The fact that you're seeing a bunch of stuff saying "attacks" and claiming it doesn't sound limited to attack rolls when attack rolls are literally what decide whether something is an attack or not is troubling for rules debates.
The fact that you're seeing a bunch of stuff saying "attacks" and claiming it doesn't sound limited to attack rolls when attack rolls are literally what decide whether something is an attack or not is troubling for rules debates.
"you choose the targets and resolve the attacks consecutively, not all at once"
Choosing targets is not unique to spells with attacks. It's a separate clause - you choose the targets consecutively, and you resolve the attacks consecutively. Even without an attack, you'd still choose targets consecutively (there would just be no attacks to resolve). I am tentatively willing to grant Hold Person has no attack. (Although see parenthetical below)
And it sure would be weird if spells with attacks choose targets differently from other spells, when the rules never say they do. The actual rules on targeting, sparse as they are, treat all spells the same. If Sage Advice says spells with attack rolls work this way, all spells need to work that way, because the rules nowhere distinguish the two on targeting.
(That said, it's also pretty obvious Crawford was being imprecise and meant to include effects that weren't attack rolls, otherwise his note about things which explicitly say they're simultaneous would be pointless, because it only covers magic missile, which does not make attack rolls. Nor is this an unusual imprecision. Colloquially Hold Person is an attack, even if it's not an Attack Roll. Further, 3.5e considered all actions which negatively impacted the enemy to be attacks, formalizing the colloquial understanding. 3.5: "All offensive combat actions, even those that don’t damage opponents are considered attacks. Attempts to turn or rebuke undead count as attacks. All spells that opponents resist with saving throws, that deal damage, or that otherwise harm or hamper subjects are attacks." That Crawford would be imprecise in exactly this way is not surprising. Indeed, 5e seems to make a distinction between 'attack' and 'attack roll', otherwise there'd be no need for the hiding rules to specifically call out "you make an attack roll" - they could just say 'attack' instead. That seems to imply that not all attacks involve attack rolls - ie, the colloquial understanding appears at least sort of valid under 5e, and aligns with the 3.5 definition.
5e's supposedly "simple" rule, " if you’re making an attack roll, you’re making an attack.", isn't so simple. It tells you that, yes, all things with attack rolls are attacks. And it tells you that all things that aren't attacks don't make attack rolls. But you cannot conclude from that that if you don't make an attack roll, it's not an attack. A => B. -B => -A. But -A =/=> -B. It would be a fallacy to claim -A => -B.)
"Even though the duration of each of these spells is instantaneous, you choose the targets and resolve the attacks consecutively, not all at once."
That doesn't sound limited to attack rolls.
I agree. It's not clear, however, if that statement is expanding the context of the original question or still within the context of the original question. However, Saving Throws have an additional rule: Damage against Multiple Targets
Damage dealt via saving throws uses these rules.
Damage against Multiple Targets
When you create a damaging effect that forces two or more targets to make saving throws against it at the same time, roll the damage once for all the targets. For example, when a wizard casts Fireball, the spell’s damage is rolled once for all creatures caught in the blast.
So the Sage Advice cannot apply to damage dealt via Saving Throws, at least in terms of rolling damage. Applying the damage/effects could still happen in sequence.
Declaring targets is a universal feature of spells, it shouldn't change based on whether the spell has an attack roll or not. They seem to be claiming this is how you normally resolve things, not a special feature of spells with attacks. (And even if there isn't an attack, this still tells you to choose targets consecutively, not all at once. In the case of upcast Hold Person, there are simply no attacks to resolve, but this would still say you choose targets consecutively).
And there may be a reason to resolve them consecutively. such as if a target had a reaction that triggered when an ally was affected by an enchantment effect (I don't know if such a reaction exists).
And Crawford's separate statement: "Multiple attacks on the same turn aren't simultaneous, unless a feature or spell says otherwise." Seems to suggest that he really does mean things beyond just 'attack rolls' when he says "attacks", because afaik, the only spell which says otherwise is magic missile, which doesn't make attack rolls. (Crawford was almost certainly the author of the quoted Sage Advice response).
