How to Tell if You Live in a Simulation
Home page: www.acampbell.org.uk
"[The Red King is] dreaming now," said Tweedledee: "and what do you think he's dreaming about?"
An ancient dilemmaAlice's dilemma is a very ancient one. The seventeenth-century Spanish playwright Calderón used it in his play Life is a Dream, and long before him the Chinese Taoist sage Chuang Tzu said on waking that he could not be sure if he was a man who had dreamt he was a butterfly or a butterfly who had dreamt he was a man.
A modern version of the dilemma: the simulation hypothesisThis theme has now been given a technological update: some recently published papers [1,2,3] by Martin Rees, Nick Bostrom and Robin Hanson on the Internet have discussed the possibility that we are all living in a computer simulation—that what we take to be an objectively existing material world is merely a virtual reality within a supercomputer.
Who is supposed to be responsible for this computer simulation? The proferred answer is: our own remote descendants, who are supposed to have taken computing power far beyond what exists today. (It seems to me equally conceivable that the simulators could be non-human beings on other planets, in other galaxies, or even in other universes; but this does not alter the essential proposition.)
Although the simulation hypothesis could be taken to be simply a modernised version of a very ancient idea, I think it does introduce something new: I mean the possibility that we might be able to discover clues to the real state of affairs, planted there by whoever is responsible for the simulation.
Ethical questionsAs Bostrom says, the simulators would have the status of gods so far as we are concerned. They created us and they could pull the plug on us if they so chose. Or could they? Even at our present relatively primitive level of computer simulation, people have seriously debated the ethics of killing off intelligent artificial life forms if such were ever created.
Might it not be the case that ethical issues would exist for our simulators too? Would they not feel responsible for our wellbeing? They might feel that, in creating us, they had created an autonomous life form to which they had obligations. This might make it difficult or impossible for them just to switch us off. Some theologians maintain that God has given us free will and cannot control our actions or eliminate us if we behave immorally; perhaps the same would be true of our simulators.
Christopher Langton, described by John Horgan as the founding father of artificial life, holds that the simulations of life run on a computer are literally alive. He said to Horgan: "I like to think that if I saw somebody sitting next to me at a computer terminal who is torturing these creatures, you know, sending them to some digital equivalent of hell, or rewarding only a select few who spelled out his name on the screen, I would try to get this guy some psychological help!".[John Horgan, The End of Science, p.200]
There is also another ethical question. Would the simulators be happy to leave us in perpetual ignorance of the real state of affairs, and of their existence, or would they wish to make themselves known to us, as God is supposed to have done? I think they might. But how would they go about it?
They could, presumably, simply alter the rules so that we were just naturally aware of the situation in which we found ourselves. We would then be living in a different kind of world, in which we all just knew, without question, that we were simulations. But this would be to change the whole scenario very radically; the simulators might be reluctant to do so or might consider it unethical to manipulate us to that extent.
If the direct manipulation of our brains is excluded, the most natural recourse for the simulators would probably be to give us occasional gentle hints, nudges in the right direction. That is, they would create occasional anomalies in our world which were intended to act as clues. These anomalies could not be too glaring or everything would break down and science would become impossible.
At the same time, they should be sufficiently incompatible with the results of our scientific inquiry that they could not be explained within a rationalist world view, otherwise they would not have any power to disturb us. They would be borderline phenomena, exceptions to the laws of nature as we experience them but occurring only rarely and unpredictably. It would be difficult or impossible to be sure that they really were exceptions; it should always be more or less possible for us to explain them away. This would require fine judgement on the part of the simulators but it should be manageable.
Bostrom does consider the possibility that anomalies might occur accidentally in the simulation and suggests that the simulators might wipe our memories to eliminate our awareness of them or might rerun the simulation by a few seconds to put matters right. This would work, so if anomalies do nevertheless occur it implies that the simulators permit them or even insert them deliberately.
Anomalous events within the simulation may give us a clueFor as far back as written records go there have been reports of what we now term the paranormal. Hauntings, premonitions, apparitions, and similar phenomena have been described throughout history. In the nineteenth century a group of British intellectuals came together and founded the Society for Psychical Research to study such phenomena; the American Society for Psychical Research soon followed (the philosopher and psychologist William James was a prominent member).
Today the emphasis has shifted from the investigation of spontaneous phenomena to laboratory studies of alleged psychological abilities such as extrasensory perception (ESP) and psychokinesis (PK); some results have been claimed but there is no agreement about what, if anything, it all means. A recurring difficulty is the absence of any plausible mechanisms to explain how ESP and PK might work. If we are living in a simulation, that problem disappears; ESP and PK work at times because the designers of the simulation decree that they should. No other mechanism is needed.
The same applies to the spontaneous material. Take the puzzling phenomena called poltergeists. Rationalists are generally content to dismiss all reports of this kind as due to gullibility, misreporting, and fraud. Their motive for taking this attitude is seldom based on a critical examination of the evidence, but rather on the view that poltergeists are so impossible to reconcile with a scientific world outlook that the reported events simply cannot have occurred.
But if, as Mike Dash has remarked in his pleasingly sceptical discussion of the alleged paranormal, although self-deception and fraud can explain the vast majority of reported anomalous events, poltergeists are an exception to this trend. The difficulty of accounting for such reports in a rationalist manner has been very well brought out by Alan Gauld and A.D. Cornell.
I concluded my review of their book by saying that "the authors have a lot of first-hand experience of the matter and I have none. I therefore have to keep an open mind about it, but, if they are right, the world must be a much stranger place than we rationalists like to think it is." Similar remarks would apply to a number of other kinds of phenomena which are difficult to explain within a rationalist world view. Most materialists cope with them simply by ignoring them, but it could be argued that this is scarcely a scientific attitude.
Post-mortem survival in a simulationIf we live in a simulation, it clearly provides for the possibility of post-mortem survival in various forms, including reincarnation. Most rationalists regard this idea as the implausible residue of religious thinking, but there have been some important philosophers and psychologists in recent times who have thought that the empirical evidence for survival, while not compelling, cannot be dismissed out of hand[6,7,8,9,10]. The main difficulty they have acknowledged is providing any plausible theory to account for the survival of personality, in whole or in part, after the destruction of the physical brain. The simulation hypothesis eliminates this difficulty completely.
To summarise: lacking the simulation hypothesis, we are confronted with the unpleasant choice of either dismissing a number of thoughtful, intelligent, and critical writers as gullible idiots, or accepting that our scientific world view has certain unacknowledged deficiencies—it doesn't join up properly at the edges. The simulation hypothesis offers a way out of this dilemma.
If the world we perceive, including its denizens (us), is a virtual reality, we may suppose that the simulators have inserted occasional anomalies such as poltergeists (not to mention ghosts, premonitions, UFOs, and similar intellectually distasteful items). Religions, especially those forms that depend on revelation, could also be included here. As has often been remarked, there is no clear dividing line between religion and the paranormal: UFOs have much in common with apparitions of the Virgin Mary. (At least one author has suggested that UFOs indicate that we live in a virtual reality; see my review of UFOs: The Final Answer?).
The cosmic intelligence testThe existence of anomalies of this kind could be construed as a kind of cosmic intelligence test, designed to make us suspect that reality is different from how it appears to us. There is nothing new about this idea. On the contrary, the notion of a universal key to unlock the secrets of the universe is an ancient one. We find it, for example, in the mediaeval Islamic sect known as the Ismailis (see my book The Assassins of Alamut).
The Ismailis believed that the Koran contained an inner significance, an esoteric secret to which they alone held the key, and they went on to construct a theosophy of extraordinary richness and complexity. Rather similar ideas are to be found in the Kabbalah, among the Cathars of the Languedoc in the Middle Ages, and in the writings of Jakob Boehme. An underlying theme in all this is the idea that the world as we experience it is unreal, a mere seeming, masking some unknown ultimate reality.
Most modern commentators would no doubt regard all this as delusive. Yet it would not be difficult to recast much of it in terms of virtual reality. For example, Bostrom suggests that there may actually be multiple layers of simulation, each layer acting as simulator to the layer beneath. The Ismailis also believed something of the kind. They had a hierarchical model of existence, with ten layers or Intelligences. Each layer was supposed to be populated by its own particular denizens, and to control the layer below while aspiring to the layer above, which it regarded as a superior Intelligence.
Elaborate mystical constructions such as Ismailism could be seen as an intuition, in theosophical and mystical terms, of what we now call virtual reality; a kind of foreknowledge of a state of affairs which only now is it possible for us to express in scientific terms. Whether we take this seriously depends, I think, on whether we take the notion of simulation seriously.
Footnote 1: But do you believe it?This is really a slightly unfair question, like asking the conjuror how he does his tricks. But, to come fairly clean, I wrote this piece as what Frances Yates called a ludibrium: a kind of intellectual game, but with a serious message contained in it.
As Kant and Schopenhauer maintained, it may be impossible for us to know the ultimate truth about the world; all we can know is appearance. And it is indeed conceivable that the appearance is constructed by gods or by their technological equivalents, an advanced society with supercomputers; after all, if cosmologists of scientific stature take it seriously it can't be total fantasy.
If you are religious you should also take it seriously. The designer(s) of the simulation would be equivalent to a creator God so far as we are concerned.
Personally, I wish to endorse rationalism and materialism, and I cherish the hope that even the most anomalous events will ultimately be explicable in naturalistic terms. Indeed, I believe that this is the case. But we are not there yet, and meanwhile it seems to me to be unscientific simply to pretend that everything is neat and tidy. The idea that these anomalies may have been planted in our world by the designers of a simulation strikes me as entertaining, if nothing more.
Footnote 2: New paperA new paper  by David Chalmers offers an excellent discussion of the simulation hypothesis and its philosophical implications.
Notes and references