I originally titled this post Robots and Aliens but I ended up discussing something completely different. That happens a lot.
I have been teaching a class this past semester on Computers and Robots in Film. We have fantastic discussions about consciousness, time, and reality. This past week we watched The Matrix, and one of the students brought up the concept of the simulation hypothesis. This is the notion that from probabilistic standpoint we are living in a simulation created by a race of hyper intelligent aliens.
Another student mentioned that the simulation hypothesis postdates The Matrix by a few years so the philosophical inspiration is probably Descartes (and Plato, but I am not going to talk about him). Descartes problem is, how can we trust our perceptions. How do we know we are living in a world, and not in a computer simulation created by aliens (or in Descartes case, the Devil). Descartes proposes that the most basic thing we can trust is that we thinks. When you are having a doubt that world is a fiction – you are afterall thinking – I think therefore I am. Whether or not this is a a correct assumption we could further discuss, but what I think is interesting is the notion of proof. How do I know that the world is real.
There is a distinction between these two approaches that mirrors a shift from scientific method to data science, and I do think that historians looking back will not place computer simulations in the same category as physical experiments and hypothesis testing.
With Descartes the question is what is true, or what is real? It is a binary question, either something is true or it is not, either something is real or it is not, the cat is alive or the cat is dead (that was a trick :).
With the simulation hypothesis, we model all of human existence and select the answer with the greatest possibility. The simulation hypothesis, is itself, the product of simulation thinking. If we run a simulation on the question of whether or not we live in a simulation, it is most likely that we live in a simulation.
The simulation hypothesis does not resort to experimental data, but to probabilisitic data. With the scientific method there is an appeal to epistemology – what do we know, how do we know something. The answer from science, is through experimental data and the creation of falsifiable theories. With the simulation hypothesis there is no notion of epistemology. We are not asking how do we know we are in a simulation, we are making a judgement that we are probably in a simulation. The question is metaphysical. – what is reality, and its proof is numerical.
I also have an intuition that this is probably related to mereology, the relation of parts to a whole that is used mostly in set theory, how that is I have not quite worked out.