I was in a few philosophical discussions this summer that involved disagreements about what it means for something to exist. Engaging in such can be a bad move, because it’s one of those words that actively resist clarification and actually get more confusing when you try to get non-confused about them. They’re the non-newtonian fluids of words.
Not wanting to get lost again in exotic, unmapped territories where ordinary communication stops working, I’m writing this to be able to refer to it if need be instead of having to improvise on such a slippery topic again. And slippery it is. This article has taken some time to finish because I’ve had to rewrite it several times as I kept getting myself confused.
In summary: the word “exist” is a classic case of trying to get more consistency, rigor and depth out of words than they can provide. Statements like “X exists” (or “X is real”, which works almost exactly the same) don’t have a single, context-free meaning. They’re shorthand, and we should focus on understanding what they’re shorthand for.
Prelude: Existing people
I’ll start with something easy: just think about what it means to say that some person exists. An obvious example, a real person whose existence everybody can’t help knowing about even if they try not to, is Donald Trump. He exists, obviously.
What about Napoleon? He died two hundred years ago. Does he exist? Not now. Not in the same sense Donald Trump does. But we wouldn’t say that he isn’t a real person or that he’s nonexistent or imaginary (common antonyms to “existing”). He isn’t made up, like, say, Sherlock Holmes is. Just like it makes sense to say that Donald Trump exists a way Napoleon doesn’t, it makes sense to say that Napoleon exists a way Sherlock Holmes doesn’t.
Sherlock Holmes clearly doesn’t exist, though. He’s fictional. That means doesn’t exist, right? I mean, that is what it means.
Not so fast. If someone asked “is there an existing character we can use for our new 19th century London-based crime story?” You could say “Yes, what about Sherlock Holmes?”.
Fictional characters do exist, as in, there’s a commonly used meaning of “exist” that applies to fictional characters and is used to differentiate between those that have been created and those that haven’t. Holmes is a real fictional character, unlike Archibald Dundermunch, abrasive crime fighter with a slight drinking problem and a complicated relationship to his ex-wife, that I just made up.
I could go further. Dundermunch has actually been made up, so he exists in a way purely hypothetical characters don’t. He’s a formed idea, like Russell’s Teapot or The Chinese Room. An existing idea of a character, but not an existing character.
I can conceive of even more generous senses of existence, like existing as a “possible person” etc. but there is no need for things to get even more convoluted. My point is that we use words in partial, extended and metaphorical senses that eventually become multiple standard meanings. We aren’t going to clear this mess up and find out that this patchwork of meanings is some low-level projection of a coherent truth in some higher-dimensional plane we can reach for. What is existence, man? That’s not how this works.
The Big List
Ok, enough game-playing. Regardless of whether historical figures, fictional characters or imagined fictional characters exist in some sense, we did agree that Donald Trump really exists, and we feel this is different. He exists not in some extended or metaphorical way like I was playing around with before, but literally. He belongs in the Big List of Things That Currently Actually-Truly-No-Shit Exists For Real.
Deep, in our *hearts*, we feel there ought to be something like such a list. Things come into existence, exist, and then stop existing. At each point in time there is therefore a complete list of existing things. This is the same thing as saying that “X exists” is true or false for any value of X.
Donald Trump is such an existing thing. He does change a little bit all the time, but that doesn’t make us think that he disappears and comes back as somebody or something else, not literally, anyway. He has a persistent identity, some fundamental essence that stays from the moment he starts existing to the moment he’s taken off the list.
But the list is just a model, and these persistent essences aren’t something found in the world. We intuitively believe, alieve, in them because it’s efficient to model things that way and we have a hard time separating model from reality — if we act as if something is the case long enough, we’ll begin to believe it.
If this feels like hair splitting, it’s because our intuitions about humans still work pretty well despite theoretical problems. Humans have relatively concrete physical boundaries and change slowly and incrementally enough for us to comfortably model them as a single, persisting thing.
The same can’t be said for other kinds of objects. Then we feel less sure about what “existence” means.
A good old classic example: heaps. Do heaps exist? If you take a fistful of candies (do “fistfuls” exist?) and arrange them into a heap, has a new thing come into existence? Does the Big List have a new entry? Like when a baby is born?
What if you define a new kind of object? I arrange exactly 13 candies in a row and then proclaim that this is called a “threap” from now on. Does it earn a place on the list when I arrange them or when I create the definition?
What if we also define “tweap” as “heap with an even number of candies”, and keep adding? Does it blink in and out of existence each time we add another piece? Is it the same tweap coming and going or a new one every time? What’s the total number of tweaps existing throughout history?
