• Michael
    9.5k
    Some of you may have heard of a new game coming out called No Man's Sky. With a complex algorithm and a single seed an entire universe (264 planets) and everything in it is accounted for (although obviously not generated). It is predetermined what you will find if you fly off into the distance – even though that distant thing doesn't actually "exist" yet. I've found an interesting summary of this here:

    In one sense, because of the game’s procedural design, the entire universe exists at the moment of its creation. In another sense, because the game only renders a player’s immediate surroundings, nothing exists unless there is a human there to witness it.

    The second sentence in particular reminded me of the Copenhagen interpretation, and it made me wonder whether or not the universe we live in is itself procedurally generated, or something similar. Perhaps the objectivity of our world is limited to an algorithm of sorts (one that allows for randomness to account for quantum mechanics) and some seed value but that the actual "rendering" is dependent on conscious interaction of some sort.

    So for discussion, do you think that a No Man's Sky-like real world is metaphysically possible, and if so is it possible for such a world to behave exactly as our world behaves?

    Personally I think at first glance it's an elegant union of realism and idealism, gaining from their respective strong points and accounting for their respective weak points.
  • tom
    1.5k
    It's not possible to account for quantum mechanics with randomness.
  • darthbarracuda
    3.2k
    It reminds me of Meillassoux's correlationism. I'm in the middle of reading his After Finitude, so I'll have to post some more later, but Meillassoux definitely has ideas about digitalization, ancestrality, and what have you.
  • Janus
    10k


    The analogy would only hold if the universe is a computer program. That begs the question as to what does it run on or in, and who wrote it.

    One answer might be that the physical world is a program running in the spiritual world; which would be in accordance with the Buddhist idea that the the world is created by collective Karma.
  • apokrisis
    5.1k
    With a complex algorithm and a single seed an entire universe (264 planets) and everything in it is accounted for (although obviously not generated). It is predetermined what you will find if you fly off into the distance – even though that distant thing doesn't actually "exist" yet.Michael

    Interesting article. But it does say that further random seeds are being generated all the time. Now it then tries to backtrack and argue that this is still deterministic because the original seed is the source of the pseudo-random generation of new seeds. And yet - if the principle of indifference does apply in the program - then there is a constant generation of randomness, for all practical purposes.

    Plus of course there is the human making decisions when playing the game. So the gamer is outside the "procedure" as the critical source of unpredictability.

    So the model kind of captures reality in imagining a few basic laws that get kicked along by randomised variables being plug in to generate the actual dynamics.

    But there are big chunks of reality missing - like entropy. There is no energy cost, speed of light restrictions, decay, and much else in the model. So the model represents a very thin view of physical structure.

    The clunky level of randomness and interaction being modeled is illustrated by...

    Minor adjustments to the source code can cause mountains to unexpectedly turn into lakes, species to mutate, or objects to lose the property of collision and plummet to the center of a planet. “Something as simple as altering the color of a creature,” Murray noted, “can cause the water level to rise.”

    So the major thing is that this world is driven from a single point of view and does not reflect a god's eye coherent point of view. It generates "more world" from wherever you have got to in the game's structure - so incorporates a local personal history. But the real world incorporates a global history in generating its every next step.

    That is what makes it possible for there to be chameleons that change their colour without it destablising plate tectonics. The constraints that form the real world have a hierarchical organisation which makes it meaningfull whether we are talking about universal conditions baked in generally from the first moment of the Big Bang or instead the very arbitrary choices a human is free to make when deciding whether to shoot or sit back and watch computer generated aliens on imaginary planets.
  • Michael
    9.5k
    The analogy would only hold if the universe is a computer program. That begs the question as to what does it run on or in, and who wrote it. — John

    I don't think it needs to have been "written" by anyone. Who wrote the rule that says when two particles meet they interact in this sort of way? Unless you're a theist, you'd just say that this is just how things behave. And so by the same token, one might say that this just is the sort of thing that is rendered during an observation.

    And you might as well ask what an electron "runs on". There might be some pre-electron architecture on which an electron is run, and so there might be some pre-rendering architecture on which a tree is rendered (when observed). Or it might be that the electron is fundamental, and so it might be that the rendered tree (when observed) is fundamental.

    The analogy isn't meant to suggest that a real world might be exactly like a computer program. It's meant to suggest that a real world might behave in the same sort of way as a computer program where the objects and events that we see in it only exist and occur during the observation (the rendering) and need not also exist and occur prior to the observation. The regularity (and intersubjectivity) of it can be explained without requiring realist metaphysics (where the things that are seen to exist and occur also exist and occur when not seen).

    You wouldn't say of a multiplayer game "if you both turn your characters around then you'll see a tree, therefore there must really be a tree behind your characters" and so you need not say of the real world "if you both turn around then you'll see a tree, therefore there must really be a tree behind you". Unless for there to really be a tree behind you (or your characters) just is that you would see one if you (or your characters) turned around (so rather than the former explaining the latter, the two are the same), but then unperceived existence is a counterfactual experience that obtains even under idealism, and so this wouldn't amount to realism.
  • Michael
    9.5k
    So the major thing is that this world is driven from a single point of view and does not reflect a god's eye coherent point of view. It generates "more world" from wherever you have got to in the game's structure - so incorporates a local personal history. But the real world incorporates a global history in generating its every next step. — apokrisis

    This is what I'm questioning. Is it necessary that a real world incorporates a global history? Or is it metaphysically possible for a real world to behave in the same sort of way where "more world" is generated from a single point of view?

    And if it's metaphysically possible, is it possible for such a world to be empirically indistinguishable from our world, or would there be some observable difference that we use to determine whether or not our world is such a world?
  • apokrisis
    5.1k
    This is what I'm questioning. Is it necessary that a real world incorporates a global history? Or is it metaphysically possible for a real world to behave in the same sort of way where "more world" is generated from a single point of view?

    So it seems possible if your existence is just a computer simulation. But if this about a material reality, what could generate the level of accumulated material history that we observe. Why would a real world have to fake a history - like you dig in the ground and find the mineralised bones of dinosaurs?

    Sure you might say some magical process just generates stuff as far as your eye can see. But why would it fake layers of entropic history rather than just generate shiny new untouched stuff?

    We account for our world in the way it actually seems to be - which is historically conditioned. If this alternative way of making worlds is always making new stuff, it adds considerable implausibility that it would go to the trouble of make the new stuff look "suitably old".
  • Michael
    9.5k
    Why would a real world have to fake a history - like you dig in the ground and find the mineralised bones of dinosaurs?apokrisis

    I didn't mean to suggest that a real world would have a fake history. I meant to suggest that (in this hypothetical world) the bones we see when we dig don't exist before we see them; instead what "exists" is the "function" that determines that when we dig in a certain spot we will see bones.
  • apokrisis
    5.1k
    I didn't mean to suggest that a real world would have a fake history. I meant to suggest that (in this hypothetical world) the bones we see when we dig don't exist before we see them; instead what "exists" is the "function" that determines that when we dig in a certain spot we will see bones.

    How is making weathered dinosaur bones not faking an entropic history?

    Sure I accept that your argument is that the bones are only made at the very instant they need to be struck by our spade. But it is the same issue as for creationists. Why go to the extra trouble of building in the look of a history if this is an essentially a-historic world creating process?

    It seems illogical for such a world to have a reason to create a look of history in contradiction of what it actually is as a "create it as you go" kind of world.
  • Janus
    10k


    But that "function" would have to be an elaborate program. How could it be plausible that such a thing, posited to be analogous with computer programs that we write only infinitely more complex, exist without having been written?
  • Michael
    9.5k
    You might as well ask how the rules that govern how the material world behaves under metaphysical realism can exist without having been written.

    It's just a fact that the world behaves in this way. Whether you want to say that the things we see exist when we don't see them or that the things we see only exist when we see them but that there is some underlying architecture that determines what we see doesn't make a difference on this point.
  • Michael
    9.5k
    You seem to be suggesting that a history is only real if it satisfies realist metaphysics. But that's wrong. A real history is whatever history actually happens. If the history that actually happens is one of a function with a variable time seed that determines what someone sees then that is the real history, and so wouldn't be fake.

