Reductionist Magic 8: Object Blueprints
In one of my previous articles, I have explained my ectoplasmic theory of summoning. Instead of creating an object directly, the spell creates tiny bubbles of ectoplasm, which are assembled into the final object. Introducing this intermediary step allows us to avoid all sorts of side effects.
I’ve described everything I considered necessary to the physics of the process; but one key question remains - what kind of objects are we summoning? Why is summon monster limited to a select list of creatures? How does magic even know what a bear looks like?
In this article I will discuss object blueprints - the patterns imposed on our ectoplasm at the very end of the summoning process to create the final object. However, do not be fooled: the question of blueprints is far more general than just summoning, and will also touch on transfiguration, healing, and even scrying spells. By designing a standard method of handling blueprints, we can improve our thinking about many schools of magic.
As usual, if you are confused about any terms, make sure to check out the glossary, and if you are looking for other posts, check out the blog map.
1. What are blueprints and why do we need them
Let’s start by clarifying my terminology.
First of all, there have been some questions in regards to my previous post. Summoning is when an object is created from nothing. Teleportation is when an object is moved from one place to another, possibly making it seem as if it was created from nothing. Plane travel is a subtype of teleportation when an object also moves across planar boundaries. Transmutation is when an object is transmuted from some locally available matter. Under this terminology, Conjuration spells of the Calling type should be seen as plane travel spells.
Now let’s talk about blueprints. Say that we have a spell that summons a sword. During the creation process the spell has to, somehow, decide on every property of this sword; otherwise, it cannot create it. For example, on a macro level, these would be length, width, type of steel used for the blade, and so on; and on a micro level, it would be everything up to the individual imperfections in the crystal structure. Information contained in the spell which is responsible for the sword coming out one way as opposed to another is what I call a blueprint.
Blueprint, obviously, has to contain all the information necessary to describe the created object. A trivial way to do this would be to write an enormous table, describing where each individual atom has to go; but of course, this isn’t the only way to do this. For example, the spell may simply define macroscopic properties of the steel, such as the alloy composition, and grow the sword as if during the “natural” casting process. Shape of the sword can be defined using mathematical equations, saving an enormous amount of effort compared to describing the position of each individual atom. Alternatively, the spell may simply “copy” an already existing object, which takes the place of a more explicit blueprint.
Some may question wherever there is any point in this - why couldn’t we say that the design of every summonable object is inherent to magic itself, and leave it at that? I wouldn’t be satisfied by that answer. The idea that core laws of magic include the blueprint for a whole horse as some kind of bizarre universal constant feels extremely weird to me - if at all possible, I would prefer to avoid that. Furthermore, summoning spells like Minor Creation can create arbitrary objects as defined by the caster. Even if we were to assume that a universal blueprint exists for every single object you could possibly create, we would still have to ask how the caster’s soul knows which universal blueprint to pick in response to the caster’s thinking; this would be, of course, equivalent to figuring out how to create a blueprint from scratch. Overall, I do not think this question should be left unanswered.
It is pretty obvious how the concept of blueprints applies to the fields of summoning and transfiguration. However, we can go further. Once we have a blueprint, we can imagine someone inverting it: instead of having a process to create an object based on a blueprint, have a process for figuring out wherever a given object could have been created using a given blueprint. Such an inversion is mathematically trivial, and not worth detailing. This allows us to design spells like Locate Object: take a blueprint, go through all the objects in range, and figure out wherever they fit.
This is just one example. If you know where to look, blueprint principles pop up all over the place.
2. Resurrection
Let’s discuss two spells: Resurrection and True Resurrection.
These spells resurrect living creatures regardless of the conditions of their remains, reconstructing their bodies in the process. Resurrection requires a body part, which had to be a part of the creature at the time of death, in order to work. True Resurrection does not even need that, and can work as long as you can uniquely identify the creature.
In order to reconstruct the bodies these spells, naturally, need blueprints. The only question is where they get them.
Those with genetics knowledge may assume that DNA is the source of the reconstruction, but this doesn’t work. Resurrection specifically requires a body part at the time of death, but DNA would not discriminate between times when it is gathered. On top of that, there are many factors that change how your body develops that are non-genetic in nature: for example, hormone and nutrient exposure in the womb, physical training, nutrition and obviously any scarring injuries you receive. Neither spell mentions that it would somehow affect how your body looks: we must therefore assume it is recovered exactly as it looked at the time of death. This makes DNA not a feasible blueprint source.
The most natural blueprint source is, of course, the creature’s soul. We do not need to postulate that the blueprint is directly written on the soul; only that if we have access to the soul, we also have access to the blueprint. All resurrection spells then follow this very natural algorithm:
Uniquely identify the creature that died, either using dead body, body part or other information. If the creature can’t be uniquely identified, the spell either fails immediately or produces weird effects at some later point in the casting.
Fetch their soul; if the soul is currently occupied (because it has been turned into an undead creature or the creature is already alive) or can’t be fetched for another reason (the soul has been destroyed, captured, is refusing to return, or something else), the spell fails. If the soul is in your setting’s version of the afterlife, this would naturally involve plane travel, and should be preventable by the usual methods.
Fetch the body blueprint by using the soul; if this process fails for some reason, the spell can’t be completed and fails too.
Use the blueprint to reconstruct the body; this is either permanent summoning or transmutation, depending on what the spell does.
Slot the soul into the body and shock it awake.
Many spells simply modify this basic algorithm to achieve unique effects. For example, Reincarnation uses a blueprint for a random species instead of the one for your original body. Astral Projection does something very similar to resurrection; only the body is temporary, and is based on a scan of your original body and equipment.
3. Summon Monster and Calling
Having discussed resurrection, we can move on to Summoning. Here, I’d like to focus on the Summon Monster line of spells, as it is instrumental in explaining the core concepts.
At first, when you read the text of the spell, you may conclude that it is simply the reverse of Astral Projection: instead of projecting yourself to another plane, the spell projects some hapless inhabitant of another plane to this one. Depending on interpretation, their original body may or may not be gone. However, I don’t think this quite works.
First of all, let’s review what Summon Monster does. When you cast the spell, you choose a creature from a list; this creature is summoned for 1 round per caster level (i.e. between six seconds and two minutes), and vanishes afterwards. There is one Summon Monster spell variant for every spell level (1-9); higher-level spells can either summon stronger creatures or several weaker ones. Summoned creatures match your alignment if possible, can’t use any teleportation or plane shifting abilities, and can’t cast any spells with expensive material components. Once summoned, they act immediately, and attack your enemies to the best of their abilities.
The central part of the spell is, of course, the summon lists, so let’s look at them in detail. I added a new sheet to the SableGM Grimoire, outlining the properties of the summons I considered notable. Because there are a thousand different guides online regarding the combat potential of Summon Monster, I skipped detailing that area of the spell. Relevant columns of the sheet are as follows:
Level: Level of the Summon Monster spell corresponding to the summons below. “Alt” refers to alternative summoning options.
Summon name: Name of the summon, as well as a link to their entry in d20pfsrd bestiary.
Sapient: Wherever the summon is non-sapient (empty cell), sapient (cross), or only somewhat above animal level of intelligence.
Low-light vision, scent, darkvision: wherever the summon has any of these abilities, with details where necessary. Default range for darkvision is 60 feet.
Blindsense, blindsight, tremorsense: same except for three other vision modes, and up to what range if so, in feet. BS = blindsense, BT = blindsight, TS = tremorsense.
Smite: wherever the summon can smite because it has fiendish or celestial templates.
List of additional abilities: Any additional abilities that I considered notable - ones I think you could use creatively or out of combat, such as spell-like abilities.
Magic/SLAs: SLA stands for Spell-Like Ability. Because these abilities are often very notable, I added a separate column to highlight their presence.
Movement (Land), Flight, Swim, Climb, Burrow: creature speeds with the respective movement modes, in feet per round.
Looking at the summon list and the spell description, we can easily deduce the following:
Most, though not all, creatures are capable of smiting.
A lot of the summons are sapient, with sapience becoming more common at higher levels.
A few summons (Alternative options at levels 7-9) are explicitly younger versions of creatures summonable at a higher level; we can thus surmise that unless explicitly noted, creatures are summoned as healthy adults.
In general, the creature is summoned pretty much exactly as it is in the bestiary.
Spell explicitly notes that the creature attacks your opponents to the best of its ability; this means they immediately know who your opponents are, and are willing to act in your interests.
