a well-known bad way to write about nonhuman characters is using mere synonyms. this is what it meant by accusing someone of “calling a rabbit a smeerp.” the writer uses a weird alien term for a familiar concept as a way of evoking its ostensible alienness. but nothing about how they use the concept is alien. it, to speak in oft-trod writing jargon, is .
another unsatisfactory method is a mainstay of bad cosmic horror, where you obliquely conjure an alien notion and at all costs refuse to explain its interiority, letting that pretense of impenetrable mystery substitute for a sense of the alien. this is faux ineffability. it’s not much less lazy, and it is of course unworkable for longer stories where the creatures take center stage, and eventually the true emptiness of the mystery will reveal itself simply by long enough exposure to it.
the remedy we all eventually settle on, once we spend any time thinking about the problem, involves a subtle fission and fusion of known concepts, a kind of interpretative alchemy. i think i will call this conceptual interpolation. a basic way of accomplishing this is by double analogy. let’s say a nonhuman species has a concept of rafmal, and we note that for one, the thrill of riding in a fast vehicles like chariots or ships is rafmal, but also the comfort of residing in a house.
at once it has become rather difficult to translate this into plain language, it already feels appropriately alien. this is what i meant, , where i claimed that xenofiction should be an exercise in conlanging. because while what we’ve accomplished right here involves just enough thought that it stands head and shoulders above so many attempts at writing nonhumans, we have still only really succeeded in making them about as alien as a foreign language speaker from distance culture.
conceptual interpolation straddles the line of unfamiliarity without unintelligibility, and it accomplishes this simply by redrawing the borders on the map. much like how different languages disagree on where (or if) the difference between “green” and “blue” exists, nonhumans can disagree on the difference between anger and sadness, family and friends, pet and stranger. they have nine difference concepts for “meal,” according to the ritual and company and contents.
but ultimately, this is nothing you can’t get from anthropology (in fact, it’s often less.) there is a shared underlying ontology; it’s the same legos in a different structure. is this the best we can do for xenofiction?
one advanced technique is iterative anti-induction. there’s a famous douglas adams quote:
There is a theory which states that if ever anyone discovers exactly what the Universe is for and why it is here, it will instantly disappear and be replaced by something even more bizarre and inexplicable.
There is another theory which states that this has already happened
if rafmal refers to both a positive feeling of being in a vehicle and a positive feeling of being in a house, well, the phrasing here already reveals the inference. can we not simply translate this concept as “well-present”? “aptly-confined”?
so let’s get it twisted. what if i clarified when the vehicle is not moving, it ceases to be rafmal, and likewise when the house has all of its doors closed. even a building with an open door is not quite rafmal if there’s only one way in and out.
the principle of construction here is easy enough to grasp, if not the actual results. first, i do some conceptual interpolation to get a difficult-to-translate concept. then, once you manage to parse through that to get a sense of what the underlying concept must be, i pull the rug out from under you. i say sike! it’s not that simple!
consider: a conversation between three or more friends is also rafmal. (assuming you’re e.g. in a house with multiple open doors.) not so if it’s just one on one, though if you only by yourself, you’re good and rafmal again. oh! but if you’re delivering a speech or performance to a group, again not rafmal.
one of the issues i’d pointed out with bad cosmic horror’s faux ineffability is that if you keep gesturing at something that can’t be conveyed in words, eventually you’ll have given enough hints for the reader to circumscribe what’s behind the curtain. you’ll unavoidably convey a vague sense of its dimensions.
but that’s not quite how it works, not every time. if you throw out random hints without a care for what it conveys — or with the deliberate intention to foil the reader making sense of it — then at a certain point rather than cohering, one can only conclude that it’s just a random assemblage of concepts and rules. the monster is whatever the author thinks is spooky.
of course, at this point i must reveal that i did have guiding principle behind what rafmal actually ought to mean. i think the idea becomes all the more clear if i propose that the species this concept originates from is a kind of prey animal with an acute sense of claustrophobia.
thus they find it nice when buildings have multiple open doors because it means you have escape routes, likewise it’s nice when you’re being carried somewhere. if three people are talking, one of them can slip away without stopping the conversation in its tracks, but that’s not the case if you’re delivering a speech.
this ties into one of my broader theses: a key to portraying nonhumans resides in giving them different needs and drives. so much of the vaunted alienness we’re pursuing will arise from applying these axioms.
but it also ties back into my assertion that xenofiction is conlanging. because part of what gives any language its weirdest and most specific words is simply history. this one has a chequered etymology, that one is an allusion to a particular story.
ultimately, the root of this exercise is the formation of concepts themselves, and tautologically, this happens when you perceive several different things and psychologically attribute some important commonality to them. you match patterns, and thus the most elegant way to make a creature alien is to start by making a fundamentally different patterns salient and important to them.
the concept of rafmal exists because of a innate wish to always be able to get the hell away in a pinch. add a new sense, a new hang-up, and you have a fruitful new concept.
