Implications of Considering Simulation Theory

In my previous post for this task, entitled Simulation Theory & Conjectures of Post-Humanity, I broke down the reasoning behind Bostrom’s hypothesis that we are living in a computer simulation. Assuming, as this theory argues, that it’s not impossible that we are living in a simulation, henceforth I will discuss the implications of considering these ideas.

Simulation theory is a somewhat hollow area of study, simply because it’s “not functional information” (McEvoy et. al 2018). Besides being aware that a simulation is happening, a discovery of this variety doesn’t matter(ibid) in that it would not physically change anything about the way we live our lives – albeit under the weight of “existential dread”. Outside of being “philosophically interesting” (Staggs 2013), there’s little that understanding this theory will actually do to benefit the human race. For me, the notion of researching something so inexplicably pointless is fascinating in itself in that these studies often lead to unforseen insight into perpendicular issues – which I’ll explain more toward the end of this piece.

Human relationships with computers

Various ideas about the relationship between humans and computers pulse heavily through the mainstream media. This media, both fictional and through news media, varies on a huge scale, from science fiction media which inspires blatant fear of technology in the minds of viewers, leading to public resistance of technology (Kitzinger 2009), to media which greatly misinterprets or underestimates the capacity of that same technology, as was seen in the reporting of human cloning research in 2013 (Condic 2013). We are already struggling with the notion that computers can mime humans online, through bots for example, and it’s difficult to determine whether you’re interacting with a sentient biological human over the internet. This has us feeling somewhat disconcerted (Magid 2017); we (as a race) designed and built computers for our own benefit, we view them as inferior to us.

 

1.jpg
We both underestimate and overestimate the potential of the average computer  (source)

 

It is only by “historical accident” that computers are so strongly associated with numeracy, when in truth they have fundamentally very little to do with numbers (Newell et. al 1959). A general-purpose computer has the capacity to read, move, generate, compare and associate symbols. If an abstract combination of these abilities is programmed into the computer, more complex processes can be computed (ibid) – including a potential human simulation. The idea that the joke is on us – somewhat akin to the plot of The Truman Show (1998) – and everything we interact with, ourselves included, may be a piece of code (or the equivalent), is unimaginable. It’d be like learning that horses have actually been racing us in the Melbourne Cup, or that stamps collect us as hobbies.

Humanic Concepts of Simulation

We humans are an exceedingly arrogant species. Obsessed with where we came from and where we’re going, we have strongly debated two core paths of existence for centuries:

  1. Evolution
  2. Creationism

Simulation theory does not discount either of these systems, and may indeed exist alongside one. If we count atheism as a belief system, we could theoretically claim that simulation theory could also exist alongside it, however some have come out and blatantly rejected this via a Reddit thread within the Change My View sub Reddit, in which a critique of science practice and how it understands the development of humanity unfolds.

 

4
We don’t know where we came from, but imagining a computed universe is a relatively new form of creationism (source)

 

When imagining a computer simulation, we tend to picture a video-game-style contract between the player with the remote control, and the digital universe locked in the screen – but this is not necessarily how a true simulated society would operate (McEvoy et. al 2018). Therefore we have an adjunct bias to the notion of binary simulation that impairs our ability to objectively consider quantum, or other,  possibilities. Similarly, when we imagine that we might exist in a simulation, our minds tend to form a picture reminiscent of The Matrix (1999) (Zimmerman-Jones 2015), which Bostrom argues is inaccurate. In The Matrix, the conscious human mind is not simulated (ibid), meaning that biologically-conscious human minds enter a simulated world. Bostrom argues that in the case of a ‘real’ simulation (and I use the word ‘real’ with trepidation), our minds would be simulated matter also (ibid), reaching a clause called substrate-independent (Bostrom 2003). This means that human consciousness would not be dependent on a biological brain and could thus be replicated by a computer.

This bias carries through our use of the word ‘simulation’. In their podcast entitled Two Cyborgs & a Microphone: Fourty-Two (sic), McEvoy and Shank unpack the way we look at the purpose of our lives through religion, and argue there is little disparity in definition between ‘simulation’ and ‘religion’ in practice:

Most of the world throughout history has believed we are in a simulation, but without using those words.

This raises the question of whether we can call simulation theory a religion.  The following points are a summation of the correlation between religion and simulation theory:

  • Many major faiths feature life on earth as an illusionary transitory path toward an afterlife. In simulation theory, illusion is similarly a key concept behind our existence (Staggs 2013).
  • Most religious models require their participants to worship the God/s in some form. This may include inciting war in the interests of the God, or otherwise to “toil as their faithful servants”. If simulation theory is true, it’s likely we are allowing our own invisible creators to learn or experiment with our existence (Staggs 2013; Bostrom 2003), which forges a similar creator/creation relationship to that of major religions.
  • Simulation theory, at least as we understand it, is impossible to prove. This means that considering it as a possibility requires at least as much faith as a typical religion (Staggs 2013; Johnson 2011):

We have no good reason for thinking that God designed our universe. . . Every philosopher knows that it can’t be proven false that our universe is not some kind of “simulation”—perhaps an illusion created by some Cartesian evil-demon, or the dream of some great being (or perhaps our own dream). It could be that we are in The Matrix or a brain in a vat. . . Although there is no evidence that such things are true, there is also no definitive evidence that they are false (hence the sceptical problem). (Johnson 2011)

The majority of creation myths feature an outside agent who builds a system comprised of intelligent beings; this is the essence of all belief systems. In this sense, simulation theory is not far off being its own religion – albeit without its subjects worshipping an elusive ‘creator’ (McEvoy et. al 2018).

Humans are proud creatures (Herbert 2007). The notion of being the product of somebody else’s keystrokes is, collectively, something we find alarming. Even the idea of evolution being so offensive to a selection of humans is evidence of this; why are we so insulted that we might have evolved from apes (Korthof 2014)? We’d like to believe we’re part of a larger plan, of some sort of meaning. 

The Construct of Death

Questions surrounding human death are underpinned by a combination of “arrogant ‘knowledge'” and logical acceptance of the limitations of our own existence. In her examination of thoughts on euthanasia, Prokipiou names four factors in the human attitude toward death: the effect of technology, authority, selfishness and the weakness of  self-criticism and self control (Prokipiou 2016). The notion of death as ‘the unknown’, which is often approached with fear and uncertainty (Carlisle 2013), gave rise to religion. The purpose of religion, according to some critics, is to both recognise and provide a “spiritual response” to these fears (ibid). If we are assuming simulation theory is akin to religious order, we must consider it simultaneously as the product of a society desperate to know its purpose and that which craves understanding in a universe of uncertainty.

 

3.jpg
Death is a fear that humans have carried with them throughout time – are religion and theories such as this one to diminish our unease about the unknown? (source)

 

Thus it can be argued that the empty questions of “moral and philosophical ambiguity” to which we have no answer are made “soft and comfortable by the warm quilt of religious belief, be it technological or mythical” (Staggs 2013). Staggs goes further to claim that both religion and simulation theory offer simplistic solutions to complex questions about life. He therefore argues we should question why we find simulation theory attractive, rather than whether or not it is true. 

In examining simulation theory, the concept of death becomes even more complex. To determine the value of death in a simulation, we first have to unpack what it means to be real – which is almost a whole new research project in itself. In 1641, Descartes, a philosopher, put forward that even if an all-powerful demon tried to convince him to think he exists when he doesn’t, he would still have to exist for this deception to occur (Descartes 1641). Similarly, Descartes put forward his maxim “I think, therefore I am” (which has subsequently been used to explore creator/creature relationships in popular culture such as Ridley Scott’s Blade Runner in 1982) in an ambitious claim that having the ability to question our reality makes us real (Bennett 2017). Thus my current premise is that even if we are mere lines of code typed by post-humans, we still exist in some way.

The notion of death, however, does require we have a ‘self’, or personal agency – which we may not possess in the way we think (Staggs 2013). In this sense, coupled with simulation theory, our ‘death’ may be nothing more than “logging out of (the) illusion” (ibid) – which could expel us from existence permanently, temporarily, or even returning to some other reality instantaneously – which, despite technological connotations, really does not differ from religious notions of death.

 

2
As the world moves along the technological road, we need to be considering the implications of our newfound knowledge and ability (source)

 

Questions of defining, sustaining and understanding life are becoming increasingly urgent as we toy with droids, robots and artificial intelligence reaching human capacity – and possibly surpassing it, with vast implications for humanity as we currently know it. This is an issue we ought to consider proactively, before it becomes a reality (Drevich 2017). Here lies the hidden value alluded to in the introduction of this piece.

Digital Artefact: Update

A skeleton of my interactive narrative is currently live. At this early stage it is not possible to access the entire story, but the brief preview is accompanied by a countdown to when the final product will be completed. I am keeping the @simulationmindy twitter account updated with my progress, as well as compiling various resources for use in this project. I have also written a script for Chattr’s video series Kraj Kraj Conspiracies,  presenting simulation theory as a conspiracy for a young, non-academic audience. This video has not yet been published, but when it is I will incorporate it into this portfolio.

4 thoughts on “Implications of Considering Simulation Theory

Tell me what you think?

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s