Existence Explained
Existence Explained
Probably the most infuriating question in philosophy is the nature of existence. I have racked my brains over it lo these many years and come up with nothing.[1] Now at last I think I have found the solution, and it’s not what you would think. It fits the facts. So, right here, right now, I will give you the solution—dim the lights, Geoffrey. But first a quick reminder of all the dud solutions, the runners-up, the heroic also-rans (let’s have a round of applause for them): to be is to be the value of a variable (or the reference of a pronoun); to exist is to have the second-order property of instantiating a property; to exist is to have an unanalyzable first-order property of existence; to exist is to be an occupant of space; to exist is to be an idea in the mind of God; to exist is to be spoken of in a text; to exist is to have causal powers; to exist is to be an object of thought; to exist is to be a sense-datum; to exist is to be socially constructed; to exist to be not imaginary; to exist is to be material; to exist is to be mathematical; to exist is to be useful. A rum lot really. Hardly worth refuting. So, let’s get to the winner, the star of the show, the incontestable champ.
And the winner is: to exist is to be detachable. Yes, I know it sounds a bit bathetic, but stay with me awhile, be patient. The intuitive idea is that an existent thing is a thing that can be detached from other things—separated, disconnected. A non-existent thing is a thing that cannot be detached from other things. This is intended as a statement of necessary and sufficient conditions, a classical conceptual analysis. We are not pulling any punches or speaking metaphorically. This is what existence is. You need to see the analysis in action to appreciate its merits. What is the paradigm of an existent thing? Material objects, of course (the paradigm of our modern sensibility anyway—not the gods or anything like that). Perceptible objects in space—substances. They can be separated from each other, detached, disconnected. They come to us in conjunction with other things, but they can be physically moved and detached in thought. They don’t depend on anything else to be what they are. The table is detachable from the chair, even though they are next to each other in actuality; they will still be there if separated spatially. They are also detachable from the perceiver. They are independent things (the old definition of a substance). To think about one, you don’t have to think about the other. Contrast these things with imaginary objects: here you can’t detach the perceiver; no perceiver, no object. If no hallucinating perceiver, then no pink rats. Pink rats are necessarily attached to perceivers (as things are), unlike brown ones. The latter exist, the former don’t.
Now consider some intermediate cases, starting with universals. Plato thinks they are detachable from particulars: this is what constitutes their clear existence for him. Aristotle disagrees, so universals for him have less than complete existence—perhaps no existence at all. Fictional entities are not detachable from the minds of their creators, so their existence is in doubt; some say they are the paradigms of the non-existent (even more so than hallucinated objects). Meinongian objects seem indissolubly connected to minds (e.g., the golden mountain), so their existence is questionable, though a dose of subsistence is sometimes allowed. Then we come to some hard cases: numbers, mental states, ethical values, space, time, God, meaning, events, shadows, holes, shapes, concepts, selves. All of these have had their existence rudely denied: the question is why. Is the number 2 separable from other things—other numbers or mathematical minds? Could it exist on its own? If yes, it indisputably exists; if not, its existence comes into question. Does a particular thought exist? That depends on its detachability from other thoughts and from the brain. Holism undermines individual existence, and so does inseparability from the brain (materialist reductionism). Ethical values strike people as not detachable from natural facts, e.g., societal norms, so their existence comes into doubt. Ethical realism is better off under ethical autonomy (ethical truth before moral agents). Space looks in danger of disappearance if it cannot be detached from the matter it contains. Time starts to look wispy if events are removed. God cannot survive dependence on human belief. Meaning becomes existentially questionable if it cannot be detached from words—a mere shadow of language. Events lose all substance if they are constituted by objects and properties and are incapable of existence without them. Shadows are too attached to the bodies that cause them to be accorded genuine existence. Holes cannot be detached from what they are holes in. Shapes go up in smoke if detached from physical things, or are thought to. Concepts are found wanting if not detachable from language. Selves can have no being if they cannot be disconnected from bodies and mental states. In all these cases, the applicability of our basic concept of existence is called into question: the more attached a thing is, the less real. Those who resist these conclusions do so by insisting that the questioned items are detachable; they picture worlds in which the items exist autonomously. They can exist in splendid isolation! I needn’t go into all this; the point is that existence turns on the detachability question. If it turns out that material objects are really constructions from sense-data, then they too will have their existence queried. If it turns out that mental states are completely dependent on brain states, then they will be in peril of being relegated to the ranks of nonbeing. The central point of the concept of existence is that the thing in question can exist separately: the more detachable, the more real. This is why the primary cases are material objects in space and imaginary objects in minds: the former can exist in the absence of minds, but the latter cannot. The idea of existence is closely linked to the idea of objectivity, and objectivity is a matter of detachability from the subjective perspective. We say something is merely subjective when it doesn’t really exist. Realism is typically an affirmation of detachment from human concerns. Anti-realism is a claim of attachment to human concerns.
One nice feature of this theory is that it applies to any category of object: we don’t end up favoring one kind of object over others—physical, mental, or abstract. The definition covers all these cases. It also prescinds from epistemological considerations; it is purely ontological. And it brings out a latent relational aspect of existence: for x to exist is for x to be detachable from y. It thus provides a genuine analysis not just a trivial synonym (“to exist is to have being”). We have fashioned a concept that encapsulates the facts of detachability. If I want to say that pink rats are not separable from a perceiver’s mind, I simply say they don’t exist (are not real). If asked to explain what I mean, I say “Pink rats are inseparable from the perceiver’s mental state and dependent on it”. The word “exists” is very much a catch-all term, topic-neutral, quite abstract; and also, essentially contestable. It is a philosopher’s term with a theoretical meaning (more so than “true”). I like the definition because it injects new content into discussions of existence and doesn’t leave us with nothing substantive to say. It gets at the deep structure of the concept, lying just beneath the surface. Is there anything better out there?[2]
[1] See my Logical Properties (2000), chapter 2, as well as several essays on this blog. I came up with the present theory by going through every possible theory I could think of until something gelled.
[2] Why has it not been thought of before? Because of a tacit commitment to empiricism (the usual culprit): there is a hankering for something close to the perceptual—an existence sense-datum. But the concept of existence is a highly intellectual concept—it connotes an abstract structure. It involves a quasi-geometrical modal structure: the mind has to detach in thought one thing from another in an act of modal separation—a possible world, in effect. The concept of existence is a modal concept in that it requires conceiving of a possibility—this without that. For example, for me to recognize that a part of an object exists I have to construct a possible world in which the part is detached from the whole and considered by itself. This is nothing like seeing red. I doubt that any animals have this concept. We also have the difficult task of deciding whether one thing can exist without another—whether, say, a property can exist without a bearer. The concept is much more complex and problematic than traditional theories have realized.
I vaguely remember a paper of Kit Fine from the 90’s on ontological dependence , on modality and essence or something like that. In any case, I am not quite clear on what you mean by “detachable.” Logical separability? Physical separability? Conceivability? A shadow is physically inseparable from its object but logically conceivable without it. Another issue leaving me unsatisfied is kicking the can down the road….instead of asking “does something exist”, should we now be asking “is something detachable?” But what does detachable mean?? Would you be able to define what “detachability” is more precisely? And the thing with numbers, yikes. Could 2 exist independently of other numbers? Numbers are necessarily relational (2 is more than 1, less than 3). Math is most definitely relational, and so should we, by your argument, conclude that math does not exist? I think our Math department would make a strong objection to such conclusion.
You do know that many philosophers of math have denied the existence of numbers as abstract entities? Formalism, nominalism, fictionalism, intuitionism. Math exists but not numbers for these philosophers (e.g., Hartry Field). All forms of separability count, but conceivable separation is the acid test (remember Kripke on the mind-body problem). If shadows can conceivably exist without things that cast them, then that is a mark in favor of their existence; if not, not.
Does your definition answer the riddle of why anything exists at all in some conceivable way, what might be called the hard problem of existence? I’m hoping it does.
Not that I can see. It’s just a definition.
“Probably the most infuriating question in philosophy is the nature of existence.”
Why is this such an infuriating question? Presumably it arises from the interest in prior questions like “What exists?” and “What is there?”, and questions about individual candidates, like “Does X exist?” and “Are there Xs?” And then the questions like “What is existence, anyway?” and “What is being, anyway?” inevitably follow. As an outsider I find the obsession of philosophers with these kinds of questions puzzling. So why do questions of “existence” and “being” like these seem to be of such fundamental importance for modern philosophers? After all, there is already enough problem with questions like, “When you ask ‘does X exist?’, what exactly are you trying to refer to out in the world when you say ‘x’?” (And don’t just repeat the word “X”.) What is the importance of this for anything, why does it matter? Is it just pure curiosity? Was it mainly because of the question “Does God exist?”, which by the way seems like a case of circulus vitiosus? Didn’t Kant provide an alternative way of approaching these kinds of questions?
Philosophers have always been interested in what exists and what does not–we call that ontology or metaphysics. They also want to know what “exists” means, i.e., what existence is. Is it a property like being red? They have also been interested in the nature of reference. This has no practical relevance; it is an intellectual question. Surely, this is easy to understand.
Of course, all of that is easy to understand, but that’s not what I was asking. I share the general interests, but I was trying to focus on “these kinds of questions”, an approach that seems to get hung up on words like ‘exist’ (or ‘existence’) and “being”, that involve quirkish aspects of English grammar. (In “ordinary language” if you ask “Do events happen?”, the answer is, “Of course they do.”; if you then ask, “Do events exist?”, you might get “I don’t know. I thought “exist” was for substantives”. Grammatically reasonable, and I’m sure a lot of people have noted that. So why not wonder where that distinction comes from?)
I don’t know what you mean by getting “hung up” on “exists”, any more than philosophers have been “hung up” on the word “knowledge”, i.e. the analysis of knowledge, and many other concepts.
‘Understanding’ is an interesting word. I’m kind of obsessed with that one.
BTW, I only said I was puzzled, a very productive state, one that I enjoy being. As someone who’s supposedly trying to do empirical science, I’m frequently puzzled by the kinds of things philosophers can say and get away with it, where if I put them in a linguistics paper it would be grounds for it getting thrown in the bin. So when I asked “Why is this such an infuriating question?” I just wanted to know what makes it so infuriating for you or other philosophers. To me ‘exist’ is used mainly to describe the presence vs the absence of some object in/at some spatial location. The object can be abstract, such as “There exists a lot of uncertainty surrounding this question”, and then people debate about whether it’s real or not. When the wondered-about object is something like “God”, and one asks, “Does God exist?”, the question becomes, what exactly are you asking about whether it exists or not?” (you can’t just say “God” again), and “where does it exist or not exist?” These are the kinds of mundane questions a linguist might have. But philosophy goes way beyond this. That’s what I want to know about.
It’s infuriating because it ought to be easy to answer but isn’t. Existence is clearly a very basic concept, but how is it to be analyzed? Is it a first-order logical predicate like “man” or “red” or is it a higher order property of a first-order property? It should be easy for you to find out about–just google it. There is no secret. I wrote at length about in my book Logical Properties.
Not only numbers and mathematics, but everything in nature is relational. A table depends on wood molecules, molecules depend on atoms, atoms depend on quarks and leptons, keep going until it bottoms out at quantum fields — themselves defined almost entirely by how they interact and relate to one another.
Further, all relationships in physics sum to zero: particles balance antiparticles, total energy sums likely to zero, momentum to zero, charge sums to zero, and so on. These conservation laws emerge from symmetries that themselves arose from nothing. At all times, these relationships sum back to their foundation — this relation-driven reality is essentially a reductio ad absurdum of your detachability framework. Your framework seems to imply that the only truly existing object is the only one independent of any relationship: namely, nothing. In other words, your detachability theory dissolves in an unexpected way — nothing is detachable because everything cancels, and the underlying reality is nothing.
Mathematics suffers the same issue. Mathematics is relational in an axiomatic sense: calculus depends on limits, limits depend on algebra, algebra depends on operations, operations depend on sets, and at the bottom of it all is the empty set — ∅, the nothing. The empty set is the only object that depends on nothing prior to it. So mathematics and physics share the same deep structure: relational complexity with nothing at the foundation. All of mathematics is the empty set rearranged — calculus, topology, number theory — all of it bottoms out at ∅.
In both cases, nothing is at the foundation, rules govern how that nothing relates to itself, enormous complexity emerges from those relations, and the total always sums back to a foundation that depends on nothing. If that’s right, existence may be neither detachability nor mathematical being, but the pattern of relations itself — which has no detachable foundation anywhere.
Kit Fine attempts to rescue dependency-based frameworks by distinguishing types of dependence — a tree depending on water is causal dependence, a shadow depending on a body is constitutive dependence — suggesting perhaps only certain kinds undermine existence. But this move only works if you can specify precisely which dependencies matter and which don’t.
That brings us to the central failing of your framework: you never precisely define the term that does all the work. If you want to salvage the detachability theory, the necessary first step is simple — define detachability.
But I did define it several times. The OED gives us “disengage (something) and remove it”. That can be physical or conceptual, i.e., conceivability apart. We could say things are detachable if they can exist apart in a possible world. This has nothing to do with whether reality is relational. I am not saying that existence is lacking relations! It is about one thing logically implying another or not. I don’t think you have much understanding of my paper and are clearly not philosophically literate.
Fair enough: modal dependence isn’t mere relationality. But if shadows, holes, pains, and tables all depend on other things in your modal sense, then I wonder whether you’re really analyzing existence itself or ontological independence. Why should dependence make something less real rather than simply derivative? And why think our ordinary concept of existence tracks degrees of detachability?
Because if something is non-detachable it is reducible, but if reducible then non-existent. This is why Plato is right about universals (they exist) if they can be detached from particulars and Aristotle is right if they can’t be detached (they are reducible and hence non-existent). Same for mental states if Descartes is right (they exist), but if materialism is true they don’t exist. Etc.
You mustn’t use “depend” in a loose way to include things like causal dependence; it must be restricted to analytical necessary connection.
Bravo! Philosophy with a feather pen.
I kept it light despite the weight of the subject. The performance was all.
Please clarify a couple of things about your theory. Pains are non-existent as are pink rats? Optical illusions (as opposed to hallucinations)?
Pains exist, pink rats don’t. Optical illusions don’t exist, though experiences of them do and the physical objects perceived (the unequal lines of the Muller-Lyer illusion don’t exist but the physically equal lines do).
While I understand that detachability criterion could be logical or physical, I don’t see how it applies to pain. How pain can be separated from a subject experiencing it? Pink rats do not exist because they can not be separated from the hallucinating person. But pains without the subject are like shadows without the object casting them. That’s my intuition. Please show what is wrong with it.
If pains were reducible to brain states and hence not detachable, they would not exist as we normally think of them. But the relation between pain and the subject is not identity, so we can conceive the two as separate–the two can be detached in thought (here is the person, there is the pain). It might be held that they are also modally detachable, because you can have the pain existing without the body or self (depending on what the self is). People have entertained the idea of floating selfless pains, but not floating selfless pink rats. My point is just that this is what is being debated when questions of existence are raised. It could certainly be argued that pains don’t exist–they might just be adverbial, say.
Absolutely Brilliant! I am going to take that one to the bar… to exist is to be detached.
Question: What does this theory have to say about jointly sufficient conditions with regard to material constitution?
I am assuming that anything that arises from jointly sufficient conditions among objects do not exist because they are attached to other objects. You give the example of a table & a chair. But doesn’t a table depend on the legs? I admit a table is detachable from a chair, but defiantly not detachable from its top. Enlighten me!
The parts of an object are detachable from each other and the whole, so they exist. The table itself is detachable from other objects. The table is not detachable from its parts, so doesn’t exist independently of them. The table does not exist without its top, so not detachable from it.
OK, I’m taking a stand! Plato was right! (Based on what you said in the post about universals and particulars.)
So, you are committing yourself to universals laid up in Platonic heaven existing prior to any particulars that instantiate them.
Thanks! I was wondering what I was committing myself to. I would say that the universal is logically prior, not necessarily temporally prior, to the particulars that instantiate it. But there can be no identity relation between any particular and the universal that generates it (creates its possibility as (potentially) actual). The universal is necessarily detachable from the particulars that instantiate it. The universal creates only the possibility of the particular. Something else is needed to make the possibility actual. Does there exist any modern philosopher who holds the Platonic view on these questions?
Frege, Godel, Russell, Penrose, Katz, and many others. There’s a comprehensive article in the Stanford Encyclopedia on platonism.