It is widely thought that our universe is highly unusual. It has certain features that make it lovely. Note that the term “lovely” is merely a term of art. It has no connotations beyond designating that our universe has certain features. These include (a) the fact that the universe is lawfully ordered; (b) the fact that the laws of nature have certain mathematical forms; (c) the values of the fundamental physical constants; (d) the fact that the universe starts in a low-entropy state. These lovely features are significant because if they were slightly different, then our universe would not contain any things that have value. The values include complexity, life, intelligence, rationality.

It is also widely thought that these lovely features require some explanation. Finding that you live in a universe with lovely features is far stranger than finding a watch on the heath. These features raise the Leibnizian question: why is the universe the way that it is? It is sometimes said that if our universe weren’t lovely, we wouldn’t be here to wonder about its loveliness. That counterfactual is a verson of the anthropic principle. Obviously, it’s true. And, what should be equally obvious, that truth does not in any way explain the existence or nature of our universe. The anthropic principle is not a hypothesis about the existence or nature of our universe. It is consistent with every hypothesis listed below.

1. The Brute Fact Hypothesis. The universe just exists. Our universe does not require an explanation. The loveliness is neither special nor rare nor significant in any way. There is no reason to provide any explanation for our universe.

Objections to the Brute Fact Hypothesis. We have regularly sought explanations for complex phenomena; and we have regularly found them. The success of the search for explanations both in science and in mathematics provides overwhelming evidence that the Brute Fact Hypothesis is wrong. So it is reasonable to search for an explanation for the existence and nature of our universe. Worse, the Brute Fact Hypothesis is itself not based on any evidence at all. It is an unwarrented hypothesis. It can be given for anything. God exists. It’s just a fact. Don’t ask why. This hypothesis celebrates ignorance.

2. The Necessity Hypothesis. The Necessity Hypothesis says that there is only one way that a universe can be. The lovely features of the universe are like variables in a big equation. There is only one solution to this equation. Since the lovely features are the only possible features, it is not surprising that our universe has them.

Objection to the Necessity Hypothesis: It is not the case that there is only one possible universe (so that if any universe is actualized, it must be that universe). On the contrary, there are many possible universes. And many of them can be actualized. Hence the Necessity Hypothesis is false.

3. The Plenitude Hypothesis. The Plenitude Hypothesis says that all possible universes are actual. This is the hypothesis of David Lewis. The library of possible universe blueprints is complete. For every way a universe can be, there is some blueprint in the library. So there is a blueprint in the library that describes our universe. Every blueprint in the library is actualized. There is no selection. So for every way a universe can be, there is some universe that actually is that way. Since our universe is one of the ways a universe can be, our universe is actual.

Objections to the Plenitude Hypothesis: The Plenitude Hypothesis seems to be unable to account for the regularity of our universe. Our universe exhibits regular patterns up to the present time – it’s like a novel that makes sense up to the current page. But in the library of all possible novels, there are infinitely many other novels that are like our novel up to the present page, but then diverge into random nonsense. So, the Plenitude Hypothesis seems to undermine one of the things it is supposed to explain: the regularity of the universe. All universes may indeed be possible; but there must be a selection of the ones that are actual.

4. The Lottery Hypothesis. The Lottery Hypothesis says that the lovely features of our universe are explained by random selection. There is a super-cosmic bucket filled with blueprints for universes. A lottery was somehow held and the winner was the blueprint for our universe. The winner was actualized.

Objections to the Lottery Hypothesis: There are at least two problems with the Lottery Hypothesis. The first problem is that universes with lovely features are highly rare. It is highly unlikely that the blueprint for our universe would be picked in a lottery. The second and much more damaging problem is that the lottery itself needs an explanation. What is its mechanism? Is it necessary? Was it designed? Did it evolve? The Lottery Hypothesis just pushes the mystery deeper. The Lottery Hypothesis is rejected.

5. The God Hypothesis. The God Hypothesis says that our universe was produced by God. The intelligence of God enables Him to find a lovely universe from among all the possible universes; the benevolence of God makes Him want to actualize a lovely universe; and the power of God enables Him to do it. Therefore, he does it.

Objections to the God Hypothesis: Suppose our universe does have a designer-creator (a DC). There is no reason to believe that the DC is personal or that it satisfies any of the definitions associated with the God of Abraham. The problem of evil makes it hard to say that the DC is maximally perfect. Any features of the DC that enable it to explain our universe are features that demand their own explanation. Since the DC requires its own explanation, it is not an ultimate necessary being. But God is supposed to be an ultimate necessary being. Our universe may indeed have a designer-creator, but it certainly does not match the features of the theistic deity. It is not God.

6. The Fecund Universe Hypothesis. The physicist Lee Smolin developed the idea that universes give birth to baby universes through black holes. When a star collapses to form a black hole in some parent universe, a baby universe pops out in some other dimension. This baby universe can give birth to its own babies. So there is an evolutionary tree of universes. Universes whose features are finely tuned for making black holes have more babies. So, generation after generation, the percentage of universes finely tuned for making black holes increases. But universes that are good at making black holes are also good at making stars, complex elements, planets, and life. So the loveliness of our universe is explained by this cosmic baby-making. The Fecund Universe Hypothesis is an evolutionary explanation for the existence and loveliness of our universe.

Objection to the Fecund Universe Hypothesis. This is a great idea. But it is a highly speculative idea – nobody really knows what goes on in black holes. So the hypothesis may well be false. And even if it is true, it faces problems. The first one is that the so-called “universes” really aren’t maximal physical wholes – they’re just parts of some bigger physical whole. And the bigger physical whole is the universe. And the hypothesis it isn’t ultimate. It depends on deeper laws of universe formation. Why are universes such that they can produce offspring? Why are black holes involved? What is the explanation for the ultimate laws of physics?

7. The Evolutionary Algorithm Hypothesis. This is a generalization of the Fecund Universe Hypothesis (and thus includes that hypothesis as a special case). Our universe is generated by a process of super-cosmic evolution. Super-cosmic evolution starts with some initial universe. This universe exists necessarily and does not depend on anything else for its existence. The initial universe produces some more lovely versions of itself. Once started, this process of universe-evolution is self-sustaining and self-amplifying. Each universe in any generation produces some more lovely successor universes. These successor universes populate the next generation of universes. The result is a series of generations of universes. Here’s the rule: for every universe, for every way to make it more lovely, there exists a successor universe that is more complex and congenial in that way. From generation to generation, the successor relation defines a growing tree of universes. As the tree grows, the universes in the tree become ever more lovely. Eventually, our universe appears.

Objections to the Evolutionary Algorithm Hypothesis. This is a highly speculative idea. As stated, it is also vague. It needs to be refined. And, since it is so abstract, it isn’t really even a hypothesis – it’s a type of hypothesis. More work must be done.

All the hypotheses in this list are speculative. They are all metaphysical, in the sense that none of them can be empirically tested. The hypotheses up to the God Hypothesis all suffer from very strong objections. The God Hypothesis and the Evolutionary Hypotheses (including the Fecund Universe Hypothesis) are based on analogies. This gives them some extra strength – they gain some weak empirical support from processes that we already know can produce things with lovely features (e.g. intelligent human design and biological evolution). However, the God Hypothesis faces inconsistencies. The Evolutionary Hypothesis is thus arguably the strongest hypothesis. Of course, it is really only a type of hypothesis – it needs to be worked out in detail and carefully studied.

Every question that used to be answered by appealing to God can be answered by appealing to some form of evolution.

I doubt that any theists would agree with that statement. And it’s worth stressing that biological evolution by natural selection is only one form of evolution. Obviously, some theists believe more or less in biological evolution by natural selection. But so what. As theists, they aren’t going to believe that every question that used to be answered by appealing to God can be answered by appealing to some form of evolution. That’s exactly the point that distinguishes evolutionary atheists from evolutionary theists.

An evolutionary atheist needs to posit all sorts of forms of evolution that are deeper than mere biological evolution. Biological evolution doesn’t answer any deeper questions about the structure or existence of the universe. It doesn’t tell you why the universe has the form that it has (why it has the laws it has, why its parameters are finely tuned in ways that allow life to emerge, etc.). And it doesn’t tell you why there is any universe at all. To answer questions about the universe, you’ll need to appeal to deeper forms of evolution. And cosmologists have discussed many deeper forms of evolution.

Evolutionary atheism is positive. It has content. It addresses the common complaint that the term atheism is highly negative. Of course, atheists do have beliefs (obviously, the belief that there is no God). But the content of that belief is negative. If you learn that somebody is an atheist, you don’t learn anything about their positive commitments. Atheism is always a derivative position. Perhaps it is derived from some positive doctrine (like materialism or Buddhism or religious naturalism or whatever). Or perhaps it is derived from mere reaction.

Some atheists may revel in being reactive and in arousing reactions – they may enjoy being provocative. I once went to an atheist meeting where everybody was mainly there just to be hostile towards religion (by which they pretty much all seemed to mean the conservative Christianity of the American religious right). I don’t see how hostility is ever helpful. I dislike hostility from atheists as much as I dislike it from religious fundamentalists. I have no interest in participating in a group whose primary purpose is the hatred of some other group. I’d prefer to build positive bridges and to expand the rational community of hope.

I suggested that atheists might better refer to themselves as “evolvers”. It’s a more positive word with genuine content. Some atheists might think this is trivial: why add another name? Who cares about mere labels? After all, look at what happened when some atheists tried to refer to themselves as “brights”. Nothing happened. Well, I’d say that’s because the term “brights” has no content whatsoever. But the term “evolvers” does. That said, I have little interest in pushing for some name. I’m advocating a proposition: every question that used to be answered by appealing to God can be answered by appealing to some form of evolution.

I am not such an agnostic because I do think that it is a matter of knowledge that there are no personal gods.  It is no harder for me to dismiss the existence of Yahwheh or a personal Allah or a divine Jesus as fictional or mythical than it is for all of us 21st Century modern people to dismiss out of hand claims that Thor or Zeus or Spider-Man or Aquaman exist.

But some atheists are reluctant to call themselves atheists because there remain the various god conceptions of the philosophers, some of which are, or may in the future be, plausibly formulated.  Might there be a single source of all being distinguishable from the known universe and fitting a certain, longstanding, traditional philosophical definition of the word “God”?  Perhaps.

