God vs The Multiverse

Click here for God vs The Multiverse: a rational argument for the Existence of One God who intelligently designed one universe.

Wednesday, July 18, 2012

God vs The Multiverse (Part 11: String Theory)

It will be very helpful for understanding of this post to watch the video on post seven, between the 34-41 minute marks.

In the past three posts, we have discussed serious problems with multiverse theory. We will now turn to refuting the "proofs" which are proposed in support of the multiverse.  We believe there is a deep relationship between the past three posts, and these next few posts.

In the Greene video/article, he cited three pillars of support for multiverse theories: (1) the apparently fine-tuned value for the dark energy (the cosmological constant); (2) the 10500 different possible universes predicted by string theory; (3) observations supporting inflation which mathematical analysis revealed to be eternal inflation.

It should be clear from the previous posts that the fine-tuned value of dark energy (as well as the other constants) is in no way a proof for the multiverse.  Rather, it is a proof for an Intelligent Designer, which forces multiverse theorists to desperately posit as many chaotic universes as they need to get out of the problem.  In this post, we will explain (and thereby undermine) the proof from string theory. After that, we will discuss the proof from inflation.

String theory is currently the most popular theory of fundamental physics, and many physicists believe it to be the best candidate for a Theory of Everything.  String theory is widely acknowledged to contain truly beautiful and elegant mathematics, which is one of its main appeals.

String theory attempts to unify all of physics (General Relativity and Quantum Mechanics) by assuming that all particles are really tiny, vibrating strings.  To get an idea how tiny they are, scroll down all the orders of magnitude on the scale of the universe interactive site.  There is a very large gap between the tiny strings and anything else we actually have evidence for.  In fact, it is far and away the biggest gap in orders of magnitude between any two things.  That is how tiny strings are.  They are the smallest fundamental objects.

(This exercise with the scale of the universe, as well as this video from 1977, will help give you a good intuitive sense for orders of magnitude and complexity, which is a basic aspect of our understanding of the universe.  It is not entirely clear if string theorists believe that there isn't anything in between these tiny strings and the known 'fundamental particles' (i.e., electron, quark, neutrino); or they just think that it isn't necessary to go step by step in discovery, and they've been able to guess the ultimate, most fundamental reality.)

In order for string theory to be mathematical consistent, it needs to posit at least 7 extra dimensions of space.  M-Theory (according to it's creator Edward Witten, the 'M' could mean Magic-Theory) is the most prevalent form of string theory and requires 8 extra dimensions.  One problem is that no one has ever observed these extra dimensions.  It is therefore assumed that these extra dimensions are curled up so small that we can't observe them.  (We would like to point out that there is another logical explanation for why we cannot observe them.)

It turns out that the geometry of these hidden dimensions is very important.  Depending on how you curl them up, the theory results in very different properties of particles and values of the constants.  These different geometries are called the string theory landscapes.  There are currently known to be so many possible different ways (10500) that really small, unobservable dimensions can be curled up and hidden away, that string theory can explain just about any universe with any set of constants.

That is not a good feature of a scientific theory trying to explain one particular universe.  It order to be tested, a theory needs to make credible, unique predictions.  Since we don't know the shape of these really tiny, never before seen dimensions, any phenomenon at all can be explained by the theory.

At first, this was a major cause for chagrin amongst string theorists (see video), and the theory came under heavy criticism.  It seemed like string theory was beautiful mathematics and interesting speculation, but it had a fatal flaw.  Until physicists could show some compelling reason to select one particular landscape out of the 10500 different possibilities, string theory is just beautiful speculation with no hopes of any empirical tests to support it.

In 2005, Leonard Susskind came out with his ground breaking book (The Cosmic Landscape: String Theory and the Illusion of Intelligent Design), in which he 'arranged a marriage' between string theory and multiverse theory.  Susskind recognized that if the multiverse theory were true, then having many different possible geometries might not be so bad.  It might even be a good thing.  Maybe every multiverse has a different geometry of these hidden dimensions and therefore has different values for the constants of nature.  It even explained the fine tuning of the constants!  And the origin of life to boot! A Theory Of Anything!

If everything is really tiny strings, and if there are 7 too tiny to be observed hidden dimensions, then maybe the 10500 different landscapes really exist, if there are at least 10500 different multiverses. (None of these things have ever been observed or experimentally supported).  It should be obvious to the reader that this does not constitute evidence for the multiverse.  Rather, the landscape problem of string theory is a fundamental flaw in the theory.  If it can't be solved, it prevents string theory from being a credible scientific theory.  A fatal flaw in a theory is not 'proof' for the multiverse.

Ultimately, what their proof amounts to is faith that string theory is true.  Since string theory is assumed to be true, and if there were not 10500 multiverses then string theory would have a fatal flaw, it must be that there is a multiverse. The belief that string theory is a true valid scientific theory necessitates the belief in the multiverse.  Somehow, they extrapolate that "String theory proves the multiverse!"

We're including a video by Lee Smolin (who had his own fecund multiverse theory), author of a great (and controversial) book called The Trouble With Physics.  Smolin discusses the scientific merits of string theory, as well as how scientific consensus should only form after a theory makes predictions that are confirmed by experiments and observations (which is something that string theory has failed to do).


48 comments:

  1. Wikipedia said that Smolin's theory was disproven by a test that he proposed about the mass of a neutron star. Does Smolin still believe in his multiverse theory now?

    ReplyDelete
    Replies
    1. We were wondering that ourselves. You would hope not, or it would cast serious doubts on the ability of his theory to make actual falsifiable predictions.

      Delete
  2. > Rather, it is a proof for an Intelligent Designer, which forces multiverse theorists to desperately posit as many chaotic universes as they need to get out of the problem

    Honestly, this growing pile of ad-hominem, and claim of scientists' desperation is making be pretty mad at this point. I'm literally biting my tongue and considering parachuting out of this discussion, perhaps that's what you want - an opportunity to preach to the converted.

    Let me state this pretty clearly one last time: you have no serious evidence that scientists are desperate to deal with god issues. This might have been a serious consideration for Darwin 150 years ago, I do not find it at all plausible in case of modern scientists. The video you posted strongly indicated that modern multiverse theories started with mathematical considerations (your little narrative of changes in theory in 2005 is very weak). This is the kind of crap you hear from the lowest bottom of the dumb kiruvnik crowd, relying on a naive audience who can't judge for themselves.

    If you can't make your argument without constantly claiming that the other side is biased, look in the mirror.

    Dr_Manhattan

    ReplyDelete
  3. Dr. Manhattan,

    Here is someone who is not a kiruvnik who seems to think that scientists are indeed troubled by things in our universe that appear to be designed (specifically second paragraph).

    http://www.nytimes.com/2006/01/15/books/review/15powell.html

    I'm not speaking to the argument of what the motivation of multiverse theory is; my only point is that this type of rhetoric is not limited to religious folk.

    (If you're curious who this author is, see the first 2 minutes of this video http://www.youtube.com/watch?v=9Cz8ZwaJe24)

    ReplyDelete
  4. > I'm not speaking to the argument of what the motivation of multiverse theory is; my only point is that this type of rhetoric is not limited to religious folk.

    David, 1) one data point 2) from a science editor and poorly-rated author 3) not a practicing scientist (has not even gone to a graduate school apparently) 4) about one scientist's specific version of multiverse

    In any case, my point is without serious proof the authors should stop throwing around "desperate scientists". If anything it lowers the authors' credibility, and my expectation of them saying anything interesting. And while there may be "non-kiruvniks" (or perhaps christian kiruvnik equivalents?) who make similar claim it is **typical** of that pseudo-intellectual sector, and people should aspire to be better than that.

    Dr_Manhattan

    ReplyDelete
    Replies
    1. Dr. Manhattan,

      Fair enough. Here's something with a little more "angst". It's a discussion between a group of scientists/mathematicians, including Carr and Ellis, about the topic at hand.

      http://www.st-edmunds.cam.ac.uk/faraday/CIS/Ellis/Ellis_Discussion.pdf

      It's worth reading the whole thing, but here's a few of quotes in case time proves limiting:

      Ellis:

      "Well, firstly, Susskind was absolutely explicit in his book that he was producing a multiverse as an alternative to theism."

      "...Martin Rees writes a lot about this issue precisely because Martin sees so clearly the problem of anthropic fine tuning. Martin has thought about it and really sees the problem of anthropic fine tuning which is why he’s bought into the multiverse."

      Carr:

      "I’d like to make a comment here because I’m organising a meeting entitled “God or Multiverse?” in a few weeks time. In 1979 I wrote a paper with Martin Rees on the anthropic fine-tunings. At the time it was regarded as quite a fringe idea and a lot of physicists
      were very opposed to anthropic arguments. The notion of the multiverse has led to a shift of opinion because some physicists see the multiverse as a legitimisation of the anthropic principle. As a result of my paper with Martin, I was quite often invited to theological or science-and-religion meetings, and I found myself in a sort of schizophrenic state, where I was saying to the theologians “Look at these coincidences - they could be seen as evidence of a creator” and to the physicists “Well you can explain them if you have an ensemble of universes” for even before the multiverse became a popular idea through string theory, we
      were talking about the possibility of many universes as an alternative to a creator. From a sociology-of-science perspective, this was important because I think part of the antipathy to
      the anthropic principle from physicists arose because they felt it smelt of theology. They were afraid that the only way to explain it would be was to invoke a fine-tuner.
      Now there is a famous quote from Neil Manson that “the multiverse is the last refuge for the desperate atheist” and I partly agree with this. For if there really is only one universe, you’ve got a problem trying to explain these fine-tunings and might well be forced into a theological direction. But if there is a multiverse, you’re not compelled to invoke God..."

      Delete
    2. Those are pretty explicit quotes. We appreciate the help.

      Incredible pdf.

      Delete
    3. David,

      Firstly, great article. Very explicit. Secondly, At this point, Dr_Manhattan is reiterating a point that was not only explained explicitly in the blog posts, but raised in the discussions by him and responded to. He doesn't need to be swayed by the arguments but the basic principles and evidence have been established and the reasoning behind the claim of desperation has been shown so that any readers can come to their own conclusions.

      Whether he likes the reasoning or not is irrelevant, it has been discussed enough to allow other readers the chance to decide. To lodge the exact same attack without new evidence is unfounded and "trolling". As such, I don't think a response is warranted.

      Dr_Manhattan,

      Your ad hominem attack on me and the rest of the readers of this blog (or "the converted") is unwarranted, unfounded, and irrelevant. The authors of this blog happen to find the conclusions of multiverse theorists desperate. I tend to agree with them.

      Their opinion has been stated and explained. And you have labeled us kiruvniks (without explanation, I should add). The point was raised previously and your objection has been noted.

      If you wish to leave the blog, I would personally miss the opportunity to elucidate its points and disprove it when possible, and you would be missed. I would also not miss your seemingly descending etiquette and substance.

      I appreciate very much a reasoned attack against this blog and hope that you will continue in such a vein. I have, though, noticed a shocking decline in your relevance and reasoning. If this conduct should end, I would not miss it at all. If however you feel that your more recent posts have been just as well reasoned, polite, relevant, and non-repeating as your original posts, I urge you to re-examine your posts and yourself or in the alternative cease.

      I hope we can see a return to the Dr_Manhattan of old, but the choice, as always, is yours.

      Delete
    4. Thanks David, interesting article from actual credible scientists.

      Let me mention a few things though.

      Even if Susskind was motivated as stated, that's 1 data point. But I'm not sure of even that much; he might well have been doing unbiased scientific work and then felt he had to combat the "god of fine-tuning" argument in his book. In similar vein Dawkins did some excellent work in biology before deciding religion is evil and deciding to use his scientific arsenal against it.

      It also happens to be the whole discussion was around theistic topics, and Ellis, Carr and Polkinghorn are very recognizable elements if the "last remaining famous theistic scientists". Essentially 1/2 the people who get the Templeton prize get it for this reason. (It also happens to be Carr is a "psychic researcher", which makes my alarm bells go off). I'm not expecting unbiased opinions in this select group, and a little mud-slinging at Susskind is almost predictable. As much as I like Dawkings and Harris they are somewhat extreme and I would not consider them unbiased opinion-sources; I would look at their ideas, not through their glasses.

      So in the end - maybe Susskind was biased? Who knows. So far this is the best evidence I've seen, and I don't thing the RAF/REZ has done as much research as you before declaring "desperate scientists"

      Dr_Manhattan

      Delete
    5. It appears that no amount of evidence will convince Dr_Manhattan of the very credible arguments that have been raised by the writers of this blog. For every attack that he has leveled, he has shown to be wrong about his assumptions. His most recent attack, that the scientists are not worried about God because of fine tuning, has been demonstrated clearly wrong by the excellent article found by David Fischbein. Now, even those scientists are biased? So, I get the point. If a scientist denies God he's not biased, but if he suggests that God exists, he is????

      As Michael has stated above, it would be preferred if the comments here were limited to specific rational questions on the ideas the authors are posing without the silly, "let me see if the scientists agree with you" type of thinking.

      Delete
  5. A little tired of the multiverseJuly 22, 2012 at 12:14 PM

    Any chance we could get some of the old style halachik series of posts (maybe in addition to the multiverse)?

    ReplyDelete
    Replies
    1. We also miss the halachik posts and look forward to getting back to them.

      However, our time and energies are currently in "God vs. the Multiverse", an involvement which we deem to be more fundamental and important than those posts.

      The Rambam (perush ha'mishna, end of Brachos) says that the most important thing for him to teach was to explain fundamentals of the Torah. He therefore put in discussions of these fundamentals whenever he encountered them.

      It is in this spirit that we decided to interrupt the normal Blogoshiur posts to take up this most fundamental topic.

      Delete
  6. Her is a quote from the lecture slides from Carr (the lecture that the .pdf discussion that David led us too is associated with):

    "Is the degree of faith required to believe in a multiverse more or less than that required to believe in a creator God?

    I argue that because of the lack of conclusive evidence in both cases, the degree of faith required to believe in either is the same. Both can be argued on the basis of reasonable extrapolation from known data. Neither is in fact provable. Despite scientific appearances, belief in a multiverse is an exercise in faith."

    The same "belief" that is required to posit an intelligent designer is also required to posit the multiverse, as both are scientifically untestable.

    ReplyDelete
  7. > I argue that because of the lack of conclusive evidence in both cases, the degree of faith required to believe in either is the same. Both can be argued on the basis of reasonable extrapolation from known data. Neither is in fact provable. Despite scientific appearances, belief in a multiverse is an exercise in faith."

    > The same "belief" that is required to posit an intelligent designer is also required to posit the multiverse, as both are scientifically untestable.