It's my understanding that the formally codified SAC is a team effort and whether Crawford's unofficial statements aligned with them is irrelevant. His separate statement might suggest RAI, but is not RAW or otherwise official. Technically, Magic Missile could still fall under the previous Sage Advice entry requiring them them to be resolved in sequence.
The fact that you're seeing a bunch of stuff saying "attacks" and claiming it doesn't sound limited to attack rolls when attack rolls are literally what decide whether something is an attack or not is troubling for rules debates.
"you choose the targets and resolve the attacks consecutively, not all at once"
Choosing targets is not unique to spells with attacks. It's a separate clause - you choose the targets consecutively, and you resolve the attacks consecutively. Even without an attack, you'd still choose targets consecutively (there would just be no attacks to resolve). I am tentatively willing to grant Hold Person has no attack. (Although see parenthetical below)
Athanar90 has convinced me. 5e has no formal context of an "attack" without the Attack action or Attack Roll; I just dug through several sections trying to find one as a counterpoint to Athanar90. As such, I must conclude that the Sage Advice is in the context of spells that require attack rolls and does not affect spells with saving throws or that have neither saving throws or attack rolls. A Fireball may be an offensive action but is not an "attack" in this context.
The fact that you're seeing a bunch of stuff saying "attacks" and claiming it doesn't sound limited to attack rolls when attack rolls are literally what decide whether something is an attack or not is troubling for rules debates.
"you choose the targets and resolve the attacks consecutively, not all at once"
Choosing targets is not unique to spells with attacks. It's a separate clause - you choose the targets consecutively, and you resolve the attacks consecutively. Even without an attack, you'd still choose targets consecutively (there would just be no attacks to resolve). I am tentatively willing to grant Hold Person has no attack. (Although see parenthetical below)
Athanar90 has convinced me. 5e has no formal context of an "attack" without the Attack action or [rules]Attack Roll[/spells]; I just dug through several sections trying to find one as a counterpoint to Athanar90. As such, I must conclude that the Sage Advice is in the context of spells that require attack rolls and does not affect spells with saving throws or that have neither saving throws or attack rolls. A Fireball may be an offensive action but is not an "attack" in this context.
Athanar90 has not convinced me, because the rules never say only things with attack rolls are attacks. It only says if it has an attack roll, it's an attack. (If it does not have an attack roll, it could still be an attack. A => B =/=> -A => -B see my last post on the logic of their sentence construction. If they truly meant for the logic to be bidirectional, they should have used "if and only if", not simply "if".
Regardless, I'm also completely unconvinced that targeting should work differently for spells with attacks, and spells without. If spells with attacks target one way, spells without attacks must target the same way.
"Even though the duration of each of these spells is instantaneous, you choose the targets and resolve the attacks consecutively, not all at once."
That doesn't sound limited to attack rolls.
I agree. It's not clear, however, if that statement is expanding the context of the original question or still within the context of the original question. However, Saving Throws have an additional rule: Damage against Multiple Targets
Damage dealt via saving throws uses these rules.
Damage against Multiple Targets
When you create a damaging effect that forces two or more targets to make saving throws against it at the same time, roll the damage once for all the targets. For example, when a wizard casts Fireball, the spell’s damage is rolled once for all creatures caught in the blast.
That only covers rolling damage. It does not cover if there's an order to the affected creatures making saving throws or suffering damage.
One situation this could matter is if two creatures were caught in it, and one of them had Warding Bond cast on the other. If the one who cast Warding Bond drops to 0hp due to the Fireball before the other makes his save and/or takes damage, the Warding Bond ends before the other takes damage. If it's simultaneous (or the recipient saves/takes damage first), the Warding Bond is still in effect for the recipient.
Regardless, we can differentiate the Fireball situation insofar as the target is the origin point, so the victims are not 'targets' (at least not in 2024). This could be simultaneous or sequential independently of the targets question. (I believe, however, the Sage Advice answer is from the 2014 era, and the victims of the Fireball were targets then).