The key factor that determines wherever a creature will be placed in a higher- or lower- level summon list is their combat ability.
Now let’s see how this would contradict our model of finding a creature from another plane.
Physical ability of the summons is always the same; this is not just a question of age, but also of health, prior injuries, training, and so on. If the spell was summoning a random creature, we would expect our summons to have a pretty wide spread of abilities.
Summoned magical beasts are awake, alert, and ready for combat; they are never summoned asleep.
Summons know who your enemies are: because they act immediately, this means that the spell has to communicate this information to them during the summoning process. Summoning takes only a couple seconds: we thus have to assume that this communication takes the form of directly dumping the knowledge into their mind.
Summons are on your side, regardless of who you are fighting. Given that summons come from all over the moral spectrum, this requires an explanation.
One may assume that the summoning spell includes some kind of mind controlling effect, predisposing the summons to follow your orders. This seems unlikely for two reasons. First of all, several summons are immune to mind control. Secondly, mind control spells almost always allow will saves to break control; no such mechanism exists for orders given to summons.
The summoner can only directly order the summon if they share a common communication method. If the summoning spell already mindrapes the summon and dumps a lot of information into their mind, this would be counterintuitive: why couldn’t it dump more information during the spell duration?
On top of this, we have the following problems related to the list of outsiders we are capable of summoning:
Not all outsiders can be summoned by default. Besides the unique named outsiders, even some fairly common demon species aren’t summonable. If the spell was looking for a random outsider specimen from a given species, we would expect every species to be summonable.
Outsider lists aren’t meaningfully ordered based on weight, size, hit dice, or any other “fundamental”, biological property; instead, they are very obviously sorted based on combat ability. It would be difficult to justify a basic search-and-summon spell doing this.
We can thus see that the Summon Monster spell line must be using a more complicated spell mechanism.
4. Divine Repository
There is a simple solution - instead of the spell fetching the soul from a random outsider, it would grab one from a centralized repository of select souls. I will call this a “Divine Repository”.
There are a myriad of ways for such a repository to be implemented. It could be managed by a singular deity, a collection of deities acting in concert, or even by an enterprising immortal archmage making a profit from the margins of the summoning spell power. Regardless of how it works, this repository is central to the functioning of a very popular line of spells; it is only natural to call it divine.
Introduction of a repository has several advantages. First of all, it neatly explains the progression in combat power as spell level increases. The owner of the repository dispatches summons proportional to the power of the spell, representing either a direct trade (magic power for combat ability) or a proportional responsibility principle (strength is proportional to how accustomed caster is to handling that level of power). Secondly, it significantly simplifies the action of individual summoning spells: instead of being capable of scanning entire planes in seconds in order to find a specific type of creature, spells only have to contact a centralized location. Managers of the divine repository can figure out where to get the souls necessary for the summoning and which creatures to stock up at leisure, as well as how to modify them to be responsive to the commands of the summoner.
Of course, this brings back one of our original questions: why even bother? How is this different from simply leaving the question unanswered, and saying that magic behaves in accordance with this weird list of summons by fiat?
I admit that this design decision is somewhat unsatisfactory, so here are three key reasons why I think it should be made.
First of them is philosophical: I believe that laws of magic should be, at the heart of it, “simple”. This is core to the entire idea of reductionism, and including long lists in the fundamental laws of magic would be going contrary to this philosophy. On the other hand, it would be completely natural for actors within the setting to create a “menu” of diverse and useful summoning options.
Secondly, I believe this makes summoning spells significantly easier to modify or extend. For example, a summoning spell may access two parallel repositories, which would explain why certain NPCs get additional summoning options available to them.
Third, this makes a core part of magic “accessible” to the players: theoretically, they now could set up their own repository, or break into the current ones to permanently remove some summoning option from all the mages in the world. Even if such an option will not see use in 99.9% of campaigns, I do have an aesthetic preference for it being available.
5. The Geistpartikel model
And yet, despite all that, I am still not satisfied. It’s just so ugly to offset most of the complexity of summoning spells into a divine black box. Isn’t there a better option?
I don’t think there is anything we can do in regards to Summon Monster: combat-orientation of the spell simply limits it too much. But I think we can salvage the idea of summoning a “generic version” of some species, which may be useful for other spells, and could be a basis for forming the divine repositories used for Summon Monster.
