The simulation argument

by Neil Rickert

In a recent post over at Scientia Salon

Mark O’Brien asks a question and gives his own answer with:

Could a computer ever be conscious? I think so, at least in principle.

As O’Brien says, people have very different intuitions on this question.  My own intuition disagrees with that of O’Brien.

Assumptions

After a short introduction, O’Brien presents two starting assumptions that he makes, and that he will use to support his intuition on the question.

Empirical assumption 1: I assume naturalism. If your objection to computationalism comes from a belief that you have a supernatural soul anchored to your brain, this discussion is simply not for you.

Personally, I do not assume naturalism.  However, I also do not believe that I have a supernatural soul.  I don’t assume naturalism, because I have never been clear on what such an assumption entails.  I guess it is too much metaphysics for me.

Although O’Brien calls that an empirical assumption, I take it to actually be a metaphysical assumption.

Empirical assumption 2: The laws of physics are computable, that is, any physical process can be simulated to any desired degree of precision by a computer.

It is not entirely clear what that assumption means.  Certainly, laws such as Newton’s f = ma are computable as mathematical functions.  That this is made an assumption suggests that O’Brien means something else.  My guess is that he is talking about a presumed but perhaps unknown set of laws that could be said to govern the universe.  That there are such laws would be another metaphysical assumption, and one of which I am quite skeptical.  In particular, QM would seem to pose a problem for those who believe that there are such laws.

My view of consciousness

I see consciousness as resulting from a biological adaptation that allows us to cope with unexpected contingencies.  If we were to program a robot, say to explore Mars, we would have to program into it the ability to cope with every possible contingency that could arise.  But we cannot know what contingencies might arise.  That’s why NASA designs its Mars explorer missions so that they can be reprogrammed as needed.  My view is that consciousness evolved because we also need to be deal with unpredicted and unexpected contingencies.  In some sense, consciousness is there because we cannot have been born with innate knowledge of the kind of metaphysics that O’Brien wants to assume.

For me, the real problem solved by consciousness and by intelligence, is that of allowing us to learn how to cope with the unexpected.

The problem with simulation

You cannot just take some aspect of the world, and simulate it.  Simulation requires a great deal of knowledge.  So if we built systems that behave like humans, and do it with simulation, then we will have to program into them an enormous amount of innate knowledge.

If the role of consciousness is to allow us to learn, then preprogramming with innate knowledge would seem to undermine that role.  For the programming would result is a system that had no need to learn, and thus no need for consciousness.

Some people argue that consciousness is an illusion.  It seems to me that, at best, the kind of simulation proposed might generate an illusory consciousness, but not a  real consciousness.

A note on dualism

I’m no expert on Descartes.  However, it is my impression that he saw Newton’s mechanics as suggesting a kind of clockwork universe.  And apparently he gave us his version of dualism because he did not see how a clockwork universe could account for the human mind.  So, if my impression is correct, we would have to conclude that Descartes’ intuition did not support the kind of viewpoint that O’Brien is arguing for.

For myself, I do not believe that we live an anything like a clockwork universe.  So I have no need to appeal to dualism.  I see the universe as such that consciousness could evolve as a biological adaptation (and probably did).

In short, I do not find the simulation argument at all persuasive.

Advertisements

51 Comments to “The simulation argument”

  1. The complexity of the human brain can’t be simulated on all the computers in existence. I think you’re dead on with your assessment of these so-called simulations, even if i disagree with you about the origin of consciousness.

    Liked by 1 person

  2. Neil, As nearly every one of your posts is a paradigm example of metaphysical musing, I have no idea what you might mean by something being “too metaphysical for you.” Certainly, for example, the question of whether computers could think is no less metaphysical than the question of what “naturalism” means. And the question of what philosophers generally mean by “naturalism” and much of what the assumption of the truth of their understanding of naturalism entails is also broadly empirical. Yet you persist in this “Metaphysics BAD–What I do here–GOOD” silliness.

    It is not unusual for those doing metaphysics to deny that they are (it’s quite common, in fact), but I don’t know of anybody (since Carnap and Schlick, anyhow) who is both so utterly attached to practicing metaphysics and so vehement about claiming to eschew it as you are.

    W

    Like

    • A basic definition of Naturalism is simply the idea that only the laws of nature (physical laws — even if we can’t know what they all are) operate in the world.

      Given that I am skeptical on whether there are laws of nature, I should continue to not take sides on the question of naturalism.

      What would make it more persuasive for you?

      Supportive evidence.

      It seems to me that our interaction with the world is mainly pragmatic, rather than rule-based. When we follow rules, these are usually rules that we have adopted for pragmatic reasons.

      The simulation argument or, for that matter, any computationalist account will provide a system which is rule-based, but where the rules are externally imposed. I don’t see it as fitting what we actually see.

      Like

      • “Given that I am skeptical on whether there are laws of nature, I should continue to not take sides on the question of naturalism.”

        Why are you skeptical of there being laws of nature? Science itself and the scientific method is dependent on such laws existing. Without them, there could be no predictions. Note that laws do not have to imply determinism, but laws are implied if there is any predictive order at all. Otherwise we would be dealing with unlikely coincidences, every time we perform a repeatable scientific experiment.

        Like

        • Why are you skeptical of there being laws of nature? Science itself and the scientific method is dependent on such laws existing.

          I see our scientific laws as pragmatic inventions, not as something that comes as part of nature.

          Note that laws do not have to imply determinism, but laws are implied if there is any predictive order at all.

          We construct useful rules for our own purposes. We can call them “laws”, but that does not make them “laws of nature”.

          Like

          • “I see our scientific laws as pragmatic inventions, not as something that comes as part of nature.”

            It may be that scientific laws are subsets of rules (or an approximate subset) derived from nature for pragmatic purposes, but they seem to be derived from the externally imposed rules of nature nevertheless. Again, if our scientific laws offer predictive capabililty, they must have a correlation with reality and the laws of nature. They may only be a subset, and they may only be an approximation, but they are correlated with and thus derived from the laws of nature if they offer any way to predict events.

            “We construct useful rules for our own purposes. We can call them “laws”, but that does not make them “laws of nature”.”

            If what we call “laws” are never violated, so far as we can tell, then those laws are laws of nature by definition, even if taken alone, they end up being incomplete or do not account for all the other laws of nature. Generally speaking, I think that our laws tend to be “approximations” to the laws of nature, but they are correlated nevertheless. The discovery of scientific laws, and newer ones to replace older ones, are merely our attempt of describing the laws of nature more accurately over time. They become better correlated over time as the laws are improved upon, and with that we increase our predictive capability in the process, indicating that we are getting closer to the actual laws of nature, even if those laws are forever out of our grasp, epistemologically speaking.

            Like

          • It may be that scientific laws are subsets of rules (or an approximate subset) derived from nature for pragmatic purposes, but they seem to be derived from the externally imposed rules of nature nevertheless.

            There is no evidence, as best I can tell, that our rules are derived from imagined (but unevidenced) externally imposed rules.

            You are sounding very much like a creationist, and you cling to your metaphysics in about the same way that a creationist clings to his “Adam and Eve” story.

            Again, if our scientific laws offer predictive capabililty, they must have a correlation with reality and the laws of nature.

            A correlation is a numeric quantity computed from two sets of numeric data. It makes no sense, as you are using that word.

            If what we call “laws” are never violated, so far as we can tell, then those laws are laws of nature by definition, even if taken alone, they end up being incomplete or do not account for all the other laws of nature.

            You seem to be making up stuff as you go along.

            If our rules are formulated in such a way as to be analytic truths (logically necessary truths), then they will never be violated. Why would that make them laws of nature? N.R Hanson argued that scientific laws (or at least some of them) are analytic, in his 1958 book “Patterns of Discovery”. C.I. Lewis argued that some scientific laws are a priori truths, in his 1923 paper “A Pragmatic Conception of the A Priori.”

            Like

          • “There is no evidence, as best I can tell, that our rules are derived from imagined (but unevidenced) externally imposed rules.”

            The evidence is all around us. If our rules give us predictive power, they are correlated with how reality operates (i.e. the laws of nature).

            “You are sounding very much like a creationist, and you cling to your metaphysics in about the same way that a creationist clings to his “Adam and Eve” story.”

            I have no idea how you arrived at such a silly assertion. Creationists ignore evidence. I welcome and embrace it. My metaphysics? There need be no metaphysics here, other than perhaps the assumption that the universe operates in a certain way (whatever that way is), and that how the universe operates is defined as the laws of nature. We discover correlations with these laws of nature whenever we make repeatable predictions based on our formulations of scientific laws, etc. I’m not sure where your “creationist” accusation comes from, as my line of reasoning has been pretty antithetical to such views.

            “A correlation is a numeric quantity computed from two sets of numeric data. It makes no sense, as you are using that word.”

            You are describing a quantitative correlation, which is not what I’m arguing exists. I never argued that we can determine how closely our laws correlate with the laws of nature, only that there is a correlation, and an increasing correlation over time. However, we can certainly quantify the correlation in some way by looking at the rate of error between our predictions based on our laws and the observed phenomena we’re trying to predict. If our uncertainty goes down, the correlation to reality goes up.

            “You seem to be making up stuff as you go along.”

            Such as?

            “If our rules are formulated in such a way as to be analytic truths (logically necessary truths), then they will never be violated. Why would that make them laws of nature?”

            I wasn’t suggesting that our laws are formulated to be analytic truths, rather I was merely pointing out that for those laws that are violated very rarely (if at all), they correlate to reality (see comments above).

