Suppose a cleric casts Zone of Truth on a captured enemy. Unbeknownst to them, the enemy is a rakshasa. Zone of Truth says (PHB, p. 289):

You know whether each creature succeeds or fails on its saving throw.

But (MM, p. 257):

The rakshasa is immune to spells of 6th level or lower unless it wishes to be affected.

Assuming the rakshasa doesn't wish to be affected by the Zone of Truth spell, does the caster know this?

More generally, does a caster know when targets are unaffected by a spell they have cast, either because they have made saving throws, or because they are outright immune to the spell or its effects (e.g., casting Ray of Sickness on a shadow demon, which is immune to poison damage and the poisoned condition)?


3 Answers 3


There's nothing RAW that indicates that players would know whether or not a creature is affected by most spells, though presumably in the instance of Ray of Sickness, players could infer whether the spell had taken effect based on a creature's behavior.

Does the hobgoblin captain slump for a moment, resting on her spear? Does the shadow demon react at all to the beam? Those are probably telling.

Zone of Truth appears to be a little different since the caster explicitly knows whether the target succeeded or failed the saving throw, which can only occur if the creature is affected by the spell in the first place; a creature which isn't affected by the spell never makes a saving throw. In this way, the caster would know if the hostage weren't affected, if only circumstantially; instead of having succeeded or failed to resist the spell, the Rakshasa would effectively be invisible to the spell, which would surely be suspicious.


On page 85-86 of Xanathar's Guide to Everything, Invalid spell targets, there's a ruling that would seem somewhat relevant to this issue:

A spell specifies what a caster can target with it: any type of creature, a creature of a certain type (humanoid or beast, for instance), an object, an area, the caster, or something else. But what happens if a spell targets something that isn't a valid target? For example, someone might cast charm person on a creature believed to be a humanoid, not knowing that the target is in fact a vampire. If this issue comes up, handle it using the following rule.

If you cast a spell on someone or something that can't be affected by the spell, nothing happens to that target, but if you used a spell slot to cast the spell, the slot is still expended. If the spell normally has no effect on a target that succeeds on a saving throw, the invalid target appears to have succeeded on its saving throw, even though it didn't attempt one (giving no hint that the creature is in fact an invalid target). Otherwise, you perceive that the spell did nothing to the target.

Emphasis mine.

The only caveat of the above rule is that it specifically refers to invalid targets, not immune ones. It even uses an example of trying to use a spell targeting humanoids on a non-humanoid in disguise, thus an invalid target by the spell's targeting rules and not due to the target's innate immunity. A rakshasa is a creature, and thus still a valid target for spells like Zone of Truth. As such, this might not be applicable strictly RAW.

As for RAI, I'd still say it sets the precedent that DM could always hide any sort of immunity they do not wish to explicitly reveal to the players behind the guise of a successful save.


In this case the character would not know by default.

All the character knows is that the spell casting succeeded or not. The rakshasa is immune but unless the character (not the player) knows this they would have no way of telling through the casting of the spell.

HOWEVER: characters may by use of passive or active perception check (or other appropriate skill) catch the drift that the rakshasa isn't exactly under the spell. Including the player who cast the spell May through this mechanic perceive that the spell didn't quite take disputed being successful.

To add: if the spell is cast from higher spell slot then the rakshasa isn't immune.

  • \$\begingroup\$ I wonder how your interpretation ties in to this: rpg.stackexchange.com/questions/73235/… \$\endgroup\$
    – Airatome
    Jan 30, 2016 at 16:33
  • \$\begingroup\$ I don't think it ties in at all. That question deals with the caster knowing the state of his concentration on a spell. This is just like knowing whether your spell casting is successful. This question however deals with knowing whether a caster knows that a successfully cast spell is actually doing its job when there are no known indicators. For instance it's pretty clear when an attack spell fails, but this one has no perceivable affect. You have to rely on other indicators like whether you can tell if the rakshasa is behaving as if he's compelled to tell the truth or not. \$\endgroup\$
    – Escoce
    Jan 30, 2016 at 16:46
  • \$\begingroup\$ I think that 'knows if a successfully cast spell is actually doing it's job or not when there are no known indicators....' bit is a good inclusion to your answer. That makes a lot of sense, as in your comparison to attack spells vs something a bit more discreet. \$\endgroup\$
    – Airatome
    Jan 30, 2016 at 16:56
  • 6
    \$\begingroup\$ I would say that in the general case your answer is correct. However, the wording on this specific spell seems to override the general case. "You know whether each creature succeeds or fails on its saving throw". If the raksasha makes no saving throw (because of immunity) the cleric gets no intuition either way for the creature, and thus should know that the spell failed to affect it at all. \$\endgroup\$
    – lithas
    Jan 30, 2016 at 17:38
  • \$\begingroup\$ The DM should roll the saving throw anyway just to prevent this sort of meta gaming. \$\endgroup\$
    – Escoce
    Jan 30, 2016 at 17:46

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .