Category Archives: computation

Descartes and Musk – On Dreams and Simulation

I’ve just begun reading through a paper of Tom Campbell’s which appeared in this video [link redacted]. In this post, though, I only aim to liken the popular simulation theory with Rene Descartes’ older “dream argument”. The former states that at least some of us are living in a virtual simulation. The latter runs as follows:

Premise 1 – If I know that I am awake, then I can eliminate as false the competing hypothesis that I am dreaming.

Premise 2 – I cannot eliminate the dreaming hypothesis.

Conclusion – I do not know if I am awake.

Elon Musk seems sympathetic toward the type of conclusion above. However, Musk’s interest has been not in dreams, but in whether a given person’s reality is provably virtual or actual. Descartes’ dream argument could be adapted in light of Musk’s challenge that we don’t know whether we’re living in virtual reality (VR):

P1′ – If I know that I exist in actual (non-virtual) reality, then I can eliminate as false the competing hypothesis that I exist in VR.

P2′ – I cannot eliminate the VR/simulation hypothesis.

C’ – I do not know if I exist in actual reality (as opposed to VR).

Testing the simulation hypothesis would minimally involve two steps. The first would be running participants through VR simulations. Following, the essential question for these subjects would become: “Did you know when your reality shifted between actual and virtual?” Naturally, experimenters could not give anything away until the end of such a study for it to be meaningful. If participants were consistently aware of when their reality changed kinds, then Musk’s simulation hypothesis would not apply: for their case, it would have to be ruled out as false. On the other hand, if subjects were generally unaware of their reality becoming actual or virtual, then Musk’s simulation hypothesis would hold true.

Campbell notwithstanding, this question presently remains unresolved.

Advertisements