            Like

          • If our rules give us predictive power, they are correlated with how reality operates (i.e. the laws of nature).

            This is plainly absurd. You are making it up.

            Like

          • “This is plainly absurd. You are making it up.”

            You can’t simply say it is absurd, and that I’m making anything up. You need to explain why it is absurd (and what exactly), and why you think it is made up. Otherwise, you are just dodging the issue here.

            Like

          • You can’t simply say it is absurd, and that I’m making anything up.

            Sure, I can.

            You are the one who is making a claim. It is your responsibility to provide the supporting evidence.

            I’m simply telling you that your claim is nonsense. This should be obvious to anyone who knows the mathematics.

            Like

          • Show what exactly is non-sense (quote me with each individual concept that is nonsense). You are evading the issue rather than refuting it.

            Like

          • You are evading the issue rather than refuting it.

            What’s to refute?

            You have made an obviously false claim with zero supporting evidence.

            I’ll quote Hitchens’ razor: “What can be asserted without evidence can be dismissed without evidence.”

            Like

          • “I’ll quote Hitchens’ razor: “What can be asserted without evidence can be dismissed without evidence.” ”

            Hitchens certainly wouldn’t approve of your “method” of argumentation. You simply assert “That’s nonsense”, when I’ve asserted my position with evidence. So heed Hitchens’ advice yourself. Provide evidence or an argument to refute the claim that predictive capability demonstrates a correlation between physical laws and how the world works.

            Like

          • Provide evidence or an argument to refute the claim that predictive capability demonstrates a correlation between physical laws and how the world works.

            Neither “physical laws” nor “how the world works” is the name of a numeric function.

            Your demand is nonsensical.

            Like

          • See my previous comments. You are not paying attention to what I’ve been saying. You are looking for something that I’m not claiming exists. You need to read my explanations more carefully.

            Like

          • You’ve not shown how our laws which have predictive power are NOT correlated to reality (and thus to the laws of nature). That they have predictive power implies that they are correlated with how the world actually works. You haven’t successfully argued how this can’t be the case. You need to find a way to successfully argue that our laws do predict phenomena yet aren’t correlated with how the world works. There’s no point in beating around the bush or using red herrings as some kind of defense tactic. Either show how laws that produce valid predictions are NOT correlated (i.e. linked) to reality (i.e. how the world works, laws of nature, etc.), or you clearly have no evidence to support your position. The evidence for my position (that laws of physics correlate with laws of nature) is demonstrated by those laws of physics allowing us to make successful predictions.

            Like

          • You’ve not shown how our laws which have predictive power are NOT correlated to reality (and thus to the laws of nature).

            It is not up to me to show that. It is your claim, so it is up to you to show that.

            I have no idea what “correlated with reality” even means. Normally, a correlation is between two random variables. Roughly speaking, that’s between two numeric valued functions. What numeric valued function does “reality” refer to?

            That they have predictive power implies that they are correlated with how the world actually works.

            “Implies” normally means that there is a logical argument. I’ve asked you to produce the logical argument, or the mathematical theorem. Why won’t you?

            And, to get back to the previous point, which numeric valued function does “how the world actually works” refer to?

            You need to find a way to successfully argue that our laws do predict phenomena yet aren’t correlated with how the world works.

            But that’s trivial. “How the world works” is not the name of any numeric valued function, so it is not a proper predicate to the verb “correlate”.

            There’s no point in beating around the bush or using red herrings as some kind of defense tactic.

            I haven’t been beating around the bush. That’s what you have been doing. You have been making claims which, on their face, are nonsensical. And when I point that out, all I get is bluster.

            You claim there’s a correlation, but you have not indicated which numeric valued functions are correlated. You claim there is an implication, but you have not provided the logic nor an appropriate citation.

            Like

          • “You claim there’s a correlation, but you have not indicated which numeric valued functions are correlated. You claim there is an implication, but you have not provided the logic nor an appropriate citation.”

            What part of my argument are you not understanding exactly? It’s pretty simplistic and you are needlessly over complicating it. I told you that I was using the term “correlated” in a general sense, that is, that physical laws “agree with” the laws of nature. You are looking for some version of correlation that I’ve not postulated. I’m using the term generally which I stated in an earlier response. If you want to quantify the correlation in some mathematical sense (which I’m not arguing you need to do, you are arguing this), then I supposed you could plot a predicted event (say the predicted time it takes for an object to hit the ground) with respect to the actual observed result (the actual time it takes for the object to hit the ground). If I do this for many data points, i.e., dropping a ball from different heights, I will get a number of data points, predicted data points and actual measured data points. One could then calculate a correlation coefficient between the two (predicted value vs. actual value). What about this is unclear? Please quote something specific I’ve said here so I can address it accordingly.

