Could the Universe be a giant quantum computer?
nature.com
nature.com
I see a lot of speculation about the possibility that our universe is a simulation. But our experience with inadvertently building Turing-complete systems makes me wonder if our universe is running on something that is just a side-effect of some other computation and not something deliberately constructed.
Someone in the enclosing universe could fix a bug and suspend or destroy our universe.
Or the computational substrate on which our universe executes is just some natural process in the enclosing universe, like sea foam or solar flares.
I wonder what the automated tests will look like to us...
It is not as simple. See Permutation City.
TL;DR; from a perspective of a person living in Game of Life, your time doesn't rewind when the computer running it rewinds the state.
Are we doing anything more than updating his question to our latest technology?
Is there a fundamental difference between this and, say, 'The Truman Show'? Other than a more sophisticated layer of MacGuffin.
[1] https://en.wikipedia.org/wiki/Evil_demon
[2] https://en.wikipedia.org/wiki/Meditations_on_First_Philosoph...
It also calls into question how much we can learn about the beginning of our universe. We are "trapped" by the physics of our universe so there may be no grand explanation of where all the energy came from. If the universe is simulated it's not like we're going to suddenly get down to the source code, there will just be a distinct point when the universe booted and all the stuff started happening.
I prefer to think that there are symmetry/conservation laws that work bidirectionally in time. Even if time travel were possible, the only events that could occur are those that don't break any symmetry/conservation laws. This could be seen as a weaker form of simulation hypothesis leaving out the concrete 'how' as an unknown.
I remember coming up with the idea of 'time legs'. A tree can't pick up and move somewhere else, and a falling leaf doesn't choose its path. Other creatures can. We just don't have the apparatus to move freely through time--our memories only form when moving forward.
We just need to figure out how to affect our relativity without needing to go insanely fast or being swallowed by a gravity well. Easy task...
But on the other hand, I can't remove the idea from my brain that maybe there is some kind of hidden variable/dimension/other-concept-we-can't-be-aware-of at the source of this phenomenon explaining perfectly why we see it that way.
Our science knew a big expansion since the beginning of 20th century thanks to some geniuses. And now I have the feeling there is no more 'science explosion' coming. Sure, there are some good intuitions, but nothing concrete or testable.
Everything nowadays lies on these fancy ideas to prove them or push them to their corners. But nothing really appealing, it's just some 'monkey testing' of our universe. LK-99 : false hope, string theory: too much variable and kinda not testable, dark matter or lambda CDM : nothing going towards the confirmation. Sure, the math/physics models predict a lot of stuff, but nothing incredible.
I have the feeling something big is missing to understand our universe, but we're not working in that direction.
Sorry, needed to vent.
Still, I like your idea.
It all stems from a certain fluctuation in a specific quantum field, so we may be able to discover properties which allow us to manipulate these as well as other quantum fields. For me it kind of boils down to the more we understand reality, the more complex it becomes, and it always has more layers. We don't truly understand what reality even is, and can only experience certain parts of it with our senses.
So have no fear, the drive to understand what is going on around us is a very human trait and we have built some of the best tools in the history of humanity in only the last 60 years. There's a lot of runway left on science :)
The problem with the universe is that, right now, we operate entirely within its confines and we know we don't even fully understand its physics, let alone any sort of 'physics' which may operate outside of it--should there be something outside of it.
Unless something changes in our view of the universe, this will very likely remain a philosophical question, likely for our entire existence and the existence of whatever intelligent life forms replace the human race several times over in the coming billions of years of this planet.
Like humans existed in some other existence, but we were going to get Gnab Gib'd, so we ensured that the next universe would inevitably rise to precisely recreate us through a convoluted 4 billion year evolutionary program in a tiny speck of dirt in one of billions of galaxies?
It's like some convoluted anime/manga plotline that occurs in the final stages of a 10-15 year run. Or heck, like the infamous "Dallas" plot element from the 1980s: THE ENTIRE SERIES WAS JUST A DREAM.
Or "Lost".
It does seem that quantum mechanics hints at the universe being some bizarre gigantic n-dimensional computational matrix, but that's probably because of the discrete quantities and rigid math and things like hard orbits/energy level jumps which seem a lot like memory assignment or cellular automata.
In other words, we have no basis for saying our universe exists, that we have physical existence, that anyone but ourselves exist, or that we exists in time.
Sure, I will grant that.
This does not warrant a leap from we don't actually know what the Universe is to therefore definitely simulation.
There are lots of unknowns there we can't quantify very well, but unless the universe is firmly finite in both time and space, I find it increasingly difficult to see a way around simulation in some form or other.
Especially because the threshold for us to find ourselves in an apparent simulated universe might well be the barest sliver of simulation: One mind, a single moment.
Even assuming we one day try to run a large simulation, I'd expect we before then will have run huge number of simulations of minds for brief slivers of time to try to figure out if we get it right.
It seems likely that if we're in a simulation, we're even more likely to be in a brief, constrained simulation rather than a full blown universe simulation.
That of course doesn't mean we definitely are - like the Fermi paradox it's more question of "why don't we have a better answer?"
I disagree. Physical laws hold throughout the Universe - or, that is the consensus null hypothesis anyway. In even an infinite Universe, assuming it is infinite, there are limits to that which is possible, and it is not yet demonstrated to my satisfaction that a computer could even theoretically simulate a Universe.
That's leaving aside my fundamental objection that the space of what the Universe can possibly be is infinitely vast compared to the singular assertion that it's a simulation. In other words, that this is the correct view is exceedingly unlikely simply because it has been thought of. Imagine that I generate a random complex number of arbitrary precision and ask you to guess it. It could be `0.46451453499412721134961376996+5.275644069844024487293819737719i` for example. It's my view that you have a far, far higher chance of guessing that complex number than we humans do of coming up with a simple, correct cosmology.
> the prevalence of computation in the universe
I also disagree that it's certain that computation or even mathematics are universal or fundamental. I suspect that these both are based in a very human outlook and perspective. Their apparent universality is an illusion deriving from our inability to see outside of our human perspective.
> In other words, we have no basis for saying our universe exists, that we have physical existence, that anyone but ourselves exist, or that we exists in time
While true, I'm not sure what this has to do with the simulation hypothesis. Really, given this premise, the Universe could be anything at all.
You assume that the physics of the simulation are equivalent to the physics outside it. We simulate simpler environments than our own all the time.
Maybe the constraints of our universe are the way they are because it reflects the tradeoffs required to be able to simulate it given the conditions the simulation runs in.
If we do assume that an observer cannot differentiate the simulation from the outside world, then we can make an interesting argument: if the rules of physics in some universe entails that this universe will simulate many other universes with the same apparent rules, then an observer of these rules is probably simulated regardless of what the root universe happens to be. This is the only way you can ground the probability calculation. A lot of things are possible. Not many things are probable.
It's a low bar because you don't need to presume the simulated entity gets the benefit of the passage of time. For it to be likely for us to be in a simulation, it's sufficient for it to be likely for the total runtime of "simulated us" to be higher than "real us", and our only knowledge of the passage of time are the momentary memories and sensory inputs right now, and we can't validate their truthfulness.
In other words: you don't need to simulate any physics to prevent a mind from differentiating. You just need to limit the scope of your simulation to a sufficiently narrow slice of time.
Personally, I tend to think that if simulations become widespread, test runs, debugging and development, research runs, or application specific runs of small slices are likely to far outpace full scale simulations in runtime, and so if simulations are probable, I'd be inclined to think we're most likely in a (very) short-lived, scope-limited simulation in a horrible groundhog day like setting.
But if your simulator is not simulating physics... what is it simulating? The mind? If you are simulating a human brain, that IS a physics simulation, and it is probably on the harder side of the scale, so we cannot readily assume it's going to be easy or fast. And if you are running some different kind of mind based on an abstract neural network or something like that, well, it's not a simulated mind, it's just a mind.
But IMO, this sort of theory makes a lot more sense in a "large" cosmology like Tegmark IV.
I'm not sure what you mean by that, but if the simulating universe is not qualitatively identical to the simulated universe, you lose the mathematical properties that make the simulation argument interesting.
> it may work if you assume that most of the universe at the moment is computationally "dead" and can be optimized out.
Our universe is chaotic and is largely defined by its edge cases sometimes leading to a global phase change (e.g. evolution). If one wishes to preserve these properties, I would argue it is not optimizable at all. If you are willing to lose these edge cases, maybe you can optimize certain things away, but it's quite possible you will lose more than you bargain for (basically nothing is computationally inert in a chaotic system). Your system will be bottlenecked by the most demanding parts of the simulation anyway.
> But IMO, this sort of theory makes a lot more sense in a "large" cosmology like Tegmark IV.
I tried to work it out mathematically a while ago, but I don't think it really works out (I forget the details, I should dig it up sometime).
If the infinity of possible mathematical structures actually exist, there's still an infinity of possible mathematical structures to simulate. And if, to simplify, 1% of mathematical structures are conformant with our universe, by default you would expect 1% of simulations (across the multiverse) to be simulations of our universe (unless you think we are somehow special and other universes specifically want to simulate us). So you would need over 50% of the multiverse to be simulation fodder in order for a given observer to probably be in a simulation, and that seems... kinda really high?
All of these conversations delve into the same uncertainties as faith. We’re trying to understand things we can’t truly ever understand or comprehend.
If you agree, then we agree.
A book can create emergent behavior in a mind that reads it (or a computer that processes it). But without those, it does nothing.
A computer doesn't need to be observed to run. You can press a button, walk away for a few minutes, and come back to find an entirely novel image generated by Stable Diffusion, for example.
Sure, all computers currently require an external person to push the button, perform maintenance, provide electricity, and so on. But that doesn't mean that all computers will always need this.
Meanwhile, all books will always just be static words, because that's what books are.
Now enumerate all the traits that the Universe has that neither a book nor a computer has. Qualia, consciousness, time, gravity, electromagnetism, life, among infinite other differences. The list is literally infinite.
The gulf between the Universe and a book is the same magnitude as that between the Universe and a computer.
And sure, books are static once printed and need an observer to run, but that doesn't mean books will always need this.
You've lost me here. Books don't run. A book is a large amount of organized text. That's the definition of a book. If you changed into into something that could also read the text, it would no longer be just a book.
That is disappointing.
Physical laws might be greatly simplified, anything you can't observe might not be simulated at all until necessary, the simulation might not be real-time, and the world running the simulation might have 6-dimensional spacetime (after all we love simulating 2d worlds).
E.g. a system might apply a computational "budget" per entity per time step in terms of "in universe operations", in which case time dilation could naturally fall out (more time spent simulating motion translating into less time available to simulate local subjective time)
That doesn't prove that is how our universe works, but it's fascinating all the same.
Those two explanations come from exactly the same impulse: explaining the weird fact of our existence in terms that are reducibly familiar to us.
If a tree falls in the forest and nobody is around to observe it, do you need to simulate it down to the quantum level? Or do you just do coarse-grained simulation, with fine-grained steps reserved to where it causes observable differences? If later somebody comes around to observe the fallen tree in too much detail you can always just simulate that part in more detail, and hide the difference behind what we call "inherent randomness at the quantum level".
If I, a being in your simulation, was to hypothetically build a simulator of the kind that you use to simulate me, it is obvious that it could not work: if it could approximate the running of itself to a sufficient level of detail, it already would be doing that. So the question is: what else wouldn't work? The closer the technology I build gets to physical optimality, the more it will act as a bottleneck in the simulation. If I build a physically optimal circuit to perform any useful calculation that I can verify, for example, the simulation speed will have to drop below one second per second.
What this means is that modern times are probably too expensive to simulate, so you would likely stick to simulating medieval times (we may also have to eschew simulating the brains themselves, if it turns out the biological ones are too close to optimality for what they do). At least that's what we would have to do. You can imagine that the universe that simulates us works differently, but hypothesizing a world that has different laws of physics doesn't really help us to understand ours. It's fun to think about, but nothing more than that.
Moreover, you have to make that assumption, or the equivalent assumption that computers can become so advanced that we are the equivalent to their middle ages.
My objection to the simulation hypothesis is that, if one must make those assumptions to make it work, why precisely and only those assumptions that make our Universe a "simulation": that we are here because of the intentions of creatures, that "simulation" applies to whatever it is they're doing, that the creatures are human enough to find simulating us useful.