Somehow it feels wrong that we can increase or decrease the number of things in the universe just by rearranging stuff and making up new categories. It’s at odds with at least part of what it feels like for something to exist. The part of my mind that says there should be a Big List also thinks that list should have a set number of entities and only change when things *actually* come into existence or are destroyed.
Part of the problem is that both kinds and things are partially or wholly overlapping: tweap is part of heap, human is part of mammal; a steering wheel is part of a car, a finger is part of a hand. But the Big List is supposed to only have one entry for each object.
Taking this intuition seriously leads to some strange consequences.
The first is that a thing can only be on the list as a member of a single category, its true kind. Other categories are abstracted and constructed — not real. “Human” feels like a real category and “tweap” doesn’t, so we feel humans have essences while tweaps don’t, and inbetweeners like regular heaps, characters, countries or chairs confuse us.
The second is that if some objects are part of a larger object, then the larger object or its parts can be on the list, but not both.
Our list-intuitions gets us to a weird place where things only truly exist if they have essences: if they are indivisible and fundamental; they do not consist of smaller parts and they are not generalized forms of more narrow categories. In this strict sense, Donald Trump — an abstraction of an everchanging composite structure — does not exist. Nor does anbody else or any ordinary thing.
Looking for the fundamental building blocks of reality eventually led to particle physics (i.e. it was a lot more difficult than we thought), and according to that our best guess of what the List actually looks like is row after row of superstrings or whatever hot new thing physicists currently think is down there at the bottom of it all.
Particle physics might be able to support a coherent conception of the Big List and therefore what “existence” means, but in getting there we’ve completely lost track of how the word is actually used.
The maps and the territory
In real life we don’t use one Big Complete List but many Smaller Partial Lists. An object being represented in some list (i.e. existing) means different things depending on the list — remember that Sherlock Holmes exists as a fictional character but not as a living person. He’s an item in some lists and not in others, and his existence depends on exactly what list we’re referring to.
Maybe lists aren’t the best metaphor. Certainly our models of reality are more complicated than that. Things just don’t hang around all alone and unstructured in lists, they have relationships to each other, consists of each other and are abstractions of each other. What we have in our heads is more like maps than sets of lists, maps of objects with properties and relationships to other objects, and the relationships are themselves objects with properties, and so on.
Database systems are philosophically interesting because they’re built according to how we view the world and not according to what it’s actually like. They make sense to us, more sense than the real world does. That’s why we build them this way.
Like our belief systems, a database design is typically supposed to model some part of reality. In fact, good belief formation is like a good database design: it’s all about building a model that captures reality well enough to be useful while being simple enough to understand and work with.
Unlike database systems, humans have multiple maps describing the same reality and they are neither consistent nor explicit. They’re complex and arbitrarily abstract. Everything under the sun can be represented by a concept token in a system of relationships. Objects, sure, and their properties. Categories, yes. Relationships themselves, too. But also even higher-level stuff like processes, structures, patterns and agents, including composite abstractions of composite abstractions like properties of processes by which patterns of relationships between structures change.
We pack these philosophical nightmares into chunks, slap a token onto them and treat them like single things.
It’s an extraordinarily powerful system.
But what the hell does it mean to say that something represented by such a token exists? Does it even make sense to say that the token represents something, some thing? If we try to learn from what we got when we looked at people and fictional characters, we realize that it really really depends.
In other words, existence depends on context. That’s a phrase destined to be misunderstood. I was once mistaken for a radical relativist when saying something like that. I was told:
What exists or not is dependent on context? What kind of relativist bullshit is that? This is what makes people look down on philosophy. No, there is an objective reality.
Yes, of course there is. It’s just not made of words and not structured the way our mental representations of it is. Whether something exists depends on context not because reality changes with context but because the meaning of “exist” does and therefore the truth value of the statement “X exists”.
Consider what it means to say that these things do/don’t exist or are/aren’t real (really do consider, they’re all very interesting questions):
a solution to global warming
infinitely many prime numbers
genders other than male and female
the color purple
the word ‘cromulent’
the word ‘irregardless’
the War on Christmas
a chance of rain
a meaning of life
the meaning of life
On a high enough abstraction level, asserting that these things exist has a clear single meaning: “this thing should be represented by a token in our map of reality”.
However, these existence claims compile into real world machine-code in different ways. When virtually everything — concrete or abstract, simple or complex, physical or structural — can be represented as object tokens, then the relationship between model and reality, between token and thing, symbol and referent, can be drastically different (not to mention ill-defined).
