Situation: A Gnome Bard sits in a prison cell, next to her in another cell sits a half-elf Ranger. There is one guard with them in the room, with two or three more in the next room. The Bard casts Sleep at a point that includes everyone but her.

My understanding of what happens: Half-elves are not immune to the spell Sleep, as the spell ignores only unconscious, undead, and immune-to-being-charmed creatures, of which half-elves are neither. Therefore, if the half-elf has sufficiently low hp to be affected, his current hp is deducted from the spell's remaining roll total before moving to the next target, but he does not fall asleep due to Fey Ancestry. This means that including any targets with Fey Ancestry in the area of a Sleep spell is just a waste of the spell's potential.

Question: Is my understanding of the situation correct? I'm not as much asking whether or not the half-elf should be affected, as I'm quite sure he should be, but more whether or not his hp would be deducted from the spell's roll.

Relevant PH fragments: (emphasis mine)

Fey Ancestry. You have advantage on saving throws against being charmed, and magic can't put you to sleep. (PH p.39)

Sleep (...) Creatures within 20 feet of a point you choose within range are affected in ascending order of their current hit points (ignoring unconscious creatures). Starting with the creature that has the lowest current hit points,(...) Subtract each creature’s hit points from the total before moving on to the creature with the next lowest hit points. A creature’s hit points must be equal to or less than the remaining total for that creature to be affected. Undead and creatures immune to being charmed aren’t affected by this spell. (PH p.276)

  • 1
    \$\begingroup\$ Hi Mark - welcome to RPG.SE. This is really good for your first question on the site. Please consider taking the site's tour and check out the help center if you have any questions. \$\endgroup\$
    – Conduit
    Commented Nov 29, 2017 at 15:36

4 Answers 4


You have it exactly correct. A (half-)elf in the area of the spell is affected in ascending HP order and sucks up the appropriate amount of the 'payload', it's just going to have no effect, and represents a waste of some of the spell's output.


I disagree. Magic can't put the half-elf to sleep; he is immune to magical sleep. The sleep spell will act as if he is not there.

A normal human who is unconscious doesn't get his HP subtracted from the spell total; I don't see the logic in subtracting the HP of someone who can't be affected by the spell's effect.

  • \$\begingroup\$ Comments are not for extended discussion; this conversation has been moved to chat. \$\endgroup\$ Commented Dec 1, 2017 at 15:43
  • 4
    \$\begingroup\$ I agree with this answer as a violation of the "legislative intent" of the spell's description, as long as it is recognized that the description itself fails to exempt "creatures immune to sleep." \$\endgroup\$ Commented Dec 5, 2017 at 20:08

Ported from the question posed by Mindwin: Are undead and creatures immune to charm excluded from the HP count of Sleep?

Looking at those two fragments:

Creatures within 20 feet of a point you choose within range are affected(...)

A creature’s hit points must be equal to or less than the remaining total for that creature to be affected.

it seems like "being affected" means having your hp deducted from the roll and being subjected to sleep, so therefore half-elves would be affected by the spell as a whole, then have their hp deducted from the total, and then be subjected to sleep, which they resist due to Fey Ancestry.

  • 1
    \$\begingroup\$ Your conclusion is that the spell becoming less effective by an amount equal to my hitpoint total is the spell affecting me? The hitpoint total is one of the things that determines who is affected by the spell. Immunity and undead are other ways. The spell specifies immunity to charm as a barrier to being affected because otherwise it's not obvious. (The spell doesn't have a "Charm" descriptor) Being immune to sleep is self-explanatory, which is why it doesn't need to be explicitly specified. \$\endgroup\$
    – Zeus
    Commented Nov 29, 2017 at 17:12

By my reading of the RAW (Half-)Elves would indeed 1: Be affected by the spell, thereby 2: Have their HP substracted from the total (on their "turn") and 3: Not actually fall asleep. This is because Fey Ancestry states that magic can't put them to sleep, not that they can't be affected by spells that attempt to do so.

That said, I believe the RAI is that they are completely unaffected by the spell and would thereby not have their HP substracted from the total. My arguments for this are:

First: The interaction is very specific (albeit not too rare) which means the odds of RAW and RAI conflicting are relatively high as the rules can't cover every possible interaction between features and spells.

Second: This question has been asked before and answered by Mike Mearls on Twitter. I know that Mearls' tweets are far from the gospel truth, but they can at least give some insight into the intent of at least one developer.

Third is a bit more of a reach and up to personal interpretation, delving into Flavour instead of Mechanics: Creatures immune to being Charmed are listed as being unaffected by the Sleep spell. Now it's true that Elves are not immune to being Charmed, but Fey Ancestry; the same feature that protects them from falling asleep, also grants advantage on saving throws against being Charmed. So it could be argued that while Elves are normally only partially resistant to Charm effects, Sleep is a specific subset of Charms that Elves are completely immune to, in which case they could be treated by the Sleep spell in the same way creatures immune to being Charmed altogether would be.


You must log in to answer this question.

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