Could God not just have built a computer with us as disembodied programs?
I was thinking about whether or not this universe is an ‘optimal’ universe design for supporting intelligent life forms. Initially my thoughts were that God would have to evolve life somehow so our universe is therefore pretty good.
Then I thought, maybe if I was God, I’d do something like Tron. Have a virtual world, with immortal ‘programs’ that never experience pain or suffering as the intelligent entities. That would just require me to crack AI.
Then I thought of a few problems:
1. Computers break down. Parts fail. High maintenance.
2. Maybe AI is not true life? Could God embody a computer program with emotions?
3. Who wants to be a disembodied program? Senses please… equals nervous system equals too complexity for God to develop, would needs to be evolved instead.
Any thoughts?
Then I thought, maybe if I was God, I’d do something like Tron. Have a virtual world, with immortal ‘programs’ that never experience pain or suffering as the intelligent entities. That would just require me to crack AI.
Then I thought of a few problems:
1. Computers break down. Parts fail. High maintenance.
2. Maybe AI is not true life? Could God embody a computer program with emotions?
3. Who wants to be a disembodied program? Senses please… equals nervous system equals too complexity for God to develop, would needs to be evolved instead.
Any thoughts?
Comments (30)
A. The technology is possible
B. Beings that powerful want to run programs like that
So plausible but not as likely as some others.
If the universe is some sort of computer, it's a computer operating on principles beyond what we currently regard as a computation. We would need new physics and a new theory of computation to go beyond the theoretical limitations of Turing machines, which is what we currently understand computers to be.
A related and contrary though: Eternal Inflation; is it not brilliant? Creation of a multiverse from nothing. If it were possible to achieve, could God resist himself from doing it?
"Since we're not programs, the question is moot. Most of the phenomena of the universe are noncomputable. An interesting example is determining if a real number is zero. That is not a computable problem. No computer can determine whether a real number is zero. If you can't say whether a given physical value is zero, you can't program the universe."
It's not computations in particular we're interested in, it's simulations. Current computers and programs are capable of doing simulations of the world's climate, groundwater and river flow, and thousands of other things. Of course the computer and programs needed to simulate an entire universe would be much greater. Do you really think a God who could create an entire universe would have trouble with a computer? Also, if she were lazy, she would only have to simulate the portions of the universe which observers have access to. Like back in the day when people thought the "fixed stars were just projections on a sphere outside of the world that we experience.
You're confusing simulations with approximations.
"You're confusing simulations with approximations."
No, I'm not. From Wikipedia:
"Computer simulation is the reproduction of the behavior of a system using a computer to simulate the outcomes of a mathematical model associated with said system. Since they allow to check the reliability of chosen mathematical models, computer simulations have become a useful tool for the mathematical modeling of many natural systems in physics (computational physics), astrophysics, climatology, chemistry, biology and manufacturing, human systems in economics, psychology, social science, health care and engineering. Simulation of a system is represented as the running of the system's model. It can be used to explore and gain new insights into new technology and to estimate the performance of systems too complex for analytical solutions."
Looks circular.
Suppose I program a computer to simulate gravity. Does the computer attract nearby bowling balls? No. A simulation is not reality.
But computer "simulations" of compex phenomena such as weather are approximations. They use techniques of linear algebra to approximate nonlinear phenomena. And as chaos theory tells us, rounding errors in approximations can lead to huge differences in the outcome.
You don't claim that some computer simulation of weather or gravity is how the universe actually works, do you? Aren't mathematical models of physical phenomena historically contingent and subject to revision, as Aristotelian gravity gave way to Newtonian and then Einsteinian gravity?
To put this another way, "simulate" is being equivocated here. In the OP, simulate means to run the universe EXACTLY as it is via a computation. But when we simulate the weather or a biological process for example, we first build a mathematical model that is itself an approximation to reality (as Newtonian and most likely Einsteinian gravity are approximations to reality) and then we FURTHER approximate by substituting linear approximations for the analytically unsolvable differential equations of our model.
The latter kind of simulation is definitely NOT what the OP had in mind when referring to God simulating the universe as a computation.
"You don't claim that some computer simulation of weather or gravity is how the universe actually works, do you? Aren't mathematical models of physical phenomena historically contingent and subject to revision, as Aristotelian gravity gave way to Newtonian and then Einsteinian gravity?"
A sufficiently detailed and complex simulation of a universe would be indistinguishable from a "real" one. I'd go further and say they are the same thing.
I can't make sense of that claim at all.
Consider the three levels of discourse here.
Level 0: Ultimate reality, the true nature of the world. Such a thing might not actually exist ("turtles all the way down") but for sake of discussion let's say it does. So the universe out there has a truth to it about how it works.
Note that any claims at this level are metaphysics and not science.
Level 1: Historically contingent theories about level 0. Aristotle's gravity, Newton's gravity, Einstein's gravity. Mathematical models that provide a descriptive and predictive framework for experiments. These days, they typically take the form of a collection of differential equations without analytic solutions.
Level 2: Computer approximations to Level 1. We can't solve the differential equations of Newtonian or Einsteinian physics, so instead we implement linear approximations on a digital computer to get a few decimal digits of agreement with experiment.
How you can conflate level 2 with level 0, I can't imagine.
A good case study is given by classical Newtonian gravity. This is a perfectly deterministic theory in which if we knew the position and momentum of every particle, we could use Newton's theory to predict all future states.
But it turns out that we can't do that! We can't solve the differential equations of the n-body problem even for the solar system. And when we program approximations to these equations into a digital computer, the rounding errors accumulate and throw the results wildly off. That's due to chaos theory, in which tiny changes in the initial conditions lead to huge changes in the eventual outcome. That's the famous "butterfly effect." This theory is quoted a lot but understood far less. It says that tiny rounding errors in a computer approximation can result in massive errors after a number of iterations of the model.
In fact we don't even know if the solar system is stable under Newtonian gravity, even though the theory is perfectly deterministic.
So how can you say that an inexact approximation to a historically contingent theory can be taken as the same as the actual truth?
https://en.wikipedia.org/wiki/N-body_problem
https://en.wikipedia.org/wiki/Chaos_theory
https://en.wikipedia.org/wiki/Stability_of_the_Solar_System
I'm not a computer guy or a physicist, but I do think you are asking the wrong question. Seems to me that you are talking about something different that I am. I'm not asking to predict future behavior of an existing universe. Seems to me that, in the case we're discussion, simulation is entirely different process than prediction.
The universe on a computer we're discussing is not a model of some other universe outside the computer. It is it's own separate phenomenon. We don't have to compare it's performance with something else to see if we got it right. It is sui generis. Thank you for the opportunity for using fancy words. We set the starting conditions and let it become what it wants.
You could read the links I gave if you chose to. Claiming ignorance isn't much of an argument.
Surely you can understand that an approximation to a historically contingent theory is not to be taken as the same as the underlying reality. A computer simulation of gravity does not attract nearby bowling balls. But worse, a computer simulation of gravity can't even accurately model gravity. Our scientific theories are expressed as differential equations that cannot be exactly solved. They can only be approximated; and the accumulated rounding errors cause the model to be wildly inaccurate.
Quoting T Clark
Yes I understand your point. But you've presented no evidence or argument that the universe is computable.
Honestly when I first suggested that the universe is not computable, I expected half a dozen people to jump in and say, "Oh yeah what about David Deutsch? So there!" You might enjoy looking him up to better understand the argument for your own position.
To me a strong meta-argument against the computable universe is the timeliness of the argument. The ancient Romans built great waterworks, and they thought the mind was a flow. In the Newtonian age we thought the mind was a mechanical device. Now in the computer age we think the mind is a computer. Any metaphysical speculation that claims the world or the mind is made up of the latest technology is suspect for exactly that reason. The theory is as historically contingent as the technology.
You apparently didn't read what I wrote. I said that the issues you describe are not relevant to the argument. The ignorance I refer to has to do with the distinction between a simulation and prediction. I was hoping someone with more knowledge would speak up and clarify.
Quoting fishfry
It's not clear to me what "the universe is computable," means in this or any context. It's not clear to me that it means anything. Here's something from the Wikipedia article on "Digital Physics."
In physics and cosmology, digital physics is a collection of theoretical perspectives based on the premise that the universe is describable by information. It is a form of digital ontology about the physical reality. According to this theory, the universe can be conceived of as either the output of a deterministic or probabilistic computer program, a vast, digital computation device, or mathematically isomorphic to such a device.
I'm not saying this is a useful way of looking at reality, I don't know if it is or not. But apparently somebody thinks it is. My problem with this approach is that its proponents, and the proponents of all the other ways of looking at reality, don't recognize that this way, and all the other ways, are metaphors, stories. They are not right or wrong. They are more or less useful.
Quoting fishfry
That's a pretty silly argument. It's probably an example of some kind of Logical Fallacy. Let me go look...How about cum hoc ergo propter hoc. It's in Latin, so it must be right. All descriptions of the universe are metaphors, stories. Of course every generation uses the metaphors that are in the zeitgeist. Among other things, classical physics talks a lot about forces, movement, power, work. That's the kind of technology that was in the forefront in the 17th, 18th, and 19th centuries. Ties right in with the industrial revolution from the mid-1700s, to mid-1800s.
Of course. But that is not what the OP claims. The OP claims that the world IS a computer. That claim is false for any number of reasons which I'm happy to debate. But now you don't even seem to be making that claim. If you're not talking about the subject of the thread, that explains a lot. You changed the subject evidently. Read the OP's post. OP thinks the world is a computer. I do not think the world is a computer. That's the point I'm discussing. If you're discussing some other point (metaphors) you are in a different conversation.
Quoting T Clark
You have already admitted to not knowing anything about computability. But that's what we're talking about. Could it be that ysomeone with more knowledge IS speaking to you and you don't know enough to recognize that? If one claims the universe is a computer, then the universe is subject to the rules of computability theory as laid down by Turing in 1936 and still operative. There are certain things computers can't do. Many things in fact. Implementing the universe is one of them.
I do acknowledge that the claim in my last sentence is an opinion not universally shared. But that's the subject being discussed. And you appear to lack the requisite knowledge to discuss it.
Thats no problem, I'd be glad to explain it to you. But with your attitude I see no reason to bother.
I've really been looking forward to responding to your smug, condescending post since I read it last night. I'm so excited. I'm trying to hold myself back. I've promised @fdrake I would try not to call people names any more, no matter how tempting it may be. I'll do my best.
Let's see, what are my major points that you have not responded to:
I did indicate I would like some help from someone better with computers than I am, which statement you have attempted to use as a rhetorical bludgeon against me.
Quoting fishfry
This statement is not true. The title of this thread is "Could God not just have built a computer with us as disembodied programs?" The OP speculates on the possibility but doesn't make any claim.
Quoting fishfry
The universe is what it is. At the bottom level it is unspeakable. As my close friend Lao Tzu once wrote "The Tao that can be spoken is not the eternal Tao." All of our descriptions of the universe are stories, metaphors. The standard by which we should judge or choose any particular story is how useful, effective, it is for a particular situation. The world IS not really anything.
Quoting fishfry
This is my favorite part of your post. When I read it I kept saying to myself, "What would fdrake do, what would fdrake do?" I've calmed down now and my heart is at peace.
Some people can dish it out but can't take it.
Quoting T Clark
The OP specifically referenced the idea of a computable universe. If you have a different definition of simulation, please supply it so that we can know what you are talking about.
Of course not all calculation is simulation. But in the context of a discussion on the computable universe, all simulation is calculation, or more precisely computation. The meaning of computation was explicated by Turing in 1936 and nobody's been able to alter or improve on his definition since then.
I propose we end this. We're not getting anywhere. One parting shot:
Quoting fishfry
This is not true. Not even close.
[These and all following boldings mine]
References to computers and programs.
Quoting Devans99
Virtual worlds refer to digital computer simulations.
Quoting Devans99
Computers.
Quoting Devans99
AI. Universally understood to refer to digital computers.
Quoting Devans99
Computer program.
Quoting Devans99
Program.
Quoting T Clark
I'm perfectly willing to allow that I might have misread the OP. But in light of the references I bolded, can you tell me what you think the OP was about, if not digital computer programs as understood by Turing and by all computer scientists ever since?
You say the OP was not about computers. Ok. Then just tell us what you think it was about.
Of course the OP and the entire thread is about computers. You didn't say the OP was about computers, you said :
Quoting fishfry
Which it didn't. Let's say the OP is "I like cake"
OP - I like cake
We all talk about it for a while, then you say:
FF - The OP specifically referenced flour and sugar.
TC - This is not true. Not even close.
FF - Yeah, but you can't make a cake without flour and sugar. What do you think the OP was about?
TC - Cake
Ok. So we agree on that after all.
Quoting fishfry
Ok. Just tell me what you think the post was about.
The point is simply that as soon as one talks about computers, one is constrained by the laws of computability. Some things are computable and some aren't. Newtonian gravity, for example, the classic paradigm of a deterministic, mechanical theory, is known to not be computable. That should make people reflect about their understanding of the limitations of computation.
It's true that some people believe that quantum physics is computable. The jury's still out but we don't know for sure either way. I don't recall reading anything about whether people believe general relativity is computable. There's a guy named David Deutsch who believes that all the physical processes of the universe are computable. I did expect half a dozen people to jump out at me with his name but nobody did. I'd be happy to debate the issue. I think Deutsch is wrong.
So the moment one says, "What if God could create a computer ..." then they are agreeing that God is constrained by the laws of computation, the theory of what is and what isn't computable and what it means for something to be computable. Those constraints are quite severe. I would honesty think that finding out that Newtonian gravity is not computable would give people pause to think.
But you keep saying that I'm misunderstanding the OP. Ok fine. I accept that for sake of discussion. So tell me what you think the post is about.
You did make the statement that you think the actual universe and an inaccurate simulation of a historically contingent model of the universe are "the same." That's the statement you made that I challenged by outlining the three levels of discourse: Metaphysical truth; historically contingent mathematical models, expressed as differential equations that we cannot solve; and computer approximations of the mathematical theory. You claimed that the first and third are "the same," your words. If I misunderstood you, perhaps you could try to explain what it is you meant.
Here's what I think the OP is about:
Quoting Devans99
All in all, your posts have been relevant to the OP. I don't have any problems on that score. My problem is that you haven't defended you position or responded to my points. Also, more than once you've claimed things that just aren't true.
Quoting fishfry
No, what I said was :
Quoting T Clark
And also:
Quoting T Clark
That's not the same as what you claimed I wrote.
Ok well that's a start. I'll read the rest of this tomorrow. The computable universe hypothesis is in the air, put forth by people smarter than me. I disagree with it. So perhaps some of my talking points are more aimed at the general subject than at what you said. I did actually think I was responding to something you said, but I gather you don't see it that way.
Quoting T Clark
For what it's worth, based on how I read your statement, I could not respond any differently than I did before. Given the statement:
Quoting T Clark
I still feel, even in retrospect, that I responded perfectly in kind. But I gather now that you don't think I understood you at all. So we do agree on the point we disagree on. I always regard that as progress.
I wasn't really defending the idea that reality is a computer simulation. I don't find that a useful way of thinking about how things work. Some qualified people with more on the line than me do find it useful. I did try to discuss my understanding of some of the metaphysical implications of such a view.
Someone a few posts back already noted that a perfect simulation of the universe would suffer from infinite regress. If I have a box on my desk that perfectly simulates the universe, then the simulation must itself contain such a box, and so on ad infinitum. This already presents scientific and philosophical problems.
Ah here it is ...
Quoting Banno
Perfectly correct.
But if by simulation you don't mean a simulation on a digital computer, what do you mean? We could play word games and say that the universe perfectly simulates itself, which of course it does. But I think you must mean something more than that. Some subset of the universe must somehow be a perfect simulation of the universe. I'm afraid I haven't got that much imagination, but I'd like you to try to further explain this concept to me.
1) If it's not a computation, what is it?
2) How can a part of the universe simulate the whole?
Well actually that second question has an interesting mathematical analog, since there are bounded subsets of the real number line, namely open intervals, that are topologically identical to the entire line. This is the idea of a continuum. So perhaps a part of a continuum can be identical to the whole. Is this where you're going?
Or perhaps a poetic metaphor, "To see the world in a grain of sand" and so forth.
Quoting T Clark
Uh-oh now I am confused again. I thought I was understanding you to say that you do NOT necessarily restrict a simulation to be a computation, in which case I ask what you mean by simulation.
But if you are trying to understand the metaphysical implications of the computable universe, that's exactly what I've been doing. If the universe is computable, then either it's constrained by Turing's work, which imposes limitations on the laws of physics; or else we need a new definition of computation. That's exactly the point I've been making.
We're not getting anywhere. We both just keep repeating the same things over and over again. I think it's time to give up on this subject for now. No reason you can't keep it going with others.
I made perfectly sensible points that were different from what I said earlier. I'm making an earnest effort to understand your point of view. I'm disappointed that you prefer not to address the direct questions I posed.
I believe you are being earnest. I never doubted it. But I don't think we are any closer to understanding each other.
Now that's not true! But no matter. I posed specific questions, you declined to respond. It's all good.