The Matrixpunk esthetic must die

Sorry, I invented a label. It’s to describe a nonsensical fad that I keep running into. It’s like steampunk: romanticizing the Industrial Revolution by putting gears on your top hat, imagining a world run on the power of steam with gleaming brass fittings, rather than coal miners coughing their lungs out or child labor keeping the textile mills running for 16 hours a day, limbs getting mangled in the machinery. Or cyberpunk, a dark gritty world where cyborgs rule and everyone is plugged into their machines, and the corporations own everything, including those neat eyes you bought. Sticking “-punk” on a term implies to me an unrealistic cultural phenomenon in which everyone adopts a faddish esthetic that they think looks cool, but quickly dies out leaving only a relic population that doesn’t realize how deeply uncool they actually are. Try to live on the bleeding edge, discover that the razor moves on fast leaving you lurking on a crusty blood clot.

So…matrixpunk. One movie comes out in 1999, and everyone is wearing trenchcoats, ooohing at deja vu, and talking about how deep it is that we’re just a simulation (and never mind the losers who are gaga over the red pill/blue pill idea — boy, that one sure drew in a lot of pathetic people). It might have been mind-blowing for a few months a score of years ago, but it’s time to move on and recognize that it’s very silly.

However, one of the core ideas that seems to have suckered in some physicists and philosophers is the simulation crap. As a thought experiment, sure, speculate away…it’s when people get carried away and think it might really, really be true that my hackles rise. Apparently, Sabine Hossenfelder thinks likewise.

According to the simulation hypothesis, everything we experience was coded by an intelligent being, and we are part of that computer code. That we live in some kind of computation in and by itself is not unscientific. For all we currently know, the laws of nature are mathematical, so you could say the universe is really just computing those laws. You may find this terminology a little weird, and I would agree, but it’s not controversial. The controversial bit about the simulation hypothesis is that it assumes there is another level of reality where someone or some thing controls what we believe are the laws of nature, or even interferes with those laws.

The belief in an omniscient being that can interfere with the laws of nature, but for some reason remains hidden from us, is a common element of monotheistic religions. But those who believe in the simulation hypothesis argue they arrived at their belief by reason. The philosopher Nick Boström, for example, claims it’s likely that we live in a computer simulation based on an argument that, in a nutshell, goes like this. If there are a) many civilizations, and these civilizations b) build computers that run simulations of conscious beings, then c) there are many more simulated conscious beings than real ones, so you are likely to live in a simulation.

Elon Musk is among those who have bought into it. He too has said “it’s most likely we’re in a simulation.” And even Neil DeGrasse Tyson gave the simulation hypothesis “better than 50-50 odds” of being correct.

Yeah, it’s a bunch of smart people (and a few hucksters) falling for the hammer-nail appeal. I’ve got a dazzlingly good hammer, or steam engine, or computer, and therefore the world must be made of nails, driven the piston of a very big steam engine, all under the control of a master computer. Or, more familiarly among the crackpots I have to deal with, watches are designed and manufactured, therefore the rabbits on that heath must also have been designed and manufactured. But how do you test your supposition? What would look different if the world did not operate analogously to your familiar technology, but was built on different rules? Why, what would it mean if rabbits lacked a boiler and a gear train in their guts?

Hossenfelder does a fine job of taking the whole idea to task. You should read that, not me, but here’s her conclusion.

And that’s my issue with the simulation hypothesis. Those who believe it make, maybe unknowingly, really big assumptions about what natural laws can be reproduced with computer simulations, and they don’t explain how this is supposed to work. But finding alternative explanations that match all our observations to high precision is really difficult. The simulation hypothesis, therefore, just isn’t a serious scientific argument. This doesn’t mean it’s wrong, but it means you’d have to believe it because you have faith, not because you have logic on your side.

Right. I would add that just because you can calculate the trajectory of an object with a computer doesn’t mean its movement is controlled by a computer. Calculable does not equal calculated. The laws of thermodynamics seem to specify the behavior of atoms, for instance, but that does not imply that there is a computer somewhere chugging away to figure out what that carbon atom ought to do next, and creating virtual instantiations of every particle in the universe.

Also, Nick Bostrom is an ass.


  1. Akira MacKenzie says

    THANK YOU! I swear that movie franchise is likely responsible for a huge chunk of the solipsism that rules political and social discourse! “How do we know what real, MAAAAAAN?”

  2. PaulBC says

    I think the esthetic itself is just an offshoot of cyberpunk. It’s unclear if it merits its own name. While Inception may not have the lasting influence of The Matrix it did spawn the “we need to go deeper” memes that capture a similar view.

    (I remember almost nothing about Inception except that I learned that at the center of the human brain, there are soldiers on snowmobiles. I think I may have too literal a take on this.) (I watched it on video and probably should have started earlier in the evening.)

  3. says

    Yeah, this is just some simulation in a grander set of “universal” laws. And what kind of Ur-universe computation does that next-level simulation run on? I get it: it’s turtles all the way down.

    Plus ça change, plus c’est la même chose.

  4. says

    Actually, the biologist Von Bertalanffy noted in the 1930s that ‘systems’ exist in which many phenomena are interdependent and ‘self organize’ through feedback loops which result in the ’emergence’ of things that we see and interpret as individuals. The existence of systems theory has been expanded to explain Cybernetics and Corporate behavior. But I’ve noticed for about 50 years that what I can see as systems effects are totally missed by the average person. One famous expert (Jay Forrester) referred to those effects as ‘counterintutitive’ and another (Myron Tribus) noted that people who think in one paradigm will see and hear explanations of another paradigm but interpret that in terms of their own paradigm and be oblivious to what they are seeing or being told. It is highly likely that the ‘Matrix’ is simply the awareness of system effects without the knowledge of systems theory.

