From my reading of the D&D 5e PHB, as well as other answers on this site (see Are melee combatants limited to standing around saying "I attack"?), it appears that martial combat is designed to meet these criteria:

  • All martial maneuvers can be resolved using the same lightweight core mechanics (attack bonus, weapon damage, [dis]advantage, etc.).
  • The only restrictions on what maneuvers can be attempted are the player's imagination and the DM's interpretation of what is physically possible.

Doing the same analysis for spellcasting, it appears that:

  • All spells can be resolved using the same lightweight core mechanics (concentration, components, area-of-effect, saves, etc.)
  • ...but many spells additionally define their own unique rules (see Does Burning Hands really require touching thumbs?).
  • The list of available spells is complete. Freedom to improvise new spells is not granted to players.

So, if it was a conscious design decision to move away from the fixed power-list approach of 4e for martial combat to an improvisational model for 5e, why would spellcasting in 5e be limited to a pre-defined list of unique maneuvers?


The analysis of the purpose and substance of the change from 4e to 5e that inspired this question is inaccurate.

Before 4th edition there was 3rd edition, and before that 2nd and 1st, and before those there were at least two other editions (maybe more) that don't neatly fit into the numbering scheme. In all except 4e, combat was more loosely defined. In the editions before 3e, combat was even more improvisational than in 5e. However, in every edition, magic was nailed down. (You can improvise some clever things using the effects of the spells, but the actual effects of the spells have always been fixed.) D&D 4th edition is the odd one out, making all PCs use a unified powers system.

So it's not that 5e added improvisation where there wasn't before while inexplicably leaving some powers out of the new treatment, it returned to the blend of improvisation and detail that was common in 2nd edition and earlier. If the intention was to take 4e as the starting point and then add improvisational combat action, then it would make sense to wonder why spellcasting in particular didn't get that treatment. But in actuality, 4e wasn't used as the base, the all-previous-to-4th editions that already had lots of improvisational combat stuff were used as the base, and a few 4e ideas were added back in.

So the mix of very improvisational combat and rigidly-defined spellcasting is because that's how D&D always was, more or less, before 4th edition, and 5e is a deliberate return to that way of playing.

  • \$\begingroup\$ So the choice for improvisational martial combat and the choice for non-improvisational spellcasting are both for historic/meta reasons, rather than for gameplay reasons? \$\endgroup\$ – starchild Sep 27 '14 at 3:46
  • 3
    \$\begingroup\$ I suppose I am really more interested in why one might design a system with both improvisational and non-improvisational martial character options but only non-improvisational spellcasting options. That is, from a purely game-design standpoint, free from the baggage of legacy expectations. \$\endgroup\$ – starchild Sep 27 '14 at 3:49
  • 2
    \$\begingroup\$ @starchild That sort of wondering about designer intent is beyond the scope of our Q&A format. But speaking to design reasons that can be known: it worked well for decades, so it's a proven design choice, whatever other reasons might motivate it. \$\endgroup\$ – SevenSidedDie Sep 27 '14 at 6:32
  • 1
    \$\begingroup\$ @starchild: Improvisational magic systems tend to require a lot of GM adjudication, in a manner that would be a major innovation for D&D family games. Since the main imperative for D&D5e was "Stop the loss of players who are switching to Pathfinder!", requiring new skills from the DM would not have been a commercially attractive idea. \$\endgroup\$ – John Dallman Nov 17 '16 at 20:13

Here's a pretty good writeup/critique of Vancian Magic that gives some good background on why D&D uses the system it does, and makes some good points that may answer your question as I've understood it.

I'm currently reading Vance's Tales of the Dying Earth. In it, Earth is nearing the end of its life (well, the Sun is...) and the great ages of magic have already passed. There are still a number of great wizards in the land, but very few people are really writing new spells anymore; most magic-users are just studying old ones.

In this world, magic taps into some kind of universal energy. Cast spells require the caster to memorize cryptic and powerful words that almost seem to have a life and energy of their own. Just memorizing the words is an act of magic in and of itself. Even one of the more powerful wizards in the story can only remember a handful of such spells at a time without losing them. In the one story, the character Cugel, who is not a magician, memorizes two spells that 'sit in his head like stones'.

Now, of course, we say that D&D's magic is 'Vancian', but they've definitely taken some liberties with it to make things more balanced, and I think 5e does a good job with cantrips to ensure magicians don't feel 'wasted' after one battle. With an unlimited scope for spells, it'd be impossible to see where holes exist that could break the system.

Physical combat uses physical weapons with a limited scope allowed for damage—you can't just make up your own weapons with unlimited features/damage. This is easily reigned in by reality—"No, Mr. Fighter, you cannot craft a 300ft tall 100d6 Sword of Flaming Glory". But given the unlimited and fantastical nature of magic, well, why can't I summon a giant creature with a 300ft tall sword to do it for me? Well, because it breaks things...


Quite simply, I think it's reasonable answer to say "because magic is magic".

We are all reasonably familiar with the physical world, and can have a good guess at what sort of martial manoeuvers might work; whether a good run-up would improve the success rate or effect; etc. But magic is kind of arbitrary.

Besides, magic in D&D has typically been extremely powerful even though it's rigidly defined. If a player could improvise magical spells, inventing the kind of effect they wanted based on the situation they were in, you'd have the classic problem of overpowered magic-users, but even more strongly than before. When even the laws of physics and the nature of the universe can be bent however you want, I don't think any DM could reasonably hold that in check. It's doubtful that making magic even more powerful and flexible could be balanced (especially when compared to martial characters) without massive changes to the rest of the game, at which point it wouldn't be D&D as we know it. (Historically, the result of making D&D magic flexible was the creation of Ars Magica and Mage: The Awakening!)

This problem of increased power doesn't happen with mundane actions, because mundane actions are inherently limited by what's mundanely possible. So it seems perfectly reasonable to allow characters to perform any mundane physical stunts they can think of, but to restrict magical effects/spells to very specific violations of reality.

  • 4
    \$\begingroup\$ That doesn't stop other RPGs from doing just that though, so that can't be the reason, or at least not the whole reason. \$\endgroup\$ – SevenSidedDie Sep 30 '14 at 22:18
  • 1
    \$\begingroup\$ Or to be more explicit: plenty of RPGs do let players create arbitrary and improvised magic, and they do so to great success without balance or the universe coming apart at the seams, so this doesn't hold much water. (Except perhaps in the D&D context which, in editions other than 4e, has made magic extremely powerful.) \$\endgroup\$ – doppelgreener Oct 1 '14 at 2:34
  • 1
    \$\begingroup\$ @SevenSidedDie That is a good point, I'm sure it could be done. I was thinking more along the lines of doppelgreener's proviso; D&D magic has typically been extremely powerful, and I have very strong reservations that this could be balanced (especially when compared to martial characters). \$\endgroup\$ – Andrzej Doyle Oct 1 '14 at 8:18
  • 1
    \$\begingroup\$ @AndrzejDoyle I think that comment makes a big difference. I put that reasoning, and a bit of related ideas (like the origin of Mage: The Awakening) into the answer itself. How's that look? \$\endgroup\$ – SevenSidedDie Oct 2 '14 at 19:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.