Pharyngula

Frank Tipler claims to have proven the existence of god with phyics. Will this be the sort of answer we’ll see in Ray Comfort’s promised 13 minute proof?

Notice that we don’t actually get to see the infamous equation. We see Tipler scribble a few words like “quantum” (does the word “quantum” actually appear imbedded in the formulas describing quantum mechanics?) and a few bits and pieces of math, and then the camera lingers on him writing “= god exists”, but there isn’t any substance shown. He says stuff like, “If you are using quantum mechanics and general relativity, you are forced to conclude that god exists,” but we never get a summary beyond the assertion.

I tried to find the equation on the net — it ought to be there, somewhere, right? He was scribbling it up quickly on a single large chalkboard, so it must be reasonably concise. Alas, the closest I could come was this plain English summary:

Tipler then concludes that life must be present to the end, using a string of complex and partly circular arguments: black holes evaporate, this would violate ‘unitarity, a fundamental law of quantum mechanics’ (nowhere in books on quantum theory, not even in Tipler’s book, do I find ‘unitarity’ mentioned), so universe must collapse, but ‘event horizons’ would force information and entropy to approach zero, this contradicts second law, thus event horizons do not exist, thus information goes to infinity, thus the universe is closed and goes to final singularity, but without life this would yield an infinitely improbable state, this contradicts second law, thus life must be present to guide universe to final singularity, thus event horizons are absent.

Biology has something to do with black holes? What? This guy is a kook. I’m not even going to bother with his claims to have proven the trinity and the truth of some brand of Christianity. You’ll also have to read Victor Stenger’s review of one of his books — Tipler is one weird dude.

Ray Comfort and Kirk Cameron are going to have to do much better than that. I didn’t even feel an urge to rise from my chair from that, let alone kneel down and worship the Cosmic Pixie.

(via richarddawkins.net)

Comments

  1. #1 Torbjörn Larsson. OM
    May 6, 2007

    This is an unwieldy subject, especially without having read Tipler’s books.

    But to start with the reviews, Bonting seems to be a theologian without sufficient background to analyze Tipler’s arguments. He doesn’t know about such elementary properties of QM as unitarity, he describes singularities as inhabiting a point in time, et cetera. When Tipler describes Bonting’s depiction as “inaccurate”, it is probably a fair assessment.

    Stenger is instead well positioned to discuss Tipler, and seems to do so well. He promptly describes the ideas as old hat, and notes that the big crunch isn’t currently expected to happen. (Cosmologists are instead discussing if the converse singularity, the big rip, is likely. See for example Rob Knop’s recent post “The Big Rip : an end to the Universe without recollapse”.)

    He also notes that Tipler’s view means a fully naturalistic description of gods, souls and afterlife, and that theologists like Bonting will probably not have any of it. It is, as Collins’ and Miller’s efforts, simply bad theology. Further, since Tipler’s rapture is equivalent to a simulation, and as Stenger notes it is much easier to posit and more likely to occur that we already lives in one.

    I like that Stenger can both trivialize Tipler’s achievements and implicitly point him out as a nitwit by a proxy. “I am not sure he isn’t pulling our legs.” Yes, I wonder too.

  2. #2 Torbjörn Larsson. OM
    May 6, 2007

    So what do we get of Tipler’s argument, without reading his book? Not much.

    Notice that we don’t actually get to see the infamous equation.

    Apparently he wrote down the standard formulation of GR as an illustration. Reading Stenger’s interview, it seems Tipler’s argument is long and complex, riding on a lot of assumptions, and not amenable for a single line statement. Assumptions that furthermore is currently believed to be wrong.

    Biology has something to do with black holes?

    So you haven’t seen Smolin’s idea of “Fecund universes” (or, inaptly, “Cosmological natural selection”), which features evolution twice!? ;-)

    First, it incorporates a mechanism similar to selection to replace the anthropic principle for multiverses. Smolin uses the idea that black holes may spawn new universes. Thus, the universes that maximizes black hole production will dominate statistically.

    Second, universes that makes black holes are obviously also good for producing life.

    That black holes can spawn baby universes was later rethought. Smolin’s idea has other problems as well.

    Btw, as Smolin himself notes, processes that are similar to evolution are attractive because they can in principle be described from local behavior and statistics. (For example the anthropic principle has problems here.) In that perspective evolution is an ingenious invention by nature. :-)

  3. #3 Torbjörn Larsson. OM
    May 6, 2007

    Part of Tipler’s argument seems to be as Stenger says old hat, but it can be illustrative to see where his assumptions (as related here) deviates from current expectations.

    black holes evaporate, this would violate ‘unitarity, a fundamental law of quantum mechanics’

    I see in the thread several claims to the effect that “quantum mechanics and general relativity results in anything but total confusion “. That is wrong, and it was the genius of Hawking to demonstrate one of the first breakthroughs here.

    First, both QM and GR have several known shared characteristics.

    They both obey thermodynamics, locally. (Locally, since GR doesn’t admit describing total energy.)

    They both obey special relativity. locally. (Locally, due to GR again. And relativistic QM formulations are somewhat difficult at times, for example as QFT’s.)

    Second, gravity can be quantized. In fact, Distler describes this in an elegant post:

    It’s often said that it is difficult to reconcile quantum mechanics (quantum field theory) and general relativity. That is wrong. We have what is, for many purposes, a perfectly good effective field theory description of quantum gravity. It is governed by a Lagrangian

    which he shows but I will spare you here. ;-)

    In other words, as an effective field theory, gravity is no worse, nor better, than any other of the effective field theories we know and love.

    The trouble is that all hell breaks loose for ? ~ 1. Then all of these infinite number of coupling become equally important, and we lose control, both computationally and conceptually.

    Third, in different reconciliations so called “semi-classical” theories are used to combine different aspects of QM and GR. Hawking’s description of black hole evaporation is one such, inflationary cosmologies is another, CMB anisotropy is yet another, et cetera.

    Since this comment is too long already, I will continue later.

  4. #4 Torbjörn Larsson. OM
    May 6, 2007

    So, where there we? Oh, yes:

    black holes evaporate, this would violate ‘unitarity, a fundamental law of quantum mechanics’

    So we can combine GR and QM in some cases. Hawking did that for black holes.

    A black hole curves space so severely by its gravitation that it is surrounded by an event horizon. This is where light or any other signals or matter future will inexorably be inside the horizon, there is no path out.

    The horizon is not a specific surface with high curvature or any other recognizable feature. But as I understand it, Hawking and others come to understand that it would also swallow some of the virtual particles that vacuum always produce temporarily. The remainder would radiate as thermal Hawking radiation.

    So not surprisingly black holes obey thermodynamics, they have a temperature and an entropy. But they have not much else besides that – a mass, momenta, angular momenta, and a charge. All other information about the matter that formed the hole or later sucked in is hidden behind the horizon. Famously, “Black holes have no hair.”

    Hmm. Seems a coffee break is coming up. The remaining information is sucked up by the black coffee horizon, for now. ;-)

  5. #5 David Marjanovi?
    May 6, 2007

    Second, universes that makes black holes are obviously also good for producing life.

    That part is testable: find a neutron star more than 1.5 times as heavy as the sun, and it’s falsified.

    That black holes can spawn baby universes was later rethought. Smolin’s idea has other problems as well.

    Could you elaborate? Smolin seems not to have been convinced.

  6. #6 David Marjanovi?
    May 6, 2007

    Man, did I underestimate the traffic here…

    Second, gravity can be quantized.

    I remember having read about an experiment where people took a neutron and let it drop a few micrometers. It fell stepwise. Cool, no?

  7. #7 Torbjörn Larsson. OM
    May 6, 2007

    Well, well. It does take time to discuss what kooks can throw out in a single sentence.

    black holes evaporate, this would violate ‘unitarity, a fundamental law of quantum mechanics’

    So we know that “Black holes have no hair.”

    Now, this is a problem. If black holes radiate their energy (Hawking), they will eventually evaporate. As Dustin and Erik explained, QM demands unitarity, essentially conservation, and especially conservation of probability. (No “poof!” in physics. :-) So where did the entropy information go?

    AFAIK the problem, closer described by Erik, is still standing. As we can see in the link above, while Hawking radiation has entropy physicists seems to know too little about QG to impress the black hole entropy on it. (Since Erik mentioned the details, IIRC they are trying to get a handle on the horizon description.)

    Some went so far to suggest that unitarity is violated. But most physicists don’t think so. Hawking himself seems to have resolved the problem in his mind lately. He choose to admit defeat in a bet about evaporation and unitarity violation, and gave out a paper with his (semi-classical) resolution.

    Specifically, I think earlier explicit entropy violation formulations for black holes with string theory has been replaced with ideas where black holes are giant coherent quantum states, fuzzballs.

    Samir Mathur of Ohio State University calculated that the event horizon of a fuzzball agrees with the current theory of black holes, but in one way it is different. The event horizon of a black hole is very precise and strict while in a fuzzball the event horizon is very much like a mist; it is fuzzy, hence the name ‘fuzzball’.

    As described earlier a fuzzball doesn’t have a prominent singularity at its center, and so the destruction of data that is the essence of a black hole no longer exists in a fuzzball. Instead the data from the fuzzball marks the strings that carry the information in vibrations. These data can be given out by the escape of Hawking radiation.

  8. #8 Torbjörn Larsson, OM
    May 6, 2007

    David:

    That black holes can spawn baby universes was later rethought. Smolin’s idea has other problems as well.

    Could you elaborate? Smolin seems not to have been convinced.

    I think the spawn process was related to the “white hole” solutions of black holes worming out to, or creating, other spacetimes. The specific details why white holes aren’t seen as a current attractive possibility I can’t remember, but I think Lubos Motls had a criticism on his blog that you can google. I have a vague memory with that it was not derived from string theory.

    The “other problems” are of the type that his environmental principle really has the same problem with admitting a descriptive statistic to make predictions with as the weak anthropic principle. In contrast to natural selection his baby universes doesn’t die. So even though we observe black holes and life “there’s no particular reason that we should be in a universe after many “reproductions” or one that is more primeval”.

    There are other environmental suggestions that may not have this problem, for example Boussou’s et al Causal Entropic Principle, maximizing entropy in the causally connected region (“causal diamond”) of their semiclassical cosmology.

  9. #9 Torbjörn Larsson, OM
    May 6, 2007

    uuups. “Lubos Motls” – Lubos Motl; in case you go googling.

  10. #10 Torbjörn Larsson, OM
    May 6, 2007

    And finally some words on Tipler’s cosmology, since it isn’t quite as spectacular as it sounds.

    this would violate ‘unitarity, a fundamental law of quantum mechanics’ so universe must collapse,

    I don’t know details about Tipler’s Omega point.

    However, that the future final condition (as a boundary condition) could decide the evolution of cosmology is not a new idea. Again, old hat as Stenger notes.

    Currently I believe Hawking and Hartle presents such cosmologies in the perspective of “the wavefunction of the universe”. However, I think their universe ends in a type of universe (de Sitter) we don’t observe.

    The cosmology and observations we know today seems to preclude a collapse. See the earlier link to Rob Knop.

    Erik:

    the issue here should perhaps be whether or not black hole evaporation is a case of unitary time evolution. AFAIK (I’m not an expert) this is an unresolved question. Has Tipler or anything else argued that black hole evaporation violates thermodynamics?

    Well, if we could trust Bonting, it seems so: “but ‘event horizons’ would force information and entropy to approach zero, this contradicts second law, thus event horizons do not exist”. Sounds a lot like it could be akin to Hawking’s very technical analysis of unitarity. Except coming to the reverse conclusion.

    Parallel to Hawking-Hartle above, a few iconoclastic physicists seem to have taken up old and new ideas around reversibility of microstates and perhaps retarded- and advanced-time solutions to fields, in discussions of the arrow of time. That would presumably have problems with 2LOT. Is that what you think Tipler discusses?

  11. #11 Torbjörn Larsson, OM
    May 7, 2007

    Erik:

    Thanks for the link! As a matter of fact, I had to complement those 10 pages with skimming a 61 (!) page rather technical paper to understand how his interpretation of QM ties into his problem with unitarity. I hope this cold dip into his production will keep me away from his books. :-)

    Since it become pretty involved, I think I will make two comments. The first here about his work in general, and some problems with that. The second around the specifics and some problems with that, answering you.

    In general, Tipler seems to me to be:

    - Technically proficient. (I see Keith agrees.)

    For this layman, he seems both necessarily erudite and able to both model and extract answers where he needs them.

    (Oh, have I forgotten things! Initially, I couldn’t wrap my head around a central point in his QM treatment involving a PDE quite like a Laplacian. A bounded answer on an unbounded volume means Liouville’s theorem gives a constant answer for the Laplacian, and it was applicable for his source-free PDE as well. Duh!)

    And now for the problems. He also seems to be:

    - A world builder.

    Nothing inherently wrong with that perhaps, many physicists are such even though it is mostly more productive to answer specific questions. But his model is more than most a house of cards. Remove a key assumption, and it will all fall.

    - Rambling.

    Both in his short general argumentation paper and the longer technical one he draws in wast amounts of nonessential details not because of its support but because he can treat it in his model.

    - Religiously inclined.

    Even though he professes to be naturalistic, he uses some of the usual methods such as large number reasoning (akin to his anthropic reasoning, I assume). He places unlimited intelligence as needed for his Omega Point treatment. “Once we give up these human ways of thinking, we can appreciate the true relation between intelligent life and the cosmos.”

    I don’t think teleology is needed for him, bit it is one of those nonessential details he injects.

  12. #12 Torbjörn Larsson, OM
    May 7, 2007

    Erik:

    OK, now to answer your question. I think the 2LOT argument is this:

    In the presence of life absence of event horizons are most likely, since it is a necessary requirement for seeing our universe as it is, in Tipler’s model.

    And since the absence of event horizons can’t be unlikely in Tipler’s model, life must be present at the final singularity.

    ————————
    Since I needed to briefly look into Tipler’s model, I will take this opportunity to put my sparse notes on the net to have them handy:

    It seems to me that Tipler’s whole model builds on the problem with unitarity in event horizons. It is such a large machine so I don’t see how to compact it. (Actually, I lost the original pdf on his technical article, and found the arxiv one. It was really 100 pages outside the journal typesetting …)

    Instead, I will go though the main points and comment them as I go.

    - By forbidding event horizons Tipler claims to be parsimoniously consistent with known physics.

    - Zel’dovich seems to have treated event horizons in parallel with Hawking. He modeled proton (i.e. hydrogen) lifetime against the possibility that the vacuum very briefly produces virtual mini black holes. (As every other possible quantum state.) Tipler can conclude that to preserve unitarity he must prevent this, and the universe must be closed.

    Note: I haven’t looked into Zel’dovich model, but I found a comment elsewhere to the effect that he assumed small event horizons are effective. IIRC, I have seen other comments where mini black holes event horizons have virtually zero cross section against particles. Perhaps time to learn how these treatments work. :-)

    - Closedness permits Tipler to make his analysis of the wavefunction of the universe. He reconciles the Copenhagen interpretation with the Many world interpretation. The latter he interprets as deterministic.

    Note: Several problems here. Using interpretations is okay as long as the models produces falsifiable predictions. (And he does.) But he portrays Many worlds as “Multiversas”.

    The determinism in MWI is in the global view, yet he claims the local universe seems deterministic. But even here QM appears as producing genuine finegrained stochasticity at observations.

    Assume he uses a global time operator to later introduce his inverse time operator. Hmm, seems to be totally opposite claim to Bousso’s causal diamond, incidentally introduced to explain observations around event horizons…

    - Tipler uses consistency between Copenhagen interpretation and Many world interpretation to claim recollapse of his “multiverse”. (That was his picture in the video, btw.)

    Note: The Copenhagen interpretation is used to pick a classic sector in the MWI – seems nice. He is also picking a specific Hilbert space (rigged) and specific initial conditions (delta functions) that enforces global classicality everywhere, even into the singularities – not so nice.

    MWI world doesn’t interact – presumably not a problem in the final collapse.

    - Determinism permits to claim future-to-past determinism (causation). This is interpreted as teleology.