  5. PaulBC says

    Anyone who had actually been reading science fiction before seeing The Matrix would not have found the ideas original. I mean, I thought it was a very entertaining movie, but it’s sad to think that anyone found it life-changing to watch (and yes, the red-pill/blue-pill thing is just tiresome and would be even if it had not been hijacked by the most repulsive people I can think of.)

    I prefer Philip K. Dick’s take in Time Out of Joint (published 1959) in which the faking of reality is a more surreal and the protagonist is kind of a nebbish. It’s just funny and human the way “cool” movies never are.

  6. call me mark says

    Simulation Hypothesis strikes me as similar in impact to Deism. I will concede a small chance that it might be true, but it makes no difference at all to how I live my life.

    Having said that, there’s the idea that if the universe is simulated entirely for the benefit of humanity (with the associated Omphalos hypothesis that the simulation may have started a lot more recently than the apparent age of the universe) then it’s possible that the simulators have had to step in at a number of points; back in 4004 BCE the simulation was just of a flat earth with a domed firmament and every advance in human knowledge since has necessitated a simulation upgrade so that we wouldn’t notice.

  7. hemidactylus says

    Baudrillard himself thought the movie went astray with his ideas. To think a French pomo indirectly influenced the dudebros. Priceless.

  8. R. L. Foster says

    When otherwise intelligent people toy with the idea of a simulated universe they are invariably talking about the simulation as it relates to living, thinking beings like us. They apply the concept to creatures who spend their entire lives on the skins of fragile planets orbiting stars or other large astronomical bodies. But they conveniently forget what the actual universe beyond earth is really like. It’s a vast, violent, radiation soaked place. Who would simulate a universe with gamma ray bursters that can strip a planet of its atmosphere and sterilize it of all life? Or a universe filled with voracious black holes that suck up entire star systems, and presumably advanced civilizations consisting of billions of inhabitants? Even here on lucky little earth the simulation did not have much use for dinosaurs. If the universe is a simulation it certainly does not seem to be designed with the best interests of biological life in mind.

  9. remyporter says

    Tautologically, the universe is a simulation in the same way a car is a scale model of a car (1:1 is a scale).

  10. kome says

    I consider taking the simulation hypothesis seriously to be one of the most clear-cut sign of hubris. To think that one infinitesimally insignificant speck inside the simulation could figure out that its all a simulation is just hilariously arrogant. It’s right up there with a flawed finite being claiming to understand the mind of a perfect infinite being. Yea, no.

  11. microraptor says

    The big thing to remember about the Matrix is that it’s really an allegory for being trans. And all the fanboys who think they’re so smart and are taking the red pill are the biggest fools of all.

  12. PaulBC says

    R. L. Foster@10 The other parts of the universe could just be generated on demand as we discover them. You don’t have to simulate an entire Andromeda galaxy, just a plausible collection of observations accessible to us on earth.

    In fact, I am convinced I am living in a simulation mostly because of new types of animals that just show up unexpectedly. :) I never heard of meerkats until seeing a documentary about them on TV in the late 80s or early 90s when I was in my twenties. Then all of a sudden everyone’s heard of meerkats and there’s even a Disney meerkat.

    Next came the tree climbing goats.

    Just the other day, a Russia-North Korea border was poofed into the simulation. You can’t fool me. I know China is in between! What? Oh, Russia sort of sneaks around east and there’s that little bit. Yeah, right. The simulators are just laying it on thicker and thicker till I finally snap.

    And tree-climbing goats? Those were the dead giveaway. Seriously, tree-climbing goats? Yeah, but y’all can’t see them in real life because they’re “in Morocco.” Kinda like your Canadian girlfriend. I can’t see her even when we both go to Toronto, because she just took a trip to Morocco… to see the tree-climbing goats.

  13. says

    According to the simulation hypothesis, everything we experience was coded by an intelligent being, and we are part of that computer code. That we live in some kind of computation in and by itself is not unscientific.

    According to the simulation hypothesis, there is a probability that we live in a simulation. That probability is based on some extremely dodgy assumptions, namely that there will be extremely large numbers of simulated humans and realities. Meanwhile, the probability that we’re in a reality remains extremely high, because everything we observe is consistent with being in a reality, whereas you have to imagine simulations so precise that they are realities. If you want to talk about probabilities: what is the likelihood that someone can build a simulation so good that the sims within it are intelligent and inquisitive and unable to determine that they are within a simulation? If we want to talk about unlikely probabilities, what’s the likelihood that whoever wrote the simulation we live in had the foresight to add an entire “universe” simulation and a cosmic microwave background simulation that is consistent with the physics simulation?

  14. says

    An underappreciated point: The entire point of trying to build quantum computers is that nature can do things that our computers can’t. The devices on which we base our understanding of what “simulation” means are too limited for physics.

    There’s a massive literature on this topic, starting with von Neumann and continuing with Gleason, Bell, Kochen, Specker … and way, way down the list, you get to people like me. Nowadays, some of us are working on the project of reconstructing quantum mechanics, of finding the cleanest possible expression of how exactly it’s stranger than everything that came before and then building the theory up from that.

    In other words, my actual day job is deriving quantum physics from the premise that Elon Musk is a fucking asshole.

  15. PaulBC says

    Blake Stacey@16 Maybe our quantum-based nature is actually a less powerful computational model than some other physics in which it’s embedded. I mean that’s not something I believe to be true, but when someone says “we live in a simulation” it does not necessarily mean we live in a Turing-equivalent simulation involving a countable set of finite state assignments.

    Alternatively, since we can only measure quantum states approximately, it’s still unclear what’s going on underneath. I have to admit it’ll be kind of annoying to me if quantum computers turn out to be a massive leap over digital computers that I understand much better. It’s unclear how much is fundamentally different and how much is just an opportunity for massive parallelism.

  16. KG says

    Daniel F. Galouye’s 1964 novel Counterfeit World (also published as Simalcron) anticipated The Matrix by several decades. I read it soon after it was published.

    As a designer and user of social simulations, I endorse Hossenfelder’s argument from a somewhat different viewpoint: the sheer amount of data and preprocessing needed to get even a simplistic simulation of some tiny aspect of social reality working (without anything remotely approaching either physics, or self-aware agents) is a daunting obstacle to the plausibility of the idea that our descendants (if any) will run scads of full-scale world simulations. But I think there’s a more fundamental objection. Boström’s argument can’t even get off the ground without the initial assumption that we don’t live in a simulation. If we do, then we can’t deduce anything at all about whether there are many civilizations, or how our own might develop in future – because these things would be up to the whim of the simulator, and might be changed during the course of the simulation. The simulation might simply stop next week, when the simulator’s parent calls them down for tea, or cycle back to the Precambrian; or the moons of Jupiter might hatch giant lifeforms that swoop down and tear the earth apart in a grand end-of-game spectacle; or white mice might be the real intelligence on earth, as aspects/puppets of the simulator, with all human beings except me being NPCs… So even if the rest of the argument was sound, leading to the conclusion that there probably will eventually be many more simulated than non-simulated conscious beings, that would not lead to the conclusion thatr we are probably simulated. The “Simulation Hypothesis” is just a slightly disguised version of Last Thursdayism. It’s not impossible, but there’s no way to assign a meaningful probability to it, and it’s unscientific by design, as it’s compatible with any set of observations whatever.

  17. raven says

    It seems to me that the Matrix simulation fails on two levels.

    .1. Occam’s Razor.
    The simplest explanation is that what we see is what we are.
    The universe we see is our universe.
    The evidence that we live in a simulation is zero right now.

    .2. What difference would it make if we did live in an undetectable simulation anyway?
    If we can’t see it or prove it, it becomes a useless idea.
    How do you tell an invisible god who does nothing and an imaginary god who also does nothing?

    PS The more useful idea that is very likely to be correct is that we are part of a Multiverse.
    Most cosmological models seem to lead to the Multiverse.

  18. consciousness razor says

    Everything is water. Deep down, everybody knows that.

    According to the simulation hypothesis, everything we experience was coded by an intelligent being, and we are part of that computer code. That we live in some kind of computation in and by itself is not unscientific. For all we currently know, the laws of nature are mathematical, so you could say the universe is really just computing those laws. You may find this terminology a little weird, and I would agree, but it’s not controversial.

    First of all, how could they not be representable with set of mathematical expressions? If it’s the case that a particular collection of them doesn’t do it, there is still every other possible set of mathematical expressions to consider.

    So, is there any genuine possibility that not a single of one of them does the job, no matter what the actual world might be like? What the fuck is supposed to make us think that?

    Also, it’s not just that the “terminology is a little weird.” And it is controversial that the laws are anything other than the Humean concept of them, because it’s not the case that anybody has somehow demonstrated that. In other words, you can perfectly well think that they’re true statements (ones that we formulate for our own uses) which unlike many other true statements are especially informative. They’re a relatively small/simple package which tells you a whole lot of true stuff about the world; they summarize the many, many particular facts in it (a.k.a. the “Humean mosaic”). This means the laws are not something extra which exists in addition to the world and make the world behave one way or another.

    You can’t get the simulation hypothesis out of it, without putting in some assumptions like that (i.e., they are in some sense telling the world what it needs to do), regarding what the laws themselves are. All we honestly need is to say that the real laws (whatever those may be) correctly describe the behavior of the real things, not that the laws are somehow something else which is somehow forcing them to do whatever they do.

  19. christoph says

    @ microraptor, # 13: I don’t get that. Nothing wrong with being trans, but The Matrix is more about revolting against our electronic overlords and blowing shit up with really big guns. Oh, and doing kung fu.

  20. lumipuna says

    This is oddly relevant to the topic of “Matrix aesthetic”, but back when the Matrix sequels were coming out and making splash, I saw an absurdist joke in comic strip form that hit me really hard. It’s paraphrased like this:

    Man walks into a blind (ie. eyeless) cave salamander that’s hanging around, looking depressed.

    Man: Well, that’s one depressed looking cave salamander.

    BCS: I too would like to go see the new Matrix film, but I have no eyes!

    Man: produces sunglasses No problem. Just put these on, and nobody will notice!

    BCS: puts on sunglasses So cool!

    (credit to Finnish comic artist Jukka Tilsa)

  21. numerobis says

    imagining a world run on the power of steam with gleaming brass fittings, rather than coal miners coughing their lungs out or child labor

    Most steampunk I’ve seen is cosplay which just takes on the look without much message underneath; or it’s dystopian stories that focus on the coal miners and children.

  22. slithey tove (twas brillig (stevem)) says

    of course, easy explanation for all these [inexplicable phenomena] we encounter, like momentary UFO sightings etc as “glitches in the matrix”. Like when Neo saw a cat repeat itself exactly, Trinity said “it was a glitch in the Matrix”

  23. PaulBC says

    numerobis @24 In fairness, The Difference Engine did get into the grittiness of the Industrial Revolution. And though I am not a Gibson fan and Neuromancer made absolutely no sense to me, maybe Sterling is a moderating influence, because I found the former engaging if not entirely plausible.

    And yes, of course a lot of people are into it for the cosplay. That’s been true of fans since I can remember and I assume so a lot further back.

  24. says

    But surely, if there’s a simulation, it’s necessarily simulating our physics, not some kind of high-level SimCity where everything is high-level objects, which means that the beings running it can’t usefully alter our reality on the macro scale — they could relatively easily pop a meteor into existence and send it careening towards the Earth, but (for example) influencing society as it exists to bring some particular political movement into being isn’t plausible. (You’d have to take a simulation which was running a generic ruleset, and then have it filter through the data — of which there would necessarily be an unbelievably vast amount — and figure out where the clusters of particles were which made up the brains of only the specific people in a particular area, and then you’d have to modify them in such a way that those brains had a different viewpoint than they normally would, but it would have to be consistent with the history of those people, without being a marked departure (because we would notice) and with each viewpoint being customized to the person in question, which would entail, in essence, running a whole extra simulation per person.) Judging from how our own physics simulations work, the utility of such a simulation would be basically nil — you couldn’t be certain in advance that your simulation would develop intelligent life, or that you would be able to detect it if it did, or that you would be able to guarantee conditions you were interested in simulating with that life if you found it. And the resources necessary to run a physics simulation of even a small area (by just hand-waving the existence of things outside it as PaulBC @#14 suggests) would be such that this would have to be a pretty significant undertaking, not just an idle thing to pass the time. Do you remember a few years back when some researchers simulated a fragment of (IIRC) a dead mouse brain, and it took a vast amount of computing power? (There was a post on this blog, if anybody wants to go look it up.) Simulating the entire mouse, alive but in a vaccuum, would have taken a significant chunk of the computing resources on Earth, and even if you assume massive increases in computing power are possible (hint: they realistically aren’t) running a realistic simulation of a few cubic kilometers of a reasonably-densly-filled region of space would require vast resources.

  25. answersingenitals says

    “The philosopher Nick Boström…claims ..there are many more simulated conscious beings than real ones, so you are likely to live in a simulation.”

    Approximately 100 billion humans have existed on planet earth. 93 billion are now dead and 7 billion are living. There is therefore a 93% probability that Bostrom was dead at the time these words were written so it is most probable that these words were never written. We humans have the ability to add N numbers and divide the result by N. Where we screw up is when we decide that the result of that calculation somehow determines reality. American families have 2.4 children on average. So I must have three children, one of whom is paraplegic. I’m going to sneak into the next meeting of the billionaires club and jack up my net worth 10,000 fold. The most egregious misuse of this probableistic argument is when someone states that a certain fact is ‘probably’ true.

    On the other hand, Schrodinger’s cat is probably dead by now.

  26. answersingenitals says

    The Vicar@29: “…but (for example) influencing society as it exists to bring some particular political movement into being isn’t plausible.”

    Or they could just make a few small modifications to the DNA of a few protohumans so that when human societies emerged, most people were ultra conservative (or ultra liberal).

  27. blf says

    The mildly deranged penguin asserts reality is a simulation of this Universe (and a few others). Most of the Universes in the multiverses (plural) are the computational framework for the simulation, most of which are themselves emulators running on simulated Universes. The trick to know yer in a real, not a simulated or emulated Universe, she asserts, is that there are no “peas” or “horses” or “zucchini”, the bugs of the simulation. (Emulators tend to glitch with “egg-plants”.)

    Fortunately for everyone — real, simulated, or emulated — the Unreality Principles show it’s impossible to have a real, simulated, or emulated Universe with all of “peas”, “horses”, “zucchini”, “egg-plants”, and “Trump” (also known as “Johnson” and several other insults) concurrently. Any such Universe is manifestly unstable, and soon Gib Gnabs. A famous Corollary points such theoretical Universes does not have slood. (Theoretically, if such a thing were possible, an almost-but-quite spherical world which moves in an almost-but-not-quite circle around a star and has a “Trump” would not have slood. This is considered so unlikely it’s commonly said to have “turtles all the way down”.)

  28. keinsignal says

    1; Credit to SMBC for the definitive refutation of the Simulation Hypothesis. I don’t have time to dig up the comic but the punchline was something like “When you say the Universe is probably a simulation, you’re making an argument. Most arguments are wrong. Therefore, your argument is wrong.”

    2: Stephen Wolfram has a serious research project in which he assumes the Universe essentially is a gigantic calculation, and tries to determine the nature of it. His initial assumption was that he’d find the Universe to be “computationally irreducible”, that is, you couldn’t simulate it in full on any device less complex than the Universe itself. An interesting twist is that while that may be true for the Universe as a whole, his theory finds that certain apects of his physics are “compressible”, that is, they can be expressed more efficiently than the original calculation that produced them. These regions of compressibility turn out to map onto existing theories of physics, like General and Special Relativity and various bits of QM. So the upshot is that reconciling QM and Relativity may be possible, but wildly impractical — they are in a sense the equivalent of incompatible archive formats.

  29. PaulBC says


    Or they could just make a few small modifications to the DNA of a few protohumans so that when human societies emerged, most people were ultra conservative (or ultra liberal).

    So you’re saying the answers really are in genitals?

  30. says

    PaulBC @ 17,

    Quantum computers don’t really work by parallelism. If they did, then they’d be good at everything, rather than just a class of problems that’s bigger than what classical computers can do efficiently. (The technical statement is that, while nobody knows for sure, the complexity class BQP sure looks smaller than PSPACE.)

  31. nomdeplume says

    “it’s likely that we live in a computer simulation based on an argument that, in a nutshell, goes like this. If there are a) many civilizations, and these civilizations b) build computers that run simulations of conscious beings, then c) there are many more simulated conscious beings than real ones, so you are likely to live in a simulation.” This is just the Kalam argument in another form is it not?

  32. PaulBC says

    Blake Stacey@36 Yeah, I sort of get that. Some time I would like to understand their capabilities better, though I may wait until there are more working examples and I have the time (which may need to be post-retirement unless someone is paying me to know it).

    It still kind of messes up my world if a machine can compute anything with something other than a sequential or parallel algorithm as I understand it and I can’t simulate it efficiently in some very clear way.

  33. hemidactylus says

    IMO M. Night Shyamalan‘s The Village does as much philosophical work as The Matrix, just in a much more mundane way. Both push a character outside the box so to speak.

    The Matrix came out roughly in the same time period as Memento and Fight Club which also had mindfucking reveals. All were philosophically driven. I didn’t take Fight Club seriously enough to pay much attention at the time. I’ve been reading some philosophical analysis that takes it more seriously than I expect would be justified.

    Memento outclasses all.

  34. Skeptical Jackal says

    KG, @18:

    Daniel F. Galouye’s 1964 novel Counterfeit World (also published as Simalcron [sic]) anticipated The Matrix by several decades. I read it soon after it was published.

    I think you mean Simulacron-3. And that novel was the basis of the movie The Thirteenth Floor. (It deviated somewhat from it, but not in a way which destroyed the idea. The reveal appears later in the movie.) That movie made more lasting impact in me than The Matrix.

  35. robert79 says

    The whole “Matrix” simulation idea isn’t even new. (Did you really think a couple of Hollywood movie makers would stumble across an entirely new philosophical idea?) They just popularised it.

    I first ran across the simulation argument as a teenager (a bit before the Matrix came out) when reading “Sophie’s world” ( I don’t remember the exact passages, but I do remember that when Descarte’s “cogito sum” was introduced I read it in that context — If we live in a computer simulation, do we really exist? Yes, I think, therefor I exist!

    So, I think the whole question is largely irrelevant…

    I do think doing kung-fu in trenchcoats is cool though.

  36. hemidactylus says

    This video on Baudrillard and The Matrix is worth watching:

    Back in the day I had thought the movie dovetailed with the way Schopenhauer imposed Vedic philosophy upon Kant. But maybe it’s more cave allegory as was The Village.

  37. John Morales says

    Related concepts: Teilhard de Chardin’s Omega Point, and Roko’s Basilisk.

    (lots of overlap between simulated reality and theology)

  38. unclefrogy says

    I see the idea that it is all a simulation to be related to and kind of backwards to the idea found in Hinduism and Buddhism that we live in a kind of illusion and in fact the idea of “I” is at the heart of the illusion. the simulation just seems to take the ego the idea of a separate ego for granted and lurking within the idea is that you might be able to have power over the simulation. grasping at “magic”
    chop wood carry water
    uncle frogy

  39. Rich Woods says

    @chigau #41:

    Has anyone consulted the white mice?

    No, but they have left a note saying, “So long, and thanks for all the cheese.”

  40. Elladan says

    The basic concept involved in the simulation hypothesis was obvious to Descartes in the 1600’s, specifically whether the reality we experience is real or just some kind of illusion created for us. I’m sure he wasn’t the first. Hell, Plato’s cave allegory hints at it too.

    He even immediately considered the case that simulation people seem to get to only after a bit of thought: OK, is anyone besides me real? Or in “simulation hypothesis” terms, does the universe exist when you’re not looking at it? Are other people thinking beings or NPC’s? If they’re real, does the rest of the universe exist when they’re not looking or is it computed on demand? Etc. Oooh very “deep” stuff.

    It’s all so tiresome. Wait, no, let me descend into a pit of solipsism and solve this: only I am real, these people spouting simulation arguments are simply dreams my mind creates to vex me, and so are you! You’re all fake! Words on a screen that I imagined in this hell of my own making! Noooooooooooooo! gak I am very smart now.

  41. says

    @#31, answersingenitals:

    Even assuming that the connection between genetics and politics were that direct, and that it was somehow easy to make that modification organism-wide on enough organisms to make it normal across a majority of the population, and that mutation and natural selection wouldn’t undo the modifications before they built a civilization, and that a civilization built by people who all artificially thought alike would be viable (which is in itself a big stretch), you’re still talking about threading a needle by dropping the thread over a waterfall when it comes to deriving any actual useful data about the resulting civilization/society — and when the simulation would necessarily be as resource-intensive as it would have to be, it remains amazingly implausible.

    (To put it another way: if you were going to run a simulation of our own world, at the level of physics, to prove some point, even if you started as late as the fall of Rome, if you modified most of humanity to have some particular political bias you still would be unlikely to be able to gain much insight about modern-day politics or society, because the history they enacted would necessarily be different from what happened in our world, you’d have to wait a lot longer than 1500 years to reach 1500 years later inside the simulation — trust me, this is not something which could run in real time — and you would have to sink so much computing power into doing it that it would not be an experiment you’d realistically be able to repeat.)

  42. hemidactylus says

    Devoid of all the computer nerd lingo isn’t simulation something Berkeley captured with his Idealism? Esse is percipi. Would be solipsism without Absolute Mind to shore it up.

  43. fentex says

    The Simulation Hypothesis is just Paley’s Watchmaker dressed in new fashions.

  44. billyum says

    Why does a simulation hypothesis imply a simulator? We already have programs that can write programs.