In order to do so, I will introduce a new model: geistpartikels1 (plural; singular geistpartikel).
Geistpartikels are tiny objects constantly emitted by souls. The only property of theirs that is important to us is information contained within them. The first geistpartikel type we would construct will contain two pieces of info:
Tiny pieces of the blueprint for the soul and the associated body.
A loose hash value for the soul blueprint; this can be seen as an “identity token” for that soul, in that it should be unique for each soul, and change if any part of the soul changes. On top of that, it should be similar between similar souls (e.g. the same species).
There are analogues to geistpartikels in the natural world. For example, pheromones are chemicals emitted by various animals, and one of their uses is kin recognition - ability of animals to recognize others who are closely related to them. Evolutionarily, this is used to avoid incest; similarities with our point 2 should be obvious.
In regards to point 1, we can look at the DNA - as close an analogy as we have to a full blueprint for a human body. DNA can, naturally, be split into pieces, and then reconstructed back into the full strand - this is done during genetic sequencing.
Overall, I think that a mechanism for the production of geistpartikels serving both functions is plausible. We can further postulate that this process suffers random errors - this would be important later on.
After geistpartikels are emitted, they spread throughout the plane via diffusion, same as smells in the air. Because we have complete control over all parameters of our model, I will fix the speed of geistpartikel diffusion at 5 kilometers per second by fiat, so that they can reach the other side of the planet within an hour or two.
Existence of geistpartikels allows us to do several very cool things.
5.1. Generic creature blueprint
First of all, we now have an emergent idea of a “generic” member of the population. Imagine full blueprint as a puzzle that is broken into a thousand pieces, with each piece being a geistpartikel. We have postulated that the emission process is error-prone, so reconstructing the puzzle for any specific creature is not possible - we will have to substitute pieces from other members of the same species.
Let’s put this into numbers. Suppose we wanted geistpartikels from every individual to be readily available everywhere on the plane. Volume of the Earth’s atmosphere is 4.2 billion cubic kilometers, or 4.2e18 cubic meters; if we emit that many geistpartikels every hour, we should, more or less, have reliable updates on the entire plane. Let’s further assume that the amount of geistpartikels necessary to reconstruct the entire soul-body blueprint is twenty four times larger. On average we could then expect every necessary geistpartikel to be emitted once a day.
P.S. I am assuming that planes have about the same size as Earth. This seems to more or less fit the settings I know of.
Let’s say that there is just 1% chance that any given geistpartikel is suitable for a blueprint reconstruction and that we need at least 80% of them to be correct in order for the resulting blueprint to be viable. These statistics could depend on many things - qualities of the spell, error rate of the geistpartikel emittance process, and so on. To reconstruct a specific soul, we would then need to wait for around 150 days - long enough for that soul to deviate from its initial structure, thus preventing us from copying individuals.
But if we look at a population of creatures, then we will only need 150 individuals worldwide to have good chances of performing a reconstruction on any given day. Such a reconstruction would have no traits unique to any particular individual - which is exactly what we want.
Naturally, the numbers here can be adjusted to reach a desired minimum population number.
5.2. Scrying
Second emergent property of geistpartikels is a convenient design of scrying spells.
Scrying is the eponymic scrying spell, but the category is, in my view, much broader. The key ability of the Scrying spell is that it can locate specific creatures across arbitrary distances. It’s thus sensible to group any other spell that would require the same ability - e.g. Sending - into the same group.
Under the locality assumptions I made in one of the previous posts, Scrying would seem to be an impossibility. To locate a creature, you would need to scan the entire plane; and because spells cannot act at a distance, the spell itself would need to be global - such as by growing to the size of the entire plane. But that, of course, would be ridiculous.
Geistpartikels resolve this conundrum. Because their concentration is greater closer to the target creature, and because we already decided that geistpartikels will include an identity token, spells have an easy way to locate creatures by simply flying against the concentration gradient. Such a pathfinding algorithm is entirely local and does not rely on any plane-wide scanning abilities. Of course, the more information you have about the creature, the more precisely you can determine their identity token, and thus the better your spells will be at finding them; this is why the will save for the scrying spell becomes harder.