    All you're really saying is that this hypothetical world (and its history) isn't a realist one. I know that. What I'm asking is whether such a world is metaphysically possible and whether it would be empirically indistinguishable from our world.
  • apokrisis
    5.1k
    What I'm asking is whether such a world is metaphysically possible and whether it would be empirically indistinguishable from our world.

    Well my point - again - is that the existence of a historical aspect to this reality would be one kind of empirical evidence against it being the case.

    So I am granting your indistinguishability claim - we can't tell if things were always there or generated as we go along.

    But then given that, if this world has a look of history, then that still counts as empirical evidence against it being generated.

    Given two options - a metaphysics that is consistent with what we experience, and one that would be in contradiction - then we have a reason to prefer the consistent story. And that is the one where the appearance of history is evidence of actual history (just as a lack of apparent history would be evidence in favour of a "generate as you go" ontology).
  • Michael
    9.5k
    Well my point - again - is that the existence of a historical aspect to this reality would be one kind of empirical evidence against it being the case.

    So I am granting your indistinguishability claim - we can't tell if things were always there or generated as we go along.

    But then given that, if this world has a look of history, then that still counts as empirical evidence against it being generated.

    Given two options - a metaphysics that is consistent with what we experience, and one that would be in contradiction - then we have a reason to prefer the consistent story. And that is the one where the appearance of history is evidence of actual history (just as a lack of apparent history would be evidence in favour of a "generate as you go" ontology).
    — apokrisis

    But the "look of history" is also given in a computer game like No Man's Sky. You see something move towards your character, you turn your character around, wait a while, turn it back, and the thing has reached your character. It doesn't then mean that there was a thing moving towards your character (in the realist sense) when your character's back was turned.

    You can't use something as evidence of A rather than B if that thing is also entailed by B. So given that this "look of history" is entailed by a first-person procedurally generated universe it can't be evidence of a realist world.
  • apokrisis
    5.1k
    Just address my actual argument. Would a generated world more naturally generate what is contemporary or what is historic?
  • Michael
    9.5k
    That depends on the underlying procedure. If it's one with an algorithm that generates more contemporary stuff than historic stuff then it's more likely to generate contemporary stuff, otherwise it's more likely to generate historic stuff.
  • apokrisis
    5.1k
    So you just won't answer my question about naturalness and the expectations we can rightfully derive from such an assumption. Bad faith.
  • Michael
    9.5k
    I did answer your question. But your question is like asking "would an equation more naturally give a negative integer or a positive integer?". The only possible answer is "depends on what the equation is".
  • apokrisis
    5.1k
    Bullshit. My question was not about some arbitrary programmer's choice but about what would be most self consistent.

    And when equations produce positive and negative roots, we know to throw the negative away when it can't make physical sense. So yeah, a naturalness test does get applied.
  • Janus
    10k


    You're missing the point, which is that you are wanting to draw an analogy between reality and a computer program, but unwilling to acknowledge that the most significant thing about a computer program is that it has been intentionally programmed to be the way it is.

    Algorithms in computer programs are created by programmers. If you posit there might be an algorithm that determines that when we dug somewhere we find fossils, implying that those fossils in no way existed prior to our digging, then your position requires an explanation for the existence of the algorithm.
  • The Great Whatever
    2.2k
    Yeah, I definitely think some of the classical transcendental idealists could be read as claiming that the universe is procedurally generated based on a starting facticity (roughly, the 'program,' the 'thing in itself') combined with a generation of the empirical world 'on the fly' as the knower's faculties come into contact with it.