but while typing all of this out, i thought of a way to really get weird with it.
what happens if you subvert the psychology of pattern-matching on its most foundational level? what if you had aliens that fundamentally do not understand categories, literalism, or monosemy? here’s my vision.
let’s say i have a concept of a kind of geometric whorl you can sculpt onto clay figures or carve into woodwork or whatever. i might show you a couple different objects that i’ve designed to exemplify this whorl. i point to a triangle with a spiral, and i call it pulalen. i point to some colorful yarn sown into a cloth, each thread curving at a certain angle, and again, it’s pulalen. i take out a paper and pencil, and draw the most bare example of the form. pulalen!
and maybe to test your understanding, you draw a circle. not pulalen, sorry. an oval, an archimedean spiral, still not it. this is not a word for any curve. so you point to one of my other designs, ornate and curvy work, but it lacks the characteristic pulalen spiral, which confirms for you that this is not a word for any ornate curve.
at this point, we’re probably both satisfied that we have a similar understanding of the concept.
such is an extrinsic definition. if i were mathematically inclined, maybe i could write out an equation defining exactly what shape it is, or failing that, at least a vague verbal description. this would be an intrinsic definition.
and both of these would be woefully inadequate for the kind of alien i’m imagining.
this alien would ask why you aren’t talking about what it is. what color is it? does it taste sweet, savory, or sour? does it vote liberal or conservative?
at this response, we would both look at this thing like something had gotten utterly lost in translation. it’s a curve, those questions are completely invalid.
the alien would be do a little better understanding me if i handed it that clay sculpture, let it lick the surface, or tap it to hear how it resounds, though it might still be a little perplexed that i could consider it ‘the same’ as the cloth with sown threads, yet i can’t describe its taste or whether it would rather swim in the sea or fly.
okay, let me finally explain how this thing really thinks. see, when we’re defining categories, we naturally try to group like things. objects of the same color, or same material. there’s a kind of purity to how we naturally draw category lines. this is the collection of pink things, this is the collection of clay figures.
it’s more confusing to us when these things are different. a bunch of pink balls is easy. if it’s a bunch of toys, paintings, raw meat, gemstones, etc. it would take a bit before we decide okay, everything here is pink on some part somewhere.
to put it in the most simple terms, if i show you X, Y, Z, and tell you they form a category, then the most natural conclusion is that there’s a single property they have in common, though it might be oblique or indirect.
and this is where you differ from the alien. they would find it completely unnatural to make that sort of inference. one can probably come to understand this is how humans think, and carefully imitate the symbolic manipulation of human!concepts, but to them it would be mechanical and unintuitive.
in order for the alien to decide a group of things form a natural group, they must be different.
the alien cannot conceive of a single “pinkness” referring solely to the visual perception. there must be pink songs, or pink days of the year, or a pink part of a speech. it’s a kind of axiomatic synesthetia.
one of the issues you run into, when creating nonhuman psychologies, is that humans think the way they do because it… makes sense. that’s how the world works. if you want to cut things, you’re going to come up with a tool that has a blade, and so many of the ways you’d go about designing a “alien knife” will just make for a worse knife. the concept of cause and effect, self and world, these are very difficult things to believe an alien would not understand, because each is a corollary of existing in a universe anything like ours.
and you might think what i’ve just described is an example of this pitfall. what sort of intelligent species is incapable of matching basic patterns? would they not be immediately outcompeted by anyone capable of noticing “all hot things burn when touched”?
but it’s not as ridiculous as it sounds. in fact, it’s arguably more accurate to reality. after all, think about the real world, devoid of your abstractions if you can (you can’t, but). no object has “qualities”. a thing that’s colored pink is also made of a material with a certain texture. it tastes a certain way, it reflects sound a certain way, it’s entangled with the world in innumerable aspects.
here’s a simple way to think about it. when we write, we often invent new terms specifically so we can refer to specific things in a precise way. each new term is meant to have a single unambiguous meaning
this alien would never operate that way. if it were to coin a new concept, as a matter of course it would give it multiple loosely-related and orthogonal definitions, simply to furnish it with intelligibility.
i mentioned it at the start: they cannot understand literalism. everything has a figurative axis. every conversation is actually about something else. the subtext is the text.
in that way, this is not unlike an idea that has piqued me for over a decade, ever since i was a teenager. it’s a concept of a complex, “multi-level” language that requires a higher, or at least different, kind of intelligence to ever fluently speak, because each and every sentence you compose is a dense system of equations that computes multiple meanings, rather than the singular parses of our every day speech.
but even as i’ve elaborated this utterly alien mode of thought, i’ve committed a sin of oversimplification. after all… how would it ever have a concept of a “word” as a distinct unit? why would such a thing be assigned an intrinsic meaning?
as i finally bring this blog post to a conclusion, i realize i have re-derived from first principles the concept of an unglossable conlang