Let’s start out on the low end of the scale with a “blessing” type spell. This kind of thing should be easy and quick to cast, but not have any spectacular (or even necessarily noticeable) effects. For the example, I’ll use a battle spell that gives the target a bonus to combat rolls. The first step is to determine the spell’s base DN:
- How overtly magical are the spell’s effects? This is easy enough, since this exact type of spell is given as an example. The spell gives the recipient a small bonus to attack rolls, which means that there are no visible effects--the target just appears to fight particularly well at the moment. DN=0.
- How powerful are the effects of the spell? This one’s also right there in the examples: Negligible effect, DN 0.
- How long does the spell last? This spell lasts for one battle. Since that can vary according to how well (or poorly) each side fights, we’ll call it a scene, which adds one to the DN.
- How far-reaching are the spell’s effects? You could theoretically argue that this spell affects the people being attacked as well as the attacker, but if you do that you’re going to end up getting into chaos theory and every spell will have a DN 5 for this category. Let’s keep it simple: The spell affects one character’s rolls, so let’s go with a DN of 1 for this one (assuming the caster isn’t casting the spell on himself), which brings the total DN up to 2.
- How well-defined are the spell’s effects? Here’s where we run into a problem. The spell has a fairly obvious game effect, but it’s also pretty well-defined, which could penalize players who limit the mechanics of a spell (since a spell that just gives a bonus to Damage Bonus is technically more “well-defined” than a spell that gives “combat bonuses”). The question here probably needs to be rephrased a bit. While one aspect of this question is indeed “how much room does the spell description leave for the GM to twist the caster’s intent” the point here isn’t really wish lawyering. It’s more about the fact that spells whose effects are finely-controlled (like the offensive spell that leaves allies unscathed) should be more difficult than spells that unleash brute magical power; it’s easier to knock down a door or give someone amnesia with magic than it is to pick a lock or alter a specific memory. So, let’s change this question to “How precise or finely controlled are the effects of the spell?” This helps make the question a bit more useful in general. For this particular spell (and all “pure mechanics” spells), a DN of 0 for “undefined” works, since the spell doesn’t have any definite game world effects. Whether the spell affects the story or characters ultimately depends on whether it changes the results of the rolls that it’s modifying.
With a DN of 2, this is already going to be a 1-round spell, but in the interest of knocking the DN level all the way down to zero, we may as well go ahead and toss in a few ritual elements. In this case, the caster will pray to Ares (or some other war god/heroic warrior ancestor appropriate to the setting) and place a drop of blood on the subject’s sword “to make it hunger for more.” Now that we know the ritual, we rate it on each of the five categories:
- Symbolic Elements: It’s a pretty vague spell, but not completely lame, so we’ll give it a -2 modifier.
- Knowledge: If the caster were a powerful warrior and the spell was framed as transferring his battle experience to the recipient, knowledge might be appropriate here. Since that’s not the case, this category doesn’t warrant any DN reduction.
- Material Elements: Blood’s always a good magical component, and the player made it make sense within the context of the spell, but since the blood isn’t specifically coming from the enemy that the spell’s target will be fighting, it’s only worth a -1 DN. If the caster uses his own blood and rolls a Bad Break, the GM should of course inform the spell’s subject of his homicidal feelings towards the caster.
- Setting: Other than “before battle” (the only time it’s really useful), there aren’t any setting elements to consider here, so the DN modifier is zero.
- Requirements: The player could probably get a -1 DN modifier by agreeing to limit the benefits of the spell to attacks by the weapon that received the blood drop, but since the base DN is low to begin with, that’s probably too limiting to bother with, so this category doesn’t change the DN.
When all’s said and done, we’ve got a -3 modifier for a ritual with a base DN of 2. Since a negative DN isn’t possible, that means the spell has a DN of 0. While the player doesn’t “get” anything system-wise for rituals that exceed the spell’s DN, the GM should feel free to award Yum Yums to players who come up with particularly good rituals.
