The Simulation Argument by Bostrom

[exogen]

This argument was put forth by Nick Bostrom who is a philosopher from Oxford University. The argument is not the same as Brain in a vat although it does have some similarities in the other all philosophical picture. It also makes use of similar philosophical issues. Niji may find this interesting given his interest in AI.

 

Here is a link to the original paper.

 

http://www.simulation-argument.com/simulation.html

 

Here is another link to something written by Bostrom

 

http://www.simulation-argument.com/matrix.html

 

In regards to the first paper it is of note that the notion of "substrate-independence" is something that should be discussed because it is one of the most crucial points that should be looked at. Even if substrate-independence is wrong and only biological machinery can create conscious experience one could still reform the simulation argument with the idea that instead of non-biological systems they could be using cybernetic systems where the biological portion would probably consist of the nervous system and any life support systems that were absolutely necessary. However that is only assuming substrate-independence is wrong.

 

In regards to the technological constraints one thing that could be utilized is the abilities of the brain itself to create the reality and having that system be tied into the collective system of the whole matrix. In this way the computational power would not need to come entirely from outside of the system, but could be gained from the brains connected to the system in the first place. In the same way brains in dream state create whole worlds, the brains could potentially be hacked to supply the computational power needed to create the "world" that is shared by everyone. The real issue would be in getting the brains to all work collectively. Genetic augmentation would no doubt be implemented here, again assuming substrate-independence is wrong. The same argument could be made if it isn't incorrect and artificial brains were the technology of choice for the simulation.

 

The crucial issue that needs to be understood is that even if substrate-independence is wrong, the underlying concept is that there is a difference between the subjective conscious experience of one's self and the experience of the world, and the world itself uninterpreted and unexperienced. This dual ontological categorization is an essential feature of this argument and is the basis of the "hard problem" in neuroscience and the related philosophy of mind issues. The simulation argument aside from arguing that we might and mostly likely are in a computer simulation, identifies this ontological distinction. In simple terms it is the idea that experience and consciousness in general is always virtual and not necessarily representative of any reality beyond the experience itself. The simulation argument highlights this point.

 

Here is a link to FAQ's about the Simulation Argument.

 

http://www.simulation-argument.com/faq.html

 

if you notice on the third question the Simulation argument is of a different sort then BIV argument.

 

Here is a link to Bostrom's home page.

 

http://www.nickbostrom.com/

 

Some criticism

 

This was the best criticism I could find at the moment. http://www.exampleproblems.com/wiki/index....mulated_reality

 

Let's look at some of these.

 

"Others see the proposition itself as having little to no real life applications, since the claim is essentially unprovable in any concrete sense, and any "evidence" that is directly observed could be another simulation itself. This is akin to an infinite regress problem with the argument. Even if we are a simulated reality, there is no way to be sure the people running the simulation are not themselves a simulation, and the operators of THAT simulation are not a simulation, ad infinitum. Given the premises of the simulation argument, any reality, even one running a simulation, has no better or worse chances of being a simulation than any other. "

 

Notice first that there is a pragmatic bias. It is true that if the Simulation Argument is correct then it is of no real practical significance to us. But that says nothing about if it is true or false. There can be many things that are of little use to use but nonetheless remain true.

 

The second part of this criticism has an empirical bias. By saying the simulation is "unprovable" is to use the word prove in a very limited way. Many notions can be established as true independent of empirical evidence, and in fact some of our most fundamental notions.

 

The infinite regress accusation is correct insofar as any empirical date that could be supplied to falsify this hypothesis could potentially be just another piece of simulated data. The idea that there could just be an infinite bunch of matrix's within matrix's is an inevitable consequence of the Simulation argument.

 

But so what? How does that change anything? What this really does is point out that the notion of reality being divided into two distinct ontological categories (appearance vs. reality-subjective vs. objective) does not work. How do I know this? If you remove the notion of this reality being a "simulation" and just talk about reality as we know it and not an imagined third-person objective sense in which reality exists appart from consciousness and forms the bedrock of it, then the argument not only goes away but ceases to have any relevance altogether. Not only that but the infinite regress implication goes away as well because "simulation" becomes a misnomer. We are then simply left with reality as experience without any need to speak of simulations.

 

This point is exemplified by the next criticism.

 

"Another criticism that can be made is that even if this is a "simulation", it is still the only "Reality" we have ever known, and we have no example of a non simulated reality to compare it to, so this is still, in a sense, "reality" as we understand it. "

 

Bingo!! This idea is right on the money. We don't have any experience of any other "reality" because "this reality" is the only one we have ever known. Another assumption of the Simulation Argument is materialism but this criticism undercuts that very idea. In saying that we have only ever known "this reality", we not only call into question the idea that we might be simulated, but also the notion that there is even a material reality to make such simulation possible, because such notions are purely abstract. To say it another way, all we have really every known is out experience of "the world" but not the world itself, if there is such a thing. The idea of the world being independent of our experience and also constituting reality is in actuality just that, an idea and therefore something imagined as opposed to being something self-evident. This criticism of the simulation argument draws this point out exactly.

 

In conclusion, my assessment is that while you can haggle over the probabilities, as many have, the argument is sound because it functions off of a philosophical notion that cannot be logically denied, that point of course of the truth that ontologically our reality is subjective, which provides the basis logically for the appearance vs. reality distinction as it relates to metaphysics and how we are to interpret our empirical experience.

 

 

Edit: I thought I would add this clip of Sam Harris talking about the Simulation Argument. He uses it here to attack a theological perspective(s).

 

http://www.youtube.com/watch?v=4RoGtWUMi4w

 

Notice how he points out that the argument assumes that consciousness is a material phenomenon, which allows for the logic of the simulation argument to unfold. I want to point out though how the Simulation Argument serves as an "Inuition Pump" as Danial C. Dennett would say insofar as it draws out the idea that there is a diference between experience, which is essentially virtual in kind, and the so called "material reality" that allows for the simulation argument to take place. Understanding that distinction then shows why one can then call into question the objective (material) reality as being "real" to begin with. So just as for all we know we might be in a simulation if materialism or something like materialism is true, for all we know, and unless we have arguments that are non-emperical in kind demonstrating why materialism is true, materialism might be false. So if materialism is true, then we are led to the Simulation argument, which then means we have no way of knowing if the reality we experience is representative of the actual material universe and not just a simulation.

 

My point is, if we abandon materialism and anything like it, we not only get rid of an extra ontological catagory which is imagined anyway and never directly aprehended in our experience, but we get rid of all these kinds of arguments that lead to an infinite regress of skeptical implications.

Edited July 1, 2011 by exogen

[Bakuhatsu Pengin]

its seriously the same thing as brain in a vat

[exogen]

Edited July 2, 2011 by exogen