D&D General A puzzle about spell casting in D&D

[Lanefan]

What about the many, many spells that don't have attack rolls or saving throws? Why does a thief player have to roll to see if his/her PC can actually perform the movements necessary to pull of a piece of stage magic, or pick a lock, or whatever; but the player of a wizard/MU does not have to roll to see if his/her PC can actually perform the movements necessary to cast a Knock Spell, or a Transmute Rock to Mud spell, or a Polymorph Self spell, or . . .

I think the question here might be one of whether the character is under any duress while performing whatever action.

A Thief is usually under some degree of duress when picking a pocket (doesn't want to get caught); a Fighter is usually under some degree of duress when swinging at a foe (needs to defend); and a caster is often under duress while casting a spell particularly in combat or when danger is near (or expected). In all of those cases it only makes sense some sort of competence check is required, yet only two of those three classes require one.

Things done under no duress e.g. a Ranger setting trip wires or a MU casting an Alarm spell near one's campsite, shouldn't need a check.

 

[Ruin Explorer]

That's more of the spell-nerf route, which IMO has already been travelled a considerable distance too far.

It's the only ultimate solution. So many level 5-9 spells are so powerful that it doesn't matter if you can screw them up.

You may also want to consider whether it would be any fun to have this system. It would mean that a lot more rounds where you try to cast a spell, and nothing happens.

This is important in combat, because right now it's already an issue, but we could easily see a situation where the caster just keeps fouling up their "casting check" and nothing happens. Non-casters have multiple attacks to lessen this issue but there isn't an equivalent for caster.

Further, you'd need to fundamentally re-design spells, because they're all designed and balanced on the basis that they're these things that automatically happen and only fail on a save or the like, and if that's no longer the case, they don't make the same degree of sense. My main feeling is that it would just make the game feel really naughty word as a caster in combat whilst still not preventing LFQW out of combat.

 

[Ruin Explorer]

I think the question here might be one of whether the character is under any duress while performing whatever action.

A Thief is usually under some degree of duress when picking a pocket (doesn't want to get caught); a Fighter is usually under some degree of duress when swinging at a foe (needs to defend); and a caster is often under duress while casting a spell particularly in combat or when danger is near (or expected). In all of those cases it only makes sense some sort of competence check is required, yet only two of those three classes require one.

Things done under no duress e.g. a Ranger setting trip wires or a MU casting an Alarm spell near one's campsite, shouldn't need a check.

The problem is that you haven't answered the question as to whether Vancian spells can be messed up by the caster. You're making un-evidenced assumption that they can be. There's no evidence in any edition that they can, AFAIK. Maybe in 3E? Or creatively interpreting 4E (but this isn't really a problem in 4E).

What makes you think they can be fluffed up? For all we know, the caster's arms and fingers are automatically moved by the spell itself. Yeah, if you were wearing armour/clothes that got in the way, the spell couldn't work, because it couldn't move you in the right way. Equally if a third party intervenes. I think that's actually the most plausible explanation based on the information we have.

Your point appears to be reliant wholly on the non-D&D-based, non-Vancian "people can screw up casting spells by being klutzes" deal.

Gygax and Vance have a lot to answer for.

 

[Lanefan]

That's only viable for Wizards, not a single other class in the game. On top of that, even for Wizards it's just turning it into X% chance of having the problem, and you then have to ensure no NPC Wizard have has his spellbook with that spell in fall into their hands. Removing problematic spells is a better solution but doesn't resolve LFQW.

Wizards are the biggest problem. Clerics are largely solved by knocking back their access to buff spells (which is where 3e went wrong) and leavign them with healing and divination; Bards are only solved by either a complete redesign or dropping them entirely as right now they're just another ho-hum caster variant.

And you're quite right, there's still an x-% chance of a Wizard getting a spell you'd rather they not, but I'll take x-% all day long if the other option is certainty.

Another thing regarding Wizards that a GM can do is to occasionally put their spellbooks at risk. Even since 1e, the odds of a character's items or possessions getting destroyed or damaged has dropped with each edition until now the chances are close to negligible; and IMO this is yet another ongoing design mistake.

 

[DND_Reborn]

I think the question here might be one of whether the character is under any duress while performing whatever action.

A Thief is usually under some degree of duress when picking a pocket (doesn't want to get caught); a Fighter is usually under some degree of duress when swinging at a foe (needs to defend); and a caster is often under duress while casting a spell particularly in combat or when danger is near (or expected). In all of those cases it only makes sense some sort of competence check is required, yet only two of those three classes require one.

Things done under no duress e.g. a Ranger setting trip wires or a MU casting an Alarm spell near one's campsite, shouldn't need a check.

Sigh... I avoided some of this logic because it opens a whole can of worms!

For example, a wizard is in combat with some trolls the party is trying to flee. The wizard, being attacked by a troll, is trying to cast Fog Cloud, hoping they can all slip away. He is clearly in a hazardous situation and under duress. Should he need to make a check?

 

[Lanefan]

It's the only ultimate solution. So many level 5-9 spells are so powerful that it doesn't matter if you can screw them up.

I've yet to play in or DM a game where anything higher than 5th-level spells were routinely PC-available other than on scrolls (rare); the game I play in is just now dipping its toe into 6th-level spell availability but only a few PCs have hit 12th level so they're not widespread yet.