OK, so it looks like, after a few adjustments, this system works pretty well for simple spells with relatively small effects. Now let’s try something on the other end of the spectrum: Raising the dead. Unless you’re playing a game like D&D where there are resurrection shops on every street corner, this should not be an easy spell to cast. Let’s see if our system properly reflects that:
- How overtly magical are the spell’s effects? Assuming that the person in question is undeniably dead, this is a solid DN 5 (cannot be rationally explained).
- How powerful are the effects of the spell? Hey, what do you know? This one’s an example on the table (funny how that works), so it’s easy to assign a DN of 4 for this step.
- How long does the spell last? Assuming the caster wants the person to stay alive until he dies naturally (if not necessarily of natural causes), this one’s permanent, so we add another five to the DN for a total of 14.
- How far-reaching are the spell’s effects? Since we’re avoiding chaos theory and philosophical games here and a dead caster can’t resurrect himself, the DN here is one, for an overall spell DN of 15.
- How precise or finely controlled are the effects of the spell? The player casting this spell has read “The Monkey’s Paw,” so he makes sure to specify that the person returns without the disease or wound that ended him, with full control of his mind, and free of any evil influences, brain damage, or other debilitating conditions. Since that’s pretty precise (and spoils all the GM’s fun), the DN goes up by another five, bringing the grand total up to 20.
So far, 20 sounds good for a base DN. It’s impossible to make without some Yum Yum expenditure, but it’s doesn’t max out the spell DN system (which runs to 25), leaving room for really crazy stuff like travelling through time and reshaping the world to your own nutty preferences. Of course, the player now gets a chance to lower the DN by designing a ritual, which could throw a monkey wrench into things.
The player comes up with the following ritual: “Since we’ve got the sword that Whedo the Mad used to kill Tudyk, I write Tudyk’s true name on the blade using paint mixed with the cheap wine that Tudyk always liked to drink. Then on the anniversary of Tudyk’s death, we dig up his grave and I have my loyal slave, Jack, remove the burial shroud and wrap it around himself. Then I place the point of the sword into what remains of Tudyk’s rib cage with the point in the heart (or whatever’s left of it), mimicking the wound that killed him. While doing this, I recite the Viaticum backwards. Then I remove the sword from Tudyk’s corpse and plunge it into Jack’s heart, transferring Tudyk’s death wound to the slave.”
That’s a pretty hard core ritual, complete with human sacrifice, so it should probably take the DN down quite a bit. Here’s what the tables tell us:
- Symbolic Elements: Everything seems pretty appropriate, and the “transferring the death wound” idea is good magical thinking, so this is worth a solid -4 to the DN.
- Knowledge: Since the caster is incorporating his knowledge of Tudyk’s death into the spell, I’ll give him a -2 DN for this one.
- Material Elements: The first question is, “does the body count as a material element of the spell?” Since it’s the thing being magically acted upon, probably not. This brings up the question of whether the spell should be more difficult if the body isn’t present, and, more generally, whether factors specific to the spell being attempted can affect DN. Assuming the body is still more or less intact, probably not--we’ll assume that if Tudyk’s body is mostly whole, he can claw his way out of the grave just like Dean Winchester, even if the spell is cast miles away. If the body’s badly damaged, though, the DN of the spell should probably go up--+1 DN for minor decomposition, +5 if the body has been more or less reduced to its component atoms (through cremation, a nuclear blast, etc). That means we probably need to make a note about spell specific DN adjustments somewhere in the magic system rules. These can range from -5 to +5. Since the body doesn’t count, that leaves with the sword (and maybe poor Jack the slave’s blood/corpse) for material elements. That’s probably worth another -4 DN, for a running total of -10.
- Setting: The spell’s being cast on the anniversary of the death and at the grave of the person to be resurrected, which makes sense. We’ll give it a -3.
- Requirements: In order to make the spell work, the caster has to kill Jack. Normally combat would probably count as a multi-round, multiple action requirement, but since Jack’s unarmed, wrapped in a burial shroud (and therefore at least partially immobilized and blinded), and probably not expecting to be stabbed, he’s going to have a severe penalty to his dodge roll, which more than likely reduces the action to a single attack roll on the caster’s part. -3 DN.
So, the ritual reduces the DN by 16, bringing it down to 4 (meaning a casting time of 1 round). For the Hex Ficton, which is the setting this particular magic system is geared toward, that seems a little bit too easy for such a powerful spell. It also introduces another issue: while I want spell casters to be able to create spells on the fly, I think that powerful spells like this one should require prep work of some sort--poring through ancient grimoires, searching for esoteric magical materials, buying Urns of Osirus on ebay, that kind of thing. That means I need to come up with some way to keep the ritual elements from lowering the DN on “big” spells too much and set some sort of limit on what kinds of spells can be cast on the fly.
My first thought is to introduce a Magic Level from 0-5 (0=no magic; 3=Hex ficton; 5=standard D&D world) and increase the DN of any spell aspect that’s higher than the magic level. So, for example, if you have a Magic Level of 3 and want to cast an overtly magical spell (DN 5), maybe the DN gets doubled to 10. If you really like math, you can just double the part over the magic level, so in this case you’d have 5 + 2x(5-3)= 9. Other than math-intensiveness, the main problem with this idea is that it ties us to the tables. Remember, I want the players and GMs to get away from the point-by-point DN assignments once they’ve got a feel for the system. So out goes that idea.
Another possibility would be to increase the range of the Magic Level to something like 0-25 and rule that if a spell’s DN is higher than the Magic Level, it can’t be cast. This frees us from the table, but introduces other problems. First off, do we use the base DN or the DN after the ritual modifiers are figured in? Using the base DN would automatically cut out some spells, even in relatively high magic worlds--if the Hex Ficton is a Magic Level 15 world, the resurrection spell that started us down this path is impossible. If we use the adjusted DN, the point is moot, since we’ve already managed to reduce the DN of a resurrection spell to 4, which would be well below the Magic Level of any world where someone would be attempting that kind of magic in the first place.
Yet a third option is to use the caster’s “Wizard” (or whatever) Number as the limiting factor. This limits the available spells based on caster knowledge rather than magic level, but this has the same problem as above if we use the adjusted DN and still makes some magic (anything with a DN over 16, if we’re using standard limits for character Numbers) off-limits to everyone if we use the base DN.
So what if we attack it from the other side: instead of limiting how high the spell DN can go, what if we limit how much a ritual can reduce a spell’s DN? This seems like it might fit what I’m going for a little better. Rather than messing with Magic Levels, let’s keep it simple and use the caster’s Number, so a caster with a “Sorcerer” Job at 12 can only reduce a spell’s DN by 12 using ritual elements. Going back to our death spell, one of the world’s great wizards (Job Number 16) can still get the DN down to 4, but an average mage (Job Number 11) is going to have to roll against a DN 9, which is going to be a tough roll (as it should be).
Limiting a ritual’s effect on the DN by the caster’s Job Number keeps powerful spells from being too easy (except for the most powerful casters), but at this point all spells are still cast “on the fly,” which means that an extremely powerful mage (Job Number of 15 or 16) can still walk up to a dead corpse and raise him from the dead in a single round (since the DN would be 5 or 4 depending on the caster’s Number). This actually brings up a new problem with our casting times, but that’s a fairly small adjustment we can get around to later. For now, we need to come up with a way to delineate what spells a caster can just make up as he goes along and what spells are going to require a little more prep work. Luckily, some of the ideas we’ve already brought up can be used here.
So far, I’ve only determined that I want some spells to require in-game preparation outside of the actual spell casting, so the first step is decide how that breaks down. So far I’ve been assuming that the caster can do any kind of spell he can come up with on the fly. I’ll call those “improvised spells”--spells that the caster can make up on the fly by just applying his knowledge of magic. The next “level” of spells are things that the caster can’t just put together on the spot. These are things the caster can figure out how to do, but only after spending some time working on the problem. The exact nature of the “work” will depend on the caster’s tradition--hermetic mages will search through ancient books of magical lore, druids will commune with nature, shamans will go on vision quests, etc. I’ll call these “crafted spells” (at least until somebody suggests a better name. The last category are spells that can only be cast with the assistance of powerful supernatural objects or entities--this is where your Urns of Osirus come in, as well as demonic bargains, legendary scrolls containing rare rituals, rare astrological events, and other things that make the spell dangerous and difficult to cast. The best name I can come up with for this type of spell is “esoteric.”