    Though these questions were already effectively raised by older video games. In a 1980's side-scroller, is what is off the right side of the screen 'there?' Does it 'exist?' When talking about the game, we talk as if it does, but also recognize a sense in which it doesn't, in which it's nothing more than a constellation of moving pixels programmed to appear depending on certain game states arrived at by certain inputs. There is still a procedure of generation, it's just far less variable because far more spelled out. What you can take the old transcendental idealists to be saying is that roughly, life is like a video game in this way. The sense in which the unseen world is 'there' is the sense in which the material off the right side of the screen is 'there.
  • darthbarracuda
    3.2k
    How would the transcendental idealist or simulationist deal with the fact that accidents happen all the time? I could get struck by lightning and never see the bolt before it kills me.

    Of course it's not a big deal if you accept a kind of primary vs secondary quality distinction. The bolt of lightning would have primary qualities regardless of whether or not I perceive the bolt.

    But transcendental idealism seems to argue that what we do not perceive is in some kind of "proto" state, or a state of pure potentiality and no effable characteristics.
  • apokrisis
    5.1k
    Yeah, I definitely think some of the classical transcendental idealists could be read as claiming that the universe is procedurally generated based on a starting facticity (roughly, the 'program,' the 'thing in itself') combined with a generation of the empirical world 'on the fly' as the knower's faculties come into contact with it.The Great Whatever

    And so as I suggest, the kind of world that does unfold before our questing gaze ought to be generated in some kind of accordance with our wishes. This becomes an empirical test (even if a reasonably modest one) of idealism as a metaphysical hypothesis.

    But I guess that is simply standard. Realism is supported by the world's recalcitrant being. If we kick a stone, our toe still hurts, even if it seems more logical that it shouldn't if our preferences actually ruled.

    So I wonder what kind of naturalism could explain the world that the idealist encounters? Why - at some deep level of metaphysical reasonableness - is this the world we generate?

    And the failure of idealism to deliver a reasonable account on this point would be further reasonable evidence against idealism.
  • The Great Whatever
    2.2k
    And so as I suggest, the kind of world that does unfold before our questing gaze ought to be generated in some kind of accordance with our wishes. This becomes an empirical test (even if a reasonably modest one) of idealism as a metaphysical hypothesis.apokrisis

    As far as I can see this doesn't follow or support realism. Idealism has never claimed, to my knowledge, that whatever you want to happen happens, nor do I see what would ever commit it to that.

    How would the transcendental idealist or simulationist deal with the fact that accidents happen all the time? I could get struck by lightning and never see the bolt before it kills me.darthbarracuda

    The same way that your character can die, even if you don't want them to. As for it all ending, that can either happen on the game's terms (you lose a life and black out), or outside of the game's terms (the console or CPU crashes).
  • darthbarracuda
    3.2k
    From what I can tell, it's that idealism accounts for accidents only because what happens outside of our perceptions is dependent upon what happens within our perception. The unperceived is still given, but it is not given to the subject. idk this shit's confusing.
  • apokrisis
    5.1k
    As far as I can see this doesn't follow or support realism. Idealism has never claimed, to my knowledge, that whatever you want to happen happens, nor do I see what would ever commit it to that.The Great Whatever

    And yet it follows that if the world is truly generated from a personal viewpoint, then there has to be some reasonable account of why that isn't the case.

    Realism justifies itself on the grounds of recalcitrant nature. It is because the world shows no sign of coming from our point of view that we should believe it to be most likely real.
  • The Great Whatever
    2.2k
    And yet it follows that if the world is truly generated from a personal viewpoint, then there has to be some reasonable account of why that isn't the case.apokrisis

    Why does that follow?

    There's clearly a massive missing premise in this argument, which once you spell out I suspect you'll see is false.
  • The Great Whatever
    2.2k
    It's your proposed argument, so I can't do that for you. You have to tell me -- clearly there is no validly drawn conclusion as it stands, and I have no idea what's going on in your head that makes you think there is.
  • apokrisis
    5.1k
    What are you talking about? The grounding premise was that consistency would be a good thing. If your counter argument relies on a malicious deceiver - the faking of ancient dinosaur bones - then already it is weaker because of a lack of explanatory parsimony.

    I agree that such arguments are avoided like hell by ontic idealists of course. :-}
bold
italic
underline
strike
code
quote
ulist
image
url
mention
reveal
youtube
tweet
Add a Comment