WHOA, DUDE, ARE WE INSIDE A COMPUTER RIGHT NOW?
September 12, 2012 9:42 PM   Subscribe

NASA Scientist suggests everything we see, touch, feel, taste, and smell could be a simulation running inside a computer.

It’s an idea that every college student with a gravity bong and The Matrix on DVD has thought of before, but [Rich Terrile] is a well-regarded scientist, the director of the Center for Evolutionary Computation and Automated Design at NASA’s Jet Propulsion Laboratory, and is currently writing an as-yet-untitled book about the subject.
posted by crunchland (268 comments total) 39 users marked this as a favorite
 
Just to get it out of the way: 42
posted by ShutterBun at 9:46 PM on September 12, 2012 [1 favorite]


There's no theism backdoor crypto-theism.
posted by mobunited at 9:47 PM on September 12, 2012 [5 favorites]


And somewhere, some Alien programmer looks at this simulation, puzzled, and thinks "how do I monetize this?"
posted by mazola at 9:47 PM on September 12, 2012 [46 favorites]


Paging Rene Descartes.

Given his rightful stature on MeFi I would poInt readers towards Stanislaw Lem's terrifying piece on the consequences of such a world.
posted by fallingbadgers at 9:48 PM on September 12, 2012 [7 favorites]


WHOA, DUDE, ARE WE INSIDE A COMPUTER RIGHT NOW?

Yes, you are. Well, to be exact, you're on the screen of my computer. Didn't you know that, little blue web site?
posted by benito.strauss at 9:48 PM on September 12, 2012 [40 favorites]


Time to start griefing.
posted by Ghostride The Whip at 9:49 PM on September 12, 2012 [5 favorites]


Hahah, I can't resist.
posted by resurrexit at 9:50 PM on September 12, 2012 [4 favorites]


Yo dawg, I heard you like The Sims, so I put your Universe inside a computer so that you can simulate Xzibit while Moore's Law....

It's Xzibits all the way down.
posted by explosion at 9:50 PM on September 12, 2012 [12 favorites]


There was any question about this at all? The minute you can simulate atoms all of that becomes possible.

If, of course, souls aren't incorporeal.
posted by Tell Me No Lies at 9:51 PM on September 12, 2012 [1 favorite]


fallingbadgers beat me to it. I'm gonna just go with Descartes on this one.
posted by jimmythefish at 9:51 PM on September 12, 2012


Yeah, yeah I saw The Thirteenth Floor too.

Unless I'm missing something, the argument for this goes as follows: Eventually we will be able to run simulations of the galaxy. If we can run simulations of the galaxy, we will run simulations of the galaxy. Given many simulations of the galaxy, the odds of our experiences being one of the simulations rather than the (presumably) single "true" universe are very, very high.

This has got to be the least convincing argument ever?
posted by Justinian at 9:52 PM on September 12, 2012 [8 favorites]


This is why I can never find any goddam spoons.
posted by SPrintF at 9:52 PM on September 12, 2012 [24 favorites]


So the people who built the computer we're in right now... are they inside a computer too? Where are the Triscuits®?
posted by not_on_display at 9:52 PM on September 12, 2012 [7 favorites]


Every so often I entertain the possibility that I'm a brain in a lab, and some researcher is laughing his ass off.
posted by Pope Guilty at 9:53 PM on September 12, 2012 [3 favorites]


BRANES
posted by stavrosthewonderchicken at 9:53 PM on September 12, 2012 [15 favorites]


NASA Scientist finally stumbles on mundane possibility that has already occurred to any pot-smoking 18 year old living in a college dorm, or anybody that has seen the Matrix.
posted by dgaicun at 9:53 PM on September 12, 2012 [9 favorites]


I've had trouble getting into these kinds of thought exercises ever since I took that blue pill.
posted by hypersloth at 9:53 PM on September 12, 2012 [3 favorites]


Burhanistan wins with his awesome Infocom reference.

The antechapel where the statue stood
Of Newton with his prism and silent face,
The marble index of a mind for ever
Voyaging through strange seas of Thought, alone.

posted by Justinian at 9:54 PM on September 12, 2012 [4 favorites]


It’s an idea that every college student with a gravity bong and The Matrix on DVD has thought of before, but [Rich Terrile] is a well-regarded scientist, the director of the Center for Evolutionary Computation and Automated Design at NASA’s Jet Propulsion Laboratory, and is currently writing an as-yet-untitled book about the subject.

Haha. Fuck, should've at least read the rest of the post.
posted by dgaicun at 9:54 PM on September 12, 2012 [1 favorite]


NASA Scientist finally stumbles on mundane possibility that has already occurred to any pot-smoking 18 year old living in a college dorm, or anybody that has seen the Matrix. -- Uh oh. Deja vu, dude. You know what that means.
posted by crunchland at 9:58 PM on September 12, 2012 [6 favorites]


There's not much meat in that article. What means would we have to test this hypothesis? The existence of the Planck length and attributing it to be a "pixel" (voxel, whatever) of the universe is not, in itself, diagnostic.
posted by chimaera at 10:01 PM on September 12, 2012 [4 favorites]


the lolcats are the glitch
posted by ninjew at 10:01 PM on September 12, 2012 [8 favorites]


Hey, this is UNIX! I know this!
posted by Blazecock Pileon at 10:04 PM on September 12, 2012 [25 favorites]


Nick Bostrom, who gets a mention in the article, has sketched out the logical argument for a simulated reality for a while, that one of the following is probably true:
1. No civilization will reach a level of technological maturity capable of producing simulated realities.

2. No civilization reaching aforementioned technological status will produce a significant number of simulated realities, for any of a number of reasons, such as diversion of computational processing power for other tasks, ethical considerations of holding entities captive in simulated realities, etc.

3. Any entities with our general set of experiences are almost certainly living in a simulation.
posted by kithrater at 10:06 PM on September 12, 2012 [8 favorites]


NASA Scientist suggests all that you touch, and all that you see, all that you taste, all you feel, and all that you love, and all that you hate, all you distrust, all you save, and all that you give, and all that you deal, and all that you buy, beg, borrow or steal, and all you create, and all you destroy, and all that you do, and all that you say, and all that you eat, and everyone you meet, and all that you slight, and everyone you fight, and all that is now, and all that is gone, and all that's to come, and everything under the sun could be a simulation running inside a computer, eclipsed by the moon.
posted by Ice Cream Socialist at 10:06 PM on September 12, 2012 [67 favorites]


We have no reason to believe that the universe is finitely rather than infinitely complex. If it's infinitely complex it's not computable. So we have no reason to think that the universe is capable of being simulated.
posted by East Manitoba Regional Junior Kabaddi Champion '94 at 10:08 PM on September 12, 2012 [2 favorites]


Why simulate everything we experience. Much more efficient to just simulate everything I experience.
posted by StickyCarpet at 10:09 PM on September 12, 2012 [8 favorites]


In 30 years we expect that a PlayStation—they come out with a new PlayStation every six to eight years, so this would be a PlayStation 7—will be able to compute about 10,000 human lifetimes simultaneously in real time, or about a human lifetime in an hour.

Wouldn't the waste heat from all these supercomputers cook us before we can set up a digital utopia?
posted by Blazecock Pileon at 10:09 PM on September 12, 2012 [3 favorites]


NASA Scientist gets high, watches eXistenZ.
posted by Sys Rq at 10:11 PM on September 12, 2012 [8 favorites]


Hey, don't put Descartes before the horse.
posted by oneswellfoop at 10:12 PM on September 12, 2012 [25 favorites]


Given his rightful stature on MeFi I would poInt readers towards Stanislaw Lem's terrifying piece on the consequences of such a world.

Only...you didn't. What piece? WHAT PIECE?!
posted by adamdschneider at 10:14 PM on September 12, 2012 [8 favorites]


It’s an idea that every college student with a gravity bong and The Matrix on DVD has thought of before, but Rich is a well-regarded scientist, the director of the Center for Evolutionary Computation and Automated Design at NASA’s Jet Propulsion Laboratory...

Yeah, I'm just going to put this out there: I bet the folks at JPL can still whip up a pretty mean gravity bong when the mood strikes them.
posted by Kadin2048 at 10:15 PM on September 12, 2012 [11 favorites]


This has got to be the least convincing argument ever?

Let's say I purchase 1 McGriddle. Using my powers of observation and deduction I am able to ascertain how to make perfect, yet counterfeit McGriddles. I set out and cook up millions of indistinguishable fake McGriddles. You then come over to my house, pick one up, and eat it. What are the chances that you ate the bona fide McGriddle?

I feel like it's more convincing with McGriddles.
posted by the jam at 10:17 PM on September 12, 2012 [13 favorites]


I think you mean anti-gravity bong
posted by ninjew at 10:18 PM on September 12, 2012 [2 favorites]


So to be clear, this is a debate about metaphysics not physics. There are no measurements we can make, no logical arguments based on experiment that we can form that are relevant. This is because everything - the laws of nature, our memories, logic, experience - are all mutable and controllable in such a universe.

The problem with the "Brains in jars" theory is that it is a dead end, at the end all we can do is agree with Descartes and move on.

And try not to think of the horror...
posted by fallingbadgers at 10:18 PM on September 12, 2012 [9 favorites]


So?
posted by fullerine at 10:19 PM on September 12, 2012 [1 favorite]


Yeah, I'm just going to put this out there: I bet the folks at JPL can still whip up a pretty mean gravity bong when the mood strikes them.

The secret ingredient is the sky crane.
posted by mazola at 10:19 PM on September 12, 2012 [4 favorites]


the jam: What about your McGriddles makes them less authentic than the McGriddle you bought? It can't be that they're mass-produced...
posted by Apropos of Something at 10:20 PM on September 12, 2012


so what
posted by TwelveTwo at 10:20 PM on September 12, 2012


that's the good news. The bad news is we think the computer is running in a larger simulator.
posted by boo_radley at 10:23 PM on September 12, 2012 [2 favorites]


I've sometimes thought so, but no matter how many times I go running through empty office buildings screaming, "TECH SUPPORT!!! TECH SUPPOOOOOOORT!!!!!" at the top of my lungs, they never show up. I've got a long list of bugs they need to address.
posted by Max Udargo at 10:23 PM on September 12, 2012 [15 favorites]


Also, the only difference between this theory and basic theism is the supposition that our protobeing has the technical capability to produce an entire complex universe but has chosen to only simulate a universe similar to his/her own.
posted by Apropos of Something at 10:23 PM on September 12, 2012 [7 favorites]


So?
posted by fullerine at 10:19 PM on September 12 [+] [!]

so what
posted by TwelveTwo at 10:20 PM on September 12 [+] [!]


How do we break out and become actual moral agents instead of grooves in a record?
posted by 5ean at 10:24 PM on September 12, 2012 [3 favorites]


Just to get it out of the way: 42

When I popped over here and saw this post, the comment count was in fact 42. Eeeeerieee.


It was 43 by the time I clicked inside to comment, but whatever.
posted by me3dia at 10:25 PM on September 12, 2012 [1 favorite]


Relevant.
posted by Wordwoman at 10:25 PM on September 12, 2012


Wake me when we can study the reality which we are a simulation of, until then-- maybe reality is but a recursive simulation of itself, and thus any discussion of 'actual reality' should be dismissed exactly as Hegel advised.
posted by TwelveTwo at 10:27 PM on September 12, 2012 [2 favorites]


I've been thinking for some time that all the quantum weirdness down at the bottom of things could be, in essence, lazy evaluation. Whatever computational substrate we're running on, to this way of thinking, simply never determines many of the answers, using approximations instead. It's only when a specific answer actually matters that the computation is fully carried out, and, if necessary, any other retroactive adjustments to spacetime are also implemented. That's why quantum measurements taken in the future are always consistent with entangled ones taken in the past -- the simulation goes back, and edits everything that way. It doesn't matter how long this takes in 'real time', if such a thing even exists; since our time is the simulated time, we can't tell how long each individual tick takes to run. To us, all ticks are the same speed, no matter how long the GodComputer takes to run each one.

Interestingly, simply watching for 'hot spots' in the simulated universe, areas that are taking lots of computation time, should inevitably lead the implementors to interesting things happening in that universe.... our particle accelerators, if we're running on a simulation, would be producing some very, very strange requests for 'CPU time'. That would be a flashing neon light that the entities running the simulation should check out that third planet orbiting that unremarkable sun in that rather plebian spiral galaxy.

Another thing that kind of supports this idea is that the universe seems to have minimum sizes and times -- the Planck length and the Planck time. And energy and matter all seem to come in discrete packets, there or not-there, which is also consistent with a simulation.

It would, of course, require computers and algorithms well beyond our capacity to ever grasp; it's unlikely that anything running in a model could ever be smart enough to understand the model. If a modeled entity could understand its own model, that means the host computer would need to simulate both the entity plus itself, which may not be logically possible. But, perhaps, we can understand enough to know that we're IN a model, even if the scope is too complex for us.

Another thought I just had: the fundamental quantum randomness might be very deliberate, a damping effect on perturbations. If the GodComputer has to go back to earlier frames and change the results of computations to match later measurements, the ripples from that change could potentially mean everything within that event's light cone would have to stop, return to an earlier frame, and restart -- a missed branch prediction, in CPU-speak. The random quantum oscillations could function as a field reducing the spread of butterfly-wing effects to a local area, so that scientists doing weird crap in a laboratory, instead of making a huge chunk of a galaxy miss a couple of beats, might just force a recomputation of their local laboratory... eventually, the ripples of difference would be swallowed by quantum noise.

Purely hypothetical bullshit, of course, but I find it interesting to think about.
posted by Malor at 10:27 PM on September 12, 2012 [73 favorites]


Not too new. In the 70's a version of this was published.

Time and space are illusions. The universe is all built, and we move through its elements like the stylus on a record. Our illusion of awarenes is us, propelled by a cosmic force, being moved along the groove, as we sense fluctuations in the cosmic egg.
posted by mule98J at 10:27 PM on September 12, 2012


Think of your current computer, riddled with viruses, Easter eggs, programming errors, random memory glitches, etc. Now magnify those to the size of a computer capable of simulating what we perceive as the universe, and the software running it. Maybe that's what we perceive as UFOs, the Loch Ness monster, ESP, miracles, etc.

Just a thought...
posted by blm at 10:27 PM on September 12, 2012 [1 favorite]


So it turns out Steve Jobs is God?
posted by mazola at 10:29 PM on September 12, 2012


All I want to know is: Where do I file my bug report?

Or even better, where do I find the sadistic SOB that managed or designed the project? I would like to wring his neck. You guys have robots, right? Please give me access to a large robot or whatever it is you use in your dimension somewhere near this entity.

More seriously: Even if reality is really "real* - we already live in the "computer simulation" of consciousness and the removal of our senses. There's a finite resolution to your eyes, your nerves - even your thoughts.

But it's also a pretty easy leap for any modern armchair philosopher/stoner to ponder the simulation argument.

There are some very strange things about our reality. The fractal nature of, well, nature. The behavior of Pi. Unresolved deep questions about space, time, physics and cosmology. Not to mention widely reported paranormal phenomena and other glitches in the matrix, as the kids say today.
posted by loquacious at 10:30 PM on September 12, 2012 [2 favorites]


Besides, if someone is running all I experience (and the world, etc) in a simulation and there is anyway I can communicate with the simulator, I will definitely have words with them.

Words, I tell you.
posted by chimaera at 10:32 PM on September 12, 2012 [4 favorites]


Yeah, I dug Horton Hears a Who, too.
posted by Sidhedevil at 10:32 PM on September 12, 2012 [2 favorites]


Following up on the Cartesian posters---nearly four centuries! Does anybody remember cogito ergo sum? Or a more modern take on the brain in a vat?
posted by phenylphenol at 10:35 PM on September 12, 2012


Just weird to see a random NASA scientist's authority elevated so much compared with a pretty solid philosophical tradition.
posted by phenylphenol at 10:35 PM on September 12, 2012 [5 favorites]


I think I can prove that he's wrong. If it were true, then position would be quantized. And the coordinate system would be absolute.

Relativity doesn't permit any absolute coordinate system. And quantum mechanics doesn't allow granularity in position. If there was an absolute coordinate system, or if position was granular, we'd have noticed it by now.
posted by Chocolate Pickle at 10:37 PM on September 12, 2012 [3 favorites]


There's a paper I read in college, on how Descartes' argument was based on a fallacy. It was written in the last 50 years, for sure.

I can't remember if this was just a myth to get us going, but we were told it was written by a guy who basically wrote three papers, and on that was granted a chair at a prestigious university. One of those papers blew the Cartesian thought experiment out of the water. Something along the lines of, the experiment is not possible, you cannot doubt everything. You must believe something to be true in order for the concept of doubt to have any meaning.

If anybody knows what I'm talking about and can give me a link, I'd shit myself.
posted by phaedon at 10:37 PM on September 12, 2012 [1 favorite]


I saw this TNG episode. Somehow we're going to have to persuade Moriarty that he can leave the holodeck and get on a shuttle or something.
posted by Chekhovian at 10:38 PM on September 12, 2012 [8 favorites]


It's a cool (if unoriginal) theory but I don't see how it could be "simulations all the way down". Even if the simulation contained a whole bunch of tricks like not simulating anything that a conscious being isn't looking at, there would be a limit on the complexity of the simulated universe imposed by the complexity of the simulated universe above it. The simulations would have to get progressively simpler each step of the way down. I do like the idea of computer programmers being gods, but I think this idea works even without computers. For example, suppose your life is actually a simulation being run inside the brain of a super-intelligent being, perhaps an alien with such a highly evolved capacity for empathy that it simulates the entire consciousness of beings it plans to make contact with some day inside its own mind. Or perhaps your lived experience is just a "dry run" of each day that you make in your sleep each night, with the actual day being experienced by a separate part of your mind, or maybe even lived unconsciously.
posted by L.P. Hatecraft at 10:42 PM on September 12, 2012


When I was young I read a sci-fi story which had as it's crux that there was proof of parallel universes (which I see as an analogue of living in a simulation). The author suggested this would lead to mass suicide, moral decrepitude and other horrors, as the inhabitants thought things like "well, I'm jumping off this roof, but in another universe, I'm not, so what does it matter?"

Likewise, if you're living in a simulator, you could come to a similar conclusion: my life is shitty, I'm miserable, whatever, why should I continue running this program if ultimately this pain is just for someone's doctoral thesis or whatever?

I couldn't see that point of view when I first read the story, but many years later I kind of can get a peek at it. Back then I endorsed the attitude of "even if it's a simulation or you're in one of many universes, why would that matter? It's real to you!" Now I can at least see "this is kind of a grind--if this is just a simulation, why bother?"
posted by maxwelton at 10:44 PM on September 12, 2012 [2 favorites]


As it happens, I write simulations of the human experience for both fun and profit. For my day job, the software I've written can simulate the travel decisions that four million people make on a typical fall weekday in about an hour. The part that looks at their interactions in creating traffic congestion runs another hour and a half or so. In my spare time, I'm poking at a simulation of all 35 million Canadians, running them through their lifetimes and seeing who gets lung cancer. (It takes 5 minutes per year.) Next year, 167 people in Trois-Rivieres are diagnosed with lung cancer, including person 22823194, a 79 year old woman who has had a pack-a-day smoking habit since she turned 18. It is discovered at stage III or IV, and is inoperable.

So I'm susceptible to this line of thinking. But here's the thing. As time goes by and computers get more powerful, it turns out that the simulations need to get more and more detailed. We used to simulate the travel that people make in a city as a pure mathematical abstraction, a matrix of numbers convoluted this way and that. Now, we simulate these decisions person-by-person, because it's actually easier to make more complex decisions that way than adding in more matrices. But we still only represent road traffic for a citywide simulation as a mathematical abstraction, numbers in a network. However, the first city-level simulations of vehicle-by-vehicle traffic are starting to be produced and maybe in 5 or 10 years will be the new standard. But then it'll be that we don't simulate the underlying economy or the decisions of developers or the relationships within the household or the demographic principles or the relationship with the larger world. Whatever we simulate, we have to have a damn good reason to do, and we have to abstract as much as possible, because there's only so much runtime to go around.

And that's just one domain; eventually, everything starts to come together. To model the demographics for a city, you need to look at mortality and the causes of death, so you need to understand cancer. And that means on one hand, you need to simulate the decisions around smoking, so you're getting into psychology and eventually the entire human brain. On the other hand, you also need to model the tumours, which means you need to model the cells in the lungs and eventually you get down to needing sub-atomic particle interactions.

And of course, if these people are interacting, they start experiencing the world; some of the ideas of emergent systems that I'm talking around came from E.O. Wilson, and so you need to simulate ants, and they move dirt and drop pheromones, so you're doing all of that. On the other hand, people are interacting in ways that involve space -- the cold war had space exploration as a major theme in the geopolitics, which fucked the world up in a billion different ways, so you can't just have a skybox or drop in a moon with a patch like on the Kerbal Space Program.

And yeah, yeah, Moore's law. (On the next simulation over, it's called Vijanishara's law. On the one after that, it's intelligent dolphins, who are having an ethical debate over whether they should be keeping primates in captivity.) It's one thing to say that computers are getting more and more powerful. But you can only extrapolate so far -- the average newborn weighs 7 pounds, and the average 1 year old weighs 3 times that much. But the average 10 year old doesn't weigh 413,343 pounds. Eventually, you wind up with a simulation so detailed you just give up and simulate every subatomic particle in the universe. Except your computer is made out of something, and something rather more complex than subatomic particles.

So the only way that it works, that we are simulations, is if it's as part of an infinitely more complex universe than the one we experience. In which case, you have to wonder why they are simulating something so vividly when it can only be a tiny subsection of what they themselves are dealing with.
posted by Homeboy Trouble at 10:46 PM on September 12, 2012 [49 favorites]


MetaFilter: everything we see, touch, feel, taste, and smell
posted by mazola at 10:46 PM on September 12, 2012 [1 favorite]


What do zombie physicists eat? BOLTZMANN BRAIIIINNNSS
posted by en forme de poire at 10:48 PM on September 12, 2012 [6 favorites]


I think I can prove that he's wrong. If it were true, then position would be quantized. And the coordinate system would be absolute.

I don't think that follows. We can imagine a universe where space isn't quantized and without an absolute coordinate system (i.e. ours), and could write a program to simulate this universe to some degree of accuracy today and some better degree of accuracy tomorrow, so it seems some vastly advanced future being could simulate it to the degree of accuracy that we're able to measure today.

Also, we know GR and QM and SM are incomplete in various ways, and I've seen serious proposals, unrelated to the universe being a simulation, that suggest space is quantized, with proposals on how to test it.
posted by blm at 10:50 PM on September 12, 2012


Maybe the universe is incomplete and these errors aren't due to our reality being but a limited simulation of a whole and perfect existence, but are instead real features of an imperfect, incomplete existence. Perhaps these ugly quirks of existence are not distortions borne of our limited intellects, these ugly quirks are not the limitations of a simulation, but instead these ugly quirks are essential features of reality itself. Our universe may simply be an ugly quirky universe, a universe where position and momentum cannot both be known simultaneously, where a set of all sets is not possible, where antimonies run wild, and P != NP.

The dream that this is all a dream is but the dream that this is a dream from which we may some day wake. But what a nightmare is the truth: we are already awake.
posted by TwelveTwo at 10:50 PM on September 12, 2012 [4 favorites]


I'd like to subscribe to your newsletter, malor.

The simulation argument would probably be completely unprovable, but it's a little different from traditional deism in that it posits a deeper layer of programming underneath the laws of physics. If we could "root" the universe and gain access to that layer, we could become gods ourselves, and rewrite the cosmos to be anything at all. Hypothetically.
posted by Kevin Street at 10:51 PM on September 12, 2012


what stanislaw lem piece please
posted by facetious at 10:54 PM on September 12, 2012 [4 favorites]


The idea that we and everything around us is a form of data isn't really that far fetched. The moral and ethical problems with this are much more interesting. As a disclaimer I don't necessarily believe in Simulism. I am however fond of philosophy.

How would simulation creators get around the problem of evil (and good)? Creating beings (they would after all be real beings) who feel and experience pain, hardship and suffering and then exposing them to a world full of random pain, hardship and suffering is a very questionable way of acting.

Looking at Nick Bostrom's three postulates there's a good argument to made that in order for a civilization to survive long enough to be able to benefit from Moore's Law to this extent they would have to avoid wiping themselves out and continue co-operating on a high enough level to support the necessary infrastructure. Some sort of ethically guided morality seems necessary for this equation to function.

It therefore seems less likely that a society of sufficient sophistication to achieve this would remain barbaric enough to torture and kill billions upon billions of people, thus making the second postulate more likely.

There is a way around this impasse however. Voluntary participation in the simulation, with full disclosure of the hazards, by simulated beings, perhaps even the simulation creators themselves, would somewhat alleviate this moral dilemma.

When you take this possibility in conjunction with potential breakthroughs in quantum computing where it might only require one volunteer to experience ALL the lifetimes in the simulation at once and things start to get really weird. In other words, in this scenario, everyone who exists and has ever existed is the same person, you.

Stick that in your bong and smoke it.
posted by Jernau at 10:57 PM on September 12, 2012 [7 favorites]


If we could "root" the universe and gain access to that layer, we could become gods ourselves, and rewrite the cosmos to be anything at all.

Combining this and my first comment above, maybe there are legitimate magicians, shamans, gurus, whatever, and they are that way because they've managed to hack the universe in some way.

Just another thought...
posted by blm at 10:59 PM on September 12, 2012 [2 favorites]


How many of you guys are high, right now?
posted by mazola at 11:00 PM on September 12, 2012 [4 favorites]



It's a cool (if unoriginal) theory but I don't see how it could be "simulations all the way down"...
posted by L.P. Hatecraft at 10:42 PM on September 12


We would have absolutely no way of knowing what the properties of a simulator's world are.
posted by Jernau at 11:01 PM on September 12, 2012


Minecraft!!!
posted by Hairy Lobster at 11:01 PM on September 12, 2012


How many of you guys are high, right now?

None of us. But many of us are programmed to *think* we are.
posted by flapjax at midnite at 11:02 PM on September 12, 2012


there is no spoon
posted by WalkingAround at 11:02 PM on September 12, 2012 [1 favorite]


But if we did root the universe we'd have to remember to write ourselves into the new program upon reboot. Many an ancient empire may have perished by neglecting that simple precaution...
posted by Kevin Street at 11:03 PM on September 12, 2012 [3 favorites]


We would have absolutely no way of knowing what the properties of a simulator's world are.

It's Sinatra's Simulator's world, we just live in it.
posted by flapjax at midnite at 11:03 PM on September 12, 2012 [1 favorite]


There's an episode of DS9 I was watching on Netflix not too long ago where Dax and Odo find a colony where people are going missing. Dax and Odo agree to help and discover the colony and its inhabitants are actually a holographic simulation that's been running for years. The computer needs repair, which is why people are disappearing, so Dax and Odo tell the colonists they're all holograms and they need to shut them off temporarily to repair them.

There's initial skepticism, and one dude turns out to be real, but the thing I really like is at the end, the holograms, knowing they're just holograms, are just happy the disappeared holograms have returned and they can continue their (simulated) lives together.
posted by DyRE at 11:06 PM on September 12, 2012 [7 favorites]


there is no spoon

Anyone who ever had to deal with The Tick wishes that were true...
posted by evilmidnightbomberwhatbombsatmidnight at 11:14 PM on September 12, 2012



We would have absolutely no way of knowing what the properties of a simulator's world are.

Actually, we would be able to know some properties. We would be able to know what was mathematically impossible in the simulator's world even if we only had access to what was possible in our simulated world. It would be a structuralist knowledge, purely negative, but it would still be a knowledge of their world.
posted by TwelveTwo at 11:14 PM on September 12, 2012


How would simulation creators get around the problem of evil (and good)? Creating beings (they would after all be real beings) who feel and experience pain, hardship and suffering and then exposing them to a world full of random pain, hardship and suffering is a very questionable way of acting.

Greg Egan's short story Dust explores this very question.
posted by blm at 11:14 PM on September 12, 2012 [6 favorites]


In which case, you have to wonder why they are simulating something so vividly when it can only be a tiny subsection of what they themselves are dealing with.

....(swap swap swap)...

For my day job, the software I've written can simulate the travel decisions that four million people make on a typical fall weekday in about an hour.

Did you do that on purpose?
posted by chimaera at 11:15 PM on September 12, 2012 [3 favorites]


There's also an interesting argument to be made as to why running simulations might be a good idea. Historical value.

A simulation based on our worlds physical rules with a known starting and ending point could, if run enough times and then statistically analyzed, provide us with some pretty exact details about our history.
posted by Jernau at 11:15 PM on September 12, 2012 [2 favorites]


Carl Sagan presented a relatively clever method for proving that a universe was an artifact -- encoding a statistically unlikely message into a transcendental number.

In that particular case it was a string of figures a few billion digits into Pi that when viewed in base 11 provided a few thousand 1's and 0's that graphically made a circle. The chance of them being random was statistically negligible.

It's interesting because it's the kind of proof that scientists can get their head around rather than relying on revelation.

In any case if we are living in a simulation there may be a signature laying around somewhere. We just have to find it.
posted by Tell Me No Lies at 11:20 PM on September 12, 2012 [5 favorites]


Oh man, there was this clique of highly religious students at my university who insisted that perfect reality simulations based on physics and science would never become perfect simulations of reality, because they wouldn't account for the influence of God.
posted by ceribus peribus at 11:20 PM on September 12, 2012 [1 favorite]


Surprised it took that long for Greg Egan to get mentioned. Highly recommend Permutation City to anyone interested in this stuff.

Also, anyone who believes that our free will would be negated by living in a computer simulation is already a theist.
posted by cthuljew at 11:21 PM on September 12, 2012 [1 favorite]


There's also an interesting argument to be made as to why running simulations might be a good idea. Historical value.

That's one reason, there are others: entertainment, experiencing alternative worlds, research into how the universe works. There are lots of "real life" examples of each.
posted by blm at 11:21 PM on September 12, 2012


That's one reason, there are others: entertainment, experiencing alternative worlds, research into how the universe works. There are lots of "real life" examples of each.

I agree, I just though using simulations as a way of gaining historical insight was a not necessarily something most folks would immediately think of.
posted by Jernau at 11:24 PM on September 12, 2012 [1 favorite]


I'm not convinced simulating the "universe" would really be that complex, would it? The simulation would start at birth and end at death, and, realistically, between A and B there isn't really a lot of data...the universe, as such, exists only as the experiences of an individual, right?
posted by maxwelton at 11:30 PM on September 12, 2012


Oh man, there was this clique of highly religious students at my university who insisted that perfect reality simulations based on physics and science would never become perfect simulations of reality, because they wouldn't account for the influence of God.

In my experience even the most diehard of atheists are liable to falter in the face of declaring the human experience to be strictly a matter of molecules bouncing around in a deterministic fashion. The idea that there is some engine that can affect the electrical impulses in our brain but that is not subject to strict physical cause and effect runs very deep.

So even putting aside God you still need to come up with a physical model for free will. There's been a tendency among scientists lately to light a bonfire and dance around it chanting "Quantum, Quantum, Quantum!" when faced with questions like this but I have to say that having looked closely I'm not seeing it.
posted by Tell Me No Lies at 11:33 PM on September 12, 2012 [2 favorites]


I did not read the article because this is one of the recurring nightmares I have. I do not know why, but about once or twice a month I dream we are actually in a computer and not real. When I wake up, it is always weird for a day or two.