            Like

          • What part of my argument are you not understanding exactly? It’s pretty simplistic and you are needlessly over complicating it. I told you that I was using the term “correlated” in a general sense, that is,

            Ah, okay. So you are using “correlated” in a sense that does not mean correlated. No wonder I had difficulty understanding you.

            , that is, that physical laws “agree with” the laws of nature.

            What, precisely is it that agrees with what?

            Next you are going to tell me that you are using “agree” in a sense that does not mean agree.

            In short, it’s all meaningless waffly feel-good nonsense.

            It is easy to come up with two function, say f and g such that either one of them completely determines the other, yet there is no correlation between them. I can do this with a Fourier transform, or I can do this with cryptography. One completely determines the other in a holistic sense. The whole of one determines the whole of the other. But it cannot be broken down to some small part of the value of one determining or correlating with some small part of the value of the other.

            If you want to quantify the correlation in some mathematical sense (which I’m not arguing you need to do, you are arguing this), then I supposed you could plot a predicted event (say the predicted time it takes for an object to hit the ground) with respect to the actual observed result (the actual time it takes for the object to hit the ground).

            But that has nothing to do with what you have been claiming. That merely shows that our predictions work reasonably well. It says nothing about whether our scientific laws are correlated to reality (whatever that is supposed to mean).

            Like

          • “Ah, okay. So you are using “correlated” in a sense that does not mean correlated. No wonder I had difficulty understanding you.”

            No, I’m using it in a general and perfectly legitimate sense as I mentioned several replies ago. You must have failed to read that in my previous reply, and you have obviously been continuing to infer that I’m implying some mathematical correlation between
            As per: http://dictionary.reference.com/browse/correlation, we can see that a synonym is “matching”, which is the sense I used the term. There is a match (or close match) between the predictions provided by our physical laws and the actual observed instances of what we predicted. Thus, there is a correlation. If you disagree with the dictionary, I suggest you take it up with the editors and publishers, not me.

            “What, precisely is it that agrees with what?”

            As I’ve said countless times now (I don’t know why it’s going in one ear and coming out the other), the predictions provided by our physical laws agrees (although not necessarily completely) with the observed instances of said phenomena. Predictions agree with observations. That’s what I’ve been saying from the beginning, which is such an obvious point, I honestly don’t know how you could take issue with it. It’s seems like a no-brainer.

            Like

          • The dictionary definition that you reference is still about a relation between two kinds of data. We do not see that where Fourier transforms are involved.

            Predictions agree with observations.

            I’ve been agreeing with that all along. I’m inclined to call that an entirely epistemic claim.

            My disagreement is with where you have made assertions of correlation between scientific laws and reality, which looked like a metaphysical claim.

            Perhaps you are now agreeing that was an overstatement.

            Like

          • “The dictionary definition that you reference is still about a relation between two kinds of data. We do not see that where Fourier transforms are involved.”

            The dictionary definition I gave has “matching” as a synonym for “correlation”. Would you like to look up the definition for “matching” and confirm that it is limited only to a relation between to kinds of data? Even if that was what you found, my claim is that there is a relation between predictive data and measured data. So either way you want to play the semantics game, it will be to no avail. I suggest you focus on the main content of the discussion and not semantics, as it is a waste of time since you clearly know what I mean (now at least).

            Like

          • The dictionary definition I gave has “matching” as a synonym for “correlation”.

            In your assertion that scientific laws are correlated with reality, which particular matches which particular.

            Even if that was what you found, my claim is that there is a relation between predictive data and measured data.

            I have been agreeing that predicted data matches measured data.

            Can I take it that you are dropping the part about a correlation between scientific laws and reality?

            I suggest you focus on the main content of the discussion and not semantics, as it is a waste of time since you clearly know what I mean (now at least).

            I’ve been trying to persuade you to explain what is the content of “scientific laws are correlated with reality.” I do not see any content in that. It just seems nonsensical.

            Like

          • “I have been agreeing that predicted data matches measured data.

            Can I take it that you are dropping the part about a correlation between scientific laws and reality?”

            That predicted data matches measured data was my assertion all along, i.e., that there is a correlation between scientific laws and reality.

            Like

          • “My disagreement is with where you have made assertions of correlation between scientific laws and reality, which looked like a metaphysical claim.

            Perhaps you are now agreeing that was an overstatement.”

            Yes, I made assertions of a correlation (a match or close match) between scientific laws (i.e. their predictions) and reality (i.e. what we actually measure). I don’t think that this is an overstatement, nor metaphysical, unless you are defining “reality” metaphysically, rather than the fairly standard parlance and use of the term.

            Like

          • Hey, guys. I’d be happy to referee this contest, or at least tell you whose posts are more convincing. (I assume you care about being convincing to others or you’d conduct this discussion via email rather than on a public blog on the internet.) But I have two conditions.

            First, both parties would have to admit that this argument reflects a metaphysical disagreement.

            Second, both parties would have to agree that my response–at least with respect to who (if either) has made a more convincing case on each issue I touch on, is final. Naturally. you may think that both your readers are confused, mistaken, etc. But as there are just the three of us,,,,

            So, who’s got cojones for arbitration?

            XXOOXX,

            Prof. Waltie

            Like

          • Thanks, Walto. But I don’t think a referee is needed.

            Anyone reading the thread and comments is welcome to make their own decision as to which arguments are more convincing.

            Like

          • I agree with Neil. Conducting this discussion via a forum/blog allows others to see various trains of thought, points of view, arguments, etc. Whether or not they are convincing is up to each individual to decide for themselves. Many times, we digress from the main argument due to semantic issues, but regardless, others may benefit from these digressions, especially if they have similar presumptions or a lack of clarity that’s important to address.

            Like

          • “I agree with Neil. Conducting this discussion via a forum/blog allows others to see various trains of thought, points of view, arguments, etc. Whether or not they are convincing is up to each individual to decide for themselves. ” I’m sorry to report this to you guys–but there’s only one other (me), and I’m afraid there will never be anybody else. And you two are obviously never going to convince each other of anything. So that leaves only the question of which, if either of you, has made any arguments or claims that are convincing to me. I mean, I wish posterity or something would take an interest…..but alas. So you can continue your nyah nyahs indefinitely, or you could learn something about whether your argumentative skills are any good according to somebody with no dog in this fight.

            And, as indicated, I’d have told you, but only under my conditions.

            Natch, however, one thing you two CAN agree on is that neither of you wants anybody (impartial or not, philosophy professor or not) telling you whether you’re making a good case for your views.

            OK, by me! It’s your absolute right and privilege to continue gainsaying each other forever! Anyhow, no doubt I’m dead wrong, and a resolution is just around the corner. After all, we’ve found you do agree on the cojones matter! 😉

            Like

          • “So that leaves only the question of which, if either of you, has made any arguments or claims that are convincing to me.”

            Actually, Neil’s opinion of my views and my opinion of his matter (unless you are claiming that our opinions on each other’s views simply don’t matter). They matter to us. Ironing out philosophical differences or even semantic problems is beneficial even if the other person refuses to budge. At the very least, the dialogue allows each of us (Neil and myself at the very least) to clarify our positions and talk about the issues. Sometimes we may never have considered that there were certain errors or unjustified assumptions in our views, unless the dialogue occurs. That is my opinion anyway. Anyone that chooses to comment on this issue or on the discourse between Neil and I, may also provide benefits (to themselves, to Neil, or to myself). If you don’t talk about these issues, then they can only stay in one’s mind, and forever remain out of 3rd party scrutiny. Keep the dialogue going always (I say)! 🙂

            Liked by 1 person

          • I don’t think the ironing out is going too well, which is why I offered to begin with. I mostly see flat denials. But, hey, if you think these “no YOU’RE wrong!” discussions are valuable–or even if you just like them–why should I be a killjoy?

            Carry on!

            W

            Like

          • Dialogue is always valuable, even if there is no mutual agreement. Often times, there is indeed agreement and new perspectives can help us establish our own views more coherently. Peace!

            Liked by 1 person

      • “Supportive evidence.”

        Such as what (for example)? Or do you not know what you would consider supportive evidence? Know that I’m mainly playing the devil’s advocate here, rather than simply siding with O’Brien (as I clearly disagreed with him on a few points, and think he should have better formulated his “assumptions” at the very least).

        “It seems to me that our interaction with the world is mainly pragmatic, rather than rule-based. When we follow rules, these are usually rules that we have adopted for pragmatic reasons.”

        I’m not sure what you mean when you say that our interaction with the world is mainly pragmatic rather than rule-based. I would say that we constantly use rules for pragmatic reasons, but that our interaction is still rule-based. It may be that some of these rules are simple heuristics and many of these rules aren’t consciously known to be operating (even though they are), but it seems most logical to assume that our brain is ultimately performing multiple levels of pattern recognition in which each pattern recognition module is a parameter correlating with the outside world, and the hierarchical organization or dynamic/plastic connectivity between the pattern recognition modules seem to represent the natural laws/rules of the universe available to us via our senses and perception. It may be that some of the rules we use have different weights to them, whereby sometimes certain rules will override others, but the rules appear to be there operating nevertheless. If you don’t think so, please give an example of people interacting with the world without the use of rules. Random, thoughtless interaction is certainly possible without rules, but I doubt you are proposing that random, thoughtless interaction has any pragmatic use.

        “The simulation argument or, for that matter, any computationalist account will provide a system which is rule-based, but where the rules are externally imposed. I don’t see it as fitting what we actually see.”

        I don’t see it as fundamentally different than our brains, which have “rules” ingrained into them via the DNA sequence that codes for a particular structural development within that brain, leading to our pattern recognition capability (which likely rewires over time as it learns and experiences based on neurological data), to consciousness, to learning, etc. I see this as fitting what we see fairly well and I’m not sure how you are arriving at your point of view.

        Like

        • I’m not sure what you mean when you say that our interaction with the world is mainly pragmatic rather than rule-based. I would say that we constantly use rules for pragmatic reasons, but that our interaction is still rule-based.

          We invent rules for pragmatic purposes. They are our rules, not nature’s rules. When we follow rules, we are following our own pragmatic rules rather than externally imposed rules.

          Like

          • “We invent rules for pragmatic purposes.”

            I’m not arguing that we don’t invent rules for pragmatic purposes.

            “They are our rules, not nature’s rules.”

            Ah, so you admit that there are rules of nature. (hehehe…I had to grab that low hanging fruit).
            However, those rules are going to be correlated with reality as some form of an approximation (the pragmatic element), and thus they will have some correlation to nature’s rules.

            “When we follow rules, we are following our own pragmatic rules rather than externally imposed rules.”

            Unless our pragmatic rules are in fact based on externally imposed rules (the laws of nature), which I think they are. That is, when we interact with the environment, we are using rules which we’ve gathered from experience, the experience of nature as governed by its rules. The laws of nature are definitely externally imposed rules, if you see us as living and interacting as a part of a much larger system. We have no way to change these externally imposed rules. All we can do is learn from them, learn what they are (a best approximation anyway), and we may derive subsets of rules for pragmatic purposes, but they seem to be derived from the externally imposed rules of nature nevertheless. If we ever use rules to solve a problem or to make a prediction, the more repeatable such an action is, the more obvious it is that our rules are in fact a subset of nature’s rules.

            Like

          • However, those rules are going to be correlated with reality as some form of an approximation (the pragmatic element), and thus they will have some correlation to nature’s rules.

            There is no meaning of “correlation” that would allow us to assert that.

            Like

          • “There is no meaning of “correlation” that would allow us to assert that.”

            Certainly there is. The correlation to reality is implied by predictive capability. If we can predict how reality operates successfully, then our predictive laws have a correlation to that reality. Cut and dried. That is it.

            Like

          • If we can predict how reality operates successfully, then our predictive laws have a correlation to that reality.

            How about you provide the mathematical proof?

            Hmm, I wonder how your “argument” would have impressed George Berkeley, given that he did not believe that there was any reality.

            Like

          • “How about you provide the mathematical proof?”

            Proof for what? Just look at the rate of error between our predictions of particular types of events, and actual measured instances of those events. That is the proof.

            “Hmm, I wonder how your “argument” would have impressed George Berkeley, given that he did not believe that there was any reality.”

            Yes, certainly we are defining “that which we experience” as “reality”. That is a reasonable starting point for the purposes of this discussion that I think you agree is reasonable. Unless you are advocating Solipsism, or some type of idealism, I think we can proceed just fine here. This is irrelevant based on the assumptions we bring into these types of arguments (such as reality existing). We assume reality exists if we are going to talk about it. Whether or not reality exists is a topic for a separate discussion.

            Like

          • I’m just glad there’s no metaphysics going on here. 🙂

            Liked by 1 person

          • Proof for what? Just look at the rate of error between our predictions of particular types of events, and actual measured instances of those events. That is the proof.

            Correlation is a precisely defined mathematical quantity. If you can’t show the math, you’ve got nothing.

            Like

          • “Correlation is a precisely defined mathematical quantity. If you can’t show the math, you’ve got nothing.”

            I just explained correlation earlier, but perhaps you didn’t read it. What I said was that you could measure an attribute of some phenomena (say, the velocity of an object dropped from the roof of a building or the time it will hit the ground), and then calculate what the velocity would be according to Newton’s laws (kinematic equations) for example. The smaller the error between what is actually measured and what the equation predicts is a higher correlation to reality (if not then why not?). If one were to use relativistic equations, we may expect to find a slightly different answer, even better correlated with the measured phenomena. If we call an error of 0% a perfect correlation, and an error of 100% a zero correlation, then we can calculate anything in between. If my error is 50%, then I could say I have a correlation of 50%. If you are looking for something other than this, then you clearly mistake what I mean by correlation (I’m using the term in a general way at the very least, so there should be a “link” between the laws of physics and what we actually measure, i.e., reality).

            Like

          • What I said was that you could measure an attribute of some phenomena (say, the velocity of an object dropped from the roof of a building or the time it will hit the ground), and then calculate what the velocity would be according to Newton’s laws (kinematic equations) for example. The smaller the error between what is actually measured and what the equation predicts is a higher correlation to reality (if not then why not?).

            Your argument seems to be of this kind:

            It sounds good to me. I can’t think of any alternative. So it just must be true.

            That’s the way that creationists argue. And that’s the way that you are arguing.

            Yes, you can make a strong case that velocity is correlated to velocity. You can even make a strong case that measured velocity it correlated to computed velocity. But you have not explained how that is related to “Newton’s laws are correlated to reality.”

            Like

          • “Your argument seems to be of this kind:

            It sounds good to me. I can’t think of any alternative. So it just must be true.

            That’s the way that creationists argue. And that’s the way that you are arguing.”

            How am I arguing that way exactly? What have I said specifically that demonstrates this? On the contrary, you are simply asserting your position with no evidence, which is rather dogmatic and much more like a Creationist. I’ve presented evidence. The evidence is the predictive power of physical laws, which means they correlate with how the world works, which means they correlate with the laws of nature. If what you’re looking for is an actual example of physical laws providing predictive power, then you are looking for something that is quite obviously all around us and in every physics experiment. A scientist makes predictions based on known laws, and tests the efficacy of those laws through experimentation. There will be some correlation (i.e. link) between the predictions and the actual results. What more evidence do you need, unless you are simply squabbling over semantics?

            Like

          • How am I arguing that way exactly?

            You are presenting a pseudo-argument. You claim an implication, but won’t show the logic. You won’t even explain what it means.

            When pressed for details, you repeat the same pseudo-argument.

            The evidence is the predictive power of physical laws, which means they correlate with how the world works, which means they correlate with the laws of nature.

            This is just bullshit meaningless nonsense. “Correlate with the laws of nature” mignt have a nice warm fuzzy feel to it, but it doesn’t mean anything until you actually produce the alleged laws of nature and do an actual computation to determine the correlation.

            A scientist makes predictions based on known laws, and tests the efficacy of those laws through experimentation.

            Sure. But what does that have to do with “correlation with reality”. Surely it only shows that the known laws are efficacious in predicting.

            Like

          • See my previous response as I’ve already addressed this.

            Like

          • “Yes, you can make a strong case that velocity is correlated to velocity. You can even make a strong case that measured velocity it correlated to computed velocity.”

            No, actually I made a strong case that “predicted velocity” is correlated to “measured velocity”.

            Like

  3. “I don’t assume naturalism, because I have never been clear on what such an assumption entails.”

    A basic definition of Naturalism is simply the idea that only the laws of nature (physical laws — even if we can’t know what they all are) operate in the world. The main reason that O’Brien mentions this is to simply exclude the idea of souls and dualism. Basically, he is assuming materialism. Since you don’t believe you have a supernatural soul, for the purposes of this discussion, you agree with O’Brien’s first assumption.

    “Although O’Brien calls that an empirical assumption, I take it to actually be a metaphysical assumption.”

    I agree. O’Brien seems to have made a mistake there. Both of his assumptions appear to be metaphysical (naturalism & physical law computability).

    “That there are such laws would be another metaphysical assumption, and one of which I am quite skeptical. In particular, QM would seem to pose a problem for those who believe that there are such laws.”

    I agree in the sense that the randomness we find in QM (even if ontologically deterministic) implies that our ability to compute physical laws is limited, however if we can compute the laws of physics within a particular range of accuracy, then for all practical purposes, the uncertainty/randomness stemming from QM may not matter for the simulation. The laws that we do know (including the Shrodinger equations for a particular system), give an incredibly accurate approximation, and may be all that is necessary for reproducing the non-local or collective events happening in the brain during consciousness. While we won’t be able to predict what exactly will happen with a simulated entity down to every atomic detail, the probability distribution of possible outcomes on collective brain processes should allow us to produce something reasonably probable and therefore correlated with reality (and for all practical purposes of a simulation). I don’t argue that this can currently be done in practice, just in principle.

    “If the role of consciousness is to allow us to learn, then preprogramming with innate knowledge would seem to undermine that role.”

    The purpose of the “preprogramming of innate knowledge” is such that the simulation has the same starting point as a conscious grown up human being. So the fact that some pre-programming of innate knowledge would seem to undermine the role of consciousness is simply because the simulated human didn’t develop as it normally would have in real life. On the other hand, if the simulation were to start from a simulated conception, presumably the ontogenic evolution of the simulated embryo into an adult (with experiences also simulated), would theoretically produce consciousness sometime during the simulated development.

    “For the programming would result is a system that had no need to learn, and thus no need for consciousness.”

    As I just mentioned, it seems that the purpose of the preprogramming, or perhaps what you are defining as pre-programming, is simply the information and states related to producing a simulation of an already grown person with memories and knowledge. If it is expected to continue to learn consciously within the simulation, then it would still need consciousness. If the simulation was only meant to produce an analogue to a conscious human, then consciousness is all that is needed, and the need to learn is negated (as you said). However the simulation isn’t supposed to produce “needed consciousness” and if it is supposed to produce consciousness and learning through traditional biological mechanisms, then the simulation would merely have to encompass that greater goal. I think O’Brien was just exercising the thought experiment of trying to reproduce any person reading the article with all their memories intact, which isn’t required to consider natural experiential acquisition processes (for the memories anyway) nor any natural selective pressures (for the simulated consciousness itself).

    “So, if my impression is correct, we would have to conclude that Descartes’ intuition did not support the kind of viewpoint that O’Brien is arguing for.”

    I think Descartes’ intuition (as well as that of most humans) is based on the fact that consciousness doesn’t allow us to experience the clockwork processes of the brain (physical causal processes that generate thoughts for example), such as processes that are in the unconscious, etc. When a thought enters the stream of consciousness seemingly ex nihilo, we are illustrating that either thoughts come from nothing, or that what they come from is outside of conscious experience (brain processes that are unconscious). So Descartes’ intuition is easily explained. And clearly his ignorance of psychology, neurology, and neuroscience which hadn’t been developed yet, led him to postulate dualism. Now in retrospect, we can see Descartes’ error, and thus the fact that it doesn’t support O’Brien’s viewpoint is perhaps a bonus point for O’Brien.

    “For myself, I do not believe that we live an anything like a clockwork universe. So I have no need to appeal to dualism. ”

    Whether the universe is like a clock or not, I have no need to appeal to dualism in either case. Descartes thought it was necessary in a clock-like universe, but that seems to be mainly because of his lack of knowledge about how the brain works (which is perfectly understandable for one living in his time).

    “I see the universe as such that consciousness could evolve as a biological adaptation (and probably did).”

    I agree.

    “In short, I do not find the simulation argument at all persuasive.”

    What would make it more persuasive for you? In order to be taken as valid, I think it requires many more assumptions which he didn’t list along with his original two. He seems to assume that consciousness is a functionalist (as opposed to type-identity limited) phenomenon. That is, O’Brien assumes that consciousness is an abstract property not dependent on particular physical constituents. At the very least, he assumes that the complexity, patterns, and organization of information are more important than the physical biological constituents of the brain that contain it. I’m definitely open to the same idea, that is, that consciousness isn’t limited to biological hardware, but rather the pattern of information and processing has only been accomplished through biological hardware thus far.

    The question comes down to whether or not we think there is something unique and special about biological materials with regard to consciousness, or rather that those particular biological materials were an inevitable precedent for consciousness due to the evolution of DNA and a certain level of complexity that it provided through presumably non-teleological means. With conscious foresight, just as we can produce properties and functions of particular objects (e.g. tele/micro-scopes and photo-detectors) with non-biological hardware that has also evolved slowly through biological processes (e.g. eyes and retinal receptors), it is likely that many other types of properties that evolved biologically (such as consciousness) could in principle be produced and accomplished through non-biological means.

    It is true that in the case of optics I mentioned (whether silicon-based photo-detectors or organic retinal receptors), that in both cases physical materials are used, but it is clear that the physical materials needn’t be the same. The final question that needs to be answered is if consciousness is the same as these types of properties (optics/light detection). O’Brien argued that consciousness was more of an abstract property rather than a physical property, so it may be the case that with consciousness, not only could multiple materials be used (like with bio- or non-bio-optics), but the levels of organization and complexity that may lead to consciousness may be accomplished in multiple formats and multiple types of arrangements.

    Like

  4. On the simulation-argument.com FAQ page Nick Bostrom writes . . . :

    “It seems likely that the hypothetical simulators, who would evidently have to be technologically extremely advanced to create simulations with conscious participants, would also have the ability to prevent these simulated creatures from noticing anomalies in the simulation.”

    ‘IF’ anyone could THINK, then what would be the best way of making sure your simulated population didn’t notice anomalies?

    In a simulated reality attempting to present entirely SOFTWARE DEFINED self aware, free thinking people then it is very obvious that the designers of such a simulation would prevent their simulated creatures from noticing anomalies by DIRECTLY managing the AWARENESS, THINKING &, EVALUATING ABILITIES of their simulated population AND they would particularly use this very cheap to implement strategy to AGGRESSIVELY manage their simulated academics and scientists.

    Evidence of this being implemented would be that ON ONE WAS ABLE TO THINK OF THIS SERIOUSLY OBVIOUS STRATEGY in the first place.

    On this page here:http://www.soul-healer.com/new-age-origins/how-would-a-simulation-designer-hide-evidence-of-ourselves-being-in-a-simulation/ are 18 points listing visible evidence that would way more than suggest that we are in a simulation.

    Like

%d bloggers like this: