Clarification on Shield Master Feat

[Yakmala]

Please let me know if I'm interpreting the way Shield Master works correctly. According to the text of the Feat...

If you take the Attack action on your turn, you can use a Bonus Action to try to shove a creature within 5 feet of you with your shield.

As I interpret this, it means that you must take the Attack action first and complete it, including all attacks associated with the action, before you can take the Bonus Action.

So, for example, a Level 20 Fighter with four attacks per round and Shield Master must take the Attack action and complete all four attacks [as they are part of the Attack action] before they can attempt to shove as a Bonus Action. They cannot declare an attack, then do their Bonus Action shove, then follow up with their attacks at advantage.

I ask this because I've run into a lot of melee types, including strength rogues, who seem to think they can do the shove first, get advantage, and then unload with their smite/sneak attack/crit fishing or other damage booster after their opponent is down.

[stoutstien]

Are we talking AL or just a home game?

[Yakmala]

Are we talking AL or just a home game?

I run both home games and public AL. I know for home games the DM can interpret the rule as they see fit, but for this conversation, I'm concerned about the RAW.

The potential damage difference is huge, depending on how the rule is interpreted, especially if the advantage is being used to offset the disadvantage of multiple attacks with GWM and/or Smite.

[jas61292]

Per the most recent official clarification, RAW is that you must make all your attacks before shoving.

It was also arguably stated though that RAI, you just need to make one attack first, but that this was not what the rule actually say.

The idea you could shove before any attack was based on an old, now overturned ruling.

[stoutstien]

I run both home games and public AL. I know for home games the DM can interpret the rule as they see fit, but for this conversation, I'm concerned about the RAW.

The potential damage difference is huge, depending on how the rule is interpreted, especially if the advantage is being used to offset the disadvantage of multiple attacks with GWM and/or Smite.

They have stated that you must finish an attack action before the ba shove. Now saying that a fighter could attack action>ba shove>action surge>attack action.

All in all, its potential damage boost isn't that high due to how easy advantage is to get.
The Barbarian already has reckless
The paladin can use wrathful smite
And so on

[DMThac0]

Well, according to the SA Compendium, it is Attack action then Shield Master.

I'm wondering how many people would accept a nuance to this or if I'm stretching things a bit too thin:

A fighter with 4 attacks on their Attack Action can move between their attacks to hit another target.

Would it be allowed, by you, to grand the SM in between those attacks or, would you stay strict to the RAI/Errata and have them finish all four swings?

[stoutstien]

Well, according to the SA Compendium, it is Attack action then Shield Master.

I'm wondering how many people would accept a nuance to this or if I'm stretching things a bit too thin:

A fighter with 4 attacks on their Attack Action can move between their attacks to hit another target.

Would it be allowed, by you, to grand the SM in between those attacks or, would you stay strict to the RAI/Errata and have them finish all four swings?

It's a common ruling in my area to allow the shove as long as a single weapon attack is made with the attack action before.

[Laserlight]

It doesn't say "complete an Attack action, then shove", just "take".

There's an argument to be made thus: If you get two swings per Attack, and you take one swing but have yet to make the second, have you "taken" the Attack Action? If so, then why can't you bonus shove? If not, does that mean you can change your mind and cast a spell instead?

I rule it that you can Shove then Swing, because I can't see any plausible reason why you have to swing, then bash, and never the other way around. Particularly when you can also shove as an attack. "Shove Swing is fine, and Swing Swing Shove is fine, but Shove Swing Swing is impossible" is a bit hard to explain--even if it does appear to be RAW at the moment.

[NaughtyTiger]

Crawford last "official" tweets said you had to complete the attack action, cuz actions were indivisible (except when they aren't)

SA compendium said you have to take the attack action, not complete. Nothing in SA compendium explicitly stated that actions are indivisible.

Crawfords "unofficial" tweets said you can sheild bash after the first attack.


Any of the 3 interpretations (before, during, after) are legit per English language and the books.

Ultimately it comes down to how closely the DM follows the word of Crawford, and which version (official or unofficial).
wait, you are the DM, so you choose which is legal. just be explicit when a sheild master shows up at the table before the game starts.

[Keravath]

The relevant rules are the following:

Sage Advice:
"Shield Master: The Shield Master feat lets you shove someone as a bonus action if you take the Attack action. Can you take that bonus action before the Attack action? No.
The bonus action provided by the Shield Master feat has a pre-condition: that you take the Attack action on your turn. In-tending to take that action isn’t sufficient; you must actually take it before you can take the bonus action. During your turn, you do get to decide when to take the bonus action after you’ve taken the Attack action. This sort of if-then setup appears in many of the game’s rules. The “if” must be satisfied before the “then” comes into play."

Nothing in this sage advice says that you must COMPLETE the attack action before using the bonus action shove. Only that you earn the bonus action after taking the attack action.

Attack Action:
The most common action to take in combat is the Attack action, whether you are swinging a sword, firing an arrow from a bow, or brawling with your fists. With this action, you make one melee or ranged attack. See the "Making an Attack" section for the rules that govern attacks.
Certain features, such as the Extra Attack feature of the fighter, allow you to make more than one attack with this action.

The attack action only requires that you make ONE melee or ranged attack for you to be considered to have "taken the attack action".

Extra attack and similar features allow you to make additional attacks as part of the attack action.

Characters can do other things between their attacks that are made as part of "taking the attack action" on their turn.

"MOVING BETWEEN ATTACKS
If you take an action that includes more than one weapon attack, you can break up your movement even further by moving between those attacks."

A character who moves between attacks has STILL already taken the attack action. They just have not finished it or used additional attacks provided by the extra attack feature.

Bonus action timing:
"You choose when to take a bonus action during your turn, unless the bonus action's timing is specified,"

Once you have earned a bonus action, you can take it at ANY time, in the middle of movement, in the middle of another action, whenever. The character chooses when to take the bonus action.

Finally the shield master feat says the following:
"If you take the Attack action on your turn, you can use a bonus action to try to shove a creature within 5 feet of you with your shield."

This only says that you have to take the attack action and says NOTHING about having to finish it. Taking the attack action only requires making ONE melee or ranged attack. If you have extra attacks they can be taken any time after the first attack. Extra Attack is also enabled by taking the attack action (thus you can't get extra attack when using the cast a spell action to cast booming blade even though you make a melee weapon attack).

The bottom line is that both RAW and also consistent with sage advice, you need to make at least one attack thus take the attack action to enable the bonus action shove with shield master. After the first attack, you have still taken the attack action but may have movement and additional attacks still available.


Finally, although not a rules reference, JC would also consider this a reasonable interpretation.
https://twitter.com/JeremyECrawford/status/1105183657877135360

So ... making at least one attack enables the shield bash bonus action option.

[NaughtyTiger]

Finally, although not a rules reference, JC would also consider this a reasonable interpretation.
https://twitter.com/JeremyECrawford/status/1105183657877135360

which is hysterical cuz previously he said that was an illogical interpretation... (but i already stirred that hornet's nest)

[Keravath]

which is hysterical cuz previously he said that was an illogical interpretation... (but i already stirred that hornet's nest)

Lol :) ... everyone can change their minds.

And although it may not make the greatest sense ... it is consistent with the rules the way they are written and avoids the issue of someone using the shield bash and then being unable to take the attack action for whatever reason which then invalidates the shield bash since they didn't take the attack action.

[Yakmala]

Requiring at least one attack, but not all attacks, before taking the bonus action shove makes sense, particularly based on the precedent of allowing movement between attacks.

[Man_Over_Game]

Requiring at least one attack, but not all attacks, before taking the bonus action shove makes sense, particularly based on the precedent of allowing movement between attacks.

Try not to go too far down the rabbit hole. There's a discussion on this every other week, and us trolls grow weary.


Some people say it makes sense that you could Bonus Action before the Action, as long as you make the Attack Action in the turn.
Some people say it makes sense that you need to make only a single attack, as the Attack Action has been "confirmed" by that point, as well as using the timing that allows you to move between attacks as evidence for this conclusion (which is the same as your opinion).
Officially, you have to complete the Attack Action before you're allowed to spend the Bonus Action.



What you do from there is all DM fiat or obeying the RAW.

[noob]

Rewriting the feat to fit its title better and have it be about giving orders to shields (something like advantage to diplomacy to do so) would considerably reduce the headache people have with interpreting it.

[NaughtyTiger]

And although it may not make the greatest sense ... it is consistent with the rules the way they are written and avoids the issue of someone using the shield bash and then being unable to take the attack action for whatever reason which then invalidates the shield bash since they didn't take the attack action.

I have trouble seeing that case arise.

Shield bash is available as an attack normally.
So if you shield bash as BA, but something happens to block you taking the attack action, then you retcon that the shield bash was your attack action.



What you do from there is all DM fiat or obeying the RAW
but of course the disagreement is about which of the 3 cases is RAW.

[Keravath]

Try not to go too far down the rabbit hole. There's a discussion on this every other week, and us trolls grow weary.


Some people say it makes sense that you could Bonus Action before the Action, as long as you make the Attack Action in the turn.
Some people say it makes sense that you need to make only a single attack, as the Attack Action has been "confirmed" by that point, as well as using the timing that allows you to move between attacks as evidence for this conclusion (which is the same as your opinion).
Officially, you have to complete the Attack Action before you're allowed to spend the Bonus Action.



What you do from there is all DM fiat or obeying the RAW.

I'm just curious what the official source for having to complete the Attack Action is?

The only reference I can find is in the sage advice which says "During your turn, you do get to decide when to take the bonus action after you’ve taken the Attack action." It doesn't say you have to have completed the attack action. The rules consider you to have taken the attack action after you have made at least one attack.

Anyway, if there is a more definitive rules reference that indicates that you have to complete the attack action and any extra attacks before activating the bonus action granted by shield master I would be interested to know it.



P.S. From a logic perspective, the requirement to complete the attack action and all extra attacks before allowing the shield bash really doesn't make sense. A 4th level fighter with shield master can make one attack then shield bash, a 5th level fighter has to make two attacks (why would they need to make two when one level earlier they only needed to make one?) and an 11th level fighter would have to make three. It makes far more sense and is consistent with both the rules and sage advice to allow shield bash after making one attack.

[Greywander]

Requiring at least one attack, but not all attacks, before taking the bonus action shove makes sense, particularly based on the precedent of allowing movement between attacks.
This is probably the best compromise. I don't know how strict AL is when it comes to this sort of thing, but I'd imagine as long as the players are happy (e.g. no "rocks fall, everyone dies") and it doesn't go beyond your table (e.g. no "You all learn how to cast Wish, at-will!" or "Here's a bunch of legendary magic items!"), I wouldn't think it would be a problem.

I'd ask your players, first of all if any of them have Shield Master, and second if they're okay with running it that way. If any disagree, you can ask someone higher up at the event (if such a person exists) for guidelines on how to run it, or default to the current "official" rule stating that all attacks must be completed.

As Man_Over_Game said, this discussion comes up fairly regularly, and it's difficult to reach any kind of agreement. I subscribe to the actual RAW as allowing the BA shove before the Attack action, so long as you take the Attack action on the same turn. I'll allow that this may have never been RAI, but my stance is firmly that it is, in fact, what the RAW says. I also have a very poor opinion of Sage "Advice", and think it should remain just that: advice, not a ruling.

Of course, not everyone agrees with this interpretation, and you'd be playing with fire by running it that way in AL. As I said, one attack then BA shove is a good compromise and might fly in AL (especially since it was the official rule at one point).

[Man_Over_Game]

I'm just curious what the official source for having to complete the Attack Action is?

The only reference I can find is in the sage advice which says "During your turn, you do get to decide when to take the bonus action after you’ve taken the Attack action." It doesn't say you have to have completed the attack action. The rules consider you to have taken the attack action after you have made at least one attack.

Anyway, if there is a more definitive rules reference that indicates that you have to complete the attack action and any extra attacks before activating the bonus action granted by shield master I would be interested to know it.



P.S. From a logic perspective, the requirement to complete the attack action and all extra attacks before allowing the shield bash really doesn't make sense. A 4th level fighter with shield master can make one attack then shield bash, a 5th level fighter has to make two attacks (why would they need to make two when one level earlier they only needed to make one?) and an 11th level fighter would have to make three. It makes far more sense and is consistent with both the rules and sage advice to allow shield bash after making one attack.

The closest thing we have to "official", from the Lead Designer:


No general rule allows you to insert a bonus action between attacks in a single action. You can interrupt a multiple-attack action with a bonus action/reaction only if the trigger of the bonus action/reaction is an attack, rather than the action. (https://twitter.com/JeremyECrawford/status/995024061267767298)

As DM, I allow the bonus action of Shield Master to happen after you make at least one attack with the Attack action, since making one attack fulfills the action's basic definition (PH, 192). If you have Extra Attack, you decide which of the attacks the bonus action follows. (https://twitter.com/JeremyECrawford/status/1105183657877135360)

The simple by-the-book way (RAW) to determine whether you've completed an action is to finish the whole action. Yet you fulfill our design intent (RAI) with the Attack action if you make at least one attack with it, since that is how we define the action in its basic form. (https://twitter.com/JeremyECrawford/status/1105204044610428929)

[DMThac0]

I'm of the mind that at least one attack has to be made (even if it's followed by extra attacks) to trigger Shield Master, and you can break up the attacks because it's more cinematic.

However, I can see two different arguments that fit options 2 & 3 of Man Over Game's points.

Argument 1: You can break up the attacks because the precedent has been set that you can move between attacks and that Bonus Actions state you can choose when to take it. The first attack (even with extra attacks) is the if conditional required to trigger Shield Master.

Argument 2: You cannot break up the attacks because it's an embedded loop inside the if conditional required to trigger Shield Master.

[DarkKnightJin]

If a player has the Feat, I would ask them if they intend to do anything other than attack after shoving.

Most characters that end up having the feat are casters anyway. If they want to use Booming Blade, they cant bash someone in the face on that turn.
If they want to stab the possibly-Prone victim 2 or 3 times(the usual follow-up)? I'd allow it for RAF and cinematic sense. Rule of Cool, I suppose.
They seem hell-bent on abusing this? I'm going to adopt the JC 'ruling' of a single attack needing to be made to satisfy the *if* condition of the feat to allow the bonus action bash to the face.

Seems reasonable, I think.

[Prince Vine]

I'm just curious what the official source for having to complete the Attack Action is?

The only reference I can find is in the sage advice which says "During your turn, you do get to decide when to take the bonus action after you’ve taken the Attack action." It doesn't say you have to have completed the attack action. The rules consider you to have taken the attack action after you have made at least one attack.

Not an official answer, but if the philosophy minor part of me HAD to explain it I came up with something.

The attack action lets you make an attack. If you have the extra attack feature you may instead make two, three or four attacks instead. If you perform an action before taking all the attacks you could have you have functional just chosen not to take those extra and ended the action.

To make a crazy metaphor, you have a magically resealing time-locked cookie bag. Once you open it you can reach in and take as many as three cookies but once you pull out your hand, no matter how many cookies you took you cannot get back into that bag until it unseals again.

[Zhorn]

Regarding the single weapon attack vs full Attack Action, I'd side with a single weapon attack to lock into the qualifier of having taken the Attack Action, but inserting a Bonus Action in between weapon attacks does not 'end' the Attack Action for the same reason you can divide up your Movement and Attack Actions between each other without negating the remainder of either.

edit: obnoxious bolding to avoid confusion on what I said

[Chronos]

The way I see it:

You choose the Attack action on your turn. Having chosen that, you can now make whatever number of attacks your class features allow you during your turn.

You have now taken the Attack action. Having done so, you are now also allowed to take a bonus action to shield bash. Go ahead and do it now if you like, because as a bonus action, it can be at any point during your turn.

Now you've bashed your enemy, and (if the dice gods smiled on you) your enemy is now on the ground. You can now make the attack(s) that you're entitled to, as a result of taking the Attack action. The attacks may well be at advantage, if your shove was successful.

If it so happens that your first target is no longer a valid target for an attack (maybe your shove caused it to slip down a muddy hill away for you, or maybe it was flying and your shove forced it to the ground far away from you, or whatever), then you might try to attack something else, possibly using your movement first if there's nothing next to you. But you can't do anything other than attacking, because you've already taken your action to attack.

[MeeposFire]

Regarding the single attack vs full Attack Action, I'd side with a single attack to lock into the qualifier of having taken the Attack Action, but inserting a Bonus Action in between attacks does not 'end' the Attack action for the same reason you can divide up your Movement and Attack Actions between each other without negating the remainder of either.

I am not speaking specifically about shield master because nobody is ever happy with anything to do with that conversation but there are some other things here that I want to address.

5e does not have a single attack or full attack actions. 5e has the attack action and some classes get an ability called extra attacks which allows you to make more attacks using that attack action, it is still even with a 20th level fighter an attack action.

More importantly do not use the movement between attacks rule to make a ruling on this. The reason is that moving in between weapon attacks is a specific rules exception in the book. It is not meant to be extrapolated for other rules as it will lead you astray. If you could break up anything there would not be a need for that particular exception and as a classic example you can move between 4 strikes of your sword but your level 20 warlock cannot move between the 4 beams he uses from eldritch blasts even though they are attacks (they are not weapon attacks for instance and so the moving between attacks rule does not apply as an example).

[LudicSavant]

Shield Master is a mess; there are at least 3 distinct versions of the Sage Advice ruling on it which are all mutually exclusive. Pick whichever one you feel is most appropriate for your table.

[Zhorn]

I am not speaking specifically about shield master because nobody is ever happy with anything to do with that conversation but there are some other things here that I want to address.

5e does not have a single attack or full attack actions. 5e has the attack action and some classes get an ability called extra attacks which allows you to make more attacks using that attack action, it is still even with a 20th level fighter an attack action.

More importantly do not use the movement between attacks rule to make a ruling on this. The reason is that moving in between weapon attacks is a specific rules exception in the book. It is not meant to be extrapolated for other rules as it will lead you astray. If you could break up anything there would not be a need for that particular exception and as a classic example you can move between 4 strikes of your sword but your level 20 warlock cannot move between the 4 beams he uses from eldritch blasts even though they are attacks (they are not weapon attacks for instance and so the moving between attacks rule does not apply as an example).

???
I feel like you're trying to call me out on things I didn't say.

For starters I didn't mention Shield Master specifically, I'm talking about qualifiers, the thing an ability requires you to do before you can use it. It could be Shield Master, it could be Polearm Master, or anything else with the paraphrased wording of "when you do x, you can then do y"

Secondly, I thought the capitalisation would be enough of a distinction, but I guess not. I'm aware 5e doesn't have a Full Attack Action, or a Single Attack Action, hence writing it as
Regarding the single attack vs full Attack Action

Third, apples to oranges with comparing Eldritch Blast. that's a single spell, is not taken with an Attack Action, not four distinct spells, and does not compare to four distinct weapon attacks. We both agree you cannot move between casting each beam, and I do get what you're trying to say. But again, I didn't imply that scenario. That extrapolation is all you.

I'm sorry if this was just a simple misread of what I wrote, and I'll go back an bold for emphasis some things if that helps to make what I said clearer. But I don't get these accusations it looks like you're saying about what I wrote. :smallconfused:

[Talionis]

The RAW says you have to take an attack action during the turn so you can order the actions how you see fit as player.

It was clarified in a tweet that you must attack first at least once.

I forget what rules AL looks at but RAW is still what is written unless there is errata.

[RSP]

RAW, outside of official SA, you can Shield bash “before” the Attack Action, assuming that’s how your DM interprets your character’s (lower case) actions that turn.

Whether in combat or not, the basis of the game is:

1. The DM describes the environment.
2. The players describe what they want to do. (Though “the DM decides what happens, often relying on the roll of a die to determine the results of an action.”)
3. The DM narrates the results of their actions.

So, essentially, if a Player describes their (lower case) actions in a way that leads the DM to determine the Attack Action is included in the description of (lower case) actions, then, strict RAW, you’ve fulfilled the requirements of the Feat.

For whatever reason, when the designers answer questions on mechanics, they seem to forget this is the basis of the game, per the RAW.

Obviously, this answer comes amazingly close to “ask your DM,” which is the answer that matters, however, I wanted to point out that strict RAW does justify the use of the BA Shove prior to initiating any of the mechanics of the Attack Action.

[ChildofLuthic]

This is one of those debates that everyone has their own thoughts on. The bonus action shove triggers on the attack action. Some will say you make all your attacks before you shove, since they're all part of the attack action. Others will say you only need to make one attack before you shove, since the extra attacks are just something you get after you take the attack action. But I don't think there's any kind of consensus.

[NaughtyTiger]

This is one of those debates that everyone has their own thoughts on. The bonus action shove triggers on the attack action. Some will say you make all your attacks before you shove, since they're all part of the attack action. Others will say you only need to make one attack before you shove, since the extra attacks are just something you get after you take the attack action. But I don't think there's any kind of consensus.

agreed. The Common English used isn't precise enough to differentiate between the 3 cases. Moreover, poor editing throughout all of the rulebooks has made it difficult to nail down.

The problem is that we see our interpretation in the RAW; therefore, every other interpretation is "illogical" (ie Crawford plays the way he called "illogical")

[Keravath]

Two quick comments.

1) the argument all starts with the wording of the shield master feat. You can use your shield to bash IF you use the attack action on your turn.

The problem as noted in the sage advice is that there are a lot of circumstances that can make it impossible for a character to take the attack action. If a character is prevented from taking the attack action then they do NOT get to bash as a bonus action.

There are three solutions to this
- require the character to actually take the attack action by making at least one (some say all) attacks before the bonus action is available
- if a character loses their option to take the attack action then just retcon the character into having used the attack action to make the original shield bash. However, this could affect any actions that had been readied based on the character using their attack. Some folks also have philosophical issues with retcon when trying to adapt to a rule.
- retcon the shield bash as just not happening and restart the players turn or house rule that shield bash doesn’t require the attack action.

2) As for bonus actions and reactions, they can be taken at any time .. including during the attack action.
If a character with multiple attacks makes one attack they could then move away .. if they are a wizard they could cast shield to block an op attack, if they are a rogue they could use the disengage bonus action then move and make another attack.

[NaughtyTiger]

The problem as noted in the sage advice is that there are a lot of circumstances that can make it impossible for a character to take the attack action. If a character is prevented from taking the attack action then they do NOT get to bash as a bonus action.

There are three solutions to this
- require the character to actually take the attack action by making at least one (some say all) attacks before the bonus action is available
- if a character loses their option to take the attack action then just retcon the character into having used the attack action to make the original shield bash. However, this could affect any actions that had been readied based on the character using their attack. Some folks also have philosophical issues with retcon when trying to adapt to a rule.
- retcon the shield bash as just not happening and restart the players turn or house rule that shield bash doesn’t require the attack action.


The "using bonus action attack at the start of your turn, then making it be your full action attack instead" (not quoting you) isn't a retcon. You are allowed to make a single sheild bash as your entire attack action. The DM doesn't have to undo any part of the narrative. Whatever happened to prevent the rest of the attacks, just happens.

I haven't found a situation that actually prevents you from taking an attack if the first attack is with a shield. If someone could layout an example where that happens, then I could maybe understand.

[LudicSavant]

Regarding the "retcon" thing, note that there are already examples of bonus actions being triggered by actions that occur after them. See Tempestuous Magic in XGtE:


TEMPESTUOUS MAGIC
Starting at 1st level, you can use a bonus action on your turn to cause whirling gusts of elemental air to briefly surround you, immediately before or after you cast a spell of 1st level or higher. Doing so allows you to fly up to 10 feet without provoking opportunity attacks.

So clearly the system is already intended to be able to handle bonus actions being triggered by future events.

This should be unsurprising, as the devs have commented repeatedly that abilities like War Magic were originally intended to be in either order (prior to JC's ruling change). And even after JC's ruling change, the Sage Advice Compendium still emphasizes that DMs would break nothing in the system by allowing players to reverse the order.


Does the “when” in the Eldritch Knight’s War Magic fea-ture mean the bonus attack comes after you cast the can-trip, or can it come before? The bonus action comes after the cantrip, since using your action to cast a cantrip is what gives you the ability to make the weapon attack as a bonus action. That said, a DM would break nothing in the system by allowing an Eldritch Knight to reverse the order of the cantrip and the weapon attack.

Note also that this is a change. Older versions of the Sage Advice Compendium simply say that it can be done in either order.

[BurgerBeast]

Per the most recent official clarification, RAW is that you must make all your attacks before shoving.

It was also arguably stated though that RAI, you just need to make one attack first, but that this was not what the rule actually say.

The idea you could shove before any attack was based on an old, now overturned ruling.

There seems to always be this underlying source of conflation in these sorts of discussions. If we're talking about Sage Advice, then it is not true that: "RAW is that you must make all your attacks before shoving."

Sage Advice is RAI, as interpreted, or as interpreted to be intended by the designers, by JC.

If we are talking about errata, then it is true that: "RAW is that you must make all your attacks before shoving."

But I'm not aware of the errata.

My personal interpretation and ruling method is that players can declare actions, and then the character can fail to execute the action, and this still constitutes "taking the action" because the character wasted their action on the action.

For example, you can attempt to cast a spell, and select an invalid target. I rule that you did take the cast a spell action, but you did not cast a spell.

So, at my table, you can take the attack action (i.e. commit to spending your action on Attack), but then never actually attack. On this basis, once you choose to attack (i.e. commit to spending your action on Attack), you can then choose to shield bash first. Your action is then spent on Attack whether you are able to actually make any attacks or not.

Just as an example: What if you declare an attack, but only complete two attacks when you're capable of three? Have you therefore not taken the attack action? - I think most of us would agree that you have taken the attack action despite not making the third.

[MeeposFire]

???
I feel like you're trying to call me out on things I didn't say.

For starters I didn't mention Shield Master specifically, I'm talking about qualifiers, the thing an ability requires you to do before you can use it. It could be Shield Master, it could be Polearm Master, or anything else with the paraphrased wording of "when you do x, you can then do y"

Secondly, I thought the capitalisation would be enough of a distinction, but I guess not. I'm aware 5e doesn't have a Full Attack Action, or a Single Attack Action, hence writing it as

Third, apples to oranges with comparing Eldritch Blast. that's a single spell, is not taken with an Attack Action, not four distinct spells, and does not compare to four distinct weapon attacks. We both agree you cannot move between casting each beam, and I do get what you're trying to say. But again, I didn't imply that scenario. That extrapolation is all you.

I'm sorry if this was just a simple misread of what I wrote, and I'll go back an bold for emphasis some things if that helps to make what I said clearer. But I don't get these accusations it looks like you're saying about what I wrote. :smallconfused:

They actually compare very well. Both are a single action that gives you 1 or more different attacks. A level 20 fighter can use eldritch blast as an action and makes 4 spell attack rolls or he could use his action to use the attack action in combination with his extra attack class feature he can make 4 weapon attacks. I find it curious that you are saying "distinct" weapon attacks as if it is any different that how eldritch blast works. They are both one action and then you make one or more attacks each eldritch blast bolt is just as distinct as each weapon attack is from a single attack action.

As a further example that being a spell itself has nothing to do with anything unless the rules have been changed in a later printing you could have a cantrip called "weapon strikes" which is an action that allows your caster to make a weapon attack and then make an additional one at level 5, 11, and 17. In that case the caster could move between each weapon attack when using that cantrip because they are weapon attacks (the fact they come from a cantrip makes no difference in the rules). That same caster could not move between eldritch blast attacks. The only important difference is whether you are making weapon attacks not what the original actions are.

That is why I am saying it is not a good general example because it is itself a very specific rules exception that only works on a very narrow set of circumstance (multiple attack roles when using a weapon with a single action) and does not even say you can break up those attacks with anything except movement (though that aspect does not seem to come up in these discussions).

[Aimeryan]

They actually compare very well. Both are a single action that gives you 1 or more different attacks. A level 20 fighter can use eldritch blast as an action and makes 4 spell attack rolls or he could use his action to use the attack action in combination with his extra attack class feature he can make 4 weapon attacks. I find it curious that you are saying "distinct" weapon attacks as if it is any different that how eldritch blast works. They are both one action and then you make one or more attacks each eldritch blast bolt is just as distinct as each weapon attack is from a single attack action.

As a further example that being a spell itself has nothing to do with anything unless the rules have been changed in a later printing you could have a cantrip called "weapon strikes" which is an action that allows your caster to make a weapon attack and then make an additional one at level 5, 11, and 17. In that case the caster could move between each weapon attack when using that cantrip because they are weapon attacks (the fact they come from a cantrip makes no difference in the rules). That same caster could not move between eldritch blast attacks. The only important difference is whether you are making weapon attacks not what the original actions are.

That is why I am saying it is not a good general example because it is itself a very specific rules exception that only works on a very narrow set of circumstance (multiple attack roles when using a weapon with a single action) and does not even say you can break up those attacks with anything except movement (though that aspect does not seem to come up in these discussions).

The lore and mechanics of casting a spell is quite different to that of making attacks with a weapon. In some situations a spell allows you to overlap the two (Booming Blade, et al.), however, those are specific situations.

Example of differences; Eldritch Blast - the spell takes time to cast, then the spell activates and resolves instantly (Duration: Instantaneous). There is no time to fit anything else in between an instantaneous duration.

[Tanarii]

Makes sense to me it should be after the first attack in an Attack Action for a character with Extra Attack. My "logic" is the first attack qualifies any character for having taken the Action, Extra Attack just adds the option of further attacks. Also that Extra Attack shouldn't make SM harder to use.

[MeeposFire]

The lore and mechanics of casting a spell is quite different to that of making attacks with a weapon. In some situations a spell allows you to overlap the two (Booming Blade, et al.), however, those are specific situations.

Example of differences; Eldritch Blast - the spell takes time to cast, then the spell activates and resolves instantly (Duration: Instantaneous). There is no time to fit anything else in between an instantaneous duration.

In my hypothetical cantrip "weapon strike" it would be duration instantaneous and yet you still could move between the weapon attacks because of the rules. If booming blade made multiple weapon attacks you could move between them (but it only has one so it does not come up).

If we had a spell with a duration of 10 rounds and every round it lets you make up to four spell attacks of force damage as a single action you still could not move between the spell attacks and that spell would not have an instantaneous duration. Duration is not actually an important distinction for this. Now if that same spell allowed you to make weapon attacks with beams of force then you would be able to move between each attack.

If the specific rule saying you can move between attacks did not exist then you would not be able to move between weapon attacks even though it does not have a stated duration. This is how it was before 5e (I frankly prefer how 5e does it and would port it back into other versions of D&D that could use it).

[NaughtyTiger]

The lore and mechanics of casting a spell is quite different to that of making attacks with a weapon. In some situations a spell allows you to overlap the two (Booming Blade, et al.), however, those are specific situations.

Example of differences; Eldritch Blast - the spell takes time to cast, then the spell activates and resolves instantly (Duration: Instantaneous). There is no time to fit anything else in between an instantaneous duration.

you can fit a shield spell, and a counterspell, (act upon a readied action for the start of a spell), move an abjurers ward, and deflect an attack to a nearby enemy. (resolves includes determining the effects of a spell)

[Aimeryan]

you can fit a shield spell, and a counterspell, (act upon a readied action for the start of a spell), move an abjurers ward, and deflect an attack to a nearby enemy. (resolves includes determining the effects of a spell)

Other than the Shield spell, all of those come into affect during the casting (which has a time component), not the resolution (which is instantaneous).

The Shield spell is just messed up however you look at it; you can wait until you are hit to use the spell to retro that you were not hit. Lore-wise, probably the Shield is cast before the attack (whether physical or magical), in anticipation.

~~~


Makes sense to me it should be after the first attack in an Attack Action for a character with Extra Attack. My "logic" is the first attack qualifies any character for having taken the Action, Extra Attack just adds the option of further attacks. Also that Extra Attack shouldn't make SM harder to use.

That would be the most reasonable reading both by RAW and RAI, as I also see it. A part of me rails at the requirement for the Attack action at all, though, especially given that by RAW the Attack action and the Bonus action need not even target the same creature (indeed, they could be 30ft away from each other).

[NaughtyTiger]

Other than the Shield spell, all of those come into affect during the casting (which has a time component), not the resolution (which is instantaneous).


are you saying that you have to declare all of your targets for eldritch blast and determine if each hits before you roll damage?

cuz if you fire off 3 EB blasts, you can fire them at Bob until he drops, and then target the next guy..... unless you have to finish casting BEFORE the resolution. I haven't seen anything in PHB describing that.

[DarkKnightJin]

are you saying that you have to declare all of your targets for eldritch blast and determine if each hits before you roll damage?

cuz if you fire off 3 EB blasts, you can fire them at Bob until he drops, and then target the next guy..... unless you have to finish casting BEFORE the resolution. I haven't seen anything in PHB describing that.

I have always seen EB ruled as calling out the (intended) target (s) before even rolling to hit.
Which kinda makes sense. You set forth 1-4 beams of energy, and they gonfor their target at the same time. Like a stronger but less accurate version of Magic Missile, basically.

[Tanarii]

I have always seen EB ruled as calling out the (intended) target (s) before even rolling to hit.
Which kinda makes sense. You set forth 1-4 beams of energy, and they gonfor their target at the same time. Like a stronger but less accurate version of Magic Missile, basically.
I agree it seems to make a kind of sense at first glance. But everyone online seems to think it should work the other way. Well, everyone except one now. :smallamused:

[Zhorn]

I find it curious that you are saying "distinct" weapon attacks as if it is any different that how eldritch blast works. They are both one action and then you make one or more attacks each eldritch blast bolt is just as distinct as each weapon attack is from a single attack action.
I say distinct because while each attack granted by the Extra Attack feature is covered under the umbrella of single Attack Action, they are individual actions(lower case). Swing the sword once for each attack, draw the bowstring once per each attack, throw a javelin once per each attack. this is the justification for

Moving Between Attacks
If you take an action that includes more than one weapon attack, you can break up your movement even further by moving between those attacks.
The weapon attacks do not happen simultaneously, they are sequential.
As others have noted, weapon attacks are not the same as spell attacks. Sure there are exceptions in the form of Booming Blade and Green-Flame Blade, but as they each specify "a weapon attack" (singular) they are not going to bring any nuance to the scenario in discussion.
Eldritch Blast being a spell oppereates under a different set of logic. The casting of the spell take an Action, but the beams it produces are all instantaneous and simultaneous, with no time in between to move.

"No general rule allows you to move between the attacks of a spell."
https://twitter.com/JeremyECrawford/status/811282788968185857
And asking "Are Eldritch Blast bolts simultaneous or sequential? Pick targets when cast or as each bolt is resolved?"

"pick when cast, resolve all at once"
https://twitter.com/mikemearls/status/630903081530560514
Eldritch Blast and similar spells are less like this
http://i.imgur.com/yLIJwtH.gif
http://i.imgur.com/yLIJwtH.gif

and instead more like this
https://media.giphy.com/media/DoDUpwqevHrKo/giphy.gif
https://media.giphy.com/media/DoDUpwqevHrKo/giphy.gif
If you have a reference to spell or official ruling to the contrary, I'm happy to consider it. But until then, I maintain the view that casting spells is not comparable to the Attack Action when moving between attack rolls.

As for

As a further example that being a spell itself has nothing to do with anything unless the rules have been changed in a later printing you could have a cantrip called "weapon strikes" which is an action that allows your caster to make a weapon attack and then make an additional one at level 5, 11, and 17. In that case the caster could move between each weapon attack when using that cantrip because they are weapon attacks (the fact they come from a cantrip makes no difference in the rules). That same caster could not move between eldritch blast attacks. The only important difference is whether you are making weapon attacks not what the original actions are.
In my hypothetical cantrip "weapon strike" it would be duration instantaneous and yet you still could move between the weapon attacks because of the rules.Is this "weapon strikes" spell a homebrew? Because I'm not finding a reference for it online. If there's an official source for it, I'd love to have a read of it directly and see what rulings there were for its interactions. But if it is a homebrew (as I suspect it may be), it's not a solid grounding to support your argument on when discussing official rules interpretations. If it was from an earlier edition, it would be worth the time as a talking point, but still not relevant to a 5e discussion.

[NaughtyTiger]

I have always seen EB ruled as calling out the (intended) target (s) before even rolling to hit.
Which kinda makes sense. You set forth 1-4 beams of energy, and they gonfor their target at the same time. Like a stronger but less accurate version of Magic Missile, basically.

Are you referencing PHB or even designers intent? Or just a logical DM ruling?

[Willie the Duck]

there are already examples of bonus actions being triggered by actions that occur after them. See Tempestuous Magic in XGtE:
<Ex.>

So clearly the system is already intended to be able to handle bonus actions being triggered by future events.


Indeed, and it would be nice for that to be how sage advice should handle it. Likewise, they could have simply reworded the feat to be unambiguous, and put that in the SA compendium (and presumably future PHB printings).

I my campaigns, I have rewording SM and TM to be allowing the user to spend their bonus action to grant their rider effects as part of the primary action.

[Keravath]

The "using bonus action attack at the start of your turn, then making it be your full action attack instead" (not quoting you) isn't a retcon. You are allowed to make a single sheild bash as your entire attack action. The DM doesn't have to undo any part of the narrative. Whatever happened to prevent the rest of the attacks, just happens.

I haven't found a situation that actually prevents you from taking an attack if the first attack is with a shield. If someone could layout an example where that happens, then I could maybe understand.

Most of the situations where you couldn't take the attack action might involve movement or held actions by opponents.

These scenarios would all start with the character using a shield bash to start with.

1) Character knocks target prone and moves away to attack another target. Runs into a trap and dies/knocked unconscious/held (pit trap/glyph of warding etc). They can't take the attack action so wouldn't have a bonus action shield bash.

2) Character uses shield bash and a held spell (hold person) is triggered by the character making an attack. The character fails the save, is held and can't take the attack action.

3) Character uses shield bash and knocks opponent prone, runs over to attack what they think are additional creatures only to discover that they are all illusions. They don't have any valid targets for an attack and can't take the attack action.

This is D&D, tons of strange things can and do happen :) ... most of the time the character would be able to take a bonus action shield bash followed by an attack action but there are lots of circumstances that could prevent that from happening.

Which is where we get back to the whole argument about the wording of the rules ... does the character expend a bonus action or an action to use the shield bash?

The character is supposed to decide this when they resolve the interaction. If they want to use an action to bash then I don't think anyone would object, they then have movement and a bonus action to spend the rest of the turn.

On the other hand, if they spend a bonus action and then can't take an attack action then the bonus action shield bash wasn't a valid choice. However, although the narration would not change in terms of the activities of the character it is still a mechanical retcon to change that the character actually spent an action using shield bash rather than a bonus action. One way to avoid this inconsistency is to ask that the character take the attack action first which means making at least one attack roll. Does it make much logical sense? Nope ... not to me :) ... but it does avoid having to rewrite the player choices afterwards to be consistent with whether they could use a bonus action or action to shield bash.

Anyway, the bottom line is that it is a DM call on how they want to run it and both interpretations are likely possible in terms of RAW as long as the DM wants to adjudicate what happens in the edge cases when an attack action isn't possible. Some folks prefer one solution, some another.

The only D&D gaming venue where RAW really matters is Adventurers League where folks try to use the same set of rules for a lot of different games. AL will typically leave the decisions in the DMs hands but it would be nice if a couple of the more contentious ones could be resolved with a AL DMG clarification as to what interpretation should be used for AL (however, I can understand the generally hands off policy in AL since enabling the DMs is a very important factor to consider)

[NaughtyTiger]

1) Character knocks target prone and moves away to attack another target. Runs into a trap and dies/knocked unconscious/held (pit trap/glyph of warding etc). They can't take the attack action so wouldn't have a bonus action shield bash.
except that a character can make a sheild bash as their attack action. so they took the attack action for shield bash, and didn't use their bonus action. no change to the narrative.



2) Character uses shield bash and a held spell (hold person) is triggered by the character making an attack. The character fails the save, is held and can't take the attack action.

Readied action is not the Attack Action, so the character cannot cast hold person AND shield bash.
Sheild bash is an attack, so other wizard could still pop hold person in response to the character making an attack. no change to the narrative


3) Character uses shield bash and knocks opponent prone, runs over to attack what they think are additional creatures only to discover that they are all illusions. They don't have any valid targets for an attack and can't take the attack action.
except that a character can make a sheild bash as their attack action. so they took the attack action for shield bash, and didn't use their bonus action. no change to the narrative.



However, although the narration would not change in terms of the activities of the character it is still a mechanical retcon to change that the character actually spent an action using shield bash rather than a bonus action.
less of a mechanical retcon than Shield spell in practice. (rarely do DMs say, "I rolled a 15 on the die" and pause for the character to say I pop shield, then say "does an 18 hit?". Usually they say "I rolled an 18" and the player interrupts to say I pop shield).
i find it hard to believe that you require the player to say "I attack the dwarf, move 10 ft, 2 attacks on the wight, move 20ft" before a single die is rolled for the turn, then hold the player to that decision.

[Xetheral]

However, although the narration would not change in terms of the activities of the character it is still a mechanical retcon to change that the character actually spent an action using shield bash rather than a bonus action.

I'm not sure that a "mechanical retcon" makes sense as a concept--it seems like an oxymoron to me. As I understand the term, only a change to the fiction itself would ever qualify as a "retcon".

More broadly, I understand why retroactively changing the fiction would be problematic (it lessens verisimilitude and challenges suspension of disbelief), but I don't see why retroactively changing the mechanical abstraction would have a similar negative impact. Why do you consider it problematic?

Even more broadly, if a character makes a single shield bash and does nothing else on their turn (e.g. due to be becoming incapacitated mid-turn) why does it even matter if the character used an action or a bonus action to make that shield bash? The end result is identical and can't affect anything else.

[Tanarii]

Are you referencing PHB or even designers intent? Or just a logical DM ruling?
Designers intent has been made clear by SA hasnt it? Eldritch Blast targets are selected and rolled sequentially, one after the other.

[NaughtyTiger]

Designers intent has been made clear by SA hasnt it? Eldritch Blast targets are selected and rolled sequentially, one after the other.

RAI is multiple spell attacks from the same spell are not simultaneous (general).
https://twitter.com/JeremyECrawford/status/614588258404597760

Therefore, Aimeryan's contention that the spell resolution is indivisible is not RAI.
That the spell casting is not (necessarily) complete before some effects are resolved even for instantaneous duration. (per RAI)
Thus the weapon attack vs spell attack analogy is stronger than Aimeryan considered.

Does that counter your earlier agreement with DarkNightJim?

[Aimeryan]

RAI is multiple spell attacks from the same spell are not simultaneous (general).
https://twitter.com/JeremyECrawford/status/614588258404597760

Therefore, Aimeryan's contention that the spell resolution is indivisible is not RAI.
That the spell casting is not (necessarily) complete before some effects are resolved even for instantaneous duration. (per RAI)
Thus the weapon attack vs spell attack analogy is stronger than Aimeryan considered.

Does that counter your earlier agreement with DarkNightJim?

RAI might indeed be that there is time between Eldritch Blast attacks to make decisions about who the next blast hits, it just doesn't match the RAW duration of Instantaneous. If this was the intention then the Duration should have been listed as Special (See Text). Depending on how the text was then wrote there may be grounds for other actions/movement between the attacks.

[NaughtyTiger]

RAI might indeed be that there is time between Eldritch Blast attacks to make decisions about who the next blast hits, it just doesn't match the RAW duration of instantaneous. If they wanted the spell to last until the end of the turn, upon which at any time you may make an attack if you have any left, they should have listed the Duration as Special.

the question I actually asked you:

are you saying that you have to declare all of your targets for eldritch blast and determine if each hits before you roll damage?
your answer is "yes, must declare targets and determine hit before you roll any damage".
i can accept that as the most correct interpretation, and my way is houserule.

and as far as "if they wanted"... i agree RAI is a crapshoot, because they revise their intent on a whim.

[MeeposFire]

I say distinct because while each attack granted by the Extra Attack feature is covered under the umbrella of single Attack Action, they are individual actions(lower case). Swing the sword once for each attack, draw the bowstring once per each attack, throw a javelin once per each attack. this is the justification for

The weapon attacks do not happen simultaneously, they are sequential.
As others have noted, weapon attacks are not the same as spell attacks. Sure there are exceptions in the form of Booming Blade and Green-Flame Blade, but as they each specify "a weapon attack" (singular) they are not going to bring any nuance to the scenario in discussion.
Eldritch Blast being a spell oppereates under a different set of logic. The casting of the spell take an Action, but the beams it produces are all instantaneous and simultaneous, with no time in between to move.

And asking "Are Eldritch Blast bolts simultaneous or sequential? Pick targets when cast or as each bolt is resolved?"

Eldritch Blast and similar spells are less like this
http://i.imgur.com/yLIJwtH.gif
http://i.imgur.com/yLIJwtH.gif

and instead more like this
https://media.giphy.com/media/DoDUpwqevHrKo/giphy.gif
https://media.giphy.com/media/DoDUpwqevHrKo/giphy.gif
If you have a reference to spell or official ruling to the contrary, I'm happy to consider it. But until then, I maintain the view that casting spells is not comparable to the Attack Action when moving between attack rolls.

As for
Is this "weapon strikes" spell a homebrew? Because I'm not finding a reference for it online. If there's an official source for it, I'd love to have a read of it directly and see what rulings there were for its interactions. But if it is a homebrew (as I suspect it may be), it's not a solid grounding to support your argument on when discussing official rules interpretations. If it was from an earlier edition, it would be worth the time as a talking point, but still not relevant to a 5e discussion.


I have a feeling this conversation will sadly stay derailed on a discussion on spell durations but I am going to try for the moment anyway and see if we can get on the same page for this discussion. Whether you are right or wrong about the point with spell durations is immaterial. It does not make an actual difference in my actual point (remember I am not arguing that EB should allow for moving between blasts so we do not need alternative reasons for why EB can not have movement in between beams).

We need to go back. What was my initial point and the reason why I brought it up.

Initially you reasoned that you believe you can put the bonus action shove in between your attacks from using your attack action specifically because you can break up your move between making your weapon attacks (if I am misrepresenting your position on this in a significant way you can correct me here but this is how I essentially see your position).

My position was that using the moving between weapon attacks as a guide for how the general rules work is flawed because the ability to move between multiple attacks from a single action is a specific rules exception and not a general rule. Moving between weapon attacks is a corner case and a very specific exception that only applies in a very narrow area. To me the fact that it is such a narrow case that the designers had to specifically make an exception for you to do actually undermines your position. If you are generally able to break up a single action with multiple parts with your movement then the weapon attack rule would not be needed, but since it does exist the implication is that the general rule must be that you cannot break up a multipart single action with your movement. So in my view your reasoning actually undermines your position hence why I thought it was not a good example for you.

Now the reason I brought up eldritch blast is because it is very similar mechanically as using the attack action with extra attack. Now for the purpose of staying on task and not getting sidetracked on a tangent let us ignore the talk about spell duration for the time being since it really is not important here (alternatively you could just use an example of a spell that can make multiple attack rolls using one action that does not have a duration of instantaneous whichever works best for you) though if it helps you can choose to think of spell duration as an additional reason for why eldritch blast beams can not be broken up with movement.

So can I move between the multiple spell attacks from a single action from a spell? No and the reason is that the rules do not allow for it. The movement between attacks rule only applies to multiple weapon attacks from a single action.

Now if you take the same spell but instead of spell attacks it had you make weapon attacks you could move between the attacks because they would be multiple weapon attacks from a single action.

As for an actual example since we are all looking for one we can use swift quiver. That spell allows you to make two weapon attacks as a bonus action and it has a duration that will not interfere with the conversation.

Can you move between the two attacks swift quiver attacks?

No.

The reason is that swift quiver is multiple weapon attacks made with a bonus action. The moving between attacks rule only applies to when you use an action. Swift quiver does not require an attack action either so we cannot even claim that the bonus action adds on to the attack action either. Since the specific exception for movement between weapon attacks only applies to actions that means it does not apply to swift quiver which means you cannot break up those attacks (well at least with movement anyway). The only argument I can think of around this is if you believe that the word "action" used in the movement between weapon attacks rule is not being used for the rule term "action" but rather a more general term of doing something but I think that is a rather weak argument considering that the part of the book we are talking about is talking extensively on the game term of "action".


The way I see it either the rule is such a specific exception that it is useless as a basis outside of its direct context or if not then it actively works against you because the implication of that rule is that by default you cannot break up multiple attacks with a single action with movement because weapon attacks get a special exception to do it.

Another reason that it does not work as a basis for a ruling is that you can really only say for sure that weapon attacks can be broken by movement with that rule but it says nothing about anything else. That is why when I brought up swift quiver I made that quip about not being able to break up the two attacks with movement. That is also why despite all of this I am not pushing any particular ruling on the shield shove itself but rather contesting this particular reasoning behind your ruling.

[Tanarii]

Therefore, Aimeryan's contention that the spell resolution is indivisible is not RAI.
That the spell casting is not (necessarily) complete before some effects are resolved even for instantaneous duration. (per RAI)
Thus the weapon attack vs spell attack analogy is stronger than Aimeryan considered.
Doesn't necessarily follow. It may be casting completed (v,s,m etc), then all effects occur with resolution after casting is completed.

Also non simultaneous resolution doesnt necessarily mean non instantaneous. I mean, it would if we were talking basic newtonian physics or something. But we're talking rules mechanics.

Regardless, RAI is clear and it works perfectly fine within RAW: EB and other such spells are sequential target then attack.

Edit: also i wasnt agreeing earlier. I was disagreeing, by saying first glance may make one think that its choose all targets first.

[NaughtyTiger]

Edit: also i wasnt agreeing earlier. I was disagreeing, by saying first glance may make one think that its choose all targets first.
thanks for clarifying, i wasn't sure.

still thinking about the other parts.

[Yakmala]

I believe someone brought this up earlier, but could [to give an example] a Fighter/Rogue multi-class that has multiple attacks and cunning action take the attack action, make their first attack, use their bonus action to disengage, move to another target and then use their second attack?

If so, this is a practical example of a bonus action occurring between attacks during an action.

[Tanarii]

still thinking about the other parts.
I generally agree that it's easier to think of resolution as taking place sequentially in-game as well as mechanically. I'm just pointing out that sequential mechanical resolution doesn't necessarily have to map to sequential in-game time.

Alternatively from the PoV of in-game time, and in support of your point (I think), "instantaneous" could just mean "a short enough time there's no need to worry about it." Not "so short it must be simultaneous".

[Zhorn]

-snip-
It is a noticeable deviation from the post topic, so I agree we should draw it back in.

On Shield Master:
I'm of the camp that abilities that say "when you (x), then you can (y)" require that (x) to actively trigger before (y).
Regarding the "fully complete" (x) part is where I differ in opinion from the clarifications that is most recent by @JeremyECrawford, mostly because it comes across as logically inconsistent (I know, logic consistency in a fantasy setting... what am I thinking?).
At 2 attacks: You can attack once, run 30 ft, attack again, then bonus action shove... But you cannot attack, bonus action shove, then attack?
At more attacks, this seems even sillier. At 4 attack, you can attack, move, shove(as an attack), attack, move, attack, and then bonus action shove, but still cannot attack, bonus action shove, then attack?

On other bonus actions:
Then there's instances where a target is killed when carrying things like Hunter's Mark, or Hex. Comes across as overly harsh telling players they cannot transfer the debuffs until they end their attack action.
How about Misty Step? Attack once, bonus action Misty Step, but none of the unspent attacks on the other end, but you could spend the rest of the turn using your full movement.

On spells:
From what we're both saying, we do seem to agree that movement should not be able to be inserted between rolls if multiple attack rolls are included. Our reasoning does come from a different place, but the end result is the same. We're probably not going to come to an agreement on our interpretations of how they behave, but for the purposes of the discussion, it's probably not worth the effort for either of us.

[Jerrykhor]

Eldritch Blast multiple bolts are definitely not simultaneous, because if they were, it would say so. Like in the case of Magic Missile.

[Aimeryan]

Eldritch Blast multiple bolts are definitely not simultaneous, because if they were, it would say so. Like in the case of Magic Missile.

Whether something is simultaneous or sequential matters not if the time taken either way is insignificant enough to be listed as 'Instantaneous'. Essentially, for a 'Duration: Instantaneous' to have any meaning nothing can come between it (otherwise it is by no definition instantaneous).

