The history of divine magic in D&D

In D&D, when a cleric casts a spell, what really happens? The rules are pretty clear on the mechanics of bless or cure light wounds but what about the story? Are the deities aloof, or do they intervene in the world? D&D has taken a number of positions on this issue.

The earliest position is that clerical spells are similar to wizard spells, but a cleric prays for each spell on a regular basis. His deity grants that spell to him, and when he wants to cast it again, he must pray again the next day. Acting against the tenets of the religion may interfere with a cleric's spellcasting powers at the DM's discretion. This is the basic position of the earliest editions through early second edition.

One alternative that is generally presented in early D&D is that a cleric may worship a philosophy or force (i.e. goodness) instead of a deity. Who 'grants' the cleric spells isn't really defined in this alternative system. This system, however, seems to be targeted at groups who want to portray clerics but don't want to get into the details of developing and portraying a fictional religion.

Over the course of second edition, a few alternative perspectives arose. Al-Qadim developed a religion of enlightened gods who represented ideals (bravery, adventure, wisdom, etc.) and let their churches and priests be differentiated based on cultural aspects (the Pantheonists only admitted 5 enlightened gods, while the Temple of 10,000 Gods admitted them all). The Dark Sun setting, instead, split clerics into elemental worshippers and those who gained their power by serving neigh-omnipotent sorcerer-kings. Elemental priests got their spells from a nebulous somewhere, whereas the templars clearly got their spells from an active agent in the world.

Third edition presents clerics much as before, though they have ways of channeling their deity's power without spells. Third edition clerics are still at the mercy of the DM if they act against the wishes of their deities.

Finally, fourth edition frees clerics from the tyranny of the DM by making the gods aloof in the world. Clerical magic (same for Paladins and Avengers, though not Invokers or Runepriests) is granted by an ordination ceremony, which allows the Cleric to channel the powers of the astral sea. They technically don't need to worship the deity for a second after that ceremony, and can continue casting spells all day long. This also differentiates divine magic from arcane magic, as divine magic arises from the astral sea. One strange effect here, however, is that all clerics draw upon the same pool of powers, so clerics of evil deities, sea deities, or darkness are also liable to be slaying their foes with holy radiance. Divorcing clerical magic from worship of the deities may seem odd, but it also means that a cleric cannot lose spellcasting powers if he does something against his faith. Given the edition's power system, this is important as any character who loses access to his powers is crippled.

Two of the more curious systems arose in the second edition Spells and Magic book. While tied to a complex spell-point system, divine magic was presented with two interesting options that DMs could use. The first is ritual prayer, and the second is conditional magic.

In the ritual prayer system, the cleric's deity isn't invested in the cleric's prior actions, but just the act of completing the ritual properly:

"In this system, the deity or power is concerned more with the priest’s show of devotion and observance of the proper form, and considers the priest’s actual situation to be irrelevant—after all, martyrs are made every day."

Here, we get a solid system for clerical magic in which the deities are aloof and seem to take no part in the affairs of the world. Clerical magic is ritualized, and rushing the ritual is likely to reduce the power of the spell. Conversely, ritual prayer made by a ritually pure priest (observant, not necessarily pious) with the proper offerings ad the proper place and conducted with care can produce spectacular effects. With the right casting conditions, even previously expended spells can be cast again! Its all about the motions, however. In this way, the ritual prayer system can be seen as a precursor to the 4e system. Moreover, it links the cosmology (aloof deities) to the magic of clerics.

These aloof deities make divine magic feel much more like the magic in the D20 Conan game and Conan universe in general. While independent sorcerers exist, many are trained in the priesthoods of these aloof and alien deities. Whether or not the god even exists is not important: it is the ritual training of the priests that give them sorcerous abilities.

The inverse system is also presented as Conditional Magic. In this system, the gods carefully answer each prayer based not necessarily on the urgency of the petitioner but on the probability to influence the world in a way the deity prefers. When a cleric casts a spell, you tally up the total of positive and negative conditions. If it is skewed heavily positive (casting a spell to smite the deity's enemies in a holy place, etc.) the spell is cast at a higher caster level or reduced in cost. If it is skewed negatively, the spell is more costly or weaker.

The conditional magic system presents interventionist deities, and their tools are their clerics. The mechanics explicitly reward clerics for advancing their deity's agenda and penalties for working against it. This codifies some of the DM fiat where by the DM may simply neuter a cleric by taking away all his spells.

The idea of interventionist deities can be clearly seen in the Dark Sun novel Rise and Fall of a Dragon King:

"O Mighty Hamanu! Lion-King, Lord, and Master, hear me!"A distant voice echoed in Hamanu's mind. The totality of his awareness raced backward, along a silver thread of consciousness through the Unseen netherworld, to the source.

"Armor! I crave invincible armor and earthquake!

"The Gray was charged with acid needles, and Hamanu's vision, when he opened his sulphur eyes above the desperate templar, was streaked with lurid colors. There was powerful magic—someone else's powerful magic—in the vicinity.

"O Mighty Hamanu! Hammer of the World! Grant me invincible armor and earthquake!

"Squinting through the magic, Hamanu made out chaos and bloodshed: a full cohort of his own templars outnumbered by ragtag brigands. Or, not brigands. Another moment's study discerned a well-armed, well-drilled force disguised for brigandage. In the midst of the Urikites' impending defeat, a militant, a human man with tears of panic streaming down his face, raised his bronze medallion and entreated the Lion-King for the third time:

"O Mighty Lion, grant me invincible armor and earthquake, lest I die!

"A wise invocation—in its way. An earthquake, if Hamanu empowered the spell to create one, would swallow everything on the battlefield, friend and foe alike, except for the invincibly armored militant. Though sacrifice was necessary in battle, the Lion-King of Urik was not in the habit of rewarding militants who'd save themselves and doom the lesser ranks and mercenaries they led. He'd have considered granting the earthquake while withholding the invincible armor—and savored the militant's death—if the netherworld turbulence wouldn't have negated any spell he granted.

If you want to portray deities like that, the basic D&D cleric system of divine magic doesn't quite cut it, though using the conditional magic system definitely helps.

These two systems nicely define and differentiate two distinct types of deities: aloof vs interventionist. It is even somewhat easy to imagine distinct classes using different systems, such as a Dark Sun game where the elemental clerics use ritual prayer while templars (and druids?) use conditional magic.

These two systems are but a small part of the magic of the cleric class (and related classes). The new edition of D&D will probably present a generic system as D&D has historically done, which can be modified somewhat to suit different worlds. I do hope, however, they they make it easy for these distinct visions of divinity to be mechanically bound to the setting.

Certain classes are even linked to the role of the deities in the game. The 4e Invoker class is essentially a prophet compared to the cleric's priest. I use these in the sense of the Hebrew bible and the ancient near east, where the priests worked the temple and prophets were charismatic leaders who claimed to speak directly for their deity, often leading small bands of devoted followers. If the deities are all interventionist and use their churches and clerics to intervene, there isn't much space for a distinct prophet class. Conversely, in a game where deities are largely aloof, a prophet who is called to reform the church, lead the chosen to safety, or combat the deity's enemies becomes very different and viable class from the ritual-prayer priest.

Ultimately, the state of divine magic in D&D is a prime case for wedding the rules to the system. I hope the designers of the new edition keep this in mind.

Classes for D&D Next

I've discussed what makes a class, and some class options here before, but when I lay my ideas out, they don't look too bad. This is the latest version of my list of the best classes to include in D&D Next, or any retroclone. I've put the list together here and broken it down into four categories: basic, intermediate, advanced, and later-on. These could have just as easily been basic, expert, master, optional, or common, uncommon, rare, unique. The labels for the four classes aren't particularly important, they just show a ranking of what is probably core to every D&D game, moves onto traditional classes, then some less-traditional or non-western classes, and some others that could be added.

There are lots of classes out there for various versions of D&D. Some are well done (thematic and make balanced or at least useful party members), others less so. Many are designed for PCs, while others are for NPCs. I've tried to narrow the list down to the classes that I think offer broad coverage of character archetypes and are appropriate for PC use. Each of these probably can be built with a customizable "build" option, so fighters might have fighting styles, rogues have their talents, clerics their domains and deities, and wizards their specialty schools. Likewise, druids might venerate specific forces, sorcerers might have their bloodlines, warlocks their pacts, ans so forth.

Just about every game will have the four basic classes. The intermediate classes add some more options, but don't really stray from the realm of standard (A)D&D fantasy. Advanced classes do dip into some non-western options, and options that might provide more moral dilemmas and may not be appropriate for every heroic game. Finally, I have a few concepts that are probably class-worthy, but might be better added in specific campaign settings. This simply follows some of the class ideas laid out by the D&D Next designers, its not my innovation at all.

Getting D&D cantrips right

I gave my opinion of the wizard class and cantrips before, and I'm still more or less of that mind. So the first playtest with boring at-will attack cantrips was a bit disappointing to me. Thankfully, Mike Mearls has indicated they are or have been considering a different approach to cantrips and that cantrips might be a little strong as they're currently implemented. The whole and most relevant quote goes like this:

mepstein73: Hello! Just wondering why the wizard's cantrips are so strong. Ray of Frost can end combat pretty quickly, and Magic Missile is very powerful if it's unlimited/day. 

Mearls: I think that for at-will abilities, we might have made them a little overpowered a bit in terms of math and feel. For instance, does it feel OK that magic missile does auto damage every round? The speed thing on ray of frost is tricky, because it can vary from being very powerful to being useless. I think getting the minor spells right will take a few iterations. 

Jeremy Crawford: When we playtest things, we prefer to start powerful and tone things down, rather than starting weak and beefing things up, hence the spells' potency.

Some friends reports and forum posts have suggested that the at-will cantrips and the auto-hitting magic missile specifically make the playtest wizard boring. The wizard isn't even rolling to hit with magic missile, whereas the sun cleric at least needs to hit with his lazers.

I really hope they tone down the cantrips in D&D Next. Hopefully in the next version of the play test. I think that these cantrips need to be really carefully crafted, as well as numerous. There should be clear low-level effects and/or guidelines in the DMG for adjudicating the low-level effects of cantrips. Low-level effects, or example, should be seen in the light spell. The light cantrip shouldn't be better than a torch. Call it torchlight, floating lantern, or whatever. Light, as a first level spell, used to double as a blinding spell (target your enemies eyes!), so that seems about right for a first level spell. Plus there's that creative use of spells, right? A sorcerer would almost never take light as a first-level spell, but creative uses makes these spells useful.

There should also be two or more cantrips per school of magic (just like at each spell level). This is both in general for variety, but also assuming that specialty wizards will reappear. I love abjurers, but if you only have one abjuration cantrip option, that's not too thematic.


Now, one problem with fantasy literature, TV, and film is that the wizard often has a role of an aloof mentor or as a bumbling fool. In terms of He-Man, the wizards are the Sorceress and Orko for the good guys. The PC wizard needs to be something a little in-between. When used creatively, and in the right situation, wizard spells should make an encounter dramatically easier for the PCs.

So I'm going to try my hand here at two possible cantrips:

Lingering Innovations: The Runepriest and Runecaster

Runes have been a source of inspiration in D&D, probably since the beginning. Nordic rune magic can be seen in the old spell lists, but the journey from a few spells to the recent runepriest class isn't strait forward.

I've been pretty critical of the runepriest class in the past. I just don't like how it was implemented. For example, why strength as the key ability score? Constitution, at least, would make some sense given the lore of sacrifice to gain the power of runes. Wisdom as well, since runes have a tradition of insight about them, rather than intellectual cunning per se. What is interesting is the choice to make rune magic a type of divine magic, and labeling its wielders priests. There is a precedent for this, however. I'm going to stick just to the D&D examples of runes here though, rather than going afield into other RPGs.

Basic D&D

The first systematic predecessor to the runepriest that I've found so far comes in 1988. The Northern Reaches (GAZ7) for the basic D&D system doesn't present a specific class of rune-users, but instead gives northern priests access to a system of rune magic. I actually think its quite ingenious for the time. Rather than introducing a new runecaster class (see below), clerics gain access to a small series of second- and third-level rune spells (2 interpret, 2 bless, 3 know, and 3 inscribe rune), which they uses to access a system of runes.

The new pace of healing in D&D

While I'm generally happy with what I've seen in the D&D Next playtest, one thing that I'm still a bit unhappy about is the rate of healing. Now, I've written on healing here before, but I think this warrents a new post. Basically, I think they've got the rate of healing wrong for the timeline.

I'd like to spell out first why the daily healing mechanic found in the D&D Next playest (and 4e) is wrong, and suggest one possible solution for this.

Recall that the adventure is the new encounter. This means that the standard of balance is adventure-based, and the adventure is the timing unit that designers are supposed to be thinking about. So fighters might do more in combat but less during exploration and interaction, and that's fine. A fighter's background and theme and race may play more of a role in the other areas to make up for his class not doing that.

But the rate of healing, along with the rate of spell recovery, is daily. The day and the adventure are not the same. So I'm really opposed to daily healing, meaning a PC recovers all or most of his health (hit points, surges, hit dice, whatever) each day.

Lingering Innovations: The D&D Shaman

Its no secret that I like the idea of having a shaman class in D&D. But I will reveal another secret: I've been updating the wikipedia page on the Shaman.

Now, its not perfect. My update probably still violates a lot of wikipedia good form and such, but the information is significantly improved. I'm trying not to go overboard with including shaman information everywhere since many of the other class pages (the whole section in general) needs some significant revision. But when I finish, it should be much more useful to people. Its hard to do this type of research though when you're not in the US with access to large piles of D&D books though. Feel free to join in on the work. Wikipedia is always a collaborative effort.

I should also note, maybe, that the shaman was listed as one of the 15 or so core D&D classes before I started messing with it. But I think wikipedia had that one right. The shaman has a long and interesting history in D&D, and I've learned a few quite interesting things thus far about the Shaman.

Flatter Math

Yesterday's Legends and Lore column rocks.

I was concerned about the flattening of the math from the playtest because the ability score modifiers are the same as in 4e and 3e. But now I think I like what they're doing with the math here. Dumping a stat completely (which often happened in 4e, maybe less-so in 3e) becomes more of a risk when each ability score does something useful for your character. So that 18 starting intelligence might make your wizard more potent, but it will probably limit your chracter in other ways.

Weapons, armor, and fighting-men

Fighters have always been the "simple" class in D&D. I don't want to argue that simple classes are bad: after playing different RPGs with different people, its pretty clear to me that some people would do better with a "simple" class while others prefer more options that a complex class offers. This doesn't mean that a fighter should only be simple, however. I think the D&D Next authors know this.

Why, however, do weapons and armor get just a few pages allocated in the rules when spells get significantly more? Though I can't find a reference anywhere, I think I read somewhere that the D&D sorcerer was developed as a second Arcane class because such a large portion of the third edition players handbook was devoted to arcane spells (i.e. the wizard). Similarly, why is such a small portion devoted to martial characters fighting implements?