This thread is really interesting and (for me) helps me understand the different ways of thinking that go into the various beliefs on this. For what it is worth, here are my 2 cents:
1) The in initial JC post was a standard ambiguous wording -->WAI type post. He has done tons of these, and this just seemed like any other.
2) The 2nd post smells of a nerf. I know the reasons they gave (if X then Y, etc), but if you take a step back and look at the fact that (1) he stated the intent in the first post, and indicated in the 2nd post that the intent did not change and (2) they are refusing to change the wording,. post an errata, then my gut tells me that the 'lets keep it consistent' thing was a facade, not their real intent. I think they looked at the bonus action knockdown/attach action combo that people were doing every round felt like abuse to them. But because he stated what the intent was in the first tweet, they know if they change the wording and it matches the 2nd tweet and not the first, everyone will know it is a nerf.
3) I think part of the reason people are so irritated with this ruling is because deep down they suspect it is a nerf and that fact pisses them off. NOTE: Im not saying the nerf was not needed. I get why it might be an issue (I also see why it might not...I dont have a strong opinion myself on the impact on gameplay/combat). My point is people dont like feeling lied too.
4) The whole weapon/improvised weapon/melee weapon/ranged weapon thing is a complete mess. Using the statement that if you attack with a melee weapon it is always a melee weapon attack but if you attack with an improvised weapon it is not just sounds silly. And saying that the only melee weapons are the ones on the list sounds even more silly. Especially when you say that improvised weapons ARE weapon during the attack action. If being on the list is the only qualification for something being a melee weapon, then you end up with some really odd edge cases. Here is an example of how slicing things this way can cause ridiculous issues.
Example: Great weapon Master
On your turn, when you score a critical hit with a melee weapon or reduce a creature to 0 hit points with one, you can make one melee weapon attack as a bonus action
NOTE: The 2nd part of GMW explicitly says 'makes a melee attack', which is different then scoring a hit with a a melee weapon per their rules.
So using their logic, if I make an improvised attack with a melee weapon (because its on the list!), it is still a melee weapon and can trigger anything a melee weapon can trigger
If you read this literally, using the info above, you can THROW your melee weapon (which would make it an improvised weapon attack) or hit someone with the pomel of your sword, but still get an extra attack if it happens to kill someone. Why do I point this out? Because if a melee weapon (on the table) can become an improvised weapon, then 'being on the table of melee weapons' is not how the attribute 'melee weapon' is set. It also doesn't jive with what a 'melee attack' is per the rules. It is ridiculous.
They have different kinds of attacks (melee/ranged/spell). They have different kinds of weapons (melee/ranged/improvised/natural/etc)
Saying 'if you make a melee weapon melee attack sounds ridiculous, but that is actually what you are doing. (vs an improvised weapon melee attack(beer bottle) or a natural weapon ranged attack(manticore stingers) or a melee weapon spell attack(green flame blade)). But it is far more accurate. and it would take away the ambiguity.
5) Saying you cannot split an attack with a bonus action when you can do both moves (attack/move/attack) and reactions (cast spell action/target counterspells/ caster counterspells the counterspell/determine the results of the casting) during actions is again both inconsistent and messy. again, the if X then Y reason does make sense IF that is what they were trying to do. But the fact that they won't errata the feat to reflect that (maybe because if they do that, they would have to admit it is a nerf?) again points to them refusing to admit they dont like the shove bonus action/attack action combo.
4) The whole weapon/improvised weapon/melee weapon/ranged weapon thing is a complete mess. Using the statement that if you attack with a melee weapon it is always a melee weapon attack but if you attack with an improvised weapon it is not just sounds silly. And saying that the only melee weapons are the ones on the list sounds even more silly. Especially when you say that improvised weapons ARE weapon during the attack action. If being on the list is the only qualification for something being a melee weapon, then you end up with some really odd edge cases. Here is an example of how slicing things this way can cause ridiculous issues.
Example: Great weapon Master
On your turn, when you score a critical hit with a melee weapon or reduce a creature to 0 hit points with one, you can make one melee weapon attack as a bonus action
NOTE: The 2nd part of GMW explicitly says 'makes a melee attack', which is different then scoring a hit with a a melee weapon per their rules.
So using their logic, if I make an improvised attack with a melee weapon (because its on the list!), it is still a melee weapon and can trigger anything a melee weapon can trigger
If you read this literally, using the info above, you can THROW your melee weapon (which would make it an improvised weapon attack) or hit someone with the pomel of your sword, but still get an extra attack if it happens to kill someone. Why do I point this out? Because if a melee weapon (on the table) can become an improvised weapon, then 'being on the table of melee weapons' is not how the attribute 'melee weapon' is set. It also doesn't jive with what a 'melee attack' is per the rules. It is ridiculous.
They have different kinds of attacks (melee/ranged/spell). They have different kinds of weapons (melee/ranged/improvised/natural/etc)
Saying 'if you make a melee weapon melee attack sounds ridiculous, but that is actually what you are doing. (vs an improvised weapon melee attack(beer bottle) or a natural weapon ranged attack(manticore stingers) or a melee weapon spell attack(green flame blade)). But it is far more accurate. and it would take away the ambiguity.
This is getting somewhat out of scope for this thread, but I just wanted to point out that while the rules on this may have been worded better, they're unambiguous as they are. They don't use "melee weapon melee attack", they use "melee attack with a melee weapon". I'm pretty sure there are instances of "weapon attack with a melee weapon" (which includes ranged attacks with thrown melee weapons, but not Booming Blade attacks, since those are "spell attacks", not "weapon attacks"), "attack with a ranged weapon" (which would include Magic Stone attacks, I think), etc. Basically, "[ranged/melee/unspecified] [weapon/spell/unspecified] attack [with a [melee/ranged/unspecified] weapon/unspecified]", where the first choice (ranged, melee, or unspecified) determines whether the attack has to be either at range or in melee to qualify, the second choice (weapon, spell, or unspecified) determines whether the attack has to be a "physical" ("weapon") attack, or a spell attack, to qualify, and the third choice ("with a ... weapon", or unspecified) determines whether the attack requires a weapon (i.e. no "raw" spells, nor unarmed attacks) to qualify, and the third choice's subchoice (melee, ranged, or unspecified) determines which type of weapon is required to qualify.
It's not exactly clear, some unfortunate wording was used (the same word for different concepts, for example), but it is unequivocal.
On your turn, when you score a critical hit with a melee weapon or reduce a creature to 0 hit points with one, you can make onemelee weapon attack as a bonus action
NOTE: The 2nd part of GMW explicitly says 'makes a melee attack', which is different then scoring a hit with a a melee weapon per their rules.
I can sorta see the point you're trying to make, relevant to Shield Master, but you have some faulty assumptions.
First, you are right that "hit with a melee weapon" and "make one melee weapon attack" are distinct things. Hitting with a melee weapon is a post-action conditional event. Making a melee weapon attack is an actual action.