~~~

Regarding the timing of the Bonus Action issue specifically, here is the text of relevance:


You choose when to take a bonus action during your turn, unless the bonus action's timing is specified...

It is quite clear that if you have a bonus action, you can take it during the Attack action because you get to choose when you take it - unless the timing is specified.

In regards to Shield Master specifically, Shield Master does not specify the timing, only the condition under which you are granted the bonus action. The issue then becomes, when have you fulfilled the condition? The condition is fulfilled if you take the Attack action in the same turn, however, the dispute lies in the 'take'. Some say you have taken the action the moment you lock yourself into that action, some say it is the moment you have made an attack, some say it is the moment you have taken all possible attacks including extra attacks.

Personally, since you can not change your action after making an attack, I see it that you have taken the action at that point. I could also see a ruling that if someone locks themselves into the action, even if yet to make an attack, then they have taken the action.

[BurgerBeast]

Whether something is simultaneous or sequential matters not if the time taken either way is insignificant enough to be listed as 'Instantaneous'. Essentially, for a 'Duration: Instantaneous' to have any meaning nothing can come between it (otherwise it is by no definition instantaneous).

This has come up before though, and it does matter, particularly when the warlock has the invocation repelling blast.

If multiple EBs hit simultaneously, then RAW the effects would not stack, so the maximum push distance is 10 feet. If the effects are sequential, then the maximum push distance is 10 feet per blast.

[Zhorn]

This has come up before though, and it does matter, particularly when the warlock has the invocation repelling blast.

If multiple EBs hit simultaneously, then RAW the effects would not stack, so the maximum push distance is 10 feet. If the effects are sequential, then the maximum push distance is 10 feet per blast.

Sequential vs simultaneous also comes down to what end of the spell is being refereed to, from the caster vs on the targets.


Sequential from the caster: targets can be declared as rolled for in sequence, allowing the caster to not waste beams on over-killing targets.

Simultaneous from the caster: declare all targets, cast the spell, no changing declared targets

Sequential onto the target(s): things like Repelling Blast stack their movement effect. Potentially knocking a target our of range of proceeding beams.

Simultaneous onto the target(s): all beams hit the target(s) together, but Repelling Blast is only triggered a single time per target.

The more I read into this, the more I come to see that the current ruling by designers is that Eldritch Blast IS resolved sequentially for both target selection and effect application (twitter is terrible for rulings with flip-flapping back and forth). So I stand corrected on many points I made regarding my understanding on Eldritch Blast's behaviour.

being said, in the games I DM, I will run it as simultaneous from the caster (because that's how I envision spells behaving) and semi-sequential onto the target(s) (hit one after the other, but close enough together that all impacts are made before the stacked movement from repelling blast is applied).

[Aimeryan]

This has come up before though, and it does matter, particularly when the warlock has the invocation repelling blast.

If multiple EBs hit simultaneously, then RAW the effects would not stack, so the maximum push distance is 10 feet. If the effects are sequential, then the maximum push distance is 10 feet per blast.

I don't see that being the case at all; the distance would be summed up, just like the damage is.

@Zhorn: The intention is clear enough, given Twitter. However, that doesn't change RAW. If something is within a timeframe described as 'Instantaneous' it cannot be a timeframe where you can wait for the blast to hit, see what result that has had, and make a decision about how that affects the next blast - several times over.

Therefore, you need not stand corrected, unless you presumed it was also the intention?

[Chronos]

As an aside, until this thread, it never even occurred to me to wonder whether eldritch blasts were sequential or simultaneous: I'd just assumed that they were simultaneous. But I suppose someone else could have just as blithely assumed they were sequential.

And realistically, the range is long enough and most battlefields small enough that the question of one blast knocking the target out of range of the others probably isn't likely to actually come up.

[Benny89]

I still follow RAW and ignore Jeremy interpretation. I spoke to my players and they want SM to stay competetive vs other feats.

RAW says you have take attack action it that turn. It doesn't say anything about before, after, in the middle. Just that you need to take attack action that turn.

So the way I see it- it's a combo. And I treat it as such. Meaning that fighter starts a sequence of attacks/manouvers that have to be chained together. So he shield bash enemy to prone him and immidietly follows up with attacks on him, while target is proned.

Therefore if my players says "I am going to shield bash him and try to prone him and then follow up with my attacks"- that is imo ok.

Without that Shield Bash will never stand next to feats like GWM, PAM or SS.

[BigRedJedi]

To fix Shield Master, replace the first bullet point:

- If you take the*Attack*action on your turn, you can use a bonus action to try to shove a creature within 5 feet of you with your shield.

With:

- While wielding a shield, if you take the Attack action on your turn and replace one or more of your attacks with an attempt to shove a creature within 5 feet of you, you can use a bonus action to make one additional attack.

(You can add the clause: "...against the same target that you shoved." If you would rather not allow just a free extra attack.)

[Benny89]

To fix Shield Master, replace the first bullet point:

- If you take the*Attack*action on your turn, you can use a bonus action to try to shove a creature within 5 feet of you with your shield.

With:

- While wielding a shield, if you take the Attack action on your turn and replace one or more of your attacks with an attempt to shove a creature within 5 feet of you, you can use a bonus action to make one additional attack.

(You can add the clause: "...against the same target that you shoved." If you would rather not allow just a free extra attack.)

Why not just make it simple like PAM?

"While wielding a shield, you can use your bonus action to try to shove a creature within 5 feet of you with your shield".

Done.

[Chronos]

Polearm Master requires that you use the attack action, too. In fact, it's even more restrictive: It requires that you attack only with a polearm (which, I suppose, rules out holding a quarterstaff in one hand and a dagger in the other, or something, and alternating them for your attacks).

[Jerrykhor]

I don't see that being the case at all; the distance would be summed up, just like the damage is.

@Zhorn: The intention is clear enough, given Twitter. However, that doesn't change RAW. If something is within a timeframe described as 'Instantaneous' it cannot be a timeframe where you can wait for the blast to hit, see what result that has had, and make a decision about how that affects the next blast - several times over.

Therefore, you need not stand corrected, unless you presumed it was also the intention?

Are you talking about the English definition of 'Instantaneous?' Because that's not how its defined in the phb. I don't care what you think it should be, we are talking about RAW here. And 'Instantaneous' simply means that it cannot be dispelled because the magic effect starts and ends in the same turn. Besides, its the spells DURATION that is 'Instantaneous', not the casting. Casting time of EB is 1 action. 1 action is definitely not super instant that you can't possibly fire multiple bolts sequentially. Besides, if a fighter can attack 10 times and in one turn, why can't a warlock fire 4 bolts of EB?

I'll agree that by RAW, one cannot move in between firing the bolts though.

[ThePolarBear]

agreed. The Common English used isn't precise enough to differentiate between the 3 cases.

Common english requires x to be a thing before y can exist in a conditional statement. ALWAYS.

This rules straight out "before "take the attack action"". There's no possible argument on this.
It might be before the attack for some meanings of "take" (to name one, declare), but it will ALWAYS happen afterwards, or at the very earliest during (for meanings like "take = attacking").


The problem is that we see our interpretation in the RAW; therefore, every other interpretation is "illogical" (ie Crawford plays the way he called "illogical")

It is illogical because there's no way that an implication refers to something happening before a condition in english.

[Laserlight]

This rules straight out "before "take the attack action"". There's no possible argument on this.
It might be before the attack for some meanings of "take" (to name one, declare), but it will ALWAYS happen afterwards, or at the very earliest during (for meanings like "take = attacking").

"First, I use my bonus action Shield Master shove. I can only use that if I take the Attack Action, so by using SMShove, I'm taking--ie selecting, declaring, committing to--the Attack Action." Valid, or no?

[RSP]

Common english requires x to be a thing before y can exist in a conditional statement. ALWAYS.


If you play football in the NFL, then you will get paid.

Is this a valid common English conditional statement?

X does not come before Y, in this case. Players get paid (and paid well) before ever actually playing.

[Boci]

If you play football in the NFL, then you will get paid.

Is this a valid common English conditional statement?

X does not come before Y, in this case. Players get paid (and paid well) before ever actually playing.

That's because your saying something technically inaccurate, which doesn't matter because everyone understands what is mean. Players don't get paid for playing in NFL, they get paid because they sign a contract which says they get paid in return for responsibilities, one of which is likely going to be playing football, but its not actually conditional on playing football.

[LudicSavant]

Common english requires x to be a thing before y can exist in a conditional statement. ALWAYS.

That is neither how if/then statements work in common English, nor how they work in sentential logic.

Which is why you can say things like "If the movie is at 6:00, then we should arrive at 5:45."

[RSP]

That's because your saying something technically inaccurate, which doesn't matter because everyone understands what is mean. Players don't get paid for playing in NFL, they get paid because they sign a contract which says they get paid in return for responsibilities, one of which is likely going to be playing football, but its not actually conditional on playing football.

Players do get paid for playing football, they just get a portion of their playing salary upfront. If a player signs a contract, gets paid his upfront portion, and then decides to never play football, the team can recoup the money paid. Why? Because the player decided to never play football, violating the conditions of the agreement.

[Boci]

Players do get paid for playing football, they just get a portion of their playing salary upfront. If a player signs a contract, gets paid his upfront portion, and then decides to never play football, the team can recoup the money paid. Why? Because the player decided to never play football, violating the conditions of the agreement.

Yes exactly, the contract and the signing of is the important part here. They are paid, because they signed a contract with specified an amount of upfront payment with an option for recouping. It starts with the contract being signed.

[Willie the Duck]

That is neither how if/then statements work in common English, nor how they work in sentential logic.

Which is why you can say things like "If the movie is at 6:00, then we should arrive at 5:45."

Last time I made that point, I was accused of believing in retrocausality. :smalltongue:

[LudicSavant]

Last time I made that point, I was accused of believing in retrocausality. :smalltongue:

It's one of those things where certain people on the internet with no background in relevant fields somehow feel that they are qualified to "correct" logicians on the point. Kind of like how certain people on the internet will tell mathematicians that they're wrong when they say .9 repeating equals 1.

[RSP]

Yes exactly, the contract and the signing of is the important part here. They are paid, because they signed a contract with specified an amount of upfront payment with an option for recouping. It starts with the contract being signed.

With the condition that they play football.

So the statement “If you play football in the NFL, then you will get paid,” isn’t inaccurate. There absolutely are more details involved, but the statement is true.

Isn’t that the same with the SM Feat? It’s a very similar process of “go ahead and take the BA Shove, but if you don’t take the Attack Action, I’m going to take back the BA and count it as one of your attacks from the Attack Action;” to “go ahead and take the signing bonus, but if you don’t play in the NFL, I’m going to take back the signing bonus.”

[Boci]

With the condition that they play football.

So the statement “If you play football in the NFL, then you will get paid,” isn’t inaccurate. There absolutely are more details involved, but the statement is true.

Not really. What if they don't play not because they change their mind but because a tragic accident ruins their career before it starts? Do they have to replay the money then? If not, then its clearly not an accurate statement, they weren't paid because they played, because they never played. They were paid because they signed a contract.

[RSP]

Not really. What if they don't play not because they change their mind but because a tragic accident ruins their career before it starts? Do they have to replay the money then? If not, then its clearly not an accurate statement, they weren't paid because they played, because they never played. They were paid because they signed a contract.

Yes really.

If you’re saying they get injured while practicing football with their NFL team, then they have fulfilled the requirement of playing in the NFL (the prior statement did not require them to play in official in-season games during the formal season), and because there’s a clause that protects them from losing their money in the event they get hurt while practicing.

So again, the statement holds true and is not inaccurate.

[Boci]

because there’s a clause that protects them from losing their money in the event they get hurt while practicing.

Right. Clause, contract, not playing. They got paid becausde they signed the contract, not because they played, because they never did play.

As I said, its a very nitpicky difference, but it is a difference.

[Alucard89]

I am not native speaker but RAW:

"If you take the Attack action on your turn, you can use a bonus action to try to shove a creature within 5 feet of you using your shield."

I don't see any indication here whenever Attack action needs to come before or after bonus action. It just requires me to take Attack action in this turn. So I can use bonus action first as long as I will after that take Attack action to complete requirement of "take attack action on your turn".


"If you take the Attack action on your turn". I do.

[NaughtyTiger]

The problem is that we see our interpretation in the RAW; therefore, every other interpretation is "illogical" (ie Crawford plays the way he called "illogical")
It is illogical because there's no way that an implication refers to something happening before a condition in english.

you made my point on this one. you can't see it from anyone else's point of view; therefore, anyone else's point of view is illogical [to you].

shield bash at the start was logical for 4 years to everyone i have played with; and the lead designer.
then it was logical that shield bash had to be at the end to the lead designer for 6 months(?)
now it is logical to the same lead designer that sheild bash can be in the middle

it is moot, now, cuz the RAW has been updated in an errata that the first case is explicitly barred.

this is not an insult or an attack , just a different point of view.

[ThePolarBear]

you made my point on this one. you can't see it from anyone else's point of view; therefore, anyone else's point of view is illogical [to you].

The fact that the exception on bonus actions with timing not being considered AND that conditionals do have a timing not being considered made the ruling ALWAYS illogical from a RAW point of view, but not from the INTENDED point of view.

It was a mistake.

People made a mistake and still make a mistake in thinking otherwise.

There's no "different" point of view here.


shield bash at the start was logical for 4 years to everyone i have played with; and the lead designer.

This is where you are wrong. It was never logical, but it was how it was stated to be intended to work due to a mistake.
This doesn't make it any more or any less logical.


then it was logical that shield bash had to be at the end to the lead designer for 6 months(?)

IT STILL IS, from a RAW point of view.


now it is logical to the same lead designer that sheild bash can be in the middle

From a RAI point of view.


it is moot, now, cuz the RAW has been updated in an errata that the first case is explicitly barred.

Still confusing things? RAW has not changed. RAW has NEVER changed.


this is not an insult or an attack , just a different point of view.

Which is in multiple places objectively wrong.

Edit:
To put this into perspective:
If today there was a release, where there's a rule that states that the new monster type "Quidfastig" all take double poison damage, on top of any weakness they might have.
In the release, there's a "Quidfastig" named "Pebble", that has poison damage immunity.

When questioned about how much damage a "Pebble" would take from a 35 poison damage attack, if the answer is "Because all Quidfastig take double poison damage, it takes 70", it would still be illogical. It might show that the intention was for "Pebble" to never have poison damage immunity, but having immunity makes it so that the damage "Pebble" would take is 0.
For Shield Master, the explanation has always been "Like most bonus actions". The illogicity was never considering that Conditionals have timings and that there's a rule about that for bonus actions to begin with.
It ended with an illogical result, which by power of RAI might have very well been the INTENDED result. It wasn't - SM was never intended to be an exception on that rule - so a mistake was corrected.

[Xetheral]

Common english requires x to be a thing before y can exist in a conditional statement. ALWAYS.

