[RPG] What happens to a Chain Lightning spell with invalid primary target and valid secondary targets

dnd-5espellstargeting

This question asks what happens when a single-target spell has an invalid target. (A target that is not legitimately permissible, not a target that is weak from illness or injury).
The answer to that question appears to be: that depends on whether one wishes to implement an older Sage Advice segment of a Dragon Talk podcast, or the more recent but optional written rules in Xanathar's Guide to Everything.

Assume for this question that I prefer XGtE.

The optional rule for resolving invalid spell targets states (XGtE, p. 85-86):

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.

Now consider the chain lightning spell, whose description says:

You create a bolt of lightning that arcs toward a target of your choice that you can see within range. Three bolts then leap from that target to as many as three other targets, each of which must be within 30 feet of the first target. A target can be a creature or an object and can be targeted by only one of the bolts.

Suppose my primary target for the spell is invalid, because it is not a creature or an object. If an example must be given, suppose it is an illusion. However, each of the three secondary targets of the spell are valid, being either creatures or objects.

Do I spend the spell slot with nothing happening at all?
Or does the slot get spent as the chain lightning impacts the illusion but does nothing to it and then leaps to the three valid targets (with the full effects given in the spell description)?


Note: I am assuming that an illusion is not an object, based largely on my interpretation that an illusion is not an "item" and on the text of the 14th-level School of Illusion wizard feature, Illusory Reality.

I am open to frame challenges that demonstrate that illusions are, in fact, objects – but such answers will be better if they then either provide a more appropriate example of something that is not a creature or an object, or demonstrate that the question is moot since everything is at least either a creature or an object.

Best Answer

I would definitely rule that the chain lightning continues on to its other targets. It does nothing to the invalid target (though in the case of the illusion this would be an “interaction” that could allow a save to see if you buy the illusion faking its reaction to being electrocuted), but the valid targets are affected exactly the same as they would be if the initial target is valid.

And frankly, if I was playing in a game where a DM ruled differently, I’d pretty strongly reconsider playing at that table.

It’s not that they don’t have a decent argument to make, based on the rules as written. It does say that it hits the initial target, and then continues on to the other targets. Xanathar’s does say that nothing happens to that target, which you could argue would include sparking the additional links in the chain. I’ve certainly played video games that have implemented things that way.

But I wouldn’t in D&D. It’s just finicky and counter-intuitive and an unnecessary “gotcha” moment. The game just does not need that. You’re already missing out on your primary attack, that’s a big loss. Compounding it by causing it to drop all the secondary attacks is just cruel, I think.

And arguing that “this is what the rules say!” is, I think, besides not really relevant—we care much more about a good game—also not totally accurate. The rules say that the spell does nothing “to” an invalid target. OK! Sure. Continuing on to target other individuals is not a thing done to the original target. Xanathar’s does not say that the spell fails as a whole, or that the invalid target is somehow no longer targeted—only that they aren’t affected. I think it’s a real stretch to nullify the rest of the spell on this basis. Not an impossible interpretation, but it is by no means a certain situation. And that only increases my feeling that ruling this way—bending over backwards to do so—is simply bad for the tables overall. You could rule that way, but you really, really don’t have to—and there’s a lot of good reasons why you shouldn’t.