The only remaining problem is determining when the spell has reached its destination. Scrying starts out as good as it will ever get at locating creatures, and only increases in the duration of the viewing as you grow in levels; it thus seems intuitive that the active part of the spell takes a larger amount of energy compared to the search. If we make this assumption, key design consideration is not triggering the active function (remote viewing) prematurely. To facilitate this, the spell would decay sensor creation until it hits the target soul. This can explain one of the reasons for the required will save: if the target soul can somehow prevent contact until the spell runs out of juice, the pathfinding algorithm will get confused, and report back that it could not find the creature.
Similar principles can be extended to other cases of determining someone's identity, such as magical locks or defenses that only let specific people through.
5.3. Locality of True Names
True names are a concept mostly used for outsiders. The basic idea is that every outsider (or, occasionally, mortal) has a name associated with them, which, if known, gives some "degree of control" over that person. The only explicitly defined effect I know is that it makes summoning that outsider easier.
Of course, to identify a specific outsider, True Names have to be globally unique. Once again we have a problem with locality.
Geistpartikels can solve it. To explain the mechanism, we have to look at the soul gestation process.
To be fully general with respect to how different species reproduce, we can say that at time T1 we have no soul and no true name (e.g. before conception), and at time T2 (e.g. birth) we have a complete soul with a true name. It may seem intuitive to conclude that the True Name is created at the same time as the soul; but this isn’t the best option.
Let us imagine a somewhat more complex timeline. At time T1 soul gestation starts: we have a “proto-soul”. At time T2 the soul grows to the point where True Name could start being established. At time T3 True Name is finalized. Finally, at time T4 gestation is complete.
By separating the gestation events in time we can design a clever mechanism for establishing a True Name:
Between T1 and T2, the soul absorbs geistpartikels from the surrounding environment.
At T2 it starts to generate a True Name, then converts it into a geistpartikel identity token. If the token matches any of the geistpartikels the soul absorbed from the surrounding environment, it changes the True Name and tries again.
Of course, during this process the soul emits geistpartikels as normal. As a result, if two souls happen to be gestating at the same time and happen to stumble on the same True Name, a name collision would be detected as soon as their respective geistpartikels manage to percolate to the other soul’s location.
By the time the True Name is finalized at time T3, the chance that the soul’s True Name is not unique is virtually nil.
Using this mechanism we can achieve both locality and global uniqueness with overwhelming probability.
5.4. Object penetration
Before, I haven’t mentioned how geistpartikels interact with objects. Broadly, we can imagine three valid options: they pass through objects freely, they work like scanning divination spells, or they are blocked by solid objects and bounce off of them. First option would make global synchronization easier; it makes the most sense for True Names. Second option is very limited - according to our radiation model, geistpartikels would be stopped by a kilometer of air - and is mostly only useful for various scanning divination spells. Last model is the one I like most, and which gives you neat features such as automatic detection of a Path of Effect via geistpartikel gradients.
Impenetrability of objects creates an interesting side effect - it implies that scrying spells stop working if your target is encased by a solid. I enjoy these kinds of spell limitations, so I see this as an upside.
Buildings shouldn’t suffice for this purpose, as they have various holes and gaps, through which geistpartikels would leak out. This is especially true of older buildings, constructed before the invention of modern window seals and precision door manufacture. You can consider my pressure calculations in the ectoplasm post for how quickly gasses - which geistpartikels effectively are - can evacuate such places. In general, if the encasing object lets through enough air to allow someone to breathe, then it is also not solid enough to prevent scrying.
This means that in practice, this is no limitation at all, unless your target is taking deliberate and severe steps to counter scrying. They would need an air-tight bunker - with seals made of solid materials, such as clay, as opposed to liquids, such as the air traps within your plumbing. This means they will also need a renewable air supply, assuming they are planning to remain hidden for a prolonged period of time. They will not be able to travel, unless their travel vehicle includes the same sort of isolation chamber - making it heavy and bulky. Obviously, all of this will make it very difficult to operate any kind of enterprise or a political scheme.
If we imagine that instead of being completely blocked by solid objects, geistpartikels can penetrate through them to the same extent as scanning divination spells, then countering scrying becomes even harder.
5.5. Property ownership
Finally, let’s tackle the magical concept of ownership. Magic often assumes that an item “belongs” to someone: for example, possessions can be used to target a scrying spell. A viable model of ownership can be designed by assuming that when the soul attempts to create a geistpartikel inside of a solid object, that geistpartikel becomes “trapped” in that exact location.
If we further assume that the soul somewhat extends out of its corresponding body, then it would obviously leave geistpartikels trapped within anything it touches or handles. The longer this goes on, the more geistpartikels get trapped; eventually their amount is large enough that they can be used for scrying. At that point, I think it is sensible to talk about the object being “owned” by that person.
To simplify things, we can define a minimum amount of handling time required to “own” an object. Of course this time can be shortened or lengthened depending on the circumstances at hand. 240 hours of handling seems like a reasonable basis for me: this is equivalent to using something for 8 hours a day for a month, or for 1 hour a day over the course of 8 months.
If a new person handles the object at a later point in time, then we have two simple possibilities: either they become the new owner after the same base amount of time (i.e. age of geistpartikels is key, later owners supercede earlier owners), or they become the new owner after their handling time supersedes the earlier owner (i.e. sheer amount of geistpartikels is key, longer owners supercede shorter owners). There is also a middle ground where a combination of factors is important.
If no unique owner can be established (e.g. two or more people handle the object over the same timeframe, such that none of them has 240 more handling hours than the others), then the object is owned communally; depending on the spell, it may be considered to be owned by either all of them or by nobody.
By playing with the handling parameters, we can define a diverse range of ownership definitions for various purposes.
Geistpartikel definition of ownership has one massive advantage over the alternatives: it replaces messy and often illogical pseudo-legal questions with a clear test based on physical proximity. Here is a sample of the sorts of questions it resolves:
Why are spells better targeted by body parts (e.g. nails) than by possessions? Because the soul covers the body much tighter, and thus the concentration of shed geistpartikels is higher. Any other object that would be covered to the same extent (such as a soul gem that housed someone’s soul for a long time) would be equally as good. Garments, due to being close to the body, also quickly reach the point of saturation.
What counts as a “possession”? For example, would a brick from someone’s house count, since they own it? No, because there was no physical proximity. It would also be no for most items in their house, since they are not handled very often; mostly, you are looking at clothing, tools, emotionally significant items, and their bed, since they sleep on it for 8 hours a day.
Could you do legal tricks, such as drafting up a gift certificate with your target as a recipient, in order to use the gifted item as part of a scrying spell? No, because legal reality has no connection to magical ownership.
Overall, I think geistpartikels are a very powerful model, capable of handling many diverse and seemingly unconnected magical effects.
6. Inanimate objects
Similarly to summoned creatures, many spells require blueprints of inanimate objects to function. Here, I can see several source possibilities:
Divine repository model. Obviously, this works for anything, but is best suited for spells narratively tied to some obvious source.
Direct construction by the caster. Caster could manually define the shape and material of the object. This sort of method is easy to understand, but quickly runs into knowledge problems: pretty much nobody understands enough material science for this. Even something as simple as a brick of steel - a homogeneous material with clearly defined atomic composition - contains a lot of complexity that most people are simply not aware of. If we assume that magic at least “understands” different materials - more on this in the Collective Unconscious section - we could use this method for blueprint construction within the bounds of the caster’s engineering competency, but anything complex is a no go.
Copying a specific object. Another simple method: magic would simply copy an item indicated by the caster. There are two obvious questions raised by this method: first of all, where is the original item, and secondly, what prevents the existence of an alternative spell, which would copy a different - but much more useful - item.
For example, take Sturdy Shelter: spell which constructs a temporary cottage. It would be natural to assume that the spell simply copies some nearby house; the spell, in fact, even says that the cottage type depends on the area. But the existence of such a spell would imply other copying spells: ones that can copy documents, books, seals, keys, clothing your target is wearing, and so on and so forth. In general, the copying toolset is very versatile, which may or may not be what the Game Master wants.
Locating the original item would also seem challenging, unless the caster provides it themselves. For example, where is the original chariot summoned by Phantom Chariot? If it is simply the closest one to the location where the spell was cast, we may run into problems if the spell is used in an area where scrying doesn’t work.
Overall, all of these methods are unsatisfactory. If we cast a spell that summons a chair, we don’t particularly want to deal with drawing up an STL model of the chair. Instead, we want the magic to understand the overall non-specific concept of “chair”, and summon an example of that. When it came to creatures, the geistpartikel concept allowed us to solve this problem; but how could we have geistpartikels for objects, which have no souls, and whose definitions depend on culture and language?
7. Collective Unconscious
The only way to make magic understand a cultural concept is to make the residents of our world a part of the definition system.
To describe the set of all blueprint-related concepts understood by magic, I will be using the term Collective Unconscious (later CU; no connection to any German philosophers). Before we can construct the mechanics of CU, we have to make some crucial assumptions. Let's begin with a mental experiment.
When I say the word “chair”, what do you think of? Maybe you imagine some specific chair, or maybe you look at the chair in your room. Regardless, you have some idea in your head of what I mean by this term.
First assumption I will make is that in your mind, there is something we can call the “concept” of a chair; and that this concept can be applied to various objects, and classify them as more chair-like or less chair-like. I will further assume that this concept is at least mostly consistent: it doesn’t change randomly based on time of day or your mood.
I don’t know whether this assumption is reasonable. I don’t know wherever people have, in their heads, some piece of mental tooling that lets them consistently classify objects into groups - it could be that human reasoning works in a completely different way.
For the purpose of illustration, we can imagine these concepts as shapes on a plane. Every dot on the plane is an object, and concepts group them into areas. More precise concepts are smaller areas, with sharp edges; vague, broad concepts are appropriately large in size, and their edges aren’t as stark.
Second assumption I will make is that these concept-areas are very similar between different people and species. This isn’t quite saying that everyone thinks in the same way: there is much more to your mind than just how you classify concepts. But it is, still, a pretty severe restriction.
With these assumptions, construction of the CU system is trivial. First, we rule that whenever any person is thinking of any concept, their soul emits geistpartikels containing that concept boundary, as well as that of related concepts. For example, say that you are thinking of a sword in your hand: your soul would thus emit geistpartikels that would describe your general “sword” concept, as well as the visual appearance, shape, heft, balance, and texture of this specific sword, and other related concepts - such as “weapon” or “sharp”.
Concepts emitted by individual people are, of course, going to be different. However, in large populations patterns will start to form as commonalities will aggregate together. Words have linguistic meaning due to diverse individuals ascribing similar meaning to them, and concepts within CU have meaning due to the statistically significant similarities in the concept boundaries of individuals. By assigning geistpartikels into groups based on similarity, we create concepts recognized by magic, once the group exceeds some minimum size.
7.1. Collective Unconscious and blueprints
Of course, concepts are not by themselves sufficient to design blueprints. Ten million people can agree on the concept of a human, but that will not let them reconstruct a single mitochondrion within their own cells.
To deal with this problem, we need to introduce some sources of data about the various materials, microscopic properties and designs of various items into the magic system. Easiest way to do this is via “item scans” - use the third blueprint creation method from section 6. A very natural source for these scans are magic items due to their intrinsic link to magic, though other sources - such as certain divination spells - might also make sense.
To transmit the item scan information we will use the same geistpartikel model. Let’s use magic item creation as an example of the item scan source. Geistpartikels can be emitted either continuously (also useful when it comes to describing how Detect Magic works) or at a singular point in the creation process. Magic can then link the detailed scans of the items to the concepts relating to those same magic items, and from there to any other concept that links to those concepts. For example, a wooden staff doesn’t just provide data on wooden staves, but also on any other thing made out of wood.
Once we have the item data, blueprint creation is a simple algorithm.
Look into the mind of the caster, and grab the concept they are using to think about their desired creation.
Fetch the corresponding concept geistpartikels.
Find the concept cluster that fits this concept best
Find the item scans from this cluster
Use the two key pieces of information - concept cluster and the item scans - to construct a blueprint representative to the cluster.
Use the blueprint for whatever you need it.
Item scans provide us with material properties, while the concept cluster provides us with the shape and distribution of those materials. Together, they make up for each other’s weaknesses, and allow us to design versatile blueprints.
7.2. Language spells
In addition to the questions of item definition, CU allows us to break another hard nut of spellcrafting: language spells.
This mostly has to do with Comprehend Languages and Tongues. These spells allow you to understand any spoken language, with somewhat different limitations. Of course, because languages are a social construct, it would seem very difficult to create reductionist spells that would deal with them.
Not so with the Collective Unconsciousness. In fact, dealing with languages requires only a very slight modification to the already established rules: we simply have to add the spoken and/or written descriptions of those concepts to the emitted geistpartkel. Language comprehension will then simply involve taking the spoken sounds, plugging them into a metaphorical database of all concepts that have words associated with them in any language, and getting an appropriately translated selection of concepts beamed directly into your brain. Everything else is, largely, an implementation detail. Speaking is the same process but in reverse, and with an illusory element that would turn your mental comprehension into sounds.
By defining language spells in this way, we can draw a clear distinction based on the number of active speakers between proper languages and various secret codes, organization lingos, cants, and other proto-languages which, as a Game Master, you may not want to be translatable.
8. Pressure and temperature
Another question CU resolves has to do with pressure and temperature.
For example, take a spell like Major Creation. Say that you attempted to create a piece of metal under 1E30 atmospheres of pressure. This would still be a piece of metal - the atoms haven’t changed, after all - it would simply be compressed to a ridiculous extent.
There is a common misconception that solids are incompressible, but that isn’t the case. They are compressible; it’s just that you need pressure higher than anything you can find in normal life to get any visible compression going. Of course, if you do manage to significantly compress a solid, the amount of stored energy is truly stupendous.
Major Creation simply defines the material, but nothing else about the created object. It would be hard - at least, I don’t see any obvious way of doing it - to prevent the creation of objects under tension or compression. Pretty much everything you ever see in life is under some amount of tension or compression, almost always both at the same time in different places; some objects, in fact, cannot exist at all without it. For example, bolted or nailed connections rely on compression to stay together. Outlawing compression as a concept is thus not really viable, and limiting it to a certain maximum amount seems fairly arbitrary.
But if you don’t outlaw compression, then someone will create a brick of steel under 1E30 atmospheres of pressure, which will promptly decompress and vaporize a city. How do we square this circle?
CU provides at least some natural help in this regard. Because it defines objects based on the collective understanding of large groups of people, various edge case ideas - extremums of energy, pressure, force or density, exotic materials, and so on - are avoided out of the box. Most people would not recognize those extremes, and so the concept wouldn’t either. Even if they did, the underlying data used to construct blueprints comes from existent physical items, and so it would be natural for the spell to fail when presented with a data point well outside of its training area.
Of course, this leaves a potential exploit of culture building. If literacy and science within your world reaches a point where most people will start to think in explicit numeric terms, then the concept-space can change to accommodate things like superheated plasma. This process would go faster if new terms are invented for these unusual cases. If science reaches a point where creating a magical item under extreme pressure becomes possible, then it will become a part of the item scan geistpartikels; this will provide magic with the necessary data required to reconstruct other item blueprints in this pressure regime.
If this is a serious concern in your world, then you may have to do more work to correct it.
CU doesn’t just solve the question of the extremes of temperature; it also helps us decide on what is the default temperature at which items are created. Because the blueprint is based on the concept, this means that items will be summoned at whatever is the most stereotypical temperature for that item. This means you could use conjuration spells as climate control by summoning items at an appropriate temperature. Magic already provides a lot of options for this, so it is going to remain a niche option.
Overall, I think that the concepts of blueprints, geistpartikels and the collective unconscious can be useful in the design of a broad range of spells, and in thinking about their limitations.
If you are looking for other posts on my blog, check out this list of all other posts, and if you enjoy what I write, you can subscribe to receive updates by email:
Article art is based on "Ghost Lake - by Dita" by Dita *is Catching Up* Actor, licensed under CC BY 2.0, as well as various blueprint drawings by Aezzy, and is itself licensed under CC BY 2.0.
Why Geistpartikel?
Because coming up with a good term that refers to “small particles emitted by a soul, primarily used to deliver information and track down the emitting soul” is very hard. My other option was “soulromone” as a portmanteau of soul and pheromone, but it sounds stupid. Geistpartikel is from German, with Geist meaning soul and Partikel meaning particles; it’s not correct grammatically - that would be Geisterpartikel - but the correct version is longer and thus worse. At least it sounds better to my ear.
A good read! I've always liked a divine repository like model but I must say the ghost particle idea is cool