This rules straight out "before "take the attack action"". There's no possible argument on this.
It might be before the attack for some meanings of "take" (to name one, declare), but it will ALWAYS happen afterwards, or at the very earliest during (for meanings like "take = attacking").

It is illogical because there's no way that an implication refers to something happening before a condition in english.

This is not always true. Depending on context, a conditional statement can imply either the possibility of the condition occuring last or in some cases the requirement that the condition occurs last. Consider the following example:


"If you bring my car back with a full tank I'll give you the keys."

This is a perfectly valid English sentence. In the context of knowing that cars require keys to operate, the speaker must give the keys to the listener prior to the condition being met.

Here's another example:


"If you buy me some more, you may eat the last cookie."

Here, in the context of knowing that purchasing more cookies will result in the "last" cookie no longer being the last one, we can infer that the speaker's most likely intention is to permit the listener to eat the last cookie and only then going to buy more cookies.

Here's a final example that mirrors the structure of the Shield Master feat:


"If you clean your room today I will give you a ride to school."

If said on a school day prior to the start of school, then, in context, this sentence implies that the ride to school will come before the condition of "cleaning your room today".

Obviously, Shield Master has entirely different context. We can (and do!) endlessly debate whether the context of the Shield Master feat permits the condition to occur last. But these examples disprove your original assertion that "Common english requires x to be a thing before y can exist in a conditional statement. ALWAYS."

[NaughtyTiger]

"The problem is that we see our interpretation in the RAW; therefore, every other interpretation is "illogical""

snip

proving my point again.

[Laserlight]

There's no "different" point of view here.

If that were the case, there wouldn't be an argument. Clearly there is an argument. Ergo...

[RSP]

Right. Clause, contract, not playing. They got paid becausde they signed the contract, not because they played, because they never did play.

As I said, its a very nitpicky difference, but it is a difference.

Again, I’m not stating there aren’t other factors involved, but it’s not an inaccurate statement.

Note this is true with SM as well.

[ThePolarBear]

This is not always true. Depending on context, a conditional statement can imply either the possibility of the condition occuring last or in some cases the requirement that the condition occurs last. Consider the following example:


"If you bring my car back with a full tank I'll give you the keys."

This is a perfectly valid English sentence. In the context of knowing that cars require keys to operate, the speaker must give the keys to the listener prior to the condition being met.

Here's another example:


"If you buy me some more, you may eat the last cookie."



Both these examples however are fallacious, since the actual condition is the promise that something will be done, not the actual act. That's the meaning behind how the phrase work. It's an implicit. And as that implicit, the condition is always fulfilled.

Upon not keeping the word, the rebuttal would be "you told me you would". Because that's all that has been asked as a condition.

As such, the condition is ALWAYS fulfilled first. It's just that what the condition actually is in common english is dependant on the contextual meaning of the phrase. You really really want that gas refill, but you can only go with the word, and accept that


Here's a final example that mirrors the structure of the Shield Master feat:


"If you clean your room today I will give you a ride to school."

Again, working on a promise, not on a fact. Furthermore, all your examples are a different kind of conditional from the shield master one.


"The problem is that we see our interpretation in the RAW; therefore, every other interpretation is "illogical""

No. It doesn't follow the requirement of logical analysis. Therefore it is illogical. The fact that someone doesn't have the same interpretation doesn't make it illogical. But critical analysis must be hard enough, while strawman are abundant and easy, right?

[NaughtyTiger]

Both these examples however are fallacious, since the actual condition is the promise that something will be done, not the actual act. That's the meaning behind how the phrase work. It's an implicit. And as that implicit, the condition is always fulfilled.

If you [promise to] take the attack action, you can take the bonus action shield bash.

every argument you made against the premise can be applied for the premise.


"The problem is that we see our interpretation in the RAW; therefore, every other interpretation is "illogical""
proving my point again.

[Willie the Duck]

But critical analysis must be hard enough, while strawman are abundant and easy, right?

If childish insults is where you have fallen to, no one is going to walk away from this thread thinking you were correct. You have not made your case, and are blaming your audience for your failure to do so. Everyone else here is just as well versed in the process of critical analysis as you are. This is freshman level stuff, no one here is more expert than everyone else. If you have not convinced your audience, it is your failure to communicate. Please step back, take a breath, realize that you might not have expressed in written word the iron-clad case you have in your head, and made a humble and reasoned further attempt. Perhaps you will see where you have failed to make your case.

[Xetheral]

Both these examples however are fallacious, since the actual condition is the promise that something will be done, not the actual act. That's the meaning behind how the phrase work. It's an implicit. And as that implicit, the condition is always fulfilled.

Upon not keeping the word, the rebuttal would be "you told me you would". Because that's all that has been asked as a condition.

As such, the condition is ALWAYS fulfilled first. It's just that what the condition actually is in common english is dependant on the contextual meaning of the phrase. You really really want that gas refill, but you can only go with the word, and accept that



Again, working on a promise, not on a fact. Furthermore, all your examples are a different kind of conditional from the shield master one.

Hold on a moment. I thought you were claiming that all conditional statements in English require the stated condition to occur first, and therefore because Shield Master is a conditional statement in English the stated condition must occur first. My examples disprove that premise.

If you're instead claiming that it's possible to map any conditional statement in English to a conditional in formal logic where a (potentially implicit) condition preceeds (or occurs simultaneously with) an effect, that's a very different premise. I actually agree with that new premise, but it doesn't support your conclusion about Shield Master because it's possible to map the text of Shield Master to the logical conditional IF (promise to take the attack action on your turn) THEN (make bonus action shove).

Whether or not that map between the English and the logic is a good interpretation of the English is, of course, what this entire debate is all about.

[jas61292]

That is neither how if/then statements work in common English, nor how they work in sentential logic.

Which is why you can say things like "If the movie is at 6:00, then we should arrive at 5:45."

I just want to talk about this example for a second, because I think it is an easy one to get caught up on. It seems to me like you are trying to say that this is an example of something in the future being the conditional for something before it, and that this is fine English. Well, it's true that this is fine English, but it is not true that the future event is a conditional.

What the conditional here is based on is the scheduled time of the movie. This is a present fact, not a future event. If at 5:59 the power at the theater goes out and so the movie never actually plays, it does not suddenly become the case that you did not go to the theater at 5:45. Things like this are phrased such as they are to imply a relation to a future event, but the actual conditional is regarding a belief about the future, not the actual future event itself.

This is similar to what ThePolarBear just pointed out about other examples being conditional to promises to take future actions, not actually the future actions themselves. And while I'm sure the game won't break if you run it this way, the rules themselves are based on actual actions, not the promise to take them later.

[NaughtyTiger]

the rules themselves are based on actual actions, not the promise to take them later.

I am not convinced the promise to the DM is excluded.

[Zhorn]

I am not convinced the promise to the DM is excluded.

Unless the DM doesn't take promises, in which case you must take the action first. But that just brings us to the "different rules for different DMs" model, in which case RAW, RAI, or commonly understood rulings have no bearing.

[LudicSavant]

I just want to talk about this example for a second, because I think it is an easy one to get caught up on. It seems to me like you are trying to say that this is an example of something in the future being the conditional for something before it, and that this is fine English. Well, it's true that this is fine English, but it is not true that the future event is a conditional.

What the conditional here is based on is the scheduled time of the movie. This is a present fact, not a future event. If at 5:59 the power at the theater goes out and so the movie never actually plays, it does not suddenly become the case that you did not go to the theater at 5:45. Things like this are phrased such as they are to imply a relation to a future event, but the actual conditional is regarding a belief about the future, not the actual future event itself.

This is similar to what ThePolarBear just pointed out about other examples being conditional to promises to take future actions, not actually the future actions themselves. And while I'm sure the game won't break if you run it this way, the rules themselves are based on actual actions, not the promise to take them later.

You are mistaken.

The wording is "should," not "will." No mention was made of scheduling, or planning, nor is any required for the conditional statement to be proper by either sentential logic or common English.

If someone arrived at the theater at 5:45, and there was a power outage at 5:59, they might hypothetically say "well, we shouldn't have even shown up then."

[ThePolarBear]

If you [promise to] take the attack action, you can take the bonus action shield bash.

every argument you made against the premise can be applied for the premise.


Go back and read what i wrote in the first post. Because that is actually ALREADY THERE. I've already stated that it depends on the meaning given on "take", and in the end, no matter the meaning, it will never be BEFORE what one means by "take" is actually fulfilled or in the course of being fulfilled (for continuous verbs).

Now, with that font of inspiration, change the perception of what you think i'm saying.

Because the "being illogical" is not about what JC says it's RAW or not. It's about not applying the conditional timing when there is a conditional, therefore not considering that conditional bonus actions do not fall under the general rule as it was previously stated to be the case.
The reason behind the ruling was illogical, making the ruling standing on illogical reasons (or simply, illogical). Not true, false, whatever. Simply "illogical".


"The problem is that we see our interpretation in the RAW; therefore, every other interpretation is "illogical""
proving my point again.

And again, apparently something is escaping comprehension, since what you are using as an argument has already been covered.


If childish insults is where you have fallen to

What? The fact that trying to look at his own advice before dismissing arguments should be something that one does?
Or that what i'm doing is actually applying logic? It is a strawman, i'm not dismissing his argument just because it is his, but because it is not valid.
See the above cut off. How is that even an insult by the way? Applying logic is hard. I'm pretty sure he can handle it. Repeating his opinion without criticizing mine is easier. And there has been at least a bit of "miscommunication" or willingly ignorance if the above comes to exist, right?

I mean, it can absolutely be a mistake.


You have not made your case, and are blaming your audience for your failure to do so.

At most, "i blame my audience" for not making arguments against it. Xerathal did. And i'm answering to their issues.


If you have not convinced your audience, it is your failure to communicate.

This is a fallacy, however, that assumes that everyone is willing to change their mind based on arguments. While i might agree, discussion is two way. There's no logical analysis possible into "it's because you say so, it's because i say so". If one is unwilling, there's nothing to be done to it but pointing to it.


Please step back, take a breath, realize that you might not have expressed in written word the iron-clad case you have in your head, and made a humble and reasoned further attempt.

I'm doing it with people willing to have a discussion. Ex. Xerathal.


Hold on a moment. I thought you were claiming that all conditional statements in English require the stated condition to occur first, and therefore because Shield Master is a conditional statement in English the stated condition must occur first. My examples disprove that premise.

No they don't, Xerathal. Your example prove that english is a language used by people for people, with implicit meanings and "non-verbal" discussions that are still part of communication. The example of the dad and the keys WILL end with "Promise?", since that is what is actually asked for, and not, if not jokingly, with a "where's my full tank?". Because if it wasn't, there would be an impossibility (not really... just a need to break a couple of pieces off :D), and the conditional would never be fulfilled to begin with!

And, going deeper, most of the time it is not a strict correlation, too. I might give you the keys anyway, and not because you came back with a full tank or an empty tank, but because i need you to pick up groceries...

You can use, in english and many other languages, constructs. Those constructs might have different readings. Conditionals always have the same structure, it's the meaning and the actual condition that varies. In your examples it's the "word given" what is being looked for. And it's not the empty tank or the unkept room that will be the reprimand point, but the broken promise. The tank and the rooms are only the evidence of that fact.


If you're instead claiming that it's possible to map any conditional statement in English to a conditional in formal logic where a (potentially implicit) condition preceeds (or occurs simultaneously with) an effect, that's a very different premise. I actually agree with that new premise, but it doesn't support your conclusion about Shield Master because it's possible to map the text of Shield Master to the logical conditional IF (promise to take the attack action on your turn) THEN (make bonus action shove).

That is EXACTLY what i stated. In the case of shield master, it's the meaning of "take" that has always been a problem:


Common english requires x to be a thing before y can exist in a conditional statement. ALWAYS.

This rules straight out "before "take the attack action"". There's no possible argument on this.

Notice that it is not "before an attack", or "before the attack action".


It might be before the attack for some meanings of "take" (to name one, declare), but it will ALWAYS happen afterwards, or at the very earliest during (for meanings like "take = attacking").

always afterwards the "take", since apparently it wasn't clear enough.

And... why not possible? You just did it!

It's a zero conditional, since it expresses a generally true fact, not a possible future or a counterfactual state. The verbs are coherent with it being a zero conditional, and "can" is used to express a possibility you are empowered should the condition come to pass, expressing that it is only then that you have the possibility, and not that you are forced or otherwise compelled to.
Analyzing the phrase itself, you can reduce it to "You can use the bonus action if you take the attack action", and that's the only meaningful reduction you can reach.
Condition and implication placement doesn't really matter, but when the condition is first there's usually a comma, and there's a comma.

There's no other possible logical derivation. And again, i've ALWAYS, FROM THE START, been a proponent that "take" can, for different meanings, be something different. But it STILL isn't about the conditional statement having the possible solution of having the implication come before the condition.

But, again, the "declaration" might as well be a meaning that people came up AFTER the ruling, not before, as an explanation on why. But i have no idea if that's the case.

It is simply not true that conditionals are "unclear". It's the CONDITION that might be unclear.
But, if we base it on intent, we know that "declaring" is not what "take" means. If we base on official clarifications, we know that X has to come before Y.


Whether or not that map between the English and the logic is a good interpretation of the English is, of course, what this entire debate is all about.

Yeah. But ignoring linguistics "just because" is... well.

[NaughtyTiger]

Go back and read what i wrote in the first post
which first post? do you mean your actual first post on the thread?


Now, with that font of inspiration, change the perception of what you think i'm saying.
i have stated that I understand your point of view, but I do not agree that it is the only correct interpretation.
you have actually stated my position well, the implicit promise bit, much better than i was, so thanks.



And again, apparently something is escaping comprehension, since what you are using as an argument has already been covered.
similarly, your argument has been covered (again, you did a great job supporting the implicit promise bit)


Repeating his opinion without criticizing mine is easier.
to be fair, i don't need to criticize your opinion.
you keep telling me, i'm wrong.
I keep saying your interpretation is also reasonable.


At most, "i blame my audience" for not making arguments against it. Xerathal did.
and since Xerathal made the same argument I did (but better), should I regurgitate his words?


Fine, I will bite, cuz you are insistent as hell.

"If you take the Attack action on your turn, you can use a bonus action to try to shove a creature within 5 feet of you using your shield."

Given your interpretation, how does the feat change if "on your turn" was removed?
"If you take the Attack action [-], you can use a bonus action to try to shove a creature within 5 feet of you using your shield."

How i read it:
The "on your turn" to me allows the promise (again thanks for putting my position in clear wordrs), it allows 'at any point during your turn'.
Without that phrase, then I would give ground on the timing[/QUOTE]

Side note, can someone give me another example of restricted timing using "If" (edit: from the PHB)? (the "when" wording is pretty easy to find, and "when" definitely implies a timing aspect to me)

[jas61292]

You are mistaken.

The wording is "should," not "will." No mention was made of scheduling, or planning, nor is any required for the conditional statement to be proper by either sentential logic or common English.

If someone arrived at the theater at 5:45, and there was a power outage at 5:59, they might hypothetically say "well, we shouldn't have even shown up then."

Exactly. You are basing your current decisions, not on what happens in the future, but what you think, based on current facts, might happen in the future. As evidenced by your second paragraph, they didn't decide to leave based on when the movie actually started. They chose to leave based on when the facts of the moment led then to believe it would start.

In other words, the original statement is not stating "if the movie actually starts at 6, we'll have arrived at 5:45." It is saying "if you go on the theater's website and it says the movie is starting at 6, then we will arrive at 5:45." The action, arriving, is based on the condition of finding out the scheduled time, not the actual time it starts.

[LudicSavant]

Your are basing you're

Please... stop... doing... this... :smallfrown:


It is saying "if you go on the theater's website and it says the movie is starting at 6, then we will arrive at 5:45."
No, it is not saying that. You have rewritten the sentence into one that is not logically equivalent. For example, "will arrive" and "should arrive" have different meanings.


The action, arriving, is based on the condition of finding out the scheduled time, not the actual time it starts.
Technically, we don't need to find out any information about the scheduled time at all in order to make a valid if/then statement of this sort.

For example, I can make the statement: "If the movie starts at any time t, I need to arrive at time t-15 in order to buy popcorn and get to my seat before the movie begins." This is a perfectly valid sentence and contains zero expectations about the time the movie starts.

If the movie does not begin for whatever reason, all that means is that I may not have needed to arrive. I may or may not have done so anyways, but it would have served no need. (An astute reader will also note that I say may not have needed to arrive. There could still be a different reason I need to arrive at that time, unless we used an "if and only if" statement instead of an "if" statement). All of this stuff about websites and the like are simply assumptions and embellishments on your part, and are unnecessary for the statement to parse.

[jas61292]

Please... stop... doing... this... :smallfrown: Sorry. I'm typing on my phone which causes me to make mistakes like this sometimes. I usually try to reread my posts a few times after posting and fix errors, but... alas, I miss some.


