Water is a terrible element (yet its still my favorite despite being horribly suboptimal in D&D)

While nominally working on my Al-Qadim Church game (I was also tinkering with some 5e stuff for the Princes of the Apocalypse game I'm playing in and working on a secret project), I've been going through a lot of old and new D&D stuff looking at water spells. My conclusion: no one can write a decent water spell to save their lives.

First problem: what sort of damage does water do? You can't really have a water elementalist without enough water spells, but you also can't really have a water elementalist in 5e without those spells doing damage. The Elemental Evil Player's Companion addresses this a bit by adding some spells that do bludgeoning damage. That makes sense: you conjure a big pile of water and whump someone with it. There was an old water blast in Al-Qadim as well, which is just a big spray of water (generally to the face). Unlike fire, however, there's really only one or two big watery face-blasts you can get away with it seems. Thus far one of my favorites is called Cone of Teeth from 2nd edition, where your watery blast takes the form of shark's teeth and rends your enemies.

Second problem: If you're not conjuring water, you're restricted to natural bodies of water. So many spells involve letting things normally on land bypass or function under water. Maybe you need to float or swim or breathe or see in the water? There's a spell for that, and a water elementalist can surely cast it. Possibly a couple different versions of it.

Third problem: Mist/fog is water, but its also air. A lot of the fog spells are listed as both air and water, and its easy to see why. Rain and weather are a combo of the two elements. So, for better or for worse, air and water need to share some spells or you have to decide to try to limit your possibly spells by restricting things to one or the other element. Related: does waterbreathing fall under air or water, or should there be a distinction between water breathing and lasting breath spells?

Fourth problem: ships are both air and water, but also neither. Ship spells are another issue all together. A lot of water spells are found in ship/pirate type games and supplements because it makes sense that they'd be useful in the setting. But is a spell to conjure rigging really a water spell? What about one that strengthens or weakens a ship's hull? Some of the ship spells are even nonsensical if you consider a water-wizard or water-sorcerer to be a "sea" type. Sea things often include storms and ships, by why the heck would a storm-sorcerer have the power to weaken a hull or conjure rigging with their innate magic? Why would a sea wizard gain any benefit for casting those spells which seem like tangentally related spells?

Fifth problem: Ice. Ice is the easy way out. Sure it makes a little sense since water is actually ice, but not when a water elementalist's primary means of attack is doing cold damage with a frostbite spell or a cone of cold. I certainly don't think a pirate in an Arabian Adventure (or, honestly, pirates in general) should be using Ice Knife and Snilloc's Snowball Swarm as their primary offensive spells till they hit 5th level. Ice sure can make for a nicely thematic set of spells though, especially if you add in wind/weather (and maybe some enchantments and northern lights type spells).

Sixth problem: Conjuration or Transmutation (or Evocation)? Related to the second problem, you can easily make many a water spell a conjuration and then you have no ridiculous need to cast it in/near a body of water. This could be the case with so many spells that it blurs the artifical lines between the codified 8 schools of magic in D&D. In 5e I can really see why they remove a lot of those sorts of limits in spells so tidal wave and wall of water don't need to rely on the surrounding environment. Still, if the schools weren't such a big thing it might be nice for them to get a boost in power when cast in the right environment.

So, what's the verdict? Its hard to concoct some original water spells. I bet everyone could come up with about a 6-12 fairly distinct ones (depending on the granularity) but everyone will basically cover the same ground with their water spells (stay dry, go underwater, get over the water, blast of water, something ice, something fog, manipulate the current/flow...). So I'm left with a few different feelings on this. One can either stretch the definitions of water a bit (and if you do that, why not add in a scalding steam spell alongside a bit of ice and fog), or accept the fact that if a player wants a water elementalist its part of the DM's job to ensure the setting supports water (i.e. ships or rivers will be involved). The latter is a bit unsatisfying, so maybe there's a few more ways one can make spells which are thematically watery (a spell that smooths and polishes objects, or alternately bloats and cracks them; scrying in a pool of water) that can be used to expand the list a bit.

Side note: all the elements are a bit repetitive, but somehow air, earth, and fire ones don't seem quite as bad. 

Revisiting D&D Schools of Magic

As I've been thinking that a D&D style game is hindered by overly generic classes (the exception being a game with only a small number of classes: maybe 3-5), and doing away with the generic Wizard or Magic-User might open up a few more doors. Replacing the generic class with the specialty versions prevents one class from stealing the thunder of its subclasses, and allows multiple of those types of characters to coexist a little better in play, I think. The natural D&D thing to replace the standard Wizard with is thematic specialist Wizards. But a few seem a bit off.

One issue, in Second and Third edition, the schools are somewhat poorly defined. Transmutation is a heinously powerful school, and Abjuration is horribly ignored. There's no contesting that an Abjurer has a terrible selection of generally useless spells (they don't even get all the defensive ones!) and Transmutation has a strange abundance where things like Burning Hands could just as well be evocation. Conjuration, too, in 3rd edition gets a strange set of ranged touch attack spells which could easily fit into evocation. This didn't matter so much in first edition where schools were largely story-based (until Dragonlance Adventures came along and divvied them up between the different orders of High Sorcery).

Another issue is reversible spells: spells that summon things are clearly conjuration/summoning, but spells to dismiss them are abjuration. Two sides of the same coin should be maybe combined into one school, no? I suppose this is just another instance of schools being poorly defined.

Yet another issue is schools are only slightly relevant. Pathfinder almost neuters the idea of an opposition school for specialists by just forcing the wizard to learn the spell as one level higher. The Illusionist can still cast Animate Dead, it just is harder. So the specialty goes from a meaningful choice to a thematic choice. Any wizard can basically cast any spell. First edition Illusionists simply cast a slightly different set of spells that normal Magic-Users had, and while not exactly weaker, they were overshadowed by the generalist Magic-User.

So what would the fix look like for a game with a healthy set of classes but without a generic ones? I think you could still use a generic Wizard (who's intelligence and prolonged study grants magical mastery) but require a specialty: no generalists allowed. Second, give a smaller set of coherent specialists. For example, a Binder or Conjurer would focus on various protection-from-X spells, wards, summon monster spells, and a few others like maybe Charm Person and Hold Person. A thematic and coherent set. An Enchanter or Beguiler might focus on charms, illusions, and people-based transmutations (think Bulls Strength or Polymorph Other).  A Necromancer would have their traditional suite of spells (death spells, animating and controlling and destroying undead, fear and some curses). An Elementalist could take the role of the war wizard and evoker throwing around fireballs and lightning bolts. An artificer or alchemist would enchant items, change and create physical items, and other genuine transmutation spells.

This set-up just did away with the abjurer (mostly by giving his spells to the conjurer) and the illusionist (by folding him into the Enchanter). I'd also throw tighten up some of the definitions and probably allow some subschools to be shared (people-based transmutations could belong to an Enchanter or an Alchemist, Fear or Curses could be had by an Enchanter or Necromancer, etc.). By properly subsetting schools into more thematic lists, a given wizard could slowly grow in mastery of his own specialty and possibly sacrificing a little depth for the breadth of choosing spells from outside his specialty list. I think it'll work as long as each class has 3-4 clearly thematic subschools.

So what would this look like? Assume you chose Enchanter: