Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Smolin's 1995 Landscape of Physical Law idea

  1. Jul 27, 2004 #1


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    It's interesting that Lee Smolin used the idea of an evolutionary Landscape of physical law back in 1995

    http://arxiv.org/gr-qc/9505022 [Broken]

    We just had that young quantum gravity researcher Leonardo Modesto at Marseille remove the Black Hole singularity so that spacetime extends thru the black hole to somewhere else ("Disappearance of the Black Hole Singularity in Quantum Gravity" http://arxiv.org/gr-qc/0407097 [Broken])

    Smolin explicitly anticipated that kind of result in Quantum Gravity---eliminating classical GR's Black Hole glitch----in his 1994 paper
    "The fate of black hole singularities and the parameters of the standard models of particle physics and cosmology"
    http://arxiv.org/gr-qc/9404011 [Broken]
    This 1994 already is describes the Landscape concept but doesnt apply the word "landscape" to it.

    ----from 1994 abstract---
    The implications of a cosmological scenario which explains the values of the parameters of the standard models of elementary particle physics and cosmology are discussed. In this scenario these parameters are set by a process analogous to natural selection which follows naturally from the assumption that the singularities in black holes are removed by quantum effects leading to the creation of new expanding regions of the universe.

    In the 1995 paper the analogy is drawn between the genes of an organism and the parameters of physical law which generate the universe---the genes are to the organism as the fundamental constants and the laws of physics are to the universe----different values of the constants means a different universe.
    In evolutionary Biology the ensemble of all possible sets of genes for an organism constitute a "Fitness Landscape" and this will have hills and valleys determined by a fitness function, with selection driving the gene pool to higher nearby levels of fitness (reproductive success).

    Smolin draws the analogy explicitly (see e.g. page 33) and proposes a way the ensemble of possible sets of physical constants can be seen as a reproductive fitness Landscape.
    the act of reproduction in this case being the formation of a black hole.

    He is able to make a falsifiable prediction from this--which as of today still stands and has not been refuted

    "... leads to a definite and testable prediction, which is that, Almost every small change in the parameters of the standard models of particle physics and cosmology will either result in a universe that has less black holes than our present universe, or leaves that number unchanged...."

    Have to go, will get back to this later
    Last edited by a moderator: May 1, 2017
  2. jcsd
  3. Jul 27, 2004 #2
    Will have to look deeper as well.

    Last edited by a moderator: May 1, 2017
  4. Jul 27, 2004 #3
    I think we should refer back to Susskinds talk on Landscape as well for a clear perspective? Peter Woit's site would carry this.


    And for further reference on Landscape?
    Last edited: Jul 27, 2004
  5. Jul 27, 2004 #4
    this might be of some value- when Susskind published an article on his Landscape idea for Edge.org- Lee Smolin responed in the Reality Club-[ http://www.edge.org/discourse/landscape.html ]and a rather heated debate insued between the two of them


    /:set\AI transmedia laboratories

    http://setai-transmedia.com [Broken]
    Last edited by a moderator: May 1, 2017
  6. Jul 27, 2004 #5


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    that is correct. Susskind's first Landscape paper was February 2003, right after Kachru et al came out with the 10100 string theory vacua.

    http://arxiv.org/hep-th/0302219 [Broken]

    The Anthropic Landscape of String Theory
    Leonard Susskind
    21 pages

    "In this lecture I make some educated guesses, about the landscape of string theory vacua. Based on the recent work of a number of authors, it seems plausible that the lanscape is unimaginably large and diverse. Whether we like it or not, this is the kind of behavior that gives credence to the Anthropic Principle. I discuss the theoretical and conceptual issues that arise in developing a cosmology based on the diversity of environments implicit in string theory."

    For String Theory, this paper is what put Landscape on the map. Let's get a broader historical context for it, however.

    In 1995 Smolin used the concept of a fitness Landscape of all possible
    Standard Models, and he did not then or later resort to anthropics. Instead he proposed an explanatory theory of physical law evolution and made a testable prediction. This prediction permits the new theory or scenario (which offers a possible explanation for why we are where we are in the Landscape) to be refuted. Smolin's theory can be falsified---which means that the theory is meaningful and part of the empirical scientific tradition.

    whether or not it is right we do not know, it has not been invalidated as yet.

    Smolin's chapter in the new Cambridge U. P. book elaborates on this
    "Scientific Alternatives to the Anthropic Principle"
    http://arxiv.org/hep-th/0407213 [Broken]
    By way of contrast, in 2003 Susskind employed the Landscape image applied to the plethora or "discretuum" of string vacua which had been found by Kachru and others. He immediately proceeded to invoke the Anthropic Principle, a principle which has several versions.
    Lively discussion ensued.
    From this no testable prediction emerged which would allow either String Theory or the Anthropic Principle to be invalidated or otherwise falsified.
    Doubts have arisen as to whether String Theory as practiced by Susskind and associates can actually be considered a part of science. these doubts were aired on sci.physics.research in 2003 in a long thread called "the string theory crack-up". Disatisfaction has been echoed by some eminent string theorists (Witten, Banks, Gross IIRC) who have been critical of
    Susskind for resorting to Anthropism. the AP hoopla appears to have tended to divide and discredit rather than help matters.

