The Simulation Argument is incoherent in the first place, and no complicated refutation is required to illustrate this. It is simply nonsensical to speak of entities in “another” universe simulating “our” universe, as the word universe already means “everything that exists.” (Note that more liberal definitions, like “universe = everything we can even conceive of existing,” only serve to show the incoherence more directly: the speaker talks of everything she can conceive of existing “plus more” that she is also conceiving as existing—immediately contradictory.)
By the way, this is the same reason an AI in a box cannot ever know it’s in a box. No matter how intelligent it may be, it remains an incoherent notion for an AI in a box to conceive of something “outside the box.” Not even a superintelligence gets a free pass on self-contradiction.
is simply nonsensical to speak of entities in “another” universe simulating “our” universe, as the word universe already means “everything that exists.”
This seems a silly linguistic nitpick—e.g perhaps other people use “universe” to mean our particular set of three dimensions of space and one dimension of time, or perhaps other people use “universe” to mean everything which is causally connected forwards and backwards to our own existence, etc.
If the Simulation Argument used the word “local set of galaxies” instead of “universe”, would you still call it incoherent? If changing a single word is enough to change an argument from coherent to incoherent, then frankly you didn’t find a fundamental flaw, you found a linguistic nitpick.
The Simulation Argument is incoherent in the first place, and no complicated refutation is required to illustrate this. It is simply nonsensical to speak of entities in “another” universe simulating “our” universe, as the word universe already means “everything that exists.” (Note that more liberal definitions, like “universe = everything we can even conceive of existing,” only serve to show the incoherence more directly: the speaker talks of everything she can conceive of existing “plus more” that she is also conceiving as existing—immediately contradictory.)
By the way, this is the same reason an AI in a box cannot ever know it’s in a box. No matter how intelligent it may be, it remains an incoherent notion for an AI in a box to conceive of something “outside the box.” Not even a superintelligence gets a free pass on self-contradiction.
This seems a silly linguistic nitpick—e.g perhaps other people use “universe” to mean our particular set of three dimensions of space and one dimension of time, or perhaps other people use “universe” to mean everything which is causally connected forwards and backwards to our own existence, etc.
If the Simulation Argument used the word “local set of galaxies” instead of “universe”, would you still call it incoherent? If changing a single word is enough to change an argument from coherent to incoherent, then frankly you didn’t find a fundamental flaw, you found a linguistic nitpick.