Declaring targets is a universal feature of spells, it shouldn't change based on whether the spell has an attack roll or not. They seem to be claiming this is how you normally resolve things, not a special feature of spells with attacks. (And even if there isn't an attack, this still tells you to choose targets consecutively, not all at once. In the case of upcast Hold Person, there are simply no attacks to resolve, but this would still say you choose targets consecutively).
And there may be a reason to resolve them consecutively. such as if a target had a reaction that triggered when an ally was affected by an enchantment effect (I don't know if such a reaction exists).
Wasn't there a subclass ability that triggered off an enemy passing a charm effect saving throw? (I seem to recall something about using Friends against someone you were in combat with to have them automatically pass and trigger the ability. I can't remember what it was an ability for...)
Upcast hold person and the like: In a situation where you're unaware of the saves of enemies, and there's multiple individuals of multiple different enemy types, you might choose a different target based on whether the first save passed or failed (basically testing the strength of their saves - especially if the DM rolls his dice in the open).
Sure would be nice if the general rules for casting spells said something about this.
Have you met our lord and savior, 3.5e? ;)
Still my preference (or 1st or 2nd AD+D). But you play what people play.
No one in rules discussions in 3.5 took Sage Advice seriously. They only ever got referenced to make fun of them or to demonstrate why they were wrong.
I think a couple of important clarifying points should be made.
-Smite is correct about the Damage Against Multiple Targets rule, it's a more specific rule that beats the general Simultaneous Effects rule.
-The Simultaneous Effects rule likely applies to spells like Hold Person as well, though how relevant that is would vary. In most cases, the result is identical.
-Magic Missile functions like a really weird AoE spell in 2014, not necessarily so in 2024 where the damage is rolled per dart. I'd have to look at the wording but I'm on my phone and feeling lazy about it, so I'll leave it at "maybe, would have to look deeper".
-3.5e considering all negative effects toward enemies to be attacks has zero bearing on 5e's rules. There's a reason the Sanctuary spell makes rulings about areas of effect and damaging spells in addition to attacks instead of lumping it all together.
-Okay, decided to look at Magic Missile after all. 2024 creates a rules exception by explicitly stating in the spell's text that the darts all strike simultaneously.
-Magic Missile functions like a really weird AoE spell in 2014, not necessarily so in 2024 where the damage is rolled per dart. I'd have to look at the wording but I'm on my phone and feeling lazy about it, so I'll leave it at "maybe, would have to look deeper".
-3.5e considering all negative effects toward enemies to be attacks has zero bearing on 5e's rules. There's a reason the Sanctuary spell makes rulings about areas of effect and damaging spells in addition to attacks instead of lumping it all together.
-Okay, decided to look at Magic Missile after all. 2024 creates a rules exception by explicitly stating in the spell's text that the darts all strike simultaneously.
The fact that you're seeing a bunch of stuff saying "attacks" and claiming it doesn't sound limited to attack rolls when attack rolls are literally what decide whether something is an attack or not is troubling for rules debates.
"you choose the targets and resolve the attacks consecutively, not all at once"
Choosing targets is not unique to spells with attacks. It's a separate clause - you choose the targets consecutively, and you resolve the attacks consecutively. Even without an attack, you'd still choose targets consecutively (there would just be no attacks to resolve). I am tentatively willing to grant Hold Person has no attack. (Although see parenthetical below)
Athanar90 has convinced me. 5e has no formal context of an "attack" without the Attack action or [rules]Attack Roll[/spells]; I just dug through several sections trying to find one as a counterpoint to Athanar90. As such, I must conclude that the Sage Advice is in the context of spells that require attack rolls and does not affect spells with saving throws or that have neither saving throws or attack rolls. A Fireball may be an offensive action but is not an "attack" in this context.
Athanar90 has not convinced me, because the rules never say only things with attack rolls are attacks. It only says if it has an attack roll, it's an attack. (If it does not have an attack roll, it could still be an attack. A => B =/=> -A => -B see my last post on the logic of their sentence construction. If they truly meant for the logic to be bidirectional, they should have used "if and only if", not simply "if".
Regardless, I'm also completely unconvinced that targeting should work differently for spells with attacks, and spells without. If spells with attacks target one way, spells without attacks must target the same way.
Please quote the 5e definition of an "attack". Could be an attack is not a rule and cannot be the basis of a RAW discussion. What is defined in an official source?
When you create a damaging effect that forces two or more targets to make saving throws against it at the same time, roll the damage once for all the targets. For example, when a wizard casts Fireball, the spell’s damage is rolled once for all creatures caught in the blast.
That only covers rolling damage. It does not cover if there's an order to the affected creatures making saving throws or suffering damage.
Yes, I said as much. "So the Sage Advice cannot apply to damage dealt via Saving Throws, at least in terms of rolling damage. Applying the damage/effects could still happen in sequence."
Wasn't there a subclass ability that triggered off an enemy passing a charm effect saving throw? (I seem to recall something about using Friends against someone you were in combat with to have them automatically pass and trigger the ability. I can't remember what it was an ability for...)
Upcast hold person and the like: In a situation where you're unaware of the saves of enemies, and there's multiple individuals of multiple different enemy types, you might choose a different target based on whether the first save passed or failed (basically testing the strength of their saves - especially if the DM rolls his dice in the open).
Countercharm, Level 7 Bard ability triggers off an ally failing a save against charmed or frightened conditions. So, if you are upcasting Charm Person, you may want to target the Bard first. Silvery Barbs could trigger on a successful save. I would think that while there may be rare benefits to resolving the effects in a certain order, the scenarios where you would be concerned with targeting a creature based on another's success or failure would be extremely niche and probably reserved for oddball third party content. The only scenario I can anticipate would be if you could retarget a creature who succeeded in order to force a failure and I can't see any official content allowing for that.*Hopes that there is no official content allowing for that.*
Sure would be nice if the general rules for casting spells said something about this.
Have you met our lord and savior, 3.5e? ;)
Still my preference (or 1st or 2nd AD+D). But you play what people play.
No one in rules discussions in 3.5 took Sage Advice seriously. They only ever got referenced to make fun of them or to demonstrate why they were wrong.
Same except for anything with THAC0. :D I miss the way you could layer multiclassing and templates to get custom concepts beyond a basic 1-20 class or even a 1-10 plus a 1-10 prestige class. I also wished that I got a chance to play a d20 modern game. But I also just miss the days of the rules (and errata) being enough for discussions because they were beautifully precise if sometimes quirky. I wish we had that for 5e but I don't think it would be as accessible if we did.
An exemple of something that's an attack despite no attack roll or Attack action is Monk's Martial Art's Unarmed Strike as a Bonus Action to grapple or shove.
An exemple of something that's an attack despite no attack roll or Attack action is Monk's Martial Art's Unarmed Strike as a Bonus Action to grapple or shove.
That is a special case because the Unarmed Strike definition defines it as a melee attack, which may or may not be a product of the 5.14 Unarmed Strike being an attack that deals damage as the only option. Whether that's the case or not, it does not establish a global definition of an "attack" outside of an attack roll or the attack action.
Interestingly, the grapple and shove options will not end Sanctuary. Arguably, shoving someone off a cliff or dragging them over spikes may, even if the grappler/shover is not dealing damage themselves, per se.
The fact that you're seeing a bunch of stuff saying "attacks" and claiming it doesn't sound limited to attack rolls when attack rolls are literally what decide whether something is an attack or not is troubling for rules debates.
