D&D 5E How does Surprise work in 5e?

[FormerlyHemlock]

I take back what I said earlier about this being ambiguous and thus up for interpretation. The above stated interpretation of the surprise rules on p. 189 of the PH is actually based on a propositional fallacy, an error in logic that concerns a compound proposition. The proposition in question, if I may paraphrase, is: if (A) a character or monster doesn't notice a threat, then (B) it is surprised. It does not follow that if A is false then B is also false, i.e.: if (not A) a character or monster notices a threat, then (not B) it is not surprised. This is called denying the antecedent, and it happens when the consequent (B) in an indicative conditional, like the statement above, is claimed to be false because the antecedent (A) is false.

It's not a fallacy when applied to the reading of rulebooks. When a rulebook says "If A then B", it means "If and only if A, then B, unless another rule applies which also imposes B." Rules are not intended to be read the same way as logical propositions.

If the rulebook says, "If you are reduced to 0 hit points, you die unless you make your death saves," the clear implication is that "As long as you are above 0 hit points, you are not dying [unless another rule applies]."

 

[Dausuul]

I take back what I said earlier about this being ambiguous and thus up for interpretation. The above stated interpretation of the surprise rules on p. 189 of the PH is actually based on a propositional fallacy, an error in logic that concerns a compound proposition. The proposition in question, if I may paraphrase, is: if (A) a character or monster doesn't notice a threat, then (B) it is surprised. It does not follow that if A is false then B is also false, i.e.: if (not A) a character or monster notices a threat, then (not B) it is not surprised. This is called denying the antecedent, and it happens when the consequent (B) in an indicative conditional, like the statement above, is claimed to be false because the antecedent (A) is false.

This argument is itself a logical fallacy--to wit, shifting the burden of proof. The normal rule is that everyone can act normally each round; surprise is an exception to that rule which applies in particular circumstances. Absent such circumstances, there is no surprise. Hence, the burden of proof is on those arguing that the PCs should be surprised. If you can't come up with a solid argument for why surprise should apply in this case, then it shouldn't.

 

[Hriston]

I would interpret the statement to say a character is surprised if and only if it doesn't notice a threat. So either both have to be true or both false for the statement to be true.

I understand what you are trying to say here, but I don't think that changing "if" to "if and only if" changes the meaning in the way I think you are intending. It still would not preclude the possibility of a creature having noticed one threat, but not having noticed another threat, being surprised because that creature has, by definition, not noticed a threat.

For it to mean what you seem to be saying, there would need to be some additional language along the lines of, "a creature who doesn't notice a threat is surprised as long as it doesn't notice any other threat." If this meaning was intended I'm sure the designers could have written it in a way that made it clear. Instead, I have to assume they chose a less stringent standard for surprise so that it would apply in cases like the ones we are discussing.

 

[Hriston]

The interpretation/argument is based on the terms used: "a threat", not by claiming that (not A) = (not B).
Any character or monster that doesn't notice [one side] is surprised at the start of the encounter.
Any character or monster that doesn't notice [one individual] is surprised at the start of the encounter.

PHB mentions sides trying to be stealthy and gives the examples [A band of adventurers] and [A gelatinous cube] as one side of the battle gaining surprise over the other.

Now that I've taken time to read up in the book about the potential 3 sides in a combat (or even more, if things get real messy):
Given your example with the trolls above, since no 3-way example is mentioned in the book, I would have to interpret the trolls as being a third side in the conflict, thus surprising both goblins and PCs. (not A = not B obviously not being true)
If the goblins and trolls were allied instead, I find it hard to see that side of the combat trying to be sneaky when one of the parts is in plain sight.

My issue with this interpretation is that nowhere do the rules define a threat as being an entire side of a conflict. In fact, they don't define threat at all, which is why I posted a dictionary definition up thread. The DM should determine whether something actually poses a threat. Personally, I'd draw the line at something that can actually do some damage to its opponents, beginning with anything with a CR above 0. If it is impossible for the unnoticed creature to actually inflict harm because of magical protections or high AC, then it is not a threat, but this is something for the DM to decide.

Admittedly, the book talks about different sides with reference to surprise. It gives some examples where, "one side of the battle gains surprise over the other," but it doesn't say that surprise can only be gained by one side or the other, in fact, mechanically no one gains surprise at all. Rather, it is inflicted upon those individuals who do not notice a threat.

It also says, "If neither side tries to be stealthy they automatically notice each other." This is different from saying that an entire side must try to be stealthy for individuals on that side who are trying to be stealthy to not be noticed.

Now to return to my examples, I'm glad you see the logic of both the goblins and the adventurers potentially being surprised by the trolls in the first scenario, but given that the adventurers in the second scenario don't know that the goblins and the trolls are working together, how could that fact influence whether they are surprised?

 

[Hriston]

Because the goblins already see a threat. In psychological terms, they are already in Condition Red: threat identified, entering kill mode.

You can still be surprised when you're in kill mode, but it's not the mechanical D&D version of "surprise" wherein you get no reaction and can't take opportunity attacks. It's just regular old surprise where the tactical situation turns out to be different than you thought: you just used your action to doff your shield, sheathe your sword, and pull our your bow in order to shoot the goblins 100' away from you, but wait! now there's a troll attacking me and I really wish I had known this was about to happen before I doffed my shield.

As for why adventurers don't walk around in Condition Red all the time: I suppose you could, but it would drive you psychotic. The door to the room opens: roll initiative. (Everyone pulls out weapons.) Oh, just kidding it was the chambermaid. Roll initiative. (Everyone readies an action in case she attacks.) She drops the meal platter and bolts. Do you take your opportunity attack? Etc.

TLDR; you can be surprised by a third party in combat, just as you can be surprised by the enemy strategy, but "surprise" from the mechanical perspective only happens to someone who "doesn't notice a threat" and is in Condition White or Yellow. (See http://en.wikipedia.org/wiki/Jeff_Cooper for more on threat conditions.)

Funny, I can't find "regular old surprise" in the rules anywhere.

But seriously, I found your references to the Color Code interesting, but ultimately irrelevant. I'd never looked into this subject, but if I'm reading the article you linked correctly, the Color Code is not a model for different levels of threat or alertness, but is rather a tool for getting yourself into the correct mindset for the use of lethal force. There's no reason to think that goblins or medieval fantasy adventurers would have access to this sort of training. Maybe they do in your campaign. Whatever training they do have with regard to alertness, I would think, would be reflected in a proficiency with perception which is already accounted for in the surprise rules quite nicely.

If you did want to think about surprise in terms of the color code, there's always the possibility that it's modeling what the marine corp calls condition black, going from white or yellow directly to red, resulting in an inability to think clearly, which i guess is what you're talking about in your last paragraph. So it's interesting, but to me there's no guarantee that everyone involved in combat is going through a condition orange and setting mental triggers for kill mode. If that were the assumption of the game, I would think the rules would somehow reflect that.

 

[Hriston]

It's not a fallacy when applied to the reading of rulebooks. When a rulebook says "If A then B", it means "If and only if A, then B, unless another rule applies which also imposes B." Rules are not intended to be read the same way as logical propositions.

If the rulebook says, "If you are reduced to 0 hit points, you die unless you make your death saves," the clear implication is that "As long as you are above 0 hit points, you are not dying [unless another rule applies]."

I'm pretty sure logic applies to rulebooks.

 

[Hriston]

This argument is itself a logical fallacy--to wit, shifting the burden of proof. The normal rule is that everyone can act normally each round; surprise is an exception to that rule which applies in particular circumstances. Absent such circumstances, there is no surprise. Hence, the burden of proof is on those arguing that the PCs should be surprised. If you can't come up with a solid argument for why surprise should apply in this case, then it shouldn't.

Specific trumps general. The particular circumstance that surprise applies to is that a creature has not noticed a threat. Once that condition is met, the burden of proof would be on those who argue for limiting surprise to a more narrow set of circumstances, regardless of whether the creature has simultaneously both noticed a threat and not noticed another threat, which is entirely possible. My argument is that there is a threat that the characters/pirates/goblins/whatever haven't noticed, therefore they are surprised.

 

[FormerlyHemlock]

Funny, I can't find "regular old surprise" in the rules anywhere.

But seriously, I found your references to the Color Code interesting, but ultimately irrelevant. I'd never looked into this subject, but if I'm reading the article you linked correctly, the Color Code is not a model for different levels of threat or alertness, but is rather a tool for getting yourself into the correct mindset for the use of lethal force. There's no reason to think that goblins or medieval fantasy adventurers would have access to this sort of training. Maybe they do in your campaign. Whatever training they do have with regard to alertness, I would think, would be reflected in a proficiency with perception which is already accounted for in the surprise rules quite nicely.

If you did want to think about surprise in terms of the color code, there's always the possibility that it's modeling what the marine corp calls condition black, going from white or yellow directly to red, resulting in an inability to think clearly, which i guess is what you're talking about in your last paragraph. So it's interesting, but to me there's no guarantee that everyone involved in combat is going through a condition orange and setting mental triggers for kill mode. If that were the assumption of the game, I would think the rules would somehow reflect that.

Why would you expect "regular old surprise" to be in the rules? It's not a rule. When the human wielding a gigantic longsword does not charge into melee but instead drops the longsword and casts Meteor Swarm, you may be surprised by his tactic and wish you had not clustered up quite so nicely, but you don't need a rule to tell you you're surprised. Instead, it is intrinsic to the tactical situation. I already gave an example (shield doffing) of how the goblin/troll ambush can still benefit from this kind of surprise, and other possibilities exist (squishy wizards moved to the back, and then trolls attack from the back).

I think you may be misreading the article. Jeff Cooper didn't invent the state of mind he calls "Condition White", he just invented the term, and I linked to it because the terms are convenient in explaining surprise. When you're reading a book in a library, and suddenly werewolves charge in and attack you, only a very special kind of person (Alert) is able to instantly transition to deadly-threat mode. Everyone else is caught in Condition White: no threat expected, mechanically "surprised" and unable to take reactions. However, if you're already in Condition Red (regardless of whether or not you call it by that name) and ready to kill some goblins at the slightest provocation, having somebody else charge you from the side will not catch you psychologically unready, so you won't freeze up, and you still get to act on your turn as normal and you get reactions. It doesn't matter if that "somebody else" turns out to actually be a third party who will kill goblins as well as you. In game turns, you were aware of a threat, so you are not surprised.

 

[fjw70]

I understand what you are trying to say here, but I don't think that changing "if" to "if and only if" changes the meaning in the way I think you are intending. It still would not preclude the possibility of a creature having noticed one threat, but not having noticed another threat, being surprised because that creature has, by definition, not noticed a threat.

For it to mean what you seem to be saying, there would need to be some additional language along the lines of, "a creature who doesn't notice a threat is surprised as long as it doesn't notice any other threat." If this meaning was intended I'm sure the designers could have written it in a way that made it clear. Instead, I have to assume they chose a less stringent standard for surprise so that it would apply in cases like the ones we are discussing.

You are right that this is a separate issue. The "a threat" techically could mean a singe threat or any threat. From context I would say the latter is intended since the former leads to absurd results (e.g. A single hidden goblin freezing a party that sees a threat of a dozen ogres).

 

[Azrothan]

Now to return to my examples, I'm glad you see the logic of both the goblins and the adventurers potentially being surprised by the trolls in the first scenario, but given that the adventurers in the second scenario don't know that the goblins and the trolls are working together, how could that fact influence whether they are surprised?

Should I ever bring a third+ side into a fight in the first round of combat that also tried to be sneaky about it, I would explain to the players that they are surprised by the fact that both sides of the (almost) ongoing encounter are startled by this sudden appearance.
As long as I follow the rules as I have interpreted them consistently, I'm not worred about the "logic" past that, as fluff reasons can always be given for the odd exception. I'm the one putting the encounters together, so I know what kind of tactics would have to be used to try and surprise the players - so in your examples the goblins would also be hidden or in some kind of disguise in order to spring the trap together with the trolls, not just stand in the open. And the players know that if they want to surprise the enemy, they all need to try and be sneaky about it to make it happen.

I have watched all seasons of Rollplay and a lot of their related stuff, and I don't think I can recall one time when there were more than 2 sides present in an encounter, let alone in a sneaky manner that would have called for a "surprise round". But there's a lot of hours and I don't remember everything in detail, so I could be mistaken.