Episode Transcript
[00:00:00] Speaker A: But on the other hand, if the universe is a work of mind, one whose purpose evidently includes beings like us and beings capable of understanding the universe, and we're not a cosmic accident. We're part of a purposeful story.
And I don't know about you, but that is hope conferring? I think so. It bears deeply on what it means to be human. The answers to these questions ID the Future, a podcast about evolution and intelligent design.
[00:00:28] Speaker B: Well, by now you're likely familiar with the multiverse hypothesis, which is actually just one of dozens of cosmological models proposed to explain away the significance of a beginning to the universe and the evidence of fine tuning. But you may not know there's actually a rigorous method to evaluate those models, which can be very helpful in understanding the flaws of each. Welcome to Idea of the Future. I'm your host, Andrew McDermott. Today I continue my conversation about evaluating cosmological models with philosopher of physics Bruce Gordon and physicist Brian Miller. Welcome back to the show, gentlemen.
[00:01:04] Speaker C: Thank you. It's a pleasure to be here.
[00:01:06] Speaker A: Pleasure to be with you again.
[00:01:08] Speaker B: Yeah. Well, by way of quick introduction, in case our audience is not familiar with you guys, Bruce is associate research director and a senior fellow with Discovery Institute's center for Science and Culture. He's also a research professor of philosophy of Science at St. Constantine College.
He's an expert in cosmology, general relativity, and quantum mechanics. He received his Ph.D. in the history and philosophy of physics from Northwestern University in Chicago. Bruce is the contributing editor of two books, the Nature of Nature and Biological New Perspectives. And he's the author of numerous articles and essays and journals and academic volumes including Minding the Brain, which is a volume that Discovery Institute Press put out a couple of years ago. Brian, on the other hand, is a senior fellow and research coordinator for the center for Science and Culture here at the Discovery Institute. He helps manage the ID 3.0 research program and helped launch the biennial conference on Engineering and Living systems, or Cells. He obtained a BS in physics with a minor in engineering from MIT and a PhD in complex system physics from Duke University.
His research focuses on thermodynamics, information theory, protein rarity, and the origin of life. He has contributed to multiple books and technical journals covering the debate over intelligent design, including the Mystery of Life's Origin and Inference Review.
So, gentlemen, let us continue a conversation that we started in a previous episode. In Part one, we touched on the Halper Meier debate before looking at key questions that the debate turns on Now. Bruce and Brian also gave us answers to some of the common objections to fine tuning. And Bruce gave us a wonderful detail and dose of complexity as he unpacked the objections to fine tuning and also some answers to that very rigorous, but worth making it through, because we really need to bite the bullet of complexity if we're going to understand why a naturalistic blind process is. Is not responsible for this stuff. Now, in this segment, we're going to turn our attention to Bruce's method for evaluating these cosmological models. He calls it a scorecard. And before I hand the reins over to Brian again, let's just start with some basics. Bruce, I'm curious, how long did you work on developing this scorecard method?
[00:03:30] Speaker A: Well, I've been thinking about issues related to cosmology and different cosmological models for a long time. But in terms of actually putting together a scorecard and a rubric for evaluating competing models in terms of their explanatory adequacy, I've really only been thinking about that for the last, oh, maybe a little over a year or so. It was catalyzed the project by discussions with Steve Meyer and Brian, actually, and some of the stuff that they were needing to respond to and wanting us to think about. And they got me going on this research project. So I've made some progress in the last year. I think I've got a basic structure in place for doing this.
[00:04:20] Speaker B: Yeah. Well, give us a rundown of how the scorecard works. And what yardstick does it use to measure just briefly.
[00:04:28] Speaker A: Well, so the focal question is going to be which model best explains the total evidence, and how do we rank them in terms of their explanatory adequacy?
And to keep everybody honest, we need a yardstick to measure things against.
And so the standard big bang model, the lambda cdm, the cosmological constant, cold dark matter standard model of cosmology, is going to be our yardstick in that respect. So what does the standard model of cosmology do? Well, on the basis of basically six numbers or so, so how much ordinary matter is there? How much dark matter is there?
What's the strength of the primordial ripples that seeded galaxies?
So there's a slight unevenness in the total energy density of the primordial fireball, if you will, which by Einstein's equations, is the same thing as the unevenness in the curvature of space.
All right, the tilt of these ripples, that is how their strength varies with scale.
And then there's a cosmic ruler.
So the apparent size of the.
On the sky, of the pressure waves that are frozen into the early universe, and it fixes the geometry and the expansion rate and something you might call a fog factor, how much the microwave background light Was scattered by the era when the stars first lit up. Okay, so those six numbers, if you will, Are the dials of the standard model of cosmology. And they're measured, all right, so you set them to their measured value, and what pops out? Well, the entire peak by peak structure of the cosmic microwave background, the abundance of hydrogen and helium and deuterium forged in the first three minutes of the universe's existence, the cosmic web of galaxies, the distances to supernova, all these things fall out of those six numbers Applied to the standard model of cosmology. And the dials cross check with each other. Ordinary matter figures read off the microwave background and agree with the independent figure from primordial deuterium. And all of these things kind of dovetail in a way that philosopher of science from the 19th century, William Whewell, called consilience independent standards of evidence from disconnected directions coming together to yield a cohesive picture. So there's a beauty about the standard model of cosmology, the lambda cdm model, and there's a reason we take it as a kind of the baseline for cosmological explanation. But there are some things that it doesn't explain.
So the question of whether there's a beginning there is left open. In fact, it's suggestive of the fact that there was a beginning.
It's geodesically incomplete to the past. And what you do with that judicial incompleteness Is an interesting question in and of itself. We're not going to have a chance to talk about that too much today.
But it's fine tuned with respect to its initial entropy.
So how much order was present in the initial explosion? And turns out this is Roger Penrose's figure was fine tuned to one part and 10 to the 10 to the 123rd power. That's extraordinary fine tuning with a special low entropy start. And then there are various constants of nature that are fine tuned as well. So it makes no pretense to trying to explain these things.
Neither does it, by trying to explain them, worsen the problem, which is what happens with some of the models in question. So we've got this fundamental honesty, and this is kind of our benchmark or our baseline then, that we use for evaluating how other models compare to this model Once we have the standards of evaluation in place. Okay, so what are the evaluative criteria that are used in our analysis? I have 10 that I've introduced into the model. The evaluative criteria divide into two tiers. The first one deals with well, does the model genuinely explain what it claims to explain?
And the second one asks the question of, well, if you accept the model, can you do science?
Is rational science even possible in light of the model? All right, so Those are the 10 criteria. Ontological economy, consilience, causal integrity, physical coherence, resolution of fine tuning, temporal direction, empirical traction, the measure problem, observer typicality and epistemic viability. All right, so 10 criteria.
Now I rank each of those criteria on a scale from 1, from 0 to 3, all right? Where a higher number is actually a worse score.
So zero means you have no significant difficulty with this criterion at all. And three is an epic decisive failure. All right, so you've got four possible points, zero to three.
And a finer scale than that would invite, I think false precision.
And a cruder one wouldn't quite be adequate. So we've got zero, no significant difficulty, three, epic fail, and in between one is, yeah, there's an issue here, but it's potentially resolvable. And two is now it's starting to look pretty serious. I mean it's not an epic failure yet. You might be recoverable from it, but we got a significant issue we need to deal with. All right, so 0, 1, 2, 3. So score the baseline itself to forestall the suspicion that maybe we've stacked the deck.
How does the hot big bang model stand?
It comes out with a weight of 14 out of a possible 60. No fatal flaws, zeros nearly across the board. One point though for unidentified dark matter and dark energy.
And for admitted breakdown at the Planck scale.
And the more substantial penalties, two points each, which weighted come to six on exactly the two things it declines to explain.
The fine tuned start and the fine tuned constants. All right, so total of 14 and that honest 14 then is the bar that every other rival has to clear. And when we do the analysis, every other model scores higher, which on this inverse scale means that it's worse.
Okay, so there's the picture.
[00:11:34] Speaker B: Yeah. Okay, so that gives us plenty of understanding of how the scorecard is applied. And Brian, I'll let you take over here with just the remaining time. We have maybe 10, 15 minutes here.
Brian, maybe you can ask Bruce to walk us through how it might perform, you know, with some of these other models.
Just pick one or two, you know, we don't have to go beyond that, but I just want to let people see how it gets applied. Bruce, you've Given us the base, which is the hot Big Bang, right they call it, which is the current model of, of the beginning to the universe.
How does it apply to some other models? Let's just pick one or two. Brian, you want to take over from here?
[00:12:20] Speaker C: Yeah, and let's just do two models for the first one. Let's talk about bounce cosmologies, because historically that's been the primary means that people try to use to eliminate the beginning of our universe.
So if you apply it to let's say, loop quantum cosmology or other bounce models, what sort of scores do you see?
[00:12:42] Speaker A: Well, if we take loop quantum cosmology for instance, we had a score of 14 for the standard Big Bang. Loop quantum cosmology scores a 28.
So let's talk about bounce models. What's the idea? You're trying to get rid of a beginning by supposing that there was a bounce, which implies a prior contractive phase. All right? So you replace the initial singularity with a bounce and you've got a contracting phase that precedes it. And that contracting phase has some problems associated with it.
First of all, you've got Penrose's fine tuning result. All right? So the state of the bounce has to be extraordinarily special.
It has to converge in such a way that you have extraordinarily fine tuned entropy coming out of the bounce, which means it has to be even more fine tuned coming into the bounce in order to have that come out of the bounce. So huge problem of fine tuning in that respect.
You're also confronted with kind of like a bunch of different theorems that function together to kind of pinch the result. It's a pincer sort of effect. So you've got Tolman entropy. So in a cyclic universe, there's an accumulation of entropy from cycle to cycle that drives the cosmos towards a featureless thermal state.
All right? Because the accumulated entropy makes such each successive cycle larger and longer, tracing the cycles backwards, then shrinks them back to a singularity, Judaism, incompleteness in the past in finite time.
So you're looking at a beginning in effect, from a Tolman standpoint anyway.
And the standard escape then is to let the universe grow a little bit between cycles so that entropy density gets diluted.
But guess what happens with that? Well, you fall right back into the situation that is governed by the Bord Gulenkin theorem. And this is what Kinney and Stein point out. The very growth that dilutes the entropy density supplies exactly the positive averaged expansion that's required to force A pass boundary.
So you have geodesic completeness incompleteness in that situation as well. And then finally, you want to invoke quantum considerations. You've got Wall's Quantum Singularity Theorem to confront in that respect.
It closes the loophole that bounces exploit. Even where quantum effects are allowed to replace the classical energy conditions and allow them to be violated, you still get past incompleteness itself.
Important to note that Wal's proof is semi classical, so it doesn't formally reach into the discrete geometry of loop quantum gravity and loop quantum cosmology, but its thermodynamic logic does.
All right, so we've got a situation of kind of pick your poison, right? Accumulate entropy and reach a thermodynamic dead end, or dilute it by growth and have it fall under the Bord Gutte Vilenkin theorem with geodesic past incompleteness. And the implication that you're confronting something that looks like a beginning, trying to avoid a beginning is very difficult.
[00:16:28] Speaker C: Well, it seems like what's happening from what you're saying is that this model for a bounce, one, it doesn't really get rid of a beginning, but two, it creates even more problems than you have with the Big Bang model. So it's a worse model and doesn't even solve the real problem from what you're saying, I gather.
[00:16:47] Speaker A: Yes. Yeah, you've worsened some of the fine tuning of the Big Bang model in having to explain how the bounce happened. So in a way, then, that would be a fatal flag associated with the fine tuning criterion, and you'd have to set that as an explanation aside. Although let me say that we haven't introduced dual scoring yet.
Okay. And dual scoring would say, let's place this not in a naturalistic context, but in a design context. How does it fare in a design context? And all of these models fare better in a design context.
It's kind of, kind of interesting.
The more you hypothesize these special mechanisms in an effort to alleviate the beginning and the fine tuning problems associated ostensibly with the standard model in cosmology, the more you need design to compensate for all of the problems that you create for yourself.
[00:17:53] Speaker C: And I think that's just really a powerful argument because what you've basically said, and Steve made this point also, is that one, all of these competing models to the standard Big Bang model are deeply problematic. But two, even if you assume they're true, they still point to design much more strongly than to a naturalistic framework. That's an incredible result.
[00:18:19] Speaker A: People will perhaps dispute how badly A model fails and whether the criteria should be weighted the way that I've weighted them. All right, but you can erase the weights and just level the playing field across all 10 criteria.
And that may reduce the cardinal difference between the evaluation defeat, spinal explanatory adequacy. It doesn't disturb the ordinal structure of the analysis. Everything pretty much still comes out in the same order as which is the better explanation and which is the worst.
So it's robust from an ordinal perspective, even though you can have discussions about whether one should assign this value or that value in various places that produces the more extreme results.
[00:19:24] Speaker C: And that's profound that you can quibble over a number here or a weighting here, but it doesn't change the fact that design always comes out on top, which is just incredible.
Could you apply your model very quickly as sort of a second example to Penrose's conformal cyclic cosmology?
And this is an important model because Penrose and Stephen Meyer had really interesting discussion back and forth critiquing each other. So could you mention that model since it's sort of front and center in the debate?
[00:19:58] Speaker A: Sure. So what is the model? It postulates that there's an endless succession of eons or universal ages. The universe expands and then cools until essentially nothing is left. Mass even fades away by hypothesis and you're left with nothing but massless radiation. And then at that point, you're able to conformally rescale reality. Okay, so there's no measuring stick that's inherent in the universe. Mass has disappeared. So you haven't got the Compton wavelength, which would establish a natural yardstick. There's nothing that restricts you from changing the scale.
And so Penrose changes the scale. He maps the massless cold future of the universe onto a hot Big Bang beginning because there's a conformal transformation that takes you from one to the other. So you conformally rescale the cold, empty far future and identify it with a hot Big Bang of the next eon.
You invoke the Weyl curvature hypothesis to keep each new beginning smooth. So it's a stipulation that gravitational entropy be near zero at each new start.
And that transition, as I mentioned, has to be massless.
So you got to get rid of all of the mass in the universe. And there's no mechanism by which known by which that would ever happen. But nonetheless, there's a mass fade out hypothesis. That's part of the picture. The crossover from one eon to the next has different equations that describe it or that can describe it. And there's no way of adjudicating among the different models.
So there's different ways of describing how that might happen.
What can we say about this general picture? The first thing I think that we can say is, well, we seem to have a violation of causal integrity. The map is not the territory. So rescaling relabels one geometry, but it's not a cause that ends one eon or starts the next eon.
Simply because we haven't got a yardstick to measure the universe doesn't mean that the universe doesn't have a measure, that it isn't a certain size.
We can't go from something that is unimaginably large to something that is infinitesimally small by conformal rescaling and saying that just happens because look, here's a mathematical transformation that enables us to map one to the other.
Now, the mathematical description doesn't create the reality. Okay, so there's a deep problem with causal integrity. I would say that that's a fatal flag up front.
But what else? Well, I mean, the entropy books aren't going to bounce, so let me spend a couple of minutes talking about that.
So you've got the gravitational entropy of one eon that really does fall near zero in the far future, but only because it's locked in supermassive black holes and then transferred by Hawking radiation and the evaporation of those, those black holes into the radiation sector. Right? So yeah, the gravitational entropy disappears, but it's transferred into radiation.
And if you look at thermodynamic entropy, then as a count of microstates, it's a pure number.
And that number is unchanged by relabeling scale. But so conformal crossover changes it by exactly zero.
All right, so the very scale blindness that lets Penrose suggest that you can stitch eons together by conformal rescaling is what makes it powerless to address this question of the scale free number of the thermodynamic entropy.
And it turns out that the next Big Bang is going to inherit about 15 orders of magnitude too much entropy in comparison to the previous eon in order for the Big Bang to ever sit at the low entropy floor that the vile condition demands. So in that respect, conformal cyclic cosmology refutes itself after one eon.
Now, what could lower the entropy?
[00:24:39] Speaker C: Well,
[00:24:43] Speaker A: information loss, okay? Genuine black hole information loss, which is not. I mean, some physicists think it could happen, majority do not. It runs against the page curve results that are now recovered within various gravitational theories itself. And it pits Penrose's model against string theory, against loop quantum gravity, against the ADs CFT correspondence and a variety of other approaches to this question.
But even if you grant it, even if you grant information loss of the sort that he needs, it targets the wrong quantity.
It bears on the fine grained reconstructability of what fell into the black holes, Whereas what the next aeon inherits is the coarse grained entropy that's radiated out and dispersed before the hole vanish.
Before the hole vanishes entirely. So unitary conformal cyclic cosmology, and there is such a thing as a physicist named Eckstein that's tried to explore that possibility, it doesn't fare any better in that respect.
So when Penrose says something like the second law isn't violated, it's transcended because the effective entropy definition just shifts down.
It's basically a decision not to count. The microscope states that are still physically present and mass fade out can't rescue.
And the inherited entropy is already in the massless sector that carries over the crossover. So the incredibly special low entropy condition required by the Weyl curvature hypothesis has to be put in by hand. It's not explained by the transition. It's imposed at every single crossover and demanded. If there are infinitely many of these things, it has to happen infinitely many times over, which is again then a multiplication of the fine tuning problem rather than a resolution of it, and you've got another epic fail of a fatal nature.
So anyway, that's a basic picture of things in terms of where it comes out on the scale.
I think it comes out at 32 orders of magnitude, which is amazingly bad.
[00:27:10] Speaker C: Now, going through just some of your presentations, I was really struck by what I see as a pattern. So would you say that when you look at the sort of big picture, all these various models out there, do you see a consistent pattern where one, they appear far less viable than the Big bang, and two, they're much, they present much more evidence of design. Is that the pattern you see?
[00:27:34] Speaker A: Yeah, I mean, does every model point to a designer? Yes, it does, in a way.
So think about the standard model for just a second. The only real hits that it takes are precisely with respect to two things that a purposive intelligence would alleviate.
The fine tuning of the entropy at the beginning and the fine tuning of various constants for the universe to be compatible with life. All right, so if you score each model twice, you do dual scoring once with an indifferent distribution, and it scores 14 on the Jeffreys scale.
If you take out the things that are really contributing to the problem, that score drops down to acceptable range of 2 on a theistic understanding of the hot big bang model.
And in fact that pattern is repeated with respect to every model, they all score better.
If you apply design considerations or allow design considerations to alleviate the fine tuning that is associated with each of them.
What can we say about that? Is that, is that putting a thumb on the scale? Well, I mean, explanatory adequacy is not going to be a context free quantity. Whether a model's inability to explain fine tuning counts as a deficiency depends on what's available to do the explaining.
So if origins trace to undirected nature of some sort of, which by its very character supplies no reason to expect life permitting outcomes, then the inability to explain them when they're there is a grave defect. All right, but if it traces to transcendent intelligent agency whose purposes include embodied conscious beings and their existence, then that same inability is far less damaging.
Context itself supplies the resources that the physical model itself lacks.
And I think that's an important insight. All right, and that deficiency generally what I see is that it grows with naturalistic ambition. All right, so the deficiency always favors mind. And the more you try to get rid of mind, the worse problem you have and the more it's alleviated by reintroducing Mott.
So the criteria that the design context relieves are definite. And there's a short list. Reifying mathematics is a cause. Magnifying fine tuning, imposing low entropy bounds by fiat, and the various pathologies that come from a vast ensemble of unobservable entities or unobservable universes. All right, so those things are alleviated by the design hypothesis.
And it doesn't excuse every defect. It only relieves those problems that concern the source of order.
It doesn't relieve things associated with a model's intrinsic costs.
And that's why this ranking of models survives the introduction of the design hypothesis. So if you do dual scoring though.
Yeah, design as the metaphysical context reduces the scores of explanatory inadequacy of every model and it reduces the one for the standard model down to an acceptable range on the Jeffreys scale. Classically considered.
[00:31:21] Speaker C: Okay, Bruce, that was a wonderful description. Could you just describe why people should care? Now we love this stuff. We were both helping Steve Meyer as consultants for his book Return of the God Hypothesis. But why should this be relevant to the non physicists, this conversation?
[00:31:38] Speaker A: Well, it's addressing a fundamental question. I mean, Hawking himself asked the Question, what breathes fire into the equations and makes a universe for them to describe? Which is an insight that points toward causal integrity, right?
Math doesn't make the world, math describes the world.
And every cosmological model is going to have to stop somewhere. And the honest question then is going to be where is it going to be? Brute, unexplained fine tuning because even the multiverse has relocated the fine tuning, or is it going to be in a transcendent rational mind? And I mean, brute factuality is deeply, deeply problematic. If you abandon the principle of sufficient reason, saying that, you know, contingent states of affairs have to have an explanation, then if it's possible for something to happen for no reason at all, that possibility becomes a competitor for every other explanation that you might think you have. And what's more, you can't say that it's improbable because you haven't got any antecedent states of affairs that would confer objective improbability on it. So you can't say it's improbable that there's no explanation and the whole edifice of science comes tumbling down. And not just that, but knowledge in general comes tumbling down. You wind up in a skeptical situation where you can't even say that it's improbable that your current perceptions aren't happening for no reason at all.
So the principle of sufficient reason is pretty important thing. And to jump to brute factuality is a huge mistake. And of course, I said earlier that physics itself is on the, the track of, or is driven by the engine of explanation. And if you give up on that, you're essentially giving up on science itself. So that much said, the implications, I think, reach well beyond physics. While the audit that we've been doing doesn't really force anyone, it does show quantitatively what the costs are and, and ultimately, you know, if we're an accident, then there's no purpose to be discovered, no story that we're a part of. In the long run, there's nothing left when the energy of the universe runs out and all we've got is this brief candle of light around us.
And you might say, in the bigger picture, then what's the point? We'll disappear from the scene and things, the universe won't care. And ultimately all of human history, as Bertrand Russell once put it in his essay, a free man's worship is destined to extinction, and he put it in the vast death of the solar system. But yeah, what's the point? He said, only on the future of unyielding despair or on the foundation of unyielding despair, can we. Can we build what we're going to build? But if it's on the basis of unyielding despair, then what's the point? But on the other hand, if the universe is a work of mind, one whose purpose evidently includes beings like us and beings capable of understanding the universe, and we're not a cosmic accident, we're part of a purposeful story.
And I don't know about you, but that is hope conferring, I think so. It bears deeply on what it means to be human. The answers to these questions.
[00:35:10] Speaker C: Yeah, well, that. That. That was beautifully said. And again, I just want to reiterate what you've shown, which is incredible, is that the standard model that fits the data the best points to design very clearly.
But every attempt to avoid the evidence of design requires models that are less plausible and point to design even more.
And I think that's not just true in cosmology, but you see it in Origin of Life, you see it in biology, you see it with our planet, that the more science advances, the more you see clear evidence of a mind behind our universe that gives us meaning. So thank you for your. Thank you for your answering our questions.
[00:35:49] Speaker A: You bet, you bet. It's been a pleasure to be here.
[00:35:51] Speaker B: Yeah, thank you, Bruce, and for all the work in developing this methodology for, you know, evaluating these proposals. As you said, there's dozens and dozens of them, and we need to have some way to say, you know, nope, sorry, fatal or no, no, that's not as good as this, and you have provided that. So that's really, really awesome work, and I'm glad we could do a flyover of that in the. In these episodes. Now, as we close, we should mention some ways that our audience can learn more about the topic.
There's obviously always ways to dive into this stuff.
In part one, I mentioned the debate between Phil Halper and Dr. Stephen Meyer that's available on YouTube. There's also Dr. Meyer's book Return of the God Hypothesis. And Brian the movie the Story of Everything brings this to life visually, doesn't it?
[00:36:41] Speaker C: Yeah, it covers a lot of this material with extraordinary animations, so it's even more accessible than just hearing it on a podcast.
[00:36:49] Speaker B: And both of you have been writing about this at Science and Culture Today, so folks can. Can look that up as well.
And of course, I'll mention. Don't miss part one of this conversation in a separate episode where we're sort of setting things up And I did have a separate conversation with Brian, too, about this and about the helper debate, so you can check that out. Well, Bruce, Brian, thanks again for your time and unpacking all of this with us.
[00:37:16] Speaker C: It's been a pleasure.
[00:37:18] Speaker A: It's been a pleasure indeed. Thank you, Andrew.
[00:37:20] Speaker B: And yeah, you're welcome. Thanks for coming on. Well, for ID the Future, I'm Andrew McDermott. Thanks for joining us.
[00:37:28] Speaker A: Visit us at idthefuture.com and intelligentdesign.org this program is copyright Discovery Institute and recorded by its center for Science and Culture.
