that they can tumble from one into the other.  For example, the subterfuge moves are specifically designed to “allow the players’ characters to get into or out of a battle on their own terms” (174).  So when you are acting as the bait or the cat or the mouse, the move is designed to shape the fictional positioning at the start of a battle (or in the case of the mouse, to avoid an disadvantageous battle altogether).  When you push the consequences of a move off into the snowball, you are saying to some extent that his action is potentially part of a larger set of actions, like a single task within a larger conflict.  Moves that result in a resolved (and changed) situation are a kind of guided conflict resolution. 
 
Of course moves will snowball even when they aren’t specifically designed to lead into or out of specific situations.  The snowball is then a result of the choices that the MC and the players make with their individual moves, riding the changing situation with escalating actions until they play themselves out to a momentarily stabilized situation.
 
All that is just a thought.  I’m not convinced it’s a very useful thought, but if I don’t write it down my brain is just going to keep playing with the idea and bugging me, so I appreciate your indulgence.  I know that the concept of task vs. conflict resolution is no longer in any way important to the way the Bakers think about RPG design, so I’m not suggesting that this was a comment that they were making through Apocalypse World.  

​The help or interfere move, being a basic move, is designed to work in any narrative situation, and as such it can easily work in battle too.  But while it’s a functional move in battle, it’s not an especially sexy move, and it means that the supporting character is doing a lot less than the character being supported.
 
Enter the tactical and support moves:

​The three moves under the “Tactical and Support Moves” banner—lay down fire, stand overwatch, and keep an eye out—all allow for the supporting player to have their own fiction-shaping move to make.
 
Because seizing by force can cover a large swath of fiction in one move once “the action of the game tips over into battle” (166), in a scene with several PCs acting in unison, the move discourages everyone from making it.  Either everyone has to attempt to seize something different (e.g. tactical position, control of an important object or person, etc.) or everyone needs something specific to do.  This structure is one of the reasons that battles in Apocalypse World are so rich and varied, simulating cinematic fights and not a mere exchange of bullets and punches.  While the collection of basic moves—going aggro, helping, manipulating, and reading situations and people—all work beautifully in battle, these additional tactical and support moves broaden the narrative elements available to the players.  And while these three moves cover a lot of how characters will regularly support each other with shit is going down, they are also meant to inspire you to create similar moves if your playgroup craves other narrative possibilities.
 
The question that these three moves raise for me is the idea of when exactly “the action of the game tips over into battle,” which is the point at which these moves technically become accessible to the PCs.  We know from the in battle move that “when you’re in battle, you can bring the battle moves into play” (141), so at what point in the action can you lay down fire or keep an eye out?
 
Let’s take a look at the keep an eye out example:

​We don’t know what events precede this example.  It could be that they are on the run from enemies so we are already in a battle, but “keeping Audrey busy with driving” suggests that the challenges the characters have been facing have not been battle-related ones.  So what has tipped the action of the game into battle that allows Dust’s player to make this move?  Have the players agreed in some what that they are “in battle”?  Or does the player asking to make the move trigger the possibilities of battle?  Or if the player said, “I think I’m going to keep my eyes out for trouble,” would there need to be a discussion about the possibility of whether they are “in battle” or not?
 
So this is what we know about the way Vincent at least (and probably Meguey too, though I haven’t read enough of her games to know for certain) writes his rules: he is exactly as specific as he means to be.  The phrases “in battle” and “tips over into battle” are purposefully vague phrases because no hard definition is actually necessary.  What makes these moves battle-specific is that the fiction they produce will assume a battle is taking place, and the rules can never know when that is the case, so it is entirely up to the players.  The results of the move involve taking harm or choices from another battle move’s list.  To trigger keeping an eye out is to say that the fiction resulting will specifically involve “an enemy” and/or “an ally.”  So, “the action of the game tips over into battle” literally means that the fiction the players are creating is combat fiction, either when the move is triggered or as a result of the move being triggered, and the game doesn’t care which.
 
In many RPGs, the moment that play tips into battle is significant because special rules kick in at that moment.  In D&D and similar games for example, the order of who acts when is critical when battle begins, but not otherwise.  The time scale can change, going from minutes to seconds, and of course issues of to hit and to avoid being hit are specific to those moments.  As Ron Edwards has observed, often surprise rounds work as a kind of patch to smooth the transition in games from non-combat to combat.  That’s not the case for Apocalypse World, which makes no distinction between combat and non-combat in the first edition, and makes the distinction in the second edition only to make sure that the fictional outcomes fit with the fictional inputs.  Moves can go in and out of combat fluidly, and are designed to do so, allowing you to make non-battle moves between battle moves and vice versa, so “in battle” only ever refers to the fiction that is being created right now with this move.