Let's have a short look at the Action Window flow charts on page 21 of the FAQ again.

We see that both step 4 (Passives) and step 5 (Responses) have arrows leading to a subset of steps numbered I-IV. This tells us that Passives and Responses also go through the same Initiation-Save/Cancel-Resolution cycle the normal action does, and that there's also a step where Passives induced by the Passive/Response in question are resolved. Such a step is obviously needed, because the regular Passives step has already passed by the time the Response is triggered. If it wasn't there, there'd be no point where e.g. CS Benjen Stark's passive ability could resolve if he was killed by a Response, say, No Quarter.

So far, so good.

Now, step 2 of the action window (Save/Cancel Responses) has no such arrow. This has to be due to space constraints, of course - we know that save/cancel responses can themselves be canceled, we do it all the time. Also, it is implied in the section on save/cancel responses on page 18 of the FAQ that saves and cancels also go through the Initiation-Save/Cancel-Resolution cycle, just like other responses.

So, if save/cancel responses go through the same steps I-III normal responses do (even if the flow chart doesn't actually have the arrow pointing to those steps), it stands to reason that they'd also go through step IV, right? That would mean that during step 2 of the action window, there's a separate step for resolving passives induced by save/cancel responses triggered during that step. That would in turn mean that there are actually passives that resolve before step 4 of the action window, no? For example, if CS Benjen Stark is killed to pay for Harrenhal, would his passive actually resolve in step 2.IV instead of step 4? That would make an actual difference in gameplay, so this exercise is not entirely academic.

But you sort of answer your own question when you say, "Such a step is obviously needed, because the regular Passives step has already passed by the time the Response is triggered" in relation to Passives and Responses. When a save/cancel is triggered and resolved, the "regular passive step" associated with the original initiation (that you are trying to cancel or save from) has not happened yet, so passives that would be initiated by the save/cancel are rolled into that passive step - the one for whatever you are interrupting.

But you sort of answer your own question when you say, "Such a step is obviously needed, because the regular Passives step has already passed by the time the Response is triggered" in relation to Passives and Responses. When a save/cancel is triggered and resolved, the "regular passive step" associated with the original initiation (that you are trying to cancel or save from) has not happened yet, so passives that would be initiated by the save/cancel are rolled into that passive step - the one for whatever you are interrupting.

Yeah, I wasn't quite sure if I answered my own question. You say that because the step isn't needed, it isn't there, and a passive induced in step 2 will just resolve in step 4. Which is logical, and I'm fine with it.

But I think the alternative is also conceivable: That there is a general structure to actions and responses, and that structure is generally followed. I mean, if 5.IV is only there because it is needed, then why is 4.IV even there? That one isn't needed - we're still in step 4, those passives could resolve just fine there without mentioning that step. In fact, the way 4.IV is worded (both in the flowchart and on page 18), my initial assumption would be that passives induced in turn by a regular passive would all resolve in 4.IV, before the next regular passive initiates. That means, even if there would be a conflict between a 4.IV passive and a regular passive, the 4.IV passive would always resolve first, rather than the order being decided by the FP.

As I said, I'm fine with your ruling, though. Passives resolve in step 4, unless they can't. It's a lot cleaner than the alternative. I just don't think it's the only conclusion that can reasonably be drawn from the FAQ.

While we're at it, there's another arrow originating from both 4.III and 5.III, leading to a box on the right with the text "Follow steps I through IV, etc." What the hell is that supposed to mean?

I mean, if 5.IV is only there because it is needed, then why is 4.IV even there? That one isn't needed - we're still in step 4, those passives could resolve just fine there without mentioning that step. In fact, the way 4.IV is worded (both in the flowchart and on page 18), my initial assumption would be that passives induced in turn by a regular passive would all resolve in 4.IV, before the next regular passive initiates. That means, even if there would be a conflict between a 4.IV passive and a regular passive, the 4.IV passive would always resolve first, rather than the order being decided by the FP.

Well, it kind of depends on how technical you want to get. Remember that all passives are supposed to happen at the same time (unless they can't). So you could argue that there is only one time in Step 4 when all available passives initiate. But if passive #1 needs to resolve before passive #2 is created, passive #2 will miss that initiation opportunity (kind of like the way that using the Response on Banner for the Storm gives a character Vigilant after the point of initiation for passives to winning the challenge). So 4.IV is there to make sure that when Harrenhal discards itself, there is no question that the "Prized" passive it therefore creates gets a chance to initiate.

It's not worth going into very deeply, but there is a technical argument to be made that 4.IV is as necessary as 5.IV in its own way because the conceptual idea is that all passives initiate together.

Remember that the flow charts were created as a graphic representation of the rules close to 10 years ago, but that the understanding of the rules has been evolving since that time. Like a map, the visual representations are not always going to be able to provide 100% of what you need to know.

Thank you for your input.

Remember that the flow charts were created as a graphic representation of the rules close to 10 years ago, but that the understanding of the rules has been evolving since that time. Like a map, the visual representations are not always going to be able to provide 100% of what you need to know.

Oh, absolutely. One of the reasons why I think the FAQ could do with a rewrite from scratch (that we'll probably never see). Anyway, thanks four your input. Always a pleasure picking your brain on these technicalities.

I still don't know what those boxes "Follow steps I through IV, etc." are supposed to mean. Any idea? Am I missing something obvious?

I take those to indicate that you follow Steps 4.I-4.IV and Steps 5.I-5.IV for each individual passive set and/or individual Response triggered.

The fact that the arrows coming off of "part 3" (and thus would come before "part 4" of the previous effect) rather than part 4 is something I chalk up to the graphic representation not keeping pace with the 10 years worth of evolution in understanding how to explain and verbalize the flow of the game.