[extropy-chat] The Simulation Argument (was: Atheists launch inquisition...)

[Harvey Newstrom]

The Simulation Argument has such obvious flaws, that I can't imagine 
why people are still arguing about it....
But here I go anyway....   :-)

1.  The Simulation Arguments makes a huge assumption without any basis: 
  That the "probability" of being born into a simulation is the same as 
the "probability" of being born into a real universe.  I am not sure 
that this statement even makes any sense.  Even if it does make sense, 
I am not sure why simulations should be given equal weight to real 
universes that have much larger volume, much longer time-spans, very 
long head-starts, and have more resources than the simulations 
contained within them.  It seems more likely that non-simulated 
universes should have much more weight in comparison.  It may even be 
plausible that non-simulated universes have more weight than all the 
simulations contained within them combined.  On the basis of space, 
time, precedence, hierarchy and other comparisons, it seems that real 
universes always win.  I can't think of any reason to give simulations 
equal or greater weight.

2. The Simulation Argument proponents seem to prefer possibility #3 
without evidence.  There is no reason to assume possibility #3 is 
likely. Why would anybody run a significant number of simulations of 
their history?  By the time they have enough information to make 
accurate simulations, they wouldn't need to run simulations.  They 
would already know how things would turn out.  It wouldn't be very 
useful for research or discovery.  For entertainment, it would be even 
less likely.  We fantasize about being in a future simulation to escape 
our boring present world in favor of the better future world.  For such 
simulations to exist, we must imagine that the future humans have the 
exact opposite desires.  They would be trying to access our primitive 
and boring universes instead of their own more advanced universes.  
Even if this were sometime desirable, I doubt such situations would 
outnumber futuristic simulations or outnumber real-life endeavors.

3. The Simulation Argument proponents seem to dislike possibility #2 
without evidence.  There are many plausible scenarios where such 
simulations are not common.  Maybe advanced humans are more interested 
in reality than simulations.  Maybe advanced humans decide that it is 
unethical to create life-forms in a simulation and not let them out 
into the real world.  Maybe most simulated civilizations find a way to 
escape or crash their universes, such that they don't exist in large 
numbers or for very long.  Maybe there is no point to simulating entire 
universes and simulating smaller games, small situations, or temporary 
settings are more common.  Maybe simulations never last long compared 
to real universes, and thus, the number of them is limited or 
distributed over time such that they never outnumber real universes.  
Maybe future humans figure out time travel or observe the past directly 
so that simulating isn't required.  Maybe future humans observe 
parallel universes via Many Worlds quantum physics and have more 
universes than they could ever explore and don't need to make 
additional boring ones.

4.  The logic is faulty.  Similar arguments can be made for other such 
claims.  For example, there are more dreams than real lives on this 
planet, since every human dreams many dreams per day, therefore we are 
almost certainly living in a dream.  There are more aliens on other 
planets than there are earthlings on this planet, therefore we are 
almost certainly  extraterrestrial aliens who merely believe we are 
earthlings.  There are more ex-workers from my company than current 
employees, therefore we are almost certainly already fired and just 
don't know it yet.  The Simulation Argument seems very similar to the 
Doomsday Argument.  It assigns faulty statistics about things we cannot 
measure or predict, and then makes predictions based on our faulty 
assumptions.  It is simply circular logic that makes assumptions to 
support the conclusion and then concludes what was already assumed.  It 
is not an example of the scientific method, Occam's Razor, logic, or 
other form of evidence.  It is an example of persuasive rhetoric that 
proves nothing.  Just because our current circumstances seem unlikely 
does not mean that we must reject the obvious reality around us in 
favor of an unseen reality that cannot be detected, tested or proven.

So, in summary, I think that the obvious solution to the Simulation 
Argument is possibility #2: "any posthuman civilization is extremely 
unlikely to run a significant number of simulations of their 
evolutionary history (or variations thereof)."  I think realistic 
simulations of history will be limited to specific events of interest, 
and that the sum total of all such simulations in a universe will not 
even come close to simulating an infinitesimally small fraction of the 
complete universe itself.  I also think that any attempt to weight 
probabilistic chances of being born into one of these simulations is 
extremely speculatory at best and meaningless at worst.  I think the 
usage of terminology such as "conclusions" or "likelihoods" or "almost 
certainly" with respect to this "argument" are simply inaccurate and 
misleading.

--
Harvey Newstrom <www.HarveyNewstrom.com>
CISSP, ISSAP, ISSMP, CISA, CISM, IAM, IBMCP, GSEC