An argument that existence cannot be the ground of itself — and an honest look at what more it takes to conclude it was caused by something outside of existence.
A topic that’s been debated for thousands of years in philosophy is whether or not our existence is caused by something else, or just is. This tends to be an important question in people’s core beliefs, since how you answer it shapes how you think about the origin of everything.
I should state upfront that this is purely an argument about causation. Through logic, I will strengthen The Kalam Cosmological Argument as far as logic honestly reaches: I will show that reality cannot be the cause or ground of itself, and that — given one further premise, the Principle of Sufficient Reason — an external cause follows. I will also be clear about exactly where logic stops and that premise takes over, because pretending the gap isn’t there would be its own kind of dishonesty.
I will not be arguing for any attributes beyond that single one: it caused our existence. Any other attributes are beyond the scope of this argument.
One of the most well known arguments for the existence of another cause is from Aristotle and his Unmoved/Prime Mover Argument.
The argument Aristotle makes can roughly be boiled down to this:
Aristotle essentially points out that one of the laws within our existence seems to exist never-endingly, and that nothing within our existence could have caused that. Within our realm of existence, anything that imparts motion onto another object must be in motion itself. Therefore since motion exists, something must have begun that motion: the unmoved mover.
This is the argument that gave birth to other cosmological arguments. One of them being the Kalam Cosmological Argument:
This argument is a deductive argument, and if the premises hold true, then it is absolutely necessary that the conclusion follows.
Note: The word “universe” in this case really means “existence”. If it turns out that there exists “multiple universes”, then this argument will simply change to encompass all universes that exist.
Now, since the conclusion to the Cosmological argument follows necessarily, objections to the argument happen within the premises:
And these are valid objections to the arguments.
Or, are they? (for illustrative purposes, please imagine Michael from VSauce saying this).
I will strengthen the argument here by refuting both objections to the first two premises. It requires a chain of rebuttals across multiple smaller arguments, so bear with me.
I am also changing the word “universe” to “reality”, and I’m defining it like this:
Reality encompasses all laws, space, and material. This includes things such as gravity, light, physics, logic, the universe or multiverses, and particles.
By definition, we can also state that the only enumerable possibilities for cause are:
Now, let’s rebut the 2nd and 3rd possibilities.
The rebuttal is a proof by contradiction. Suppose, for the sake of argument, that reality has no external cause. Then only two possibilities remain:
If both can be ruled out, the supposition fails and an external cause follows. The first can be ruled out by logic alone. The second, as we’ll see, can only be ruled out with the help of an additional premise — and I’ll be upfront about exactly where that premise enters, rather than smuggling it in.
The concept of an internal cause creates a logical contradiction:
We have to be careful here, because “no cause” actually hides two very different ideas, and they need separate treatment:
Reality cannot ground itself. A cause, or a complete self-explanation, would have to be in some sense prior to or independent of the thing it explains. Reality cannot be prior to or independent of itself. The same obstacle shows up more formally: for reality to fully account for itself, it would have to contain a complete and consistent description of itself from within. Any system rich enough to do that runs into self-reference limits — this is the neighbourhood of Gödel’s Incompleteness Theorems, which show that a sufficiently expressive, consistent, formally-described system cannot prove every truth about itself from within its own axioms.
(I should be honest that applying Gödel to “reality as a whole” assumes reality is the right kind of formally-describable system, which is a strong assumption in its own right. I lean on it only as illustration; the informal point — that nothing can be prior to itself — already does the work.)
Either way, the result is the same:
Reality cannot be self-causing or self-grounding.
But “cannot ground itself” is not the same as “must be caused.” This is the step I originally got wrong. Ruling out self-grounding leaves us with two options, not one:
The self-reference argument has no grip on option 2, because a brute fact makes no claim to explain itself. When we show “reality cannot account for its own existence from within,” the defender of brute existence simply agrees: “Correct — there is no account, internal or external. That is what ‘brute’ means.” So self-reference refutes self-grounding; it does not refute brute-ness. (Note too that uncaused is not the same as self-caused — a brute reality performs no causal act on itself; it just isn’t caused at all. The self-reference fallacy only touches the second.)
Can we rule out brute existence? Honestly: not by logic alone. The statement “X exists with no explanation” contains no contradiction. It is deeply unsatisfying, but it is not illogical. To rule it out, we need an extra premise:
The Principle of Sufficient Reason (PSR): everything that exists, and every fact that obtains, has an explanation or sufficient reason for being so rather than otherwise.
If we accept PSR, the argument goes through:
So the conclusion does follow — but it is exactly as strong as PSR, no stronger, and it is worth being clear-eyed about what that means.
PSR is not a theorem of logic; it is a principle we choose to adopt. The best case for it is that we already rely on it everywhere else. In science and in ordinary reasoning, we never accept “it just is, for no reason” as a real answer — we treat an unexplained fact as a problem still to be solved. To allow exactly one brute fact — the existence of all of reality — can look like special pleading.
The honest counter is just as real. The existence of reality as a whole is not an ordinary fact sitting inside reality; it is the precondition for there being any facts at all. So it may not be the kind of thing that has, or needs, an explanation. Someone can consistently accept PSR for everything within reality while denying it for the totality, and they cannot be convicted of a logical error — only of a different starting intuition.
Two further honesties, even if we do accept PSR:
First, PSR cuts both ways. If nothing may be brute, then the external cause may not be brute either — so it would need its own ground, and that ground its own, without end. To stop the regress we must allow exactly one ungrounded terminus somewhere. But the moment one brute terminus is permitted, reality itself becomes a candidate for being it. Indeed, most accounts of explanation agree that explanatory chains must bottom out in something fundamental and itself unexplained. So the real question is not “is there a brute terminus?” — there must be at least one — but “is reality that terminus, or is something beyond it?”
Second, even if the terminus is external, we cannot describe it. As argued in Categorizing Reality, anything outside reality is not bound to our reality’s laws or logic. Our concept of “cause” may only approximate whatever relation such a thing bears to us — or may not apply at all. The most we can honestly say is that reality does not and cannot ground itself; the rest, by its very nature, lies beyond what we can know.
Stripping the argument down to what actually survives:
So the strongest logical claim is not the flat “an external cause must exist,” but this: reality cannot account for itself, so it rests either on a brute fact or on an external cause — and if we hold that nothing is truly brute, it must be the latter.
That is weaker than a proof, but it is far from nothing. It rules out the comfortable middle position that reality is somehow self-explaining or self-causing. It forces a genuine choice between two uncomfortable options — an unexplained brute reality, or a cause beyond reality that we cannot comprehend — and it shows that confidently denying any external cause is itself a bet that PSR fails, not a neutral default.
Common objections that are illogical and aren’t core to the main argument fall here. It’s worth touching upon them as they’re often used to dismiss the argument.
If an external cause exists, what caused it? Would this require an infinite number of causes?
This is a categorical error and touching on this argument is great for understanding higher-order categorization.
The failure here has to do with assuming that an external cause must follow our reality’s laws. This is not true. By definition such a cause would be outside of our reality. If reality is externally grounded, that ground need not follow our reality’s laws at all; it could have entirely different rules. Furthermore, it is impossible for us to comprehend what they may or may not be since we can’t break out of our own reality. Our logic works within our own reality but doesn’t necessarily apply to anything above it. In particular, we cannot know whether the external ground itself requires a cause, so a chain of causes is possible, but nothing forces it to be necessary.
For example:
A video game developer creates a game world, and inside the game it has the following law:
In the eyes of a video game character, they become more healthier by eating cakes. Their health is a simple point system. The video game character is programmed in a way where this is the only possible way it could understand health. Its understanding of health however does not apply to the developer, who can’t just eat infinite cakes and has to do other things to care for their health.
Therefore, it’s only safe to say that it is possible that there could be a chain of causes, but not necessary. It is not logically comprehensible for us to know what the answer is here.
This objection misunderstands the nature of the argument presented here. The classic “argument from the gaps” fallacy occurs when someone uses “we don’t know” as evidence for a particular conclusion. However, this argument is not based on ignorance — it reasons from the structure of reality plus an explicitly stated premise.
Science helps us understand our observable reality and it does sometimes replace older explanations, but it cannot solve everything. Since Science deals with what we observe, it struggles to deal with things we can’t observe. Science cannot answer questions about what exists outside of reality itself, because:
Science operates within reality’s framework: All scientific methods, observations, and experiments occur within reality. Science can tell us about how reality works, but it cannot investigate what exists outside of reality’s laws, space, and material.
The question is meta-scientific: The question “what caused reality?” is not a scientific question—it’s a philosophical and logical one. Science studies phenomena within reality, but the existence of reality itself is not a phenomenon that can be observed or tested scientifically.