About Rationally Speaking

Rationally Speaking is a blog maintained by Prof. Massimo Pigliucci, a philosopher at the City University of New York. The blog reflects the Enlightenment figure Marquis de Condorcet's idea of what a public intellectual (yes, we know, that's such a bad word) ought to be: someone who devotes himself to "the tracking down of prejudices in the hiding places where priests, the schools, the government, and all long-established institutions had gathered and protected them." You're welcome. Please notice that the contents of this blog can be reprinted under the standard Creative Commons license.

Thursday, April 26, 2012

Rationally Speaking podcast: Live at NECSS, David Kyle Johnson on the Simulation Argument

In this special live episode recorded at the 2012 Northeast Conference on Science and Skepticism, Massimo and Julia discuss the "simulation argument" -- the case that it's roughly 20% likely that we live in a computer simulation -- and the surprising implications that argument has for religion.

Their guest is philosopher David Kyle Johnson, who is professor of philosophy at King's College and author of the blog "Plato on Pop" for Psychology Today, and who hosts his own podcast at philosophyandpopculture.com.

Elaborating on an article he recently published in the journal Philo, Johnson lays out the simulation argument and his own insight into how it might solve the age-old Problem of Evil (i.e., "How is it possible that an all-powerful, all-knowing, and good God could allow evil to occur in the world?"). As usual, Massimo and Julia have plenty of questions and comments!


  1. With the uncertainty discussed in the podcast in coming up with the 20% figure, it seems that the error bars would include 0% and 100%.

    1. Well, one would never put error bars on an epistemic probability - it's not as if it's representing some physical Probability out there in the world that we are trying fallibly to measure!

      I think your intuition comes from the fact that you are far from sure if Bostrom's reasoning in getting to 20% is correct (as am I). But in that case, one should either factor the uncertainty of the line of reasoning into the probability itself, or else bracket it out by saying "given that this line of reasoning is correct, 20% probability."

      Moreover, 0% and 100% are precisely the two numbers that this probability estimate *cannot* include, though it can approach arbitrarily closely to them.

      P ∈ (0, 100), not [0, 100].

    2. I agree with your response, Ian. I was being a bit sarcastic and not thinking clearly when I decided on the wording of that. I saw the episode at NECSS, but did not listen to the podcast. I enjoyed the speculation and discussion, but wished there was a bit more back-and-forth between everyone. I realize that becomes tricky with a live audience.

  2. the case that it's roughly 20% likely that we live in a computer simulation

    There's no such case, and it discredits both you and philosophy to make this claim repeatedly.

    1. E-mail Andrew Gelman and post his response here.

  3. That was a really fun episode.

  4. This comment has been removed by the author.

    1. This comment has been removed by the author.

    2. This comment has been removed by the author.

  5. I was at NECSS and I've gotta say, this was the most provocative and lively session. Lots of speculation, yes, but fun. Thanks for having David Kyle Johnson as a guest. I hope there'll be a video of this sometime soon -- If anyone knows of one, let me know.

  6. It's always fun to watch atheists sneak into church, and make no mistake, this episode was more theology than skepticism. Bostrom, Chalmers, Kyle and now Julia and Massimo are all engaging in spiritualism. And as when any atheist enters church, there is a lot of excitement from the pews. But just because an argument has entertainment value, that doesn't make it a worthwhile argument.

    Massimo did a decent job of trying to place the discussion in the proper philosophical context, but if you look at the presumptive reason for presenting this topic in the first place, then I think it's clear that he's slipping back into the "mysteries of faith" that he so abhorred as a child. Or, at best, he is still stuck in a debate with those priests that really aught to be left in his childhood.

    Kyle's added value to the whole Bostrom project is only a clever "gotcha" argument for academic theists. This begs the question: is playing "gotcha" with theists (as rabble rousing as it might be) a useful use of a public intellectual's time? We don't want to play their game, do we? Then why spend so much time trying to point out to them that their game has faulty rules? How about we play a whole new, non-God game?

    To be clear, I don't think Church can really be avoided. I think Church is hardwired into us. But as skeptics we need to be more transparent and responsible about when and how we use Church.

