PvI, Science, and Natural Theology
April 5, 2010 — 17:10

Author: Andrew Moon  Category: Existence of God Problem of Evil  Comments: 21

I’ve been slowly working through Peter van Inwagen’s The Problem of Evil, and I gave pause at the following remark:

I believe that science has made only two contributions to the data of natural theology. The discovery of this fact [that animals have been suffering long before the arrival of humans] is one of them; the other is the discovery that the physical world does not have an infinite past. (112)

One could question PvI’s claims both about what have been contributions to natural theology and what have not. Regarding the former, I didn’t think the finitude of the physical world was an established scientific discovery.
Regarding the latter, I wonder whether there are counterexamples to PvI’s claim. Maybe we can make a list and consider why he doesn’t think they are contributions to natural theology. Given my respect for PvI, the fact that he doesn’t think X is a contribution to natural theology gives me strong reason to think it isn’t.
To start off the list, there has been the discovery that the conditions of the universe shortly after the big bang are finely-tuned so that life can exist; the probability that those conditions would be life-permitting rather than life-prohibiting is very small. Surely, this is a contribution to natural theology? I can’t see why he would think they weren’t.
Any other examples?

Comments:
  • Ted Poston

    Hi Andrew,
    Re finitude of the universe: The recent Borde Guth Vilenkin theorem establishes that any universe must have a past space-time boundary. Here’s a quote from Vilenkin:
    “It is said that an argument is what convinces reasonable men and a proof is what it takes to convince even an unreasonable man. With the proof now in place, cosmologists can no longer hide behind the possibility of a past-eternal universe. There is no escape, they have to face the problem of a cosmic beginning” (Many Worlds in One [New York: Hill and Wang, 2006], p.176).
    Bill Craig talks about this in his new version of Reasonable Faith and here: http://www.reasonablefaith.org/site/News2?page=NewsArticle&id=6115
    Also, re the PvI quote: he talks about contributions to the data of natural theology. I don’t think the FTA “data” is up to the same high level as the two that PvI mentions.

    April 6, 2010 — 8:04
  • Andrew Moon

    Hmm, not that I understand all that, but thanks for the links on the past finitude of the universe. =)
    Why don’t you think the Fine-tuning data is as strong as the evidence for the past finitude of the universe? I thought the fine-tuning data was well agreed upon.

    April 6, 2010 — 10:12
  • Let me give another shot at a scientific discovery that makes a difference to natural theology. I forget who it was (Ayer? Dawkins?) who said that Darwin made it possible for him to be an intellectually fulfilled atheist. Without the knowledge of contemporary evolutionary biology, an argument from biological design seems very powerful. Contemporary evolutionary biology significantly weakens the biological design argument, doesn’t it?

    April 6, 2010 — 10:14
  • I think it’s an overstatement to say that there must be a past boundary by the Borde, Guth and Vilenkin work. In their abstract, they only say: “Here we offer a simple kinematical argument, requiring no energy condition, that a cosmological model which is inflating — or just expanding sufficiently fast — must be incomplete in null and timelike past directions. Specifically, we obtain a bound on the integral of the Hubble parameter over a past-directed timelike or null geodesic. Thus inflationary models require physics other than inflation to describe the past boundary of the inflating region of spacetime.” http://arxiv.org/abs/gr-qc/0110012
    The last sentence does rather qualify the claim.
    Moreover, I suspect that the result is compatible with the hypothesis that our universe arose from an earlier universe, and so on.

    April 6, 2010 — 11:26
  • Ted Poston

    Alex,
    My knowledge of the BGV theorem comes through Bill Craig. Craig explicitly says their result shows that even a Multiverse must have had a beginning. I don’t have the book with me but it’s in his new edition of RF in his discussion of the kalam.
    Andrew,
    I don’t think the claim that *the probability of the FTA data is extremely low* is itself as epistemically secure as the two pieces of data PvI mentions. Why? Short answer: observation selection effects.

    April 6, 2010 — 12:11
  • I think the claim of past finitude depends on the exact referent of the phrase ‘the physical world.’ The Big Bang Theory implies that our universe has only a finite past. Some highly speculative work in physics has suggested that our universe came from another universe, or something like that, and you might use the term ‘the physical world’ broadly enough to refer to include the parent universe, or you might not. Even if the source universe is included, it is not clear that that gets you an infinite past, since time (or, at least, our time) begins at the Big Bang. So there is good reason for saying that physicists have discovered that the physical world does not have an infinite past. It is possible to wriggle out of the conclusion, but only with significant ingenuity, and at the expense of construing the words in a questionable way and incurring some odd physical and metaphysical commitments.

    April 6, 2010 — 15:15
  • Alex and Ted,
    I don’t know, Ted’s quote seems to make it really clear that they think that science shows that the universe does not have an infinite past. I didn’t understand Alex’s quote, but if it does say that they think science leaves open there being an infinite past (or it somehow implies this), then it looks like Vilenkin is contradicting himself.
    Ted,
    How do observation selection effects make the FTA data less epistemically secure? I’m not sure what sort of observation selection effects you are talking about.

    April 6, 2010 — 16:09
  • Mike Almeida

    I don’t think the claim that *the probability of the FTA data is extremely low* is itself as epistemically secure as the two pieces of data PvI mentions. Why? Short answer: observation selection effects.
    What’s the selection effect that’s distorting our assessment of the probabilities? To get a selection effect you’d have to have the observation consistent with both a high and low probability assignments. For instance, the observation that there is life on Earth consistent with both a high probability that earthy planets give rise to life and a low probability that they do. And that’s certainly true. But how could the probablity of life on the noted parameters be either high or low (though we mistakenly believe it’s low)? Is the effect that the probability that there’s life in some universe or other under the given parameters is high, even though the probability that it arises in any particular universe of this sort is low?

    April 6, 2010 — 16:41
  • Mathis

    You know, I’d rather not rely on Craig when it comes to cosmology. I’d rather rely on James Sinclair, who was his co-author on the chapter on the Kalam argument in the Blackwell Companion To Natural Theology (he wrote about the scientific considerations) – because Sinclair actually is a cosmologist.
    From the BCtNT, page 141
    Theorists wondered whether this process could be infi nitely extended into the past. Interestingly, Guth himself, along with collaborators Alexander Vilenkin and Arvind Borde, has likely closed the door on that possibility. In 2003, Borde, Guth, and Vilenkin published an updated singularity theorem far grander in scope than the Hawking–Penrose theorems.
    They explain,
    — Our argument shows that null and time-like geodesics are, in general, past-incomplete in infl ationary models, whether or not energy conditions hold, provided only that the averaged expansion condition Hav > 0 holds along these past-directed geodesics. (Borde, Guth, & Vilenkin 2003, p. 3)
    A remarkable thing about this theorem is its sweeping generality. We made no assumptions about the material content of the universe. We did not even assume that gravity is described by Einstein’s equations. So, if Einstein’s gravity requires some modifi cation, our conclusion will still hold. The only assumption that we made was that the expansion rate of the universe never gets below some nonzero value, no matter how small. This assumption should certainly be satisfi ed in the infl ating false vacuum. The conclusion is that past-eternal inflation without a beginning is impossible. (Vilenkin 2006, p. 175) —
    Vilenkin affirms that any universe (including universes modeled by higher dimensional cosmology, pre–Big Bang cosmology, and so forth,) which, on average, expands has to connect, in a fi nite time, to a past boundary (pers. comm., March 4, 2004).

    He goes on to name 4 models that try to escape this and argues against them.

    April 6, 2010 — 18:04
  • I guess I’m still not clear on the implications of all this. Grant that every possible universe must be temporally finite. What is the empirical evidence that makes it even a tad less likely than not that there is a beginningless series of such temporally finite universes?

    April 7, 2010 — 5:09
  • Ted Poston

    The article that Craig cites in his discussion is “Inflation is not Past Eternal.” As the quotes Mathis provide and the quote I gave from Vilenkin evince this result closes the door to eternal inflationary models. Craig talks briefly about a quantum gravity model that introduces imaginary numbers for the time variable and evidently that avoids an initial singularity but Craig argues that there’s no realist interpretation of this model.
    Re the observation selection effect: the effect is that the only kinds of universes we’d observe are life permitting ones. This weakens our epistemic position with respect to the probability claim because we can’t rule out there’s some hidden variable or other oddity that significantly constrains the values of the constants in the laws (ditto for the nature of the initial conditions). I still think the FTA constitutes a good reason to raise the probability of theism but I don’t think it’s right to consider the probability claim a piece of data. Kosh?

    April 7, 2010 — 8:20
  • “Some highly speculative work in physics has suggested that our universe came from another universe, or something like that, and you might use the term ‘the physical world’ broadly enough to refer to include the parent universe, or you might not. Even if the source universe is included, it is not clear that that gets you an infinite past, since time (or, at least, our time) begins at the Big Bang.”
    True. However, presumably the main relevance to natural theology of the claim that our past is finite is that there is no unbounded backwards infinite chain of causes. But if our universe came from another universe, and so on ad infinitum, then even if each universe has its own time, we still get an unbounded backwards infinite chain of causes, and that is what is relevant.

    April 7, 2010 — 8:31
  • Mike Almeida

    . . .the observation selection effect: the effect is that the only kinds of universes we’d observe are life permitting ones. This weakens our epistemic position with respect to the probability claim because we can’t rule out there’s some hidden variable or other oddity that significantly constrains the values of the constants in the laws
    Sure, the only universes we’d observe are life-permitting universes. Now that might affect my assessment of the chances of there being living things only if I’ve overestimating the kinds of universes that are possble. Suppose I’m mistaken in thinking that any universe with different initial conditions is so much as possible. In that case the chances of life in some universe of other seems reasonably high. I shouldn’t be surprised that there’s life. But it seems evident that there are many more possible universes whose initial conditions are not life-permitting than universes whose conditions are life-permitting. Suppose we agree on that (unless I’m now begging the question). Then it does seem a bit surprising (not that improbablity always elicits surprise or should) that a life-permitting universe is actual.

    April 7, 2010 — 9:23
  • Ted Poston

    Hi Mike,
    I think that it’s conceivable that the mathematical expressions of the laws have different values for the constants. Simple example: Let it be a law that y=Fx+2. Clearly, we can conceive that the expression may be y=Fx+4. But I think the OSE *weakens* move from this to the claim that its robustly possible that the constants for the laws have many different values. Again, my main point here is not that this isn’t a reasonable move but that it’s not evident, not data that science provides.

    April 7, 2010 — 12:39
  • Presumably PvI is including data which favors atheological arguments in natural theology; ancient animal pain doesn’t help the tradition view of of the Fall, for example.
    I would say that genetic (and to a lesser extent environmental) contribution to human behavior, as well as general cognitive science, challenges at least some positions natural theology might hope to support, e.g. the existence of human souls or freedom of the will. Certainly, many atheologians purport to have counter-evidence along these lines. Do you folks think these fields present challenges to natural theology?

    April 7, 2010 — 13:15
  • Mike Almeida

    Clearly, we can conceive that the expression may be y=Fx+4. But I think the OSE *weakens* move from this to the claim that its robustly possible that the constants for the laws have many different values. Again, my main point here is not that this isn’t a reasonable move but that it’s not evident, not data that science provides.
    The OSE affects (or might affect) our modal knowledge? I wonder why that would be. Does it affect modal knowledge generlly, or just modal knowledge relevant to this issue?

    April 7, 2010 — 17:15
  • BlakeG

    Ted,
    “observation selection effect”:
    — To my knowledge, the observation selection effect (or the weak anthropic principle) is only going to work if one presupposes the relevant kind of multiverse. That being said, if an atheist invokes a multiverse simply to stave of the unlikelihood of the fine-tuning, he’s committing the inverse gamblers fallacy (e.g. If I roll double six’s on my first roll, I shouldn’t say “ah, well, these dice must have been rolled many times before” or “ah, well, there are people rolling dice in adjacent rooms”.)
    “some hidden variable or other oddity that significantly constrains the values of the constants in the laws”:
    1. So? Even if there were some meta-law that constrains the variables, that just pushes the unlikelyhood (on atheism) up a level. No better than saying “yeah, so what if a human, fully clothed with a watch on, materializes in New York from a collision of molecules? There was a physical law that mandated such a thing happen!” That hardly takes away the epistemic surprise on any non-design hypothesis. So in the words of the prominent Bernard Carr & Martin Rees “even if all apparently anthropic coincidences could be explained [in terms of some grand unified theory], it would still be remarkable that the relationships dictated by physical theory happened also to be those propitious for life”
    2. It’s unlikely that there is such a meta law (that is going to explain the constants, force values, and initial conditions all at once). Paul Davies (prominent physicist; been a prof. at 6 Universities [Cambridge, London etc.]): “there is nothing in present ideas about laws of initial conditions, remotely to suggest that their consistency with the laws of physics would imply uniqueness, far from it, it seems then that the physical universe does not have to be the way it is, it could have been otherwise.”

    April 7, 2010 — 22:20
  • It’s looking like I may have been wrong about the result only showing OUR universe to have a finite past.

    April 8, 2010 — 19:14
  • Ted Poston

    Alex: Thanks for the update.
    Mike: The OSE affects (or might affect) our modal knowledge? I wonder why that would be. Does it affect modal knowledge generlly, or just modal knowledge relevant to this issue?
    One quick clarification: The probability claim in the FTA is defined over an outcome space and the nature of the outcome space isn’t to be identified with conceptual space. Rather the outcome space is similar to physical possibilities, though the standard def of physical possibility doesn’t give the right space at issue. Re your question: I think OSE affects modal knowledge generally. Suppose we become convinced that there’s an OSE for our ability to conceive of possibilities. Then that undermines our knowledge of the nature of modal space. Similarly for the FTA data: OSE is an undermining defeater for our ability to conceive of the nature of the (relevant) physical-possibility space.
    Blake: I agree with most of what you say. It’d just enter a mild qualification regarding the notion of presupposition. I’m not sure what you mean by ‘presupposing’, but I don’t think the OSE requires a premise to the effect that there is multiverse. As I understand it, the OSE is fundamentally an undermining defeater regarding our ability to accurate envision the relevant modal space.

    April 9, 2010 — 8:08
  • Mathis

    I think that it’s conceivable that the mathematical expressions of the laws have different values for the constants. Simple example: Let it be a law that y=Fx+2. Clearly, we can conceive that the expression may be y=Fx+4. But I think the OSE *weakens* move from this to the claim that its robustly possible that the constants for the laws have many different values. Again, my main point here is not that this isn’t a reasonable move but that it’s not evident, not data that science provides.
    Well, you also have string theory, which is at least mathematically consistent and it allows to, if I recall correctly, 10^500 different sets of constants.
    I see no reason for thinking a mathematically consistent theory should be metaphysically impossible.

    April 11, 2010 — 15:54
  • Robert Gressis

    I certainly think that if science showed that we don’t have libertarian free will as conclusively as it showed that animals were around and felt pain before people were around, then this would be an interesting result for natural theology. This would matter because it may rule out some or all free will defenses, which would change the complexion of the PoE debate.
    I don’t know how much it would matter if we disproved the existence of immaterial souls; there could still be an afterlife, after all, although perhaps some find the possibility of the resurrection of the physical body to be less plausible than the continuation of an immortal soul.

    April 13, 2010 — 18:58