If the planets orbiting Trappest-1 nominally exist (ie they are discoverable in our future), then surely they need to be simulated now so that there are no inconsistencies when we do find them.
That aside, If we are part of a simulation, why should we suppose it is anything to do with us? Maybe the simulation is of a set of initial conditions and we are just an irrelevant consequence.
Even if such a simulation were possible isn’t the probability that we happen to be living in the timeframe that such a simulation was created and maintained fairly slim?
I think it’s worth making a distinction between the two claims:
Claim 1 is really nothing new of course – doubting our reality has been a fundamental part of philosophy, across many cultures, and for millenia.
However, there is still some value in this new rendering of the concept, because it is including somewhat of a proposed mechanism.
So, many people who would be dismissive about speculation like “maybe this is all a kind of dream”, are more inclined to take the simulation idea seriously.
Also, it prepares people for hyperreal VR, which I think one day will indeed have a profound impact.
Claim 2 I think is debatable, and gets very interesting when you try to fully take it on and think about counter-arguments. It’s a little dismaying to see countless technologists and even philosophers jump on board without applying this kind of skepticism.
Here are some counter-arguments I thought of:
a. Part of the premise, is that when humans look down a microscope, or through a telescope, some of what we see is smoke and mirrors, because it would not be possible to really simulate billions of galaxies down to the level of the atom.
–but–
This of course throws up a whole can of worms. We’re essentially saying that the simulation is in some sense “forced”; it is not an open simulation, being allowed to evolve entirely naturally, but has at least some answers fed in (the conscious inhabitants must not guess that it is a simulation). This throws up problems like: How much is forced – we have no reason to assume that fooling us about the nature of the simulation is the only possible motivation. Is everything I’m seeing and hearing forced (i.e. something like solipsism)? What’s the probability of living in a forced vs unforced reality – we might expect the latter to far outnumber the former, since they require less effort to produce (there’s no rule that a simulation must pretend to have billions of galaxies).
b. How many universes don’t make hyperreal simulations? Part of the premise seems to be that there is one real reality that a conscious agent might find herself in, but that’s something we don’t know. If you combine simulation theory with many worlds, you get…an unholy mess. And certainly not one where there are enough knowns to declare we know that living in a simulation is almost certain.
btw I put “real” in quotes because I think it’s actually a misconception to consider real as some kind of attribute that an object has. But that’s a reasonably big tangent on its own, so I am parking it for now.
That’s not running a simulation of the world, that’s running a simulation of one brain and giving it fake input to make it think there’s a world around it. The idea that someone can make a ‘brain in a jar’, whether using an actual flesh brain in a jar or a detailed simulation of a brain, doesn’t seem to be what the original question was about.
Sure, it’s possible in theory, but other than the novelty why do it? And especially why would these future people with access to huge amounts of computing run ever more and more of these ‘brain in a jar’ sims and not something that would be useful or entertaining to them? Simulating a universe from starting conditions and rules is an excellent way to learn how the rules work and interact, and how changing them affects things, how life in your own universe might have developed, and the like. Simulating a brain and giving it a scenario doesn’t lead to any understanding like that - you’re making the intelligence up from the start, not letting it develop. And you don’t test out any theories of the formation of the cosmos, because you’re just making up what the cosmos looks like, not applying rules and seeing what results.
You’re not going to be able to tell anything about how a particular human reacts unless you have had a chance to scan their brain in detail to make an accurate simulation of it. The processes we’ve come up with for that are destructive, though maybe they’ve got some near-magic ‘learn where every particle is without tearing the brain up’ tech. And if you’re tossing around Jupiter Brains for computation, you’ve got computers that vastly outthink human beings so you’re not learning something useful about a meaningful opposition.
In many religions and faiths there are parallels to simulation theory. In that our world, that which we dwell in, is a illusion of some sort, either constructed by self or set up by a deity or more. In many of those faiths there is a way to ‘break the matrix’ , and see the illusion (simulation) in various ways.
Now if this illusion was created on silicon or some other means I don’t see how it matters, as it still would be a simulation that was created by someone that we are in.
There is the concept of universal laws, also called dharma, in Buddhism. It is the belief that the universe works on some supreme guidelines that govern existence of sentience which is the true basis for the universe and not matter and energy in spacetime. It is sentience that causes the matter/energy spacetime universe as a side effect and not the other was around. As such the universe is both existing and non-existing. So in that already you have simulation theory, and the ability to create it.