    And the probabilistic arguments are highly suspect. If there are many more sentient beings within simulations (a big assumption right there) than those who are not, we cannot say that “I” am more likely to be in a simulation. Who is this “I” who could be any sentient being? The only “I” we know is who we are. I have to be me.

  45. drsteve says

    I saw The Matrix in theaters right around the time I turned 17, which was the ideal way to experience it. I remember that the marketing did an excellent job of selling the movie without really tipping off the premise, so there was an exhilirating experience of working out just what was going on by the point Neo woke up in his vat of goo. It’s a bummer that so many people apparently lost sight of appreciating the movie for what it is: a fun, pulpy action movie classic that made use of some philosophy as raw material for its cyberpunk aesthetic. Treating it as a serious philosophical text in its own right seems to me to just miss the point. Also, the focus on the literal manifestation of ‘The Matrix’ crowds out more subtextual readings, like the one mentioned above that sees the story as an expression of trans experience. In grad school 12 or 15 years ago I recall being very tickled by an essay in Science discussing the Agents as a metaphor for the immune system fighting off pathogens.

    Now I have to shout out to this blog for first putting Iain M. Banks on my radar, as I think he had a lot of smart ideas on simulations among other things — one of my favorite bits of prose I’ve read of his so far is a passage in Surface Detail painting a picture of civilizational progress in terms of ever more sophtisticated ways of modeling and simulating, tracing all the way back from theory of mind as manifested in the early hominid brain.

    Anyway, he explicitly used the simulation hypothesis in The Algebraist, as a minor detail of the worldbuilding. As I recall it was the central dogma of a sort of state religion of the relatively unobtrusive variety that everyone was expected to profess belief in, but which had no impact on most people’s lives or, indeed, on the novel’s story. So I suppose that if I were forced to believe in a version of this idea, it would be the one where the simulation is being generated by the mind of a clever Scotsman who has programmed us to just get on with living our own adventures as we would in any case.

  46. DanDare says

    Consider a minimal thing in our universe, maybe a fundamental particle.
    How much information do we need in a simulation of it, and how much of this universes materials are required to store and process that information?
    Any universe is going to be limited in simulating itself to a universe that is degraded in someway due to simulation over head.
    In each simulated universe this holds true, so their own simulation construct will be further degraded.
    There is a bottom. A universe without the resources to simulate even 1 bit of itself.
    But that is not the true limit because the simulation is taking resources from the folks that are running it. If they destroy themselves building the sim then it will not last long. Otherwise it must be even more degraded than the theoretical limit.
    Thus the number of simulated entities vs the number of real ones is likely to be very small indeed.
    Compression allows the reality around the sims to seem real but they can probably spot the compression if they try. Regardless compression is itself a degradation of the sim.

  47. John Morales says

    DanDare, thing is, there’s no need to simulate the entire universe, only the sims and their sensory inputs.

  48. answersingenitals says

    The Vicar@50

    Come now vicar, you are really taking this flight-of-fancy discussion at face value? Once you assume God, you can postulate anything with equal weight. Once you assume this Supercalifragil…. coder/simulator you can postulate anything with equal credibility (and equal incredibility). This whole blog is like discussing Johnathan Swift’s “A Modest Proposal” on a recipe site.

  49. springa73 says

    I think that humans have a long history of drawing analogies between the natural world and their own most advanced technologies, whatever those happen to be. From ancient times when some people conceived of the gods as creating the world and humans and other animals from clay, to the eighteenth and nineteenth century when the universe was thought of as a very finely-made clock or engine, to modern times when some people think of the universe in terms of a computer simulation, there has been a similar pattern. Some people take these analogies too literally.

  50. PaulBC says

    The more I consider the premise that “we live in a simulation” the more I find solipsism even more plausible. Why would “we” live in a simulation? If it’s a simulation, wouldn’t it be a lot easier if I was the only conscious entity being simulated and the rest was just whatever roughly consistent set of stimuli to keep me engaged? Physics as such doesn’t even have to work. I just have to be persuaded that it does to the extent that I understand it at all. That’s a hell of a lot easier than simulating all those people who supposedly know about it.

  51. says

    @#57, John Morales

    DanDare, thing is, there’s no need to simulate the entire universe, only the sims and their sensory inputs.

    That doesn’t work, because the sensory inputs need to be synchronized and consistent with the physics the sims expect, and the easiest way to do that is… to actually simulate the physics of all the stuff in the environment. Unless, of course, your universe’s physical state is procedurally generated… but if it is, then you don’t need to run the simulation at all, just express the procedure with the right time inputs, and the state of the sims will be embedded in it automatically.

    @#59, answersingenitals:

    Come now vicar, you are really taking this flight-of-fancy discussion at face value?

    Since the objections don’t actually require me to do much work to point them out, why not do so? We already had a thread arguing about Elon Musk wanting to send people to Mars even though we don’t have any way of shielding people from radiation once they leave the Earth’s magnetosphere, making the whole enterprise a moot point anyway. If we can waste a zillion comments arguing about whether a Musk-led Martian colony is ethical, there’s no problem with pointing out reasons why a universe-simulator would not be a realistic thing to build.

  52. John Morales says


    That doesn’t work, because the sensory inputs need to be synchronized and consistent with the physics the sims expect

    What a feeble objection; what the sims perceive is “synchronized and consistent with” the physics they have inferred from what the sims have perceived — how could it be otherwise?

  53. John Morales says

    My take on it is that current trends indicate VR technology is only going to get better. For example, I can certainly foresee a functional haptic suit for VR gaming, sometime in the future. Heck, maybe with a scent “printer” for yet another sensory modality. Maybe with, um, customisable codpieces. :)

  54. DanDare says

    @57 John, simulating just the sims and their surrounds is a form of compression. As their perceptions move about the sim must put consistent things in view, so it needs a mod erl of the unobserved. Either this model grows and grows or the sims see the artificial boundaries.
    As I said the sim universe is less than the one it simulates.