    This is not the same as what the authors are claiming, btw. They clearly state that "any reasonable person would be convinced" by their argument (which I am not sure I have seen yet - there are so many posts). But of course it's wonderfully convenient to use these scientists' credibility to undermine Susskind, but pretty inconvenient to follow these esteemed persons (even though they're sympathetic to deism Templeton winners) to the conclusion that multiverse is a valid possibility.

    All in all it's not a "logical contradiction", I'm just pointing out the methodology of deferring to scientists to get us 90% of the way to this "proof" and then implicitly declaring that all scientists are "not reasonable". Or maybe they just haven't seen this new proof? I suggest it be submitted to them, even these 3, for critical review.

    Dr_Manhattan

    ReplyDelete
    Replies
    1. The reason we accept scientists 90% of the way, and them reject them at the end is that the first 90% is real, tested, science. The last 10% where they go off into speculative philosophy is where we reject them, based upon the numerous problems raised in these past few posts.

      The following is an excerpt from a future post (there is a limit to how much you can put in each post):

      The value of scientific consensus as an authoritative position has applicability when it is about science. This requires that almost all scientists claim that a theory has made credible predictions confirmed through observation. In that situation, it makes no sense to doubt their authority, as it is something that can be revealed to be a lie. For a layperson to doubt scientific consensus that has developed after experimental confirmation, is bordering on conspiracy theory. However, we hope it is clear by now that multiverse theory does not come under this category at all.

      Delete
    2. On the contrary, I believe that the authors are making this point exactly. Because the multiverse theory is speculative and untestable, it cannot be considered science. The fact that credible scientists make recourse to it implies that they are willing to abandon the scientific method. What makes them do so? From the majority of sources that I have read, it is because it is anathema to most of them to accept the idea of an intelligent designer/creator.

      What is being noted is that at this point in time, our scientific knowledge has reached a stage in which it is now plausible/reasonable to accept the idea of an intelligent designer. This is an amazing concept in of itself!

      We all realize that this is far from over. Perhaps at some point in the future, our knowledge will advance and new concepts will emerge. That has happened throughout the ages. But at this point in time, scientific knowledge certainly points us in the direction of an intelligent designer of the universe.

      Delete
  8. How many more posts are planned? I sort of want to tune out until then, until a coherent argument is fully laid out.

    Dr_Manhattan

    ReplyDelete
    Replies
    1. If you already have a proper idea of One God, then we think that the essence of the proof is already laid out. The only major point left is showing why eternal inflation is no support for a multiverse. So unless you're hanging your hat on eternal inflation, you should already be in a fairly good position to choose freely what you believe to be true.

      What about the argument do you think is not coherent?

      Delete
    2. I' afraid that your argument will never satisfy our good 'ol Dr_Manhattan. He is looking for a scientific proof of God's existence -- such as those that are verifiable by experiments. This is, of course, impossible. If the proof of God's existence must be inferred from the information presented, he will remain unmoved. There will always be some alternative explanation for those unwilling to accept the idea of a designer -- even if the explanation is not scientifically valid. One can always claim that "perhaps in the future this will somehow be verified" This, then, gives the outward appearance of a scientific endeavor. But, as you have demonstrated well, such a hypothesis is outside the realm of science.

      Delete
    3. > If you already have a proper idea of One God, then we think that the essence of the proof is already laid out.

      What is the idea you're referring to? I'm somewhat familiar with Maimonidean ideas, which you seem to be inclined to. Not clear how this plays in your proof, since it wasn't yet mentioned (or I missed it).

      > So unless you're hanging your hat on eternal inflation

      Not really hanging my hat on anything. I have some objections to what I currently see steps in this proof, but I have limited time and don't want to be in the situation of "oh, we explain it in the next post" or stuff like this. I just want to see the whole thing laid out, then we can duke it out as they say :)

      Jeffrey Beer: no, I do not only care about experimental proof. If anything, the opposite, in a sense that I see a continuum between philosophy and what I see as properly methodological science (based on Bayesian probability).

      Dr_Manhattan

      Delete
    4. Dr_Manhattan I am still confused as to how you use Bayesian analysis for generating accurate probabilities. You used an earlier example of the probability for ghosts (I believe you set it a 1 in 100000) How did you generate this probability? why not a different number? If I offered you 1$ for a 99999$ bet you would take the other side than for a 100001$ bet? Similarly how did you arrive at 80% for the multiverse (why not 83% or 76%) I have read some of the material you pointed to about Bayesian analysis, but I still don't see how you arrive at numbers which can then be plugged into equations (other than a general qualitative sense of the numbers (1 in a million is effectively impossible, 2% is low but plausible, 80% is likely but still open 99.9% is almost certain etc) and the rule that probabilities must add up to 1)? Similarly how do you determine how much to change the probabilities based on each new piece of evidence? If it is only a qualitative sense, how does stating it numerically help?

      (I realize that this discussion started on an earlier post, but I wasn't clear as to how you generated the probabilities and didn't want this question to be lost in the myriad of posts and comments)

      Delete
    5. Yaakov - what material have you read?