Because we represent drastically different things with the same kind of symbolic tokens, we can use a single vocabulary structure to say drastically different things while having it sound like we’re saying the same kind of thing.
Even a single statement can be compiled in different ways. Note that the meaning “this thing should be represented in our map” can be rephrased into “a map representing this thing fits reality better than a map not representing this thing”. Its a claim of fit between reality and a map.
This fit goes both ways. It has two degrees of freedom. You’re making two kinds of claim at once, coexisting in a kind of superposition. We can see this by holding one degree of freedom fixed at a time:
If we hold the map fixed, we make a claim about reality. In the Kurdistan example, we can take the local definition of existence as a given (“is a recognized state with functional sovereignity”) and say that empirically, Kurdistan is not that. It’s not in the list of such things.
If we on the other hand hold reality fixed, we make a claim about maps. We take concrete facts about Kurdistan as a given and say that it exists. By doing that we insist that “exist” in this context (to be a country) should mean something that includes Kurdistan, like “being a geographical area dominated by an ethnic group with common culture and identity that we think deserves self-determination”.
Normally, we don’t specify what end we’re holding fixed and in many cases we don’t even know (hence “superposition”). If you say “Kurdistan doesn’t exist” in normal conversation you are both saying something about the properties of Kurdistan and what you think it should mean for a country to be considered existing (in this case, some way that doesn’t include Kurdistan).
In the same way, “free will exists” (to pick up a discussion in the comments of an earlier post) isn’t just a statement about how our wills work, it’s also an implicit endorsement of a particular conception of “free will”, specifically one that makes its existence statement come out true.
Saying that other genders than male and female do or don’t exist is another prime example: they’re about how people’s psychologies actually work, sure, but at least as much about how the concept of gender ought to be defined and used.
The others on the list are left as an exercise for the reader.
Note that “superposition” is not unique to existence statements. It’s a feature of almost any claim made in natural language, which further illustrates my point that “exist” is not a special word with a special defintion allowing it to be part of a special kind of claim. The bolded paragraph above is especially relevant to a whole family of highly general, abstract, for-map-use-only words that we struggle to define in terms other than themselves. They take their concrete meaning from their context and tend to disintegrate if put in a microscope for rigorous study (resulting in centuries of philosophical epicyclemancy). Members of this family include: good, bad, true, false, right, wrong, different, same, possible, impossible, more, less, object, subject, being, fair, valid, inherent, necessary, explain, power, cause, real and free.
If we can’t rephrase and “compile” an existence statement, something is amiss and its truth may not be a meaningful question. Does consciousness exist? Rephrase it in concrete terms, please. Do universals, numbers or possible worlds exist? Rephrase it in concrete terms, please.
“Exist” is a tool for making shorthand that can be expanded if need be. It’s a handy way to talk, and that’s pretty much it. Use it that way. Don’t use it in a way that isn’t shorthand or without knowing what it’s shorthand for.
• • •
In this case we see the limitations of the model and what kind of problems arise when real world issues depend on the exact fit between model and reality. Legally, the Big List of Existing People may get a new item when someone has their birth certificate made out, but when is the real list updated with a new person? In other words, when does a fetus become a person? If the Big List model corresponded perfectly to the world there would actually be such a moment, as many people believe. A soul is postulated to play the role of the essence, the clear-cut marker of list membership.
Like with “free will”, it’s a situation where following our intuitions leads us away from those very same intuitions because they are inconsistent. In fact, this happens so often I might want to call it “philosophy disease”. A section of my The Good, the True and The Undefined talks more about it.
Many philosopher- and scientist-types are bothered by this vagueness and compartmentalization and try to construct a complete, integrated map. This leads to flawed ideas like there being a single meaning of words (and a single correct world map) that we can apprehend by thinking really hard. I certainly have this impulse, which is why I’m so obsessed with resolving contradictions, defusing disagreements and integrating systems of ideas. I’ve just come to understand that it can’t be done by finding all the right answers.
Think of concepts like democratization, credentials, professionalism, usefulness, institutional inertia, politeness, centralization, treaty, postmodernity, unemployment or bank.
That’s if it even compiles, which typically requires manual interpretation. One can see the history of philosophy as centuries of attempts to create compilers that bridge the gap between words and whatever reality is made of.
Since the tokens are all interconnected, claiming that something exists doesn’t just invite us to use one particular token, but the whole system it’s part of. To use an example from the list: “sin exists” implies not just acceptance of some disconnected fact but the whole Christian worldview it’s situated in. “Patriarchy” is similar, as is a zillion other ideologically charged concepts.