You may also want to consider whether it would be any fun to have this system. It would mean that a lot more rounds where you try to cast a spell, and nothing happens.

Perhaps. IME it's more common that the caster spends multiple rounds NOT trying to cast and instead doing something else, as casting would be too risky.

This is important in combat, because right now it's already an issue, but we could easily see a situation where the caster just keeps fouling up their "casting check" and nothing happens. Non-casters have multiple attacks to lessen this issue but there isn't an equivalent for caster.

In no-duress situations I very rarely if ever make them roll.

Further, you'd need to fundamentally re-design spells, because they're all designed and balanced on the basis that they're these things that automatically happen and only fail on a save or the like, and if that's no longer the case, they don't make the same degree of sense. My main feeling is that it would just make the game feel really naughty word as a caster in combat whilst still not preventing LFQW out of combat.

Heh - I've already gone through the 1e spells (plus a bunch added since) from top to bottom and re-done the lot. Very tedious. But, worthwhile; and work I only really had to do once which will last me for as long as I run this system (i.e. probably forever).

The problem is that you haven't answered the question as to whether Vancian spells can be messed up by the caster. You're making un-evidenced assumption that they can be. There's no evidence in any edition that they can, AFAIK.

In 1e interrupting a Vancian spell is pretty easy. I agree that without an interruption casting is automatically successful, but I'm fine with that. What I'm not fine with is that in so many cases auto-success on the casting leads to auto-success* on the effects, particularly in combat and-or for AoE spells.

What makes you think they can be fluffed up? For all we know, the caster's arms and fingers are automatically moved by the spell itself. Yeah, if you were wearing armour/clothes that got in the way, the spell couldn't work, because it couldn't move you in the right way. Equally if a third party intervenes. I think that's actually the most plausible explanation based on the information we have.

Your point appears to be reliant wholly on the non-D&D-based, non-Vancian "people can screw up casting spells by being klutzes" deal.

Well, not quite.

@pemerton is, I think, targeting auto-success in casting as being the issue; I'm instead more targeting auto-success* in the effects that casting has.

* - where success is defined as at least forcing a save from the maximum amount of potentially-affected targets.

 

[Lanefan]

Sigh... I avoided some of this logic because it opens a whole can of worms!

For example, a wizard is in combat with some trolls the party is trying to flee. The wizard, being attacked by a troll, is trying to cast Fog Cloud, hoping they can all slip away. He is clearly in a hazardous situation and under duress. Should he need to make a check?

IMO if he's being attacked by a Troll he shouldn't be able to cast* at all. Auto-fail. Period. End of story.

* - unless he wants to do nothing except generate a wild magic surge and hope for the random best.

 

[Ruin Explorer]

It looks to me like the Doylist and Watsonian explanations for this whole issue may align.

I've yet to play in or DM a game where anything higher than 5th-level spells were routinely PC-available other than on scrolls (rare); the game I play in is just now dipping its toe into 6th-level spell availability but only a few PCs have hit 12th level so they're not widespread yet.

In 1e interrupting a Vancian spell is pretty easy. I agree that without an interruption casting is automatically successful, but I'm fine with that. What I'm not fine with is that in so many cases auto-success on the casting leads to auto-success* on the effects, particularly in combat and-or for AoE spells.

Re: 5th+ spells. Okay, but LFQW isn't a big problem until you get to 5th-level spells and higher. That was why people didn't discuss it much in 1E/2E, because so few groups played at that level, and tended to be bedecked in magic items if they did.

Re: "interrupting", that's missing the point, sorry if I was unclear.

I'm asking a more fundamental question.

Can the caster (not a third party) of a Vancian spell, accidentally (not intentionally) screw up the casting of that spell?

This entire thread and your comments seem to be premised on the assumption that they can. With non-Vancian spells, with many magic systems, absolutely the caster can screw it up. Harry Potter's approach for example has people screwing up spells all the time. It's a very common trope as I mentioned. But is it present in the Dying Earth books? No-one has answered that. I suspect the answer may be "no".

My suspicion is, and this 100% fits the D&D model, AFAIK, is that Vancian spells are essentially "self-casting". Once you've "uploaded" them into your head, they're self-executing programs that move your limbs as necessary. This is why there's no failure chance from the caster themselves. Only third parties interrupting the spell, or restrictions on movement that stop the limbs moving as the program attempts to execute can screw up the spell.

EDIT - Because nobody else seems to have read the Dying Earth books (which seem incredible, but whatever), I've just downloaded a collection of them on to my Kindle and we shall see I guess. Should be a trip if nothing else.

 

[DND_Reborn]

IMO if he's being attacked by a Troll he shouldn't be able to cast* at all. Auto-fail. Period. End of story.

* - unless he wants to do nothing except generate a wild magic surge and hope for the random best.

Perfectly fine if that is what you want, but it isn't supported in 5E.

IMO, I like the old idea of provoking OA when casting a spell in combat, and further in older editions with casting times so a spell could be disrupted, but that isn't the current flavor of the game.

 

[Lanefan]

Perfectly fine if that is what you want, but it isn't supported in 5E.

A good reason not to play 5e, I'd say; or play it with some major modofications.

IMO, I like the old idea of provoking OA when casting a spell in combat, and further in older editions with casting times so a spell could be disrupted, but that isn't the current flavor of the game.

Ditto, and were I ever to run 5e those elements would be introduced in a heartbeat.