    Note: Several problems here. Teleology it becomes later, by introducing intelligence. He doesn’t really use teleology later. He glosses over classical chaos and how it produces coarsegrained randomness.

    - Life speed constraint forces life, surviving by being rare by the Fermi paradox, but wasteful by evolution, to come up with means to empty the universe of matter to produce energy.

    Note: Tipler is arguing against morality and/or intelligence as prohibiting waste and ruthless consumism, merely noting that life speed constraint prohibits “galactic government”.

    He is also mentioning a large number argument by Ayala that intelligent life is improbable. Channeling Dembski.

    Yes, monocellular life is the most successful. No, evolution being symmetric (going any which way) doesn’t preclude asymmetric initial conditions (monocellular life) diffusing to complex niches. A probabilistic model is probably too hard to set up. But, another treatment to check.

    The total matter-to energy conversion relies on ‘his’ unitarity producing an inverse time operator and used on his interpretation of baryogenesis.

    - Tipler uses a result from Penrose that closed universes have zero total energy.

    Note: Unknown treatment to me, vaguely familiar discussion. GR doesn’t permit total energy treatment, so I think Penrose’s result is problematical. In any case, two other treatments to eventually look at.

    - By constant total energy Tipler claims life is forced to resolve the final singularity, which also explains why it must survive that long.

    Note: He must finetune his model so every Hubble volume has one intelligence. He must also assume that the matter-to-energy conversion is highly (totally?) efficient in each volume.

    Final analysis: Funny how all his assumptions in his house of cards model of “Life, the Universe, and Everything” reminds of Ayalas improbability argument…

  13. #13 Torbjörn Larsson, OM
    May 7, 2007

    This must be my day of mega commentaries. Time to cut back now. ;-)

    Erik:

    And since the absence of event horizons can’t be unlikely in Tipler’s model, life must be present at the final singularity.

    Sorry, I forgot: life must be present, since it’s actions prevent the final singularity to become an event horizon.

    Chuck:

    In case Tipler hasn’t heard, the universe is NOT closed. In the late 1990s they discovered that the rate of expansion of the universe is actually increasing. Cosmology has been grasping for explanations, and trotted out the old cosmological constant, or dark energy, as a possibility.

    Tipler is familiar with these developments. In fact, if you look into the paper Erik linked to you will see that he uses the acceleration in expansion and the cosmological constant.

    Among other things, as evidence for his baryogenesis – ‘anti-baryogenesis’ treatment. In fact, he claims that his 1994 book predicted an acceleration, but in his recollapse phase (presumably prolonging what he thinks is inevitable).

    Ollie:

    As already briefly mentioned, if he was somehow able to use both quantum mechanics AND general relativity, he’d be up for the nobel prize. This would mean achievement of the Grand Unified Theory (not to be confused with the Theory of Everything).

    GUT theories seems to combine gauge descriptions except GR, and some predicts the GUT scale – when the effective forces except gravitation comes together. ( http://en.wikipedia.org/wiki/Grand_unified_theory )

    So I don’t see how theories describing this combines QM and GR.

    Quantum gravity (QG) is the usual shorthand for combining GR with QM. TOE is the usual shorthand for combining all known effective forces (with QM). In string theory, QG and TOE is probably the same thing – it seems you can’t do one without the other.

  14. #14 Torbjörn Larsson, OM
    May 7, 2007

    I still cannot make sense of Tipler’s argument concerning event horizons (how does the event horizons of the black holes disappear? apparently they neither remain nor evaporate…).

    I think he doesn’t let them evaporate, and relies on “intelligent life” to resolve the final singularity. But as you I find him hard to follow here, I can’t find any explicit claim.

    On the general definition I sort of agree, but Tipler could make an operative definition because he specifies in detail what he demands.

    Incidentally, your question makes me realise a consistency problem in his treatment. He notes that intelligence is rare because evolution can go any which way. We observe intelligence today, but it could disappear later.

    Now, in his ‘teleology’ he needs intelligent life, and speculate or assume ‘more intelligence’. Where did the “any which way” go?

    In any case, it was interesting to see what he did. It wasn’t as ill considered as I thought, but it was still impossibly tenuous.

    And far from main stream and where current observations and models takes us. Nothing interesting, I think.