Sage Advice Compendium Update 1/30/2019

[Asgorath]

I think the word Now does a nice job of differentiating those ideas. You declare the action NOW. Or you declare the action for later. In your concept the action must be declared now. So then even if the attack action is a discrete sequential event separate from the attacks it grants then it must immediately precede them and doing anything else between the attack action and the attacks would mean that you didn't actually follow through with your declaration and so you didn't actually take the attack action.

I might could get behind that concept. Actions are discrete events that must be declared immediately as you are using them. Then their effects immediately follow. This interpretation would allow any bonus action to be used inbetween extra attacks. It would allow movement to be used after taking the disengage action. *It would allow the shield master shove attack to be used after the first attack but not before it.

I agree with everything here except the final sentence. The trigger for Shield Master’s bonus action is the Attack action, not part of the Attack action. If you split the Attack action and perform other legal activities between attacks such as movement, you must still complete the Attack action before you can do something that is triggered by it.

 

[FrogReaver]

My proposition allows for actions to be be divisible by movement.

"...using some of your speed before and after your action." does not exclude movement during the action. It just points out permissions for those other two situations. So that rule does not show that actions are not divisible by movement.

Moving between attacks is not intuitive. Many will think you can, and many will think you can't, so it needed a rule to make clear that the game does allow it for sure in that situation. The existence this rule does not mean that this is the only time you may move in the middle of an action, so it is not a rule that shows that actions are not divisible by movement.

Okay, but the existence of the rule is strong evidence that you cannot move during an action. I prefer not to rely an interpretation that makes me disregard strong evidence provided by the rules.

The Dash action adds to movement, so it makes crystal clear sense that you should be able to move during that action, so there was no need to create an extra rule for it like they did with Attack, and which is not excluded by "...using some of your speed before and after your action."

RAI you are absolutely correct, RAW is a different story though. The contradiction I keep pointing to is a contradiction about what your interpretation forces upon RAW compared to what we all know the RAI is.

Now, I've argued before, and still believe, that just because something is not excluded by the rules, does not automatically make it included in the rules. However, the natural reading Dash and other movement actions, as well as the apparent reasoning behind why Attack is called out separately, and the fact that actions being divisible by movement is not excluded, strongly indicates to me that actions are divisible, and movement is one of those things that many actions are divisible by.

Your argument is:
1. The disengage action gives you a benefit if you move (raw)
2. The action lasts till the end of your turn (your interpretation)
3. The RAI is that you can get a benefit to your movement (rai)
4. Conclusion Therefore RAW must be that the disengage action is divisible by movement.

You do realize that what you are saying is only true if the action lasts till the end of your turn. I've offered proof that it cannot. So your argument here doesn't have any bearing on mine unless you can show the disengage action must last till the end of the turn. If you can't do that then your argument doesn't invalidate my proof. In fact, my proof would invalidate your argument, because my conclusion would result in one of your premises being incorrect. By all means, provide an argument that one of my premises is incorrect.

 

[FrogReaver]

I agree with everything here except the final sentence. The trigger for Shield Master’s bonus action is the Attack action, not part of the Attack action. If you split the Attack action and perform other legal activities between attacks such as movement, you must still complete the Attack action before you can do something that is triggered by it.

If the attack action is a discrete sequential event then it must occur somewhere within our sequence. Is there anywhere in the chain of discrete sequential events that the attack action can be placed such that you must take the shield master shove after both your discrete sequential attacks? I don't think that's possible.

Your position is actually that the attack action isn't a discrete sequential event but rather that it is composed of discrete sequential events. My argument is that it's better to consider the attack action and all other actions as discrete sequential events that are declared NOW.

 

[FrogReaver]

A declaration in the common usage in D&D, how JC is using it, and how we are using it here, is a statement of future intent.

Player: "DM, on my turn I am going to move 5 feet, then cast misty step, then use my Attack action, then move 25 feet." That would be declaring what your action will be.

When you are informing the DM of what you are doing at this moment, it's a statement of action. Yes, it's technically a declaration, but it is not the kind of declaration that we are discussing. At no time, though, whether using the kind of declaration everyone else is talking about, or the kind of declaration you using on this technicality, is the declaration valid within the game rules. Being entirely informal, it has no mechanical game validity whatsoever. Nothing can trigger off of a declaration of any kind. Only when you are actually engaging the action does anything begin to trigger off of it.

It's the kind of declaration I've been discussing. I've made that abundantly clear. Why do you keep talking to me about this other type of declaration when it's not what I'm talking about?

 

[Maxperson]

Your argument is:
1. The disengage action gives you a benefit if you move (raw)
2. The action lasts till the end of your turn (your interpretation)
3. The RAI is that you can get a benefit to your movement (rai)
4. Conclusion Therefore RAW must be that the disengage action is divisible by movement.

You do realize that what you are saying is only true if the action lasts till the end of your turn. I've offered proof that it cannot. So your argument here doesn't have any bearing on mine unless you can show the disengage action must last till the end of the turn. If you can't do that then your argument doesn't invalidate my proof. In fact, my proof would invalidate your argument, because my conclusion would result in one of your premises being incorrect.

First, it doesn't need to last until the end of the turn. It only needs to last until the end of movement, which is a different. Second, on #2, I've already shown how Attack and Cast a Spell are both non-instant actions. They do in fact have a duration. Others seem to as well.