This or a number of other concepts one might formulate are, or at least could be, sufficiently conceivable and sufficiently consistent with our best science and metaphysics as to be reasonably plausible.  One might for one reason or another disfavor belief in one or another of the possible metaphysical god concepts and yet still not think of one’s disbelief as raising to the level of knowledge.

I consider someone like this open to deism and agnostic about deist gods.  I myself fit this description.  But then why call myself an atheist if I am philosophically agnostic about whether some single “divine” principle which originates existence exists?  Should I not just call myself an agnostic?

No, I should not just call myself an agnostic.   The reason I am an atheist is because I am the opposite of a theist, not the opposite of a deist.  And I am not agnostic about theism, I am agnostic about deism.  I am gnostic about theism.  I know Yahweh and a divine Jesus and a personally construed Allah, etc. do not exist with the same kind of sufficient evidence and degree of certainty that I know Aphrodite and Green Lantern do not exist.

An impersonal metaphysical principle which somehow accounts for existence or order, etc. is an entirely different thing.   Proving such a thing exists would get you no closer to proving the existence of Yahweh or a divine Jesus than proving numbers are real entities would prove the real existence of The Count from Sesame Street.

Theists are quite fond of arguing for a deistic god and assuming that it is the same thing as providing evidence for a theistic one, when it’s not.  Or, a bit more modestly, other theists think that the inconclusiveness and relative philosophical open endedness of the deist god question translates into a plausibility of the theist god.

But it does not.

I do not feel like participating in the equivocations of theists by granting them that lack of knowledge about whether a deist god may exist translates to lack of knowledge about whether the personal gods upon which religions are actually based exist or not.

The believer in a personal god is a theist.  I do not believe in any personal gods, so I am an atheist.  My considered reasons for disbelief raise to the level of knowledge by our ordinary standards for knowledge claims.  Therefore, I am a gnostic atheist.  That the existence of some deistic, impersonal god may very plausibly be proved to me is irrelevant.  Even were I to come to be a deist by affirming the existence of a “divine” impersonal metaphysical principle of some sort, I would remain a gnostic atheist. The positions are entirely compatible.

Your Thoughts?

It sometimes surprises me, although it shouldn’t, how religious devotees feel the need to regularly reinforce their own convictions in groups of like-minded individuals. I suppose this is the purpose of regular Sunday church services, for example, to reinforce the community of belief in between the rest of the week when the real world may show no evidence of God, goodness, fairness, or purpose.

Nevertheless I was not prepared for the self-congratulatory hype that I have seen spouted on the web, and have received in emails, including a typically disingenuous email from Wiliam Lane Craig to his followers regarding a debate I had with him in North Carolina last week. While carrying out the debate in the first place was something that broke my normal rules–as I said during the debate, I far prefer civil conversation and discourse as a way of illuminating knowledge and reality–I will break another rule and write this blog-like note on my own perspectives, in the hope that it may circulate and counter some of the nonsense that has propagated in the fundamentalist and religious blogs of late. Perhaps Craig will post this on his blog and send it out as well.

I believe that if I erred at all, it was in an effort to consider the sensibilities of the 1200 smiling young faces in the audience, who earnestly came out, mostly to hear Craig, and to whom I decided to show undue respect. As I stressed at the time, I did not come to debate the existence of God, but rather to debate about evidence for the existence of God. I also wanted to demonstrate the need for nuance, to explain how these issues are far more complex than Craig, in his simplistic view of the world, makes them out to be. For this reason, as I figured I would change few minds I decided also to try and illustrate for these young minds the nature of science, with the hope that what they saw might cause them to think. Unfortunately any effort I made to show nuance and actually explain facts was systematically distorted in Craig’s continual effort to demonstrate how high school syllogisms apparently demonstrated definitive evidence for God.

Let me now comment, with the gloves off, on the disingenuous distortions, simplifications, and outright lies that I regard Craig as having spouted. I was very disappointed because I had heard that Craig was more of a philosopher than a proselytizer, but that was not evident the other evening.

Craig began with an attempt to demonstrate his scientific and mathematical credentials by writing a rather meaningless equation on this first slide, which he then argued would be the basis for his ‘evidence’. The equation, in words said that if the probability, given the data, gave one a greater than 50% likelihood for God’s existence, then this was evidence. He even presented this as a pseudo- Bayesian
Argument.

The problem is that using mathematical probabilities in this fashion ONLY makes sense if you have a well defined probability measure, and if one can check that the conclusions one draws are not sensitive to one’s priors. He did not explain this at all, nor do I think he understood it when I tried to explain it to him. For the rest of the evening Craig simply proceeded to spout his claimed evidence, and then proceeded to state that each gave him a greater than 50% belief in God. The whole purpose of the mathematical nonsense at the beginning was to give some kind of scientific credibility to a discussion which was anything but. It was disingenuous smoke and mirrors. (Moreover, as I tried to explain, in modern scientific experiments, merely finding an unexpected result, with say only a 20% chance of being wrong, is not sufficient to establish evidence. One needs to go to much higher levels of confidence, especially if the claim being made disagrees with all other evidence. It is hard to think of a grander claim than evidence for a divine being who creates the universe without apparent purpose, dominated by dark matter and dark energy and containing hundreds of billions of galaxies, lets it evolve untouched for billions of years, and then roughly a million years into human evolution decides to intervene at a time before Youtube or any other objective recording and archiving tool was available.)

Next, if one is going to frame the argument scientifically, as I argued is essential when discussing empirical evidence, which Craig later took great pains to disavow, one must point out that in science when one is trying to explain and predict data, one tries to explore all possible physical causes for some effect before resorting to the supernatural. Happily it is precisely this progress in our natural philosophy that ended such religious atrocities as the burning of witches. In each and every case the actual syllogism that one ended up with was:

Craig either doesn’t understand how something could happen, or instead believes that events happened that confirmed his pre-existing belief system.
In the absence of understanding physical causes or exploring alternatives, this implies evidence for the existence of God.
Therefore there is evidence that God exists.

This is what I framed as the “God of the Gaps” argument and I continue to view, upon reflection, most of the claims of Craig as falling in this well-known theological trap.

Let me work backwards through his 5 “arguments”:

The resurrection of Jesus, and that fact that the followers of Jesus were willing to die for their beliefs provides evidence of God: I admit that this claim is so sloppy and fatuous that in an effort to demonstrate some margin of respect for Craig I tried to avoid it for as long as I could. Craig argued that most New Testament scholars believe in the resurrection. Even if this were true, though Craig provided no evidence of this, this of course is simply proof that New Testament scholars have an a priori faith that guides them. It is like claiming that most Islamic scholars may believe that Mohammed actually ascended to heaven on a horse. In the first place, there are no definitive eyewitness accounts of these events, and in the case of the claimed resurrection the scriptures were written decades after the claimed event, and the different accounts are not even consistent. Not only are there serious theologians who doubt the resurrection, there are historians who doubt the historical existence of Jesus himself. Whatever one’s views in this regard, however, one must ask oneself the simple question: Is it more likely that all known physical laws were suspended so God could demonstrate divinity–and moreover demonstrate this in a hackneyed way that recreated previous resurrection myths, down to the number of days before being raised from the dead, of several previous, and now long-gone religious cults—or is it more likely that those who were preaching to convert fabricated a resurrection myth in order to convince those to whom they were preaching of Christ’s divinity? Finally, the remarkable, and completely trite claim that the fact the Christians were willing to die for their beliefs demonstrates the validity of these beliefs would be laughable, if it weren’t so pitiful. Especially, as I indicated during the event, in light of the fact that people were recently willing to fly planes into skyscrapers because of their beliefs in a religious framework that I know Craig has openly disavowed. Throughout history people have been willing to die for their beliefs, and it is often the beliefs one is willing to die for that are most suspect. Did Roman soldiers believe in Romulus and Remus. Did Viking warriers believe in Thor. Did Nazi soldiers believe in the superiority of the Aryan race. I found and still find Craig’s statement not only facile, and not even worthy of a high school debater, but I find the claim offensive.

FineTuning: The appearance of design is one of the most subtle and confusing aspects of our Universe. Charles Darwin, with his Origin of the Species, brilliantly and masterfully explained how the modern world, with its remarkable diversity of life forms may have the appearance of design without any design at all. It was one of the greatest and most striking scientific discoveries of all time, and it is the basis of modern biology and medicine, leading to countless other discoveries that have continued to save countless lives. Craig is aware, from his superficial reading of cosmology, of fine tuning problems in Cosmology, which he then immediately argued requires the existence of intelligent life, implying purpose to the universe. Not only does he fall prey to the same fallacy that those who, before Darwin enlightened us, ascribed design in biology fall prey to, he also continually misrepresented the nature of any apparent fine-tuning of quantities that we currently may not understand from first principles. I tried to explain to him that the current entropy of the universe is not fine tuned, nor need the initial entropy be fine tuned, because Inflation provides a mechanism to wipe out initial conditions and produce huge amounts of entropy, without God. I tried to explain to him that the Cosmological Constant, which is perhaps the most confusing finely tuned parameter we know of in the Universe, is fine tuned in a mathematical sense, compared to the naïve value we might expect on the basis of our current understanding of physical theory. While it is also true that if it were much larger, galaxies would not form, and therefore life forms that survive on solar power would not be likely to form with any significant abundance in the universe, I also explained that if the Cosmological Constant were in fact zero, which is what most theorists had predicted in advance, the conditions for life would be, if anything, more favorable, for the development and persistence of life in the cosmos. Finally, even if some parameters in our currently incomplete model of the universe do appear fine tuned for human life to be possible, (a) we have no idea if other values would allow other non-human-like intelligent life forms to evolve, since we have no understanding of the locus of all possible intelligent life forms. And, beyond this, just as bees are fine tuned to see the colors of flowers which they can pollinate as they go about their business does not indicate design, but rather natural selection, we currently have no idea if the conditions of our universe represent a kind of cosmic natural selection. If there are many universes, for example, as may be the case, and as are predicted in a variety of models, none of which were developed to address God issues, we would certainly expect to find ourselves only in those in which we can live. All of these are subtle and interesting issues worthy of discussion by knowledgeable and honest intellects. I found Craig to be lacking in both of the qualities during his discussion of this issue.

