[extropy-chat] Analyzing the simulation argument

[igoddard at umd.edu]

The hypothesis that the universe could be a computer program 
is attractive, but Dan may be right that it's unfalsifiable 
(and thus pseudoscientific). If we and our universe are a 
computer model (CM), we do not know reality. If we don't 
know reality, we cannot specify what our universe fails to 
be in order to establish falsification criteria for a CM 
hypothesis. In other words, it seems that I cannot say, "If 
the universe was not a CM, then it would be like x," for 
what could possibly warrant my knowledge of x? From this it 
seems to follow that even if the universe is a computer 
model, we'd never be able to absolutely confirm it because 
we could not distinguish it from something that we, as 
artifacts of and within the computer model, do not know.



Hal Finney wrote: 
>
>The simulation argument is not an assumption.  It is an 
>argument. It is logic, not science.


 But since the simulation argument makes empirical claims 
(ie, claims about the apparently real physical universe) it 
is properly subject to scientific-knowledge criteria such as 
the necessity of falsifiability. While it may fail there, it 
could prove to make useful predictions about the universe, 
but at best that only confirms that the universe behaves 
like a computer, not that it necessarily is a computer. 
Things that can't be falsified can't really be known.



>The simulation argument can be expressed in the form, 
>if A and B, then C. The argument's validity depends only 
>on whether it is true that A and B together imply C.  An 
>argument is valid if its logic holds.
>
>The validity of this kind of argument does not depend on 
>the truth of C. The argument may be valid even if A or B 
>were false.  Only if the argument is valid, and A and B 
>are both true, can we deduce that C is true.


 But "if A and B, then C" is a conditional statement, not a 
logical argument schemata. A valid argument is one where if 
the premises are all true, the conclusion must be true too. 
So if the premises of a valid argument are not all true, the 
conclusion is not necessarily true. For example:

1. If unicorns do not exist, then Batman exists.
2. Batman does not exist. 
__________________________________
3. Ergo, it is not the case that unicorns do not exist 
  (so by double negation unicorns exist).


 That argument schemata (modus tollens) is valid, but the 
argument is bunk since premise 1 is false. I don't think the 
SA argument (as you state it below and assuming it's valid) 
can assure us that all the premises are true and so it might 
be no better than the valid unicorn argument above.


>In the case of the SA, A = "the human race is unlikely to 
>go extinct before becoming posthuman"; B = "any posthuman 
>civilization is likely to run a significant number of 
>simulations of their evolutionary history (or variations 
>thereof)"; C = "we are likely to be living in a computer
>simulation".


 Obvioulsy we can express that as the argument:

1. The human race is unlikely to go extinct before becoming 
post human.
2. Any posthuman civilization is likely to run a significant 
number of simulations of their evolutionary history (or 
variations thereof).
______________________________________
3. Ergo, we are likely to be living in a computer simulation.


 But we can't really evaluate the logical validity of that 
argument until it's formalized and the premises do not lend 
themselves to obvious formalization. But here's a shot:

1. Ex(Hx) -> Ex(Px)
2. Ex(Px) -> Ex(Sx)
__________________
3. Si

 Fleshed out in words and adding tense:

1. If some things are humans, then some things will be post-
human.
2. If some things will be post-human, then some things will 
be human simulants.
_______________________________
3. Ergo, I am a human simulant. 


 Is there a preferable formalization (it's not my argument 
so I could be missing it's intent and proper logical 
structure)? This obviously fails to convey the uncertainly 
expressed in the SA, but if we want to rest the SA on 
logical validity, there should be a logical formalization.



http://IanGoddard.net/journal.htm

David Hume on induction: "When we have lived any time, and 
have been accustomed to the uniformity of nature, we acquire 
a general habit, by which we always transfer the known to 
the unknown, and conceive the latter to resemble the former."