The simulation argument is that if you observe a universe X, and you work out that a real universe X will simulate two universes just like it, then someone who observes that they are in a universe that looks like X is probably in a simulation. That is because for every real X you have two simulated X: even if you assume that there exists a truly real and physical X, it still follows that most observers of X are simulated. I don't think our universe will simulate more observers than there are real observers, but at least this is something we can meaningfully argue about.
If, in contrast, you say that there is a universe Y, and Y simulates X, there are no interesting conclusions you can draw from that. You are comparing "X is real" to "Y is real, and Y simulates X", but there's no reason to think Y is more likely to be real than X. It's possible, but a lot of things are possible. It's not interesting.
Personally, I don't find the argument compelling because of its unexamined, embedded assumptions. What is meant by computer? What is meant by a simulation? What is meant by Universe? What is an observer? How is it falsifiable? By the time these terms are pinned down, the argument disappears in ontological haze.
In a nutshell, some people believe in Moore's law more than they believe in the laws of physics: the failure of the former would surprise them more than the failure of the latter.
> What is meant by computer? What is meant by a simulation?
These are especially interesting questions once you think about the low hanging fruits for optimizing a world simulator. If you want to simulate the interaction between two systems as efficiently as possible, you want the states for these two systems to be located as closely together as possible, in order to minimize communication overhead. And if you are trying to simulate a 3D universe, and calculate what would minimize internal communication delays, the best computer architecture to do this is, well... it's going to be isomorphic to what you are simulating.
That is to say, if you travel around the planet in the most efficient computer simulator, an external observer would probably see a bunch of photons and electrons travelling in a physical circle inside the simulator. And the ensemble of transistors that simulate you at any given moment would probably physically look like you.
When your simulator physically looks like the universe it is simulating, is it still a simulation? Or is it in the middle of some spectrum between real and simulated? Ontological haze indeed.
> …In that Empire, the Art of Cartography attained such Perfection that the map of a single Province occupied the entirety of a City, and the map of the Empire, the entirety of a Province. In time, those Unconscionable Maps no longer satisfied, and the Cartographers Guilds struck a Map of the Empire whose size was that of the Empire, and which coincided point for point with it.
https://m.youtube.com/watch?v=lH7H9d28BR8
Although he talks about Gods and Creation, I think the same argument applies to the Simulation Hypothesis aswell.
This is as much speculation as a statement that such a system does exist.
I guess we can't rule out that someone is using Earth as a computer, but if they were, you'd expect the mice to get better cheese.
But, if we take that the universe has rules, and all particles are following a wave function, and interacting. Then it is ok to think of our universe as 'executing functions'. So our universe is self executing, like a computer.
There is no randomness, just rules we haven't discovered. See Super Determinism.
First, what a computer is cannot be identified with any physical system, because computation is observer-relative. There is no fact of the matter that some physical system is performing computations. Computation is not a natural kind. We can harness physical phenomena "computer-wise", that is, use physical phenomena to simulate computation, but it makes no sense to identify computation with physical phenomena as such. That physical phenomena involve quantities that can be computed does not mean the phenomena are performing computations as such. A simpler example of something analogous is a book. The print in a book, physically, is just a series of pigmented spots on sheets of something made of plant fiber. Physically, it isn't even "writing" per se, as writing is not a natural kind, but what we call the result of the act of writing. The pigmented spots on plant fiber are the only content, physically, of what is in a book. The content we think of when we speak of books is brought to the book by the reader who interprets the marks on the pages according to what he already has in his head. This is what reading is: interpretation, assigning meaning. The same is true of computers. Physically, no computation is occurring.
Second, computation is about something. When a physicist or engineer computes the current in a circuit, he isn't manifesting the circuit. He is computing something about the circuit. Computation is an abstract activity of the mind. (We can use this in an argument against materialism.)
Third, part of the reason why this identification of computation with physics is likely occurring is because the methodology or methodologies of physics necessarily strip reality of much of its substance and leave us with a mathematical, formal model expressed in computational terms. Guess what else is a formal model that involves computation? This allows us to talk about the formal equivalence or subsumption or whatever of physical models and computational models. That such a relationship exists isn't that surprising in retrospect.
Alas the zeitgeist is conducive to "bits will eat the world". Here is a more radical idea: Could the Universe be a giant Universe? (= a recursive function). Maybe its Universes all the way down?
[1] Apparently yes