No, it quite clearly is not saying that. You have completely rewritten the sentence into one that is not logically equivalent. For example, "will arrive" and "should arrive" have different meanings. Yes, I technically did rewrite the sentence. But that is because English is not formal logic. What I wrote is one probable implied meaning of the phrase in standard English. You are right that I should have used "should," but that is really besides the point. The actual point is that the casual relationship is between the scheduled start time and the action of arriving, and not involving the actual start time. Like in all such situations, the cause happens before thre effect.


Yet again you are wrong. Technically, we don't need to find out any information about the scheduled time at all in order to make a valid if/then statement of this sort. Technically true, but irrelevant. It doesn't matter if you know the actual start time or are just guessing or whatever. What matters is that the action if arriving is predicated on a present belief about future events, not on the actual future event.


For example, I can make the statement: "If the movie starts at any time t, I will need to arrive at time t-15 in order to buy popcorn and get to my seat before the movie begins." This is a perfectly valid sentence and contains zero expectations about the time the movie starts. Yes, this is perfectly sound logic. But again the action is predicated on knowledge or assumptions or whatever in the present. Not the events of the future.

None of this says that the statements in question are not logically sound. What it says is that they still work under normal causality. An event at time t=x can only be caused by an event at time t=y where y<x. If the feat in question stated that the intent or promise to take the necessary action was the trigger, it would be a situation like these examples and it would be fine. But how I, and many others, interpret it is that it is saying the action itself is the trigger, not the intent thereof, and thus allowing the bonus action before the trigger would be like retroactively allowing you to decide when to leave for a movie based on when it actually started.

And like some have said, most of this hinges on the interpretation of "on your turn." Some take this phrase to be an implicit implication that an intent is fine. But others such as myself see it as nothing of the sort, and find the idea that a whole intent mechanic, that is nowhere else in the rules, would be hidden implicitly in a three word phrase in an optional rule, to be absurd.

[LudicSavant]

Well, it's true that this is fine English

Yes, this is perfectly sound logic.

Well, I agree with both of these statements, at least. :smalltongue:

And since it is indeed both fine English, and perfectly sound logic, I am having a hard time figuring out the basis for your objection.

Perhaps it would be helpful to offer examples with game rules. Say...
"If moving your knight would put your king in check, then you may not move your knight." Note that the knight never actually needs to move (and indeed, cannot) for this statement to work.

[Tanarii]

"The problem is that we see our interpretation in the RAW; therefore, every other interpretation is "illogical""


proving my point again.
Ah yes, "logical" to nerds on the internet and off ... just another way of saying "doesnt match my preferred way of thinking."

Its one of the more abused words, along with literally.

[jas61292]

Well, I agree with both of these statements, at least. :smalltongue:

And since it is indeed both fine English, and perfectly sound logic, I am having a hard time figuring out the basis for your objection.

Perhaps it would be helpful to offer examples with game rules. Say...
"If moving your knight would put your king in check, then you may not move your knight." Note that the knight never actually needs to move (and indeed, cannot) for this statement to work.

The issue with a statement like your chess one is that it is not really much of a conditional. I mean, it is, but not really in the same sense as the D&D rule in question. Ultimately it is just a fancy way of saying "you may not move your own pieces in such a way that an opponent's piece could thereby capture your king." Or, more simply "if a move is illegal, you may not make it." This is true before you ever move. A move does not suddenly become an illegal move because you tried to make it.

It is the same thing with the movie example. Your claim that you always want to be at the theater 15 minutes before start time in order to buy snacks or whatever is just a statement of fact. It can be stated as a conditional, but all you are actually saying is "it takes 15 minutes to take care of the stuff you want to take care of before a movie."

But in both these cases, the facts are true without any actual conditions. You are just using abstract examples that rely on implicit knowledge as the antecedent, rather than explicit knowledge, with the intent of obfuscating the fact that this antecedent comes before the consequent.

A move being illegal is what prevents you from making it. Attempting to make it is not what retroactively renders it illegal.

[LudicSavant]

The issue with a statement like your chess one is that it is not really much of a conditional. I mean, it is, but not really in the same sense as the D&D rule in question.

Let's not mix up our goalposts here. The statement I was addressing was...


Common english requires x to be a thing before y can exist in a conditional statement. ALWAYS.

...which as we've established, isn't true.

[jas61292]

Let's not mix up our goalposts here.

I mean, I'd say it is true. Every counter example I've seen involves some sort of abstract thing to be the X, in such a way that it may appear with a casual glance to not come before Y, but in all those instances, X is an idea, concept or thought about some future thing, and that idea, concept or thought does exist before Y.

[ThePolarBear]

i have stated that I understand your point of view, but I do not agree that it is the only correct interpretation.

You stating makes no difference nor demonstrate your actual understanding, and you not agreeing doesn't make you correct.


to be fair, i don't need to criticize your opinion.

But you did. Or better, you attempted to belittle it, by appearing to do so with a fallacy. You are under no compulsion to do so, certainly.


you keep telling me, i'm wrong.

What should i tell you? You are correct? No matter what, if you use "banana" to mean "cat", i'm going to point that out.


I keep saying your interpretation is also reasonable.

Which is a different thing. On a murder case, with only one murderer, it doesn't matter if it appears that there might be three different murderers, prehaps all working together. Any one could reasonably be the murderer, yet, there's only one "correct" solution if the intent is to put only the guilty person in jail.


and since Xerathal made the same argument I did (but better), should I regurgitate his words?

Because you could have done it before Xerathal, since you answered before them? If that was to be your argument, why not make it?


Fine, I will bite, cuz you are insistent as hell.

"If you take the Attack action on your turn, you can use a bonus action to try to shove a creature within 5 feet of you using your shield."

Given your interpretation, how does the feat change if "on your turn" was removed?
"If you take the Attack action [-], you can use a bonus action to try to shove a creature within 5 feet of you using your shield."

How i read it:
The "on your turn" to me allows the promise (again thanks for putting my position in clear wordrs), it allows 'at any point during your turn'.
Without that phrase, then I would give ground on the timing

It reads "at any point after you take the Attack Action you can..." when outside of context.
But, "on your turn" is a limitation. It must be on your turn. Which still doesn't cancel the sequentiality of the "if" construction used in the rest of the text. Until you take the Attack Action, no matter when you take it if you remove the "on your turn", you still cannot do your bonus action. There's still a timing: one thing before the other, no matter what value you give to "take".

The "promise", "Declaration", whatever, can still exist even with "on your turn". It's still only because "take" is not clear in what it means. It has nothing to do with the conditional statement in construction and meaning.

Continuing to change what "take" means doesn't change my position. "Take" is not clear in RAW in meaning, even if at the time official tweets made it clear that it was never meant to have a "declaration" meaning, and as explained later on it meant to actually DO what the action tells you to do. So, you can attribute to "take" a moltitude of meanings and still be inside a reasonable context.

HOWEVER, a conditional sentence IS a timing. Since the conditional sentence has a timing, this statement was illogical. (https://twitter.com/JeremyECrawford/status/557816721810403329) It was built on incorrect premises.

The reason you think why it was logical is incorrect. Factually so, EVEN in the case where conditional construction with "future" could screw with the sequencing, because the conditional used is not one with a future in the condition anyway. So it is a logical leap, EVEN ASSUMING THE ABOVE, to apply it anyway.

You can rationalize it, but the reason was stated, and it was illogical years ago like it is now. The conclusion simply doesn't follow what is there.
RAI it would be fine ANYWAY, since RAI can go against what is actually written in any case.


Side note, can someone give me another example of restricted timing using "If"? (the "when" wording is pretty easy to find, and "when" definitely implies a timing aspect to me)

Any "if" conditional construction is time restricted.

"If i die, you can take everything from my body." - said no adventurer ever :D
It has no meaningful difference with when. Both If and When express the same thing in a conditional sentence. You end up with the same final meaning.
"If you take", "when you take". There's no real change, and to further restrict timing there's need for further words to reinforce it: "If you take, immediately", "when you take, immediately"


...which as we've established, isn't true.

Faulty examples do not make a demonstration! Again, all those example work off an expectation of something happening: an expectation that comes to pass BEFORE the actual implication resolves.

If whomever you gave the keys to were to come back with an empty tank, would you not feel duped?

That's all there is to it. It's THIS simple. You weren't basing your decision on an actual filled tank, but on a promise you took for granted. THAT was the condition you acted upon, no matter HOW you stated it.

This is true for every and any "future" in conditional sentences placed in the condition. It always end up into the belief that something will work, will happen... in an expectation.

It's the very concept of trust or faith (the non-religious one) that allows you to make such a statement to begin with.

[NaughtyTiger]

But, "on your turn" is a limitation. It must be on your turn.

can you take your attack action outside of your turn? no. so "on your turn" is either redundant, sloppy editing, or has specific meaning. i read it as specific meaning that coupled with "you can take a bonus action at any time during your turn" means "before your attack action is valid"

you put a lot of emphasis on the word "take" but none on the phrase "on your turn".


as far as your temper tantrum.
not engaging you is not belittling you,
it is just i don't like your tone.
you are mad that i didn't make my point clearly BEFORE Xeratheral or you.
you are mad that i didn't repeat my point AFTER Xeratheral and you did a better job.
you are mad that I didn't engage you despite you being aggressive.
you are mad that I am not pleading my case before you, despite you making it clear that you have made up your mind from your first post on this thread.



to be clear, the following is belittling you.
you dont need to tell me anything. nothing you have said or done makes you a damned expert in my view.
you wrote " it depends on the meaning given on "take", and in the end, no matter the meaning," seriously what analyitical mind says it depends on the meaning of take, but no matter the meaning of take?
you also wrote that "sheild bash at the end of Attack Action is RAW", no it isn't. your analytical mind should know that NOWHERE does it saw attack action is atomic. but you saw an opportunity to attack me, ineptly.

i asked you a simple question, in bold, and you couldn't even answer it straight (you tried to answer it, you used related words, but you didn't freaking answer it)
you told me how it doesn't change.

i guess that ends my tantrum...


"How does the feat change if you remove "on your turn"?



Side note, can someone give me another example in the PHB of restricted timing using "If"?

[Benny89]

snip

I am lurking this discussion and man- you have been proven few times that what you said is incorrect- just admit that your statement of "ALWAYS" was a mistake. It's not a shame to admit that you were not wrong. Guys really tried to explain that to you in polite and easy to understand way.

[Jamesps]

The best way I've been able to interpret the action timing rules is with the idea that taking the attack action ques up your attacks, which you can then use at any legal part of your turn.

So a character with shield master would take the attack action, gaining an attack in their que (or more than one if they have extra attack). They could then use shield master to shove. Afterwards they can opt not to attack, but their action has already been spent, they can't use it for anything different. This doesn't cause anything to trigger that specifically states "after attacking" like the monk martial arts or dual weapon style, but anything that simply requires the attack action to be taken could be used this way.

This interpretation seems to solve a lot of the timing issues that crop up with 5e actions.

[LudicSavant]

Again, all those example work off an expectation of something happening: an expectation that comes to pass BEFORE the actual implication resolves. No, they don't, no matter how many times you randomly capitalize something.


"If the movie starts at any time t, I need to arrive at time t-15 in order to buy popcorn and get to my seat before the movie begins." This is a perfectly valid sentence and contains zero expectations about the time the movie starts.

Here are some more examples:
"If winter comes at some future date, then the statement "winter is coming" is an accurate prediction."
"If winter does not come at some future date, then the statement "winter is coming" is not an accurate prediction."

Expecting the prediction to be accurate or inaccurate is not required or relevant to whether or not these are valid conditional statements. And winter coming certainly doesn't come to pass prior to the prediction being made.

The notion that if/then statements require some expectation of the If part being true is entirely false. In fact, you can even make valid if/then statements with the exact opposite expectation. For example:
"If you are really going to repay me, then there's no reason for me to not lend you the money. However, I do not believe that you will."

[Laserlight]

Crawford originally ruled "as with most bonus actions, you choose the timing, so the Shield Master shove can come before or after the attack action."

He later changed his mind: "if a feature says you can do X as a bonus action if you do Y, you must do Y before you can do X. For Shield Master, that means the bonus action must come after the Attack action. You can decide when it happens afterward that turn."

I think that makes it clear that:
a) there are two points of view
b) "do it whenever you feel like, as long as you also take the Attack action" is a valid reading of RAW
c) "do it after you complete the Attack action" is a valid reading of RAW
d) per Crawford, B is now deprecated and C is now official
e) Crawford isn't your DM so you can ignore D if you wish.

[Aimeryan]

The best way I've been able to interpret the action timing rules is with the idea that taking the attack action ques up your attacks, which you can then use at any legal part of your turn.

So a character with shield master would take the attack action, gaining an attack in their que (or more than one if they have extra attack). They could then use shield master to shove. Afterwards they can opt not to attack, but their action has already been spent, they can't use it for anything different. This doesn't cause anything to trigger that specifically states "after attacking" like the monk martial arts or dual weapon style, but anything that simply requires the attack action to be taken could be used this way.

This interpretation seems to solve a lot of the timing issues that crop up with 5e actions.

Agreed; I think this is the way many people on this side of the argument look at 'take the Attack action' - it is not about attacking, but about bearing the cost of the action.

~~~


Crawford originally ruled "as with most bonus actions, you choose the timing, so the Shield Master shove can come before or after the attack action."

He later changed his mind: "if a feature says you can do X as a bonus action if you do Y, you must do Y before you can do X. For Shield Master, that means the bonus action must come after the Attack action. You can decide when it happens afterward that turn."

I think that makes it clear that:
a) there are two points of view
b) "do it whenever you feel like, as long as you also take the Attack action" is a valid reading of RAW
c) "do it after you complete the Attack action" is a valid reading of RAW
d) per Crawford, B is now deprecated and C is now official
e) Crawford isn't your DM so you can ignore D if you wish.

You are missing an option: "do it after taking the Attack action". Funnily enough, that is near enough the exact RAW wording (with the addition of the 'after'). The issue is that taking the Attack action can mean three things in itself:

1) You bear the cost of the taking the action, that is, you no longer have that action available for other uses. The idea here is that you have taken the action by paying the cost - you need not have yet received any of the benefits of paying that cost. An analogy would be paying the shopkeeper for some goods, but having yet been given the goods. On the counter is a stand of lollipops; you are allowed to take one when you have bought something at the shop - you have paid for something at the shop, so being able to take the lollipop should be fine.

2) You bear the cost of taking the action, plus you must have received at least some of the direct benefits. The idea here is that there is a more firmer connection to having taken the action. The same analogy as the above, except some of the goods have been handed over, and then you take the freebie, with the remaining goods to follow.

3) You bear the cost of taking the action, plus you must have fully completed the transaction. The idea here is that it is unarguable that you have gone through with the action and some people like such simple solutions. The analogy above is changed here to you only being able to take a lollipop after every thing else has been handed over, for some reason.

[Tanarii]

Crawford originally ruled "as with most bonus actions, you choose the timing, so the Shield Master shove can come before or after the attack action."

He later changed his mind: "if a feature says you can do X as a bonus action if you do Y, you must do Y before you can do X. For Shield Master, that means the bonus action must come after the Attack action. You can decide when it happens afterward that turn."

I think that makes it clear that:
a) there are two points of view
b) "do it whenever you feel like, as long as you also take the Attack action" is a valid reading of RAW
c) "do it after you complete the Attack action" is a valid reading of RAW
d) per Crawford, B is now deprecated and C is now official
e) Crawford isn't your DM so you can ignore D if you wish.
Didn't Crawford say he had made a mistake, or something to that effect? As is, b is not correct and never was. (That also invalidates a if it actually means "there are two points of view, b & c.")

[NaughtyTiger]

Crawford originally ruled "as with most bonus actions, you choose the timing, so the Shield Master shove can come before or after the attack action."

He later changed his mind: "if a feature says you can do X as a bonus action if you do Y, you must do Y before you can do X. For Shield Master, that means the bonus action must come after the Attack action. You can decide when it happens afterward that turn."

I think that makes it clear that:
a) there are two points of view
b) "do it whenever you feel like, as long as you also take the Attack action" is a valid reading of RAW
c) "do it after you complete the Attack action" is a valid reading of RAW
d) per Crawford, B is now deprecated and C is now official
e) Crawford isn't your DM so you can ignore D if you wish.

C is not official.

The PHB and DMG does not state that the Attack Action is atomic, there is plenty of evidence that it isn't. The errata explicitly sidestepped the atomic issue.
A common interpreation (now that B is deprecated) is that the bonus action shove must come after the Attack Action has started. It can be performed before the Attack Action is completed.



Didn't Crawford say he had made a mistake, or something to that effect? As is, b is not correct and never was. (That also invalidates a if it actually means "there are two points of view, b & c.")
Crawford made multiple mistakes, the first was saying that B is wrong (my opinion). the second was saying C was right (atomic actions).

[Tanarii]

C is not official.

The PHB and DMG does not state that the Attack Action is atomic, there is plenty of evidence that it isn't. The errata explicitly sidestepped the atomic issue.
A common interpreation (now that B is deprecated) is that the bonus action shove must come after the Attack Action has started. It can be performed before the Attack Action is completed.
Given that the Attack Action grants one attack, it can be considered taken when the first attack occurs.

But yes, I agree a is a misstatement of possible views. There are actually 3:

a) There are at least three possible readings of RAW
b) Do it whenever you want, as long as you take the Attack Action on your turn at some point.
c) It must occur after you take the Attack action, which means after you take the one attack it grants. Extra Attack doesn't change this, since the Attack Action is still one attack before modification as a base rule.
d) It must occur after you take the Attack action, which means completing the attack. Since Extra Attack modifies the Attack Action, to take the newly modified thing, they must also be completed.

[Laserlight]

You are missing an option: "do it after taking the Attack action".

I wasn't saying "other readings don't exist", I was saying "we know these two are both legitimate readings of RAW because Crawdaddy said so, at one time or another."

The point being that the ambiguity is in the text; Crawford has issued a ruling, so we know what the Official Way is now (unless he changes his mind again); and unless we're at Crawford's table we can do it the original way or the new way or anything in between. Beyond that, it's just arguing over how many angels can dance on a pinhead.

[ThePolarBear]

can you take your attack action outside of your turn? no. so "on your turn" is either redundant, sloppy editing, or has specific meaning. i read it as specific meaning that coupled with "you can take a bonus action at any time during your turn" means "before your attack action is valid"

YES YOU CAN! You can take the Attack Action outside your turn! If you ready an action, aren't you readying an action? Prehaps the Attack one? You know, "Extra Attack, can't do it outside your turn" thing?
Even if that isn't an actual possibility, can you exclude that in the future there will be something that allows you to do so?
Even then, however, you still cannot use your bonus action outside your turn! So... what if that "on your turn" refers to the condition, and when that condition has to happen? But then, without the "on your turn", couldn't you use the bonus action on subsequent turns?


you put a lot of emphasis on the word "take" but none on the phrase "on your turn".

Because it's irrelevant for the demonstration anyway! Whether or not conditionals have timing is not dependant on what the condition is.
And it's irrelevant in determining the illogicity of the first ruling.


not engaging you is not belittling you,

engaging and dismissing is.


you are mad

I'm not mad.


you dont need to tell me anything. nothing you have said or done makes you a damned expert in my view.

And again, it doesn't make you any more or any less correct in your views. And the point is that i can do so, not that i need to, and as long as that possibility exists i can choose to use it. Exactly as you can choose to ignore or to answer.
I wouldn't try, again and again, if i didn't think there would be a point in doing it, don't you think?
Don't worry, however. This is it. Last try.


you wrote " it depends on the meaning given on "take", and in the end, no matter the meaning," seriously what analyitical mind says it depends on the meaning of take, but no matter the meaning of take?

Because you are decontextualizing again? The meaning of "take" changes when you can take the Bonus Action. But regardless of what meaning you give it, the condition has to be satisfied before the implication can happen.

If you give "declare", you need to declare. If you give "promise", you need to promise. And so on. If you give it "must at all costs finish all the attacks and climb three walls while singing", then you need to do that before.

And in context, since you replied to "how conditionals work", what the conditional is is really really inconsequential to conditional having or not a timing. Even changing which is, however, not changing "exists".


you also wrote that "sheild bash at the end of Attack Action is RAW", no it isn't.

Yes it is? If it is a possible, reasonable reading of the rules as written, isn't it RAW? Otherwise, wouldn't every other possible, reasonable reading ALSO not be RAW?


your analytical mind should know that NOWHERE does it saw attack action is atomic.

And again, irrelevant. And even without making the argument "it doesn't say you can split it", it being atomic or not is not part of my argument. It states "take". And if "take" means "complete", then all the attacks need to be completed. The fact that it's atomic or not is completely irrelevant: my argument stands true even if "take" means "while doing", or "declare to". You still need to be doing, or to have declared, to be allowed the bonus action.


but you saw an opportunity to attack me, ineptly.

Not attacking YOU. Attacking a statement that you made that was incorrect was not attacking you. I might be attacking your way to discuss. Your arguments and how you conduct them. Their form, integrity, validity.


i asked you a simple question, in bold, and you couldn't even answer it straight (you tried to answer it, you used related words, but you didn't freaking answer it)

HOW DOES IT MATTER?
Again: It makes so that if you take the attack action at any point, you can use the bonus action. The restriction on bonus actions still apply, because without exceptions those are still in place.
There's no longer the limitation on "take the attack action" to be "on your turn". That's it.


you told me how it doesn't change.

... the fact that the conditionals have or have not a - by definition - timing. Thus not changing the order. And, in the end, it doesn't change the illogicity being not bound by my interpretation. Or anyones, really.


"How does the feat change if you remove "on your turn"?

"at any point after you take the Attack Action you can..." This is how it changes. A limitation is lifted. You are no longer bound to your turn if you somehow can do it when it is not your turn. That's it. Exactly for the same reason that you cannot use Extra Attack when you Ready. Because it is not "Your turn".


Side note, can someone give me another example in the PHB of restricted timing using "If"?

"If you roll a 7 and a 1, for example, the number rolled is 71." You really cannot have 71 as a result of a roll without rolling first.


I am lurking this discussion and man- you have been proven few times that what you said is incorrect- just admit that your statement of "ALWAYS" was a mistake.

And again, those are all improper examples.

Conditionals have 4 forms + some exceptions.
Not a single one of those forms deals with a "will" as a future in the condition causing it to be after the implication. Because it is NONSENSICAL for an implication to come before the condition. And that is it.

When does this nonsensical situation happens? When you use two futures, one in the condition and in the implication, to describe something that MIGHT happen (first conditional) or SHOULD happen (zero conditional) (or any other possible case) after an event that is REALLY not bound on the condition for it to happen. So, it is not a condition.

"If you'll fill the tank, i'll give you the keys". Whether or not the tank will actually be filled, the keys will be given - if we take it as a possible future, or as a direct factual consequence.But in reality, there's an "or not" always looming. So, can you REALLY call this a conditional?
Only if there's the "word" given, implicit, that it is something that HAPPENS before the keys are given. The expectation of the filled tank is what you are "contional" about, not the actual tank.

"will" is modal, it expresses an intention here. And it should not be cut into "'ll", formally.

There's no difference in meaning if we pose the case as "Will you fill the tank? Here, have the keys". It's clearer that the "will" here expresses the request of confirmation of intentions, an actual human "will" in this example. And that is what is relevant for the condition.

All there's here its a showing of examples that ALL fall in this category, which i PLENTY made arguments against, but all but one (that i noticed, at least) argued in response, but just dismiss or just go off a tangent.


A It's not a shame to admit that you were not wrong. Guys really tried to explain that to you in polite and easy to understand way.

This is the thing: they didn't. "You are wrong. Unrelated reason". Or simply "You are wrong, unrelated example".


No, they don't, no matter how many times you randomly capitalize something.

Care to explain why?


Here are some more examples:
"If winter comes at some future date, then the statement "winter is coming" is an accurate prediction."
"If winter does not come at some future date, then the statement "winter is coming" is not an accurate prediction."

... and neither of those can happen before the winter comes? When the winter finally comes, you'll have the answer? The condition has to come, before one or the other ends up realizing?

The prediction will be accurate or inaccurate only after the condition manifests.
I'm confused?


Expecting the prediction to be accurate or inaccurate is not required or relevant to whether or not these are valid conditional statements. And winter coming certainly doesn't come to pass prior to the prediction being made.

Except that the implication is not that "a statement has been made", but "is an accurate prediction" (or not)?
Let me help:

"If winter comes at some future date, then the statement "winter is coming" exists"

This is not a conditional, because the condition doesn't matter for the implication. It's a mistake.

Just like 2 + 2 > 0 is not an equivalen


The notion that if/then statements require some expectation of the If part being true is entirely false.

No, it's just english.


In fact, you can even make valid if/then statements with the exact opposite expectation. For example:
"If you are really going to repay me, then there's no reason for me to not lend you the money. However, I do not believe that you will."

Again, whether or not you believe it influences b. Your expectation not coming to pass is what prevents the lending. Exactly like intended.
I'm still confused.


Crawford originally ruled "as with most bonus actions, you choose the timing, so the Shield Master shove can come before or after the attack action."

He later changed his mind: "if a feature says you can do X as a bonus action if you do Y, you must do Y before you can do X. For Shield Master, that means the bonus action must come after the Attack action. You can decide when it happens afterward that turn."

I think that makes it clear that:
a) there are two points of view

I've posted the link where there is the reason WHY he ruled in one way. There are also posts on the explanation on why he ruled otherwise. The first ruling being illogical might or might not have influenced the second, but it is still illogical, and demonstrably so. There are unlimited points of view. There are only a limited amounts of reasonable ones. In conversation, either you are going for an hyperbole to emphasize a point you already made, or you really really really should stick with reasonable ones.


b) "do it whenever you feel like, as long as you also take the Attack action" is a valid reading of RAW
c) "do it after you complete the Attack action" is a valid reading of RAW
d) per Crawford, B is now deprecated and C is now official

And this is again the mistake. b) was not the reason given for the ruling. The reason was "because like most bonus action, it can go whenever." Except it is not "like most bonus action", the rule in the phb, but "unless there's a clear timing", another rule in the phb. If said timing allows you to go before, because take can mean "declare" or whatever, then you can go at any point "as long as you take [...]", because the timing specified allows to".

The fact that JC corrected two things doesn't make the illogicity of the first ruling dependant on what "take" means, just on that "conditionals have timing".
He also cleared what that timing IS, but that's secondary.
This is prehaps the clearest i can make it.


e) Crawford isn't your DM so you can ignore D if you wish.

Agreed.

[Zhorn]

...
c) "do it after you complete the Attack action" is a valid reading of RAW
...
C is not official.
:smallconfused: Not really your call to make on that one.


http://dnd.wizards.com/articles/sage-advice/sage-advice-compendium
"We gather your D&D rules questions and occasionally provide official answers to them in the Sage Advice Compendium."

https://media.wizards.com/2019/dnd/downloads/SA-Compendium.pdf
Page 1:
"Official Rulings
Official rulings on how to interpret rules are made here in the Sage Advice Compendium by the game’s lead rules designer, Jeremy Crawford"

Page 8
"Shield Master
The Shield Master feat lets you shove someone as a bonus action if you take the Attack action. Can you take that bonus action before the Attack action?
No. The bonus action provided by the Shield Master feat has a precondition: that you take the Attack action on your turn. Intending to take that action isn’t sufficient; you must actually take it before you can take the bonus action. During your turn, you do get to decide when to take the bonus action after you’ve taken the Attack action. This sort of if-then setup appears in many of the game’s rules. The “if” must be satisfied before the “then” comes into play."

Now you don't have to like this ruling, that is perfectly fine. There are a lot of people that disagree with it.
You don't have to play by it, and that is also fine. WoTC often encourage people to take the game's framework and make it their own. Add house rules, change rulings, throw out entire sections if you want. All the more power to you.
But if the company that owns and publishes the intellectual property says something is official, it's official, and it comes across as childish to deny that just because it's not the ruling you want. You're a smart cookie, and that sort of behaviour should be beneath you.

Now, I agree that the language in the book is a little open to interpretation (it's why this question pops up so often). I'm also aware of the official ruling being flip-flopped over the past few years. These points do not invalidate the validity of the most recent ruling.
It could change in the future to be different. It might stay the exact same. But the point is, the current official ruling IS the Attack Action must come first.


unless we're at Crawford's table we can do it the original way or the new way or anything in between. Beyond that, it's just arguing over how many angels can dance on a pinhead.
Heh, @Laserlight I like your closer :smallbiggrin:

[NaughtyTiger]

Not really your call to make on that one.

Totally think we got crossways on that one.

C - suggested that you had to complete the Attack Action
That is not required by the text you specified.

1) sheild bash before the attack specifically forbidden by errata
2) complete the Attack Action, then shield bash
3) perform the first attack, then sheild bash, then finish the rest of the attack

[NaughtyTiger]

I'm not mad.

nothing says "not mad" like a 3/4 page post.

[Aimeryan]

I wasn't saying "other readings don't exist", I was saying "we know these two are both legitimate readings of RAW because Crawdaddy said so, at one time or another."

The point being that the ambiguity is in the text; Crawford has issued a ruling, so we know what the Official Way is now (unless he changes his mind again); and unless we're at Crawford's table we can do it the original way or the new way or anything in between. Beyond that, it's just arguing over how many angels can dance on a pinhead.

Crawford tweet isn't official. Besides, something can be RAW without need Crawford to hold your hand.

The text in the Sage Advice errata modifies the RAW, and that only clarifies that you can't just intend to take the Attack action, you have to commit. Which then leads to the three readings that I mentioned:

1) You bear the cost of the taking the action, that is, you no longer have that action available for other uses. The idea here is that you have taken the action by paying the cost - you need not have yet received any of the benefits of paying that cost. An analogy would be paying the shopkeeper for some goods, but having yet been given the goods. On the counter is a stand of lollipops; you are allowed to take one when you have bought something at the shop - you have paid for something at the shop, so being able to take the lollipop should be fine.

2) You bear the cost of taking the action, plus you must have received at least some of the direct benefits. The idea here is that there is a more firmer connection to having taken the action. The same analogy as the above, except some of the goods have been handed over, and then you take the freebie, with the remaining goods to follow.

3) You bear the cost of taking the action, plus you must have fully completed the transaction. The idea here is that it is unarguable that you have gone through with the action and some people like such simple solutions. The analogy above is changed here to you only being able to take a lollipop after every thing else has been handed over, for some reason.

[Zhorn]

Totally think we got crossways on that one.

C - suggested that you had to complete the Attack Action
That is not required by the text you specified.

1) sheild bash before the attack specifically forbidden by errata
2) complete the Attack Action, then shield bash
3) perform the first attack, then sheild bash, then finish the rest of the attack

I sincerely apologise if I have come across as misrepresenting your opinion on the matter. I hadn't included the previous post because I was only intending to respond directly to the "not official" component. I had been catching up of the several posts since my last and have seen the continued stance of

i read it as specific meaning that coupled with "you can take a bonus action at any time during your turn" means "before your attack action is valid"
coupled with

I am not convinced the promise to the DM is excluded.
as an indication you are/were still pushing that the bonus action shove can occur before the Attack Action.

A 'Valid' Attack Action vs 'Complete' Attack Action, there is nuance there, and I may have jumped the gun on responding quickly (relative to my perspective from reading-to-posting).
I did take a cursory look for an Sage Advice ruling on Bonus Actions inside Attack Actions, but so far have only come across the twitter-only responses, which do contradict each other on whether the Bonus Action 'ends' the Attack Action (regardless of completeness), or can be slotted inside without 'completing' the Attack Action (if anyone can site a recent source; that'd be awesome).

[LudicSavant]

No, it's just english. It has been explained to you quite thoroughly by multiple posters why this is not the case.