"When you take the Hide action, you make a Dexterity (Stealth) check in an attempt to hide, following the rules in chapter 7 for hiding." This is clearly not an instant effect. You don't just take the hide action and then the effect is X. It says that when you take the hide action, you make the dex check in the attempt to hide. It takes time to get behind something to hide as you don't just instantly vanish behind an object. Help is an aid to the action of another, including Attacking which we know takes time. To help with an attack, you have to feint or act in some other credible manner with your weapon to make the enemy believe you are attacking and give your ally advantage. That would take time similar to the Attack action.

I don't see why we should just assume that Dash, Disengage and Doge are somehow instantaneous when the other actions are not. 1 action is a length of time in combat per RAW, so it would seem to me that the actions would all also have a 1 action length of time associated with them. Unless you can show where 1 action has a length of time here, but a time of instant there.

 

[Maxperson]

It's the kind of declaration I've been discussing. I've made that abundantly clear. Why do you keep talking to me about this other type of declaration when it's not what I'm talking about?

Okay, but what you are talking about has no mechanical game validity at all. No declaration does. A declaration of any sort cannot trigger any abilities.

 

[FrogReaver]

First, it doesn't need to last until the end of the turn. It only needs to last until the end of movement, which is a different.

For precision that's fine but it's uncessarily wordy and not of much importance to this discussion. All I would have had to do was add the caveat "as long as you still have movement" and there would be no objection from you. So let's not get distracted by this as there are actual important points to discuss.

Second, on #2, I've already shown how Attack and Cast a Spell are both non-instant actions. They do in fact have a duration. Others seem to as well.

Well, you got half way there. There's a notion that the action happens and then it's effects. You haven't offered anything to show that notion is incorrect. As such the cast a spell action may occur and then the casting of the spell may occur. I'm not saying that's the case just that such a possibility hasn't been ruled out.

"When you take the Hide action, you make a Dexterity (Stealth) check in an attempt to hide, following the rules in chapter 7 for hiding." This is clearly not an instant effect. You don't just take the hide action and then the effect is X. It says that when you take the hide action, you make the dex check in the attempt to hide. It takes time to get behind something to hide as you don't just instantly vanish behind an object. Help is an aid to the action of another, including Attacking which we know takes time. To help with an attack, you have to feint or act in some other credible manner with your weapon to make the enemy believe you are attacking and give your ally advantage. That would take time similar to the Attack action.

When is not an indicator that an action and it's effects last together in tandem. Otherwise the shield master shove that says "when you take the attack action", would definitely allow you to shove between the attacks. In fact it wouldn't allow you to shove any other time...

I don't see why we should just assume that Dash, Disengage and Doge are somehow instantaneous when the other actions are not. 1 action is a length of time in combat per RAW, so it would seem to me that the actions would all also have a 1 action length of time associated with them. Unless you can show where 1 action has a length of time here, but a time of instant there.

Because RAI fails to work by RAW when you interpret them as lasting the effect duration. It all comes back to my proof. Which isn't a proof of instantaneousness, it's a proof that actions and their effects have different durations. That premise then leads to the belief that the disengage action is instantaneous but it's a separate conclusion from what I prove.

 

[FrogReaver]

Okay, but what you are talking about has no mechanical game validity at all. No declaration does. A declaration of any sort cannot trigger any abilities.

Of course it's mechanically valid to the game. Try to play the game without declaring the disengage action and see how far you get.

 

[Asgorath]

If the attack action is a discrete sequential event then it must occur somewhere within our sequence. Is there anywhere in the chain of discrete sequential events that the attack action can be placed such that you must take the shield master shove after both your discrete sequential attacks? I don't think that's possible.

Your position is actually that the attack action isn't a discrete sequential event but rather that it is composed of discrete sequential events. My argument is that it's better to consider the attack action and all other actions as discrete sequential events that are declared NOW.

Why is it not possible?

1) Move
2) Attack (1/2)
3) Move
4) Attack (2/2)
5) Move
6) Shield Master shove

You can place the shove anywhere after 4, because that's the point where your Attack action is complete. The Attack action is unique in that there are rules exceptions that allow the action to be broken into smaller pieces, with things like movement in between. Those pieces must all still be processed before the action can trigger something else like Shield Master's shove. I'm not sure why the concept of "are you done attacking or not" is that hard to figure out? If you are still making attacks from the Attack action, you haven't finished the Attack action yet.

 

[DND_Reborn]

Why is it not possible?

1) Move
2) Attack (1/2)
3) Move
4) Attack (2/2)
5) Move
6) Shield Master shove

You can place the shove anywhere after 4, because that's the point where your Attack action is complete. The Attack action is unique in that there are rules exceptions that allow the action to be broken into smaller pieces, with things like movement in between. Those pieces must all still be processed before the action can trigger something else like Shield Master's shove. I'm not sure why the concept of "are you done attacking or not" is that hard to figure out? If you are still making attacks from the Attack action, you haven't finished the Attack action yet.

Because for the bonus action for Shield Master their is no specified timing, only the condition that you take the attack action. My interpretation is you can shove anytime after 2 since you have used your attack action. Flurry of blows is a different example because it specifies immediately after you take the attack action. Does that mean it has to be complete if you have Extra Attack? Maybe, maybe not, that is up to the table's interpretation.