    Bostrom is not being honest with himself or us when he says there is only a 20 percent chance that we are in a simulation. By his model (and even more obviously on a psychological level) there is a 100% chance we are in a simulation. Because simulated beings can certainly develop simulations themselves (their very existence proves it possible) so the vast majority of beings will be simulated beings on the zillionth level of nested simulations. If there is a 20 percent chance, it has to become a certainty. Actual beings are an incredible minority! And more simply put, tell anyone there is a 20 percent chance of something as mind-blowing as this, and you have effectively made it a certainty.

    At the end of the day, the Matrix was a bad movie with good special effects. Just as it fails to provide a coherent vision of the aliens (they need us for "power"? So we hallucinate the world why?) The Matrix/simulation argument is just a loophole for people who don't want to admit they are pushing all their metaphysical uncertainty (and ultimate fear of death) onto some mysterious super other being. But God is God even if you want to dress him up as an alien or a post-human of the future. And we are going to die. As skeptics we should be working hard to figure out how to live without God and with death.

    It could be just too dang hard for us to do. If Julia and Massimo can't resist the siren call of the choir, who can?

    1. I agree that the application of the simulation argument to theology is, while clever & amusing, a bit over the top. I didn't get the whole "atheists going to church" vibe though, and certainly not from Julia & Massimo.

      On the other hand, the simulation argument itself I find very interesting, as will all lines of broadly anthropic reasoning. Not least because it brings out the intuition in me very strongly that there is no principled difference between living in a simulation & living in the 'real world.' I have a wonderful proof of that, which this margin is too small to contain.

    2. I have the same intuition, Fermat, and kept returning to it listening to the podcast. Although I would need to hear a rather fantastic argument to even allow that it's not obviously true.

    3. Julia and Massimo are taking Bostrom seriously, as you seem to be too! Say it aint so.

      Let's paraphrase Bostrom to draw out the theology of the argument.

      Simulation = world,
      Post Human = God
      Some humans might become gods. Some of those gods might create worlds. Therefore, it is possible that our god is a former human.

      If you are talking about the nature of God, you are engaging in theology, right?

      The Naturalistic Method seeks explanations for things in this world from within this world, right?

      There is no "simulation" that fits the bill for this fantasy. I know they have a holodeck on the Enterprise, but a) that aint real and b) that's a simulation entered into by beings from that actual world (of Star Trek, for crying out loud!). That's not what Bostrom proposes. He proposes and entirely digital simulation in which there is an entirely digital being. It's hard to conceive of a purely digital simulation in which one part of the program "experiences" another part of the program. If such a thing is possible, then you aren't talking about a simulation, because then the part of the program that is experiencing the other part of the program is actually experiencing it. It's not a simulation. It is a world.

    4. Here is a better way to put Bostrom to bed. I believe the following argument is identical in all important ways to Bostrom's "Simulation Argument."

      At any given moment, half the humans on the planet are asleep. They might be having very vivid dreams about beings. They might have multiple such dreams a night. Therefore, it is as or more likely that I am a dream being. That is to say, I am either dreaming or in a dream, statistically speaking.

      He claims his argument is not rehashed Descartes, for precisely this reason. However, I'm not buying it. I'm disappointed that Julia and Massimo did buy it.

      Do you dismiss this "argument" as a foolish? If so, how is my argument different from Bostrom's?

    5. OneDay,

      sorry to disappoint you, but I don't even see the parallel between your example and Bostrom's argument. The latter is based on a certain assumptions about computability and philosophy of mind, which have no parallel in the case of dreams.

      Incidentally, I made clear in the episode that I actually don't buy one of Bostrom's crucial assumptions, a sufficiently strong version of the computational theory of mind. So I only entertain the simulation hypothesis as an interesting philosophical exercise, not as a real possibility.

    6. Massimo, for me, the assumptions about computability and philosophy of mind are exactly the parallel I'm trying to draw. They are HUGE assumptions. Unless the exercise varies significantly from Descartes, I wonder what the value of the exercise is? Yes, we have SIMS, just as Descartes had dreams. For the avatars in SIMS, is their "experience" different than dreaming? Can we intelligibly talk about SIMS characters having "experiences." For me this is all the stuff of dreams.

    7. I still don't see the parallel at all. Dreams are not even close to be as realistic as what we are talking about. And no, nobody is thinking that the sims are conscious. I see a broad analogy there, but nothing that really impinges on Bostrom's arguments (which, as I said before, I actually don't buy because of the computational assumption).

  7. "Wilson-type consilience is actually a scientistic and anti-intellectual enterprise." Perhaps wrong but why anti-intellectual? According to Wikipedia, "Wilson held that with the rise of the modern sciences, the sense of unity gradually was lost in the increasing fragmentation and specialization of knowledge in the last two centuries. He asserted that the sciences, humanities, and arts have a common goal: to give a purpose to understanding the details, to lend to all inquirers "a conviction, far deeper than a mere working proposition, that the world is orderly and can be explained by a small number of natural laws." Wilson's concept is a much broader notion of consilience than that of Whewell, who was merely pointing out that generalizations invented to account for one set of phenomena often account for others as well."
    http://en.wikipedia.org/wiki/Consilience Is the Wikipedia entry (as quoted) correct?

  8. David,

    no, it isn't. If you read the book, Wilson is after wholesale reduction of the humanities to the sciences.

  9. I've got a...disagreement(?) with Johnson, and MP, from this podcast. You both assumed something as obvious that I don't think is so obvious. When mocking the "natural evil is necessary" argument, MP mocked it by saying something like...the massive floods in India could not possibly be offset by the good done in reaction to them.

    But this is ignoring the slightly subtler (and more workable) explanation that general growth, development, learning, even at the most primitive cognitive levels, occurs only in reaction to challenges. Of course, an omniomni god could easily create brains with similar structures without the suffering, but couldn't the _memory_ of suffering still necessarily be present, making this an odd form of Last Thursdayism? And a less than omniomni god could definitely find it impossible to create structures of such interesting computational complexity from scratch WITHOUT actually doing it...

    Just thinking here, but if you're not considering an omniomni God, maybe something like the Lisa Simpson God from Treehouse of Horror VII, or if you're just thinking about the real world, then opposition becomes a good thing, with the twin reasons of being new experiences, and the strengths we learn in overcoming it. Which also deals with your argument about "why do they try to fix any of these problems, if the problems themselves are good?"

  10. BubbaRich,

    Let us grant everything you just said. It would then follow that an omnipotent / omniscient / omnibenevolent god is compatible with ***any*** state of moral affairs. So, e.g., one could claim a world maximally evil, i.e., a natural world with the greatest amount of suffering possible, is compatible with god because we simply do not if that world conduces to bringing about some greater good.

    I think at this point terms such as 'good' and 'evil' lose their meanings and we can no longer claim, in any coherent sense, that some agent is 'good' or 'bad', to include even god, since before we ascribe such properties we must first understand what those terms mean and under what circumstances they are to be applied. So, e.g., if there is a world where every infant is raped and tortured and yet one can insist (as your comments entail that they can) that such a state of affairs is, in the end, morally necessary and thus 'good', I no longer have any idea what 'good' and 'evil' mean.

  11. I liked OneDayMore's dream analogy because you have no way of knowing you are in either a dream or a simulation. Dreamologists will disagree and point to 20 to 30 ways you can tell the difference between reality and a dream but as others say, life is but a dream.

    The important aspect of it all is that there seems to be precious little we can do about it if we are indeed pieces of a large program. The simulation idea makes a lot of sense, but I have no opinion about the probabilities. My take is simply that seeing that we made computing machines, how could we possibly have NOT been similarly created.

    Was lucky enough to meet one of my favorite authors last night at a talk. He is a novelist who brushes his pages with 'post-modernist' musings about science and philosophy but at the end of the day, like many people here, drives philosophy to new places but does not directly extend scientific knowledge.

    I asked him where he stood on science, given these musings (money - the entire money supply - as a very very large object, a world of backwards time flow where a guy who punches you in the face is a good person because he instantaneously heals your cuts and bruises, or even panpsychism: "Albert tired of his food. Albert's food tired of him" and so on. He replied that the only branch he felt especially close to was cosmology; he felt that it was a depressing business these days, and brought up the sorry state of string theory, (good product, no buyers), and started on how the universe will always be cleverer than us. What the scientists HAD done, says he, is prove god's existence, even if you postulate god as an intelligent universe. He could have gone on but the talk had ended, he had to autograph books, and I had to go home.

    "The universe will always be cleverer than us". Hmmm, a moving target.


Note: Only a member of this blog may post a comment.