Roger Penrose and Cosmology

Roger Penrose and Cosmology

Famous Cosmologist Roger Penrose is known for the Conformal Cyclic Cosmology model. Dr. Craig defines it and interacts.

Transcript Roger Penrose and Cosmology

KEVIN HARRIS: Does Dr. Craig get it wrong? I mean, he’s not perfect. There is a video on YouTube that claims he gets it wrong, at least on one issue. So let’s take a look. Welcome to Reasonable Faith with Dr. William Lane Craig. I’m Kevin Harris. Let me play some audio for you from a video produced by Skydivephil. You are probably like me, you don’t fully understand the conformal cyclic cosmology that Roger Penrose has developed, but let me assure you that it is fascinating. Even if you are not that much into cosmology I think you are going to find this interesting, and you are going to pick up some important principles that apply to other areas like the importance of correctly interpreting someone’s views and being generous when doing so. It looks like over 55,000 people have viewed this video in which Roger Penrose himself is asked about Dr. Craig’s particular statements on the conformal cyclic cosmology. That is just one video related to this. There are also other social media outlets making an impact about this.

We want Reasonable Faith to continue making an impact all over the world using video, these podcasts, articles, and Dr. Craig’s debates and public speaking and more. So please help us make that important impact if you stand for the importance of Reasonable Faith. Right now any gift that you give will be doubled up to [$250,000] thanks to a generous donor. Take advantage of this. It is only offered until the end of the year so act quickly. You can double the impact of your gift and keep the impact of Reasonable Faith. Go to and give quickly and easily online. We so appreciate it. Please pray for Reasonable Faith as we seek to continue coming into your life in the coming year.

Now here is the audio from the Skydivephil interview with Roger Penrose.[1]

INTERVIEWER: Sean Carroll, he is soon going to be debating a theologian named William Lane Craig. A key difference between their viewpoints is whether or not the universe could exist eternally into the past. When asked about CCC on his podcast, William Lane Craig stated, and I’ll read you this thing that he actually said just so we can check the accuracy, “It is not clear at all that these other cycles are chronologically prior to our big bang. Rather it seems what is described here is more like a multiverse model in which you have twin expanding universes coming out of an origin point so that you do not have one universe chronologically prior to the other. Rather they both share a common beginning before which there is not anything and then you have a sort of branching or multiverse structure.” Our impression is that this is not an accurate representation of the CCC model. I just want to be sure of that. Is that correct?

DR. PENROSE: You are correct. That is very inaccurate. I should say that CCC is definitely not a multiverse model in the form that people put it. They have parallel universes. These are sequential.

KEVIN HARRIS: Dr. Craig, Roger Penrose is a fascinating man. We look to him a lot for cosmology and his study. You’ve been a researcher of his work as well. He has been talking a while about conformal cyclic cosmology. Several people have sent emails to me to check out some videos where it is alleged that you misinterpret Roger Penrose on this. Perhaps some clarification is needed.

DR. CRAIG: Yeah, good, Kevin. Anyone who is interested in this needs to look at the article that I co-authored with James Sinclair which is called “On Non-Singular Space-times and the Beginning of the Universe.” This essay appeared in the book Scientific Approaches to the Philosophy of Religion in the year 2012. In this article Jim Sinclair considers four possible responses to the implication of the Borde-Guth-Vilenkin theorem that implies that the universe had a beginning. He lists four such attempts to evade the implications of the Borde-Guth-Vilenkin theorem. First would be an infinitely contracting universe which then bounces back into an expansion phase. So the universe contracts from infinity and then bounces back into an expansion. The second possibility would be what is called an asymptotically static universe. In this view the universe is basically static from infinity past and then begins to expand at some point.[2] The third type of model would be an eternally cyclic universe, and here Penrose’s conformal cyclic cosmogony would be an example of a universe which is eternally going through a cycle of expansions. Finally, number four would be an attempt to have models based on a reversal of the arrow of time, and the model championed by Sean Carroll – the Carroll-Chen model – would be an example of this.

Jim offers in this article a response to each of these four attempts to evade the implications of the Borde-Guth-Vilenkin theorem. In response to the Penrose conformal cyclic model he offers a number of criticisms. Here I think that Skydivephil and these videos have simply misunderstood what Sinclair is saying. In the video, she asks Penrose, “Is this an accurate representation of your theory?” Jim accurately exposits the theory, but what he then offers is a critique of the theory. It is very important to distinguish between what a theory asserts and what a theory implies. For example, the typical cyclical models of the universe that the universe is like an accordion expanding and contracting from eternity actually implied the beginning of the universe because entropy is conserved from cycle to cycle and so each cycle was bigger than the one before it and took longer than the one before it. So as you trace the cycles back in time they got smaller and smaller until you came to a beginning of the universe. These old cyclic models of the universe actually implied the beginning of the universe. But of course they didn’t assert the beginning of the universe. These were attempts to evade the beginning of the universe! So there is a difference between what a theory asserts and what it implies.

Similarly, with regard to the Carroll-Chen model, in my debate with Sean Carroll what I argue is his model based on a reversal of time actually does not imply an eternal universe, but it really implies a universe with a beginning. How is that? On his model spacetime is sort of like an hourglass that, as you go back in time, the universe shrinks down to this narrow neck and then it re-expands out again. The top of the hourglass is a sort of mirror image of the lower half of the hourglass. In particular, the direction of time in the model runs from the narrow neck into the expanding part. So time does not continue through the neck of the hourglass to the lower half as you go back in time. In no sense, on the Carroll-Chen model, is that lower half of the hourglass “earlier than” or “before” the upper half of the hourglass. This is true even though you can draw a geometrical line right through the hourglass neck from the lower half to the upper half. The arrow of time runs in different directions. They are mirror images of each other. What Sinclair has plausibly argued, I think, is that the Carroll-Chen model is not really a model of an eternal universe. It is a model of a sort of branching universe that has an origin in the neck and then you have two universes emerging each with its independent arrow of time from that common origin point. Aron Wall, as I point out in the debate – a cosmologist – makes the same point. He thinks that the Carroll-Chen model involves a thermodynamic beginning of time which is just as problematic as a metaphysical or philosophical beginning of time.

This is the very similar point that Sinclair is making about the Penrose model. On the Penrose model, his argument is that you can’t really say that these universes that are stacked up are sequentially ordered in time. Rather, there is time in each universe that runs from the Big Bang to infinity, and there isn’t a kind of hypertime in which these are all then sequentially ordered.[3] Sinclair writes this, “We may as well just interpret his model as having a reversed arrow of time when looking to the past of a Big Bang singularity so as to obey a second law of thermodynamics. Extendability by this understanding would be a technical artifact rather than an indication of past eternity.” He is saying that even though you could trace an extendible line through these different sequences or universes, they don’t really stand in temporal order to each other. It was very interesting in the interview with Penrose – did you notice what he said? He said as you go into the past there becomes a time at which mass first becomes relevant. Then he says, “before that” mass was not relevant and so there was no time. On the face of it, that is logically incoherent. It is logically incoherent to say before that there was no time. Sinclair thinks that this same problem attends the conformal cyclic cosmology.

Let me read again from the article, “If two stages of a cosmogony did not stand on one and the same time axis then it would be false to ascribe to them an earlier-than or later-than relation even though extendibility existed between the relevant universe phases. Recall for example that Penrose denies a real time value to his boundary as a physical necessity. When Penrose says ‘the remote future of one phase of the universe becomes the Big Bang of the next’ his statement is arguably incoherent in the context of his model.” So I think that this is a criticism well worth considering, and not just dismissing. It is not an inaccurate representation of the model because it is not a representation at all. It is talking about what the model implies rather than what the model asserts. I think the interview was just misconceived.

KEVIN HARRIS: Bill, as your co-host and friend, I can defend you all day that you are not being disingenuous and that you didn’t re-define Penrose’s model here. But I don’t have to, because so many commentators have pointed this out to people. So many bloggers and commentators said, no, Dr. Craig and Jim Sinclair are critiquing the model or evaluating the model. They are not mis-defining it. All over the Internet I’ve seen your defense here.

DR. CRAIG: Good.

KEVIN HARRIS: People are coming to your defense. I’m glad because when you read headlines like one blogger – “William Lane Craig is either lying or getting things very wrong” – it is inflammatory. It is not very gracious.

DR. CRAIG: Uncharitable. This violates the scholarly principle of interpreting your opponent as charitably as possible – to treat him fairly and give him the benefit of the doubt. I know, for example, that Jim Sinclair is infallibly fair and accurate in representing the cosmological models that he evaluates and critiques. So to say one is lying or disingenuous I think is not scholarly discourse. This is worthy of the Internet but not taking seriously.

KEVIN HARRIS: This particular blog wraps up that he thinks that Penrose’s model - conformal cyclic cosmology – would eliminate the kalam cosmological argument. Do you think as it stands that it would either change or eliminate the second premise?

DR. CRAIG: No, because the second premise is based first and foremost on philosophical arguments of which the scientific evidence is confirmatory. But, secondly, just offering a model isn’t enough to overturn that second premise. You would have to show that this model is true! That it has the evidence of contemporary cosmogony in its favor as opposed to models involving a beginning. It simply doesn’t. It is just a possibility, but there is no grounds for thinking that this model is plausibly true. But Sinclair’s critique is powerful because what he is saying is, first, the model is based on a number of mathematical tricks that are dubious in their physical implications, and that, secondly, even if the model is correct it still doesn’t avoid the beginning of the universe.[4]

KEVIN HARRIS: We’ve done podcasts on the multiverse. The state of the multiverse. I was joining many others in saying there is just not enough evidence for a multiverse. We don’t have to worry about it. Put it on the back burner. Justin Brierley, you’ve been on his show Unbelievable many times, he produced something that said the same thing – multiverse is very interesting but there is not a lot of evidence for it. But it seems that it is not going away quietly into the night. People are still hanging on to it – the cosmologists still want to affirm this model. It is still in the literature apparently is what I’m saying.

DR. CRAIG: Oh, absolutely. This is the main alternative to a cosmic designer – to say that we are just a random member of a multiverse in which all of the various constants and quantities are explored and take various values so somewhere in the multiverse fine-tuned universes like ours will appear by chance.

KEVIN HARRIS: Like it or not, this Skydivephil video – his videos – are well produced. They’ve had quite an impact and lit up the blogosphere on some things. I was reading this blog – it says, “The general thrust of the argument centers on the claim that Penrose’s model does not describe consecutive cycles of the universe but rather that the cycles are concurrent sharing a common beginning. In Skydivephil’s film of an interview with Roger Penrose, Penrose directly refutes this notion stressing that the conformal cycles of his cosmology do not originate at a common beginning and instead are sequential.”

DR. CRAIG: Clearly the blogger has failed to understand the difference between what a theory asserts and what a theory implies. He is mistaken when he says that Sinclair’s critique suggests that the cycles are concurrent. That would be to imply, again, some sort of overarching time in which these universes all exist simultaneously. Sinclair’s point is there is no such time on this. These times within each universe are incommensurable. So you can’t say that these are concurrent anymore than you can say they are sequential. Penrose doesn’t refute this idea. Rather what he does is he corrects the claim that his model does not assert a sequential ordering of these, and that is quite right. It does assert a sequential ordering of these, but this is not to respond to Sinclair’s critique that that part of the model is without physical justification.[5]

[1] See - this portion of the interview starts at the 20 minute 5 second mark. The interviewer quotes from the Reasonable Faith podcast “Truth, Free Will, and Cosmology” found here: (links accessed December 20, 2016).

[2] 5:00

[3] 10:06

[4] 15:15

[5] Total Running Time: 18:19 (Copyright © 2016 William Lane Craig)