Any If X, Then Y statement in English requires no presumption that statements X or Y are true statements, or that X or Y happen in any chronological order, only that statement Y is a true statement if statement X is a true statement (which it may or may not be).

Also, if you care about speaking English properly, it's "English" not "english," "perhaps" not "prehaps," and "dependent" not "dependant." I can't even chalk this sort of thing up to typos because you do it so consistently. Not to mention... stranger errors.


No, it's just english.

english

english.

Prehaps

prehaps

prehaps

dependant

dependant

english is dependant

which i PLENTY made arguments against

[NaughtyTiger]

All good.


as an indication you are/were still pushing that the bonus action shove can occur before the Attack Action.


Ah. No that is a very specific thing.
I had conceded that BEFORE was no longer an option because of errata. PolarBear wouldn't let me concede that.




it is moot, now, cuz the RAW has been updated in an errata that the first case is explicitly barred.
Still confusing things? RAW has not changed. RAW has NEVER changed.

[jas61292]

Any If X, Then Y statement in English requires no presumption that statements X or Y are true statements, or that X or Y happen in any chronological order, only that statement Y is a true statement if statement X is a true statement (which it may or may not be).
This right here tells me that we are dealing with semantics here. Yes, it is absolutely possible to have a grammatically correct English sentence that involves an "if/then" conditional where the "then" clause chronologically comes before the "if" clause. That is because being grammatically correct has nothing to do with logical validity.

"If it is raining, then elephants rampaged down the street yesterday."

That is a nonsensical statement. It is proper English, and clearly a conditional where the consequent happened first, but it is not valid logic. The premise does not necessitate the result.

You will find that this will be true for any case where the result comes first chronologically, because rather than having a result come from an event, it is assuming the cause of an event that already occurred. The only way that a statement like that can be valid is if it is not simply an "if/then" statement, but an "if and only if" statement. Furthermore "if and only if X, then Y" is the logical equivalent of "if X then Y AND if Y then X."

In other words, "if X then Y" where Y comes before X, can only be valid when "if Y then X" is also valid, meaning that, despite the way it is being displayed, the past event is still what caused the future one.

[LudicSavant]

This right here tells me that we are dealing with semantics here. The definition given is actually the one from logic, not grammar. It just happens to also be grammatically correct.

Incidentally, semantics means "the branch of linguistics and logic concerned with meaning." Consider that when you are downplaying semantics.




"If it is raining, then elephants rampaged down the street yesterday."

That is a nonsensical statement. It is proper English, and clearly a conditional where the consequent happened first, but it is not valid logic. The premise does not necessitate the result.

You will find that this will be true for any case where the result comes first chronologically, because rather than having a result come from an event, it is assuming the cause of an event that already occurred. The only way that a statement like that can be valid is if it is not simply an "if/then" statement, but an "if and only if" statement. Furthermore "if and only if X, then Y" is the logical equivalent of "if X then Y AND if Y then X."

In other words, "if X then Y" where Y comes before X, can only be valid when "if Y then X" is also valid, meaning that, despite the way it is being displayed, the past event is still what caused the future one.

This is a fallacious argument. And also a logically invalid one, in the actual, formal sense of the phrase (the conclusion does not follow from the premises).

Your conclusion does not follow from your premise, because the ability to construct an invalid statement does not demonstrate an inability to construct a valid one.

Moreover, numerous examples of valid ones were already given to you, by numerous posters. You could find more by opening up any logic textbook and flip to the explanation of how if/then statements work.

[jas61292]

The definition given is actually the one from logic, not grammar. It just happens to also be grammatically correct.

Ok. So your point is that you can formulate logical statement that are not logically valid. This is true. Irrelevant, but true.

[LudicSavant]

Ok. So your point is that you can formulate logical statement that are not logically valid. This is true. Irrelevant, but true.

You are misusing the term "logically valid," and misunderstanding how a conditional statements works in logic.

The material conditional statement p → q does not conventionally specify a causal relationship between p and q; "p is the cause and q is the consequence from it" is not a generally valid for interpretation p → q). It merely means "if p is true, then q is also true" such that the statement p → q is false only when both p is true and q is false. (https://www.fecundity.com/codex/forallx.pdf)

[jas61292]

This is a fallacious argument. And also a logically invalid one, in the actual, formal sense of the phrase (the conclusion does not follow from the premises).

Your conclusion does not follow from your premise, because the ability to construct an invalid statement does not demonstrate an inability to construct a valid one.

Of course that statement is logically invalid. That is the entire point of me saying it. The entire reason I wrote it was to show that, yes, you can formulate an if/then statement in English that is not valid.

What you seem to be missing is what has been said to you numerous times: No, there had not been a single valid example given where the past event was the "then" clause. Some may appear to be like that, but every one is either not valid, or is obfuscating the fact that the premise comes first by using English and the implied meanings that come with it, rather than logical notation.

[LudicSavant]

Of course that statement is logically invalid. That is the entire point if me saying it. I didn't say that the statement is logically invalid, I said that your argument is, and explained why.

Additionally:
The material conditional statement p → q does not conventionally specify a causal relationship between p and q; "p is the cause and q is the consequence from it" is not a generally valid interpretation for p → q). It merely means "if p is true, then q is also true" such that the statement p → q is false only when both p is true and q is false. (https://www.fecundity.com/codex/forallx.pdf)

Additionally:
Please note what logically valid and logically invalid mean. They actually mean something very specific (https://en.wikipedia.org/wiki/Validity_(logic)), and do not apply to any single statement, but a set of premises and conclusion.

:vaarsuvius:

[jh12]

No, there had not been a single valid example given where the past event was the "then" clause. Some may appear to be like that, but every one is either not valid, or is obfuscating the fact that the premise comes first by using English and the implied meanings that come with it, rather than logical notation.

How about one from the Shield Master feat itself?

"If you are subjected to an effect that allows you to make a Dexterity saving throw to take only half damage, you can use your reaction to take no damage if you succeed on the saving throw, interposing your shield between yourself and the source of the effect."

"If/Then" statements, and especially "If" statements, are about relationships between the clauses, not just about chronological relationships. "If the jury finds me guilty, it doesn't change the fact that I didn't do it" is both grammatically and logically correct. So is, "if you use your Action to Attack, you can make a melee attack, a ranged attack, cast a spell, or use your bonus action to shove someone with you shield." It just doesn't happen to be the way the official rules work now.

[jas61292]

Also, you are continuing to use the phrase "logically invalid" wrong! It actually means something very specific (https://en.wikipedia.org/wiki/Validity_(logic)), and does not apply to any single statement, but a set of premises and conclusion.

I know what it means and that is exactly how I'm using it. The only thing I said that is arguably non valid is that the lack of any given example being valid means there are no valid examples. But that was not the full extent of my argument. My argument went beyond basic logic to the idea of causality.

Yes, P→Q does not mean that P caused Q. It means that P being true means that Q is true. What I meant by going beyond basic logic is that In looking at the universe and acknowledging that any given event can have a multitude of different causes, and so the states predating that event can have many different forms. Looking at a given state, it is impossible to know what caused it, and thus impossible to know what states predated it, unless this particular state has only one possible cause.

So if P implies Q (even if it doesn't cause Q) and Q comes before P, then either Q implies P as well (the aforementioned "if and only if" scenario) or the statement is invalid because the state that causes P does not necessarily involve Q, and thus P could be true without Q having been being true, invalidating the original statement.

Now this is all assuming we're not entertaining the idea of retrocausality. But seeing as that only exists is philosophy and fiction, I feel comfortable rejecting it.

[LudicSavant]

My argument went beyond basic logic to the idea of causality. Which is why your argument is wrong.

"It is important to remember that the connective ‘→’ says only that, if the antecedent is true, then the consequent is true. It says nothing about the causal connection between the two events. (https://www.fecundity.com/codex/forallx.pdf)"

If p→q necessitated a causal relationship such as the one you describe, logicians would not keep stressing that it does not necessitate a causal relationship and that you should be very careful to avoid thinking that it does.

Note: For those unfamiliar with the study of logic, the notation ‘→’ means "If (thing before arrow), then (thing after arrow)." Also "antecedent" is the if statement, and "consequent" is the then statement.

So what this is saying is that it is important to remember that there is no causal connection between the events. Or, in other words, that what PolarBear and JAS are saying is false. As basically everyone else in the thread has already pointed out.

Please, if you have not studied logic, please stop spreading disinformation about how logic works. For the love of all that is good and factually accurate. :vaarsuvius:

Edit:

Here is yet another example of an if/then statement where the antecedent statement occurs after the consequent statement:

"If 2+2=4 when the sun explodes, then 2+2 equaled 4 when the sun was born."

"2+2=4 when the sun explodes" takes place after "2+2 equaled 4 when the sun was born." And it does indeed logically follow that if the antecedent is true, then the consequent is true, therefore making it a proper material conditional statement.

QED. Again. :vaarsuvius:

[jas61292]

Which is why your argument is wrong.

It is important to remember that the connective ‘→’ says only that, if the antecedent is true, then the consequent is true. It says nothing about the causal connection between the two events. (https://www.fecundity.com/codex/forallx.pdf)

Note: For those unfamiliar with the study of logic, the notation ‘→’ means "If (thing before arrow), then (thing after arrow)." Also "antecedent" is the if statement, and "consequent" is the then statement.

So what this is saying is that it is important to remember that there is no causal connection between the events. Or, in other words, that what PolarBear and JAS are saying is false.

No, this right here is why you are wrong. All you are saying is that it does not necessarily imply causation. That is true, but it is not the whole picture. It can, and often does involve causation. And when you look at situations like this, causation is necessary. Not necessarily between P and Q, but between some state and both. Maybe Q happens if and only if X happens and X also causes P, so that even though Q does not cause P, it necessarily implies it.

And again, the reason for this is the timeline. There are four possible courses of events involving P and Q where Q would come before P. They would look something like this (arrows are time flow):

1. >>>> Q >>>> P >>>>
2. >>>> >>>> P >>>>
3. >>>> Q >>>> >>>>
4. >>>> >>>> >>>>

P→Q, which can also be expressed as "[not P] or Q" is valid only if situation 2 cannot happen. But as I expressed in an my previous post, in this universe, there are typically multiple states that can proceed any given state. It is impossible to know from a state alone which states preceded it, unless there is a specific, necessary set of states that must always proceed it, one of which will be the cause.

Specifically, either we know that the states that must necessarily proceed P include Q (in which case Q→P), or there is a set of states that can proceed P that does not involve Q, in which case situation 2 CAN happen, and thus P does not imply Q.

This is only true when Q comes before P because when it does not, the states that must necessarily proceed P are irrelevant.


Please, if you have not studied logic, please stop spreading disinformation about how logic works. For the love of all that is good and factually accurate. :vaarsuvius:

I have taken (and done well in) multiple college logic courses. I know what I'm talking about.

[LudicSavant]

No, this right here is why you are wrong. All you are saying is that it does not necessarily imply causation. That is true

If all I am saying is that it does not necessarily imply causation, and that is true, then what I am saying is necessarily not wrong.

That statement is even provable in formal deduction :smalltongue:


but it is not the whole picture. It can, and often does involve causation.

Which is irrelevant when discussing the truth value of the claim, and I quote:

"english requires x to be a thing before y can exist in a conditional statement. ALWAYS."

If there is even one exception, the statement is false. In this case, there are exceptions (per your own admission), therefore the statement is false. QED. That should be the end of the discussion.


I know what I'm talking about.

If that were true, then you would realize that any exception falsifies an "always" statement, and that "there can be cases where p precedes q" is not a logical defense of the position "p ALWAYS precedes q."

[jas61292]

"english requires x to be a thing before y can exist in a conditional statement. ALWAYS."

If all you are actually trying to defend if that this statement is false, then I stand by my previous statement of "This is true. Irrelevant, but true." I was never trying to defend that statement.

What I've been trying to defend is "x must be a thing before y can exist in a valid conditional statement. ALWAYS." That it what my last post shows.

Again the fact that you can construct English sentences that are invalid conditional statements is meaningless.

[LudicSavant]

If all you are actually trying to defend if that this statement is false, then I stand by my previous statement of "This is true. Irrelevant, but true."

That is indeed all I have been saying since page 3. As long as you are willing to admit that PolarBear's statement is wrong, our conclusions are in agreement.

[jas61292]

That is indeed all I have been saying since page 3. As long as you are willing to admit that PolarBear's statement is wrong, our conclusions are in agreement.

Fair enough.

Edit: I do want to add that there is one kind of exception to my argument that you and another previous poster made me think of: the consequent being a tautology.

"If P then Q" is true, even if Q is in the past and Q does not imply P, if Q is simply the logical equivalent of True. As the if then statement is equivalent to "[not P] or Q", if Q is always necessarily true, then the statement is always sound. But in that situation, the statement is always meaningless as well.

[Chronos]

"If it is snowing right now in my city, then yesterday elephants rampaged down the streets."

This statement is both logically valid, and true. It is not snowing right now in my city. Elephants also did not rampage down the streets yesterday, and that doesn't matter. The only way that an if-then statement is false is if the antecedent is true but the consequence is false.

And how did we end up down this rabbit-hole, anyway?

[Tanarii]

1) sheild bash before the attack specifically forbidden by errata
2) complete the Attack Action, then shield bash
3) perform the first attack, then sheild bash, then finish the rest of the attackMuch more concise than the way I wrote the three.

Nitpick on 1), but isn't the Shield Master clarification merely a matter of Crawford ruling via tweet? Or did it make it to Sage Advice? Either way it's not errata.

I guess I could look ... :smallamused:

(Edit: ah okay, it's in the Sage Advice Compendium.)


And how did we end up down this rabbit-hole, anyway?Probably because initially the word "logical" seemed to be tossed around to justify a personal preference, without actual logic to back it up.

Otoh I learned something new with jas61292 last point out about "always Q = True" making "If P then Q" a meaningless (or at least not useful) statement. I'd never considered that before.

[OverLordOcelot]

This discussion in particular indicates that we cannot take Jeremy Crawford's tweets as indicative of the designer's intent for determining RAI (Rules As Intended), since he's made clearly contradictory statements about what the intent supposedly is. WOTC has a weird policy of trying to issue rule changes through sage advice but calling them 'clarifications' instead of changes, and he follows that, which means that his job depends on calling rule changes 'RAI' when they clearly weren't what the designer was thinking when the rule was written.

OP, if you're reading, everyone I know who runs AL games believes that the intent of the shield master feat is that you can knock someone down with your bonus action, then attack them while prone. No one thinks that the person who designed the feat sat down and thought "I want to create a feat that allows to to knock enemies prone to help other people's attacks but not your own", or the weird "I want to create a feat that allows people with multiattack to knock enemies prone after they do one regular attack" that Crawford is pushing now. Since it's pretty clear that Crawford's contradictory RAI claims are just weird corporate speak and the 'declare that you're doing an attack action, knock them down first' is consistent with actual RAW, no one loses any sleep over it and shield-using characters are happy.

[Tanarii]

OP, if you're reading, everyone I know who runs AL games believes that the intent of the shield master feat is that you can knock someone down with your bonus action, then attack them while prone.
Interesting. Because my AL personal experience was that before Crawford made a ruling, pretty much everyone ruled as you must take the attack action first before the bonus action. Crawfords initial "ruling" caused an uproar both online and in AL before it was finally generally (but not universally) accepted as the standard by DMs. And of course it never helped that ruling make sense in the face of the EK's War Magic ruling the exact opposite, that you must use the cantrip first.

Personal experiences clearly are personal.

[Xetheral]

"If it is snowing right now in my city, then yesterday elephants rampaged down the streets."

This statement is both logically valid, and true. It is not snowing right now in my city. Elephants also did not rampage down the streets yesterday, and that doesn't matter. The only way that an if-then statement is false is if the antecedent is true but the consequence is false.

And how did we end up down this rabbit-hole, anyway?

Your statement is true, but it's a conditional statement that is not intended to be an argument with a premise and a conculsion, so the concept of logical validity doesn't apply.

The rabbit hole appeared when the claim was made that all conditional statements in English have the antecedent temporally preceding the consequent. But it turned out that the claimant instead meant that for all causal conditional statements in English, it is possible to identify a (potentially implicit) antecedent that temporally preceeds the consequent. Since then the different sides of the debate have been talking past each other, with one side contesting the originally-stated claim and the other side defending the intended claim.

[Boci]

Interesting. Because my AL personal experience was that before Crawford made a ruling, pretty much everyone ruled as you must take the attack action first before the bonus action. Crawfords initial "ruling" caused an uproar both online and in AL before it was finally generally (but not universally) accepted as the standard by DMs. And of course it never helped that ruling make sense in the face of the EK's War Magic ruling the exact opposite, that you must use the cantrip first.

Personal experiences clearly are personal.

True. Doesn't hurt to quantify your AL expirience though. How many AL groups where you involved with where this issue needed to be ruled on by a DM? (Ditto for OverLordOcelot, how many is "everyone I know who runs AL games", ballpark).

[Tanarii]

True. Doesn't hurt to quantify your AL expirience though. How many AL groups where you involved with where this issue needed to be ruled on by a DM? (Ditto for OverLordOcelot, how many is "everyone I know who runs AL games", ballpark).My personal experience is three different game stores, although I've mostly DMd my personal (non-AL) campaign at them for the last several years. Not every game had Shield Masters, for that matter I can't even recall if I was an AL player or running my own campaign at the time. But I remember the ruling being heavily discussed by various AL DMs at all three at the time, who I always talk to as a matter of course about gaming matters.

My other personal experience is the new ruling seems to be upsetting AL players that have built characters around it far more than AL DMs. I consider that a fair concern on their part.

But again, personal experience. I merely share it as an example of how personal experience can differ. It's not supposed to be an argument about the validity of any interpretation.

[LudicSavant]

And of course it never helped that ruling make sense in the face of the EK's War Magic ruling the exact opposite, that you must use the cantrip first.

War Magic was previously ruled to work in either order, too (as well as it being said that the design and balancing intent was for such abilities to be used in any order). That got changed years later just like Shield Master did.

[Tanarii]

War Magic was previously ruled to work in either order, too (as well as it being said that the design and balancing intent was for such abilities to be used in any order). That got changed years later just like Shield Master did.
Yup. June 2016 said before or after for EK, revised when SA returned in Aug 2017.

More importantly there was no Shield Master ruling at that time. So I must be remembering debates based on Crawford tweets and the EK ruling.

[OverLordOcelot]

True. Doesn't hurt to quantify your AL expirience though. How many AL groups where you involved with where this issue needed to be ruled on by a DM? (Ditto for OverLordOcelot, how many is "everyone I know who runs AL games", ballpark).

A dozen and a half DMs, including me (total of about 40, but I don't know the opinions of the whole set). It's also a running joke for DMs and rulesy people to laugh at the idea of Crawford being authoritative. In any AL group where there's someone who takes the feat, the issue has to be ruled on by a DM, even if they just go with whatever the player wants to do.