The Purpose Of Our Simulation

The 1 February 2020 issue of NewScientist contains a collection (here and here, but behind a paywall, I should imagine) of short articles which, essentially, express the anguish quantum physicists feel as they continue to try to figure out how to connect gravity with quantum mechanics, understand the Copenhagen Interpretation of quantum mechanics, i.e., what the hell does it mean to say that it requires an observer to force the probabilistic wave function to collapse to reveal the exact location of a quantum particle, hell, to even know if there is a reality.

One seldom considered question, however, is how exactly models ought to seek to explain reality. Some, such as general relativity, take some known quantities about nature – the position of a planet, say – and predict what will happen next. Quantum theory takes a different philosophical approach, assigning probabilities to future outcomes we might see. …

Some physicists are now exploring whether a similar approach can help us make headway. For example, constructor theory starts from the idea that the essence of reality is information, and then sets out what kinds of things are possible and impossible. It is early days, but it has already made predictions in circumstances that defeat other theories, such as the behaviour of quantum particles in a gravitational field.

I was awfully darn tired – and a little sick – when I started reading through this one morning, and hazily connected it to my own private theorizing that we are, ourselves, a computer simulation. Then I came across this:

IN OUR quest to understand reality, there is an elephant in the room. How do we know that the reality we are in is real? The suggestion that we could be living in a computer simulation isn’t just a Matrix-style science-fiction idea. It is a hypothesis that has been discussed and debated by philosophers and physicists since Nick Bostrom at the University of Oxford floated it in 2002. If its startling but logical conclusion is correct, it renders decades of intellectual endeavour obsolete and, ironically, takes us back to the beginning.

Nice to know that other people harbor weird paranoias like mine. Oh, wait, they take their’s to new extremes:

Bostrom’s simulation argument says that if humans could one day create simulations of the universe populated with conscious beings, then in all likelihood we are living in such a computer-generated universe. The argument assumes that, eventually, enough computing power will exist to create simulations of human history that are detailed enough for the simulated people in it to be conscious. If so, then, statistically speaking, we are more likely to be living in a simulation, because simulated people would vastly outnumber unsimulated ones. That is especially true if simulated people make their own simulations ad infinitum in an endlessly nested reality.

Nice, I’m almost certain a simulated creature! Although I’m not sure our limited experience bears out Bostrom. Think of our attempts to simulate a chemical reaction – get much beyond 3 or 4 atoms and the calculations become too laborious.

Skipping over my trivial objections, I’d like to move on to what my sickly, tired mind came up with that morning. We run simulations for tangible reasons, from weather forecasting to predicting chemical reactions to archaeological modeling to, yes, quasi-entertainment and learning skills, such as flying the latest aircraft.

So, if we are ourselves entities in a quasi-computational simulation of some sort, what is the goal of the entities who created this environment? What are they trying to achieve? (Let’s skip over the quasi-ethical question of whether we should comply with their/its purpose, which makes my head hurt thinking about the number of people who think the answer should automatically be Yes!).

Before burping up my thought, let’s add in one other area of interest, that of the semi-related, poorly labeled field of artificial intelligence, aka machine learning (ML). I gave my ad hoc definition of ML here back in 2018:

My observations of ML, on the other hand, is that ML installations are coded in such a way as to not assume that the recipe is known. At its heart, ML must discover the recipe that leads to the solution through observation and feedback from an authority entity. To take this back to the deferment I requested a moment ago, the encoding of the discovered recipe is often opaque and difficult to understand, as the algorithms are often statistical in nature.

One of the largest known stars, Betelgeuse, has recently dimmed and become warped. Will it soon go supernova? Such a thought is still more comforting than wondering if we’re all just a big simulation by someone trying to figure out the nature of their own reality.

So, is it possible that the purpose of our simulation is to discover the nature of reality? Just as we have difficulty understanding reality, is it possible that the creating entity has encoded the observed facts on their ground into our simulation and then manipulated us into trying to figure out how reality really works?

And what if that entity got an observation improperly encoded?

I think I’d better go back to watching Xena: Warrior Princess, because the existential angst these physicists are feeling is as nothing compared to my existential angst at thinking I may be a … bug. Flaw. Miscue. Eeeek.