I sent Part One to Professor Bostrom at his personal email address. No surprise: I haven’t heard from him and don’t expect to.
If you’re wondering why Professor Bostrom and his theories deserve attention, here’s a bit from a very long and detailed essay on him from The New Yorker Magazine:
‘Within the high caste of Silicon Valley, Bostrom has acquired the status of a sage.’ Whoa! A Silicon Valley Sage!
They go on…
‘Bostrom’s sole responsibility at Oxford is to direct an organization called the Future of Humanity Institute… Bostrom runs the institute as a kind of philosophical radar station: a bunker sending out navigational pulses into the haze of possible futures.’
‘The haze of possible futures,’ indeed! And further…
‘… when his arguments are challenged he listens attentively, the mechanics of consideration nearly discernible beneath his skin. Then, calmly, quickly, he dispatches a response, one idea interlocked with another.’ [my emphases]
Hey, Prof, how about ‘dispatching a response’ to my argument. Are you listening attentively? (Anyone who wants to goose the good Professor can forward Part One to email@example.com.)
An Open Letter to Professor Nick Bostrom, Part Two
Since writing Part One I’ve read Civilization One by the authors of Who Built the Moon?, the provenance of both the Megalithic Yard and the metric system is examined. The inarguable facts/observations/numbers put me way past the tipping point regarding a ‘higher intelligence’ actively fiddling with ‘our reality’ – by imparting information to our ancestors — and doing so in a way that would result in our knowing of its existence ‘scientifically.’ It’s an important book for those who believe in following the evidence.
In other words, given the choice between the existence of said higher intelligence and the belief in several billions-to-one coincidences (multiplied by each other), acceptance of the former would seem the rational choice.
As far as I know, numbers do not lie, Professor Bostrom: the sun, moon, and earth have been ‘put here,’ presumably for our… benefit (assuming existence is a benefit); given the role of the gas giant planets as ‘asteroid vacuums’, possibly the solar system as a whole was ‘designed.’ Knight and Butler have merely collected the numbers (in both books); the choice we are left with is some form of Intelligent Design, with or without caps, or some version of your SH (possibly both).
Let’s look at the evidence, see which way it swings us.
The star visibility question would seem to indicate SH by a non-supernatural power, given that a Strong I.D. (SID) would not likely need to cut corners. (The very definition of a SID might be based on this observation: ‘God’ would have infinite computing power.)
One might also see quantum physics’ ‘observation problem’ as another corner-cutter in the sense that if ‘nothing exists until it is observed,’ a lot of computing power/memory is saved. (I’m simplifying.)
The SH would seem to answer the old one, ‘If a tree falls in the forest and no one is there, does it make a sound?’ in the negative; if no observer is present, there is no tree to fall, no forest for it to fall in, no nuttin’. However, imagine this: We leave several recording devices at a place in the forest and come back some time later to find that a tree has fallen, and it’s all on our mechanical devices, our cameras. Isn’t there a problem here for SH, for its algorithm?
One might reply (as you have) that there is no problem because all that is needed is the image (and sound) of the tree falling for the later observer. But this is to avoid the issue. Assuming several devices (cameras with sound, and they all match content), it would appear that the ‘system’ needs to ‘know’ a lot more about simulation events than we would expect is possible, via this ‘later conscious observer’ assumption. Does ‘the program’ continually ‘watch’ every conscious entity to see if it’s planting a recording device to ‘observe’ something like a tree falling? Does it know if the camera works? (Or even what one is?) Or if the ‘observer’ is going to come back, get it, then watch the tape? How else could multiple cameras record the event? Is it possible, in principle, to create an algorithm that can make these qualitative judgments? Surely you don’t have ‘someone’ watching every human all the time, on the lookout for anomalous glitches! From a careful read of your paper, it would appear that you do. You write:
“…a posthuman simulator would have enough computing power to keep track of the detailed belief-states in all human brains at all times. Therefore, when it saw that a human was about to make an observation of the microscopic world, it could fill in sufficient detail in the simulation in the appropriate domain on an as-needed basis. Should any error occur, the director could easily edit the states of any brains that have become aware of an anomaly before it spoils the simulation. Alternatively, the director could skip back a few seconds and rerun the simulation in a way that avoids the problem.”
As fascinating as your paper is – I’m after all taking quite a bit of time to react to it – I found some of your language usage imprecise, with non-trivial implications. Aside from the above excerpt, looking once again at your Abstract…
ABSTRACT. This paper argues that at least one of the following propositions is true: (1) the human species is very likely to go extinct before reaching a “posthuman” stage; (2) any posthuman civilization is extremely unlikely to run a significant number of simulations of their evolutionary history (or variations thereof); (3) we are almost certainly living in a computer simulation. It follows that the belief that there is a significant chance that we will one day become posthumans who run ancestor-simulations is false, unless we are currently living in a simulation. A number of other consequences of this result are also discussed.
… terms like ‘very likely’ or ‘extremely unlikely’ or ‘almost certainly’ or ‘significant chance’ or (worse) ‘significant number’ seem to avoid important points, just one being that it would only take one surviving advanced civilization – if it (and only it) ran billions or trillions of simulations – to make it likely we are ‘living in a simulation’. (This is in reaction to #1, the ‘very likely’ imprecision, and in #2 where you say ‘extremely unlikely’.) This sort of problem runs throughout your paper.
An example of imprecision of a different sort is evidenced in the above excerpt. The first sentence glosses over the meaning of ‘keep track’ as it refers to ‘detailed belief states’ of all human brains. Even notwithstanding the distinct possibility that not all aspects of consciousness take place inside our skulls (see Rupert Sheldrake’s work), it also misses the point of ‘chaos theory’, which shows that certain complex systems are unpredictable in principal – quantum physics’ uncertainty principal might also apply on the level of consciousness. You also (by implication) seem to rule out free will, without a mention of having done so (by definition, free will is not predictable).
You blithely refer to ‘the director’ as if it were a conscious entity, rather than part of an algorithm. A rather busy conscious entity, I would estimate, given that he’d be simultaneously keeping track of seven billion belief states (plus possibly every recording device on the planet, including CCTV, of which there are hundreds of millions) – keeping in mind it may be impossible to ‘keep track’ of even one belief state. In short, I would humbly point out that you don’t seem interested in the matter of how things would actually work, and I don’t mean technically, but in principal.
I’ve taken the time to view most if not all your Youtube talks/podcasts and might here point out the one in which, in a thought experiment, you posit a choice made by a minor tyrant, to not kill a certain woman, who then goes on to be the great great (etc.) grandmother of Adolf Hitler; the implications of the tyrant’s ‘belief state’ would of course be immense for world history. You even mention the ‘butterfly effect’ (related to chaos theory, mentioned above). The point you were making is mine exactly: ‘keeping track’ of all the repercussions of any given event/belief state is impossible in principal. Aside from negating your own SH assumption regarding ‘omniscience,’ this observation will be vital when we discuss possible motives behind theoretical simulations.
(Keep in mind we are assuming huge but not infinite computational power in the Base Program.)
Speaking of possible inefficiencies or glitches in the Base Program, one possibility is the so-called ‘Mandela Effect.’ Although an in-depth analysis of this phenomenon is not called for here, my research indicates that it is real, although how we define ‘real’ is the precise crux of the matter. (It may be a psy op run by humans, for example, although if it is such, a huge amount of resources have been poured into it, with dubious effect.)
If you’re thinking that – from ‘what you’ve heard’ – this isn’t the sort of ‘theory’ a serious scientist/thinker would deal with, please keep in mind that your Simulation Hypothesis (SH) does not rule out that our universe is the result of some kid dropping a quarter into an arcade game (in another branch of reality). If you haven’t given the Mandela Effect a look, I urge you to do so. Here’s the most persuasive of the videos I’ve seen.
But the real persuasion is my own memory. Although many of the examples are iffy, some are not, and only possible via something very strange going on. One of those somethings could be a ‘glitch’ in the Base Program, i.e., more evidence for SH. Yet I haven’t come across a mention of it in your work.
Another question you don’t deal with nags: How would the base culture ‘observe’ the running of the simulation? Presumably, it could ‘run time’ at any speed it wished, to get to a ‘point of interest,’ say, but, again, what point of view would be available, even in principal? I mean, a laptop screen, a big flat one, or sense-around goggles? What, in principal, might they perceive?
Point being, what are they going to do, watch CNN to see how history is playing out? (Yes, irony!) Think of it this way: You and I ‘know’ what’s going on here on earth because we live here; wrong as we may be, we have a sense of this reality deep in our consciousness. Assuming a different ‘evolution’ of life or even just human history, what means of information gathering (experiencing the simulation) will the Simulators have?
Which brings me to the Big Problem of the SH concept itself: In running a simulation of any sort, ‘initial conditions’ are everything. Given that the present (mainstream) physics is based on the big bang, we have two possibilities. One, the big bang is untrue to begin with, which will equal the ultimate ‘garbage in’ scenario, or if it is true (which is highly unlikely), upon what algorithm would the program bring us to ‘the present’, or any version of ‘humanity’ at all? In my view, this is not a problem of our present inferior knowledge or technology but rather a version of Heisenberg’s uncertainty principal, and/or chaos theory. In your paper you write:
While it is not possible to get a very exact estimate of the cost of a realistic simulation of human history, we can use ~1033 – 1036 operations as a rough estimate. As we gain more experience with virtual reality, we will get a better grasp of the computational requirements for making such worlds appear realistic to their visitors.
What is missing here (and in your paper’s over all assumptions regarding SH and AI in general) is this problem of ‘initial conditions.’ As hinted at above, running an ‘ancestor simulation’ presupposes a sort of knowledge that we likely can never have, even notwithstanding unlimited computational power. (Chaos theorists have pointed out that even the weather may be unknowable after a limiting period of time. On a slightly different level, examine ‘The Game of Life’; after a certain number of ‘moves’ it becomes impossible to reverse-engineer back to the beginning – or to calculate the simple initial rules — such is the seemingly random nature/infinity of possibilities of the progression.
In other words, an ancestor simulation would have no predictive or explanatory power, possibly on even simple levels; predicting paths of biological evolution may be a good example (especially if you believe in Darwinism’s ‘random’ causal mechanism). Human history is another obvious one. Absent any predictive/explanatory power, what is the point of a simulation?
To put it still another way: It seems to me that given the appalling present state of cosmology/theoretical physics, plus the rampant untruths inherent in ‘history,’ plus the insuperable problem of computer simulations – best summed up as ‘garbage in – garbage out’ – plus ‘chaos theory/the ‘butterfly effect,’ there is no ‘quest for knowledge’ rationalization for running an ancestor simulation. The ‘scientific’ or ‘knowledge’ value would be zero.
Which brings me to what I refer to as the Injection Idea. The only reason for running simulation(s) of any sort, but particularly one that would ‘wind up’ with ‘our present’ is Entertainment. (Maybe survival but let’s leave that one alone for now.)
The best example of a simulation for entertainment that does not involve direct participation of the simulators would be the film The Truman Show. As good a movie as it was, can anyone picture a reality wherein a whole life is watched/viewed/experienced, on TV or any other way? Unedited? Picture it: Truman is born and then for three years does nothing but shit and cry. This is what a non-participatory simulation would be like. Who’s gonna watch that? Who the fuck cares? Storytelling involves dramatic structure! As a successful screen and TV writer: Hello! Anyone listening? (IMO, there is more to be learned about the motives and mentality of the powers-that-be [PTB] from Hollywood than from any other source. In researching this essay I recently viewed several feature films that bear upon this subject. More to come…)
Entertainment (or the separate subject of survival) would involve injection of entities from the Base Reality to interact with the denizens of the simulated reality. (Assuming this is theoretically possible, which is a helluvan assumption.) If so, and given the state of our world, the war, misery, greed, and above all, deceit – all those involved in this creation – the Simulators and the Injected Entities (IEs) — must be best characterized as… well… evil. To simplify: They are here to enjoy themselves at the ultimate theme park, at the expense of its inhabitants. They are here to wield power. The word ‘psychopath’ comes to mind.
When I lived in the outback of Costa Rica, some of us had a list of the Types who come down to live there, such as Explorers, both True and Half-assed, Homesteaders (ditto), Space Cadets (who can’t figure out how to get home), and Big Fish in a Little Pond Types. This last one comes to mind in SH (I battled almost to the death with one in CR, a true psychopath. I wrote a book about that experience… not that I think of him as an IE; if he were one, I wouldn’t be writing this).
Almost by definition, simulations for entertainment would be for the elite (of the Base Reality) only; and we may assume that events are either fashioned for the benefit of the IE’s or somehow supervised to ‘keep them out of trouble’ while they reek whatever selfish havoc they like. I assume that for the full ‘enjoyment of the IEs, the inhabitants of the simulation would have to be genuinely self-aware and capable of pain and pleasure, which would explain why we are so blessed.
So if SH is true, we are likely in a simulation run by psychopaths. Or just one. However many is not the point, although the term ‘Elite’ does again come to mind…
A final question (referred to previously) before we move on from SH: What is the relationship between the (presumed) Base Programmer and the intelligent agent (w/without caps) who left the astronomical-numerical clues referred to above and in Part One, and who likely created the earth/sun/moon arrangement? Are they one and the same?
ARTIFICIAL INTELLIGENCE AS IT STANDS NOW AND MIGHT DEVELOP
This part of my essay ran long so I’m making it Part Three, which (finally) gets to the real meat to the AI matter. Give me a couple days…
(I might as well spill it here: I do not believe we are living in a computer simulation. I’ll get to my reasoning in Part Three. And by the way, Professor Nick Bostrom doesn’t believe it either – he says – which means… for one thing… his head is up his ass, as a philosopher. But that’s not the worst of it with the good Prof…)
I’ve gotten some interesting replies to Part One. You got something to say, fire away. Use the Reply button or my direct email, acwdownsouth at good ol’ yahoo.com. If you want me to publish it in Part Three, let me know.
I’m writing these words from Death Valley. Here’s my view last night: