Sage Advice Compendium Update 1/30/2019

[DND_Reborn]

The key issue Arial Black and other supporting that position have is this: because the Attack action and bonus action shove can be simultaneous, only the order of resolving them needs to be determined. He stated this ordering is determined by the player, which is slightly incorrect as it is the DM who controls the narrative, but since the DM often defers to the player's wishes, this is a very minor oversight really.

The problem with even allowing the shove to be resolved first, assuming the Attack action is simultaneous but the DM allows it to be resolved second, is scenarios where after you resolve the shove you cannot attack anything and thus never take the Attack action. If you never take the Attack action, you never satisfied the condition of Shield Master which allowed you to shove. This leads to paradox and how do you handle that?

Example. You party has been engaged in a battle with an archmage. Everyone else is unconscious or dead, so it is just you and him now. On your turn, you employ your bonus action to shove, deciding to resolve it before resolving your attack from your Attack action. Unknown to you, the archmage's contingency was if he is knocked prone by an enemy, he teleports to his hideout. So, the DM says as soon as you knock the archmage prone, he disappears. But, you have no target to attack with your Attack action, so how can you take it on your turn???

Now, you might argue that since your actions were simultaneous, you should be able to resolve your attack before he teleports away. The DM agrees, thinking this will avoid the paradox, so you make your attack roll and do so with advantage (after all, you shoved the target, right?). The DM puts his hand up before you roll, telling you that you don't have advantage because the actions are simultaneous, since even though he is allowing you to resolve one and then the other you agreed they occur at the same time. So, since they are simultaneous you cannot benefit from the effect gained by shove until the simultaneous actions are both resolved. If you argue against this, you are not in fact doing the actions simultaneously. If you try to benefit from the prone target via Shield Master, it in fact came first and the attack followed second. If one is first and the other is second, they aren't simultaneous.

The DM could tell you to roll two d20's at the same time since the actions are simultaneous. One d20 is for the Strength (Athletics) check versus the archmage's Dexterity (Acrobatics) check for the shove, and the other d20 is for your attack roll versus his AC. Since they happen together, you don't have advantage on the attack roll.

If you claim they are simultaneous, the order of resolution is not important because you cannot benefit from the shove (advantage on attack) since the opponent is not prone when you simultaneously make the attack roll. At best, you could resolve it as "shove, attack (no advantage because the shove is simultaneous), attack (with advantage since the shove is now complete)". This is no different from "attack, shove, attack (with advantage due to shove)". In short, it is easier to resolve the order as: "attack (no advantage), shove, attack (with advantage)", since this makes it more obvious when you can attack with advantage than resolving them as "shove, attack (no advantage), attack (with advantage)". You have the exact same actions with the exact same benefits, only the order changes between the shove and first attack.

This is actually pretty important in another way if you think about this: suppose you only have one attack for your Attack action (no Extra Attack)?

By the argument of simultaneous actions, that the shove and attack must occur together, you can never gain the benefit from the prone target because the actions are simultaneous! It becomes irrelevant if you shove then attack or attack and then shove when the actions are simultaneous.

I hope at this point no one is still arguing you can use the bonus action and THEN use the Attack action, that they need not be simultaneous or that the Attack action must come first. The only two logical interpretations IMO are:

1. The bonus action Shove can be taken simultaneously to the Attack action.
2. The bonus action Shove must be taken after the Attack action.

Either way, unless you have the Extra Attack feature, you can never benefit from the shove on the first attack made on your turn. Of course, as I have shown in other posts, this goes against JC's official stance from SA and tweets, but that is only of concern to your table if you value his position.

 

[Hriston]

The rule you quoted does not say what you wish it said.

During your turn there are two states of actionhood.

1. You have not taken your action this turn.

2. You have taken your action this turn.

If you wait until 2/3 of your turn is done before taking your action, up until that moment, you are in state 1. You have not taken your action this turn. At that moment when you take your Attack action, the state your character is in switches from "Has not taken the Attack action this turn." to "Has taken the Attack action this turn," and remains that way for the last 1/3 of the turn, allowing you to use your Shield Master bonus action.

The rule allowing you to move before and after your action does not change that fact.

I don't need a rule to change that because it's irrelevant to the condition set up by the feat. The condition for using a bonus action to shove a creature is not that "you have taken the Attack action on your turn," so it doesn't matter which of these two states prevail when you use the bonus action the feat gives you.

 

[DND_Reborn]

But I may satisfy it later on my turn. If I take the Attack action later on my turn, then I took the Attack action on my turn but you didn't allow me to use a bonus action to shove a creature. That isn't following the rule.

Then again, you may not, which leads to paradox since the condition required to gain the bonus action was that you take the Attack action on your turn.

Then you only shoved on your turn. How is that a problem? It isn't breaking anything.

Again, the problem is you are employing a bonus action granted by a feature and didn't satisfy the requirements of the feat to gain the bonus action. That is breaking the rule of gaining bonus actions allowed by satisfying conditions to gain them.

Of course not! The shove attempt is always resolved the same way whether you use a bonus action to do it or not.

I never said it wasn't and that is immaterial to the argument.

You can shove a creature without using a bonus action any time you like, though. The benefit of the feat is it lets you shove a creature AND take the Attack action on your turn. It's an added benefit!

Accept that to gain the bonus action shove you must take the Action. I have shown numerous examples of how it leads to paradox if you use your bonus action to shove before you attack because you can be denied the attack later on.

Right, which invokes the general rule for bonus actions. "You choose when to take a bonus action during your turn, unless the bonus action's timing is specified," so following the RAI established by Jeremy Crawford's original ruling on the Eldritch Knight's War Magic feature, you choose when during your turn to take the bonus action shove because its timing isn't specified.

And if you look at the SA response, he states that allowing you to make the attack before casting the cantrip would probably not break anything. Clearly he felt it could break things if you allow the shove to precede the Attack action via Shield Master.

No, you'd still need to take the Attack action on your turn for your shove-attempt to use a bonus action.

And what if you can't? Then what? You have benefited from a bonus action granted by a feat and never met the requirements to gain said bonus action.

No, that isn't my reasoning at all. My reasoning is that the conditional language of the feat requires that the shield master's total activity for his/her turn is considered when fitting it into the action economy. This doesn't create a paradox because all that happens in a shove-first scenario is either that the shield master attempts to shove a creature, or that the shield master attempts to shove a creature and also takes the Attack action. There's nothing paradoxical about either of those situations.

Again, you are either not understanding that bonus actions require you to satisfy the conditions needed to have them or there are none. In this case, the requirement is the Attack action, if you never attack, you should NOT have gained the bonus action. You continue to believe you can use the bonus action first even though you could be denied your Attack action later on. If you want to ignore the requirements for Shield Master in situations when you have NO Attack action, that is up to you. Play on!

My point is that although the official interpretation is one of at least two valid interpretations, it goes against the stated RAI for bonus actions of this sort, and when given a choice, I'll take playing the game as it was intended to be played, rather than go with an unintended interpretation just because it's considered to be more literal. I can see why Jeremy Crawford might express such an interpretation, but I don't quite understand why anyone would want to play that way. On the other hand, as I said, it wouldn't be a deal-breaker.

Well, we disagree on that obviously. It is your opinion as to what was intended on how play was suppose to be, but obviously the lead game designer feels otherwise. Ignore him if you want.

I found this tweet from Jeremy dated March 10, 2016:

Does Shove qualify you to use the bonus attack in Two Weapon Fighting or Martial Arts? The shove and grapple options don't involve an attack with a weapon or an unarmed strike, so no.​

Well, a shove-attempt could satisfy the condition for using a bonus action to make another shove-attempt, but the point I was making is only tangentially related. The point I was making has to do with the idea of being committed to taking the Attack action. If you shove a creature, then you're either taking the Attack action to make that shove-attempt, or you're using a bonus action conditioned on taking the Attack action, so either way you're committed to taking the Attack action. The Eldritch Knight's War Magic feature doesn't have this quality because making a weapon attack in no way commits you to casting a cantrip.

Sure, but if you use your Attack action and your attack granted by that action to attempt to shove, that is using the Attack action and allows the bonus action to Shove from Shield Master. But notice you are in fact using your Attack action before you gain the bonus action. But you can't use the bonus action shove first to attempt to shove unless it is preceded by the Attack action.

I'm sorry, but if your goal here is to clear things up, you're barking up the wrong tree. I already understand the official ruling, and I disagree with it. No amount of explaining it to me is going to change that. I also disagree that it requires a house-rule to play out of accordance with the official ruling.

LOL from everything you say you obviously do NOT understand the official ruling, whether you agree with it or not. However we agree, no amount of me explaining that to you will change your understanding if you refuse to change it. Perhaps face-to-face we would be more successful at meeting a mutual resolution, but it obvious though posts the only resolution is agreeing to disagree. Play on!

 

[Asgorath]

...

I hope at this point no one is still arguing you can use the bonus action and THEN use the Attack action, that they need not be simultaneous or that the Attack action must come first. The only two logical interpretations IMO are:

1. The bonus action Shove can be taken simultaneously to the Attack action.
2. The bonus action Shove must be taken after the Attack action.

Either way, unless you have the Extra Attack feature, you can never benefit from the shove on the first attack made on your turn. Of course, as I have shown in other posts, this goes against JC's official stance from SA and tweets, but that is only of concern to your table if you value his position.

