Historical Simulations

   

My new paper on historical simulations has finally appeared in print in the August issue of the Journal of Futures Studies. George Dvorsky has posted some comments about it on Sentient Developments, where it spawned an interesting little thread on the stacked simulation issue, (simulations within simulations) involving Robert Bradbury and others. George quoted an earlier passage (written prior to my paper) from Bradbury stating that simulated realities will run slower than the host reality, therefore allowing at least "trillions of trillions of trillions" of simulations and therefore the universe is unlikely to run out of computing power due to stacked simulations. 

This quote from Bradbury also indicated that if we live in a simulation, then we cannot extrapolate the size of the host or "basement" universe from examining the size of the simulated universe that we inhabit, since future societies might simulate "weird" universes with different physical laws and dimensions than the real one they occupy. George thus concludes that it is arbitrary and unfounded to conclude that our simulation will terminate around 2050 when we start on a widespread basis to create historical simulations of our own. There are several problems with George's concerns:

1) He ignores the point in the paper that there is another reason for the simulation to be unplugged when it reaches 2050, i.e. that a simulation depicting a period after that date would lose the element of deception of its inhabitants necessary to produce genuine behavior for observational and experimental purposes. Robert Bradbury's comments in the thread discussing my paper even concede that it is possible that we could be in a simulation designed "to study evolutionary paths of advanced civilizations up to the point where they begin to simulate the evolutionary paths of advanced civilizations" in which case we may be indefinitely "suspended" (nice euphemism for "terminated") at that point.

2) George's quote states that due to the simulations running much more slowly than the host reality, the universe would be capable of at least "trillions of trillions of trillions of simulations". However, this is merely 10³⁶ simulations, which still doesn't come anywhere close to the 10⁹⁶simulations that would be produced after only 16 generations of stacking starting with an initial one million simulations.

World of Warcraft currently has at least 6 million subscribers, so it is easy to project a starting point of at least 1 million fully realistic simulated worlds by 2050. In any event, Robert concedes in his comments that if the hardware is intentionally engineered to facilitate simulations, (e.g. a nanotechnology or femtotechnology graphics card!) it will run at close to the speeds of the reality being simulated. Presumably if a future society wants to create simulations, then it will dedicate special technology to the task and won't use a generic form of hardware. Robert also notes in his comments that the simulation could develop abstractions (shortcuts or heuristics) that would enable computational speedups which exceed the reduction in speed inherent in the simulation.

3) In his comments on the paper, Robert also notes that there is the issue of whether the abstractions or shortcuts developed in the simulation can be translated back into the reality that produced the simulation. If the rules in the simulation are vastly different, (e.g. the "weird universe"), then extracting the inventions of the simulation may be impossible. Robert then wonders whether the "gods" would intentionally design simulations from which extraction is difficult or impossible. I think that the answer to this question must be in the negative. If a future society goes to the trouble of creating a historical simulation, it must be for the purpose of extracting information from it. If we are in a simulation now (which we probably are), then it is likely not a depiction of a "weird" universe that is fundamentally different from that of the host reality. We can therefore extrapolate from observing the simulated universe that we inhabit what the total computational power of the host universe is likely to be.

I am looking forward to further comments. The stacked simulation problem, which was only a small part of my Historical Simulations paper, could easily justify a paper of its own.

Comments

[Ryan]

I'm not quite convinced of this theory. Correct me if I'm wrong, but you're saying basically that because we can imagine creating a historical simulation, it is thus likely that we are one? I don't see much difference between that and the argument that because one can imagine a perfect or greatest being (god), it must exist.

Another thing is, what makes the 'host' civilization different from us? That is, from their perspective, aren't they likely a simulation as well? Even if you work your way up to the one civ. which really *isn't* a sim., how would they themselves possibly know? Presumably there is no way to test for this, your host will either allow you to know or not. So even the real hosts would have to assume they are simulations.

[Peter S.]

Ryan: I think that he's saying what makes the host civilization different from us is that they are the ones holding the cards, so to speak. If they were the first ones to reach the simulation stage of civilization, they are the ones watching all of the simulations and deciding when to end them.

I'm not certain that suspension and termination are the same. Wouldn't it make more sense to suspend a historical simulation and resume it when the civilization which has created it reached a new level of understanding above the historical simulation point?

I do that all the time while playing Total War games.

It seems like the simulation (read: us) would never notice a thing, but would continue to indefinitely simulate a period one or two eras behind the creating civilization.

[Ryan]

Well, I get that point. I just wonder how anyone would conclude that they are the first to reach the simulation stage of technology.

Simply running your own simulations wouldn't seem to be evidence that you are real. Even now for example in The Sims, your characters within the game can play computer games themselves. So those running the simulation we inhabit could easily be a simulation themselves, and so on.

[Peter S.]

I don't think that he means that the sole criterion for a 'real' society is the ability to run your own simulations. From the abstract, 'A future society will very likely have the technological ability and the motivation to create large numbers of completely realistic historical simulations and be able to overcome any ethical and legal obstacles to doing so.

It is thus highly probable that we are a form of artificial intelligence inhabiting one of these simulations.'(line break mine)

The conclusion appears to be that it's more likely than not that we are simply a simulation being run by a more advanced society, not that the ability to run such simulations is what makes a reality real.

And from those premises, he's saying that to avoid the stacking of simulations, i.e. your sims example to the extreme, the most advanced civilizations cut off historical simulations which reach the point of simulation. And if all civilizations follow this logic, then by definition any civilization which passes the accurate historical simulation threshold is real.

Please correct me if I'm wrong, but that's how I interpreted the abstract.

What I don't understand is why you would cut off the simulation at the point where Super-accurate historical simulation becomes feasible. I see no difference in principle between an extremely accurate simulation of historical events and a book detailing historical events.

While one is a much more thorough version of the other, the principle is the same. Not only that, but you would need to watch the simulation from a perspective, giving a watcher the same biases as anyone simply reading notes from that perspective.

And while it seems like a huge devotion of resources for us to imagine accurate historical simulation, i'm sure the writing of the first history texts was also perceived as a mammoth undertaking. Histories of the world are still mammoth undertakings. And yet, we write history books about history books. Nobody seems to worry that at some point a God is going to end them for writing those. Maybe they should, according to your premises.

If a simulation of historical events is your goal and you're worried about stacking, you should cut off the simulation at the point where writing arises in your simulatees, not super-simulation. A Simulation is just a book with obscene amounts of detail.

Perhaps an oral tradition of history brings instant obliteration? Heck, any brain function allowing memory is risky.

[Peter S.]

I don't think that he means that the sole criterion for a 'real' society is the ability to run your own simulations. From the abstract, 'A future society will very likely have the technological ability and the motivation to create large numbers of completely realistic historical simulations and be able to overcome any ethical and legal obstacles to doing so.

It is thus highly probable that we are a form of artificial intelligence inhabiting one of these simulations.'(line break mine)

The conclusion appears to be that it's more likely than not that we are simply a simulation being run by a more advanced society, not that the ability to run such simulations is what makes a reality real.

And from those premises, he's saying that to avoid the stacking of simulations, i.e. your sims example to the extreme, the most advanced civilizations cut off historical simulations which reach the point of simulation. And if all civilizations follow this logic, then by definition any civilization which passes the accurate historical simulation threshold is real.

Please correct me if I'm wrong, but that's how I interpreted the abstract.

What I don't understand is why you would cut off the simulation at the point where Super-accurate historical simulation becomes feasible. I see no difference in principle between an extremely accurate simulation of historical events and a book detailing historical events.

While one is a much more thorough version of the other, the principle is the same. Not only that, but you would need to watch the simulation from a perspective, giving a watcher the same biases as anyone simply reading notes from that perspective.

And while it seems like a huge devotion of resources for us to imagine accurate historical simulation, i'm sure the writing of the first history texts was also perceived as a mammoth undertaking. Histories of the world are still mammoth undertakings. And yet, we write history books about history books. Nobody seems to worry that at some point a God is going to end them for writing those. Maybe they should, according to your premises.

If a simulation of historical events is your goal and you're worried about stacking, you should cut off the simulation at the point where writing arises in your simulatees, not super-simulation. A Simulation is just a book with obscene amounts of detail.

Perhaps an oral tradition of history brings instant obliteration? Heck, any brain function allowing memory is risky. All of these require (compared to their contemporary actions) immense amounts of resources, and yet seem paltry now.

I just don't get why that specific point would be so darn horrible that no simulation will ever be allowed past it. Sure, it'd be pretty bad for a civilization that had just gotten to that point (She Can' Take Any More, Cap'm!), but a few decades and it should be easy peasy.

[Peter S.]

not only was that a ridiculously long post, i did it twice. My bad.

[Giulio Prisco]

As many of those who posted comments, I agree that we may well be living in a simulation running on some supercomputer in "a higher level of reality". But I don't think we have enough information to assign any probability to this possibility, and I don't agree with the conclusion that the simulation would probably be terminated as soon as its conscious inhabitants develop the capability to run their own equivalent simulations of "lower levels of reality". This would make the original simulation more interesting, wouldn't it? Creating an endless cascade of realities may even be the *objective* of the original simulation.

[Peter S.]

Good point: it might be an interesting examination to see how the constraints of a simulation affect the results of stacked simulations it creates.

[Jim Self]

I also agree that it's pointless to guess whether or not we're a basement civilization. If we're told by our God that we're a simulation (and perhaps we have been) then we'll know, otherwise it can be nothing other than a 50/50.

There's no worries about an endless series of sims-in-sims either. If each sim has a decreasing capacity to simulate, you would finally reach a sim incapable of producing its own. If that premise is false (simulations are less efficient than reality) then it isn't a problem either.

My biggest worry would be a scenario that a sim would create a sim with less complexity but far greater speed. That sim could conglomerate into a single superintelligence and began to impose itself on the creator sim before they could respond. The old "AI taking over the world" story taken to the next level. That would be an even worse problem given efficient simulation.

[Richard Franks]

"This means that by 2050 it would be feasible to have a completely realistic historical simulation running on every desktop, and that these simulated worlds would out-number the real one by a factor of millions or even billions to one. This makes it almost certain that we live in one of the simulations if a future society has the motivation to create them and does not encounter any insurmountable ethical and legal obstacles to doing so."

If there are millions (or even billions) of potential historical simulations of the universe running then would each one then have to be unique for it to have a purpose, or would it be a handful of simulations, each testing different variables?

If you have millions/billions of unique 'world' simulations running, then who populates the intelligences in them? If we assume that the 'zero hour' is 50-100 years in the future, then it can't be humans - humans get more meaning from 'real' interactions with each other (which is why MMORPG's are so popular in the first place), plus there simply aren't enough of us to fill these worlds - assuming that our reality is based upon the greater realities' history.

This seems important to me, as our base assumptions here determine the 'almost certain' probability that we exist in a simulation. Thus it matters whether a future society would create millions of simulations or just a few or just one.

Well before we have the ability to simulate an entire world including all the intelligences to roam it on a single computer, it follows that we would have the ability to simulate a single intelligence in a smaller geographical area on a single computer. Would this 'world' simulation not occur around that time using many different networked computers? All of our long-distance travel options seem to fit nicely into the model of being able to easily pass resources off to another server.

This type of simulation would almost certainly not be capable of 'stacking' to the degree discussed - as it happens much earlier than the 'universe simulator' described in the paper.

Which brings our chances of existing in such a simulation down to a much-less-scary 50/50!

[Richard Franks]

It's also interesting to note that quantum mechanics seem only to give us a certain 'probability' of matter/energy existing at a certain location in spacetime -- to simulate an entire universe for the benefit of a sentient intelligence with limited perception, one wouldn't need to simulate every last particle/super string/etc, only those being directly observed at the time. Simulating the 'macro' world which we spend most of our lives in, would take significantly less resources than our universe appears to use, to get the same result.

[Stephan Kinsella]

Well, if they are going to terminate when we figure out the deception--I think you have doomed us all. :)

When I was young, I once had this idea: a good God would not create beings that He knew would choose to sin and go to hell; such a creature did not ask to be born; and would have been better off never having been born and put in the position of having to choose (and choosing wrong); infinite torture outweighs any finite joys in life. So a good God would never create people who would end up sinning. Yet there are apparent sinners--so obviously they are mere flesh-robots placed here by God to tempt the rest of us, who are guaranteed to choose the right thing and go to heaven. Until, that is, some brash smartass spills the beans and ruins his experiment. Uh-oh!

Seriously, this is an interesting and imaginative paper, but I think it is obviously ridiculous. Peter, let me ask you--aside from your interesting hypothesis, what do you *really believe*? Do you really believe we are in a simulation? Or that this is seriously possible? If you had to bet something serious on this--how would you *bet*? Seriously--?

[Peter S. Jenkins]

Stephan - Yes, I really believe it is seriously possible, otherwise I wouldn't have written the paper. I will make you a gentleman's bet of $1, since I don't believe in wagering substantial sums on important issues, (it diverts attention away from the issue itself) and as you are aware, it would probably be impossible for me to collect if I am proven right. I should point out that if you accept this bet, you will essentially be admitting that the theory is a scientific one, i.e. falsifiable in the Popperian sense, rather than merely a metaphysical one.

[Stephan Kinsella]

I don't think betting is admitting anything. And anywya the problem w/ your view is not that it is not "scientific"--I don't care if it is or not. The main this is: is it correct. I think it's obviously ridiculous. NOt saying it's not a valiant effort, or interesting, or nicely argued, or worth discussing. I just think at bottom, it's ridiculous, and what's more, you and I both know it is.

I'm not sure how we could bet--what are the determinable conditions that would tell who has won? You want to just say that you have to pay me $1 in 2055 if we are still around?

And notice, I asked if you really believe it, and you did not asnwer. You said it's "seriously possible." But do you BELIEVE it? Seriously believe that we are in a simulation? I will tell you honestly that if you say that, I either think you are not telling the truth, are you are half nuts.

How about my god-robot idea?

and what is your response to the idea that our creators would end the simulation even earlier than 2050, since smartypants like you are now spilling the beans and no doubt before 2050 it will be common knowledge that we are just a simulation?

[Peter S. Jenkins]

Stephan - Given that the world as a simulation concept has been around at least since Plato's Cave, and has not yet gained general acceptance, it is unlikely that it will do so until we are making large numbers of historical simulations (2050), at which point it will be evident to most people that we likely live in a simulation ourselves.

[Stephan Kinsella]

Peter: I notice you have answered this question: do you really believe your theory *is* true?

[Ivo Jansch]

If you are having difficulty with, or counterarguments to, the possibility that we might be living in a simulation, I invite you to join my wiki to read and discuss the idea, at http://www.simulism.org.