In dharma you have the concept of karma, which would govern your thief question, in the rules there will be a net loss and pain to live that model. That loss and pain will guide the soul to seek out relief and a better way.
And the two reasons that pop up in my mind when I hear this are, (1) the fact that it is developed and defended in the traditional style of conspiration theory fanatics, and (2) it does not pass the test of Occam’s razor, in my opinion.
The way I see it, a Jupiter brain would create such simulations as an ordinary byproduct of its thinking. Simulating the future actions and reactions of another living being is part of every thinking being’s normal behaviour; a spider models the future behaviour of its prey, a gazelle models the future behaviour of the cheetah that is chasing it. A human models the behaviour of any other human she wants to trade with, or to mate with, or to murder, and so on.
Jupiter brains, and matrioshka brains, and indeed any large-scale sentient entity, would model entire populations of sentient beings to an arbitrary level of detail, just to predict what they might do in the future. This would be an ordinary, indeed an indispensable, part of their behaviour. Forewarned is forearmed.
Would these gigantic entities bother with modelling the past? I don’t know, but perhaps they would. Modelling the past gives the modeller an opportunity to check the realism of their modelling techniques, since they can check their results against what actually happened. Indeed, I doubt that a sufficiently-advanced thinking entity could even begin to contemplate historical events without modelling it at some level of detail. We all might be the unintended consequence of an intelligent super-object idly contemplating 21st century history.
I disagree. We already model other people’s minds and behaviours in our own minds, whenever we attempt to anticipate their future actions. This process does not include ripping them apart neuron-by-neuron. A sufficiently-advanced thinking being could make an adequately-accurate model of any arbitrary human, just by observing them from the outside.
It’s what historical novelists do all the time. Hilary Mantell didn’t need to submit Thomas Cromwell to a microtome in order to recreate his mind tolerably well; neither would a future Jupiter Brain, if it wanted to recreate the great Coronavirus epidemic of 2020.
Disagree all you want, you’re simply wrong. You’re not actually running a simulation of a person’s mind if you’re not actually simulating what happens, you’re running a model based on guesses. The idea that ‘sufficiently advanced’ intelligences could make models of people’s brains based on observing them or just descriptions that are good enough that they’d be indistinguishable from the real thing is just straight-up fantasy.
It’s pretty typical for this kind of discussion to start off with ‘hey, you might actually be living in a simulation’ and turn into ‘well, you could write a story about a historical character and guess how he’d react to a situation, and that counts as a simulation!’.
Yes, absolutely. All simulations of the past would be based on guesses. I don’t believe in Tipler’s absurd recreation of past lives.
But a simulation based on guesses is the best we can do, it is the best a spider or an antelope can do, and it is the best a Jupiter brain could do. But a simulation made by a Jupiter brain or comparable entity would be entirely convincing to the entities contained within, and would presumably produce useful results (or it wouldn’t be worth doing).
Don’t mistake me for someone who thinks the Simulation theory is a religion, or has any relevance to our current lives and existential condition. As many people have pointed out, we can’t disprove it (at our current level of development). Only by building our own Jupiter Brain(s) will we ever find out it there is any truth to the matter. If we never do that we’ll never know, so we should just carry on as normal.
Solipsists always grossly underestimate the effort it would take to feed a single person a lifetime’s worth of sensory information. When you hear about, for example, scientists discovering that space is big by noticing that it doesn’t behave like a painted dome, the thing feeding you sensory information has to have figured out what scientists studying the sky would see - particularly since you at any time could decide to go check out at book and read more about it. (Or, well, look up.) Every little thing you come across, like say the shirt you’re wearing, it has a huge history behind it. It came from a department store, with employees, and you could go to the store and stalk the employees and learn that they all have fleshed out lives, minds, and histories, and the shirt itself comes from a factory in another country run by more people. Watch the evening news and the ‘simulation’ is scrambling to invent stories all over the world about events and people all over the world - and they have to be consistent with all the past stories. The information feeding in through your senses draws on a huge, and extremely detailed world.
And solipsism posits that whatever is feeding you this information is deliberately coming up with all this information in the most difficult, error-prone way possible.
Suffice to say, entities in a simulation are going to experience their reality in the full detail it provides - if all the trees are cardboard standees, they’re going to notice that sooner or later. If the ‘tree’ simulated objects are composed out of basic molecular data that is composed of basic atomic data executing basic physical interactivity laws, then they’re going to figure that out sooner or later too. Whatever the mechanics of your simulation, eventually they’re going to notice them.
People who actually run simulations know this; they try to create the world as completely as possible. And any failures and shortcuts that they take in doing this, the simulated entites notice. Because the world they exist in is the world they experience, and your shortcuts are their laws of physics.
One way that the amount of simulated data that a universe simulation could be considerably reduced in quantity would be to use the real universe as a source of information. So you don’t have to model the planets of Trappist-1, or the stars in the Leo supercluster; you only have to observe them, and translate those observations into data the simulated entities can use.
Sure, such a simulated universe would be comparatively boring compared to one where every star and planet was imaginary, but those kinds of imaginary universe might turn out to be less useful than reality-based ones.
If the rest of the universe were based on real observations, the only environment you’d have to model in any detail would be the 500 million kilometres of Earth’s surface, plus the ISS and any other spacecraft; this is the only environment humans can experience in any detail. I’d guess that you could model this expanse using a set of processors roughly the same size, using the same amount of energy as falls onto the Earth (if you couldn’t do that, it would be a poor show). Assuming a Dyson Sphere around the Sun at I AU, you could model half a billion Earths.
Of course you’d have to dismantle Jupiter and all the other planets to do this,and incorporate them into the model, but at our current level of observational data I think this could probably be done on a laptop.
And this is the thing with simulation theory. As it progresses it gets deeper and deeper into science fiction. It starts positing requirements and developments for which there is no evidence that they are probable or even possible. If you need to keep adding elements to make something possible, let alone likely or inevitable, then it is a very weak position.
There’s an invisible pink elephant in my living room.
Ok. How come you never run into him?
Oh well, he’s very agile.
Ok. How come he never crushes anything?
Ahh, well you see, that’s additional evidence that’s he is clearly very agile.
Ok. I have a very sensitive gravity meter here. How come there’s no gravity disturbance?
Ahh, well you see, clearly being invisible also prevents or disrupts gravity waves.
And so on, and so forth.
The thing I find the most amusing about simulation theory is the inverse similarity to creationism. One common argument in creationism is that the odds of the universe forming so exactly right to allow us to exists implies a creator! Simulation theory says the odds of the universe being real is so low that it must have been created! 
On the plus side, it also implies that one day, we will be the creators ourselves.
It wouldn’t have to simulate the entire universe. Just the part you can view and interact with. And only to such a level that you observe it. IOW, your “universe” is simply a detailed rendering of the areas you frequent with an appropriate “LOD”. The simulation doesn’t need to render every detail of the Andromeda Galaxy, Mars, or Boise, Iowa (unless you live there).
And I say “you” because once you accept the conceit that your universe has artificially been simulated for you, there is no logical reason to believe anything outside of your perception is “real”. For all you know, all your interactions with other “people”, whether real or online, are really with highly sophisticated bots or avatars.
The argument can even be taken so far that “you” aren’t necessarily “real” either and simply exist as an AI or digital copy of what was you in some sort of Black Mirror style artificial reality.
*Joseph Smith has entered the chat
About as likely as characters in a book writing an actual book on their own.
Surely it’s more likely than that, because the characters in a book aren’t self-aware and don’t possess free-will. I mean, assuming we have free-will at all.
Agreed!
Simulations require a system that is larger than the device simulated. To simulate our solar system would require a computer that has more components than are contained in the current solar system. The presence of such a “computer” would have a serious recursive effect that would distort the target simulation.
To begin design of such a system. we would have to - define the problem. We do not have enough data to define the molecular structure of the solar system or a brain or all of the microbes that are in our gut and on our skin… If we are a simulation we would need to include the simulator as part of the problem definition.
Current numerical computers are just adding machines that can be very big and go very fast, They are impressive tools for investigation of the universe, but they do not simulate the brain, Neural nets perform multilevel regressions that are brain like but they do not ‘simulate’ the brain.
What would be the motivation for our society to create such a simulator? Why would it be funded? There are lot’s of neat ideas proposed every year that go unfunded. Why would a venture capital group back a people simulator - perhaps a pending doll show or they wish to create a slave society to replace labor?
Sometime around my 50th birthday I realized my zeal for implausible theories was a great waste of time - UFOs, Simulation, organized religions, bigfoot, Churchwood’s Continent of Mu, Libertarian politics, lost treasure of Point Reyes etc - all went into the burn before reading file.
Theft - To the extent that I do not steal, I am complying with a social norm, like I don’t pee in elevators.