My next step is to decide how to determine what spells fit into which categories. Base DN (before ritual modifiers) seems to be the way to go here, so I just need to break it down. Improvised spells are easy--if the base DN is less than or equal to the caster’s Number, he can cast it spontaneously.
For the others, I’m going to go back to the idea of a 0-5 Magic Level for the setting. If a spell’s base DN is higher than the caster’s Number but less than or equal to Caster’s Number + Magic Level, it has to be cast as a crafted spell. If the base DN is higher than Caster’s Number + Magic Level, it’s esoteric, which means the caster’s going to have to complete some kind of story goal (binding a demon, finding an ancient artifact, etc.) in order to cast the spell.
The difference between spontaneous and crafted spells is really one of down time, so I probably need a quick mechanic to determine how much time the caster has to spend researching/vision questing/whatever to work out how to do the spell. For that, I’ll use an extended roll, where the player can add multiple rolls together in order to reach a target number. The target number for the roll is the DN of the spell. Each roll represents 1 hour of research or meditation or whatever. A successful roll adds to the tally, a failed roll means the character wastes that hour. On a Lucky Break, the roll only takes half an hour to complete, but a Bad Break means that two hours are wasted.
Example: Bob the Wizard has as Job Number of 12 and the world’s Magic Level is 2. He wants to cast a spell with a base DN of 13, which means he’s got to hit the books. His first roll is a 6, so he spends another hour studying. Next, he rolls a 14, so he’s up to to hours but still has a running total of 6. The next roll is 20, a Bad Break, which wastes 2 more hours. Finally, Bob rolls a 12 (a Lucky Break), which gets him over the target number of 13 and only takes half an hour. In the end, it takes him 4.5 hours to figure out how to make the spell work.
That leaves esoteric spells. They’re going to have to be researched (or meditated on or whatever) just like a crafted spell, and the mechanics will work using the same extended roll, but for esoteric spells each roll takes 4 hours. Esoteric spells also require that the caster tap into serious mojo by harnessing supernatural forces or engaging in difficult or dangerous magical practices. The details of this are up to the GM based on how high the spell’s base DN is in relation to the caster’s Number and the world’s Magic Level. Examples of things the GM might throw at the characters include:
- The spell requires an unusual and potentially difficult to obtain material component (powdered unicorn horn, a miser’s dying breath, a relic of St. Christopher).
- The spell will only work at a specific time or place (inside a church, on St. Swithin’s Day, when Taurus is ascending in Jupiter).
- The caster or subject must complete a task or pass a test (solve a riddle, defeat a monster in hand-to-hand combat, resist temptation).
- Spell requires a sacrifice (can be a blood sacrifice, sacrifice of material goods, years off the caster’s life, or anything else the GM can dream up).
- The spell requires the assistance of a supernatural entity (demon, fairy, god, ghost). The caster will have to locate or summon the creature and then convince it to help.
One obvious question about crafted and esoteric spells is whether the caster has to do the research every time. Once he’s researched a spell once, does he “learn” it, allowing him to skip the research in the future? I’m going to say no to that. I have practical reasons for this (to avoid having to keep up with every spell the caster has used and to keep the extra hoops for powerful spells in place), but can also back it up philosophically with the words Isaac Bonewitz (from Authentic Thaumaturgy by Steve Jackson Games):
“Furthermore, while magicians are sometimes as lazy as anybody else, and will tend to stick with tried-and-true methods, many will make deliberate alterations in each ritual used (though this is less likely with formal religious ceremonies). These alterations are for two main reasons: firstly, because the universe is changing every millisecond and each ritual must be tailored to the specific situation at hand; secondly, because the better magicians are artists and can’t resist tinkering with their techniques in (not always successful) attempts to produce better results.”
While casters will have to research crafted and esoteric spells anew each time they’re cast, the GM should feel free to give the player bonuses when he’s researching spells that are similar to those he’s used before.
While I’m tweaking things, I want to switch up the casting times a bit. The resurrection spell, when cast by a caster with a 16 Job Number, would take 1 round under the old rules, but the actions described in the ritual alone would take longer than that. So I want to raise casting times on the low end. Also, now that high-DN spells can involve hours of research or meditation or Tantric sex or whatever, it probably wouldn’t hurt to lower the casting times for more powerful spells. For spells with a (final, after-ritual modifiers) DN of less than 10, I’ll make the casting time equal to the DN in rounds (minimum casting time 1 round). If the DN is 11 or more, casting time is 10 minutes + 10 minutes for every point above 10, or 10 + (DN-10)*10. So DN 11 takes 20 minutes, DN 12 takes 30, etc. In addition to changing the casting times, I’m going to specify that the GM may rule that performing certain ritual actions (especially those that require additional rolls, like killing the servant in the resurrection spell) do not count as part of the casting time. For example, if a caster is casting a DN 8 spell whose ritual involves wrestling a bear, the casting time is 8 rounds plus however many rounds it takes him to take down Yogi. (Note to self: Come up with a spell that requires caster to rassle a bear).
Let’s see what these rules do to the resurrection spell (they shouldn’t affect the combat spell, since its DN was so low). This time around, we need to know the caster’s wizardly Number, which I’ll say is 13 and the world’s Magic Level, which I’ll set at 3. Since the base DN is 20, the spell falls into the “esoteric” category, which means that the caster’s going to have to spend some time figuring out how to make the spell work. Once that’s done, he tells the GM his ritual, which we keep the same as above. Even though the ritual is worth 16 points, the caster only benefits up to his Number of 13. That means that the spell’s DN is 7. So the spell will take 7 minutes to cast, plus however long it takes to end Jack (probably a round, since he’s pretty much helpless). Also, since it’s an esoteric spell, the GM needs to come up something to make the spell even more difficult or dangerous. He briefly considers letting the human sacrifice serve for the esoteric requirement, but since the ritual is substituting the slave’s soul for Tudyk’s, the GM decides it doesn’t count. At this point, he could make up some artifact that the character has to find, but that’s a little lame for a spell that raises the dead. Instead, the GM decides to go with an informal version of the “test” idea--when the caster opens the gates to the afterlife to swap Tudyk and Jack, something (maybe hellhounds or evil spirits). The creatures will continue to pursue the caster until he’s dead or they’re sent back to hell.
Now that we’ve tried spells on both ends of the spectrum and tweaked accordingly, let’s try something in the middle: A sleep spell (the standard D&D type, not the Snow White poison apple kind).
- How overtly magical are the spell’s effects? This one kind of depends on the situation. If the intended victims are watching TV, it would fit under “unremarkable coincidence.” If they’re in the middle of combat, it’s going to at least be a DN 2. Assuming we have no way of knowing what the victims will be up to when the spell is cast, we’ll go with the worst case. DN 2.
- How powerful are the effects of the spell? The caster could get the spell effect by slipping the victims some roofies, so we’ll call this DN 2 as well.
- How long does the spell last? The spell will last a number of hours equal to the caster’s Success Degree, so the DN here is 1.
- How far-reaching are the spell’s effects? This is an area of effect type spell (“immediate”), so the DN is 2 again.
- How precise or finely controlled are the effects of the spell? The caster isn’t worried about friendly fire here, so the spell will affect everyone in its path. The spell’s “path” will depend on the form the ritual takes, but for now I’ll assume it’ll be something rather inexact (“sandman dust” or something), for a DN 2.
That gives us a base DN of 9 for this one. Assuming the caster is at least average (Job Number 11), he can cast the spell on the fly and bring the DN all the way down to zero if he’s got a good enough ritual. The caster decides that he’ll focus the power of the spell into a candle that he’ll dowse with wine and rub with Valerian root . After that, he’ll carve the symbol of the god of dreams into the candle. While doing these, he’s going to sing a lullaby he remembers from his childhood. When the candle is lit, anyone who breathes the smoke will be subject to the spell’s effects (this fits the “inexact” definition for the spell’s precision, so the DN of 9 stands).
- Symbolic Elements: Not terrible, but not particularly inspired. I’ll give it a -2.
- Knowledge: The spell isn’t specific to any particular subject, so no modifier here.
- Material Elements: Wine and Valerian root are both things that make people sleepy, so we’ll be generous and give this part a -3.
- Setting: Unspecified. No modifier.
- Requirements: Lighting a candle doesn’t require a roll, so -1 DN for this one.
The grand total for ritual modifiers is -6, for a final DN of 3. That means it’ll take the caster 3 rounds to dowse the candle, rub it with wine, and carve the dream god’s symbols into it (all while singing). The candle can then be lit whenever it’s needed.
This brings up another question: When a spell is focused into an item as described in the sleep spell, how long does the magic hang around? Since the system is meant to handle magic item creation using the same rules as spell casting, and since I don’t want to get stuck doing magical item half-life accounting, I’m going to rule that the magic lasts indefinitely. Of course, as noted earlier, enchanting an item doesn’t make it any more durable or resistant to damage than a normal item of its type. If players start hauling around tons of sleep candles and other general use spells, the GM can always use the forces of entropy (not to mention destructive encounters) to keep them in check. Magic wine can go sour, potion bottles can break, candles can melt or rust, etc.
While we’re talking about magical items, let’s take a look at the difference between focusing a spell into an item (like the candle above) and creating multi-use magical item. In our case, let’s look at three different variations on a magical sword that will kill the king of Regalia.
In the first variation, the magic of the sword is being focused into the sword and will only work once. The spell effect is released the first time the spell draws blood from the target (so the magic isn’t spent if the wielder swings and misses, but if he hits, he needs to make it count) and does additional damage equal to the caster’s DN.
- How overtly magical are the spell’s effects? Killing someone with a sword shouldn’t raise any eyebrows, but even a weak strike could potentially kill the target, so we’ll go with a DN of 1 here.
- How powerful are the effects of the spell? Since there are many ways to kill someone without using magic, the DN here is 2.
- How long does the spell last? While death is permanent, the spell doesn’t guarantee a kill, it just increases the damage of a single attack, which is instantaneous (for a DN of 0).
- How far-reaching are the spell’s effects? This affects a single target, DN 1.
- How precise or finely controlled are the effects of the spell? Since the wielder of the sword will probably have to fight his way through guards, the spell needs to be set up to only work when it hits the king, and for our purposes it’s geared to the title, not the individual. We’ll call that “average” precision for a DN 3.
That gives us a final DN of 7 for the “focused” version of the spell. What if instead we want the spell to work multiple times (just in case the first hit doesn’t do the job) or to permanently hold the enchantment (so it can be used to kill Regalian kings for generations to come)? According to our current rules, the DN for either of those options would still be a 7, which means we need to rethink our rules when it comes to magical items. Fortunately, that’s easy enough:
- If the spell effect focused into an item is only intended for a single use, the normal DN for the spell’s effect is used.
- If a spell focused into an item can be used multiple times, add 3 to the DN. The total number of “charges” is equal to the caster’s Success Degree (roll - DN). The GM may also allow casters to use this mechanic to make “batches” of minor single-use items, like healing potions.
- If the caster wants to permanently enchant the item, increase the DN by 5.
- If the spell’s enchantment can be used multiple times, but is limited by some other factor, use the DN from the appropriate duration table. So if the sword will cause damage every time it hits the king, but the magic wears off at dawn, the GM might increase the DN by 2 (since the evening’s adventure is based around the assassination attempt on the king), for a DN of 9. If the spell will work multiple times but the duration of the enchantment is based the caster’s Success Degree in hours, the DN would be 8 (7 + 1). If the enchantment lasts until the current bloodline is overthrown, the GM would probably add 4 to the DN, for a grand total of 11.
So to give our sword multiple “king killing” charges, the DN would be 10, the DN for permanent enchantment would be 12, and the DN for multiple uses in a limited time period would be somewhere in between. As previously noted, the caster will need to use additional spells if he wants to make a permanently enchanted item durable enough to last; otherwise it will be just as susceptible to damage and decay as a non-magical item.
Now that I’ve actually pseud-playtested the system by coming up with a few spells, I can go back and refine the rules I’ve got so far. I’ll get to work on that and post the final rules here (along with any additional changes) sometime in the near future.