Slashdot Mirror


No, We Probably Don't Live in a Computer Simulation, Says Physicist (gizmodo.com)

Science doesn't have all the answers. There are plenty of things it may never prove, like whether there's a God. Or whether we're living in a computer simulation, something proposed by Swedish philosopher Nick Bostrom. From an article on Gizmodo: This kind of thinking made at least one person angry, theoretical physicist and science writer Sabine Hossenfelder from the Frankfurt Institute for Advanced Studies in Germany. Last week, she took to her blog Backreactions to vent. It's not the statement "we're living in a simulation" that upsets Hossenfelder. It's the fact that philosophers are making assertions that, if true, should most certainly manifest themselves in our laws of physics. "I'm not saying it's impossible," Hossenfelder told Gizmodo. "But I want to see some backup for this claim." Backup to prove such a claim would require a lot of work and a lot of math, enough to solve some of the most complex problems in theoretical physics.

20 of 418 comments (clear)

  1. In Other Words by MightyMartian · · Score: 5, Interesting

    In other words, "the universe is a simulation" is an unevidenced assertion, much like the multiverse. Yes, there may be some extrapolations of the underlying math that might point in such a direction, but at the moment, it's simply a cool-sounding idea with absolutely no experimental evidence at all. Of course, I feel the same way about string theory, though one thing string theory has produced is some pretty useful mathematical tools, so even when a theory is wrong or indemonstrable, it can still be of some use.

    --
    The world's burning. Moped Jesus spotted on I50. Details at 11.
    1. Re:In Other Words by MightyMartian · · Score: 3, Interesting

      The problem comes when a theory such as this is "abused" as it were to justify a whole bunch of metaphysical claptrap. It's like every New Age fruitcake using the word "quantum" in sentences to make the word salads and bullshit they spew somehow sound "sciency". The fact is that demonstrating the Universe is a simulation is very far out of reach at this stage, and really, as the article makes pretty clear, there's little point to direct inquiry since the problems that need to be solved to make it a viable claim are problems that need to be solved anyways. Unlike String Theory, which has produced some good tools and new conceptual innovations, I don't see any great new tools being produced by simulation theory. It will become evident at whatever point we solve a lot of the big open questions in physics whether the simulation claim makes any sense or not.

      --
      The world's burning. Moped Jesus spotted on I50. Details at 11.
    2. Re:In Other Words by Nutria · · Score: 3, Funny

      But Elon Musk say it's a billion to one likelihood that we live in a simulation! He can't be wrong, can he????

      --
      "I don't know, therefore Aliens" Wafflebox1
    3. Re:In Other Words by doug141 · · Score: 4, Interesting

      In other words, "the universe is a simulation" is an unevidenced assertion, much like the multiverse. Yes, there may be some extrapolations of the underlying math that might point in such a direction, but at the moment, it's simply a cool-sounding idea with absolutely no experimental evidence at all.

      You are wrong about there being no evidence for "the multiverse." There's actually more evidence for the existence of level 1 multiverses than there is for just our one being the only: https://arxiv.org/pdf/astro-ph...

    4. Re:In Other Words by ShanghaiBill · · Score: 3, Interesting

      It isn't even a theory though, as there isn't anything to support it.

      There is no conclusive proof, but there is plenty of evidence that the universe is a simulation. In many ways, the universe appears to be designed to be easy to simulate. If you were designing a universe simulation, what would you do?

      1. Due to limited computational resources, the simulated universe would be granular or "quantum".
      2. To limit computation, reality would be held in a fuzzy probabilistic "superposition" state until it is actually observed, similar to how a GPU running OpenGL will skip the generation of hidden polygons.
      3. The maximum speed of information transfer would be finite, to limit the propagation of changes through the universe.

      All of these are actually true in our universe, ergo, we are very likely a simulation.

    5. Re:In Other Words by FatdogHaiku · · Score: 5, Funny

      It isn't even a theory though, as there isn't anything to support it.

      There is no conclusive proof, but there is plenty of evidence that the universe is a simulation. In many ways, the universe appears to be designed to be easy to simulate. If you were designing a universe simulation, what would you do?

      1. Due to limited computational resources, the simulated universe would be granular or "quantum". 2. To limit computation, reality would be held in a fuzzy probabilistic "superposition" state until it is actually observed, similar to how a GPU running OpenGL will skip the generation of hidden polygons. 3. The maximum speed of information transfer would be finite, to limit the propagation of changes through the universe.

      All of these are actually true in our universe, ergo, we are very likely a simulation.

      The whole shebang may be sitting on a table at a science fair with the label:
      Kinetic Sculpture with Self Aware Components
      By G. Hova

      --
      You have the right to remain sentient. If you give up the right to remain sentient, you will be elected to public office
    6. Re:In Other Words by ShanghaiBill · · Score: 3, Insightful

      Umm no and here's the rebuttal: http://backreaction.blogspot.c...