Further, "with a melee weapon", "melee attack", and "melee weapon attack" are NOT the same thing. When an ability/action description states that the conditional is "with a melee weapon", it means that the ability/action just needs to involve using the melee weapon (USED as a melee weapon) for the ability/action to function. It is written this way because there are more ways to score a critical hit "with a melee weapon" than just the Attack, and it is clear that all of different ways you can crit with a melee weapon fulfill the conditional statement of GWM.Green-Flame Blade is an example of this.
"Melee attack" does not necessarily require a weapon be used; unarmed strikes, natural weapons, and attacks with melee weapons are all melee attacks.
"Melee weapon attack" does require a weapon be used, that the weapon actually be a melee weapon, and is no more than a baseline standard attack.
So using their logic, if I make an improvised attack with a melee weapon (because its on the list!), it is still a melee weapon and can trigger anything a melee weapon can trigger
If you read this literally, using the info above, you can THROW your melee weapon (which would make it an improvised weapon attack) or hit someone with the pomel of your sword, but still get an extra attack if it happens to kill someone. Why do I point this out? Because if a melee weapon (on the table) can become an improvised weapon, then 'being on the table of melee weapons' is not how the attribute 'melee weapon' is set. It also doesn't jive with what a 'melee attack' is per the rules. It is ridiculous.
An object that bears no resemblance to a weapon deals 1d4 damage (the DM assigns a damage type appropriate to the object). If a character uses a ranged weapon to make a melee attack, or throws a melee weapon that does not have the thrown property, it also deals 1d4 damage. An improvised thrown weapon has a normal range of 20 feet and a long range of 60 feet.
I would not count a pommel strike as an improvised melee weapon. That's not an uncommon way to use the weapon, and frankly a ridiculous analogy because any "pommel strike" is just a flavorful description of a normal attack with the melee weapon. Mechanically, there are no instances where you would be able to "pommel strike", yet not be able to make a normal attack with the weapon. You can trigger GWM from this.
With Green-Flame Blade, you are taking the [Tooltip Not Found] (no such thing as "melee weapon spell attack"), that involves making a "melee attack" "with a weapon". If the melee attack results in a critical hit (or the creature drops to 0 HP), and the weapon used is an actual melee weapon used as a melee weapon, you can absolutely trigger GWM off it.
Throwing the weapon is completely different. You are improvising if the weapon does not have the thrown property. Weapons with the thrown property are ranged weapons. You are not making a melee attack, nor are you using the melee weapon as a melee weapon. You are making a ranged attack with an object as an improvised ranged weapon. You cannot trigger GWM from this.
5) Saying you cannot split an attack with a bonus action when you can do both moves (attack/move/attack) and reactions (cast spell action/target counterspells/ caster counterspells the counterspell/determine the results of the casting) during actions is again both inconsistent and messy. again, the if X then Y reason does make sense IF that is what they were trying to do. But the fact that they won't errata the feat to reflect that (maybe because if they do that, they would have to admit it is a nerf?) again points to them refusing to admit they dont like the shove bonus action/attack action combo.
Back to being on-topic, nobody is saying you can't split the Attack if you have Extra Attack. You can. What is pertinent to Shield Master is whether or not you have to actually "make an attack" before you can attempt to shove as a bonus action. TL;DR yeah, you do.
Attack
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.
Making an Attack
Whether you're striking with a melee weapon, firing a weapon at range, or making an attack roll as part of a spell, an attack has a simple structure.
1. Choose a target. Pick a target within your attack's range: a creature, an object, or a location.
2. Determine modifiers. The DM determines whether the target has cover and whether you have advantage or disadvantage against the target. In addition, spells, special abilities, and other effects can apply penalties or bonuses to your attack roll.
3. Resolve the attack. You make the attack roll. On a hit, you roll damage, unless the particular attack has rules that specify otherwise. Some attacks cause special effects in addition to or instead of damage.
If there's ever any question whether something you're doing counts as an attack, the rule is simple: if you're making an attack roll, you're making an attack.
I do think Crawford basically got caught with his pants down twice over this, and the "If X then Y" schema stands.
I don't like it any more than anyone else, but that's how it is. Homebrew to hearts' content.
Rollback Post to RevisionRollBack
You don't know what fear is until you've witnessed a drunk bird divebombing you while carrying a screaming Kobold throwing fire anywhere and everywhere.
After getting caught up on the thread, I'd like to wade in here; not trying to say anyone else is wrong, but rather how I take the RAW and make a ruling.
The rules of the Attack action include the following: “With this action, you make one melee or ranged attack.”
I'd say that the Making an attack section is irrelevant because it applies only to an attack that occurs, whether it be ranged, melee, or spell, not to the Attack action itself which can have any number of attacks or other activities granted by features.
Also, I feel that the RAW does not inarguably define that the one attack granted by the default Attack action must be ordered first amongst all the other possible feature-granted attacks or activities that may fall within the same Attack action.
The SA ruling also does not address whether the bonus action may be taken during the Attack action, only that “you do get to decide when to take the bonus action after you’ve taken the Attack action”. Again, not that the Attack action must be completed to any extent, only that it be “taken”.
The movement-between-attacks rules also indicate that the attacks may be staggered at the player's discretion amongst movement, further supporting the interpretation that the attack is not unalterably immediate once the Attack action has been declared.
Why is this relevant? Because the feat does not specify when the bonus action must be taken, only that the clause “If you take the Attack action on your turn” must be satisfied. From that moment, it seems to be fair game to execute at the players’ pleasure.
Thus, here is how I rule it for now:
If you have irreversibly declared the Attack action, then you may at any point thereafter, including immediately, activate the feature of Shield Master and shove using the bonus action.
Afterward, the Attack action continues to resolve with however many attacks remain.
Always happy to hear critiques and counterpoints! It's all up for interpretation and the conversation is good stuff!
On your turn, when you score a critical hit with a melee weapon or reduce a creature to 0 hit points with one, you can make onemelee weapon attack as a bonus action
NOTE: The 2nd part of GMW explicitly says 'makes a melee attack', which is different then scoring a hit with a a melee weapon per their rules.
I can sorta see the point you're trying to make, relevant to Shield Master, but you have some faulty assumptions.
First, you are right that "hit with a melee weapon" and "make one melee weapon attack" are distinct things. Hitting with a melee weapon is a post-action conditional event. Making a melee weapon attack is an actual action.
Further, "with a melee weapon", "melee attack", and "melee weapon attack" are NOT the same thing. When an ability/action description states that the conditional is "with a melee weapon", it means that the ability/action just needs to involve using the melee weapon (USED as a melee weapon) for the ability/action to function. It is written this way because there are more ways to score a critical hit "with a melee weapon" than just the Attack, and it is clear that all of different ways you can crit with a melee weapon fulfill the conditional statement of GWM.Green-Flame Blade is an example of this.
"Melee attack" does not necessarily require a weapon be used; unarmed strikes, natural weapons, and attacks with melee weapons are all melee attacks.
"Melee weapon attack" does require a weapon be used, that the weapon actually be a melee weapon, and is no more than a baseline standard attack.
So using their logic, if I make an improvised attack with a melee weapon (because its on the list!), it is still a melee weapon and can trigger anything a melee weapon can trigger
If you read this literally, using the info above, you can THROW your melee weapon (which would make it an improvised weapon attack) or hit someone with the pomel of your sword, but still get an extra attack if it happens to kill someone. Why do I point this out? Because if a melee weapon (on the table) can become an improvised weapon, then 'being on the table of melee weapons' is not how the attribute 'melee weapon' is set. It also doesn't jive with what a 'melee attack' is per the rules. It is ridiculous.
An object that bears no resemblance to a weapon deals 1d4 damage (the DM assigns a damage type appropriate to the object). If a character uses a ranged weapon to make a melee attack, or throws a melee weapon that does not have the thrown property, it also deals 1d4 damage. An improvised thrown weapon has a normal range of 20 feet and a long range of 60 feet.
I would not count a pommel strike as an improvised melee weapon. That's not an uncommon way to use the weapon, and frankly a ridiculous analogy because any "pommel strike" is just a flavorful description of a normal attack with the melee weapon. Mechanically, there are no instances where you would be able to "pommel strike", yet not be able to make a normal attack with the weapon. You can trigger GWM from this.
With Green-Flame Blade, you are taking the Cast a Spell (no such thing as "melee weapon spell attack"), that involves making a "melee attack" "with a weapon". If the melee attack results in a critical hit (or the creature drops to 0 HP), and the weapon used is an actual melee weapon used as a melee weapon, you can absolutely trigger GWM off it.
Throwing the weapon is completely different. You are improvising if the weapon does not have the thrown property. Weapons with the thrown property are ranged weapons. You are not making a melee attack, nor are you using the melee weapon as a melee weapon. You are making a ranged attack with an object as an improvised ranged weapon. You cannot trigger GWM from this.
5) Saying you cannot split an attack with a bonus action when you can do both moves (attack/move/attack) and reactions (cast spell action/target counterspells/ caster counterspells the counterspell/determine the results of the casting) during actions is again both inconsistent and messy. again, the if X then Y reason does make sense IF that is what they were trying to do. But the fact that they won't errata the feat to reflect that (maybe because if they do that, they would have to admit it is a nerf?) again points to them refusing to admit they dont like the shove bonus action/attack action combo.
Back to being on-topic, nobody is saying you can't split the Attack if you have Extra Attack. You can. What is pertinent to Shield Master is whether or not you have to actually "make an attack" before you can attempt to shove as a bonus action. TL;DR yeah, you do.
Attack
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.
Making an Attack
Whether you're striking with a melee weapon, firing a weapon at range, or making an attack roll as part of a spell, an attack has a simple structure.
1. Choose a target. Pick a target within your attack's range: a creature, an object, or a location.
2. Determine modifiers. The DM determines whether the target has cover and whether you have advantage or disadvantage against the target. In addition, spells, special abilities, and other effects can apply penalties or bonuses to your attack roll.
3. Resolve the attack. You make the attack roll. On a hit, you roll damage, unless the particular attack has rules that specify otherwise. Some attacks cause special effects in addition to or instead of damage.
If there's ever any question whether something you're doing counts as an attack, the rule is simple: if you're making an attack roll, you're making an attack.
I do think Crawford basically got caught with his pants down twice over this, and the "If X then Y" schema stands.
I don't like it any more than anyone else, but that's how it is. Homebrew to hearts' content.
I think we are on the same page on most of this actually.
My comment about the melee weapon thing was in regard to this tweet by JC. It implies that the only thing that makes a melee weapon a melee weapon is whether it is on the table. Which is patently abusrd (thus my absurd thrown sword thing)
As for the bonus action between acitons thing, I believe you and I both agree that IF you say it requires an attack action, and you have mulitiple attacks, you can use the shove after one attack. This is not supported by comments online though.
If you just read the first tweet, then you can read is simply as 'you need to make an attack to do the shove, so you can do a bonus action after 1 attack'. Thats how i read it, and I think thats how you read it. But the second tweet comment say that you have to COMPLETE the attack action before you trigger the X then Y for shield master. It all goes to the wording 'take the attack action'. As someone earlier in the thread pointed out, that is super ambiguous because what does it mean to take an attack action. Do I have to say I am attacking, make an attack, complete all attacks? Any of those could be valid with the wording. according to folks online (yes I know, dont trust the internet) that was determined to be after completing the action. The RAw/RAI on this is just ugly, even after it was 'clearedd up'. Twice.
And that sort of makes my 1st point. It would NOT be hard for them to reword this in the eratta to make it crystal clear. The fact that they are refusing to do so combined with the reversing of the ruling makes me think they are choosing not to for a reason. And the only reason I can come up with (I admit I might be missing something) is that they would end up nerfing the feat explicitly, which would cause a much larger backlash (vs the twitter storm and us rules nerd talking about it on forums which is what happens now.
If you just read the first tweet, then you can read is simply as 'you need to make an attack to do the shove, so you can do a bonus action after 1 attack'. Thats how i read it, and I think thats how you read it. But the second tweet comment say that you have to COMPLETE the attack action before you trigger the X then Y for shield master. It all goes to the wording 'take the attack action'. As someone earlier in the thread pointed out, that is super ambiguous because what does it mean to take an attack action. Do I have to say I am attacking, make an attack, complete all attacks? Any of those could be valid with the wording. according to folks online (yes I know, dont trust the internet) that was determined to be after completing the action. The RAw/RAI on this is just ugly, even after it was 'clearedd up'. Twice.
And that sort of makes my 1st point. It would NOT be hard for them to reword this in the eratta to make it crystal clear. The fact that they are refusing to do so combined with the reversing of the ruling makes me think they are choosing not to for a reason. And the only reason I can come up with (I admit I might be missing something) is that they would end up nerfing the feat explicitly, which would cause a much larger backlash (vs the twitter storm and us rules nerd talking about it on forums which is what happens now.
Wow, you're right about that second tweet. Unfortunately, that makes everything even worse than it already is... and actually makes the entirety of Making an Attack the central premise now.
The Attack action is not resolved until "making an attack" has been resolved.
"Making an attack" is not resolved until the attack has been rolled.
Ergo, the Attack action is not resolved until the attack has been rolled.
Bonus actions cannot interrupt multiple attacks during the Attack action if the trigger to do so is the Attack action itself.
Ergo, the Attack action is not resolved until ALL attacks has been rolled (or by forfeiting any remaining attacks, I would suppose).
Ergo, the bonus action Shove from Shield Master cannot be used until the ENTIRE Attack action has been resolved.
In the GWM analogy, you would be able to interrupt your Attack action for the bonus action because the trigger is based on the individual attack (via critical or KO), not the Attack action itself, but Shield Master is SoL.
Maybe we should start a running tally of how many times Crawford gets caught with his pants down when it comes to this feat...
You don't know what fear is until you've witnessed a drunk bird divebombing you while carrying a screaming Kobold throwing fire anywhere and everywhere.
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.
He is saying that there is no general rule allowing bonus actions to interrupt the Attack, and that prohibition is the default state, unless a specific feature allows sequence breaking.
Rollback Post to RevisionRollBack
You don't know what fear is until you've witnessed a drunk bird divebombing you while carrying a screaming Kobold throwing fire anywhere and everywhere.
Ahh, I get you! I do think it's important to note that Crawford's tweets are not rulings. Both he and WotC have clearly stated that to be the case. Only published SAs carry the weight of actual rules. Crawford's posts have been relegated to "advice" and "preview of rulings that will appear".
With that in mind, the official SA ruling states the following: “you do get to decide when to take the bonus action after you’ve taken the Attack action”. To me that means everything hangs on whether "taken" is interpreted to mean completed or something else, for example as I rule at my table: irreversibly declared.
Really hope they publish an errata of some sort to this effect! (Or have they that I've overlooked?)
2) The 2nd post smells of a nerf. I know the reasons they gave (if X then Y, etc), but if you take a step back and look at the fact that (1) he stated the intent in the first post, and indicated in the 2nd post that the intent did not change and (2) they are refusing to change the wording,. post an errata, then my gut tells me that the 'lets keep it consistent' thing was a facade, not their real intent. I think they looked at the bonus action knockdown/attach action combo that people were doing every round felt like abuse to them. But because he stated what the intent was in the first tweet, they know if they change the wording and it matches the 2nd tweet and not the first, everyone will know it is a nerf.
That's silly. He can't nerf the game through tweets; if the way Shield Master worked was considered a problem it'd be changed in errata. And if you read enough of his tweets or watch a couple of Dragon+/Dragon Talk episodes with him it's pretty obvious that he's not very concerned with what min/maxers do as long as most players find the game enjoyable.
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.
He is saying that there is no general rule allowing bonus actions to interrupt the Attack, and that prohibition is the default state, unless a specific feature allows sequence breaking.
Official Rulings Official rulings on how to interpret unclear rules are made in Sage Advice. The public statements of the D&D team, or anyone else at Wizards of the Coast, are not official rulings; they are advice. One exception: the game’s rules manager, Jeremy Crawford (@JeremyECrawford), can make official rulings and usually does so in Sage Advice and on Twitter.
Only published SAs carry the weight of actual rules.
I know this is a nitpick but since this is a thread about dissecting the rules, I have to point out that Sage Advice Compendium has never carried the weight of actual rules. The only official rules for D&D are the 3 core books. Sage Advice is basically an FAQ.
2) The 2nd post smells of a nerf. I know the reasons they gave (if X then Y, etc), but if you take a step back and look at the fact that (1) he stated the intent in the first post, and indicated in the 2nd post that the intent did not change and (2) they are refusing to change the wording,. post an errata, then my gut tells me that the 'lets keep it consistent' thing was a facade, not their real intent. I think they looked at the bonus action knockdown/attach action combo that people were doing every round felt like abuse to them. But because he stated what the intent was in the first tweet, they know if they change the wording and it matches the 2nd tweet and not the first, everyone will know it is a nerf.
That's silly. He can't nerf the game through tweets; if the way Shield Master worked was considered a problem it'd be changed in errata. And if you read enough of his tweets or watch a couple of Dragon+/Dragon Talk episodes with him it's pretty obvious that he's not very concerned with what min/maxers do as long as most players find the game enjoyable.
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.
He is saying that there is no general rule allowing bonus actions to interrupt the Attack, and that prohibition is the default state, unless a specific feature allows sequence breaking.
Official Rulings Official rulings on how to interpret unclear rules are made in Sage Advice. The public statements of the D&D team, or anyone else at Wizards of the Coast, are not official rulings; they are advice. One exception: the game’s rules manager, Jeremy Crawford (@JeremyECrawford), can make official rulings and usually does so in Sage Advice and on Twitter.
Only published SAs carry the weight of actual rules.
I know this is a nitpick but since this is a thread about dissecting the rules, I have to point out that Sage Advice Compendium has never carried the weight of actual rules. The only official rules for D&D are the 3 core books. Sage Advice is basically an FAQ.
Yes, and Shield Master has been clarified as to how if <-> then situations work in the context of the game rules, and anything that has an 'if' limiter must be completed before the 'then' can be performed. Because of that, bonus actions that require the completion of some 'if' have to be completed before being granted the bonus action, hence why you are not able to use it in between an attack; it isn't that there is a specific rule on bonus action timings, rather than the fact that most bonus actions have an 'if' limiter that automatically controls when they can be used.
Because of this, we can fully understand the design intent of the feat; It follows the theme of slowly pushing back your foe in a narrow hall, slowly gaining ground by pushing the assailants back with your shield over time.
It isn't and wasn't supposed to be a way for min-maxers to give themselves advantage on all their attacks, but rather supposed to be a defensive feat in all aspects.
Ahh, I get you! I do think it's important to note that Crawford's tweets are not rulings. Both he and WotC have clearly stated that to be the case. Only published SAs carry the weight of actual rules. Crawford's posts have been relegated to "advice" and "preview of rulings that will appear".
With that in mind, the official SA ruling states the following: “you do get to decide when to take the bonus action after you’ve taken the Attack action”. To me that means everything hangs on whether "taken" is interpreted to mean completed or something else, for example as I rule at my table: irreversibly declared.
Really hope they publish an errata of some sort to this effect! (Or have they that I've overlooked?)
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
The above is the up to date SA clarification: Intending to make the attack is not enough, you must fully take the attack action per the area I highlighted...such a clause would be unneeded if taking the attack action didn't require actually attacking.
I can definitely appreciate your interpretation of it, AtlaStar! Were I playing at your table, I'd not argue.
That said, as a DM, however, I interpret differently whether "taking an action" means specifically that said action must be completed beginning to end to satisfy the condition of Shield Master. I do not see anywhere in the published rules or in SA where it does so inarguably, so I won't decree to my players that they can't do it.
I hardly see it as a problem for min-maxers, because they will always find ways to eek out numbers. Rather I see the opposite as an arbitrary and prohibitive interpretation of what would otherwise be a sensible ability. 20th level fighter: "I can't bash with my shield first because why?!" Me: "Oh, you know: you have to swing your sword 4 times first."
And whether it's a defensively themed feat is debatable as well--though major kudos for raising that discussion point! I've seen it brilliantly used offensively to create attack angles for allies as well as to clear a path to attack a big bad. Being able to double shove is also astoundingly powerful, but no one ever tries that!
I can definitely appreciate your interpretation of it, AtlaStar! Were I playing at your table, I'd not argue.
That said, as a DM, however, I interpret differently whether "taking an action" means specifically that said action must be completed beginning to end to satisfy the condition of Shield Master. I do not see anywhere in the published rules or in SA where it does so inarguably, so I won't decree to my players that they can't do it.
I hardly see it as a problem for min-maxers, because they will always find ways to eek out numbers. Rather I see the opposite as an arbitrary and prohibitive interpretation of what would otherwise be a sensible ability. 20th level fighter: "I can't bash with my shield first because why?!" Me: "Oh, you know: you have to swing your sword 4 times first."
And whether it's a defensively themed feat is debatable as well--though major kudos for raising that discussion point! I've seen it brilliantly used offensively to create attack angles for allies as well as to clear a path to attack a big bad. Being able to double shove is also astoundingly powerful, but no one ever tries that!
If that is a problem, they can always use their first of many attacks to shove. Halving movement or causing forced movement is super strong, and knocking prone while getting all fighter attacks breaks the action economy...it is ironic though, because a lot of people find 5e to trivialize death and be easy, but half the reason it is too easy is that DM's allow things at the table that aren't intended.
If the feat said “taken” or “have taken” this wouldn’t be an issue. It says “take”, not “taken”.
If it is supposed to be an “If-then” statement, why leave out “then”.
I accept Crawford’s advice. If I want to change when the bonus action can be made, I accept that it would be a house rule.
I will not accept that take and taken are synonymous or that an invisible “then” exists in the description of the feat.
"if X, do Y" is an if <-> then logic statement. You have to take your action, not declare that you are taking it. Your argument is basically that having intent to do something is the same as doing it, which is not the case...if you take the attack action you make attacks and until you have made those attacks you haven't taken anything, just declared your intent to take that action which is not the same. It'd be like me saying "I am going to rob a bank" and getting arrested for robbing a bank...stating intent isn't the same as doing.
As to the specifics of why the "then" was left out, "if you take" is a future action...simple causality makes the implication clear that when the if is satisfied, the other conditional attached to the if is the "then." But yes, this is an issue with having the rules written in a natural language, but very few people liked 4th edition which was more rules and mechanics focused by removing the natural language aspects.
There are several ways that the writers could have written this so it would be unambiguous.
If you have taken ...
After you have taken ...
If you take ... , then ...
There is no ambiguity; the feat says "if you take the attack action on your turn" and nothing else.
Therefore you must take the attack action, which has a sequence of events you have to follow as part of the rules on making an attack...which includes resolving the attack.
If you don't roll to make an attack, you haven't taken the attack action on your turn. Period.
I'm now convinced that you don't understand the difference between present and past tense.
You don't need to have "taken" an attack action... you need to "take" one.
Words are hard, I get it... but please, buddy, just for a minute stop and really wrap your head around the difference between a present tense conditional trigger and a past tense conditional trigger. Shield Master is written with a present tense trigger. Ergo, the attack action DOES NOT NEED TO BE RESOLVED. You must be taking it. Take. Present.
According to the rules you linked, you must simply start the action by 1. Choose a target. Pick a target within your attack's range: a creature, an object, or a location.
So, you Choose yourself a target. At this point you have initiated the attack action. Then you immediately use a bonus action to Shove, resolve shove, and then resolve the original attack action. That's as RAW as the written rules get.
If state = [take attack action], then enable [bonus action shove] until end of turn.
If [take attack action], then sequence
Choose Target
Determine Modifier
Resolve
Changestate [taken attack action]
If you were to program the equivalent of the RAW about the relevant rules... you'd be able to take the shove bonus action at any of those sequence bullets. It'd be enabled from the moment the attack action sequence initiated because the conditional trigger is the present tense [take attack action] point and not the past tense [taken attack action] point.
If state = [take attack action], then enable [bonus action shove] until end of turn.
If [take attack action], then sequence
Choose Target
Determine Modifier
Resolve
Changestate [taken attack action]
If you were to program the equivalent of the RAW about the relevant rules... you'd be able to take the shove bonus action at any of those sequence bullets. It'd be enabled from the moment the attack action sequence initiated because the conditional trigger is the present tense [take attack action] point and not the past tense [taken attack action] point.
That is incorrect, because those 3 parts would be subroutines if anything. Those 3 parts make up an attack, as such all subroutines would have to return as the attack function was called.
So if programming it per RAW, you'd declare you were taking the attack action which would call the attack function. It would then call the target selection function, you'd select targets and it'd return with the object you selected which would be used to determine modifiers based on the target. After all that, it'd resolve. If extra attack is available, it'd give you the option to make additional attacks repeating the previous steps. Then and only then, does the attack function return.
You haven't taken an action until you start the process of completing the action. This is 100% how it is written, and I don't understand why people try to argue it says otherwise especially given that the Sage Advice makes it very clear that you don't get the bonus action until you actually make the attack.
There are several ways that the writers could have written this so it would be unambiguous.
If you have taken ...
After you have taken ...
If you take ... , then ...
There is no ambiguity; the feat says "if you take the attack action on your turn" and nothing else.
Therefore you must take the attack action, which has a sequence of events you have to follow as part of the rules on making an attack...which includes resolving the attack.
If you don't roll to make an attack, you haven't taken the attack action on your turn. Period.
I'm now convinced that you don't understand the difference between present and past tense.
You don't need to have "taken" an attack action... you need to "take" one.
Words are hard, I get it... but please, buddy, just for a minute stop and really wrap your head around the difference between a present tense conditional trigger and a past tense conditional trigger. Shield Master is written with a present tense trigger. Ergo, the attack action DOES NOT NEED TO BE RESOLVED. You must be taking it. Take. Present.
According to the rules you linked, you must simply start the action by 1. Choose a target. Pick a target within your attack's range: a creature, an object, or a location.
So, you Choose yourself a target. At this point you have initiated the attack action. Then you immediately use a bonus action to Shove, resolve shove, and then resolve the original attack action. That's as RAW as the written rules get.
You don't mix tenses...if you said if you have taken X on your turn, it then puts the entire context of the sentence into the past tense, including the part discussing your turn...being granted a bonus action once your turn has ended is pretty useless, hence to have a grammatically correct sentence you need to use present or future tense to indicate that once X has occurred, do Y.
To take the attack action means going through every step, otherwise you haven't taken that action, you've only shown the intent to take that action...once again, if I said "I am going to rob a bank," which is a statement of intent, it doesn't mean I deserve to get arrested for robbing a bank as the action hasn't actually occurred.
And finally, under the rules for making an attack.
If there's ever any question whether something you're doing counts as an attack, the rule is simple: if you're making an attack roll, you're making an attack.
If choosing a target was enough to be considered an attack, why include the fact that you need to be making an attack roll to be making an attack? So if you want to argue that you can use the bonus action in between attacks, fine go ahead...but it is pretty cut and dry that unless you are rolling to attack, you aren't making an attack.
EDIT: And since you wanted grammar pedantry, making is the present participle which can only be referring to something that is currently happening or happened in the past continuous tenses...and since you can not continue 'making' an attack roll as it is an instantaneous event where you roll and it completes, until you perform the act of rolling for an attack, you aren't attacking.
I agree with your conclusion but changing words and adding words in order to support your argument is not persuasive.
The text of Shield Master could be interpreted in different ways. Crawford has written in Sage Advice on how to interpret it in the context of the entire set of rules. If a DM doesn’t agree with Crawford, the DM can make a house rule for their own game.
That is really all that needs to be said.
Rollback Post to RevisionRollBack
To post a comment, please login or register a new account.
This thread is really interesting and (for me) helps me understand the different ways of thinking that go into the various beliefs on this. For what it is worth, here are my 2 cents:
1) The in initial JC post was a standard ambiguous wording -->WAI type post. He has done tons of these, and this just seemed like any other.
2) The 2nd post smells of a nerf. I know the reasons they gave (if X then Y, etc), but if you take a step back and look at the fact that (1) he stated the intent in the first post, and indicated in the 2nd post that the intent did not change and (2) they are refusing to change the wording,. post an errata, then my gut tells me that the 'lets keep it consistent' thing was a facade, not their real intent. I think they looked at the bonus action knockdown/attach action combo that people were doing every round felt like abuse to them. But because he stated what the intent was in the first tweet, they know if they change the wording and it matches the 2nd tweet and not the first, everyone will know it is a nerf.
3) I think part of the reason people are so irritated with this ruling is because deep down they suspect it is a nerf and that fact pisses them off. NOTE: Im not saying the nerf was not needed. I get why it might be an issue (I also see why it might not...I dont have a strong opinion myself on the impact on gameplay/combat). My point is people dont like feeling lied too.
4) The whole weapon/improvised weapon/melee weapon/ranged weapon thing is a complete mess. Using the statement that if you attack with a melee weapon it is always a melee weapon attack but if you attack with an improvised weapon it is not just sounds silly. And saying that the only melee weapons are the ones on the list sounds even more silly. Especially when you say that improvised weapons ARE weapon during the attack action. If being on the list is the only qualification for something being a melee weapon, then you end up with some really odd edge cases. Here is an example of how slicing things this way can cause ridiculous issues.
Example:
Great weapon Master
NOTE: The 2nd part of GMW explicitly says 'makes a melee attack', which is different then scoring a hit with a a melee weapon per their rules.
So using their logic, if I make an improvised attack with a melee weapon (because its on the list!), it is still a melee weapon and can trigger anything a melee weapon can trigger
If you read this literally, using the info above, you can THROW your melee weapon (which would make it an improvised weapon attack) or hit someone with the pomel of your sword, but still get an extra attack if it happens to kill someone. Why do I point this out? Because if a melee weapon (on the table) can become an improvised weapon, then 'being on the table of melee weapons' is not how the attribute 'melee weapon' is set. It also doesn't jive with what a 'melee attack' is per the rules. It is ridiculous.
They have different kinds of attacks (melee/ranged/spell).
They have different kinds of weapons (melee/ranged/improvised/natural/etc)
Saying 'if you make a melee weapon melee attack sounds ridiculous, but that is actually what you are doing. (vs an improvised weapon melee attack(beer bottle) or a natural weapon ranged attack(manticore stingers) or a melee weapon spell attack(green flame blade)). But it is far more accurate. and it would take away the ambiguity.
5) Saying you cannot split an attack with a bonus action when you can do both moves (attack/move/attack) and reactions (cast spell action/target counterspells/ caster counterspells the counterspell/determine the results of the casting) during actions is again both inconsistent and messy. again, the if X then Y reason does make sense IF that is what they were trying to do. But the fact that they won't errata the feat to reflect that (maybe because if they do that, they would have to admit it is a nerf?) again points to them refusing to admit they dont like the shove bonus action/attack action combo.
This is getting somewhat out of scope for this thread, but I just wanted to point out that while the rules on this may have been worded better, they're unambiguous as they are. They don't use "melee weapon melee attack", they use "melee attack with a melee weapon". I'm pretty sure there are instances of "weapon attack with a melee weapon" (which includes ranged attacks with thrown melee weapons, but not Booming Blade attacks, since those are "spell attacks", not "weapon attacks"), "attack with a ranged weapon" (which would include Magic Stone attacks, I think), etc. Basically, "[ranged/melee/unspecified] [weapon/spell/unspecified] attack [with a [melee/ranged/unspecified] weapon/unspecified]", where the first choice (ranged, melee, or unspecified) determines whether the attack has to be either at range or in melee to qualify, the second choice (weapon, spell, or unspecified) determines whether the attack has to be a "physical" ("weapon") attack, or a spell attack, to qualify, and the third choice ("with a ... weapon", or unspecified) determines whether the attack requires a weapon (i.e. no "raw" spells, nor unarmed attacks) to qualify, and the third choice's subchoice (melee, ranged, or unspecified) determines which type of weapon is required to qualify.
It's not exactly clear, some unfortunate wording was used (the same word for different concepts, for example), but it is unequivocal.
I can sorta see the point you're trying to make, relevant to Shield Master, but you have some faulty assumptions.
First, you are right that "hit with a melee weapon" and "make one melee weapon attack" are distinct things. Hitting with a melee weapon is a post-action conditional event. Making a melee weapon attack is an actual action.
Further, "with a melee weapon", "melee attack", and "melee weapon attack" are NOT the same thing. When an ability/action description states that the conditional is "with a melee weapon", it means that the ability/action just needs to involve using the melee weapon (USED as a melee weapon) for the ability/action to function. It is written this way because there are more ways to score a critical hit "with a melee weapon" than just the Attack, and it is clear that all of different ways you can crit with a melee weapon fulfill the conditional statement of GWM.Green-Flame Blade is an example of this.
"Melee attack" does not necessarily require a weapon be used; unarmed strikes, natural weapons, and attacks with melee weapons are all melee attacks.
"Melee weapon attack" does require a weapon be used, that the weapon actually be a melee weapon, and is no more than a baseline standard attack.
Lots of faulty assumptions... Improvised Weapons
I would not count a pommel strike as an improvised melee weapon. That's not an uncommon way to use the weapon, and frankly a ridiculous analogy because any "pommel strike" is just a flavorful description of a normal attack with the melee weapon. Mechanically, there are no instances where you would be able to "pommel strike", yet not be able to make a normal attack with the weapon. You can trigger GWM from this.
With Green-Flame Blade, you are taking the [Tooltip Not Found] (no such thing as "melee weapon spell attack"), that involves making a "melee attack" "with a weapon". If the melee attack results in a critical hit (or the creature drops to 0 HP), and the weapon used is an actual melee weapon used as a melee weapon, you can absolutely trigger GWM off it.
Throwing the weapon is completely different. You are improvising if the weapon does not have the thrown property. Weapons with the thrown property are ranged weapons. You are not making a melee attack, nor are you using the melee weapon as a melee weapon. You are making a ranged attack with an object as an improvised ranged weapon. You cannot trigger GWM from this.
Back to being on-topic, nobody is saying you can't split the Attack if you have Extra Attack. You can. What is pertinent to Shield Master is whether or not you have to actually "make an attack" before you can attempt to shove as a bonus action. TL;DR yeah, you do.
I do think Crawford basically got caught with his pants down twice over this, and the "If X then Y" schema stands.
I don't like it any more than anyone else, but that's how it is. Homebrew to hearts' content.
You don't know what fear is until you've witnessed a drunk bird divebombing you while carrying a screaming Kobold throwing fire anywhere and everywhere.
After getting caught up on the thread, I'd like to wade in here; not trying to say anyone else is wrong, but rather how I take the RAW and make a ruling.
The rules of the Attack action include the following: “With this action, you make one melee or ranged attack.”
I'd say that the Making an attack section is irrelevant because it applies only to an attack that occurs, whether it be ranged, melee, or spell, not to the Attack action itself which can have any number of attacks or other activities granted by features.
Also, I feel that the RAW does not inarguably define that the one attack granted by the default Attack action must be ordered first amongst all the other possible feature-granted attacks or activities that may fall within the same Attack action.
The SA ruling also does not address whether the bonus action may be taken during the Attack action, only that “you do get to decide when to take the bonus action after you’ve taken the Attack action”. Again, not that the Attack action must be completed to any extent, only that it be “taken”.
The movement-between-attacks rules also indicate that the attacks may be staggered at the player's discretion amongst movement, further supporting the interpretation that the attack is not unalterably immediate once the Attack action has been declared.
Why is this relevant? Because the feat does not specify when the bonus action must be taken, only that the clause “If you take the Attack action on your turn” must be satisfied. From that moment, it seems to be fair game to execute at the players’ pleasure.
Thus, here is how I rule it for now:
If you have irreversibly declared the Attack action, then you may at any point thereafter, including immediately, activate the feature of Shield Master and shove using the bonus action.
Afterward, the Attack action continues to resolve with however many attacks remain.
Always happy to hear critiques and counterpoints! It's all up for interpretation and the conversation is good stuff!
I think we are on the same page on most of this actually.
My comment about the melee weapon thing was in regard to this tweet by JC. It implies that the only thing that makes a melee weapon a melee weapon is whether it is on the table. Which is patently abusrd (thus my absurd thrown sword thing)
https://twitter.com/jeremyecrawford/status/1038805695875801093
As for the bonus action between acitons thing, I believe you and I both agree that IF you say it requires an attack action, and you have mulitiple attacks, you can use the shove after one attack. This is not supported by comments online though.
https://twitter.com/jeremyecrawford/status/994993596989300736?lang=en
https://twitter.com/JeremyECrawford/status/995024061267767298
If you just read the first tweet, then you can read is simply as 'you need to make an attack to do the shove, so you can do a bonus action after 1 attack'. Thats how i read it, and I think thats how you read it. But the second tweet comment say that you have to COMPLETE the attack action before you trigger the X then Y for shield master. It all goes to the wording 'take the attack action'. As someone earlier in the thread pointed out, that is super ambiguous because what does it mean to take an attack action. Do I have to say I am attacking, make an attack, complete all attacks? Any of those could be valid with the wording. according to folks online (yes I know, dont trust the internet) that was determined to be after completing the action. The RAw/RAI on this is just ugly, even after it was 'clearedd up'. Twice.
And that sort of makes my 1st point. It would NOT be hard for them to reword this in the eratta to make it crystal clear. The fact that they are refusing to do so combined with the reversing of the ruling makes me think they are choosing not to for a reason. And the only reason I can come up with (I admit I might be missing something) is that they would end up nerfing the feat explicitly, which would cause a much larger backlash (vs the twitter storm and us rules nerd talking about it on forums which is what happens now.
Roger that. 👍
Wow, you're right about that second tweet. Unfortunately, that makes everything even worse than it already is... and actually makes the entirety of Making an Attack the central premise now.
The Attack action is not resolved until "making an attack" has been resolved.
"Making an attack" is not resolved until the attack has been rolled.
Ergo, the Attack action is not resolved until the attack has been rolled.
Bonus actions cannot interrupt multiple attacks during the Attack action if the trigger to do so is the Attack action itself.
Ergo, the Attack action is not resolved until ALL attacks has been rolled (or by forfeiting any remaining attacks, I would suppose).
Ergo, the bonus action Shove from Shield Master cannot be used until the ENTIRE Attack action has been resolved.
In the GWM analogy, you would be able to interrupt your Attack action for the bonus action because the trigger is based on the individual attack (via critical or KO), not the Attack action itself, but Shield Master is SoL.
Maybe we should start a running tally of how many times Crawford gets caught with his pants down when it comes to this feat...
You don't know what fear is until you've witnessed a drunk bird divebombing you while carrying a screaming Kobold throwing fire anywhere and everywhere.
Feels like I'm missing something: where is the rule which prohibits using a bonus action during the sequence of attacks from the Attack action?
It's that second tweet you linked: https://twitter.com/JeremyECrawford/status/995024061267767298
He is saying that there is no general rule allowing bonus actions to interrupt the Attack, and that prohibition is the default state, unless a specific feature allows sequence breaking.
You don't know what fear is until you've witnessed a drunk bird divebombing you while carrying a screaming Kobold throwing fire anywhere and everywhere.
Ahh, I get you! I do think it's important to note that Crawford's tweets are not rulings. Both he and WotC have clearly stated that to be the case. Only published SAs carry the weight of actual rules. Crawford's posts have been relegated to "advice" and "preview of rulings that will appear".
With that in mind, the official SA ruling states the following: “you do get to decide when to take the bonus action after you’ve taken the Attack action”. To me that means everything hangs on whether "taken" is interpreted to mean completed or something else, for example as I rule at my table: irreversibly declared.
Really hope they publish an errata of some sort to this effect! (Or have they that I've overlooked?)
That's silly. He can't nerf the game through tweets; if the way Shield Master worked was considered a problem it'd be changed in errata. And if you read enough of his tweets or watch a couple of Dragon+/Dragon Talk episodes with him it's pretty obvious that he's not very concerned with what min/maxers do as long as most players find the game enjoyable.
Worth pointing out that he was talking specifically about bonus actions that have triggers.
Up until a few months ago they were. He only changed his mind on that in the very last update to Sage Advice Compendium, because he didn't want people to have to dig through tweets for official answers. If you look at older version of the Sage Advice PDF you'll find this:
I know this is a nitpick but since this is a thread about dissecting the rules, I have to point out that Sage Advice Compendium has never carried the weight of actual rules. The only official rules for D&D are the 3 core books. Sage Advice is basically an FAQ.
The Forum Infestation (TM)
Yes, and Shield Master has been clarified as to how if <-> then situations work in the context of the game rules, and anything that has an 'if' limiter must be completed before the 'then' can be performed. Because of that, bonus actions that require the completion of some 'if' have to be completed before being granted the bonus action, hence why you are not able to use it in between an attack; it isn't that there is a specific rule on bonus action timings, rather than the fact that most bonus actions have an 'if' limiter that automatically controls when they can be used.
Because of this, we can fully understand the design intent of the feat; It follows the theme of slowly pushing back your foe in a narrow hall, slowly gaining ground by pushing the assailants back with your shield over time.
It isn't and wasn't supposed to be a way for min-maxers to give themselves advantage on all their attacks, but rather supposed to be a defensive feat in all aspects.
The above is the up to date SA clarification: Intending to make the attack is not enough, you must fully take the attack action per the area I highlighted...such a clause would be unneeded if taking the attack action didn't require actually attacking.
I can definitely appreciate your interpretation of it, AtlaStar! Were I playing at your table, I'd not argue.
That said, as a DM, however, I interpret differently whether "taking an action" means specifically that said action must be completed beginning to end to satisfy the condition of Shield Master. I do not see anywhere in the published rules or in SA where it does so inarguably, so I won't decree to my players that they can't do it.
I hardly see it as a problem for min-maxers, because they will always find ways to eek out numbers. Rather I see the opposite as an arbitrary and prohibitive interpretation of what would otherwise be a sensible ability. 20th level fighter: "I can't bash with my shield first because why?!" Me: "Oh, you know: you have to swing your sword 4 times first."
And whether it's a defensively themed feat is debatable as well--though major kudos for raising that discussion point! I've seen it brilliantly used offensively to create attack angles for allies as well as to clear a path to attack a big bad. Being able to double shove is also astoundingly powerful, but no one ever tries that!
If the feat said “taken” or “have taken” this wouldn’t be an issue. It says “take”, not “taken”.
If it is supposed to be an “If-then” statement, why leave out “then”.
I accept Crawford’s advice. If I want to change when the bonus action can be made, I accept that it would be a house rule.
I will not accept that take and taken are synonymous or that an invisible “then” exists in the description of the feat.
If that is a problem, they can always use their first of many attacks to shove. Halving movement or causing forced movement is super strong, and knocking prone while getting all fighter attacks breaks the action economy...it is ironic though, because a lot of people find 5e to trivialize death and be easy, but half the reason it is too easy is that DM's allow things at the table that aren't intended.
"if X, do Y" is an if <-> then logic statement. You have to take your action, not declare that you are taking it. Your argument is basically that having intent to do something is the same as doing it, which is not the case...if you take the attack action you make attacks and until you have made those attacks you haven't taken anything, just declared your intent to take that action which is not the same. It'd be like me saying "I am going to rob a bank" and getting arrested for robbing a bank...stating intent isn't the same as doing.
As to the specifics of why the "then" was left out, "if you take" is a future action...simple causality makes the implication clear that when the if is satisfied, the other conditional attached to the if is the "then." But yes, this is an issue with having the rules written in a natural language, but very few people liked 4th edition which was more rules and mechanics focused by removing the natural language aspects.
I'm now convinced that you don't understand the difference between present and past tense.
You don't need to have "taken" an attack action... you need to "take" one.
Words are hard, I get it... but please, buddy, just for a minute stop and really wrap your head around the difference between a present tense conditional trigger and a past tense conditional trigger. Shield Master is written with a present tense trigger. Ergo, the attack action DOES NOT NEED TO BE RESOLVED. You must be taking it. Take. Present.
According to the rules you linked, you must simply start the action by 1. Choose a target. Pick a target within your attack's range: a creature, an object, or a location.
So, you Choose yourself a target. At this point you have initiated the attack action. Then you immediately use a bonus action to Shove, resolve shove, and then resolve the original attack action. That's as RAW as the written rules get.
I'm probably laughing.
If state = [take attack action], then enable [bonus action shove] until end of turn.
If [take attack action], then sequence
If you were to program the equivalent of the RAW about the relevant rules... you'd be able to take the shove bonus action at any of those sequence bullets. It'd be enabled from the moment the attack action sequence initiated because the conditional trigger is the present tense [take attack action] point and not the past tense [taken attack action] point.
I'm probably laughing.
That is incorrect, because those 3 parts would be subroutines if anything. Those 3 parts make up an attack, as such all subroutines would have to return as the attack function was called.
So if programming it per RAW, you'd declare you were taking the attack action which would call the attack function. It would then call the target selection function, you'd select targets and it'd return with the object you selected which would be used to determine modifiers based on the target. After all that, it'd resolve. If extra attack is available, it'd give you the option to make additional attacks repeating the previous steps. Then and only then, does the attack function return.
You haven't taken an action until you start the process of completing the action. This is 100% how it is written, and I don't understand why people try to argue it says otherwise especially given that the Sage Advice makes it very clear that you don't get the bonus action until you actually make the attack.
You don't mix tenses...if you said if you have taken X on your turn, it then puts the entire context of the sentence into the past tense, including the part discussing your turn...being granted a bonus action once your turn has ended is pretty useless, hence to have a grammatically correct sentence you need to use present or future tense to indicate that once X has occurred, do Y.
To take the attack action means going through every step, otherwise you haven't taken that action, you've only shown the intent to take that action...once again, if I said "I am going to rob a bank," which is a statement of intent, it doesn't mean I deserve to get arrested for robbing a bank as the action hasn't actually occurred.
And finally, under the rules for making an attack.
If choosing a target was enough to be considered an attack, why include the fact that you need to be making an attack roll to be making an attack? So if you want to argue that you can use the bonus action in between attacks, fine go ahead...but it is pretty cut and dry that unless you are rolling to attack, you aren't making an attack.
EDIT: And since you wanted grammar pedantry, making is the present participle which can only be referring to something that is currently happening or happened in the past continuous tenses...and since you can not continue 'making' an attack roll as it is an instantaneous event where you roll and it completes, until you perform the act of rolling for an attack, you aren't attacking.
AtlaStar,
I agree with your conclusion but changing words and adding words in order to support your argument is not persuasive.
The text of Shield Master could be interpreted in different ways. Crawford has written in Sage Advice on how to interpret it in the context of the entire set of rules. If a DM doesn’t agree with Crawford, the DM can make a house rule for their own game.
That is really all that needs to be said.