  55. John Morales says


    As I said the sim universe is less than the one it simulates.

    Which, according to the simulation hypothesis, means our universe is less than the one where the simulation is running. :)

    I mean, we’re already talking some system that can simulate human-level consciousnesses. Who knows its limits?

  56. voidhawk says

    Most -punk fiction is deliberately about the oppression of people within their world. That’s what the ‘punk’ is all about. The ‘gears on hats’ aesthetic is just that, an aesthetic.

    The only exception to this I know is Solarpunk, which is a deliberately utopian subgenre, with the ‘punks’ in the fiction deliberately fighting to create a more sustainable world (activists taking over a self-sustaining building to house the city’s homeless, people building sustainable raft cities on the flooded ruins of the past, solar-powered robots purifying the land and water, settlers fighting back desertification…

    As for the simulation hypothesis, it’s a bit like Hard Solipsism. Sure, I might be just a brain in a jar, but if the jar is all that I have and there’s no possible way of ever interacting with anything outside the jar, then I might as well spend my efforts on learning how the jar works and making it as nice a place to live as possible. ‘Real’ is relative.

  57. says

    The question of whether reality is real certainly didn’t arise with the Matrix.
    People have been asking that question for thousands of years, it’s only recently that the concept of simulation has been introduced, as we begin building technologies which can run simulations.

    It’s important to remember that the simulation hypothesis does not require that each simulation achieve the same fidelity as its parent. How do we measure the absolute fidelity of our own reality? We don’t. We can’t. The only information available to us comes from within this universe, and if there is another which is causing ours to exist, it would be impossible to know. The limits of computation we measure in this reality can’t be assumed to hold true elsewhere. The hypothesis therefore starts simply with the fact that simulations exist. It is uninterested in their level of complexity.

    As far as steampunk goes, I’ve always understood it as a yearning for a world in which the technology which makes our lives possible is comprehensible to the average person. It’s relatively simple to explain a steam engine. It’s much harder to understand the operation of a computer, and trying to comprehend the systems which operate them is now functionally impossible. Steampunk recognises this and offers an appealing fantasy in which everything modern technology can do is possible with machinery run on simple steam and gears.

  58. says

    So I was a Goth-punk in 1999. Trench coat and everything. My senior year in high school and then COLUMBINE. I was a troubled teen and was already on the list. Then the Matrix and I was cool again. Then the third movie and I couldn’t be bothered. I have no words to explain how fucking weird 1999-2001 was for me. America lost it’s god damned mind.

  59. blf says

    @72, Yeah, I can see how the show could be an early “steampunk” — not so much in the “gears on tophats” fashion vein, but in the steam-powered fantasy tech weird situations vein (with an added does of goofy (as-in played for laughs) implausibility).

  60. snarkrates says

    Probably the most relevant philosophical progenitor of such ideas was the Idealism of Bishop Berkley, although there are echoes in some of the Cynics in Ancient Greece. However, I think the ideas have been given a veneer of respectability by some interpretations of quantum mechanics, where the ultimate reality is the quantum state vector, and this is, after all just information. Some of the quantum physicists were ultimately idealists–Heisenberg might have fallen into this camp. Others, however, seemed to believe in some sort of underlying reality, which we could not perceive directly, and so we had to piece together disparate glimpses based on very different types of experiments–Bohr’s Complementarity, for example.

    Ultimately, the best refutation of Idealism was Johnson’s, who kicked a stone with his foot and limped off saying, “I refute it, thusly!”

  61. says

    My problem with the simulation hypothesis is it doesn’t take into account how the known simulations of reality, i.e. films, novels, video games, Tv shows, scientific models, etc. etc., actually look and feel; the simulations we do use are more stylized and ordered than lived day to day experience. This might sound flippant but I legit think the best response to this argument is to ask Ok where are the credits? Or why on Earth the did the writers of 2020 introduce murder hornets only to drop them at season’s end?

  62. christoph says

    @ microraptor, # 27: I read the article you linked to. If Lilly Wachowski intended to make the Matrix a trans movie, I think she fell a little short. There was one character in the original script that could have been trans, but they edited her out of the movie. I haven’t been following her career, but I hope she’s since made a better trans movie.
    BTW, “Transamerica” and “A Girl Like Me: The Gwen Araujo Story” are both great trans movies. And they have trans people in them.

  63. says

    And speaking of steampunk, here’s a video from Japanese band Fate Gear, featuring a guest appearance from Jill, violinist for the Japanese metal band Unlucky Morpheus.

  64. profpedant says

    I’ll consider the idea that we live in a simulation if we are invaded by alien teddy bears with a 15th century technology who give us hyperdrives and contragravity engines and it turns out that human-compatible life-bearing planets are common across the universe. In short, fantasies are true would be some good evidence for simulation.

  65. Owlmirror says

    I’ve been pondering a counterargument to the simulation hypothesis, but it isn’t quite fleshed out yet. The bones of it have to do with entropy, and economics.

    If we’re in a simulation, what is powering it? How is that power being paid for? Who paid for the design and debugging? Who pays for maintenance and troubleshooting? What keeps the whole thing running, and why? What do the simulators actually get out if it to justify the investment?

    It’s all very well to suggest the idea, but what is so often left out is the whole complicated multicomponent infrastructure and economy and ecosystem that a simulator has to have for it to have come into existence and continue to function in.

    [Needs work, as noted.]

  66. Vreejack says

    The easiest way to simulate an entire Universe is to create an entire Universe. The word “simulation” implies that it is lacking some feature of the real thing; perhaps the physics is approximated, or time must proceed at a much slower pace than the “real” world. In the first case, you have a sucky simulation that works badly, and in the latter case… why? Maybe someone wants to run simulations to try out what-if scenarios. But you have to have some pretty serious resources on hand to make that work.

  67. PaulBC says

    Owlmirror@82 I think the assumption would have to be that the “simulation” is running in a post-scarcity meta-universe if which the simulation of our own universe did not incur a prohibitive cost. For that matter, our laws of “physics” including quantum effects could be easily simulated in their pataphysical supercomputers that run 10^(10^100) miniuniverses like our own without a single floating point error. It seems a little ridiculous to place fixed constraints on beings who are already assumed to be doing something far beyond any available human technology and beyond physics as we know it.

    But like I said, why should I believe they are simulating the whole universe when they could just simulate me sitting on couch typing into my laptop? The “tree-climbing goats of Morocco” already look photoshopped to me. If I were to get it into my simulated brain to visit this so-called “Morocco” then they could either add a quick plugin to their simulation or just use other means to persuade me. (Maybe a “global pandemic” that prevents travel or at least makes their simulatee a lot less interested in it.)

  68. raven says

    I’ll consider the idea that we live in a simulation if we are invaded by alien teddy bears with a 15th century technology who give us hyperdrives and contragravity engines and it turns out that human-compatible life-bearing planets are common across the universe.

    I read that one.
    I even managed to find the title using Google this time around.
    Since it is from 1985.

    The Road Not Taken (short story)
    From Wikipedia, the free encyclopedia Analog November 1985 Author Harry Turtledove

    Plot summary
    The story is told through limited third person point of view, with most of the story concerning a single Roxolani captain, Togram. During a routine journey of conquest, they happen upon Earth. The Roxolani anticipate a simple and rewarding campaign, as they can detect no use of gravity manipulation, the cornerstone of their civilization. Humanity is awed by the invaders, as the maneuverability granted by that technology suggests the rest of their civilization is equally impressive. But as they begin their assault, things take a turn for the absurd—the Roxolani attack with matchlock weapons and black powder explosives. Humans retaliate with automatic weapons and missiles. The battle is short, and most of the invaders are killed. A few are captured alive.

    It turns out the invaders had mastered antigravity, which is a very simple technology that humans missed.
    They had not yet discovered electricity though.

    It’s a good story, find it and read it.

  69. raven says

    Owlmirror@82 I think the assumption would have to be that the “simulation” is running in a post-scarcity meta-universe if which the simulation of our own universe did not incur a prohibitive cost.

    I read a theoretical discussion in Nature a few decades ago, that says in theory, that computation can be either very low cost or even free in energy terms.

    The idea here is that at the end of our universe, at the time of heat death when even the Black holes have evaporated, the post-humans can still eke out an existence in…a simulation.

    IRRC, the question was whether computation by itself had an energy cost or not.
    In theory since ours definitely, does since I’ve got power cords running into my computer.

  70. raven says

    Nature News
    The unavoidable cost of computation revealed
    Physicists have proved that forgetting is the undoing of Maxwell’s demon.

    Philip Ball07 March 2012
    Even if you’re not burning books, destroying information generates heat.

    Forgetting always takes a little energy. Eric Lutz at the University of Augsburg, Germany and his colleagues have found experimental proof of exactly how little. They present their result in Nature today1.

    In 1961, physicist Rolf Landauer argued that to reset one bit of information — say, to set a binary digit to zero in a computer memory regardless of whether it is initially 1 or 0 — must release a certain minimum amount of heat, proportional to the ambient temperature.

    “Erasing information compresses two states into one,” explains Lutz, who is now at the Free University of Berlin. “It is this compression that leads to heat dissipation.”

    His work now seems to confirm that Landauer’s theory was right. “Landauer’s principle has been kicked about by theorists for half a century, but to the best of my knowledge this paper describes the first experimental illustration of it,” says Christopher Jarzynski, a chemical physicist at the University of Maryland in College Park.

    This isn’t it but it is a related question.
    Maxwell’s demon is impossible because it needs to release some energy each cycle.

  71. profpedant says

    #85 There is a ‘sequel’ (which was written first) called “Herbig-Haro”.

  72. PaulBC says

    raven@87 Computation and reversibility has been studied pretty extensively. Fredkin gates are as good a starting point as any. There is really no question that if you assume the laws of physics are reversible, that there is no way to “erase” information without dissipating heat. In fact, what we think of as heat in statistical mechanics (in a reversible particle system) has not really “forgotten” that information at all. You could restore it by running the system in reverse. The “heat” consists of particles whose precise trajectories carry all the information about the original state.

    But if you literally erase it, then the system is not reversible. Models like Fredkin gates or reversible cellular automata simply do a more explicit job of bookkeeping where this “erased” information actually goes.

    I don’t know Landauer’s formulation, but the basic idea should not be in doubt. It’s easily provable (again assuming reversible physics). I think we do a disservice to anyone learning science that we do not emphasize the significance of reversibility, because it really does change the way you have to think about things. Of course, many events appear irreversible and are in practice, but in a classical deterministic system, this is just a matter of having a feasible way to compute it. (And I have no idea whether the actual universe is irreversible, but it does not need to be to see something that looks exactly like a reversible system with heat dissipation.)

    A good popular reference that ties in Maxwell’s demon is The Recursive Universe by William Poundstone. I don’t think he really gets into the question of reversibility as much, but I would have to check again.