      That is a weak rebuttal. Basically it says "I don't like this theory therefore it must not be true". A simulation would not have to compute the state of every quark in 13.85B-LY^3 with planck-time granularity. It would just need to compute the sensory input to a single brain.

    7. Re:In Other Words by rgbatduke · · Score: 5, Insightful

      1. Due to limited computational resources, the simulated universe would be granular or "quantum".
      2. To limit computation, reality would be held in a fuzzy probabilistic "superposition" state until it is actually observed, similar to how a GPU running OpenGL will skip the generation of hidden polygons.
      3. The maximum speed of information transfer would be finite, to limit the propagation of changes through the universe.

      All of these are actually true in our universe, ergo, we are very likely a simulation.

      And this, sir, is why you really need to consider taking a course in formal logic and maybe learn about logical fallacies.

      None of these assertions, even if they were true in some useful way, constitute a statistical or logical argument for the conclusion. This is true at an openly embarrassing level. Suppose one were designing a rock because you wanted to build a rock wall and for some reason didn't want to use actual rocks. Due to the cost of raw materials, rocks would be finite in size. Because you don't want the wall to be boring, rocks would come in many different colors, sizes, and shapes. Because you don't want the fake rock wall to fall down, rocks would be solid, as opposed to liquid, glass, plasma, gaseous.

      All real rocks are actually finite in size, come in many different sizes, shapes, and colors, and tend to be solid to the point where "rock solid" is a standard metaphor in human speech. Ergo, all rocks are obviously designed.

      Not!

      Teleological arguments are pure bullshit, which is what the physicist in question (as well as myself, also a physicist) are happy to point out.

      When one actually looks at rocks or Universes, there is an utter lack of either evidence or a plausible, consistent, evidence linked chain of reasoning that increases the probability that the notion/hypothesis "Rocks are designed" or "We are living in a computer simulation" is/are true from their rightful place (so far) of 0.0000.....(0 until you get bored with writing 0's)...001 to something with a tiny smidgen of actual measure.

      These are not independent assertions, by the way. If you take the assertion that the Universe is a simulation seriously, then rocks ARE designed objects, even though there is absolutely nothing about rocks to suggest that they actually are designed.

      One could then deconstruct the truth of each of your statements individually. For example, there is nothing in quantum theory that limits computational requirements -- quite the opposite. Indeed, quantum theory is built on top of complex, non-discrete numbers in every quantum textbook ever written -- C-numbers. That is, quantum objects are described in general by (at least) TWO real numbers, not just one. If you attempt to represent the quantum state of a very simple -- the simplest -- two level quantum system such as |\psi> = A|-> + B|+>, one discovers that it requires two continuous degrees of freedom and that the states of the system map nicely into points on a 3D spherical hypersurface. If you try to describe the most general quantum state of N such 2 level objects, it requires 2^N or so continuous degrees of freedom. Consequently, we are limited in our solutions or simulational studies of fully correlated quantum systems to a tiny, tiny handful of e.g. "two level atoms" -- perhaps 20 to 30 of them -- because one very quickly runs out of computational resources to perform even very small general computations.

      Second, you are building a whole mountain of assumptions into what appears to be a misinterpretation of the Planck length. To quote Wikipedia's page on this topic:

      There is currently no proven physical significance of the Planck length...

      so you are quoting something for which there is no direct evidence as evidence in a bad teleological argument for something for which there is no evidence at all.

      You also don't address the actual numbers associated with the Planck leng

      --
      Even when the experts all agree, they may well be mistaken. --- Bertrand Russell.
    8. Re:In Other Words by squiggleslash · · Score: 3, Interesting

      The counter point, for me, is that's a very complicated way to produce the universe we actually live in. What is the purpose of the simulation? Is it to be similar to Conway's "Life", in which case why build something so convoluted, and why cheat as apparently the programmer did with 2 and 3?

      If not, if the aim was to create sentient beings (well, me at least, I can't speak for you idiots), then, again, why create a system that requires fourteen billion years to actually produce them, with them being around for a mere 50,000, and each having a life span of (almost always) less than 100 years?

      And if you're about to argue the universe was created ten seconds ago, well, no, because there's apparently information in it covering about fourteen billion years. To ensure the system is stable, the logic and current state has to fit that fourteen billion years AND has to be stable right now. One off-by-one error and the Earth will go spiralling into space, or get sucked into the Sun, or just disintegrate, or turn into a black hole for a split second, or...

      I can explain why John Carmack created a number of "virtual reality" (Doom and onwards) games, but he didn't create some overly complex physics model, just the bare minimum to work for the observer. Our universal engineer, however, appears to have created this enormously convoluted system for no apparent reason. I'm not seeing the reason to assume intelligence when a more likely reason for the things you note is that our universe is more complex than you want to believe it is.

      --
      You are not alone. This is not normal. None of this is normal.
  2. The objection ignores Bostrom's basic argument by JoshuaZ · · Score: 5, Insightful

    The objection in question ignores Bostrom's basic argument. Bostrom's primary argument for being in a simulation boils down to the observation that it is very likely that an advanced civilization would have the ability to run very accurate simulations. Moreover, one of the things they'd be obviously interested in would be their own past ancestors; if that's the case, then over the very long period that such civilizations will exist one will expect many more "copies" of people on ancient Earth than any of the originals, unless one expects civilization to die out well before we get to that technology level. If the laws of physics are simulated badly enough that we can notice, then they aren't doing an effective ancestor simulation, so the objection here doesn't make sense.

    There are a lot of issues with Bostrom's argument; for example, one might question whether simulations of that level of detail will ever be able to be made on a large scale. But the argument being made here doesn't grapple with the fundamental issues.

    1. Re:The objection ignores Bostrom's basic argument by WrongMonkey · · Score: 4, Insightful

      it is very likely that an advanced civilization would have the ability to run very accurate simulations

      That is an unfounded assumption. You're taking a mere 50 years of computational progress and extrapolating to infinity. But there are physical limitations to computational density and mathematically intractable problems (like the many-body problem) that don't go away no matter how many iterations of Moore's law that you throw at them. Even simple, well-defined sets of differential equations, like the Navier–Stokes equations, are a struggle to simulate.

    2. Re:The objection ignores Bostrom's basic argument by Billly+Gates · · Score: 3, Insightful

      But in a real universe the laws of physics are different than a virtual programmed one like ours

  3. Nice try by Wraithlyn · · Score: 5, Funny

    This is exactly what an Agent would say.

    --
    "Mind, as manifested by the capacity to make choices, is to some extent present in every electron." -Freeman Dyson
  4. God vs Computer Programmers by WDot · · Score: 5, Insightful

    A previous slashdot commenter (I don't remember which, sorry) said it best. It's apparently ignorant and backwards to believe that God created the universe, but quite forward-thinking and intellectual to think a computer programmer did.

    1. Re:God vs Computer Programmers by __aaclcg7560 · · Score: 4, Funny

      Depends on what programming language that the programmer used to create the universe. That alone would start a few religious wars.

  5. EM drive by Cyberglich · · Score: 4, Insightful

    Personally the fact that recently we have found the EM drive may actually work says were are in a simulation. We just found a bug in the physics engine. That or we need to rethink many of our basic assumptions of physics.

  6. Explain Trump by goombah99 · · Score: 3, Interesting

    If the universe is a simulation then one can speculate on the purpose of the simulation. A good bet, based on our own world, would be it's a role playing game. If so the "players" are presumably the Elites in the game. A logical conclusion for any non-player character, such as yourself would then be that your highest calling in life is to become a groupie. That role is the only role that has any meaning beyond window dressing.

    --
    Some drink at the fountain of knowledge. Others just gargle.
    1. Re:Explain Trump by Gorobei · · Score: 4, Funny

      If the universe is a simulation then one can speculate on the purpose of the simulation. A good bet, based on our own world, would be it's a role playing game. If so the "players" are presumably the Elites in the game.

      Ah, so we just need to look for player characters who picked a generic white-male avatar, blundered around because they picked "easy mode," but still wound up doing and completing some of the fun missions and sidequests in the game. For example:

      1. Started life as a player character with extra gold
      2. Flew fighter planes
      3. Managed a baseball team
      4. Caught fish in the "pond on own private ranch" cliche
      5. Became president of USA. Bonus for second term election.

      Hmm, someone like that would be unbelievable and stick out like a sore thumb.

  7. Uncertainty principle! by Ihlosi · · Score: 3, Insightful
    Heisenbergs uncertainty principle is the observable effect of the simulation running with high, but limited, numerical precision.

    Oh, and it's simulations all the way down.

  8. FFS the summary goes against method by burtosis · · Score: 3, Interesting

    Science does not, nor has ever, "proven" anything. The article linked demonstrates a fundamental lack of understanding as well. The scientific method entails formulating a hypothesis and failing to disprove it, only becoming useful as a theory when it leads to new discovery and understanding.

    There isn't even a need for each particle to have a specific value, a simulation isn't limited to a binary operation in this universe much less any conceivable one. Nor is computational complexity a way to explain it away as there would be no way to tell how much time flows on this side of the simulation with respect to the other from our perspective. All that could ever be accomplished is to show that the simulation would need to be at least of complexity/size X and run for Y operations which at this point would likely correlate with the surface area of the visible universe. Further this would have to come with a caveat that it's only basic rule based and not sentient based as it would be trivial to then overwrite peoples minds/data or start it from any point at which point it generates no useful information, even if correct.

    Tldr pseudoscience