      As far as arriving at my personal probabilities... First I wouldn't call them "accurate" because that gives the feeling of great precision, which is not there for many of them, especially things in the etreme-tiny-probability bucket such as ghosts. Second, arriving at them is a process of trying to weigh existing evidence, problems with evidence, self-calibration as in "what things have I believed nearly that impossible, that proved possible". As a small example I currently suspect that Chinese medicine gets some things correct, despite being very similar to witchcraft in thinking process (using entities I do not believe exist). I used to think "no freaking way" but many reports from seemingly sane people seem to show the opposite. So I calibrate and lower my confidence in completely rejecting things that sound crazy (I still think things like Chinese medicine or the potential "ghosts" have underlying reductionist scientific explanation, even if we do not know it). But honestly with this kind of variance 1 in 1,000,000 or i in 100,000,000 is almos the same to me.

      As far as the multiverse, first of all I am not in physics, so my opinion is that of an observer, 2d hand at best. But multiverse seems like a decent theory, and it's not particularly jarring since it seems we're already in a level 3 multiverse within what we consider our universe (as the best explanation for QM - Copenhagen interpretation implies some really weird causality from observer to phenomena, which I am not buying. It also anwers Einstein's dice). I do not see that the theory's existence was primarily determined by just a desire to escape design; even people in the industry who are theistically inclined would not claim it in all instances, since they admit multiverse as a possibility, if not a greater one than design.

      I could recommend some general reading in this area, but like I said not sure what you have read already.

      Dr_Manhattan

      Delete
    6. Thank you,
      I have read a variety of things, including your links from post 4, and am currently in the middle of "probability theory, the logic of science" by E.T. Jaynes who seems to have a similar view to yours.
      I see the benefit of using probability in certain areas (for example decision making and medical research) However I don't understand its use in fundamental science and philosophy.
      First, since there is no real mental difference between 1 in 1,000,000 or i in 100,000,000 why not just leave it at the qualitative "very unlikely" which would need extensive evidence of an effect plus removing the plausibility of deception. What do we gain from reformulating it quantitatively?

      Second, I don't see how appropriate prior probabilities can be developed. If there is no rational systematic way of choosing the prior probabilities than how can this system be used?

      Back to the example of the post, using Bayesian statistics, in order to decide the posterior probability for the god of Deism, based on the evidence of fine tuning you would need a prior probability for that idea. How do you decide what that prior probability is?

      Lastly, it seems that much of science actually works through intuition, which by definition means the person is unable to identify all of the factors (or evidence) which cause them to pursue a certain theory, therefore it cannot be reframed into a probability. Both when the evidence is incomplete and in answering difficult questions and coming up with new ideas.
      This 'problem' is specifically true when it comes to philosophy. Hence you could have Einstein and Bohr extensively argue, since the underlying evidence for each world view was incompletely identified.

      Delete
    7. In terms of the level 3 multiverse
      a. Everett's underlying idea does not necessarily imply multiple universes (it only implies the non-collapse of the wave function, which some, including Tegmark, understand as multiple universes (see e.g. http://plato.stanford.edu/entries/qm-everett/)).
      b. Even if one interprets the existence of a non-collapsed wave function as implying multiple worlds (as opposed to one universe/ wavefunction, of which the part we observe is relative to the observations/interactions we make/are involved in) It does not allow for different constants since all of the 'worlds' are part of quantum mechanics, whereas variation of constants would be logical prior to any specific wavefunction.

      Delete
    8. Yaakov, you make good points about how probabilistic calculations work. Here are some incomplete answers
      > First, since there is no real mental difference between 1 in 1,000,000 or i in 100,000,000 why not just leave it at the qualitative "very unlikely"

      I might have, and it would reflect my mind state fairly well, but there are rhetorical reasons not to, as people take "very unlikely" and eliminate it from the set of possibilities altogether. Also it's a reminder that thought process is, however crudely (due to limitations of the instrument), stitched together with math. Also, there are situations you may be forced to call the odds on something unlikely and make serious decisions (e.g. invest in asteroid surveillance). Finally, common way to communicate statistical information is with mean and variance; so this is sort of my "mean" estimate, with a couple of orders of magnitude being variance.

      > Lastly, it seems that much of science actually works through intuition, which by definition means the person is unable to identify all of the factors (or evidence) which cause them to pursue a certain theory, therefore it cannot be reframed into a probability

      Intuition if crucial in generating a hypothesis.

      As far as evaluating a hypothesis, it is true that intuition is used at some stages, but this seems to be a limitation of the instrument. Let's say you have a fairly intelligent factory robot that's gotten a bit rusty and slightly miscalibrated. It still gets a lot of things right, on average, (most widgets come out ok). It can even do some self-repair and replace some of its older parts, but it can't get into the CPU and diagnose it. The robot knows though that the CPU by and large follows boolean logic, despite not being able to see it and sometimes getting wrong results from it (that do not follow boolean logic).

      In a similar fashion I think human scientists approximate Bayesian probability, and update on evidence, though only approximately. It seems clear to me that "scientific method" is an approximation of Bayesian truth-finding.

      So, prescriptively, we should be using formal probability, as much as possible, as long as premature attempts to do it do not interfere with early creativity in hypothesis generation.

      There are some formal frameworks for coming up with priors, like Solomonoff's Universal Prior. I'm not sure it's quite right, but it's very objective, and probably a huge step in the right direction. I plan on using it when discussion priors for design hypothesis later on.

      Dr_Manhattan

      Delete
    9. >Also it's a reminder that thought process is, however crudely (due to limitations of the instrument), stitched together with math.

      Do you mean the mind is completely material? I believe most of the blogs readers would disagree.

      >Finally, common way to communicate statistical information is with mean and variance; so this is sort of my "mean" estimate, with a couple of orders of magnitude being variance.

      Agreed, my question isn't if it is common method for statistics, but whether statistics is the best method for communicating this information, or whether using mathematics gives us a false feeling of precision (thereby hurting our decision making).

      >In a similar fashion I think human scientists approximate Bayesian probability, and update on evidence, though only approximately. It seems clear to me that "scientific method" is an approximation of Bayesian truth-finding. So, prescriptively, we should be using formal probability, as much as possible, as long as premature attempts to do it do not interfere with early creativity in hypothesis generation.

      Why does this seem clear to you? (i.e. it seems like this is one point where our priors differ) Could you point me to some fundamental breakthroughs made using Bayesian reasoning? (I agree it is useful in many areas such as testing a new drug where statistics are often grossly misused).

      >There are some formal frameworks for coming up with priors, like Solomonoff's Universal Prior. I'm not sure it's quite right, but it's very objective, and probably a huge step in the right direction. I plan on using it when discussion priors for design hypothesis later on.

      I am not familiar with this Solomonoff's Universal Prior, could you point me to something which explains it? (if it relates more to a later point in the discussion, I can wait till then).

      Delete
    10. Also, to make sure that I am understanding your framework correctly, are you assuming all knowledge is the ability to make successful predictions? (e.g. who would you say has more knowledge, one who understands the essential concepts of physics, but is unable to calculate the results of a particular situation, or one who can make accurate predictions, based on a detailed set of equations which are called upon in various circumstances, but does not understand the ideas?) Is the measure of knowledge winning the most (potential) bets?

      Delete
    11. Another question I have about a Bayesian view of science:
      How does Baysien Analysis model cause and effect? From the standpoint of Bayes rule, there is no difference between A causing B, B causing A or A and B both being effects of C. How does the idea of cause enter a probabilistic analysis?

      Delete
    12. > >Also it's a reminder that thought process is, however crudely (due to limitations of the instrument), stitched together with math.
>Do you mean the mind is completely material? I believe most of the blogs readers would disagree. 

      I happen to believe that, but I don't think it's relevant to the argument that the tool is crude. The mind is at the very least partially material (obvious from brain damage cases, Alzheimer's etc.). Also modern psychology demonstrates many ways in which the mind is systematically flawed (books by Kahneman, Ariely and others have good examples). If you believe mind's development was driven by evolution there are obvious reasons for the crudeness: first it takes time for evolutionary pressures to affect populations (so if intelligence is advantageous, it might still take very long to produce perfect intelligence and we might not be at the end of the process), second there are costs to intelligence (bigger brains for instance, which made it harder to give birth to mature protohumans - human newborns are much more immature and need more protection than other primates), third evolution itself is not a perfect "designer" (see this http://blogs.discovermagazine.com/loom/2010/08/30/dawkins-gets-inside-the-giraffes-neck/ for instance).
      
>>Finally, common way to communicate statistical information is with mean and variance; so this is sort of my "mean" estimate, with a couple of orders of magnitude being variance.

>Agreed, my question isn't if it is common method for statistics, but whether statistics is the best method for communicating this information, or whether using mathematics gives us a false feeling of precision (thereby hurting our decision making).
      I think the "feeling" of precision is in the eyes of the perceiver :). As long as you understand variance, you know the estimate is not precise. OTOH there are cases where Gaussian models give false feeling of precision, not because of mean/variance but because they're basically the wrong models altogether. Nasim Taleb wrote quite a bit about this. 

>Why does this seem clear to you? (i.e. it seems like this is one point where our priors differ) Could you point me to some fundamental breakthroughs made using Bayesian reasoning? (I agree it is useful in many areas such as testing a new drug where statistics are often grossly misused).

      I didn't mean exactly "using Bayesian statistics", I meant more in informal sense. I can't really prove it, but it seems scientists operate with an idea of priors and updating.
      
> I am not familiar with this Solomonoff's Universal Prior, could you point me to something which explains it? (if it relates more to a later point in the discussion, I can wait till then).
      You can start here http://www.scholarpedia.org/article/Algorithmic_probability - there are book links there if you want to dive deeper. Also the fairly easy to read http://www.amazon.com/Meta-Math-The-Quest-Omega/dp/1400077974 will explain some related math in accessible fashion.

      Delete
    13. …continued
      > Also, to make sure that I am understanding your framework correctly, are you assuming all knowledge is the ability to make successful predictions? (e.g. who would you say has more knowledge, one who understands the essential concepts of physics, but is unable to calculate the results of a particular situation, or one who can make accurate predictions, based on a detailed set of equations which are called upon in various circumstances, but does not understand the ideas?) Is the measure of knowledge winning the most (potential) bets?

      I'm not sure if prediction is the end of it. Theoretically I'd like to "predict" stuff beyond my own perceptual reach in space/time (things beyond my light cone). As far as concepts vs. precision, I prefer concepts for personal enjoyment, but precision for practicality. Not sure if that helps.


      > Another question I have about a Bayesian view of science:How does Baysien Analysis model cause and effect? From the standpoint of Bayes rule, there is no difference between A causing B, B causing A or A and B both being effects of C. How does the idea of cause enter a probabilistic analysis?

      Good question. Judea Pearl (who got the Turing award this year) did research on causality. I don't believe Bayesian approach "solves" causality, but it doesn't contradict it either. It's objective is to deal with uncertainty, and as far as you will still have some uncertainty determining causality from evidence Bayesian approach will apply.

      Dr_Manhattan

      Delete
    14. >The mind is at the very least partially material (obvious from brain damage cases, Alzheimer's etc.). Also modern psychology demonstrates many ways in which the mind is systematically flawed (books by Kahneman, Ariely and others have good examples)

      Agreed

      >I think the "feeling" of precision is in the eyes of the perceiver :). As long as you understand variance, you know the estimate is not precise.

      understanding and knowing isn't enough. Based on Kahneman, Arielly etc. we need to be careful to formulate our ideas so as to discourage cognitive bias (in ourselves and others). Using mathematics often creates a false sense of being scientific and precise (hence much of the nonsense in the social sciences)

      >I didn't mean exactly "using Bayesian statistics", I meant more in informal sense. I can't really prove it, but it seems scientists operate with an idea of priors and updating.

      agreed,


      > As far as concepts vs. precision, I prefer concepts for personal enjoyment, but precision for practicality. Not sure if that helps.

      Which one is the framework for thinking about an intelligent cause?

      thank you for the sources

      Delete
    15. >> As far as concepts vs. precision, I prefer concepts for personal enjoyment, but precision for practicality. Not sure if that helps.
      > Which one is the framework for thinking about an intelligent cause?

      Seems like a leading question and I'm not sure how my answer will help here. Basically the two frameworks should not diverge - but you might have to trade them off in certain cases. E.g. if I know precise recipe for making a certain drug, and I have good statistical knowledge that it works, I'll settle for the recipe to heal somebody rather than waiting for a more conceptual explanation. Still I believe there is an underlying conceptual explanation to the drug's workings. As far as the question of intelligent designer there is no practical aspect to the question that I see so far so it's mostly of conceptual interest. Like I said, not sure if that helps.

      Dr_Manhattan

      Delete
  9. At the risk of oversimplyifing things is the essential proof the spaceship argument (Post 3)?

    ReplyDelete
    Replies
    1. There are two essential parts to the proof. The first is like you are saying; seeing the design manifest in the fine tuning and inferring an Intelligent Cause to explain the order. (The spaceship argument)

      The second part is realizing how and why all other possible theories fail for the various reasons. That is why we spend so much time disproving the multiverse. In addition, understanding what it is that makes these great scientists say multiverse, helps you see clearly why the two theories of Stage One can not solve the mystery.

      Delete
    2. For clarity can you bullet point the Categorical reasons why the various other theories fail

      Delete
    3. Can't you just read the posts?

      Delete
    4. I did that's y I'm asking the question- but thanks for your reccomendation

      Delete
    5. Here's a brief (albeit a little incomplete) list. You can also try to reread post 6 which summarizes some of these points.

      The two theories of Necessary Existences and Master Mathematical Equation fail to solve the mystery of post 2. How could arbitrary numbers be a fundamental part of reality? And if they weren't fundamental, what kind of fundamental mathematical equation could possibly have determined the constants?

      In addition, the 2 theories can be totally rejected based on the fine tuning of post 3, as according to both theories the fine tuning is totally coincidental.

      Master Mathematical Equation has the addition problem of having to perfectly determine every number down to the last decimal place.

      So far (we're not done) we have mentioned the following flaws in Multiverse theory: A theory that explains anything, explains nothing (Multiverse of the Gaps); falsely claimed to be science when in fact it is philosophy and mathematics; undermines the entire scientific endeavor of discovering knowledge, certainly in its extreme forms; gets involved in a thorny philosophical issue of actual physical infinities; the alleged proofs turn out to be great flaws upon closer examination; no evidence, predictions, tests and observations etc.

      Delete
    6. Kind of a side point, but I don't think it's fair to require a master equation to predict the constants down to the last decimal place. Some of the most accurate physical theories out there (QCD, for example) can predict experimentally observed quantities to very high but not perfect accuracy. We would say here that the theory is most probably true though there are some higher order corrections that we have not taken into account. I think if there was a master equation out there that was able to make good though not exact predictions it would be pretty promising. That is, I think if such equations and the corresponding approximations existed then there is a good chance that there is something real there and as we learned more the predictions would get more and more accurate.

      Delete
    7. It's not a side point. It goes to the heart of the matter.

      We mean that the Master Equation theory holds that in principle there is an equation which does determine every last decimal place. We don't mean that it practically would have to predict every decimal place for it to be compelling.

      If some equation would determine them for 10 decimal places, we could certainly agree that at the very least it was a good approximation of some truly Master Equation we have yet to discover.

      However (and this is the critical point), the theory maintains that there really is that Master Mathematical Equation which does determine the constants to the very last decimal places. This becomes hard to accept with numbers like the cosmological constant.

      Of course, even if you had that equation, the fine tuning problem of post 3 would still exist.

      Delete
  10. I am looking forward to the series of posts on God's unity because I suspect that most physicists are not familiar with our concept and, were they to understand it, would reconsider the "impossibility" of an intelligent designer. They are looking for a simple, elegant first cause. Maybe they haven't realized that our minds cannot fully know this first cause, but can only detect its existence. Just a thought.

    ReplyDelete
  11. Whether or not there is some sort of "smoking gun" evidence that scientists are "biased" toward naturalistic explanations is unimportant. Scientists are biased. Methodological naturalism is the historic guiding principle of science. Methodological naturalism requires, a priori, that scientific theories exclude non-naturalistic causes in explanations. Further, it turns out that philosophical naturalism has dominated the academy. This is sort of undeniable. Now, MN is certainly not a tool exclusive to physicalists/naturalists. Theists understand that if God ever acts as primary cause, those occasions are few. For the most part He created the universe to operate according to secondary/natural causes. MN is a useful tool regardless of one's ontology. However, PN motivates the elevation of MN to a constraint on all explanations, regardless of how logical or reasonable alternative explanations may seem. It is because of PN's dominance that academics who challenge MN are labelled fringe frothy-mouthed radicals in both philosophy and natural science departments. Apart from an ontology, there is no sufficient motivation for such exclusivity.

    Multiverse theory happens to provide an excellent framework to explore the deeper question: should methodological naturalism's role in science be understood as a universally exclusive constraint? If so, what arguments support this? In the case of fine-tuning, MN spits out Multiverse theory. Currently, it is not the best of a set of theories, it is the ONLY theory available when forced to look for natural causes. MN requires us to accept explanations as the "most reasonable" just because one was generated at all. It forces scientists to reject reasonable explanations simply because they offer causes that might not be physical or naturalistic.

    ReplyDelete
    Replies
    1. Good point.
      Please tell us if this properly summarizes your comment: If scientists' premises (MN and furthermore, PN) lead them to the conclusion that a theory as difficult as the multiverse is true, then maybe they should consider dropping (or adjusting) their premises.

      Delete
    2. Do methodological and philosophical naturalism in fact produce the Multiverse theory? It doesn't seem testable enough.

      Delete
    3. Good question!

      It would seem that the multiverse can come under naturalism as it deals with physical (though unobservable) things. Naturalism usually outrules the supernatural, things that aren't governed by the 'laws of nature'. It doesn't seem that testable is ordinarily part of the definition (though maybe it should be).

      However, two quotes from the wiki on naturalism may indicate the contrary:

      1. "Methodological naturalism is thus "a self-imposed convention of science." It is a "ground rule" that "requires scientists to seek explanations in the world around us based upon what we can observe, test, replicate, and verify."[18]"

      2. "Allowing science to appeal to untestable supernatural powers would make the scientist's task meaningless, undermine the discipline that allows science to make progress,..."

      Delete
  12. I know this is a little off topic, but out of curiosity, are you aware of any other theories (or even variations of string theory) that don't require a multiverse?

    As far as I can understand it, string theory has multiple possibilites. But is it at all still possible that one out of those 10^500 geometries are true? Or does the mere fact that there are 10^500 geometries make it a non-workable theory?

    Your arguments are excellent, but I am curious what your views are as to possible theories in physics that would reconcile relativity and quantum mechanics.

    ReplyDelete
    Replies
    1. The possibility of 10^500 geometries for string theory, undermine its ability to make any falsifiable prediction, and hence preclude it from being subject to the scientific method.

      It is possible that one day there will be a way to show how one particular version of string theory is unique, and therefore subject to prediction and experiment, but that day is not here yet and no one can tell you if it will ever come.

      In terms of reconciling general relativity with quantum mechanics, there are a few different approaches out there (loop quantum gravity being one of the more popular alternatives to string theory), but none of them are truly successful (like relativity and quantum are each independently).

      We would suggest a very interesting and readable book by Lee Smolin entitled "Three Roads to Quantum Gravity" written about 12 years ago. As far as we are aware, its still a pretty current account of where modern physics is in its attempt to reconcile the two.

      Delete

In the words of Agur bin-Yakeh: "We welcome all comments, questions, contributions, and critiques - but if you insist on posting anonymously, PLEASE use a pseudonym rather than posting as "Anonymous," since this makes it much easier to carry on a normal discussion. Thank you!"