I should know better than to read MeFi late at night.
posted by SuzySmith at 11:37 PM on September 12, 2012 [1 favorite]


Twaddle. Moore's "law" was merely an observation, that's already levelling out, and besides which reality is not a computable function.

Computer people are supposed to learn this in first year. The field has well understood limits.
posted by ead at 11:41 PM on September 12, 2012 [1 favorite]


In my experience even the most diehard of atheists are liable to falter in the face of declaring the human experience to be strictly a matter of molecules bouncing around in a deterministic fashion.

Not really. Free will is pretty retro these days. Nobody that's anybody really thinks it exists, but they keep quiet about it as it upsets the economists at the Vice-Chancellor's cocktail parties.

Economists, of course, require free will to explain why they haven't come up with anything that atually works.
posted by Sparx at 11:42 PM on September 12, 2012 [1 favorite]


Oh man, there was this clique of highly religious students at my university who insisted that perfect reality simulations based on physics and science would never become perfect simulations of reality, because they wouldn't account for the influence of God.

That is a totally awesome idea. Has there ever been a science fiction work based on this? I can imagine some futuristic monastic order pursuing the perfect computer simulation so that they can do a diff against reality to see the hand of God. Or a meteorologist who, in the course of perfecting a weather prediction system, accidentally proves the existence of God.
posted by XMLicious at 11:45 PM on September 12, 2012 [7 favorites]


So even putting aside God you still need to come up with a physical model for free will....

Well looking at it from their perspective, their point was that even if you could simulate an accurate human complete with free will, the universe that the person was experiencing wouldn't be accurate without simulating the man behind the curtain, so to speak. Including how that entity would react to the level of faith and piety that your simulacrum demonstrated, reacting by affecting the nature of the universe, etc, etc.

Each side quickly became so exasperated with the other that unfortunately the discussion didn't get much further.
posted by ceribus peribus at 11:46 PM on September 12, 2012



Economists, of course, require free will to explain why they haven't come up with anything that atually works


Do they? I didn't think they did.
posted by TwelveTwo at 11:46 PM on September 12, 2012


I did not read the article because this is one of the recurring nightmares I have. I do not know why, but about once or twice a month I dream we are actually in a computer and not real. When I wake up, it is always weird for a day or two.

Try and be comforted by the thought that living in a computer wouldn't make you any less real. It would just make you slightly different than you had perceived yourself.
posted by Jernau at 11:46 PM on September 12, 2012 [1 favorite]


I thought free will was why economic models don't work.
posted by TwelveTwo at 11:47 PM on September 12, 2012


The problem with economic models is that they presume rational agents.
posted by ceribus peribus at 11:48 PM on September 12, 2012 [1 favorite]


The Lem story is probably "The Seventh Sally", from The Cyberiad.
posted by speednoisemovement at 11:48 PM on September 12, 2012 [7 favorites]


For those interested in exploring such ideas in a contemporary science-fiction context, try Greg Egan's Diaspora or Peter F. Hamilton's Void series.
posted by jet_manifesto at 11:50 PM on September 12, 2012 [2 favorites]


The problem with economic models is that they presume rational agents.

Rational is the opposite of free. Given a particular scenario a rational agent will behave identically to any other rational agent in the same scenario. That is causality veiled underneath warm terms like 'preference' and 'values.'

But then reality fails to conform to these rational agent models and economists by habit say: well, free will is why our science fails here.
posted by TwelveTwo at 11:51 PM on September 12, 2012


Our brain is a lump of meat inside a bone box. It doesn't see light, it doesn't hear sound, it doesn't feel heat or cold or pressure. It invents all those terms from limited stimuli that it gets from a few sensory aparati which themselves are able to sample only the tiniest, most miniscule effects from the world around us. Our version of reality is a poor proxy, and even that was gained after about 60,000 years of discussion to compare notes on the trivia we are pleased to call reallity. Everything comes down to conjecture. We kill to preserve the version we like best.

Now and then we hear odd scratchings and bumps, sense the movement of subtle tides, and our collective intuition trips over a few of the many, many loose ends of our conjecture, so we take a moment from our recursive melodramas to wonder what it means. Is something else there? We hold our breath a moment, peer inward for some relevent pattern, but in the end we decide, no, it's just the echo of our hearts thumping at our tympania. No, no, I have conquered the void. It's not turtles all the way down, because, you see, I have thought of everything.
posted by mule98J at 11:53 PM on September 12, 2012 [27 favorites]


"this is kind of a grind--if this is just a simulation, why bother?"

It's the structure that's important, not the matter. So if you kept the structure and simulated the matter, it wouldn't change anything. Just like it doesn't change anything to know we're actually made of atoms, and those are mostly empty.

Surprised it took that long for Greg Egan to get mentioned. Highly recommend Permutation City to anyone interested in this stuff.

Man Greg Egan sure takes this stuff pretty far - Permutation City isn't really even science fiction... it's more ... number theory fiction?

The idea is if you can simulate a universe, by making a set of rules and running them iteratively on a starting state, then you don't need to actually calculate all the steps though yourself. Just like the set of prime numbers is infinite and is not itself affected by whether or not we calculate the 10^47th prime number, the complete future of this simulation "exists" mathematically, regardless of whether anyone outside the simulation bothers to compute it. To make things in any way relevant to the reader, Greg Egan actually has a human in his real world get some investors and buy up a ton of computer time to kick off the first few initial states of this simulation, but mathematically, there's no reason that's important at all.

And that's deeply interesting to me, because I do agree with the premise (that mathematical simulations can model complex and subjective experiences), and well... it's hard to even begin to understand what the implications of this are.
posted by aubilenon at 11:57 PM on September 12, 2012 [1 favorite]


Wait, pixillation of the smallest pieces of matter? What is he talking about here, because I'm pretty sure we can only 'see' to the atomic level and that 'pixillation' is more an artifact of seeing X-rays interacting with the electron orbitals, if one could actually see the orbitals themselves (through whatever means)... well, wouldn't it all just sort blend together? God damn, I never wanted to think of p-chem again but there it is.
posted by Slackermagee at 11:57 PM on September 12, 2012


"That is a totally awesome idea. Has there ever been a science fiction work based on this? I can imagine some futuristic monastic order pursuing the perfect computer simulation so that they can do a diff against reality to see the hand of God. Or a meteorologist who, in the course of perfecting a weather prediction system, accidentally proves the existence of God."

Then he looked up. And overhead, without any fuss, the stars were going out.
posted by Kevin Street at 11:58 PM on September 12, 2012 [8 favorites]


Slackermagee, I think they're talking (in an imprecise way) about the various Planck Units. There's a smallest possible fraction of space, a smallest amount of mass, a tiniest charge, a finest gradient of temperature, and (most intriguing of all) a smallest possible sliver of time. It isn't turtles all the way down.
posted by Kevin Street at 12:03 AM on September 13, 2012 [1 favorite]


And overhead, without any fuss, the stars were going out.

That's wasn't an accident.
posted by Tell Me No Lies at 12:06 AM on September 13, 2012 [1 favorite]



[ :: Start message from: Psyop :: ]
[- World Server is going offline -]
[- Make way to a safe camp point -]
[- Settings have been auto saved -]
[- Updates: FREEWILL, HANG TOWEL -]
[- NEW POLIT & RELIG FRACTIONING -]
[- DOWNGRADING PETROLM RESOURCES -]
[- COMMON SENSE UPDATE ROLLBACKS -]
[- Server will reboot after frag -]
[- Expected downtime is 27:00hrs -]
[- Connections open at 43:94 BLT -]
[- Reporting data avail 19-93-84 -]
[- Go & enjoy the big green room -]
[- Queries? Refer @MasterControl -]
[ :+: End message from: Psyop :+: ]

posted by tilde at 12:06 AM on September 13, 2012 [12 favorites]


Kevin, so they're all Plancking?
posted by tilde at 12:08 AM on September 13, 2012


So even putting aside God you still need to come up with a physical model for free will.

Free will? What does that even mean?

It's unquestionably true that to a large extent my thoughts and actions are determined by my initial state compounded with my experiences*... If that's completely true then there is no free will. But anything that is not based on those things is ... arbitrary. Free will just meaning "I have a random number generator in my head" - well that doesn't seem like anything special to me.

Really, though, I think free will is a model already - a model for the fact that humans are complex systems that are very difficult to predict. It's not a very good model, in terms of predictive power, but it's very difficult to contradict, and often it's helpful to remind ourselves of the limits of our complex theories. But I'm not sure the concept of free will as a absolute trait that people either have or don't is a very helpful one.

* I feel like it's important to note that I am not limiting this to the material - if we have souls, those statements still are equally valid, right?
posted by aubilenon at 12:13 AM on September 13, 2012 [1 favorite]


That's wasn't an accident.

And didn't involve simulating or searching for anything.
posted by XMLicious at 12:18 AM on September 13, 2012


On the one after that, it's intelligent dolphins, who are having an ethical debate over whether they should be keeping primates in captivity.

In case anyone is listening: No. Sorry about the fishing industry, especially the thing with the tuna nets because, wow ... clearly we weren't thinking that one through. And, oh yeah, apologies for Sea World.
posted by krinklyfig at 12:37 AM on September 13, 2012


Regarding free will, let me just be lazy and paste in my comment from an earlier thread:

The only way I can make sense of the notion of an entity 'having free will' is in the form of 'having free will with respect to some other entity'. If agent X has the means to predict and manipulate*, to some extent, the outcome of the computations inside agent Y, then to that extent Y lacks free will with respect to X.

*Whether by manipulating the environment that Y uses as an input or manipulating the computations within Y.
posted by Anything at 12:40 AM on September 13, 2012


Philosophy has been here before. What is reality? Is it a dream? Am I a man dreaming that I am a butterfly, or a butterfly dreaming that I am a man?

What's new are the references. Am I playing Grand Theft Auto IV, or is it playing me?
posted by twoleftfeet at 1:03 AM on September 13, 2012


  • It's unquestionably true that to a large extent my thoughts and actions are determined by my initial state compounded with my experiences*
  • I feel like it's important to note that I am not limiting this to the material - if we have souls, those statements still are equally valid, right?

  • In the way that I think of them there is a crucial difference, namely that I can demonstrate that material is made up of components that behave in completely deterministic ways. The concept of "you" and "thoughts" are high level abstractions that have nothing to do with particles bouncing around.

    I know nothing about the world of souls. Strict determinism may or may not be possible there.

    Free will? What does that even mean?

    I'm using it to mean agency, as in the ability to make choices. For example, a rock has no observable free will -- everything it does is entirely determined physical law. When you throw life into the equation things start to get complicated. Does a flower choose to unfurl its petals or are inexorable chemical processes driving the whole thing? We're pretty sure it's the latter.

    Really, though, I think free will is a model already - a model for the fact that humans are complex systems that are very difficult to predict.

    I agree that it's a very useful model that way. However I meant "model" in the computer science sense, which calls for you to understand something well enough to simulate it and get believable results. (e.g. a Climate Model). A model for free will would be awfully close to A.I., and no one knows how to do that yet.
    posted by Tell Me No Lies at 1:07 AM on September 13, 2012


    Here's the thing: if we're a simulation, we might, or might not, be able to know that, according to the rules that run the simulation. If the simulation is designed for infinite complexity, then it's theoretically possible that simulated agents might "figure themselves out". If the simulation is designed for strict rules of generation, there is no way we could figure everything out with enough confidence to re-create such a simulation - i.e. assuming Godel's incompleteness Theorem holds true - i.e. there exist problems within every formal system that that system cannot solve.

    Also, think of the combinations and permutations possible if we combine the idea that we might be in a simulated universe with the theoretical idea of the metaverse (or infinite multiple/parallel universes. Here, were left with the potential that we're a simulation, but that every possible simulation of a simulation exists, along with every possible combination and permutation of the simulator.

    Like someone said above, Decartes! I'll add to that Wittgenstein, who claimed that it was mostly a losing battle to try to figure out where we came from; why we're here;; where we're going; how we got here; etc. etc. Instead, claimed Wittgenstein, it's most worthwhile to wonder THAT we're here. That adds a little bit of spice to Decartes.
    posted by Vibrissae at 1:10 AM on September 13, 2012


    SMBC
    posted by Tell Me No Lies at 1:25 AM on September 13, 2012 [4 favorites]


    WHOA, DUDE, ARE WE INSIDE A COMPUTER RIGHT NOW?

    I thought we were inside a Magic:The Gathering game?
    posted by Admira at 1:31 AM on September 13, 2012


    For example, a rock has no observable free will -- everything it does is entirely determined physical law. When you throw life into the equation things start to get complicated. Does a flower choose to unfurl its petals or are inexorable chemical processes driving the whole thing? We're pretty sure it's the latter.

    I dunno, rocks having free will seems just as plausible. Maybe the Andes are taller than the Appalachians because they have stronger moral fiber and simply chose not to erode. Perhaps the radioactive decay of an atom is orgasmic and every Uranium 235 atom wishes it could continuously be reincarnated as Carbon 11 so that it happens every few minutes instead of every few hundred million years.
    posted by XMLicious at 1:33 AM on September 13, 2012 [2 favorites]


    There are no measurements we can make, no logical arguments based on experiment that we can form that are relevant. This is because everything - the laws of nature, our memories, logic, experience - are all mutable and controllable in such a universe.

    Assuming active interference with experiments, this is true, but if we are in a simulation and the simulators are passive - the rules were set and left that way as the universe runs on and on - well, in computers there are ways to detect virtual machines from bare hardware. So theoretically in a long-distant future...

    From cstross's Accelerando:
    "And then there's the weirdness beyond M31: According to the more conservative cosmologists, an alien superpower – maybe a collective of Kardashev Type Three galaxy-spanning civilizations – is running a timing channel attack on the computational ultrastructure of space-time itself, trying to break through to whatever's underneath."
    posted by dragoon at 1:34 AM on September 13, 2012 [1 favorite]


    Yeah, so lag and local data constrains will mean the simulation can't be cohesive across sufficiently large, or sufficiently dense, areas of fabric, and realities will diverge.

    Now will someone please resynch the Romney egobot?
    posted by Devonian at 2:01 AM on September 13, 2012


    We would have absolutely no way of knowing what the properties of a simulator's world are.
    posted by Jernau at 11:01 PM on September 12 [+] [!]


    In a "simulations all the way down" scenario, each world (with the exception of the top-most world) would be both a simulation running in the world above it and - eventually - a simulator of the world(s) below it. We know enough (I think) about the properties of our world to say that any simulations we run would have to be less complex than our world, and so on from this point down in the world-chain. So there has to be a bottom.

    Incidentally this is what Raelians believe. Rael claims to have met the simulators (or rather their avatars in this world which is a simulation running in theirs).
    posted by L.P. Hatecraft at 2:25 AM on September 13, 2012


    As has been pointed out before, if we were living in a computer simulation, it would be fairly obvious because there would be constraints placed on the universe to keep the processing time from accidentally lagging too much. The universe might, for example, have a maximum speed, or physical properties which could only be measured down to certain computable tolerances. So, really, I wouldn't worry about it.
    posted by kyrademon at 2:26 AM on September 13, 2012 [20 favorites]


    There's been a tendency among scientists lately to light a bonfire and dance around it chanting "Quantum, Quantum, Quantum!" when faced with questions like this but I have to say that having looked closely I'm not seeing it.

    This is a feature, not a bug. It's impossible to determine where the bonfire is or how loud the chanting may be, so the soundwaves don't have enough information to be able to propagate. Instead they mill around looking sheepish.
    posted by talitha_kumi at 2:42 AM on September 13, 2012


    fuck weed, take speed
    posted by This, of course, alludes to you at 3:03 AM on September 13, 2012


    I have simulated on a computer
    the nasa scientist
    that was in
    the icebox

    and which
    you were probably
    saving
    for breakfast

    Forgive me
    he was delicious
    so sweet
    and so cold
    posted by sebastienbailard at 3:05 AM on September 13, 2012 [3 favorites]


    Dude, what if we are in a computer simulation, and the computer that is doing all the simulating is itself just another simulation inside a bigger super-computer, which itself is...
    posted by twoleftfeet at 3:13 AM on September 13, 2012


    Have not read through all the comments, but in case nobody has said it yet, this sounds like a variation on the conversation among stoned college students at 2:00 AM, Deep Thoughts about "Dude, nothing is real, nothing to get hung about....." I'm starving, let's go out to the all-night Taco Bell!
    posted by mermayd at 3:35 AM on September 13, 2012


    Why did I come in here looking for intelligent discussion of the topic? All the "have you really LOOKED at your hand" and skynet jokes are depressing.
    posted by DU at 4:31 AM on September 13, 2012


    Weak sauce. Here is a nice smack down of the--now looking rather trite--brain in a vat argument. Its by Diego Cosmelli and Evan Thompson.

    Brief extract:

    When we take into consideration the functional and structural interdependence of brain and body that evolutionary, developmental, physiological, and behavioral evidence suggest, then the philosopher’s naïve view of the brain in a vat simply will not do. The body is not just some kind of container, replaceable by a vat, that supports a commanding brain. The body is an active partner in the immensely complex and wide biological computations that the organism as a whole engages in while encountering an unpredictable world and maintaining its identity through time... Hence any “vat” capable of coupling with the brain in the requisite way must be able to duplicate these
    complex bodily processes.

    posted by stonepharisee at 4:32 AM on September 13, 2012 [3 favorites]


    Metafilter's own Hugh Howey wrote a terrific short book on the subject - The Plagiarist. I highly recommend it to any one still following the thread. It's a lovely riff on the idea of a world-within-a-world.
    posted by olya at 4:32 AM on September 13, 2012


    Hence any “vat” capable of coupling with the brain in the requisite way must be able to duplicate these
    complex bodily processes.


    The claim is not that our brains are real and the reality they perceive is simulated. The claim is that everything, including the brain, is simulated. Far from being a "smackdown" the above statement explicitly acknowledges that this would work.
    posted by DU at 4:44 AM on September 13, 2012 [1 favorite]


    Kevin Street: no, Planck units are not smallest values of things. The Planck mass is really quite big - it's about the mass of a flea.
    posted by edd at 4:51 AM on September 13, 2012


    Beg to differ. His argument rests on this fairly standard neurocentric view of consciousness:

    Unless you believe there’s something magical about consciousness—and I don’t, I believe it’s the product of a very sophisticated architecture within the human brain—then you have to assume that at some point it can be simulated by a computer, or in other words, replicated.

    The Cosmelli and Thompson argument would dispute this.
    posted by stonepharisee at 4:59 AM on September 13, 2012


    Yes, but the "brain in a vat" argument is NOT that view of consciousness. Descartes' thought experiment was that the brain was real while the reality was fake. That would indeed be hard to do. But if you can simulate the brain as well, you can omit a lot of the work by pretending to do it. The brain won't miss anything because the brain only expects what you tell it to expect and only perceives what you tell it it perceived.

    Telling a brain that it perceived no anomalies is much easier than presenting a reality in which a real brain perceives no anomalies.
    posted by DU at 5:09 AM on September 13, 2012 [4 favorites]


    Ummm lords and ladies but doesn't Descartes's conclusion of a verifiable reality rely on his notion of a benevolent God?
    posted by shakespeherian at 5:23 AM on September 13, 2012 [1 favorite]


    That was Descartes's solution, yes.
    posted by DU at 5:37 AM on September 13, 2012


    Yeah, they were probably thinking of "The Seventh Sally: Or How Trurl's Own Perfection Led To No Good" from The Cyberiad - but there was an Ijon Tichy "brain in a vat" story too.
    posted by Egg Shen at 5:37 AM on September 13, 2012


    Have you ever really looked at your hand?
    posted by chavenet at 5:37 AM on September 13, 2012


    I don't always Reddit, but when I do I like using Random. This subreddit seems relevant.
    posted by charred husk at 5:56 AM on September 13, 2012 [1 favorite]


    I love how many people basically recapitulate the least subtle parts of theistic philosophy when the simulator idea comes up:

    Of course the simulators would have to have the same concept of Evil and Good that we do, because we have those concepts.

    Of course the reason for the simulation is to simulate us; it's just obvious that the purpose of it is entirely anthropocentric and the vast span of galactic space around our tiny planet is just window dressing or a big skybox.

    Of course we were manufactured de novo in the Creator's image; the simulated universe was built around our brains (in jars), not the other way around.

    And so on.


    It's not too surprising, I guess, since the whole idea is philosophically identical to theism; all it really does is swap in a Simulator in place of God.

    In Young Earth Simulatorism we were created 8000 years ago with fossils and so on built in as the initial parameters. The Simulator used to spend a lot of time directly interacting with the simulation but has been on a snack break for the last 2000 years or so. In Big Bang Simulatorism the simulation is of the parameters of our basic physics; it was set in motion billions of years ago and we're an emergent phenomenon the Simulator may not even know about.
    posted by ook at 6:05 AM on September 13, 2012 [7 favorites]


    kyrademon: As has been pointed out before, if we were living in a computer simulation, it would be fairly obvious because there would be constraints placed on the universe to keep the processing time from accidentally lagging too much. The universe might, for example, have a maximum speed, or physical properties which could only be measured down to certain computable tolerances. So, really, I wouldn't worry about it.

    ...except that that does, in fact, seem to be the case. Quantization is a fundamental part of Quantum Mechanics.
    posted by mountmccabe at 6:22 AM on September 13, 2012 [1 favorite]


    The problem with the "Brains in jars" theory is that it is a dead end, at the end all we can do is agree with Descartes and move on.

    I once had a job salvaging iron pipe in the city morgue. There were indeed brains in jars, or parts of brains, and an entire stainless steel sink full of brains. I quit that job.
    posted by ennui.bz at 6:23 AM on September 13, 2012 [1 favorite]


    My other issue with Descartes's solution is that it assumes reliable memory of cognition.
    posted by shakespeherian at 6:30 AM on September 13, 2012


    What I want to know is, if this is a simulation, how can you tell the difference between an NPC and a player?
    posted by crunchland at 6:35 AM on September 13, 2012 [1 favorite]


    This is fascinatingly weird, it's like Descartes, the Matrix and The Hitchhiker's Guide, all at the same time. But if someone in the future 'designed our reality to simulate ancient history', that implies that there are two timelines, so to speak. The first is the one that actually accomplished that kind of computing power, and the second is our, supposedly simulated, one. Hmm. Sounds like... reality TV. And what's the nature of that original timeline? With 'simulations all the way down', we just end up with the same problem.

    Another thing though, I couldn't really follow his argument, especially the part where, y'know, he actually says, 'therefore!' - (Is it passé to ask if someone can explain what exactly he means?)
    ...And if it only behaves in a finite way when it’s being observed, then the question is: Is it being computed? Then there’s a mathematical parallel. If two things are mathematically equivalent, they’re the same. So the universe is mathematically equivalent to the simulation of the universe.
    posted by undue influence at 6:38 AM on September 13, 2012


    What I want to know is, if this is a simulation, how can you tell the difference between an NPC and a player?

    Easy. If nobody ever gives you quests, you're an NPC. If your friend repeats himself a lot, he's an NPC. (So I'm definitely an NPC.)
    posted by AugieAugustus at 6:39 AM on September 13, 2012 [1 favorite]


    mountmccabe ...except that that does, in fact, seem to be the case.

    Way to not get the joke.

    Oops.
    posted by mountmccabe at 6:43 AM on September 13, 2012 [1 favorite]


    We're living in a simulation of Metafilter in 2002.
    posted by vacapinta at 6:46 AM on September 13, 2012 [1 favorite]


    polychora: Just read this article, and I think it somewhat misses the point.

    Yeah, it's pretty lame. It seems to rest on a few assumptions that are not necessary parts of the "brain-in-a-vat" theory:

    1. That the vatted brain is physiologically similar to a human brain;
    2. That consciousness is distributed around the entire body;
    3. That no possible technology can cut into this distributed conscious system at any point and convince part of the system that another part of the system exists when it does not.

    They go into a lot of detail about 1 without ever getting around to explaining why this assumption is necessary, or even relevant. Why can't the vat-brain have a completely different structure, with the apparent structure of the human brain being an artefact of the simulated world?

    2 and 3 seem to be a bit of a stretch. Is a person who has lost a hand less conscious than a person who hasn't? Apparently yes. Could a prosthetic hand deliver the same sensory signals to the arm with the result that the rest of the organism operates exactly as if it still had a real hand? Apparently this is not merely technologically impracticable but epistemologically impossible. If it were possible, their argument would have failed to account for the further possibility that the entire body is a computer simulation of such a prosthesis.

    Not very convincing.
    posted by A Thousand Baited Hooks at 6:50 AM on September 13, 2012 [1 favorite]


    The Seventh Sally by Stanislaw Lem. I always liked Smullyan's Is God A Taoist? from the same collection.
    posted by urbanwhaleshark at 6:53 AM on September 13, 2012 [1 favorite]


    Whoa. Deja vu!
    posted by clvrmnky at 6:54 AM on September 13, 2012


    LOL unix jape!
    posted by Mister_A at 6:55 AM on September 13, 2012


    Allright, I'm going to solve this once and for all.

    *Jumps twice, squats twice, leans left, right, left, right*

    Uh, anyone know what "A" and "B" are?
    posted by mrgoat at 7:01 AM on September 13, 2012 [3 favorites]


    Uh, anyone know what "A" and "B" are?
    posted by mrgoat at 10:01 AM on September 13 [+] [!]


    We'll have that talk when you're a little older, son.
    posted by Uther Bentrazor at 7:09 AM on September 13, 2012 [4 favorites]


    Vibrassae:assuming Godel's incompleteness Theorem holds true

    What if GIT is an artifact of simulation, and it doesn't hold in 'real reality'?
    Whoa.
    posted by monkeymadness at 7:09 AM on September 13, 2012


    "Greg Egan's Diaspora"

    This is seriously one of my favorite books of all time. One of the greatest books of any kind, ever. Egan actually postulates a physically plausible rationale for a god/gods who actually need our belief and then in virtually the next paragraph basically disproves the notion of God entirely.

    It's a really phenomenal, very underrated book. At the outset, it looks like yet another cyberspace sci-fi thing, but it's SO not.
    posted by Eideteker at 7:11 AM on September 13, 2012 [5 favorites]


    Doesn't explain how the Universe always seems to know on which side the toast is buttered.
    posted by deo rei at 7:18 AM on September 13, 2012


    I once had a job salvaging iron pipe in the city morgue. There were indeed brains in jars, or parts of brains, and an entire stainless steel sink full of brains. I quit that job.

    Yea, like, really. Too much snack food everywhere!
    posted by Goofyy at 7:24 AM on September 13, 2012


    What if GIT is an artifact of simulation, and it doesn't hold in 'real reality'?

    That's a fun notion, but the incompleteness theorem strictly applies only to formal mathematical systems - i.e. systems of axioms and self-consistent rules build upon them. It's about pure abstract logic, and as such it's hard to envision how a simulated reality could have different laws of reason to a "real" reality.
    posted by phl at 7:29 AM on September 13, 2012


    People who think the Universe is a computer simulation are overestimating the power of computers and underestimating the complexity of the Universe.

    When gears and clocks were high-tech, the whole universe was supposed to be like clockwork. Now that computers are all the rage, they tend to be overused as metaphor: for the human brain, or for the Universe - "it's all a computer program!". This kind of thinking will surely go out of fashion, just like the "clockwork Universe" did.
    posted by crazy_yeti at 7:32 AM on September 13, 2012 [4 favorites]


    One of the best treatment's I've seen of this (though in a roundabout way) is Programming the Universe, by Seth Lloyd, who is a quantum mechanic at MIT. One of the things he touches on is the simulations of interactions on a quantum level and his argument is that to fully simulate the universe, it would require at least the (amount of matter and information in the) universe. It is an absolutely phenomenal read for anyone who thinks about these sorts of things; it really changed the way I looked at the world.

    "All interactions between particles in the universe, Lloyd explains, convey not only energy but also information—in other words, particles not only collide, they compute. And what is the entire universe computing, ultimately? “Its own dynamical evolution,” he says. “As the computation proceeds, reality unfolds.”

    Interview with Wired:
    W: Would it be fair to say the universe is a mind?
    SL: You could use that metaphor. And if you did, then you and I and my cat are its thoughts. But the vast majority of the universe's thinking is about humble vibrations and collisions of atoms.

    W: You've jumped from working on quantum computers to saying, oh, by the way, the universe is a gigantic quantum computer.
    SL: When you zap things with light to build quantum computers, you're hacking existing systems. You're hijacking the computation that's already happening in the universe, just like a hacker takes over someone else's computer.

    W: What is the universe computing when we are not hijacking it for our own purposes?
    SL: It computes itself. It computes the flow of orange juice as you drink it, or the position of each atom in your cells.

    W: Um, how many times have you seen The Matrix?
    SL: Sadly, only once. In The Matrix, what you see is fake - a simulation of bits - which is only a facade of what is real beneath it. But our universe is a simulation so exact that it is indistinguishable from the real thing. Our universe is one big honking quantum mechanical computer.

    W: You seem to be saying that the concept of the universe as one huge quantum computer is not just a metaphor - it's real.
    SL: Absolutely. Atoms and electrons are bits. Atomic collisions are "ops." Machine language is the laws of physics. The universe is a quantum computer.

    posted by nTeleKy at 7:34 AM on September 13, 2012 [4 favorites]


    phl, that's kind of my point. What if even our logic is restricted? It's something I've considered for years, and even as a practicing mathematician I'm not able to completely dismiss it. Could there exist a reality with alternate mathematical logic? Not just a different contrived system but a breakdown of modus ponens or something. Of course my gut says 'no', but it's fun to imagine.
    posted by monkeymadness at 7:35 AM on September 13, 2012


    When gears and clocks were high-tech, the whole universe was supposed to be like clockwork.

    Seriously. In actuality, the universe is like a giant Kindle Fire.
    posted by shakespeherian at 7:36 AM on September 13, 2012 [1 favorite]


    Maybe a sand box.

    I keep going back to Moving Mars, which has a fictional but us future world with human exploration to Mars pretty well settled into a colonial state (and the inherent issues of being so disconnected from Mother Earth, socially, politically, economically, scientifically).

    One of the ideas presented (spoliers!) is that the universe, in a sense, is an editable construct, and there is a way to adjust it, though it's not easy, nor are there no consequences. I think that's about as generic as I can get. I liked it, though I read it without reading the back and no idea what it was about (sometimes that pays off!) and never having read that author.
    posted by tilde at 7:49 AM on September 13, 2012


    In Accelerando, MeFi's own Charlie Stross has a throw-away (?) line about some hackers trying to run a timing attack on the universe itself to prove that it was a simulation. Now that's a neat idea. If this is a simulation, can we prove it from the inside?
    posted by It's Never Lurgi at 7:50 AM on September 13, 2012


    Consider the amount of information you take in via books, television, web sites, etc. about the world we live in, how it works, and everything going on. You have knowledge of places you've never been, history you weren't around for, a future we haven't achieved yet, the stories of thousands of people you've never met and never will, intimate details of celebrities' lives, the machinations of political parties in our country and others, the travels of voyagers I and II through space, interpersonal connections with people you've only connected with by reading their thoughts on a web site.

    Yet when you look at your day-to-day life, there isn't much to see, there isn't much to do, and there isn't much different -- substantially different -- from day to day, at least compared to what's going on in our heads. When their is a significant change, it is considered stressful (including positive stress) and you really can't take too much of it; mostly, you meander through a relatively simple life with relatively small boundaries, while your brain paints a picture of a huge and glorious universe that spans all space and time.

    In short, each and every one of us is running a simulation of the universe -- this larger universe all around us that we've never directly experienced -- inside our heads right now. Just sayin'.
    posted by davejay at 7:57 AM on September 13, 2012 [4 favorites]


    But the vast majority of the universe's thinking is about humble vibrations and collisions of atoms.
    Well, on the small scale, you've got vibrating atoms (described by quantum mechanics), and on the large scale you've got space-time structure - gravity - galactic clusters - black holes - the overall shape of the Universe (described by General Relativity). Nobody has yet reconciled these two theories, and it may be that they are mathematically incompatible. It may be that the Universe is not actually describable in terms of mathematics, at least not in a consistent way.
    posted by crazy_yeti at 7:59 AM on September 13, 2012 [1 favorite]


    People who think the Universe is a computer simulation are overestimating the power of computers and underestimating the complexity of the Universe. -- Consider this: if the universe is a simulation, the computer could take a million years to render each second, but the program only allows us to see the result pages... kind of like the movie Toy Story ... it took 30 hours to process and render each frame, but the end result only appears on the screen for 1/24th of a second in playback.
    posted by crunchland at 8:03 AM on September 13, 2012 [1 favorite]


    If this is a simulation, can we prove it from the inside?

    If this is a simulation, and we're about to prove it, what do the folks running the simulation do?
    posted by never used baby shoes at 8:04 AM on September 13, 2012 [1 favorite]




    If this is a simulation, and we're about to prove it, what do the folks running the simulation do?
    posted by never used baby shoes at 8:04 AM on September 13 [+] [!]


    According to the fictional universe cobbled together in Iain M Banks' The Algebraist, if enough of us become aware of the simulated nature of reality it might compromise the data, at which point the powers that be could decide to terminate the entire experiment.
    posted by Stagger Lee at 8:11 AM on September 13, 2012


    Interestingly, simply watching for 'hot spots' in the simulated universe, areas that are taking lots of computation time, should inevitably lead the implementors to interesting things happening in that universe.... our particle accelerators, if we're running on a simulation, would be producing some very, very strange requests for 'CPU time'.

    I wish to subscribe to your nnneeewwaaa *becomes agent*
    posted by Durn Bronzefist at 8:16 AM on September 13, 2012 [1 favorite]


    Perhaps in the next ten to 30 years we’ll be able to incorporate artificial consciousness into our machines.

    OK, but if consciousness is merely "the product of a very sophisticated architecture within the human brain" then wouldn't any consciousness we create in a machine be just as real as our own? Is artificial consciousness the best term to use? I get that he's playing off of artificial intelligence, but it seems like he's talking about something a little bit different here. Machines with "artificial consciousness" wouldn't just be mimicking human consciousness the way that artificial intelligence attempts to mimic human intelligence, they'd actually be self-aware, wouldn't they? Maybe I'm approaching this too philosophically; if someone with more knowledge of AI/machine learning can chime in and explain things for me I would love it.
    posted by asnider at 8:17 AM on September 13, 2012


    Also, I love The Algebraist.

    But where was I reading just a snippet of an idea about space ships off somewhere doing “reality testing”? It could have been a book, or a short story, or just an off-the-cuff imaginative remark on MeFi. Wish I could remember.
    posted by Durn Bronzefist at 8:18 AM on September 13, 2012


    This reminds me of the very funny take on this idea that the CBC radio sci-fi comedy series Canadia 2056. Worth checking out...
    posted by fimbulvetr at 8:18 AM on September 13, 2012


    When I was 14 I spent enough time thinking about this idea that I started believing it. Fortunately, this was in the 70's, and I soon discovered pot and LSD, which convinced me that the brain can believe almost anything it wants to and not to pay too much attention to it.
    posted by doctor_negative at 8:28 AM on September 13, 2012 [1 favorite]


    Anyone seen this movie, the matrix
    posted by MangyCarface at 8:31 AM on September 13, 2012


    It explains all the advertising.
    posted by SpacemanStix at 8:50 AM on September 13, 2012 [3 favorites]



    Of course the simulators would have to have the same concept of Evil and Good that we do, because we have those concepts.

    I would actually argue that this is unlikely as our current views of morality and ethical beliefs most likely have too wide a standard deviation in order to avoid a doomsday scenario for long as the pace technology continues to advance at, and our ability to destroy ourselves, is far more rapid than the pace of our ethical understanding and moral will to implement ethics held by our general population. I would argue however that some sort of ethical (rational system of beliefs) based morality (will to implement those beliefs) would be necessary to avoid a doomsday scenario.
    posted by Jernau at 9:03 AM on September 13, 2012 [1 favorite]


    The amusing bit are the minds that dream of one day uploading themselves to a non-corporeal state, never guessing that they worked dearly to pay for this brief vacation in the flesh...
    posted by Durn Bronzefist at 9:08 AM on September 13, 2012 [5 favorites]


    text got cut off in last post...

    I would argue however that some sort of ethical (rational system of beliefs) based morality (will to implement those beliefs) would be necessary in order to survive to the point of being able to become a simulator.
    posted by Jernau at 9:08 AM on September 13, 2012


    Creating beings (they would after all be real beings) who feel and experience pain, hardship and suffering and then exposing them to a world full of random pain, hardship and suffering is a very questionable way of acting. -- Jernau

    So you're an antinatalist, then?

    Most people implicitly reject this moral principle you suggest (by having kids); once it becomes possible to do it with computers, I don't see why we'd adjust our morals.
    posted by jepler at 9:19 AM on September 13, 2012 [1 favorite]


    Just noticed that an unreleased chapter of Programming the Universe which goes over the math/science behind the theory is available online.
    posted by nTeleKy at 9:21 AM on September 13, 2012


    East Manitoba Regional Junior Kabaddi Champion '94: We have no reason to believe that the universe is finitely rather than infinitely complex. If it's infinitely complex it's not computable. So we have no reason to think that the universe is capable of being simulated.
    You're assuming no shortcuts and assumptions were used in order to get the thing to run, or at least in order to get to go home this weekend instead of working through it.

    I, on the other hand, have worked in a lab.
    posted by IAmBroom at 9:37 AM on September 13, 2012 [3 favorites]


    So you're an antinatalist, then?

    Most people implicitly reject this moral principle you suggest (by having kids); once it becomes possible to do it with computers, I don't see why we'd adjust our morals.
    posted by jepler


    I'm not familiar enough with Antinatalists to subscribe to it as a philosophy. I will say I don't have children and my partner and I have no intent to have any, and that humans have a lot of built in biases that aren't necessarily rational, and could be potential stumbling blocks to reaching a post-human society capable of actually creating a simulation. A good example of how intrinsic selection based biases could hobble us before we can reach such a stage might be found in Famine, Affluence, and Morality.
    posted by Jernau at 9:42 AM on September 13, 2012


    blm: How would simulation creators get around the problem of evil (and good)? Creating beings (they would after all be real beings) who feel and experience pain, hardship and suffering and then exposing them to a world full of random pain, hardship and suffering is a very questionable way of acting.

    Greg Egan's short story Dust explores this very question.
    As does the Bhagavad Ghita.
    posted by IAmBroom at 9:50 AM on September 13, 2012 [3 favorites]


    From cstross's Accelerando:
    "And then there's the weirdness beyond M31: According to the more conservative cosmologists, an alien superpower – maybe a collective of Kardashev Type Three galaxy-spanning civilizations – is running a timing channel attack on the computational ultrastructure of space-time itself, trying to break through to whatever's underneath."


    YESSSS, that is it. Thank you.

    Also there needs to be way more allegory up in this cave.
    posted by Durn Bronzefist at 9:58 AM on September 13, 2012


    ceribus peribus: The problem with economic models is that they presume rational agents.
    Not always.

    Some fail because they're horseshit.
    posted by IAmBroom at 9:58 AM on September 13, 2012


    HUMANS WILL STOP THINKING ABOUT SIMULATIONS. HUMANS WILL CONTINUE UPDATING FACEBOOK STATUS AND POSTING ON TWITTER.
    posted by panboi at 10:01 AM on September 13, 2012 [1 favorite]


    Kevin Street: Then he looked up. And overhead, without any fuss, the stars were going out.
    Cite: Arthur C. Clarke - The Nine Billion Names of God

    See also: Jorge Luis Borges - Tlön, Uqbar, Orbis Tertius, and also his story The Circular Ruins.
    posted by IAmBroom at 10:04 AM on September 13, 2012


    Where's Philip K Dick? He's gotta check this out.
    posted by Liquidwolf at 10:12 AM on September 13, 2012


    Liquidwolf: Where's Philip K Dick? He's gotta check this out.
    He's outside, looking in.
    posted by IAmBroom at 10:32 AM on September 13, 2012


    Liquidwolf: Where's Philip K Dick? He's gotta check this out.

    He's outside, looking in.


    You mean...................................... he's the programmer. Mind= Blown
    posted by Liquidwolf at 10:43 AM on September 13, 2012


    Of course the simulators would have to have the same concept of Evil and Good that we do, because we have those concepts.

    I would actually argue that this is unlikely as our current views of morality and ethical beliefs most likely have too wide a standard deviation in order to avoid a doomsday scenario for long as the pace technology continues to advance at, and our ability to destroy ourselves, is far more rapid than the pace of our ethical understanding and moral will to implement ethics held by our general population. I would argue however that some sort of ethical (rational system of beliefs) based morality (will to implement those beliefs) would be necessary to [survive to the point of being able to become a simulator.]


    I think that that line of thinking is tied to a situation that has only been relevant in our society for the past ~100 years, and it's quite a leap to assume it necessarily holds true in a space about which we can know nothing (except possibly for pure mathematics).

    Assuming we're in a simulation, we have no way of knowing whether the physical laws it is embodied in even allow for "doomsday scenarios". "Hey, I'm running this cool simulation in which I model an imaginary concept I call 'entropy', and in which the timelike dimension is only unidirectional! Kind of unstable, but pretty interesting, eh?"

    Even assuming the simulation matches the same basic physical laws as the "real" world outside, it seems rather limited to assume that systems of ethics and morality that would be at all recognizable to us are the only possible way for a society to develop sufficient technology, or that it is necessarily a society that would be at all recognizable to us as such, or that it is necessarily technology as we think of it.

    (F'rinstance: Greg Egan's Diaspora has been rightly namechecked quite a few times here already -- there's an entity in it dubbed "Wang's Carpet" which plausibly shows an entire ecosystem naturally evolving as simulation within, basically, a layer of pond scum.)


    This kind of thing is the root of what annoys me about this whole simulationist argument -- once you get past the basic idea that it's possible -- which, sure, it is -- it doesn't tell you anything useful at all. Any theorizing we could do is the equivalent of counting angels on pinheads: we're just making it all up based on our own provincial prejudices.
    posted by ook at 10:44 AM on September 13, 2012


    ....and then there's the question of how did Kelly LeBrock escape her confines in order to help out Anthony Michael Hall and the other dude?

    I guess what I'm asking is, should I wear a bra on my head?
    posted by not_on_display at 10:53 AM on September 13, 2012


    There was any question about this at all? The minute you can simulate atoms all of that becomes possible.

    If you simulate photosynthesis in a computer do you have real photosynthesis?

    These computational consciousness theories make no sense to me. Numbers and computation are phenomenal. In a computer there is just a bunch of electrons moving around in Silicon. Nothing in a computer understands anything. How would you compute the experience of the color red? If you do the right combination of multiplications, shifts, and divisions or whatever, do the "numbers" start experiencing color? Or the silicon in the ALU and the accumulators start experiencing color?

    It's so annoying to me that NASA's name has to be put on bogus science fiction like this. NASA is supposed to represent science. How about trying to understand what is truly happening in the brain first. Compared to what scientists will know in a few hundred years, I doubt scientists today know hardly anything of what there is to know about an individual neuron or synapse even.
    posted by Golden Eternity at 11:01 AM on September 13, 2012 [1 favorite]


    Major spoilers for Permutation City follow:

    Man Greg Egan sure takes this stuff pretty far - Permutation City isn't really even science fiction... it's more ... number theory fiction?

    My favorite part of that story is how they create a simulated universe in a computer by fiat, but when they try to communicate with the simulated inhabitants, the inhabitants refuse to believe in their "creators" and come up with an alternate explanation for their existence which becomes retroactively true.

    It's like the opposite of the creationist idea that 4000 years ago God created from nothing a billions-of-years-old universe. In this case a "young" created universe retroactively becomes older than it really is (was?). Egan sets us up to think we're going to see an intelligent species deal with the idea that they were created in a lab experiment as entertainment for (relative) immortals, only to have them turn the ontological tables and actually become more real than their creators.

    (Sometimes I suspect Egan of being a closet theist writing reductio ad absurdum refutations of materialism.)
    posted by straight at 11:06 AM on September 13, 2012 [1 favorite]


    One thing that I haven't seen in this thread is the notion that the universe could be a human-centric simulation or even a me-centric simulation.

    This would allow for a significant reduction of complexity in terms of whatever the "hardware" and/or "software" is that contains and runs the sim. If the purpose of the simulation was to simulate humans on earth or just me then you could do clever culling of detail depending on location and sensory cones of us all (or just me).

    You only increase resolution for all variables where humans are or where I am just like game engines often only load environmental detail if you go to a particular place. So if a telescope is pointed into a particular direction... you increase detail there, if a probe is sent to Mars... you increase detail there. The rest would consist of coarse voxels with characteristics that are reasonable averages of the detail that would emerge if you looked a little closer.

    In conclusion: this is all about me and you're all just a bunch of strings popping up on a blue background.
    posted by Hairy Lobster at 11:06 AM on September 13, 2012 [1 favorite]


    Golden Eternity, do you believe it's theoretically possible for a computer to simulate a single neuron? That is, create an interface so that a machine receives the chemical input that a neuron receives and uses a computer model to create the exact same output that a neuron would? And that you could possibly use this whole setup to replace an actual neuron in vivo? Would a person with a single artificial neuron still think and feel and have subjective experiences?

    If so, then what happens when you simulate two neurons? Ten neurons? A thousand neurons? What if you could replace half the neurons in someone's brain with artificial neurons. Would that person still think and feel? What if you replaced all the neurons in a person's brain with simulated neurons?
    posted by straight at 11:14 AM on September 13, 2012


    If you simulate photosynthesis in a computer do you have real photosynthesis?

    Within the confines of the simulation, yes. That's sort of the whole point: that "real" isn't necessarily a bedrock concept; a bunch of electrons moving around in silicon isn't necessarily fundamentally different from whatever it is that our world is embodied in.

    Nothing in a computer understands anything. How would you compute the experience of the color red?

    Substitute "a physical brain" for "a computer" and you've succinctly described exactly the unanswered question of how consciousness works at all. How do neurons embody the experience of the color red? Is there a reason electrons in a computer would or would not be able to embody the same experience? (Answer: Reply hazy, ask again later)
    posted by ook at 11:21 AM on September 13, 2012


    Liquidwolf: Where's Philip K Dick? He's gotta check this out.

    IAmBroom: He's outside, looking in.

    Liquidwolf: You mean...................................... he's the programmer. Mind= Blown


    I think the programmer didn't expect him to escape.
    posted by tilde at 11:27 AM on September 13, 2012


    If this is a simulation, and we're about to prove it, what do the folks running the simulation do?

    High fives all around as they just took first place in the time trials at the Fifth Annual Consciousness Creation Competition at their university, beating out two teams of grad students. One teams's simulation is just entering Industrial Revolution and the other is hung up in something it calls the Snarg Age.

    Next up, the Nationals!
    posted by Tell Me No Lies at 11:29 AM on September 13, 2012 [3 favorites]


    If this is a simulation, and we're about to prove it, what do the folks running the simulation do?

    They start to realize that they're also inside a simulation, and so and on.
    posted by Liquidwolf at 11:37 AM on September 13, 2012


    I think that that line of thinking is tied to a situation that has only been relevant in our society for the past ~100 years, and it's quite a leap to assume it necessarily holds true in a space about which we can know nothing (except possibly for pure mathematics).

    First, existential threats have existed for a very long time. Society created, civilization destroying events are also not new, look at the Romans and the dark ages.

    As far as us not knowing about the simulators, your absolutely correct. We can't really know much about them, I said as much about 150 comments ago. That doesn't mean we can't try and apply the concepts of the hypothesis and attempt to use it as a gateway into understanding a potential post humanity for ourselves. In other words, well we can't know anything about simulators, but if we were potential simulators how would we get there and once there how would we do it.
    posted by Jernau at 11:37 AM on September 13, 2012


    If so, then what happens when you simulate two neurons? Ten neurons? A thousand neurons? What if you could replace half the neurons in someone's brain with artificial neurons. Would that person still think and feel? What if you replaced all the neurons in a person's brain with simulated neurons?

    To me, this will be the only way we can ever really settle many of these questions: an experiment to do exactly what's suggested above. Replace an active, living brain's function, neuron by neuron, until there's nothing left but simulated neuronal activity. Of course, you'd probably have to do this on animal test subjects first. Or the immediately terminally ill. Not sure how you could perform the subsequent follow-up testing to confirm the test subjects' minds have made the transition intact. If you could figure out a way to wire up a toad's body to a remote controller running a simulated brain maybe...
    posted by saulgoodman at 11:44 AM on September 13, 2012 [1 favorite]


    Golden Eternity, do you believe it's theoretically possible for a computer to simulate a single neuron?

    God knows. I'm sure you could simulate many different aspects of a neuron, but the simulations would not be neurons, anymore than a simulation of photosynthesis actually converts physical carbon dioxide into physical glucose with the aid of photons, or a 24 bit SRAM location in a computer containing 3 bytes (R=255, B = 0, G = 0) is the color red.

    Maybe someday they will find the exact isomorphism between physical states in the brain, and conscious experiences such as color, when/if they do I am thinking the isomorphism will not be to a computation (perhaps "computation" is itself essentially phenomenal), it will be to something physical. Maybe not something "magical" as the author states, but something probably very much unknown, currently.
    posted by Golden Eternity at 11:44 AM on September 13, 2012


    Is artificial consciousness the best term to use?

    Artificial sweeteners still make things sweet.
    posted by aubilenon at 11:54 AM on September 13, 2012 [1 favorite]


    Well, I just asked my Magic 8 Ball 6 times in a row "are we living in a computer simulation" and the only answers I got were "yes" "it is decidedly so" "as I see it yes" or "without a doubt, so that settles that. Next question?
    posted by fimbulvetr at 12:01 PM on September 13, 2012


    You only increase resolution for all variables where humans are or where I am just like game engines often only load environmental detail if you go to a particular place.

    I used to think, if we could find out when waveform collapse happens, it would be a nifty test for consciousness. Just point your AI/dolphin/advertising exec at it and later read the result. Now I think Schrodinger was just a canny game/universe designer.
    posted by Durn Bronzefist at 12:06 PM on September 13, 2012 [1 favorite]


    We must believe in free will. We have no choice. -- Isaac B. Singer

    Anyone who is interested in the question of "free will" should read about the Conway-Kochen "Free Will" theorem.

    This shows, rigorously, that "if some experimenters are able to behave in a way that is not completely predetermined, then the behavior of elementary particles is also not a function of their prior history". In other words, if we have free will, so do particles. Now this might be used as a reductio ad absurdum argument against free will, but Conway (yeah, that Conway) takes it in the other direction - we ALL have free will, people and particles. (As a reader and admirer of Schopenhauer this resonates with me.)

    Quote:
    When the floor was opened for questions, one member of the audience questioned Dr Conway's use of the term "Free Will". She asked whether Dr Conway was "confusing randomness and free will".

    In a passionate reply, Dr Conway said that what he had shown, with mathematical precision, that if a given property was exhibited by an experimenter than that same property was exhibited by particles. He had been careful when constructing his theorem to use the same term "free will" in the antecedent and consequent of his theorem. He said he did not really care what people chose to call it. Some people choose to call it "free will" only when there is some judgment involved. He said he felt that "free will" was freer if it was unhampered by judgment - that it was almost a whim. "If you don't like the term Free Will, call it Free Whim - this is the Free Whim Theorem".

    In concluding Dr Conway said that he believed he did have free will. Holding up a piece of chalk, he said he felt he could choose whether or not he would drop it or continue to hold it. His theorem he said leads him to accept that the universe is teeming with free will. He also said that while he did not have any proof for it, he believed that the cumulative free will of particles is the source of his free will as a person.


    Now, if someone can reconcile this with simulationism, I'll be impressed.
    posted by crazy_yeti at 12:10 PM on September 13, 2012 [2 favorites]


    Ook says, "I love how many people basically recapitulate the least subtle parts of theistic philosophy when the simulator idea comes up. . ."

    Indeed. The overwhelming flaw in so much of this discussion is the assumption that *we* must be important in some objective way.

    It's the same flaw which cripples any anthropic arguments except boring tautologies. The universe may be perfectly fine-tuned for us, but it's also perfectly fine-tuned for all sorts of specific chemical and physical processes which, measured in energetics, information content, or visibility, play a much bigger role in the observed universe than we do.

    Leaving aside the question of how original Terrile's statements are, the real problem is that they're so limited. Sure, we could be a simulation of the ancient past of a civilization that grew from one just like ours in a universe just like ours. But, why would anyone assume that? (Except, of course, that it's a fun idea.) Of all the entities that could run a simulations, and all the things they could simulate, it's quite a leap to assume they're like us and we're the point of the thing. Sure, people play Sim City. But, they also play pacman and mine-sweeper. (As well as games such as the Millenium Run and "find the next prime number.")

    Though, I suppose that if we *are* the point, it makes it feasible to either communicate with our simulators or to hunt for intentional clues. That's kind of neat, and certainly it makes for more interesting fiction.

    If one hopes to say something about reality, though, it seems just as likely that somewhere a grad-student from some super-advanced civilization is going to be pissed off when the result from her batch jobs are returned and she realizes that space-faring life has contaminated her study of dust grain formation in different stellar environments.
    posted by eotvos at 12:10 PM on September 13, 2012


    Shit. I am just a computer simulation. What to do now? Oh, I know. Just keep going through with this experience called life.

    Changes nothing.
    posted by thetoken at 12:16 PM on September 13, 2012 [1 favorite]


    These computational consciousness theories make no sense to me. ... Nothing in a computer understands anything. How would you compute the experience of the color red? If you do the right combination of multiplications, shifts, and divisions or whatever, do the "numbers" start experiencing color? Or the silicon in the ALU and the accumulators start experiencing color?

    The problem is that nothing we can point to in a human brain -- on the scale of a computer circuit -- understands anything either, or otherwise does a thing that computers categorically can't do. So if you say that a brain understands things and a computer can't, you're assuming a difference between brains and computers that no one has demonstrated yet. It's like saying that a building could never be built as high as the moon. Maybe it's true, maybe it's not, but we currently know of no physical laws preventing it -- we know of nothing that exists in a brain that couldn't exist in a computer.

    I wonder if people invoke this hypothetical physical property or law because it's so fundamentally disruptive to separate consciousness from individuality. I mean, there can't be anything more fundamental to our history than the fact that each person exists in one place at a time, continuously, in an identifiable physical form, and then is gone.

    If you assume that brains are merely physical structures that could be replicated, those assumptions stop applying. To wit:

    (1) Let's say we scan an adult human brain and nervous system into a computer and simulate its electrical and chemical interactions in realtime, the way we can simulate a small network of rat neurons now. And let's say we make a robot body with video cameras and microphones and pressure-sensitive skin and motorized limbs and so on, so we can hook up all those inputs and outputs to the simulated brain and nervous system, the same way we hook up video cameras to the nerves of blind people now. And let's say that when we turn this thing on, the simulated brain is capable of controlling the robot body, having conversations, walking around museums, etc. The simulation is successful enough that people who know the original person whose brain was copied are unable to tell the difference during a text-based conversation. Both the person and the simulation claim that they experience the color red.

    This may or may not ever happen, but we don't know of any physical laws right now that make it impossible. So you can't rule out that such a thing might one day exist in reality.

    So suppose that simulation did exist. Does the simulated brain in that case really experience the color red? Or is it lying and doing something fundamentally lesser than you do when you experience the color red -- like a Speak'n'Spell that can say "I experience the color red!" but isn't doing anything of the kind? If so, what makes the original person's experience legitimate and the simulation's fake?

    (2) Now suppose we think the simulation does experience the color red -- or at least its friends and family are convinced. But suppose that instead of being hooked up to a robot, we now hook its inputs and outputs up to an avatar in Second Life. So instead of a robot pointing at a painting in a museum and saying that it is red, we have a virtual avatar pointing at a virtual painting in a virtual museum and saying that it is red. And maybe the person whose brain was scanned could also bring their Second Life avatar to the same virtual museum, and point at the same virtual painting and say that it is red. Is one doing something different from the other?

    (3) And now suppose that instead of connecting to Second Life, the simulation connects to its own private server, where it can point at its own private painting in its own private virtual museum. It says "that painting is red," but only it can hear. We can verify that the simulation is still running on the computer -- the only difference from the Second Life scenario is that no human is monitoring what is being simulated. Is that no longer real, even though it's exactly the same thing the simulation did with its avatar in Second Life, which is exactly the same thing it did with its robot in the real museum? When did the change occur?

    (4) And finally, let's say we decide that the experience of the ongoing simulation in the computer is no more or less real than the person it was copied from has when they're playing Second Life. Suppose we freeze the state of the computer, copy its data to pencil and paper, and run the same calculations by hand, taking many years per cycle. Is that less real than the computer simulation? Suppose after 40 years, we copy the paper back into a computer, and they start walking around as a robot again with an extra second of memory from the time they spent on paper. Is their remembered experience from when they were simulated on pencil and paper real? If so, would someone who burned up the paper be guilty of murder? Does it matter whether there's a backup copy?

    When you get all the way there -- that the conscious me could be reduced to numbers on paper and photocopied, live its life that way and then re-emerge into the world with memories of the simulated world intact, and that in fact there would be no identifiable difference between the me on paper and me in a locked room -- the assumptions we live by are so fundamentally undercut that it's natural to assume something must have gone wrong along the way. It's natural to propose that there must be some fundamental, unknowable difference between the me in the room and the me on the paper. But I don't think we've seen evidence of that difference so far.
    posted by jhc at 12:19 PM on September 13, 2012 [12 favorites]


    aubilenon: Artificial sweeteners still make things sweet.
    Yes, but as you may remember from this recent thread, they also make people really sick.
    posted by crazy_yeti at 12:26 PM on September 13, 2012


    crazy_yeti,

    Looking over at an insulin stick...so do natural sweeteners.
    posted by effugas at 12:38 PM on September 13, 2012


    Shit. I am just a computer simulation. What to do now? Oh, I know. Just keep going through with this experience called life.

    Changes nothing.


    This is precisely why I gave up (after 12 years including a painfully philosophical adolescence) on Solipsism. Sure it could be true, but so what? It's *boring*.
    posted by Tell Me No Lies at 12:45 PM on September 13, 2012 [1 favorite]


    jhc: It's like saying that a building could never be built as high as the moon. Maybe it's true, maybe it's not, but we currently know of no physical laws preventing it

    You might want to talk to an architect or a structural engineer about that.

    Let's say we scan an adult human brain and nervous system into a computer and simulate its electrical and chemical interactions in realtime, the way we can simulate a small network of rat neurons now. And let's say we make a robot body with video cameras and microphones and pressure-sensitive skin and motorized limbs and so on, so we can hook up all those inputs and outputs to the simulated brain and nervous system, the same way we hook up video cameras to the nerves of blind people now. And let's say that when we turn this thing on, the simulated brain is capable of controlling the robot body, having conversations, walking around museums, etc.

    And let's say that's a mighty tall order!!!
    posted by crazy_yeti at 12:45 PM on September 13, 2012


    When you get all the way there [...] the assumptions we live by are so fundamentally undercut that it's natural to assume something must have gone wrong along the way.

    Agreed. As hinted in my last comment, I think it goes wrong right at the first step.
    posted by crazy_yeti at 12:49 PM on September 13, 2012


    jhc, that's really well put. I don't think there's anything wrong with your logic once we assume materialism. I'm not willing to grant you that, but loads of people will. Reminds me of this and this.
    posted by monkeymadness at 12:58 PM on September 13, 2012 [2 favorites]


    It's *boring*.

    My argument against solipsism, and to a lesser degree a simulated existence, is that if this was my universe there would be spaceships in it.
    posted by thecaddy at 1:02 PM on September 13, 2012


    Indeed. The overwhelming flaw in so much of this discussion is the assumption that *we* must be important in some objective way.

    This assumes that the argument is entirely human based. I don't think it is. The term Post-Human could apply to anything with greater processing capacity comes after humanity. Only some form of advanced sentience seems to be an actual requirement. I always try and use the term being(s) when I talk about this. The concept still works in a non-human centric equation. Look at Bostrom's argument and you'll see.
    posted by Jernau at 1:09 PM on September 13, 2012


    "if some experimenters are able to behave in a way that is not completely predetermined, then the behavior of elementary particles is also not a function of their prior history" ... Now, if someone can reconcile this with simulationism, I'll be impressed.

    I think there's two different questions here -- can you simulate atoms with math, and can you simulate atoms with other atoms?

    If you have a system that involves taking a bunch of symbols and applying logical processes to them with no external input, and the current state is always a logical consequence of the prior state, then yeah, by definition you can't perfectly simulate a system in which the current state is not a function of the prior state.

    But our simulator isn't going to be made of math. It's going to be made of ... elementary particles, just like the experimenter in your example. So we only have a rigorous logical problem if it is rigorously, logically impossible for a computer to modify its calculations based on the behavior of some of the elementary particles it's made of.

    Otherwise, we have an engineering problem -- how much non-deterministic behavior of what kind do you have to include in your simulation in order to get useful results? Can we measure the motion in a really hot cup of tea, duplicate that randomness throughout our artificial cosmos and be done with it? Or do we need so much randomness that we run into some kind of physical limitation trying to collect it? Or is the randomness not actually that important, and quantum pseudo-non-determinism generates results that are just as good on a macro scale?

    I seriously question whether you can logically prove that it's impossible to simulate the behavior of atoms, using more atoms. And if you think it's cheating to use the behavior of atoms as an input to the simulation, I question whether you can prove that the behavior of particles is and must be truly non-deterministic (as opposed to, say, the product of a pseudo-random-number generator that spits out a behavior for each particle in turn).

    As hinted in my last comment, I think it goes wrong right at the first step.

    I mean, I get that. You seem to (forgive me if I'm overstating) firmly believe that some quality of the human brain makes it physically impossible to simulate. Lots of people do. But there's no real evidence for (or against) that quality existing. I used the example of believing that, no matter how much time and material you had and what technology you used, it would be impossible to build a tower to the moon. We know we can't do it now. And maybe there's some quality of matter that means it can't ever be done. But we don't know of one right now -- we just know that we don't have fine enough control over enough stuff to try it out.

    This is in contrast, say, to the idea that you can't instantly teleport to another place; experiments show that's impossible unless we're wrong about the physical laws of the universe. There is no experiment showing that you can't simulate a brain unless we're wrong about the laws of the universe. We have no way right now of saying, yep, we just tested this thing and it can't ever work.

    So if we can't answer the question right now, and if people are going to pick an answer to firmly believe in without evidence, here's my real question: why choose in particular an undiscovered physical law that prevents computers from doing what human brains do? Why not firmly believe that the randomness of a random number generator equals that of a hydrogen atom? Or if you believe in an immaterial quality of the human mind, why not believe that when a computer becomes sufficiently complex it attracts an immaterial quality of its own, just as the immaterial soul you carry with you must at some point have taken root in your brain?

    If you believe in an immaterial element beyond the reach of science to predict, explain, analyse or simulate -- how can you also make confident predictions about where and when it appears and what it can and cannot do?
    posted by jhc at 1:45 PM on September 13, 2012 [2 favorites]


    monkeymadness -- yeah, once I got to step four I realized I was kind of retreading that XKCD comic. It was fun to work my own way through though.
    posted by jhc at 1:48 PM on September 13, 2012


    Metafilter: "Nobody that's anybody really thinks it exists"
    posted by illovich at 1:50 PM on September 13, 2012


    First, existential threats have existed for a very long time. Society created, civilization destroying events are also not new, look at the Romans and the dark ages.

    A change in government for a small part of the globe, and a brief slowdown in technological progress for a different small part of the globe. That you and I are using this medium to discuss this is a reasonably good existence proof that those weren't "civilization destroying events".

    Existential threats: sure, we could all get wiped out tomorrow by an asteroid or something, but I don't see that that tells us anything about the requisite nature of a hypothetical entity who may be simulating us (and the asteroid.)

    We can't really know much about them, I said as much about 150 comments ago. That doesn't mean we can't try and apply the concepts of the hypothesis and attempt to use it as a gateway into understanding a potential post humanity for ourselves.

    Which is exactly what I've been complaining about all this time: projecting our own particular circumstances onto a Creator (whether theist or simulationist) and declaring he/she/it must necessarily share them. We can't know anything about them, but we're going to think about them as just like us anyway, because.

    I see no reason to believe that any entity -- theistic, nonhuman, or posthuman -- capable of simulating or creating our universe would necessarily hold to an ethical or moral system that we would recognize. Posthumans, maybe, if only out of tradition... although I have reservations about even this: most of our ethics revolves around managing scarcity and coping with our mortality; neither are likely to be relevant issues for entities capable of simulating entire universes (including, presumably, themselves.)
    posted by ook at 1:58 PM on September 13, 2012 [1 favorite]


    Metafilter: "Nobody that's anybody really thinks it exists"

    Well *I* won't be part of this existential sham.
    posted by Tell Me No Lies at 2:02 PM on September 13, 2012


    In case anyone is listening: No. Sorry about the fishing industry, especially the thing with the tuna nets because, wow ... clearly we weren't thinking that one through. And, oh yeah, apologies for Sea World.

    BTW, don't suck up to the dolphins. They're dicks.
    posted by Tell Me No Lies at 2:09 PM on September 13, 2012


    jhc: But our simulator isn't going to be made of math. It's going to be made of ... elementary particles, just like the experimenter in your example.

    Right above, you said: "Suppose we freeze the state of the computer, copy its data to pencil and paper, and run the same calculations by hand, taking many years per cycle." That sure sounds to me like it's made of math. Some set of calculations which could be run on a computer, or by hand.

    Now, using some sort of analog/quantum computer to simulate our physical reality is less troubling to me. But, if I want to simulate a hydrogen atom, why not just use a proton and an electron to build the simulator? See what I mean? Why build this other, more elaborate structure "behind the scenes" as it were, a "hidden conspiracy"? I invoke Occam's Razor...

    You seem to (forgive me if I'm overstating) firmly believe that some quality of the human brain makes it physically impossible to simulate.

    Well, not exactly the brain... I believe some quality of human consciousness is impossible to simulate. The subjective, experiential aspect. The "suchness of the such".

    I realize, this opens a lot of doors. Is our consciousness more than just the physical processes of our nervous system? I don't exclude that possibility. Also, we really don't yet understand how the brain works, other than at the most rudimentary level. Not like the way we understand, say, internal combustion engines or CPUs. So I'm a little wary of any statements about the human brain. Maybe there are some kind of effects like large-scale quantum coherence. (I realize the Penrose-Hameroff "microtubule" theory has never found favor, but I think there's a possibility that something like that might be happening). Either that, or else maybe there really is a "soul".

    I realize that at some point this becomes a religious argument, and I'd like to avoid that.

    jhc, I have respect for your opinions and arguments, but I just have different inclinations. I reject strict materialism, and have vaguely panpsychist leanings.

    But more than anything else, I think the Universe is incomprehensibly more complex than any of our feeble attempts to understand it. We're just dogs howling at the moon.
    posted by crazy_yeti at 2:16 PM on September 13, 2012 [1 favorite]


    Now, using some sort of analog/quantum computer to simulate our physical reality is less troubling to me. But, if I want to simulate a hydrogen atom, why not just use a proton and an electron to build the simulator? See what I mean? Why build this other, more elaborate structure "behind the scenes" as it were, a "hidden conspiracy"? I invoke Occam's Razor...

    Two things: 1) What approach for simulating a hydrogen atom is best depends a lot on what you're trying to learn from the simulation. 2) If you use a proton and an electron you are just building a hydrogen atom, not a simulation, and thus you lose whatever benefits you were hoping to gain from using a simulation (e.g., slowing down time, isolation from gravity, possible to observe without affecting, etc)
    posted by aubilenon at 2:38 PM on September 13, 2012


    So our entire physical reality, instead of just being made out of the matter it seems to be made out of, is made out of some other matter, so that some alien researchers can slow down time and study it? That's a strange cosmology. You are free to believe in it of course. That may be what's going on, and the joke is on me.
    posted by crazy_yeti at 2:44 PM on September 13, 2012 [1 favorite]


    If this is a simulation, can we prove it from the inside?

    I've been working on that for several years, and I'm just about to fini
    posted by no relation at 2:45 PM on September 13, 2012


    "These computational consciousness theories make no sense to me. Numbers and computation are phenomenal. In a computer there is just a bunch of electrons moving around in Silicon. Nothing in a computer understands anything. How would you compute the experience of the color red? If you do the right combination of multiplications, shifts, and divisions or whatever, do the "numbers" start experiencing color? Or the silicon in the ALU and the accumulators start experiencing color?"

    Experience is related to the question of the existence of a self. To speak of an experience suggests that there should a subject/self having an experience. As far as I can gather from what I've read over the last few years in terms of scientific literature and articles related to the fields of the neuro sciences as well as my own subjective experiences this is not necessarily the case. Historically Buddhist philosophy has intuited the probably emergent and non-constant nature of what we refer to as our self but lately science appears to be increasingly point into a similar direction as well. If self is an emergent phenomenon than the exact nature of the system it emerges from shouldn't matter.

    It's so annoying to me that NASA's name has to be put on bogus science fiction like this. NASA is supposed to represent science. How about trying to understand what is truly happening in the brain first. Compared to what scientists will know in a few hundred years, I doubt scientists today know hardly anything of what there is to know about an individual neuron or synapse even.

    Considering the types of brain/machine interfaces that have been emerging from the labs over the last few years I'm not sure we're as far away as you seem to think. There's mind controlled prosthetics, cameras allowing blind people to see, increasing precision when it comes to "mind reading" (remember the video created from a readout of a cat's visual cortex? There was a recent report about some success in identifying letters or words someone thought of... can't recall the details unfortunately) etc etc.

    I think we're about to experience a whole wave of mind/machine interfacing that will shed a lot of light on all these questions.
    posted by Hairy Lobster at 2:45 PM on September 13, 2012 [1 favorite]


    Ook: Societies rise and fall, just because they are all human doesn't make them all the same society.

    Bostrom's simulation hypothesis differs greatly from the standard brain in a vat, evil demon versions in that it operates from our current place in time forward rather than some unknown point. Bostrom's work operates on a post human basis. Is it possible a flying spaghetti monster created the simulation? Sure why not, But what I referenced in my original post was Bostrom's hypothesis. As we've both pointed out repeatedly we have no other valid starting point to work from.

    Also I don't see the connection to theism here, certainly not in a traditional sense. The simulation argument no more shares the concept of religion based magical creator than you would with your parents. if anything it's less so given the amount of non-involvement a potential simulators would seem to have in our lives.
    posted by Jernau at 2:57 PM on September 13, 2012


    Also ook, most ethics revolves around interaction between beings. Scarcity and resource management is a level or two above that.
    posted by Jernau at 3:00 PM on September 13, 2012


    blm: "Think of your current computer, riddled with viruses, Easter eggs, programming errors, random memory glitches, etc. Now magnify those to the size of a computer capable of simulating what we perceive as the universe, and the software running it. Maybe that's what we perceive as UFOs, the Loch Ness monster, ESP, miracles, etc.

    Just a thought...
    "

    Don't forget Republicans.
    posted by symbioid at 3:40 PM on September 13, 2012


    Everybody talks about the simulation, but nobody does anything about it.
    posted by Twang at 4:21 PM on September 13, 2012


    Ook: Societies rise and fall, just because they are all human doesn't make them all the same society.

    True. It's not at all clear to me why you think that's relevant to this conversation, however. If you're trying to relate this to Bostrom then the only sort of collapse that matters is one complete enough to prevent that species from ever becoming capable of developing simulation-level technology; the rise or fall of a single society hardly rates.

    Bostrom's simulation hypothesis differs greatly from the standard brain in a vat

    I think you're wildly overinterpreting Bostrom. He's pretty much just doing a Drake's Equation for simulationism: his hypothesis doesn't say anything at all about what the simulators would be like, or whether they are human-derived or otherwise. All it is is an attempt to quantify what the unknown variables are. (Which, as with Drake's equation, turns out to be most of them.)

    most ethics revolves around interaction between beings.

    That's a terribly vague statement, so I'm not clear exactly what you're trying to get at... and I'm hard put to think of any significant interaction which requires ethics or morals yet does not relate to mortality or scarcity. But even assuming such exists, practically any interactions between members of a simulation-capable society would be so different from our own as to be basically unrecognizable.

    Also I don't see the connection to theism here, certainly not in a traditional sense. The simulation argument no more shares the concept of religion based magical creator

    Don't you really? Theism postulates a religion-based magical creator. Simulationism postulates a science-based magical creator. We know exactly as much about the particulars of a simulator as we do about a theistic god, which is nothing at all. If you really don't see the connection then I think I'm done pounding this particular nail into this particular brick wall.
    posted by ook at 5:11 PM on September 13, 2012 [2 favorites]


    I have enjoyed this thread immensely.
    posted by stavrosthewonderchicken at 5:15 PM on September 13, 2012


    I have enjoyed this thread immensely.

    You were programmed to do so!
    posted by flapjax at midnite at 6:17 PM on September 13, 2012


    Freewhimmed -- that's me all over.
    posted by stavrosthewonderchicken at 6:20 PM on September 13, 2012 [1 favorite]


    I'm skeptical of any argument which relies on an appeal to authority ie NASA scientist proposes x. As line scientific inquiry it does not seem to meet the requirements of a hypothesis because there are no proposed experiments or tests for the speculative statement. Furthermore the proposed proof by probability seems to simply recast Pascals wager. This notion also seems very close to intelligent design.
    posted by humanfont at 7:14 PM on September 13, 2012


    Well, this is a very old problem. Three points:

    1. The funny thing is: If the world CAN be a simulation, it likely is. Why? Mathematics. Similar to the German Tank Problem. or the doomsday argument. Simply, there are much more simulations than the real world. Hence, if simulations exists, you are just more likely to live in one than not

    2. Matrix was just a terrible movie compared to the original. And yes, the real cigars are getting smoked by someone else.

    3. The idea of living in a matrix is not metaphysics, it might be testable scientifically. If we are living in a simulation, the "real" world will likely be bound to some similarities.
    a) mathematics
    b) limitations of energy etc.
    Assuming that 3a and 3b are correct, they also will have limited resources. And while they may be able to run a simulated world or two or ten. They can not allow the simulation to run a simulation. The simulation may run again another 10 simulations or 100 and this simulations again. This problem is called "stacking". If the real world is bound to limitations of mathematics (exponential growth) and energy limitations, they must prevent stacking to happen.


    Until the, enjoy the ride!
    posted by yoyo_nyc at 7:35 PM on September 13, 2012


    "The other interesting thing is that the natural world behaves exactly the same way as the environment of Grand Theft Auto IV. In the game, you can explore Liberty City seamlessly in phenomenal detail. I made a calculation of how big that city is, and it turns out it’s a million times larger than my PlayStation 3. You see exactly what you need to see of Liberty City when you need to see it, abbreviating the entire game universe into the console. The universe behaves in the exact same way. In quantum mechanics, particles do not have a definite state unless they’re being observed. Many theorists have spent a lot of time trying to figure out how you explain this. One explanation is that we’re living within a simulation, seeing what we need to see when we need to see it."

    The idea that the universe is a simulation is very old. But the link of "to be observed" to "to be computed" is an interesting one. I like it.
    posted by yoyo_nyc at 8:09 PM on September 13, 2012 [1 favorite]


    I think it was especially fiendish of the people who created the simulation we all live in to give us The Sims.
    posted by crunchland at 8:30 PM on September 13, 2012


    But the link of "to be observed" to "to be computed" is an interesting one.

    So...if a tree falls in the forest and noone is there to observe it, it doesn't make a sound?
    posted by ceribus peribus at 8:34 PM on September 13, 2012


    Not only that, it doesn't even fall. It's just already on the ground when you show up to take a look around.
    posted by speedo at 9:21 PM on September 13, 2012


    There are more than 200 comments and no one has mentioned Scott Adams yet? I know he has a google thingy that tracks stuff relevant to his interests. I wonder if mention of his name on this website would be sufficient to bring back that account he set up.
    posted by ericales at 9:22 PM on September 13, 2012


    The problem is that nothing we can point to in a human brain -- on the scale of a computer circuit -- understands anything either, or otherwise does a thing that computers categorically can't do....Maybe it's true, maybe it's not, but we currently know of no physical laws preventing it -- we know of nothing that exists in a brain that couldn't exist in a computer.

    I suspect we don't understand all (or maybe even much) of what brains do. Even if something physical - a hydrogen atom - could be simulated perfectly such that all of the properties of it we know about are perfectly represented by variables or whatever this still doesn't seem to me to make the simulation identical with a hydrogen atom.

    Nothing we do understand about the universe explains why anything in it should be conscious, but we do know that brains are conscious. This suggests to me that there is something big we don't understand, and the place to look for it is the brain. I don't think we have that much reason to believe that consciousness happens in a computer.

    This wikipedia article on digital physics, "the hypothesis that the universe can be conceived of as either the output of a computer program, a vast, digital computation device, or mathematically isomorphic to such a device," is pretty good IMO. It seems to me a computational theory of consciousness basically relies on this ontological view? I really don't see how this can be considered "materialism" though - maybe "informationalism" or "computationalism?"

    In the article David Chalmers comments, "We are led to a conception of the world on which information is truly fundamental, and on which it has two basic aspects, corresponding to the physical and the phenomenal features of the world."

    (1) ...The simulation is successful enough that people who know the original person whose brain was copied are unable to tell the difference during a text-based conversation.


    I think that Searle's Chinese Room argument successfully shows that even if a computer could pass a Turing test it does not prove that a computer is conscious or understands anything. There are already robots used in elder care facilities in Japan used to keep elderly patients company, and the patients seem to connect with them emotionally somewhat. I don't think these robots are conscious or feel anything; I don't think Big Blue understands chess or Watson understands Jeopardy questions.

    Considering the types of brain/machine interfaces that have been emerging from the labs over the last few years I'm not sure we're as far away as you seem to think. There's mind controlled prosthetics, cameras allowing blind people to see...

    Yes, instead of the Turing test, I think this is the way to make the science of consciousness testable as saulgoodman suggested. From what I've read all of the current "brain prosthetics" are very primitive, and the brain is exceedingly complex obviously so I am not as optimistic that we will get there in our lifetimes.

    Imagine that we thought we understood everything about the experience of color vision. And this understanding revealed that there were other primary colors that could exist, but the human brain is not capable of seeing. Perhaps we could make a special prosthesis that could be wired into the visual cortex where it actually integrates into your consciousness and could cause you to see certain objects in your visual field in a brand new never seen before color. This, it seems to me, would be a way to test our understanding of the mind. If you guys are right, such a prosthesis could perhaps just be a computer interfaced to the brain with synthetic synapses, but I think it would have to be something like synthetic neurons that are able to access consciousness (which to me seems like it must somehow be a fundamental property of existence) the way our neurons do.

    So if we can't answer the question right now, and if people are going to pick an answer to firmly believe in without evidence, here's my real question: why choose in particular an undiscovered physical law that prevents computers from doing what human brains do? Why not firmly believe that the randomness of a random number generator equals that of a hydrogen atom?

    If we can't answer the question right now, why answer it? Isn't this the trouble that religion gets into? What is wrong with admitting we don't know what we don't know? I recently came across this talk by Chomsky, "The machine, the ghost, and the limits of understanding," in which he makes a good point that there is no reason to believe the human brain is capable of understanding much of the universe. A cat is not able to understand arithmetic - they have no need for it. Why should we believe the human brain should have evolved to understand everything.
    posted by Golden Eternity at 10:15 PM on September 13, 2012 [1 favorite]


    Even if something physical - a hydrogen atom - could be simulated perfectly such that all of the properties of it we know about are perfectly represented by variables or whatever this still doesn't seem to me to make the simulation identical with a hydrogen atom.

    Okay, we might not be able to, because there's obviously things we don't yet understand about physics, but are you asserting here that it is outright impossible for any entity to ever simulate a hydrogen atom? Because, remember, we're talking about intelligence levels that we would probably consider godlike; if hydrogen atoms are already simulations, whatever's doing the simulating must be far, far smarter than we are.

    Or is it just a naked assertion that a hydrogen atom is a unique thing that can't ever be duplicated, just because?

    It seems like that whole post is an argument, pretty much, that you have faith that there's something that's outside visible physics, so therefore physics can't be a computer program. But couldn't that invisible something itself be the program, running on a GodComputer?

    Why should we believe the human brain should have evolved to understand everything.

    Well, my personal guess is that we evolved to understand each other, and the mechanism to do that is general enough that we can use it for other things, as well.
    posted by Malor at 11:40 PM on September 13, 2012


    Well, my personal guess is that we evolved to understand each other, and the mechanism to do that is general enough that we can use it for other things, as well.

    I do like these lines of thought, a similar one being we had to create model of other minds to anticipate behaviour, and happened to get a model of our minds along for the ride. A chicken-and-egg one, that.
    posted by fightorflight at 7:03 AM on September 14, 2012


    I want to publicly apologize to Jernau for my little 'hammer and brick wall' jab above. That was totally uncalled for, and I regret posting it.
    posted by ook at 7:48 AM on September 14, 2012 [2 favorites]


    The Last Question (Asimov)
    posted by aniola at 8:08 AM on September 14, 2012


    Well, my personal guess is that we evolved to understand each other, and the mechanism to do that is general enough that we can use it for other things, as well.

    Including, if Scot Bakker and others are right, understanding ourselves.
    posted by adamdschneider at 10:36 AM on September 14, 2012


    I'm sure you could simulate many different aspects of a neuron, but the simulations would not be neurons, anymore than a simulation of photosynthesis actually converts physical carbon dioxide into physical glucose with the aid of photons

    That sounds like you misunderstood the thought experiment. I was asking if you could imagine a machine controlled by a computer simulation that would take in the chemical and electrical inputs a neuron receives and convert them to the exact same electrical and chemical outputs that a neuron would. One such machine is merely an artificial neuron. But two of them linked together could communicate with each other by exchanging symbolic information without converting it to chemical/electrical outputs. Replace all the neurons this way and you'd have a machine that simulates the activity in the brain symbolically. It's hard to see what stage in that process a brain would stop having subjective experiences.

    I think that Searle's Chinese Room argument successfully shows that even if a computer could pass a Turing test it does not prove that a computer is conscious or understands anything.


    Searle's Chinese Room argument makes no sense to me. In the experiment, he imagines a computer program that can pass the Turing test in Chinese. Then he imagines himself running the computer program manually and complaining that he himself wouldn't understand anything because it's all going by in Chinese. But in that analogy, Searle isn't replacing the computer program that understands Chinese. He's replacing the electrical switches that run the computer program that understands Chinese. No one claims that the electrical switches in an intelligent computer would understand and experience what the program understands and experiences.
    posted by straight at 11:44 AM on September 14, 2012 [4 favorites]


    Didn't Konrad Zuse originate this idea more than 40 years ago? He was hardly a pot-smoking college student.
    posted by alex_reno at 11:45 AM on September 14, 2012


    There are more than 200 comments and no one has mentioned Scott Adams yet?

    Perhaps that should tell you something about the regard he is held in as a physicist or a philosopher. Somewhat lower than the Wachowski brothers or that one stoner guy who never left the lounge in college, apparently.
    posted by Tell Me No Lies at 11:59 AM on September 14, 2012 [2 favorites]


    I was asking if you could imagine a machine controlled by a computer simulation that would take in the chemical and electrical inputs a neuron receives and convert them to the exact same electrical and chemical outputs that a neuron would. One such machine is merely an artificial neuron....Replace all the neurons this way and you'd have a machine that simulates the activity in the brain symbolically. It's hard to see what stage in that process a brain would stop having subjective experiences.

    It seems to me this assumes that consciousness is happening either in the chemical and electrical input/output signals or is somehow contained in or identical with the "computation" the system is performing. If consciousness is happening in bio-chemical-electrical signals themselves, then maybe subjective experience still happens. But what if consciousness is something else happening in neurons, alongside the computational function, that is irreducible? Then the purely computational neurons wouldn't work, but perhaps synthetic neurons that are able to access/create consciousness the way our hypothetical neurons do would still be possible.

    He's replacing the electrical switches that run the computer program that understands Chinese. No one claims that the electrical switches in an intelligent computer would understand and experience what the program understands and experiences.

    Actually, a computer is really nothing but electrical switches, granted some of them are configured to hold state information. What if the program and input were all contained on punch cards and the computer only performs one instruction at a time, so there is very little information contained in the switches at any moment. I agree that the electrical switches (or transistors) are not intelligent or conscious. I don't think the punch cards or the ink on the punch cards are conscious either. So what understands? Where is the consciousness? My understanding of the Chinese Room argument is the man in the room is simply receiving Chinese characters he doesn't understand and following a set of English instructions (a computer program, the man is acting as the CPU) to output a set of other Chinese characters, but he has absolutely no understanding of what any of the characters mean. I think a response to the Chinese Room argument is that the entire system, the man and the instructions together, do understand Chinese - I think you could say the entire system has 'machine intelligence' (?) but doesn't understand anything and isn't conscious.

    are you asserting here that it is outright impossible for any entity to ever simulate a hydrogen atom?...Or is it just a naked assertion that a hydrogen atom is a unique thing that can't ever be duplicated...?

    I was thinking of a simulation as nothing but symbols and symbolic operations - and hydrogen atoms as something physical - and thus they are not identical, and how would you interface the symbolic hydrogen atom to other real atoms and space-time the way a normal hydrogen atom resides in the universe? I guess the idea is that you could simulate the entire universe and it would be indistinguishable from our own. To me this still doesn't explain consciousness though. I guess it is basically just saying the foundation of our reality is mathematical/symbolic. Assuming our universe is a simulation, why is it that the symbols in the simulation that represent my brain are conscious and actually experience the simulated world? Our experiences are not symbols, they are things like colors and sounds and feelings, etc. I wonder if the fact that it seems like we can frame anything we understand about the material world as a symbolic state machine (simulation) is actually saying more about our cognitive abilities than the universe itself - the only way we are able conceive of the world is as a simulation and thus naturally we conclude that that is what the universe is.
    posted by Golden Eternity at 3:10 PM on September 14, 2012


    I think it likely the universe to be a computation, but I think it unlikely to be a simulation, but even it was, I would doubt that it's a simulation of what we know as the universe. No, we're a probably a side effect.
    posted by wobh at 8:58 PM on September 14, 2012 [1 favorite]


    That's a strange cosmology. You are free to believe in it of course.

    I believe in ALL the strange cosmologies. ALL OF THEM
    posted by krinklyfig at 10:56 PM on September 16, 2012


    It is my fond hope that the final resolution of all these questions turns out like this
    posted by Tell Me No Lies at 1:07 AM on September 17, 2012 [2 favorites]


    But what if consciousness is something else happening in neurons, alongside the computational function, that is irreducible?

    That's pretty much classic Cartesian dualism. And that, in my opinion anyway, is really just the idea of a 'soul' with the traditional monk's habit traded in for professorial robes.

    The idea that consciousness might depend on some "irreducible," non-simulatable property of the human brain (but which somehow manifests itself in almost every properly-formed human brain, and can be affected/destroyed by trauma to the human brain) seems like an unnecessarily complex argument, absent any compelling evidence for it. There's a sort of "god of the gaps" quality to the idea, where the non-physical part of consciousness is constantly shrinking beyond the horizon of what science can observe at a particular time. And that doesn't seem to me like the mark of a particularly robust theory, but rather an apologia for a preexisting position.
    posted by Kadin2048 at 10:57 AM on September 17, 2012 [2 favorites]


    But what if consciousness is something else happening in neurons, alongside the computational function, that is irreducible?

    Why do you think that might be a possibility?
    posted by straight at 12:31 PM on September 17, 2012


    The idea that consciousness might depend on some "irreducible," non-simulatable property of the human brain

    It still seems to me like there is a misuse of the idea of simulation as I understand it. If we come up with a mathematical model that can predict all known properties of a hydrogen atom, it is not equal to a hydrogen atom - not because there is something mysterious in the hydrogen atom, but because a simulation is a set of numbers and relations between numbers, and a hydrogen atom is a hydrogen atom. I'm still objecting to the apparent assumption that a hydrogen atom (the physical world) is nothing but a particular set of numbers. By "irreducible", I didn't mean that we wouldn't find properties in neurons or elsewhere in the brain that are "simulatable" and directly isomorphic with a conscious experience, I mean that the simulation would not be identical with the physical process being simulated and therefore the simulation itself would not be conscious.

    Some consciousness experiences, the color red for example, seem to be irreducible, so it would make sense to me for it to attach to an irreducible physical state that is uniquely identifiable as the physical state of the phenomenal color red. One problem with saying that color is some sort vector space where either the numbers themselves experience the color red, or the "medium" that is "storing" the numbers is experiencing the color red, how do the numbers/medium know that in this case they represent a color and not something else entirely? Is there just a certain set of numbers out of all possible that has been arbitrarily assigned phenomenal experiences such as the color red or something? If so this would be "magical" wouldn't it? Without a physical/phenomenal world to represent or interface with, a simulation is just numbers.

    Why do you think that might be a possibility?

    It seems to me consciousness must involve some sort of non-local physical states. A conscious experience is unified. There is one me (subject of experience) seeing one monitor with many shapes and colors and words as a unified experience. But if you look at the parts of the brain that are involved with filtering light frequencies, identifying shapes, words, etc, there are many different neurons/synapses in different locations separated by fairly large spaces creating one unified experience. Where/when is this experience taking place and how is it unified/integrated? It seems to me there must be some sort of conjunction in space between the neurons involved that create a single unified experience, and this is no the case or necessary in a computer. I thought this work I came across by Moshe Gur On the unity of perception was an interesting presentation on this problem.
    posted by Golden Eternity at 12:22 AM on September 18, 2012


    th filtering light frequencies, identifying shapes, words, etc, there are many different neurons/synapses in different locations separated by fairly large spaces creating one unified experience. Where/when is this experience taking place and how is it unified/integrated? It seems to me there must be some sort of conjunction in space between the neurons involved that create a single unified experience, and this is no the case or necessary in a computer.

    Where inside the computer is the unified image that gets displayed on screen? Where in the guitar or the player is the resonant sound of a strummed G major chord? Where in a bicycle is the part that makes it a bicycle and not just a collection of chains, wheels and other parts?

    What you're talking about are emergent phenomena in a functional space. It's where the rubber meets the road, and all the parts come together to function within some larger rule-governed context. There's no place in the brain or body where you'll find conscious experience, because conscious experience happens in the functional space of the world as a whole, not just in our heads: consciousness seems to be a complex physical phenomenon analogous to fire. Just as there's more than one way to ignite and sustain a fire, there may be more than one way to ignite and sustain consciousness as a phenomenal process.
    posted by saulgoodman at 7:01 AM on September 18, 2012 [2 favorites]


    But if you look at the parts of the brain that are involved with filtering light frequencies, identifying shapes, words, etc, there are many different neurons/synapses in different locations separated by fairly large spaces creating one unified experience. Where/when is this experience taking place and how is it unified/integrated? It seems to me there must be some sort of conjunction in space between the neurons involved that create a single unified experience

    But how big could that conjunction be? Two neurons? A hundred neurons? A 2-cubic-centimeter region of the brain? Why couldn't it stretch across the entire brain?

    Your description of the brain makes it sound like the visual cortex and the auditory cortex are separate systems from the system of conscious thought, that sense information is processed in one place and then "experienced" in another. Doesn't it seem more likely that the visual cortex processing images is itself part of the total system that is our conscious experience?
    posted by straight at 7:58 AM on September 18, 2012 [1 favorite]


    Why couldn't it stretch across the entire brain?

    I think it would.. Otherwise there would be multiple independent subjects of awareness/experiencers/selves. Somehow everything must be integrating together. Perhaps there is something to this: Recurrent thalamo-cortical resonance is an observed phenomenon of oscillatory neural activity between the thalamus and various cortical regions of the brain. It is proposed by Rodolfo Llinas and others as a theory for the integration of sensory information into the whole of perception in the brain.

    Your description of the brain makes it sound like the visual cortex and the auditory cortex are separate systems from the system of conscious thought,

    It seems to me they would be. I don't see why neurons involved in creating color have much direct connection to neurons involved with word recall/internal speech. Visual processing probably evolved long before language/semantics.

    that sense information is processed in one place and then "experienced" in another.


    I was actually thinking the opposite - that if an "experience" is isomorphic with a physical state then the location of the physical event is its location. I just wanted to make the point that even on the smallest scale it could be argued that non-local conjunction is required and integrated circuits don't do this. "Where" something is experienced and "what" is experiencing are interesting questions - but I don't see that they are incoherent or can't be asked for some reason. Perhaps this is where we "run against the boundaries" of what is knowable.
    posted by Golden Eternity at 12:31 AM on September 19, 2012


    Another possible ending to this discussion.
    posted by Tell Me No Lies at 5:16 PM on September 22, 2012


    '"Greg Egan's Diaspora"

    This is seriously one of my favorite books of all time. One of the greatest books of any kind, ever. Egan actually postulates a physically plausible rationale for a god/gods who actually need our belief and then in virtually the next paragraph basically disproves the notion of God entirely.

    It's a really phenomenal, very underrated book. At the outset, it looks like yet another cyberspace sci-fi thing, but it's SO not.'


    Arrgh, I meant Permutation City. I believe Diaspora is still on my "to read" shelf.
    posted by Eideteker at 7:16 AM on September 26, 2012


    « Older Pass the Spoon: a sort-of opera about a television...   |   Politicopsychopathology Newer »


    This thread has been archived and is closed to new comments