I don’t understand how P(Simulation) can be so much higher than P(God) and P(Supernatural). Seems to me that “the stuff going on outside the simulation” would have to be supernatural by definition. The beings that created the simulation would be supernatural intelligent entities who created the universe, aka gods. How do people justify giving lower probabilities for supernatural than for simulation?
At least part of it is that a commonly endorsed local definition of “supernatural” would not necessarily include the beings who created a simulation. Similarly, the definition of “god” around here is frequently tied to that definition of supernatural.
I am not defending those usages here, just observing that they exist.
The word “supernatural” often means “something that is not describable by physics” (ugly definition, I know) or “mental phenomena that is not reducible to non-mental phenomena”. Both definitions are such that it is hard to imagine a world in which there exists something they describe. “Simulation” is, on the other hand, at least imaginable.
A simulator permits interventions that do not follow from the laws of simulated physics, and arising outside the ‘natural’ from the point of view of the simulation, hence supernatural. Likewise, mental phenomena in a simulation may not be reducible to non-mental phenomena within the same simulation. A simulation postulates existence of a specific type of higher domain, super-natural relatively to our nature. And the creators of the simulation are a specific kind of gods.
I think it is a sort of conjunction fallacy, where very specific supernatural and theological beliefs are incorrectly deemed more probable than more general forms of such, because the specific beliefs come with an easy to imagine narrative while the general beliefs leave creation of such narrative as an exercise for the reader. When presented with an abstract, general concept, people are unable to enumerate and sum specific possibilities to assign it the probability consistent with the probabilities they give to individual specific possibilities.
edit: not that I think conjunction fallacy has much to do with conjunctions per se. E.g. if I ask you what is the probability that there is a coin in my pocket, or I ask you what is the probability that there is 1 eurocent coin from 2012 in my pocket, the probability that there is a coin as described may legitimately be higher conditional on me giving a more specific description of the coin.
Mostly, the “not describable by physics” part, as I and maybe many others see it, is a logical impossibility, because physics is what describles real things. Laws of simulated physics can be manipulated, but it will still be within the ‘real’ physics of the real reality. Thus, not supernatural. At least, not in the sense that I understood the question when I answered it.
As for
Likewise, mental phenomena in a simulation may not be reducible to non-mental phenomena within the same simulation.
Mostly, the “not describable by physics” part, as I and maybe many others see it, is a logical impossibility, because physics is what describles real things.
It’s sort of like answering the question about multiverse based on the sophism that multiverse is logically impossible because “universe” is meant to include everything. Clever, but if you’re seeing a “logical impossibility” you probably missed the point.
From wikipedia:
Physics (from Greek φυσική (ἐπιστήμη), i.e. “knowledge of nature”, from φύσις, physis, i.e. “nature”[1][2][3][4][5]) is the natural science that involves the study of matter[6] and its motion through space and time, along with related concepts such as energy and force.[7] More broadly, it is the general analysis of nature, conducted in order to understand how the universe behaves.
In the context of the simulation, it may be impossible for the simulated beings to conduct any form of study of the laws of the parent universe. It definitely has been impossible for us so far, if we are in a simulation.
Furthermore, those who run the simulator can break the internally-deducible laws of physics—e.g. a donut can appear in front of you in the air in a way that is not even theoretically predictable through any studying of nature that you can possibly do. Thus, super-natural, not ever describable by physics as it is defined by dictionary.
Can you expand this one?
Bots in any videogame are not reducible to some contraptions built within the game. Most game worlds do not even allow contraptions complex enough to replicate some bot AI’s behaviour.
As for reducibility in the superior universe, reducibility is sort of like Earth being on a turtle, which is standing on an elephant… eventually you will get down to something that’s not reducible. In our universe, the low level objects that are not further reducible are rather simple (or so it seems), but that needs not be true of the parent universe. Needs not be false, either.
In the context of the simulation, it may be impossible for the simulated beings to conduct any form of study of the laws of the parent universe. It definitely has been impossible for us so far, if we are in a simulation.
Of course, when I say ‘laws of physics’, I don’t mean ‘human study of laws of physics’. I mean the real laws that govern stuff. Even if a donut appears in front of my face, that just means that The Rules say not “physics”, but “what humans know is physics but what is actually arbitrary rules written by beings controlled by their own, this time really physics”.
Anyway, that’s just arguing definitions. The original point goes like that:
You: Why do people assign higher probability to ‘simulation’ than to ‘supernatural’?
Me: I don’t know about other people, but I can say why did I do that, and suppose that I am not the only one. My line of thinking at that moment (sorta):
When I am asked to assign a probability to ‘simulation’, I imagine a world when ‘simulation’ is true (our universe is run on a computer, and I can anticipate stuff like donuts appearing in front of my face or Morpheus texting me about the white rabbit), then I imagine a world where it is not true (our laws of physics are true laws of physics, I cannot ever anticipate any of violations of them), occam-conjure priors for both, see what fits my expierences better, yada yada, and decide on the balance of probability between those two.
When I am asked to assign a probability to ‘supernatural’, I try to imagine the world in which it is true, which means that there happens some stuff that True Rules of the Universe and Everything say can not happen. But if the stuff happens nevertheless, then they are not true. Smells like a logical contradiction, and I wholeheartedly assign it the same probability as I assign to 2+2=5, which, given the restrictions of the test, is equivalent to punching in 0.
So, even if the reasoning is not valid, or if author of the question had another thing completely in mind when he said ‘supernatural’, that’s the explanation why I, personally, assigned a higer probability to ‘simulation’ than to ‘supernatural’. Hope this can give you a hint why the average lesswronger did so.
I don’t understand how P(Simulation) can be so much higher than P(God) and P(Supernatural). Seems to me that “the stuff going on outside the simulation” would have to be supernatural by definition. The beings that created the simulation would be supernatural intelligent entities who created the universe, aka gods. How do people justify giving lower probabilities for supernatural than for simulation?
At least part of it is that a commonly endorsed local definition of “supernatural” would not necessarily include the beings who created a simulation. Similarly, the definition of “god” around here is frequently tied to that definition of supernatural.
I am not defending those usages here, just observing that they exist.
The word “supernatural” often means “something that is not describable by physics” (ugly definition, I know) or “mental phenomena that is not reducible to non-mental phenomena”. Both definitions are such that it is hard to imagine a world in which there exists something they describe. “Simulation” is, on the other hand, at least imaginable.
A simulator permits interventions that do not follow from the laws of simulated physics, and arising outside the ‘natural’ from the point of view of the simulation, hence supernatural. Likewise, mental phenomena in a simulation may not be reducible to non-mental phenomena within the same simulation. A simulation postulates existence of a specific type of higher domain, super-natural relatively to our nature. And the creators of the simulation are a specific kind of gods.
I think it is a sort of conjunction fallacy, where very specific supernatural and theological beliefs are incorrectly deemed more probable than more general forms of such, because the specific beliefs come with an easy to imagine narrative while the general beliefs leave creation of such narrative as an exercise for the reader. When presented with an abstract, general concept, people are unable to enumerate and sum specific possibilities to assign it the probability consistent with the probabilities they give to individual specific possibilities.
edit: not that I think conjunction fallacy has much to do with conjunctions per se. E.g. if I ask you what is the probability that there is a coin in my pocket, or I ask you what is the probability that there is 1 eurocent coin from 2012 in my pocket, the probability that there is a coin as described may legitimately be higher conditional on me giving a more specific description of the coin.
Mostly, the “not describable by physics” part, as I and maybe many others see it, is a logical impossibility, because physics is what describles real things. Laws of simulated physics can be manipulated, but it will still be within the ‘real’ physics of the real reality. Thus, not supernatural. At least, not in the sense that I understood the question when I answered it.
As for
Can you expand this one?
It’s sort of like answering the question about multiverse based on the sophism that multiverse is logically impossible because “universe” is meant to include everything. Clever, but if you’re seeing a “logical impossibility” you probably missed the point.
From wikipedia:
In the context of the simulation, it may be impossible for the simulated beings to conduct any form of study of the laws of the parent universe. It definitely has been impossible for us so far, if we are in a simulation.
Furthermore, those who run the simulator can break the internally-deducible laws of physics—e.g. a donut can appear in front of you in the air in a way that is not even theoretically predictable through any studying of nature that you can possibly do. Thus, super-natural, not ever describable by physics as it is defined by dictionary.
Bots in any videogame are not reducible to some contraptions built within the game. Most game worlds do not even allow contraptions complex enough to replicate some bot AI’s behaviour.
As for reducibility in the superior universe, reducibility is sort of like Earth being on a turtle, which is standing on an elephant… eventually you will get down to something that’s not reducible. In our universe, the low level objects that are not further reducible are rather simple (or so it seems), but that needs not be true of the parent universe. Needs not be false, either.
Of course, when I say ‘laws of physics’, I don’t mean ‘human study of laws of physics’. I mean the real laws that govern stuff. Even if a donut appears in front of my face, that just means that The Rules say not “physics”, but “what humans know is physics but what is actually arbitrary rules written by beings controlled by their own, this time really physics”.
Anyway, that’s just arguing definitions. The original point goes like that:
You: Why do people assign higher probability to ‘simulation’ than to ‘supernatural’?
Me: I don’t know about other people, but I can say why did I do that, and suppose that I am not the only one. My line of thinking at that moment (sorta):
When I am asked to assign a probability to ‘simulation’, I imagine a world when ‘simulation’ is true (our universe is run on a computer, and I can anticipate stuff like donuts appearing in front of my face or Morpheus texting me about the white rabbit), then I imagine a world where it is not true (our laws of physics are true laws of physics, I cannot ever anticipate any of violations of them), occam-conjure priors for both, see what fits my expierences better, yada yada, and decide on the balance of probability between those two.
When I am asked to assign a probability to ‘supernatural’, I try to imagine the world in which it is true, which means that there happens some stuff that True Rules of the Universe and Everything say can not happen. But if the stuff happens nevertheless, then they are not true. Smells like a logical contradiction, and I wholeheartedly assign it the same probability as I assign to 2+2=5, which, given the restrictions of the test, is equivalent to punching in 0.
So, even if the reasoning is not valid, or if author of the question had another thing completely in mind when he said ‘supernatural’, that’s the explanation why I, personally, assigned a higer probability to ‘simulation’ than to ‘supernatural’. Hope this can give you a hint why the average lesswronger did so.
Means tnot describable by the pseudo-phsyics within the simulation.