Absolute Morals: Craig argued that the existence of absolute morality gives evidence for God. Once again this is simple minded. Indeed in a meeting we convened at my Origins Project of distinguished philosophers and neuroscientists we debated the subtle issues of morality and human evolution, the possible variants of morality, and a host of other issues, without once ever resorting to God. As I tried to explain to Craig, paraphrasing fro Steven Pinker, if there were a God, either God would have the choice to determine what is right and wrong or not. But in this case, if God determined that raping and murdering 2 year-olds is morally acceptable would it be so? If not, as reason and experience suggests, then God really has to resort to other considerations, kindness, compassion, etc (except for the Old Testament God!), on which to base God’s decisions. But if that is the case, why not just dispense with the middle-man? Lastly, if there is evidence that God provides absolute Morality, it is missing from the world of our experience, where different religious groups, all of whom claim divine inspiration, have incompatible moral views, often leading to horrendous and violent acts against women and children, for example. Indeed, the Old Testament is full of such acts.

Contingency: Frankly the argument that humans or the universe do not have to exist but they do as providing evidence for God is something I find unfounded, so I will not devote any more words here to this subject. Many ‘contingent’ phenomena occur by natural causes, from earthquakes to snowflakes and I do not have to invoke God’s will to explain them. What applies to earthquakes and snowflakes applies to the Universe. Just because I cannot yet explain the origin of the Universe does not imply the existence of God…again God of the Gaps.

Our Universe had a beginning, therefore God must have created it: Actually the issue of the beginning of the Universe is the only truly interesting question worth discussing here. A host of scientific arguments need to be discussed here, and there is no doubt the question of chicken and egg is a vexing one for cosmologists as well as theologians. However, let me make a few points here: (1) All things that begin may have a cause, even if the cause is rather obscure and purposeless. However, what is important to note is that every known physical effect whose cause we understand has a physical cause. There is no reason therefore to assume the same will not be true of our universe itself. (2) There are no arguments that our universe need be unique and not derived from something pre-existing, or even eternal. Indeed, the Ekpyrotic Universe promoted by Turok and Steinhardt, which I don’t find compelling, argues for potentially eternal periods of expansion and contraction. Craig doesn’t understand the physics. (2) I continued to try and explain that quantum gravity may imply that space and time themselves are created at the moment of the big bang. This is a rather remarkable statement if true. But if it is true, in the absence of time itself, how one can ascribe arguments based on causality is unclear at best.

This last point illustrates what I tried hardest to explain. Classical human reason, defined in terms of common sense notions following from our own myopic experience of reality is not sufficient to discern the workings of the Universe. If time begins at the big bang, then we will have to re-explore what we mean by causality, just as the fact that electrons can be in two places at the same time doing two different things at the same time as long as we are not measuring them is completely nonsensical, but true, and has required rethinking what we mean by particles. Similar arguments by the way imply that we often need to rethink what we actually mean by ‘nothing’, from empty space, to the absence of space itself.

What I hoped I could convey to the truly open minded intellects in the audience, of which of course Craig was not one, was that the amazing effort to understand how the universe works reveals wonders far more remarkable than those presented by Bronze age myths, developed before we had any clear understanding of how the universe works. Simply arguing that one doesn’t understand the results, or doesn’t like the results and therefore one has to resort to supernatural explanations, which was the crux of Craig’s rather monotonous repetition of his syllogisms, is indeed intellectually lazy, as I did say at the time.

I have taken great effort to describe our actual understanding of the Universe and its implications for understanding how it might be possible for something to come from nothing, i.e. non-existence, in my new book, which will come out in January of 2012.

Your Thoughts?

According to Leibniz, God selects our universe for actualization because God knows that it is the best of all possible universes. But how does God know that? It’s not sufficient to say that God just knows it. God’s knowledge of the best universe requires an explanation. And Leibniz gives one: God runs a search algorithm. The search algorithm starts with a sorting algorithm. God compares possible universes with respect to goodness:

The infinity of possibles, however great it may be, is no greater than that of the wisdom of God, who knows all possibles. . . . The wisdom of God, not content with embracing all the possibles, penetrates them, compares them, weighs them one against the other, to estimate their degrees of perfection or imperfection, the strong and the weak, the good and the evil. It goes even beyond the finite combinations, it makes of them an infinity of infinities, that is to say, an infinity of possible sequences of the universe, each of which contains an infinity of creatures. By this means the divine Wisdom distributes all the possibles it had already contemplated separately, into so many universal systems which it further compares the one with the other. The result of all these comparisons and deliberations is the choice of the best from among all these possible systems. (Leibniz, Theodicy, sec. 225)

The output of this sorting algorithm is an ordered series of equivalence classes of universes. All universes in the same class are equally good. Leibniz uses an architectural metaphor to illustrate the output of the sorting algorithm. The totality of possible universes is like a library. The library is organized into levels. Each level is an equivalence class of possible universes. Higher levels of the library have better universes. Leibniz refers to the library as the Palace of the Fates. Leibniz describes this Palace in a story in which the goddess Athena takes a priest Theodorus for a tour of the mind of God:

You see here the Palace of the Fates . . . Here are representations not only of that which happens but also of all that which is possible. God, having surveyed them before the beginning of the actual universe, classified the possibilities into universes, and chose the best of all. . . . These possible universes are all here, that is, as ideas [in the mind of God] . . . . They went into other rooms, and always they saw new universes. The halls of the Palace rose in a pyramid, becoming even more beautiful as one mounted towards the apex, and representing more beautiful universes. Finally they reached the highest hall which was the most beautiful universe of all: for the pyramid had a beginning, but one could not see its end; it had an apex, but no base; it went on increasing to infinity. That is because among an endless number of possible universes, there is the best of all . . . but there is not any one which has not also less perfect universes below it: that is why the pyramid goes on descending to infinity. (Leibniz, Theodicy, secs. 414-417).

According to this Leibnizian story, the divine designer runs a search algorithm that involves two steps: first, it sorts the universes; second, it actualizes the universe that comes out on top in the sort. But if that’s all the divine designer does, then it isn’t God. All the design arguments show is that there is some computer running a trivial search algorithm. The computer is not a person; it isn’t conscious; it doesn’t have free will. It isn’t God. It isn’t even any of the Gods of the Philosophers. (Or, if it is, the term “God” is so trivial that it’s not worth arguing about, not worth worshiping, not really worth much of anything.) The designer is simply an abstract mathematical machine. It converges to a solution (to a final or halting state). The solution is the universe.

Design arguments are wonderful! They justify the existence of a platonic computer that is in no way divine. Atheists ought to use design arguments against theists. Why don’t they?

The theistic version of the Fine Tuning Argument goes like this: (1) The Fine Tuning Argument is sound. (2) If the Fine Tuning Argument is sound, then there is a Tuner. (3) Therefore, there is a Tuner. (4) If the Tuner is God, then God exists. (5) The Tuner is God. (6) Consequently, God exists.

The atheistic version of the Fine Tuning Argument goes like this: (1) The Fine Tuning Argument is sound. (2) If the Fine Tuning Argument is sound, then there is a Tuner. (3) Therefore, there is a Tuner. (4) If the Tuner is not God, then God does not exist. (5) The Tuner is not God. (6) Consequently, God does not exist.

The fourth premise in the atheistic version follows from the definition of God. If God is a tuner, then God is essentially a tuner. It’s not possible for God to exist and for God to fail to be a tuner. The theist, on learning that the Tuner is not God, can’t say that God’s busy doing other things or that fine-tuning the universe falls outside of God’s job description. On the contrary, it’s part of what it means to be God.

The fifth premise is what the atheist has to justify. The atheist has to prove that the Tuner is not God. This can’t be done by just blithely saying that the Tuner might not be God. It isn’t enough to just point to some other alternative possibility. Mere possibility is not what is asked for here. The atheist has to demonstrate that the Tuner is not God.

The way to do this is to present an alternative that is clearly a better explanation for the fine tuning than God. The argument goes like this: (1) The Tuner has a feature F if and only if F is required for tuning the universe (say, for life). (2) If the Tuner were God, then the Tuner would have additional features. (3) Therefore, the Tuner is not God.

Finely-tuning universes for life doesn’t seem to be very demanding. The assumption behind all the fine tuning arguments is that tuning involves setting the values of some finite number of numerical parameters. If there are n parameters, then their value ranges are the coordinate axes of an n-dimensional space. Each point in this space has the form (v1, . . . vn). The Fine Tuning Arguments always seem to assume that these values are plugged in to some fixed system of equations E. Thus E(v1, . . . vn) is the form of some possible universe. It’s usually also assumed that the equations apply to some initial conditions. They act as an operator on the initial conditions i. Any initial condition is a member of some set of initial conditions I. So the full form of a universe is (E(v1, . . . vn))(i). This form has to be tested for life. Here’s the algorithm:

    algorithm FineTuneForLife() [
        for every point (v1, . . . vn) in the space (P1, . . . Pn) do [
            for every i in the set of initial conditions I do [
                if (the form (E(v1, . . . vn))(i) permits life),
                    then make a universe with that form;]]]

Either there is some machine (some computer) that can run the algorithm FineTuneForLife, or else there is not. If there is no such machine, then perhaps the Tuner is God. If there is such a machine, then the Tuner is not God. The Tuner is merely a computer. More precisely, the Tuner is the simplest machine that can run the algorithm FineTuneForLife.

I doubt that any finite state machine can run FineTuneForLife. I’m not sure that any Turing machine can run FineTuneForLife. But Turing Machines are not very impressive; there are far more powerful transfinite computers. The outer limits of computation are said by some mathematicians to be equivalent to the constructible hierarchy of sets (also known as L). If there are models of our physics that fall within the constructible hierarchy of sets (also known as L), then there is a machine that can run FineTuneForLife. So the atheist has to defend this claim: there are models of our physics that lie in L. And I think that’s a claim that is very easy to defend.

It’s worth pointing out that FineTuneForLife is a simple brute force search. There are almost certainly very powerful ways to optimize that search. More optimal searches usually require less powerful computers.

It’s also worth pointing out that human beings already know how to finely tune toy universes for artificial life. Small numbers of human beings, working for only a few decades, have already produced impressive software universes that contain life-like structures. So, how hard can it be to run something like FineTuneForLife? Does it really require an intellect whose power transcends that of every logically possible machine? The successes of artificial life suggest that fine tuning isn’t all that hard. If it can be done by a machine, then God does not exist.

The Effectiveness Argument goes like this: (1) Mathematics is effective in science. (2) The best explanation for this effectiveness is that physical reality is grounded in mathematical reality. (3) So, by inference to the best explanation, all physical reality, including our universe, is grounded in mathematical reality – in pure mathematics.

The second premise in the Effectiveness Argument is supported by a variety of writers. Dipert (1997: 332) argues that “the very possibility of a clear understanding of the world requires the possibility that it is a simple mathematical structure”. Steiner (1998: 4 – 5) puts it even more powerfully like this:

The strategy physicists pursued . . . to guess at the laws of nature, was a Pythagorean strategy: they used the relations between the structures and even the notations of mathematics to frame analogies and guess according to those analogies. The strategy succeeded. . . . The success of the Pythagorean strategy might lead the reader to conceptual Pythagoreanism, the view that the ultimate properties or ‘real essences’ of things are none other than the mathematical structures and their relations. More radically, one might adopt metaphysical Pythagoreanism, which simply identifies the Universe or the things in it with mathematical objects or structures. (Some physicists write as though an elementary particle just ‘is’ an irreducible group representation, or even that the entire universe is.)

Steiner (1998: ch. 4) brilliantly discusses many examples in which the pythagorean strategy of identifying physical things with mathematical things is successful. His cases include: Maxwell’s study of electromagnetism; Schroedinger’s study of wave mechanics; Dirac’s study of the positron; Schwarzschild’s solution for the equations of general relativity (i.e. black holes); Heisenberg’s study of the symmetries of nucleons; Kemmer’s study of pions; Gell-Mann’s and Ne’eman’s study of particle systems with unitary spin and the consequent discovery of quarks; Einstein’s inference of the field equations for general relativity; the Heisenberg-Born-Jordan derivation of matrix mechanics; Schroedinger’s derivation of the Klein-Gordon equation; the derivation of the Yang-Mills equation; the study of analytic continuations in crossing symmetries.

As a continuation of Steiner’s reasoning, Tegmark (1998: 44) says: “the usefulness of mathematics for describing the physical world is a natural consequence of the fact that the latter is a mathematical structure.” Accordingly, Tegmark (1998: 46-47) simply collapses the distinction between mathematical and physical existence:

One might say that wherever there is light, there are associated ripples in the electromagnetic field. But the modern view is that light is the ripples. One might say that wherever there is matter, there are associated ripples in the metric known as curvature. But Eddington’s view is that matter is the ripples. One might say that wherever there is physical existence, there is an associated mathematical structure. But according to our TOE [theory of everything], physical existence is mathematical existence. (The italics are Tegmark’s.)

Dipert, R. (1997) The mathematical structure of the world: The world as graph. Journal of Philosophy 94 (7), 329-358.

Steiner, M. (1998) The Applicability of Mathematics as a Philosophical Problem. Cambridge, MA: Harvard University Press.

Tegmark, M. (1998) Is ‘the Theory of Everything’ merely the ultimate ensemble theory? Annals of Physics 270, 1-51.

Wigner, E. (1960) The unreasonable effectiveness of mathematics in the natural sciences. Communications on Pure and Applied Mathematics 13, 1-14.

These first cause arguments are debates about the structure of contingent physical existence. It’s fun to reason backwards in time along causal chains, but that reasoning remains entirely within the system of contingent physical things. The deeper questions are these: Why is there a universe rather than no universe? Why are there any physical things rather than no physical things? Why are there any contingent things rather than no contingent things? No first cause argument (or atheistic reply) even tries to answer those questions.

Aquinas’s Third Way and Leibniz’s Sufficient Reason Argument are much deeper arguments. Theists and atheists both ought to study them carefully. Leibniz’s Sufficient Reason Argument is especially interesting. Here it is:

(1) Neither in any single thing, nor in the total aggregate and series of things, can the sufficient reason for their existence be discovered. (2) Let us suppose a book entitled The Elements of Geometry to have existed eternally, one edition having always been copied from the preceding. (3) Although you can account for the present copy by a reference to the past copy which it reproduces, yet, however far back you go in this series of reproductions, you can never arrive at a complete explanation; (4) You always will have to ask why at all times these books have existed, that is, why there have been any books at all and why this book in particular. (5) What is true concerning these books is equally true concerning the diverse states of the universe, for here too the following state is in some way a copy of the preceding one (although changing according to certain laws). (6) However far you turn back to antecedent states, you will never discover in any or all of these states the full reason why there is a universe rather than no universe, nor why it is such as it is. (7) You may well suppose the universe to be eternal; yet what you thus posit is nothing but the succession of its states, and you will not find the sufficient reason in any one of them, nor will you get any nearer to accounting rationally for the universe by taking any number of them together; (8) The reason must therefore be sought elsewhere. (9) Things eternal may have no cause of existence, yet a reason for their existence must be conceived. . . . (10) Hence it is evident that even by supposing the universe to be eternal, the recourse to an ultimate reason for the universe beyond this universe . . . cannot be avoided. (11) The reasons for the universe are therefore concealed in some entity not in the universe, which is different from the chain or series of things, the aggregate of which constitutes the universe. (Leibniz, 1697)

As it stands, this argument has some well-known problems (it isn’t really even an argument, it’s just a proto-argument). But the argument can be rebuilt in ways that make it incredibly strong. Rebuilding it is mostly tedious logic. There’s no need to do that here. I’m going to assume that some rebuilt version of the argument is sound. What’s most interesting about this argument is what it says about existence.

Leibniz says the universe is the totality of physical things. It’s a spatially, temporally, and causally closed system. The entire universe is contingent – it might exist but it might not exist. Why does it exist? Why is there a universe at all? You can’t answer that question by appealing to anything that is internal to the universe. You can’t answer that question by appealing to any entity that participates in any spatial, temporal, or causal relations. This isn’t an inference back in time to a first cause. If there is a first cause, then it falls within the scope of the question. If there’s a first cause, it’s just another part of the universe – and thus it needs to be explained. And Leibniz is perfectly happy to say that the universe has always existed – no first cause at all. Leibniz says that “the reasons for the universe are concealed in some entity not in the universe”. Call this entity the ground.

Contrary to theists, the ground isn’t any concrete god. It isn’t the god of the Old Testament. It isn’t the creator of Genesis. It isn’t Yahweh or El-Elyon. And it isn’t any of the gods that have appeared in any of the mythologies of old paganisms. It isn’t Zeus or Thor. All those old gods are just concrete physical things – they participate in spatial, temporal, and causal relations. And since creation seems to entail causing an effect at some time, the ground isn’t a creator at all. For the theists, it just gets worse. Since the ground doesn’t participate in spatial, temporal, or causal relations, it can’t be a person. The ground doesn’t have any psychology. The ground doesn’t perceive the universe or intervene in it. The doesn’t have any thoughts, no beliefs, no desires. And the ground isn’t the god of deism. After all, that god is a first cause. The ground is deeper than all those gods.

What about the gods of the philosophers? Well, the ground exists. So it can’t be Plato’s form of the good; it isn’t the One of Plotinus. All those old philosophical gods are somehow beyond existence. And the ground isn’t Tillich’s ground of all being; on the contrary, it’s just the ground of the physicality of our universe. What about Spinoza’s god? I have to confess that I don’t entirely understand what that god is supposed to be – which makes me doubt that it’s Spinoza’s god. Anyway, the argument from evil entails that the ground certainly isn’t all-powerful and all-good and all-knowing. So the ground can’t be the big 3O god of classical theism. Leibniz’s argument doesn’t seem to support theism at all.

Onwards, then, to the atheists. Assuming that the ground isn’t one of those old-fashioned religious or theological entities, what would it be? Well, the ground isn’t any physical thing or structure or event. The ground isn’t the big bang or the cause of the big bang. It isn’t space-time or some quantum field or some black hole or any other exotic physical thing. It isn’t any physical thing at all. It’s important to understand the scope of this assertion.

It may very well be true that our observable cosmos, including everything that we can measure or empirically detect, is a simulation running on some alien super-computer. But if that’s true, then our observable cosmos isn’t the universe – it’s just the part of the universe that we can observe. The whole universe is a much bigger place. If our universe is running on some alien super-computer, then the Leibnizian question applies to that super-computer and to the aliens that made it. Why do those contingent physical things exist? The ground isn’t the super-computer or the alien civilization. The ground explains the aliens and their artifacts. Perhaps our universe contains many smaller cosmic domains (as in inflationary cosmology, or Smolin’s fecund universe hypothesis). If it does, then that entire multiverse is a contingent thing. Why is there a multiverse rather than no multiverse? The multiverse needs to be explained. If our universe is a big foam composed of lots of cosmic bubbles, then the ground explains that foam.

Given all this metaphysics, here’s the test question: Should atheists affirm or deny the existence of the ground?

Leibniz, G. W. (1697) On the Radical Origination of the Universe. In P. Schrecker & A. M. Schrecker (Trans.) (1988) Leibniz: Monadology and Other Philosophical Essays. New York: Macmillan, 84-86. The translation is slightly edited for consistency.

The theists say: (1) If the reasoning in the cosmological argument is correct, then God exists. (2) The reasoning in the cosmological argument is correct. (3) Therefore, the argument proves that God exists.

The atheists say: (1) If the reasoning in the cosmological argument is correct, then God exists. (2) The reasoning in the cosmological argument is not correct. (3) Therefore, the argument fails to prove that God exists.

Atheists and theists both agree on the major premise: if the reasoning in the cosmological argument is correct, then God exists. Why the agreement? Why grant that the cosmological argument is an argument for God? Sometimes atheists do point out that it might not be an argument for God – it might be an argument for something else. But I’ve never seen that possibility seriously explored. And it’s too bad.

On the one hand, atheists can attack theism by showing that the classical arguments for God are logically flawed. On the other hand, atheists can attack theism by showing that those very same arguments are arguments for things that are not God. Which attack is deeper?

I think it’s clear that the second line of attack is much deeper – it’s much, much more threatening. When your enemies attack your arguments, well, you can always deal with that. But when your own arguments turn against you, you’re in big trouble.

So I’m going to encourage atheists to look at the classical arguments to see what else they might be used for. Fix them up, make them all shiny, and use them to drive to some new place. For an illustration, stay tuned . . .

Your Thoughts?