5 / 06
Image of birds flying. Image of birds flying.

#280 What Does One Mean by “the Universe”?

August 26, 2012

I have a question about contemporary cosmology and how the word "universe" is often used. As you know when a proponent of the Kalam argument uses the word universe" they mean "the entire space time manifold." However a while ago on the forums and on the link at the very bottom somebody pointed out that when physicists like Vilenkin use the word "universe," they are using it differently than proponents of Kalam are. He then gives a quote (via email correspondence apparently) from Vilenkin saying

It is certainly more than what we can have access to. Regions beyond our cosmic horizon are included. But if there are other universes whose space and time are completely disconnected from ours, those are not included. So, by “universe” I mean the entire connected spacetime region.--Alex V.

If this is true then what does this do for the scientific evidence that you and Craig use to defend premise (2) of Kalam? Thanks for your time.

Source of the quote comes from here: http://debunkingwlc.wordpress.com/2012/03/03/universe-kalam-and-equivocation/


Flag of United States. United States

Photo of Dr. Craig.

Dr. craig’s response


This question was actually sent to my colleague James Sinclair, who kindly forwarded his response as a good candidate for a Question of the Week. Always glad to familiarize our readers with Jim’s work, I concur. But before I share his response, I want to give the simple answer, which seems to me altogether adequate.

Vilenkin and I are using the word “universe” with exactly the same meaning. Whether there are other causally disconnected universes is entirely irrelevant to the soundness of the kalam cosmological argument. Think about it: suppose, for the sake of argument, that there are other universes causally disconnected with ours. They are for that very reason irrelevant to the conclusion of the kalam argument. In particular, they cannot be the cause of the universe spoken of in the argument’s conclusion. So all my arguments concerning the properties of the cause of the universe go through as before.

Thus, the non-theist cannot avoid God by the metaphysical conjecture that other disconnected universes might exist. Of course, if the non-theist conjectures that these other universes are causally connected to ours, then they are comprised by the universe and we are right back to Vilenkin’s arguments that there is no tenable model of the universe, so defined, which does not have a beginning in the finite past. The so-called “null topology,” discussed by Jim below, if construed as physically real, is by definition part of the universe and so in need of an explanation of its coming into being.

Now for Jim’s more scientifically sophisticated answer, read on:


Thanks for the question. Let me start by commenting that the issue at hand wouldn't affect the kalam argument’s second premise


2. The universe began to exist.

at all, let alone the scientific evidence that WLC & I bring to bear. The question seems to deal with the conclusion of the kalam argument

3. Therefore, the universe has a cause.

Here the blogger is accepting kalam's conclusion that the universe has a cause, but is disputing the identity of the cause. So I think we want to explore that.

In Vilenkin's reply, I note that he is using the language of the General Theory of Relativity (GTR). He is talking about spacetimes, which means that we're taking seriously the existence of spacetimes as fundamental entities. So for now we'll talk GTR as well.

If spacetimes are disconnected, they have nothing to do with each other, which would include causal relations. It would be an interesting theological question to pursue whether God (WLC’s uncaused, changeless, timeless, extremely powerful, and immaterial Creator) is responsible for these other universes. But it wouldn't impact kalam's conclusion (or (2)). In that case the relevant definition of “universe” is Vilenkin's.

Let's be charitable to the opponent, however, and try to think of a case which he may be imagining. Now, space alone can become disconnected, as in the case of the Guth-Farhi process, whereby a mother space gives birth to a baby. A wormhole connects the mother and baby which is eventually “pinched off,” not unlike the way the umbilical cord is cut in a human birth. Here the mother is the cause of the baby. So a material “universe” is responsible for another “universe.” Eventually mother and baby become completely causally disconnected. But the full spacetime is still connected. Here I would define “universe” as the full spacetime including baby, mother, its mother, and so on.

Suppose we lived in a Guth-Farhi baby (which is the active ingredient of the Sean Carroll-Jennifer Chen model). How would we pursue the argument? In that case there is a very good thermodynamic argument, from Brett McInnes, that babies must get their thermodynamic arrow of time from their mother. If a mother universe is itself another's baby, it could not have obtained its arrow itself. So there is an ultimate mother universe to this arrangement, which McInnes calls “Eve.” We would then apply our usual arguments with respect to “Eve” to argue that the full spacetime has a transcendent origin. Or we could argue that a past eternal universe still leaves the question of the arrow unresolved. This would be similar to Stephen Hawking's criticism against past eternal universes; they all have arbitrary boundary conditions imposed on them “at” past infinity. In the case of the Carroll-Chen model, “Eve” is a De Sitter vacuum (a space with nothing in it except a certain type of dark energy). This is the paradigm case for the application of the Borde-Vilenkin-Guth theorem. It has a beginning to the finite past unless you suppose (as Carroll & Chen do) that the arrow of time eventually reverses.

Okay, so what if the arrow reverses? In that case WLC and I argue that the part of the spacetime with the reversed arrow is in no sense the past of the portion of the spacetime we live in. Rather, as with Victor Stenger's “bi-verse” proposal, you really have two universes with a common origin. If you try to argue that it is the past, you end up with a contradiction: it is both earlier and not earlier than our present universe. Alternatively (as Anthony Aguirre has suggested as a possibility regarding his own similar model), it may be that you have merely a redundant description of just one single-arrow spacetime after all.

If you leave GTR behind and go to a quantum gravity model (like Vilenkin's “tunneling from nothing” model), then what? In this case material reality springs from “nothing” into being. The question is, what does “nothing” mean? As far as the model is concerned, “nothing” is a “point.” It is a topology that has zero dimensions. We live in a 4-dimensional space that has come from this 0-dimensional “thing.” So what is it? Is it just a model? Or it supposed to represent something real in the true fact-of-the-matter way? The options seem to be something like:

1) It is just a model; it doesn't represent something real but is merely of instrumental value.

2) It represents physics inadequately trying to construe “nothing” in its proper sense of “absence of anything.” In this case, physics can’t help but construe “nothing” as some type of “thing.” Yet the moment you do that, you have “something.” (Lawrence Krauss wrote a whole book on the subject and still doesn't get it!)

3) It is a mathematical entity.

4) It is physical.

Now I tend to think (1) & (2) are adequate explanations of the meaning (or lack thereof) of the null topology. (3) fails because, as mathematician John Lennox has explained in multiple books in response to Stephen Hawking, mathematical entities are abstract universals, and such things have no causal powers. The “laws of quantum mechanics” are therefore not the cause of the universe. That leaves “quantum mechanics” as descriptive of a power that some concrete entity has. Now, as the blogger suggests, could (4) apply here? Could the cause of the universe be a physical thing?

Philosopher Quentin Smith took the null topology literally and argued for (4) in his paper, “Time Was Created by a Timeless Point.” I don't think the blogger was angling for Smith’s view, but it's best to take your opponent's case and improve it to see how good it can be.

Well, if a dynamic (or tensed) view of time is true, then (4) fails as an explanation. As WLC has argued, if the Creator is an impersonal thing, then the Creator and created must coexist together in timeless eternity. BTW, when folks such as Lawrence Krauss and Victor Stenger say that “nothing is unstable,” the only intelligible way to interpret their claim is through a dynamic process. So, to the degree they have thought the issue through, they seem to be endorsing a dynamic view of time.

I think that leaves an immaterial entity as the best explanation for kalam's cause. Notice I don't put that on the list. I think it is dangerous to identify an entity in a physical model with God. I think, for example, that Frank Tipler's honest attempts to square pure physicalism with Christianity fail and unintentionally lead to a pantheistic view of reality.

If the blogger has another example in mind (if he puts some meat to the claim), drop me a line and let me know about it.

Best regards,

Jim S.

- William Lane Craig