"you choose the targets and resolve the attacks consecutively, not all at once"
Choosing targets is not unique to spells with attacks. It's a separate clause - you choose the targets consecutively, and you resolve the attacks consecutively. Even without an attack, you'd still choose targets consecutively (there would just be no attacks to resolve). I am tentatively willing to grant Hold Person has no attack. (Although see parenthetical below)
And it sure would be weird if spells with attacks choose targets differently from other spells, when the rules never say they do. The actual rules on targeting, sparse as they are, treat all spells the same. If Sage Advice says spells with attack rolls work this way, all spells need to work that way, because the rules nowhere distinguish the two on targeting.
(That said, it's also pretty obvious Crawford was being imprecise and meant to include effects that weren't attack rolls, otherwise his note about things which explicitly say they're simultaneous would be pointless, because it only covers magic missile, which does not make attack rolls. Nor is this an unusual imprecision. Colloquially Hold Person is an attack, even if it's not an Attack Roll. Further, 3.5e considered all actions which negatively impacted the enemy to be attacks, formalizing the colloquial understanding. 3.5: "All offensive combat actions, even those that don’t damage opponents are considered attacks. Attempts to turn or rebuke undead count as attacks. All spells that opponents resist with saving throws, that deal damage, or that otherwise harm or hamper subjects are attacks." That Crawford would be imprecise in exactly this way is not surprising. Indeed, 5e seems to make a distinction between 'attack' and 'attack roll', otherwise there'd be no need for the hiding rules to specifically call out "you make an attack roll" - they could just say 'attack' instead. That seems to imply that not all attacks involve attack rolls - ie, the colloquial understanding appears at least sort of valid under 5e, and aligns with the 3.5 definition.
5e's supposedly "simple" rule, " if you’re making an attack roll, you’re making an attack.", isn't so simple. It tells you that, yes, all things with attack rolls are attacks. And it tells you that all things that aren't attacks don't make attack rolls. But you cannot conclude from that that if you don't make an attack roll, it's not an attack. A => B. -B => -A. But -A =/=> -B. It would be a fallacy to claim -A => -B.)
I agree. It's not clear, however, if that statement is expanding the context of the original question or still within the context of the original question. However, Saving Throws have an additional rule: Damage against Multiple Targets
So the Sage Advice cannot apply to damage dealt via Saving Throws, at least in terms of rolling damage. Applying the damage/effects could still happen in sequence.
And there may be a reason to resolve them consecutively. such as if a target had a reaction that triggered when an ally was affected by an enchantment effect (I don't know if such a reaction exists).
It's my understanding that the formally codified SAC is a team effort and whether Crawford's unofficial statements aligned with them is irrelevant. His separate statement might suggest RAI, but is not RAW or otherwise official. Technically, Magic Missile could still fall under the previous Sage Advice entry requiring them them to be resolved in sequence.
Have you met our lord and savior, 3.5e? ;)
How to add Tooltips.
My houserulings.
Athanar90 has convinced me. 5e has no formal context of an "attack" without the Attack action or Attack Roll; I just dug through several sections trying to find one as a counterpoint to Athanar90. As such, I must conclude that the Sage Advice is in the context of spells that require attack rolls and does not affect spells with saving throws or that have neither saving throws or attack rolls. A Fireball may be an offensive action but is not an "attack" in this context.
How to add Tooltips.
My houserulings.
Athanar90 has not convinced me, because the rules never say only things with attack rolls are attacks. It only says if it has an attack roll, it's an attack. (If it does not have an attack roll, it could still be an attack. A => B =/=> -A => -B see my last post on the logic of their sentence construction. If they truly meant for the logic to be bidirectional, they should have used "if and only if", not simply "if".
Regardless, I'm also completely unconvinced that targeting should work differently for spells with attacks, and spells without. If spells with attacks target one way, spells without attacks must target the same way.
That only covers rolling damage. It does not cover if there's an order to the affected creatures making saving throws or suffering damage.
One situation this could matter is if two creatures were caught in it, and one of them had Warding Bond cast on the other. If the one who cast Warding Bond drops to 0hp due to the Fireball before the other makes his save and/or takes damage, the Warding Bond ends before the other takes damage. If it's simultaneous (or the recipient saves/takes damage first), the Warding Bond is still in effect for the recipient.
Regardless, we can differentiate the Fireball situation insofar as the target is the origin point, so the victims are not 'targets' (at least not in 2024). This could be simultaneous or sequential independently of the targets question. (I believe, however, the Sage Advice answer is from the 2014 era, and the victims of the Fireball were targets then).
Wasn't there a subclass ability that triggered off an enemy passing a charm effect saving throw? (I seem to recall something about using Friends against someone you were in combat with to have them automatically pass and trigger the ability. I can't remember what it was an ability for...)
Upcast hold person and the like: In a situation where you're unaware of the saves of enemies, and there's multiple individuals of multiple different enemy types, you might choose a different target based on whether the first save passed or failed (basically testing the strength of their saves - especially if the DM rolls his dice in the open).
Still my preference (or 1st or 2nd AD+D). But you play what people play.
No one in rules discussions in 3.5 took Sage Advice seriously. They only ever got referenced to make fun of them or to demonstrate why they were wrong.
I think a couple of important clarifying points should be made.
-Smite is correct about the Damage Against Multiple Targets rule, it's a more specific rule that beats the general Simultaneous Effects rule.
-The Simultaneous Effects rule likely applies to spells like Hold Person as well, though how relevant that is would vary. In most cases, the result is identical.
-Magic Missile functions like a really weird AoE spell in 2014, not necessarily so in 2024 where the damage is rolled per dart. I'd have to look at the wording but I'm on my phone and feeling lazy about it, so I'll leave it at "maybe, would have to look deeper".
-3.5e considering all negative effects toward enemies to be attacks has zero bearing on 5e's rules. There's a reason the Sanctuary spell makes rulings about areas of effect and damaging spells in addition to attacks instead of lumping it all together.
-Okay, decided to look at Magic Missile after all. 2024 creates a rules exception by explicitly stating in the spell's text that the darts all strike simultaneously.
Excellent call out regarding Sanctuary.
I only mentioned 3.5 jokingly as the system, while it had its own flaws, was more precise in the language.
Regarding Magic Missile, do you think there is a difference between the darts striking simultaneously and those strikes being resolved simultaneously?
Please quote the 5e definition of an "attack". Could be an attack is not a rule and cannot be the basis of a RAW discussion. What is defined in an official source?
Yes, I said as much. "So the Sage Advice cannot apply to damage dealt via Saving Throws, at least in terms of rolling damage. Applying the damage/effects could still happen in sequence."
Countercharm, Level 7 Bard ability triggers off an ally failing a save against charmed or frightened conditions. So, if you are upcasting Charm Person, you may want to target the Bard first. Silvery Barbs could trigger on a successful save. I would think that while there may be rare benefits to resolving the effects in a certain order, the scenarios where you would be concerned with targeting a creature based on another's success or failure would be extremely niche and probably reserved for oddball third party content. The only scenario I can anticipate would be if you could retarget a creature who succeeded in order to force a failure and I can't see any official content allowing for that.*Hopes that there is no official content allowing for that.*
Same except for anything with THAC0. :D I miss the way you could layer multiclassing and templates to get custom concepts beyond a basic 1-20 class or even a 1-10 plus a 1-10 prestige class. I also wished that I got a chance to play a d20 modern game. But I also just miss the days of the rules (and errata) being enough for discussions because they were beautifully precise if sometimes quirky. I wish we had that for 5e but I don't think it would be as accessible if we did.
How to add Tooltips.
My houserulings.
That is a special case because the Unarmed Strike definition defines it as a melee attack, which may or may not be a product of the 5.14 Unarmed Strike being an attack that deals damage as the only option. Whether that's the case or not, it does not establish a global definition of an "attack" outside of an attack roll or the attack action.
Interestingly, the grapple and shove options will not end Sanctuary. Arguably, shoving someone off a cliff or dragging them over spikes may, even if the grappler/shover is not dealing damage themselves, per se.
How to add Tooltips.
My houserulings.