And we can simply let the PHB guide us to the correct answer here. The PHB talks about before and after. Movement can come before or after your action. Movement can come between weapon attacks, implying the attacks happen before and after the movement. The Ready action talks about the reaction happening after the trigger. At no point does the PHB talk about these elements happening simultaneously. Thus, the game elements themselves must be resolved discretely and in order.

The DM and players can turn this discrete sequence of game elements into a narrative as part of the ongoing story, but that doesn't change the fact that the underlying elements of the combat system must happen in a sequence. For example, the DM might narrate your attack doing no damage because the Wizard cast Shield as a reaction, and the blade of your weapon was stopped an inch short of their flesh by a magical barrier of force energy. That doesn't change the fact the attacker had to actually make an attack by performing the 3 steps listed in the PHB, the Wizard taking a reaction when the condition of "which you take when you are hit by an attack or targeted by the magic missile spell" became true, and the temporary effect of +5 AC turned the hit into a miss.

 

[Hriston]

Then again, you may not, which leads to paradox since the condition required to gain the bonus action was that you take the Attack action on your turn.

What paradox? If I take the Attack action on my turn, I have satisfied the condition for using a bonus action to shove a creature. If I don't take the Attack action, then I have merely shoved a creature (which consumes my action). What's paradoxical about that?

Again, the problem is you are employing a bonus action granted by a feature and didn't satisfy the requirements of the feat to gain the bonus action. That is breaking the rule of gaining bonus actions allowed by satisfying conditions to gain them.

Let's say I shove a creature. There's no reason to think I must necessarily use a bonus action to do so. So I'm not breaking any rules by shoving a creature. If I then go on to take the Attack action, well then I have satisfied the condition for making the shove-attempt using a bonus action. Otherwise, I didn't use a bonus action at all!

I never said it wasn't and that is immaterial to the argument.

You asked if the shove-attempt should remain unresolved because you can't use a bonus action to do it. That's a stark difference from how a shove-attempt is resolved, whether you use a bonus action or not.

Accept that to gain the bonus action shove you must take the Action. I have shown numerous examples of how it leads to paradox if you use your bonus action to shove before you attack because you can be denied the attack later on.

There's nothing paradoxical about being denied the ability to attack. It happens all the time in the game without anyone thinking it's paradoxical. The incongruity seems to arise from assigning a bonus action to the shove-attempt before the condition has been met for using one. I would recommend not doing that.

And if you look at the SA response, he states that allowing you to make the attack before casting the cantrip would probably not break anything. Clearly he felt it could break things if you allow the shove to precede the Attack action via Shield Master.

They're both SA, but only the original ruling expressed RAI. What evidence do you have for how Jeremy Crawford thinks allowing the shield master shove to come first affects game-balance? In the Sage Advice segment of the 2/1/19 Dragon Talk, he said decisions on the timing of bonus actions were made not for balance reasons, but for smooth game-play. Besides, considering how the Eldritch Knight's War Magic could potentially interact with Eldritch Strike if allowing the bonus action weapon attack to come first, I doubt he thinks allowing the shield master shove to come first is game-breaking if he doesn't think the same thing about War Magic.

And what if you can't? Then what? You have benefited from a bonus action granted by a feat and never met the requirements to gain said bonus action.

How exactly have you benefited, though? Without the feat, you can shove a creature, so that in itself isn't a benefit. No, the benefit of the feat is that you can shove a creature AND take the Attack action on the same turn, and if you can't take the Attack action for whatever reason, then you haven't benefited.

Again, you are either not understanding that bonus actions require you to satisfy the conditions needed to have them or there are none. In this case, the requirement is the Attack action, if you never attack, you should NOT have gained the bonus action. You continue to believe you can use the bonus action first even though you could be denied your Attack action later on. If you want to ignore the requirements for Shield Master in situations when you have NO Attack action, that is up to you. Play on!

I agree that if you never take the Attack action, then you can never use a bonus action to shove a creature. You can still shove a creature without using a bonus action, though, so there's that.

Well, we disagree on that obviously. It is your opinion as to what was intended on how play was suppose to be, but obviously the lead game designer feels otherwise. Ignore him if you want.

No, what was intended (RAI) has been stated by the very person you say feels otherwise. On July 6, 2015, he said, "The intent is that the bonus attack can come before or after the cantrip." That's a clear statement of the intent with which the game was designed. The more recent change in the official interpretation is motivated not by a desire to reveal the RAI, but rather to elevate a literalistic interpretation of the RAW over the RAI. A RAI interpretation is still possible with the existing language, though, so I can understand the decision not to issue errata for this. What I don’t like, however, is WotC’s tendency to then defend their uncorrected, ambiguous text by doubling down on the most literalistic interpretation possible.

Sure, but if you use your Attack action and your attack granted by that action to attempt to shove, that is using the Attack action and allows the bonus action to Shove from Shield Master. But notice you are in fact using your Attack action before you gain the bonus action. But you can't use the bonus action shove first to attempt to shove unless it is preceded by the Attack action.

That’s your interpretation. My interpretation accords with the feat’s intended lack of a timing specification.

LOL from everything you say you obviously do NOT understand the official ruling, whether you agree with it or not. However we agree, no amount of me explaining that to you will change your understanding if you refuse to change it. Perhaps face-to-face we would be more successful at meeting a mutual resolution, but it obvious though posts the only resolution is agreeing to disagree. Play on!

I’m curious what part of the official ruling you think I don’t understand? You seem to think that understanding it makes it impossible to disagree with.

 

[Asgorath]

What paradox? If I take the Attack action on my turn, I have satisfied the condition for using a bonus action to shove a creature. If I don't take the Attack action, then I have merely shoved a creature (which consumes my action). What's paradoxical about that?



Let's say I shove a creature. There's no reason to think I must necessarily use a bonus action to do so. So I'm not breaking any rules by shoving a creature. If I then go on to take the Attack action, well then I have satisfied the condition for making the shove-attempt using a bonus action. Otherwise, I didn't use a bonus action at all!

Can you show me where it says you can just shove a creature on the list of valid actions in the "Actions in Combat" section of the Combat chapter of the PHB? That section provides the rules for what you can do on your turn, I've read it a bunch of times and I haven't seen anything that says you can just shove someone. The rules do quite clearly say that you start your turn with movement and an action, and that a valid choice for an action would be the Attack action, and you can make a special melee attack to shove someone. Can you show me where it says you get to defer the decision about how you did something until the end of your turn?

 

[Mistwell]

 

[Arial Black]

The "time travel" comes in with the claim that if you use the bonus action first and then are prevented from ever taking the action, that somehow the action was taken before hand or during the bonus action anyway OR that the bonus action then becomes an action. The first is time travel, the second is Schrodinger's action, both of which are 100% unsupported in RAW.

As the examples of 'time-travel'-like rules show, seeming to time travel at the gaming table is not an issue at all! What matters is that no time travel occurs in the game world.

Actions, bonus actions, reactions, the Attack action, the Dash action, 6-second turns....all these are things the player does; the game mechanics at the gaming table.

But in the game world? There are no such things as 'bonus actions', or any other perception of the 5e game mechanics in play. The creatures in the game can have no idea that they are merely our avatars in a game.

In the game world, there are no such things as 'the Attack action' or 'bonus action shove'. No, in the game world there are just attacks, shoves, ripostes, shield bashes...

There are no 6-second turns in the game world. No 'indivisible Extra Attack actions', just a series of attacks and shield bashes. The world doesn't care in what order those two sword slashes and shield shove occurs in terms of where the shield shove is 'allowed' to be.

And, really, neither do the game mechanics. All the game mechanics require is that those attacks and that shield shove occur on the same turn.

 

[Maxperson]

I don't need a rule to change that because it's irrelevant to the condition set up by the feat. The condition for using a bonus action to shove a creature is not that "you have taken the Attack action on your turn," so it doesn't matter which of these two states prevail when you use the bonus action the feat gives you.

You do need a rule to change it. Why? Because there is no ability to declare actions in any way that has any mechanical meaning whatsoever. It simply does not exist by RAW. If the player tells you, "I am going to take an attack action n my turn," that statement is 100% informal and cannot trigger anything mechanical at all. The only way to know for the purposes of triggering the shove whether or not an Attack action is going to be taken, is to take it. Until then, because you cannot mechanically declare that you will be taking one, the trigger cannot happen.

 

[Arial Black]

Except that there is nothing to coincide with until after the first attack concludes. The Attack action does not begin until you hit step one of your first attack. Before step 1 of that first attack, no Attack action has even begun, let alone been taken. Once you have begun that first attack, you must resolve it before you do anything else, unless a special rule such as Sanctuary prevents you from doing so. Since no Attack action has even begun until the first attack starts and completes, there is no trigger for the bonus shove until that first attack completes.

IF the Attack action does not begin until you hit step 1 of your first attack, then it would perforce also be true that the bonus action shield shove does not begin until you hit step 1 of the shield bash.

But since they can be simultaneous (because "statements of causality require the antecedent to precede or coincide with the consequent in time") then neither begins until you hit step 1.

Fortunately, the rules provide for this possibility: if two things occur simultaneously, the acting creature chooses the order in which those things are resolved.