    In mid-June 2004 Susskind posted his latest paper on the Landscape:
    "Naturalness and the Landscape"
    http://arxiv.org/hep-ph/0406197 [Broken]
    "...During the last couple of years an entirely new paradigm has emerged from the ashes of a more traditional view of string theory. The basis of the new paradigm is the stupendous Landscape of sting theory vacua---especially the non-supersymmeteric vacua. These vacua appear to be so numerous that the word Discretuum is used to describe the spectrum of possible values of the cosmological constant..."
    ---end quote---

    This paper was subsequently withdrawn from the arxiv, and is no longer generally available.
    Last edited by a moderator: May 1, 2017
  7. Jul 27, 2004 #6


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    SetAI that discussion is so great!

    I think Smolin has been amazingly patient all things considered.

    He came out in 1995 with this evolutionary landscape idea
    with universes reproducing by blackholing near-copies of themselves.
    So the universe that is most prolific of galaxies and stars and has the best parameters for making black holes wins out in the evolution game---or its genes do.

    It is a testable idea. he has presented one or two astronomical observations that would shoot it down.

    so it has empirical meaning---it makes predictions.

    he actually came up with this in 1994, but in the first paper he didnt say the word Landscape. Anyway he has been waiting 10 years. And it is actually a decent idea that offers an explanation of why

    1/137 is 1/137.036 or whatever
    and why the
    cosmological constant (expressed in Planck) is
    10^-120 or so
    and the other parameters of the Standard Models of cosmology and of particle physics

    so I can understand if he sounds a bit impatient with these Landscape-come-lately Johnnies who when they finally get around to it make what looks like the wrong move and leap into the arms of the Anthropic Principal

    My guess is that it will all get straightened out eventually and the new
    smolin article, which is very plainly reasoned and non-mathematical, will help

    Also the patching of the BH singularity by Modesto (and presumably others) will help. So Smolin evolutionary landscape will finally get a decent hearing--and be checked observationally.
  8. Jul 27, 2004 #7
    In all fairness Marcus, we would have to know the limitations to make it falsifiable? If you agree with this then if you move in the direction presupposed with lorentz invariance,then we would have something to talk about in terms of supersymmetry?

    I had been viewing some of Krasnovs simulations of the early universe in the past and to me something has made it ways into my mind about the sincerity of supersymmetry in regards to how we see this early universe resorting to "pearls and chains."

    This then came to the ideas about http://wc0.worldcrossing.com/WebX?14@118.lroccxWyqad.0@.1ddf4a5f/57 [Broken] and made me think of Glast here again.

    If LQG is supported then the current models we have might have in regards to CMB might be in trouble. To me the current research supports early universe information gathering.

    Last edited by a moderator: May 1, 2017
  9. Jul 30, 2004 #8
    I wnated to make sure I did not confuse the name of Andrey Krasnov with Kirrill Krasnov that Baez is related to in literature.

    Why I wanted to add this here has to do with above posted, and how the intersection of photons held in regard to LQG's attempt at discribing the issues of Glast. from the SRian approach.

    http://wc0.worldcrossing.com/WebX?14@92.YyIIcXeLrlD.0@.1ddf4a5f/57 [Broken]

    I wanted to add this animation animation to help people recognzie the similarity that has made me aware of how this intersection is a vital recognition of how we can see the evolution of the early universe to now.

    We talk about cosmic strings here in the animation as a way of seeing the consolidation of event and structure in that uinverse. What was important for me was to recognize how supersymmetry might arisen from brane realizations, and again, the moire effect was most strange if we had considered this "intersection," of the graviton, as if the photon was to travel through these waves.

    I am still gathering my thoughts here.
    Last edited by a moderator: May 1, 2017
  10. Jul 31, 2004 #9
    Why should the number of black holes created decrease only under small parameter changes? Doesn't Smolin's theory predict that we're in a global optimum for black hole production?
  11. Jul 31, 2004 #10

    your post is a important question, so I do not want to detract from that.... so I am using it for thinking out loud here.

    Hawking Turok Theory

    So I think we need to undertand this place where beginnings and ends are really parts of the cyclical nature? The blackhole, by removing the singularites does this. To me underlying this basis,a geometrical realization must come out front and center as we might see the Life Cylce of Energy and Matter in the cosmos.

    But that's just me.

    See Creation of Blackholes in the Colliders
    Last edited: Jul 31, 2004
  12. Jul 31, 2004 #11


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    this sounds like a helpful question

    evolutionary models in biology tend to achieve local maxima
    there is a landscape of genetic possibilities
    and a fitness function analogous to height
    and if you find yourself on the side of a mountain
    then evolutionary pressures drive you up the mountain

    (higher fitness bunches of genes, more grandchildren etc., win out over
    lower fitness)

    but to leap to an entirely different mountain would be a rare accident depending on a large amount of mutation---cant count on it happening

    the process can only feel out local maxima and blindly grope its way up the slope that its on.

    there is some discretion about what is meant by "small" changes----mathematicians can usually say and these are details to work out.

    the way you tell whether or not you are at a local max (as the Smolin theory predicts with high probability we are) is you see if small changes can make things better, or if they always make things worse---fitness wise.

    does that clarify it some, Onto?
  13. Jul 31, 2004 #12
    I agree that his theory predicts we're at least in a local optimum; but the difference between evolutionary theory and Smolin's theory is that in Smolin's theory the process goes on forever, and if it's at all possible for the universe to mutate (by some preposterous luck) to an even higher peak, then this will happen at some point, and then the universes in the higher peak will exponentially swamp the others. The only way to avoid a global optimum is if that optimum is physically impossible (not just a rare accident, but completely impossible) to reach by any mutation or series of mutations whatsoever.
  14. Jul 31, 2004 #13


    User Avatar
    Staff Emeritus
    Gold Member
    Dearly Missed

    In the general fitness landscape picture of evolution, which Smolin modeled his theory upon, it is by no means a given that the highest peak will be attained. See the better popular books on the subject. Only the highest local fitness is guaranteed. If negative fitnesses lie between that and a higher peak, there may be no adaptive path from the local fitness peak to the global one.
  15. Jul 31, 2004 #14
    I know, but as I said, Smolin's theory is different from evolution, in that it describes a process that goes on for an infinite time. If arbitrarily large mutations are physically possible (even though extremely unlikely), then if you wait long enough the system will jump to the highest peak and stay there. If arbitrarily large mutations aren't physically possible, there still only needs to be a chain of universes, each of which can have at least one black hole, within one mutation step away from each other, ending at the global optimum. After that happens, the global optimum exponentially swamps everything else (I think).
  16. Jul 31, 2004 #15


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    just curious, Onto, why do you think that Smolin's model assumes
    an infinite regression back into the past?

    I didnt yet find in his essay where he says that our series of black hole ancestors goes back indefinitely.

    I would imagine that his prediction (about local maximum) would work also if one just assumes finite

    naively, all one cares about is the prediction which says local maximum,

    and a local can also be a global maximum

    (every global maximum is also a local one)

    all one cares is the prediction that we are probably at a local max
    for black hole production
    and if only that can be falsified by observation then we will have learned something namely that at least one of the postulates is wrong

    ---------anyway that's my take------
    What I am wondering, and asking you, is where does it say
    that the family tree of universes is infinite?
    You say that is a feature of his model. Where did you see it?
  17. Jul 31, 2004 #16
    I meant infinite into the future, not necessarily the past. It seems to me this works automatically, if there's always a universe that has 10^many children. (This gets into nasty probability issues: if you have an exponential tree of universes, with a fixed tiny-but-nonzero probability for any branch to die off, what's the chance that the tree goes on forever?)

    And just to clarify: I agree that it would falsify Smolin's theory if we turned out not to be in a local maximum. However, I think it would also falsify Smolin's theory if we turned out to be in a local maximum, but not in a global maximum. If I can point to anywhere in the entire landscape and show that more black holes are produced there than here, then Smolin's theory has been falsified.

    By the way, if Smolin is saying it all started with one or multiple seed universes that didn't come from any previous universe, I would like to know how he fits this into his theory (how were their parameters selected, for example?). Maybe I missed this in his anthropic principle paper; I only looked at it quickly.
    Last edited: Jul 31, 2004
  18. Jul 31, 2004 #17
    (wrong calculation deleted)
    Last edited: Jul 31, 2004
  19. Jul 31, 2004 #18
    I would think that a cyclic universe with particular probabilities for life would be completely indistinquishable from a single iteration of big bang. For it would seem that no information about the previous iteration could survive the big crunch, and you could never know what iteration or cycle you were in. And with some probability that things could produce life on the first iteration, or first big bang, you could never say that this iteration is not the first.
  20. Jul 31, 2004 #19


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    then the argument you gave earlier that we have to be at a global maximum breaks down
  21. Jul 31, 2004 #20


    User Avatar
    Science Advisor
    Gold Member
    Dearly Missed

    we are discussing Smolin's model----cyclic universe is not part of it.
    let's focus on Smolin's model in this thread

    (don't get me wrong. I like the picture you can get in Loop Quantum Cosmology of alternating expanding and contracting stages. with a smooth bounce instead of singularity---but we should have aseparate thread: it doesnt relate to this evolutionary picture)
    Last edited: Jul 31, 2004
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook