William Lane Craig on facts, tracts, and things abstract

I’m grateful to Alex Rosenberg and William Lane Craig for taking the time to respond to my post, “Fact-checking the Craig-Rosenberg debate“. I edited in a few of Rosenberg’s comments from correspondence, but Craig’s public reply, “Fact-checking the fact-checker“, is more in-depth, and deserves a response in its own right. I’ll single out two points for special attention: historical methodology, and the idea of immaterial causation.

___________________________________________________________________

Scripture and scholarship

Craig writes of my

[…] breezy dismissal of N. T. Wright’s scholarly work because Wright is “a Christian apologist and bishop” and of the work of New Testament historians in general because they are allegedly Christians […]

I didn’t dismiss Christian scholarship. What I wrote was:

Craig doesn’t note that most New Testament scholars are Christians. (Are we to take it as evidence for the truth of Christianity that a lot of Christians happen to be Christian?)

Now, of course being a Christian doesn’t make it impossible for you to evaluate Christianity in a fair and skeptical way. I believe very strongly that the Earth is round, but that doesn’t mean that I’d be hopelessly biased in a debate with flat-Earthers. Agnosticism does not imply objectivity, and objectivity does not imply agnosticism. If anything, we’d be worried if most New Testament scholars weren’t Christians, since that would suggest that the historical evidence tended to make people less religious than the general populace.

But it’s also worth noting that Christian orthodoxy is not generally considered by historians the only possible objective interpretation of the evidence of the Gospels. And appealing to scholarly consensus here is misleading inasmuch as it has the guise of an appeal to independent authorities, as opposed to authorities who already came into the field accepting Christianity.

The charge was not that being Christian invalidates one’s scholarly work on Christianity. It was that, in the context of a debate with non-theists, it’s misleading to appeal to the authority of historians qua historians without mentioning that most of them came into the field already accepting the conclusion for which you’re arguing. (From childhood, no less!)

Suppose you’re debating a Muslim theologian who asserts that we can be confident that Muhammad is a prophet because virtually all Qur’anic scholars accept historical claims that provide powerful inductive evidence for Muhammad’s lofty status. If in the process he does not mention that most Qur’anic scholars are (and always have been) committed Muslims, then his argument risks deceiving people into thinking he’s adducing wholly independent grounds for accepting Islam. That’s so whether or not you ‘breezily dismiss’ Qur’anic studies itself.

If Craig’s point had merely been ‘There are a lot of very smart Christians who have carefully studied Christianity and still believe in it,’ I would have had no objection. Likewise, I have no objection to citing the specific historical arguments of Christian scholars, which can then be evaluated in their own right, without any need to consider the personal beliefs of the arguer. But when you’re citing the people themselves as authorities, their religious precommitments do start to become relevant, in the cases of Christian and non-Christian religions alike.

CraigHe thereby displays his unfamiliarity with New Testament studies and with the skepticism with which these scholars — which include among their ranks non-theists like Bart Ehrman and Jewish scholars like Geza Vermes who concur with my three facts — approach their sources.

I never suggested that all New Testament scholars are Christian. But Craig is doing what I wanted him to do in the debate, which is citing non-Christian authorities to strengthen his case — so I thank him for that.

That said, I should note that Craig is mistaken about Ehrman. Ehrman did claim that Jesus’ empty tomb was a historical fact in a 2003 lecture, but in a 2006 debate — a debate with Craigavailable on Craig’s site — Ehrman said that he had changed his mind. Quoth Ehrman:

Paul said he [Jesus] got buried; he may simply have been tossed into a communal grave. I should point out that in some of Bill’s writings, he’s quoted a lot of my writings, and he’s taken them out of context, as I’ll show in a few minutes, because what he’s saying I’ve changed my mind to, I don’t agree with. […]

We don’t know if Jesus was buried by Joseph of Arimathea. What we have are Gospel stories written decades later by people who had heard stories in circulation, and it’s not hard at all to imagine somebody coming up with the story. We don’t know if his tomb was empty three days later. We don’t know if he was physically seen by his followers afterwards.

And Craig recognized this during their debate, saying,

Insofar as Dr. Ehrman now chooses to deny the honorable burial, the empty tomb, the appearances, he is in the decided minority of New Testament scholarship with regard to those facts.

We should keep in mind that Ehrman doesn’t deny “the appearances“, provided that dreams or visions would qualify as “appearances“. But in any case, Ehrman tells me he’ll give more details (and explain why he changed his mind) in his upcoming book, How Jesus Became God.
 

 
There are a number of further ambiguities that led to my charge of “misleading”. To keep Craig’s claims in context, I’ll quote much of his argument from the debate, adding numbers where I have questions or comments below.

CraigGod is the best explanation of the historical facts about Jesus of Nazareth.[1] Historians have reached something of a consensus that Jesus came on the scene with an unprecedented sense of divine authority, the authority to stand and speak in God’s place.[2] He claimed that in himself the Kingdom of God had come. And as visible demonstrations of this fact, he carried out a ministry of miracle-working and exorcism. But the supreme confirmation of his claim was his resurrection from the dead.[3] If Jesus did rise from the dead, then it would seem that we have a divine miracle on our hands, and thus evidence for the existence of God.[4] Now, I realize most people think that the resurrection of Jesus is just something you accept — by faith, or not. But there are actually three facts recognized by the majority of historians today which I believe are best explained by the resurrection of Jesus.[5] […] Naturalistic[6] attempts to explain away these three great facts, like “the disciples stole the body” or “Jesus wasn’t really dead,” have been universally[7] rejected[8] by contemporary scholarship.[9]

1. The best possible explanation, or just the best one anyone has yet come up with? And if the latter, is Craig further claiming that this is a good historical explanation, or merely that it’s not as bad as the alternatives?

2. It’s very unclear what’s being asserted here. Is Craig saying that no one prior to Jesus had ever claimed to speak in the name of a supreme deity?

3. Craig began by saying that “historians have reached something of a consensus“. But he doesn’t indicate where his summary of that consensus ends and his own views begin. If Craig doesn’t intend to suggest that there is a historical consensus that Jesus worked real miracles and was raised from the dead, then he should draw the line between the two more explicitly. And since there isn’t such a consensus — and if there were, it would make Craig’s subsequent argument superfluous! — drawing that line can only improve the clarity and persuasiveness of Craig’s real point.

4. This claim is too weak for Craig’s purposes. Craig needs the resurrection to not just be evidence for God, but exceedingly strong evidence for God. Framing the question as ‘Is this evidence or not?’ risks trivializing the discussion, since most things that make claims likelier only do so by trifling amounts. Perhaps that sounds nitpicky, but it’s especially important to make the strength of one’s claims clear when discussing probabilistic arguments.

5. In the past, Craig has conceded that among historians “it is controversial whether the resurrection of Jesus is the best explanation of those facts“. But he doesn’t mention this in the debate. Nor does he explain why, if historians understand the evidence Craig is citing so well, they are so reluctant to endorse Craig’s conclusion as the most reasonable historical hypothesis.

6. Be wary of false dilemmas. Craig’s hypothesis has to beat rival supernatural explanations, not just natural ones.

7. “Universally“? Is this hyperbole, or is it being claimed that no historian of early Christianity endorses any non-theological explanation of the facts Craig cites?

8. What does “rejection” mean here? Careful historians will assign rough probability estimates to hypotheses before picking some threshold that counts as ‘acceptance’ or ‘rejection’. So Craig might mean that historians assign a very low probability to each one of the “naturalistic” hypotheses to date — they don’t think any one is likely to be true. Or he might mean that historians who have looked at these hypothesis don’t assign a high probability to any of  them.

In the latter case, they may not have even considered whether they’re probably false, if they’ve only examined the evidence enough to determine whether they’re especially likely to be true. A paper ‘rejecting’ some hypothesis might simply be concluding that the evidence is too inconclusive to endorse the hypothesis, relative to general historical standards or relative to the rival hypotheses. If this is the case, then Craig’s argument will fail, since certainly ‘historians have not singled out any one naturalistic hypothesis as unusually plausible’ does not imply ‘each one of the naturalistic hypotheses is likely to be false’.

But there’s a further problem: Even historians who grant ‘each one of these hypotheses is likely to be false’ need not grant ‘it is likely that all of these hypotheses are false’. To make that leap is a probabilistic fallacy.

Consider a detective who thinks, ‘I’m sure that the killer is either the butler, the maid, or the professor; but I have no idea which of them did it!’ The detective might be extremely confident that the culprit is among those three candidates, but not at all confident in the guilt of any particular one. Or suppose I flip a fair coin ten times. The probability of any particular sequence of heads and tails (e.g., TTHHTTTTHH) is less than one in a thousand. But to conclude that it is likely for no sequence to occur, from the fact that it is not likely for any particular sequence to occur, would be absurd. In the same way, it is perfectly open to the naturalist to grant that no specific natural explanation is likely, without granting that a set-theoretic union of all the natural explanations (tomb robbers, or the women got lost, or the whole story came to an overenthusiastic follower in a dream, …, …) is unlikely too.

9. Lastly: Craig presents this as an argument for the existence of God. If we take ‘God’ to signify the Christian God, then one way for him to make his case would be to presuppose that there is some sort of deity, on the basis of his other seven arguments. The form of the historical argument would then be: ‘Given the anomalies surrounding Jesus, plus the fact that we know that some sort of intelligence created our universe, it is reasonable to conclude that this intelligence probably directly intervened in the events described by early Christians.’

On the other hand, if Craig thinks this historical argument could be used to independently conclude that some intelligence crafted the cosmos, then he can’t appeal to the other arguments as premises, and the inferential leap he’s making — from a few ancient manuscripts to the structure and origin of the entire universe — will become quite a bit harder to motivate.

___________________________________________________________________

Alexander Vilenkin

Immaterial causes and the Kalam argument

Craig: [O]ur blogger mistakenly thinks the theorem applies only to inflationary models, which is inaccurate, as the paper referenced above shows.

Craig is right. My thanks for pointing this out! And my apologies to any readers who took away from my post that Borde, Guth, and Vilenkin’s conclusion in “Inflationary spacetimes are not past-complete” applies only given inflation. It holds more generally of any model in which the universe expands on average.

In the debate, Craig presents the Kalam cosmological argument as follows:

1. The universe began to exist.

2. If the universe began to exist, then the universe has a transcendent cause.

3. Therefore the universe has a transcendent cause.

By the very nature of the case, that cause must be a transcendent immaterial being.

Rosenberg focused his attack on premise 2, but I would note that premise 1 remains deeply controversial among physicists. In response to the question “Did the universe have a beginning?”, physicist Sean Carroll writes, “Mithani and Vilenkin are […] willing to be honest about our state of ignorance: thus, ‘probably’ yes. I personally think the answer is ‘probably no,’ but none of us actually knows.” Carroll elaborated in correspondence:

[T]he BGV theorem refers to classical spacetimes, and the universe is not classical. That’s all that really needs to be said. Alex Vilenkin takes this classical result as a strong indication that the true quantum description of the universe also must have a beginning, but at best it’s suggestive. It’s absolutely plausible (and much more likely, in the view of many of us) that the actual universe is eternal, and the BGV result tells us that the classical description must break down, not that the universe must have had a beginning.

Carroll also notes, “The definition of ‘singularity in the past’ is not really the same as ‘had a beginning’ — it means that some geodesics must eventually come to an end. (Others might not.)” Craig has strongly disputed this. However, Vilenkin agrees with Carroll, though with the qualifier “most” in place of “some”. In response to Vic Stenger’s question “Does your theorem prove that the universe must have had a beginning?” (in The Fallacy of Fine-Tuning), Vilenkin responded,

No. But it proves that the expansion of the universe must have had a beginning.

More specifically, Vilenkin wrote,

The theorem says that if the universe is everywhere expanding (on average), then the histories of most particles cannot be extended to the infinite past. In other words, if we follow the trajectory of some particle to the past, we inevitably come to a point where the assumption of the theorem breaks down — that is, where the universe is no longer expanding. This is true for all particles, except perhaps a set of measure zero. In other words, there may be some (infinitely rare) particles whose histories are infinitely long.

Still, my main interest is not in disputing Craig’s premises, but in clarifying what accepting his conclusion would really mean. Since Craig bases much of his argument on the work of Vilenkin and his colleagues, it’s important to keep in mind that Vilenkin himself thinks that we can physically explain our universe’s beginning. In “Creation of universes from nothing“, Vilenkin posits that an empty geometry, devoid of time, space, matter, and energy, could give rise to the universe as we know it.

Previously, Craig has objected that this emptiness would not count as “literally nothing“, hence that Vilenkin fails to explain “being’s coming from non-being“. But Vilenkin is free to grant that the physicalist has no such account. In the context of the Kalam discussion, the physicalist’s burden is to explain, not how something could come from nothing, but how a universe with a beginning could come from an unintelligent but beginningless source. Since Vilenkin’s vacuum is atemporal, it has no beginning. Hence the Kalam argument cannot be reapplied to it. Perhaps some other philosophical objection can show theism to be superior to this hypothesis. But it will still be the case that the Kalam argument fails, at least in the sense that it cannot motivate theism on its own.

Two other potential sources of serious misunderstanding are Craig’s appeal to “transcendent” and “immaterial” causes. There is an obvious sense in which all causes ‘transcend’ their effects — because no event is self-causing. But theorists might wish to deny premise 2 if the premise is taken to mean that something past-eternal couldn’t cause our universe by becoming our universe.

Physicists like Vilenkin are also likely to be wary of the imprecision of the term “immaterial“. This term is pivotal in Craig’s argument, particularly since for him the term “universe” is defined in terms of the material, as “the whole of material reality“. When I raised this concern, Craig responded that he was quite clear:

I am using the word in the ordinary language sense to mean “not material” or “non-physical.”

… Well, sure. My problem wasn’t with the ‘im-‘ prefix. It was with what we’re considering ‘material’ or ‘physical’ in the first place. What general criterion can we use to tell material things apart from immaterial ones? I’ll run through a variety of options:

  • (a) By “material” Craig means ‘made of matter‘, in the sense used in physics. So the universe is the totality of things with spatial extent and mass.
    • Objection: This would make most of physics — spacetime, light, and gravitation, for starters — immaterial. Craig clearly doesn’t mean this, because he wants to exclude physicsy things like these as possible causes for our world.
  • (b) By “material” Craig means ‘nonmental‘. So the Kalam argument simply says that all nonmental things have a beginning, and everything with a beginning must have a cause, so the first nonmental things must have a mental cause.
    • Objection 1: This would make Craig’s position on the mind-body debate trivial, since his rejection of physicalists’ claims that mental processes are ultimately physical would then be merely definitional. (If it weren’t definitional, that would mean he allows the possibility that something could be both material and immaterial, which is, to put it mildly, confusing!)
    • Objection 2: This would render incoherent the distinction between two categories of immaterial thing Craig recognizes: Minds, and abstract objects. If ‘immaterial’ just means ‘mental’, then we can’t even meaningfully talk about neither-mental-nor-physical things like numbers. So this can’t be what Craig has in mind.
  • (c) By “material” Craig means ‘part of our spacetime manifold‘. This matches Vilenkin’s own definition of “universe” as the totality of spacetime regions connected to our own.
    • Objection: This allows that other, disconnected spacetimes might be candidate causes for our universe. Craig might simply deny, on grounds of parsimony, that there are any such spacetimes. But it still seems strange to say that such things, if they existed, would be ‘immaterial’.
  • (d) By “material” Craig means ‘spatial and/or temporal‘. So other spacetimes, if such there be, are included in what Craig calls the “universe”.
    • Objection: Human minds are temporal, hence would count as ‘material’ in this sense. This isn’t inappropriate if the Kalam cosmological argument is meant to explain all of Creation (including the mental parts of Creation), but it does contradict Craig’s stated views on the nature of mind.
  • (e) By “material” Craig means ‘spatial‘. This captures well Craig’s intuition that abstract objects and minds (both human and divine) seem immaterial, as well as his claim that branes are “physical“.
    • Objection 1: Vilenkin’s arguments at most show that “material reality” has a beginning if “material reality” is defined in terms of (c). Vilenkin’s argument does generalize to expanding multiverses, but he is silent on the issue of whether all completely disconnected physical structures, if such there be, have beginnings. So if Craig has (d) or (e) in mind when he speaks of “material reality“, he will need new, independent arguments to show that this reality too must have a beginning.
    • Objection 2: What exactly does ‘non-spatial’ mean? If it means ‘lacking spatial extent’, then point particles might count as ‘immaterial’. If it means ‘lacking spatial location’, then human minds might count as ‘material’. (This will be especially problematic if we cash out divine omnipresence in terms of spatial extension or location.)
  • (f) By “material” Craig means ‘describable in the language of physics‘.
    • Objection 1: What gets to count as ‘the language of physics‘? If we define this too strictly, then we risk calling the posits of slightly nonstandard variants of physics ‘immaterial’. On the other hand, if we define it too laxly, we start to lose any principled way to deny materiality of, for example, the mental.
    • Objection 2: What about physical laws? Craig considers such laws abstract (hence immaterial), but it’s not clear in what sense they could be foreign to physical description.

Of these, I think criterion (e) is the best option, despite its problems. It gets a lot of work done and yet is very simple. But Craig explicitly rejects (e) in the “spatially extended” sense, so his view may be closer to (f). In that case, we can restate his Kalam argument:

1. Every existent describable by an adequately physicslike theory began to exist.

2. If all such things began to exist, then they must ultimately have a cause that is not physicslike.

3. Therefore there is something un-physicslike that is the ultimate cause of everything physicslike.

Expressed this way, in terms of (f), Vilenkin himself strongly rejects premise 1. Likewise if we revised this argument to unpack “material reality” through definition (b). In the (a) and (c) variants, Vilenkin would accept premise 1, but conclude that his empty geometry is an ‘immaterial cause’ in the requisite sense. And if we replaced the argument with one appealing to (d) or (e), Vilenkin would probably maintain agnosticism about premise 1, but would again insist that his empty geometry, being non-spatiotemporal, is an adequate ‘immaterial cause’ as defined. So all of these ways of formulating the Kalam argument either make one (or both) of the premises scientifically dubious, or make the conclusion acceptable to non-theists.
 
The Music of Gounod - a Thought Form from Thought-Forms, by Annie Besant & C.W. Leadbeater
 
Still, for the sake of argument, suppose we granted something akin to the (f) version of the Kalam argument above, and concluded that something alien to contemporary physics (like a mind, number, or free-floating law) were causally responsible for the physical world. Would this suffice for establishing that a mind is the cause?

Craig thinks so. He reasons that we know that numbers and laws are “abstract objects“, and abstract objects have no causal effects. Since no one has been able to think of an immaterial object that is neither mental nor abstract, the only reasonable causal candidate is mental. When I suggested that there might be other immaterial causes to choose between, like the Forms of Plato, Craig responded:

Platonic forms and free-floating laws are abstract objects, so I just have no idea of what other world-transcending causes he’s talking about. If he can give us such a candidate, I’ll add it to the list of candidates to be considered, but I have yet to see such a candidate suggested, much less one that is more plausible than a transcendent mind.

This response surprised me. Craig has written a great deal about what’s nowadays called ‘platonism,’ or realism about abstract objects. But Craig’s assertion here reflects a lack of familiarity with the core doctrine of Plato himself, the doctrine that the sensible world is a product of the eternal Forms. Against Craig 2013, I cite Craig 2009:

By the way, what passes for Platonism today shouldn’t be identified with what Plato himself actually believed. For Plato, the Forms do not seem to be at all causally impotent but shape the world to be as it is. The debate over so-called abstract objects is actually a very recent development of contemporary philosophy which arose only in the late 19th century.

The source of Craig’s latter-day lapse is likely an ambiguity in the terms ‘platonism’ and ‘abstract’. By ‘abstract object’ philosophers (including I and Craig) usually mean ‘something non-spatiotemporal and causally inert’. But some  philosophers use the term more loosely, to refer to anything non-spatiotemporal. Plato’s Forms are abstract in the latter sense, but not in the former sense; and it is only the former sense that is relevant to Craig’s rejection of abstract objects as causes. As Gideon Rosen writes: “Plato’s Forms were supposed to be causes par excellence, whereas abstract objects are generally supposed to be causally inert in every sense.”

A second source of confusion is that even though belief in abstract objects is often called ‘platonism’ or ‘platonic realism’, Plato himself was a nominalist, and not a platonist or realist. (Paul Spade notespp. 56-61, that Plato is probably a nominalist, not just about abstracta, but about universals as well. Plato’s Forms, as usually presented, are potent particulars.)

Most metaphysicians these days consider the actual Forms of Plato so implausible as to be of merely historic interest, in contrast to the vibrant debate surrounding abstract objects. Since these abstracta have a superficial resemblance to the Forms, and are taken more seriously, the name of Plato is appropriated as a colorful way of picking out abstracta. Whence Craig’s conflation of the two.

But why do modern philosophers dismiss the Form of  Duality in favor of the abstractum 2? Simply on the grounds that our universe is causally closed. Plato’s actual views are dismissed with a chuckle, while abstract-object ‘platonism’ is vigorously attacked and defended, because Plato’s Forms purport to ‘spookily’ intrude upon our everyday lives and in the very existence of our cosmos, while abstract objects kindly recuse themselves from the realm of empirical science.

But this is precisely the assumption someone arguing for a universe-begetting intelligence cannot grant. Either Craig is illicitly assuming the causal closure of the physical when it harms rival doctrines and then rejecting it when the focus shifts to his preferred posit, or he simply hasn’t taken the time to seriously assess any hypotheses invoking unintelligent immaterial causes.

My point in all this isn’t to defend Plato’s doctrines, or for that matter Vilenkin’s. It’s merely to suggest that Craig is far too hasty in moving from his conclusion of the Kalam argument to an invocation of transcendent minds, divine or not.

___________________________________________________________________

Just the facts

CraigThis blog is not really fact-checking (which would have involved alerting readers to factual mistakes like my ascribing a quotation to Penelope Maddy instead of Mary Leng or my giving the date of Caesar Augustus’ death as AD 17 rather than AD 14) so much as it is entering into the debate itself in assessment of the arguments.

That’s true to an extent. I generally limited myself to evaluating the soundness of Craig and Rosenberg’s arguments, and not to putting forward novel arguments of my own for the broader topics under dispute. For instance, I didn’t weigh in with my own view on the historicity of Jesus, on the right interpretation of quantum mechanics, or for that matter on the existence of a deity. (The main exception: I provided an argument of my own in §10, mainly to give an example of what deductive arguments from evil should look like.)

So whatever I was doing, it wasn’t prototypical ‘fact-checking’, but it was still decidedly from the sidelines. And I think you can tell from the tone that I was mainly using the ‘fact-check’ idiom as a fun way to spice up a relatively long post. (After all, one of my checks was just an excuse to make a Scientology pun.)

For all that, I’d be very interested to see a deeper discussion about where to draw the lines between (neutral? objective?) ‘fact-checking’ and personally entering the fray. Is a fact-checker allowed to evaluate the validity of arguments, or only the truth of premises? Can she only evaluate trivial claims, or can she also question premises that are central to a debater’s whole case? How uncontroversial or obvious does a truth have to be in order to count as a ‘fact’? I don’t have easy answers to these questions myself.

However this discussion started, it’s now moving into increasingly interesting and important philosophical waters. I’d love to hear Craig’s and others’ responses to the new historical, methodological, cosmological, and metaphysical issues raised so far.

___________________________________________________________________

Further reading
Craig, William Lane (2008). “Current Work on God and Abstract
Objects” Reasonable Faith.
Guth, Alan (2002). “The Inflationary Universe“. Edge.
Rodriguez-Pereyra, Gonzalo (2011). “Nominalism in Metaphysics“. SEP.
Rosen, Gideon (2012). “Abstract Objects“. SEP.

God is no libertarian

So the world was made by a perfectly benevolent, compassionate, loving God. Yet suffering exists.

Why would a nice guy like God make a world filled with so much nastiness? All these wars, diseases, ichneumon wasps—what possible good purpose could they all serve?

We want God to make our lives meaningful, purpose-driven. Yet we don’t want that purpose to be super depressing. ‘God is a nice guy from his own perspective, but a total asshole by all human standards’ would be a pretty unsatisfying theodicy, and a terrible way to fill the pews. So how do we square a good God with a wicked world?

The standard response is that being truly good requires that one love freedom. God is so good that he won’t interfere with human freedom by preventing suffering. That certainly sounds nice; we don’t want to make autocracy the highest good. But how can this work in practice?

The idea seems to be that we are somehow to blame for our suffering. God, then, is off the hook. We’re free to blame ourselves (rather than God) for whatever evil things befall us. What’s more, we’re free to credit God (rather than ourselves) for whatever good things we accomplish. In this way we can, if we wish, preserve the pure wretchedness of man and the pure excellence of God. We are free to translate the complexity of human experience into a crisp conflict between total sin and total virtue. But there are deep problems with this approach: The shape of our world seems profoundly unlike the shape we’d expect from a libertarian architect.

____________________________________________________________________

First Problem: Natural evil limits freedom.

It’s clear that not all suffering stems from human action. If God had protected the 230,000+ victims of the 2004 tsunami, how would this have interfered with human freedom? Would it not, if anything, it have increased our freedom, by giving the tsunami’s victims a chance to live out their lives?

One might respond that the tsunami’s destruction could have been greatly reduced by human actions. Perhaps God gave us just enough power to save ourselves, and we simply did not employ it.

But blaming the victims simply does not work here. No matter what we had done, we could not have saved every life. And if some people were to blame for the level of devastation, surely those people should have been punished, not innocent bystanders. Which brings us to…

Second Problem: Human evil limits freedom.

If God loves freedom, why does he let people obstruct and enslave one another? Why does he allow oppressors more freedoms than the oppressed? Why not give us just enough freedom to control our own lives, so long as it does not infringe upon the freedom of others?

The problem of evil raises special concerns for individual freedom. You might claim, for example, that humans (and not God) are responsible even for natural disasters, because Adam and Eve introduced suffering and death into the world when they disobeyed God. But that is not a crime committed by every human being, such that every human deserves punishment for it. It is a crime committed by two particular humans. How can we justify punishing someone else for a perfect stranger’s crime? Certainly it is not my fault if I was born to a sinful father. We can’t choose our parents.

(To my knowledge, Origen is the only theologian to have ever resolved this problem. Unfortunately, later thinkers generally consider Origen’s views heretical, and even Origen falls victim to religion’s standard “blame the victim” mentality.)

Third Problem: Our freedom is physically limited.

It’s easy to say that God loves freedom by counting the hits (look at all the things he lets us do!) and ignoring the misses (the things we can’t do). But of course we aren’t free to do whatever we want. God created us in a very specific way, strictly limiting what we can will ourselves to do. We can’t fly merely by flapping our arms. We can’t will aches and pains to go away. We can’t even go directly to Heaven merely by willing it.

So what? What’s the problem? Well, we’ve granted that freedom isn’t absolute, that a good God would make beings free in some respects, but not in others. But now we are forced to explain why God limits our freedom in the particular way that he does. Why give us the freedom to make sandwiches and fire guns, but not the freedom to cure all diseases or teleport away from natural disasters?

If we can’t even begin to explain this, then ‘God loves freedom’ ceases to be a viable justification for suffering. The question is now why God loves this particular freedom (the ‘freedom’ to suffer even when we’d prefer not to) more than he loves rival freedoms (the freedom not to suffer!). A generic appeal to ‘freedom’ can’t even begin to address this question.

Fourth Problem: Our freedom is epistemically limited.

This is the problem of ignorance, a far deeper and thornier issue than the standard problem of evil. What can it mean to say that God respects freedom, when he obviously doesn’t respect informed freedom?

Freedom, in fact, seems quite meaningless when it is not informed. Imagine a child told to pick between two closed doors. Behind one door is a fierce tiger, and behind the other door is chocolate. If the child chooses the door that happens to have a tiger, can we blame the child for his messy death? Surely not.

Yet we, too, live in a world we scarcely understand. It is often claimed that God hides himself from us in order to give us the freedom to doubt him, to choose our beliefs for ourselves. But in fact God’s hiddenness has the opposite effect; it takes away from us our freedom—our freedom to make an informed choice. Since we do not know which religion, if any, is the correct one, we can hardly be blamed if we err. Yet theists assert that those who fail to find God will suffer (e.g., in Hell or merely ‘the absence of God’), and that they deserve to suffer.

Being forced to play Russian roulette, and then losing, is not the same as committing suicide. The freedom to guess is not freedom. It’s just a slavery to chance. Only the freedom to choose between options whose consequences we fully comprehend is genuine freedom, because only then do we really know what option we’re choosing. Yet clearly God did not create beings who fully understand their actions’ consequences. Least of all in the realm of religion.

____________________________________________________________________

The notion of ‘freedom’ favored by our allegedly well-meaning deity, then, ends up looking extremely peculiar. God evidently only loves freedom when it can infringe upon (and be infringed upon by) others’ freedom, and when it is severely limited in seemingly arbitrary ways, such that we are not free to escape suffering in this life or to make informed choices. After qualifying what God prizes in so many strange ways, what evidence remains for the supposition that these preferences even slightly resemble what we call “morality” or “compassion” in the case of humans?

I can think of four possible responses.

  • To Problem 1: Perhaps God created a perfectly orderly world, and in such a world it was inevitable that some disasters would arise.

This doesn’t explain why God created the particular world he did, or why he created at all. It also doesn’t explain why God prizes abstract “order” more than human welfare. Couldn’t he create a world that naturally has typhoons, yet still intervene to save the people victimized by his natural order? The fact that buildings inevitably fall down sometimes doesn’t make it any less immoral to choose not to save people from falling buildings if you’re able.

  • To Problem 2: It’s not God’s fault that humans hurt one another.

The issue isn’t that God’s to blame for everything humans do. It’s that God chose to limit human freedom in one way, but not in another. He made us free to harm one another, but not free to be safe from others’ harm. What makes the former freedom more important than the latter? Why is the villain’s freedom prized above the victim’s? Even if God doesn’t directly cause every human action, he still chose which possibilities to leave open. That calls for explanation.

  • To Problem 3: If we could do anything, we’d be God.

This relies on a false dilemma. It’s not that case that God needs to either make humans omnipotent, or deny them the ability to escape suffering. He could easily give them that one ability, while continuing to deny them other abilities. This on its own would radically decrease the suffering in the world, and radically increase people’s freedom.

And, as an aside: What’s wrong with being God? God sure seems to like it!

  • To Problem 4: If we knew everything, we’d be God.

Again, we don’t need to be omniscient merely to know the consequences of our actions. God chose to create beings that are ignorant of almost everything. If such beings sin without fully understanding the consequences, they cannot ethically be held more responsible than God for what ensues.

Fact-checking the Craig/Rosenberg debate

This 2/18 post first appeared on the IU Secular Alliance and Philosophical Society blogs.

On February 1, Christian apologist William Lane Craig and philosopher of science Alex Rosenberg debated the relationship between theology and ethics, cosmology, metaphysics, and a range of other topics at Purdue University. And, good golly, they covered a lot. In the interest of deepening this already-broad conversation, I’ll assess the merits of a smattering of their assertions, both scientific and philosophical.

But I’m not going to weigh in on who won. Because I do agree with a fundamental point raised by Rosenberg, not about the debate’s topic but about formal debate itself:

Philosophy and theology don’t proceed by courtroom-style debate. We’re engaged in a cooperative search for the truth, both theists and atheists, not an adversarial contest for victory. […]

But that’s the problem with this kind of a debate, and this kind of a format. It doesn’t work. It doesn’t work because what I’d like to be able to do is ask William Lane Craig a question, and listen to his answer, and formulate a reply, and listen to his answer. And then give a view, and listen to his question. Which is the way in which philosophical dialogue proceeds, and which enables us at least to find out where the crucial issues are between us, and how we could mutually agree to adjudicate these matters.

Rosenberg’s request is simple. He wants to talk to Craig. He wants a real-time back-and-forth, a friendly and open exchange of ideas rather than a stiff gladiatorial combat. If there is a battle of any significance here, it is between all of us and the forces of ignorance and error. Inasmuch as the debate was enlightening, both debaters won; inasmuch as it is was muddled or superficial, both debaters lost. As did we all.

But that battle continues. Just because the debate is presented as highbrow sumo wrestling doesn’t mean we can’t exploit it to open up a richer dialogue. I encourage you to join the discussion, and let me know which of my points you agree or disagree with!

___________________________________________________________________

1. God Hypotheses

Craig: “Now there’s only one way I can think of to get a contingent universe from a necessarily existing cause, and that is if the cause is a personal agent who can freely choose to create a contingent reality. It therefore follows that the best explanation of the existence of the contingent universe is a transcendent, personal being. Which is what everybody means by ‘God.’

Assessment: Misleading

Perhaps that’s part of what a lot of people mean by ‘God’. But it’s not everything that is meant by ‘God’. If you learned that this transcendent, personal cause of the universe were ignorant or mad, or that it annihilated itself in the course of making the universe, or that it were a cruel tyrant, it’s unlikely that you would even think of calling this Lovecraftian absurdity ‘God’. Certainly you wouldn’t think that it was your god.

In general, attempts to prove that something has some of the interesting properties you ascribe to your god, although not irrelevant, need to be heavily qualified when there is a great swarm of hypothetical beings that you would never worship but that meet the same requirements. There are thousands of conditions a deity has to meet, above and beyond transcendence and personhood, before it can even begin to approximate the God of the Bible.

Note also that Craig is giving an argument to the best explanation. But the best explanation may not be a very good explanation, if all the options we’ve thought of are unlikely to different degrees. If I ask ten randomly selected people to give their best guess as to the value of -4⁴, I shouldn’t be all that confident that the least unpopular answer is the right answer. The real question is: Would we expect anyone to have thought of the right answer by now? If not, then we may have reason to doubt that the ‘best’ explanation is worth very much. We also have to be wary here of appeals to ignorance; “there’s only one way I can think of…” only matters if everyone else shares your ignorance and if we would strongly expect anyone to have thought of the right answer by now.

___________________________________________________________________

2. The Beginning of the Biggening

Craig: “Because we don’t yet have a quantum theory of gravity, we can’t yet provide a physical description of the first split second of the universe. But the Borde-Guth-Vilenkin theorem is independent of any physical description of that moment. Their theorem implies that the quantum vacuum state which may have characterized the early universe cannot be eternal in the past, but must have had an absolute beginning. Even if our universe is just a tiny part of a so-called multiverse composed of many universes, their theorem requires that the multiverse itself must have had an absolute beginning. […]

But then the inevitable question arises, why did the universe come into being? What brought the universe into existence? There must have been a transcendent cause which brought the universe into being.

Assessment: Misleading

This is almost right, but requires the added stipulation that the multiverse in question be inflationary. I asked Alexander Vilenkin what he thought of Craig’s characterization, and he wrote:

“This is accurate. But note that the theorem assumes that the universe was on average expanding in the past. The conclusion can be avoided if the universe was contracting prior to the expansion. But contracting universes have problems of their own. They are highly unstable, so the contraction is not likely to be followed by an expansion (which we now observe).”

Another possible source of confusion is that Craig’s conclusion — that our universe must have a “transcendent cause” — is not generally endorsed by physicists who do grant that it had a beginning. Vilenkin comments, “I don’t think the cause should necessarily be transcendent.”

What Borde, Guth, and Vilenkin proved is that new, non-inflationary physics is required to “describe the past boundary of the inflating region of spacetime”. Maybe that new physics will have a ‘God’ term; maybe it won’t. But this theorem does not obviously rule out immanent explanations.

___________________________________________________________________

3. Immaterial Causes

Craig: “By the very nature of the case, that cause [of the universe] must be a transcendent, immaterial being. Now, there are only two possible things that could fit that description. Either an abstract object, like a number, or an unembodied mind or consciousness.

Assessment: Implausible

Note that Craig gives no argument here that only causally inert abstracta and minds could transcend our universe. Yet he asserts that we not only haven’t come up with such an entity yet, but that such a thing is impossible. This in spite of the many philosophers, from Plato to the present day, who have posited unconscious immaterial causes. Lacking any proof of the impossibility of such things, we must conclude that the argument fails; whereas Craig’s earlier argument-to-the-best-explanation was much more persuasive, though its conclusion was also much weaker.

We have to be especially wary of the fallacy of equivocation here. Craig uses ‘immaterial’ to mean ‘outside the universe’ (like God), but he also uses it to mean ‘not spatially extended’ (like ordinary human mental states). But my mind is in the universe; more specifically, it’s in the United States. My present hunger, for example, isn’t nowhere. (Nor everywhere!) It’s at the particular place where I am. But this means that we don’t know of any minds that are nonphysical in Craig’s sense, and it isn’t obvious that there could be such minds. Likewise, minds as we know them are all temporal; it’s not clear that we have any coherent idea of a thought or sensation existing outside time itself. Insofar as we do have some vague sense of such a mind, surely we might also have a vague sense of branes, Platonic Forms, free-floating Laws, or other world-transcending causes.

___________________________________________________________________

4. Anthropic Arguments

Craig: “By far, most of the observable universes in a world ensemble would be worlds in which a single brain fluctuates into existence out of the vacuum and observes its otherwise empty world. Thus, if our world were just a random member of a world ensemble, we ought to be having observations like that. Since we don’t, that strongly disconfirms the world ensemble hypothesis.

Assessment: Mostly Right

Max Tegmark has proposed that we can explain why our universe seems ‘fine-tuned’ for complex mathematical and biological structure by positing that we’re just a small part of a much larger multiverse of randomly varying mathematical objects. Since we would only expect living things to emerge and notice how nice and friendly their surroundings are in the parts of this giant ‘ensemble’ that make life possible (well, yeah), our universe’s observed hospitability then becomes a lot less surprising.

It’s an interesting idea, but, as Craig suggests, it seems to have some absurd consequences: we should expect all our memories to be an illusion formed out of a chaotic flux. This is because, on Tegmark’s view, most universes are chaotic mishmashes. If I think I’m a brain in a randomly selected universe in Tegmark’s ensemble, then I should expect to be one of the billions of brains randomly and momentarily arising from chaos (complete with fake memories!), rather than one of the rare brains produced by a huge, physically simple chunk of spacetime that lawfully produced me and ancestors like me over millions of years. This is the problem of the Boltzmann Brain.

The easiest response is that we occupy a multitude of relatively simple worlds with just a few randomly varying physical constants (e.g., the fine-structure constant), enough to account for apparent fine-tuning; but that multitude is not so diverse that it has a preponderance of ‘chaotic’ universes generating Boltzmann Brains. This may seem like a somewhat ad-hoc answer, and further serious debate about anthropic reasoning is certainly warranted. Anthropic multiverses like Tegmark’s will have to contend not only with life-selecting mechanisms like Craig’s, but with heavy-element-selecting mechanisms like Lee Smolin’s cosmic evolution.

________________________________________________________

5. Ancient Miracles

Craig: “There are actually three facts recognized by the majority of historians today which I believe are best explained by the resurrection of Jesus. Fact #1: On the Sunday after his crucifixion, Jesus’ tomb was found empty by a group of his women followers. 2: On separate occasions, different individuals and groups of people saw appearances of Jesus, alive, after his death. And, 3: The original disciples suddenly came to believe in the resurrection of Jesus, despite having every predisposition to the contrary.

Assessment: Misleading

The only evidence Craig cites is N.T. Wright’s claim that these three propositions are “virtually certain”. What Craig doesn’t mention is that Wright is not only a historian, but a Christian apologist and bishop. For that matter, Craig doesn’t note that most New Testament scholars are Christians. (Are we to take it as evidence for the truth of Christianity that a lot of Christians happen to be Christian?)

Now, of course being a Christian doesn’t make it impossible for you to evaluate Christianity in a fair and skeptical way. I believe very strongly that the Earth is round, but that doesn’t mean that I’d be hopelessly biased in a debate with flat-Earthers. Agnosticism does not imply objectivity, and objectivity does not imply agnosticism. If anything, we’d be worried if most New Testament scholars weren’t Christians, since that would suggest that the historical evidence tended to make people less religious than the general populace.

But it’s also worth noting that Christian orthodoxy is not generally considered by historians the only possible objective interpretation of the evidence of the Gospels. And appealing to scholarly consensus here is misleading inasmuch as it has the guise of an appeal to independent authorities, as opposed to authorities who already came into the field accepting Christianity.

As for the claims themselves, before we can even begin to evaluate ancient miracle accounts, we need some training in historical methodology and knowledge of the relevant cultural context. This talk is very informal, and is addressed to a nontheistic audience, but provides a nice introduction to those two topics:

________________________________________________________

6. The Great Chain of Becausing

Rosenberg: “Many of the arguments that Dr. Craig gave tonight [… rest] on, of course, the Principle of Sufficient Reason, the principle that everything that exists must have a cause.

Rosenberg: “We know that alpha particles come into existence for no reason at all every moment in this room. Why should we assume that the universe is any different? Why should we assume that purely quantum-mechanical fluctuations — symmetry breaking, which we understand is the explanation for why there’s matter in the universe and not antimatter — why this process which produces the characteristic features of our universe and does so without there being a cause for its happening one way or the other, why the symmetry gets broken one way or the other, couldn’t be the nature of reality as far back as we can possibly dig in cosmology?

Assessment: False

Craig does not appeal to a principle as strong as ‘everything has a sufficient reason/cause/explanation independent of itself’. Were he to do so, his arguments for God would backfire, since God would then need to be caused or explained in its own right. Instead, Craig claims (a) that physical events and things always require an explanation (and the universe, of course, is physical), and (b) that contingent things always require an explanation. Rosenberg questions (a), and we could also question (b), or ask how we know that anything is really contingent. But it’s important not to conflate these three claims.

It seems that just as Craig is arguing from  ’every physical event has a cause’ to ‘the universe must have a cause’, Rosenberg is arguing from ‘many physical events lack a cause’ to ‘the universe lacks a cause’. Neither of these inferences seems very strong to me. (EDIT: Rosenberg tells me that he ratheris arguing from ‘many physical events lack a cause’ to ‘[it’s] possible that the universe lacks a cause.[‘]”)

Vilenkin suggests, in correspondence:

“This is not very clear, but it seems that what he [Rosenberg] is referring to is the creation of closed universes from ‘nothing’. The possibility of such a process is indeed suggested by quantum cosmology, but the word ‘nothing’ should be interpreted with care. Here, it is taken to mean a state with no matter and no classical space and time. But the origin of the universe is described by the laws of physics, so the laws are assumed to be ‘there’ as an input.  Mr. Craig may argue that the laws must be provided by God. I am not sure this explains anything; we could just as well say that the laws have always been ‘there’. However, in fairness I should admit that so far physics offered no explanation for the laws. Why these laws and not some other? Why any laws at all?”

Physics graduate student Jeffrey Eldred provides a defense of Rosenberg’s general approach, though he notes that Rosenberg is mistaken in thinking that physicists look to spontaneous symmetry breaking to explain the matter-antimatter disparity:

“Rosenberg[‘s claim] ‘…quantum-mechanical fluctuations, symmetry breaking, which we understand is the explanation for why there’s matter in the universe and not anti-matter…’ is not generally accepted by physicists and cosmologists. Physicists already have experimental confirmation of matter-antimatter asymmetry in the properties of quarks, and there are experiments underway expecting to find the remainder of the asymmetry. […] I don’t know what Rosenberg was thinking about. Perhaps he was jumping the gun and […] looking to the theories that would explain matter-antimatter asymmetry in the event we didn’t find it in the neutrino sector, or maybe he was uncritically endorsing remarks reportedly made by Einstein. […]

“Spontaneous symmetry-breaking is the idea that an unstable symmetric system will be forced to break the symmetry in an arbitrary direction. Classically if you balance a perfectly [cylindrically] symmetric, perfectly sharp pencil perfectly on its point [then] it will never fall over. Quantum-mechanically, random fluctuations in the particles that make it up would force it to become slightly asymmetric and then cause it to settle into a stable asymmetric state (lying on the table pointing in a random direction). Whatever your interpretation is, the way the symmetry will break cannot be known from our perspective and the consequences of those fluctuations can be lasting. […]

“Inflationary theories are supported in part by Cosmic Microwave Background evidence that shows the distribution of matter in the universe fits the model of quantum fluctuations between close particles and then subsequent inflation. Inflation theories can explain in a similar way any parameter of the universe which depends on the distribution of matter, the mechanism of inflation, or could vary slowly over scales larger than our universe. The original [arrangement] of matter could be empty but then spontaneous symmetry breaking of the unstable vacuum state could cause it to become populated with matter.

“I’m not sure […] how Rosenberg is linking this to other parameters of the universe such as the gravitational constant or if he is even trying to explain them. Is he assuming that there is a different but analogous process for those parameters or is he saying that they are created by the same mechanism? Here’s how they could be created by the same mechanism. Let’s say for instance the gravitational constant varied over space in the very early universe (ie the multiverse) and subsequently inflation took place in a very small region of that space which would eventually get inflated into our universe. That would mean our universe would have an effectively constant gravitational constant because the gravitational constant wouldn’t vary much in such a small original space, and our gravitational constant could be picked effectively at random from the true possible variation in the gravitational constant if there was nothing special about the space that would become our universe. We don’t know that to be true about the gravitational constant but if inflation is right than we might never […] know if it is true about the gravitational constant or any other parameter. We might try to analyze if our universe is a typical random (or typical anthropically selected) universe from the possibilities, but we might not even be able to know what a typical universe is since we can’t observe any outside our own universe.”

________________________________________________________

7. Anthropics Revisited

Rosenberg: “To begin with, this is terrible carbon chauvinism. If these constants had been slightly different, maybe there would be intelligent life in the universe that’s germanium-based or silicon-based.

Assessment: Implausible

In fact, silicon- or germanium-based life may very well exist in our universe. But, as Craig correctly notes, the sorts of radical tinkerings that fine-tuning arguments appeal to would generally make all stable atoms impossible, not just carbon. So, although the suggestion that life might be possible in universes with very different physical constants would be a powerful anthropic rejoinder, a lot of work will need to be done to make it credible. Until then, the best anthropic arguments will appeal to some sort of multiverse.

________________________________________________________

8. Space Opera

Rosenberg: “Scientology, that claims 8 million adherents throughout the world, tells us that 75,000,000 years ago somebody named Zeno brought spaceships to Earth that look like DC-8s.

Assessment: False

The guy’s name was Xenu.

… Otherwise, yeah, that’s right.

(Though it’s worth noting that while Scientology claims 8 million adherents, the actual numbers are smaller by an order of magnitude or two.)

(Also, no way would Zeno have finished that trip.)

________________________________________________________

9. Testimony

Rosenberg: “Think about this: 53 of the first 62 DNA exonerations of people who turned out to be innocent of charges of capital crimes in the United States were convicted on eye-witness testimony. We know from cognitive, social science how unreliable eye-witness testimony is today. Why should we suppose that eye-witness testimony from 33 AD is any more reliable? This, as an argument for God’s existence, seems to me to be bizarre.

Assessment: Mostly Right

This is a very important point. Wells, Memon, and Penrod note: “Analyses of DNA exoneration cases since 1992 reveal that mistaken eyewitness identification was involved in the vast majority of these convictions, accounting for more convictions of innocent people than all other factors combined.”

What’s potentially misleading here is the suggestion that we have eye-witness testimony of any event from Jesus’ life. As Rosenberg later notes, the Gospels are generally dated to 40-60 years after Jesus’ death, and none of them even claims to be an eye-witness account.

________________________________________________________

10. The Problem of Evil

Rosenberg: “Logically speaking, if God is omniscient, and God is omnipotent, and God is truly benevolent, has a totally good will and would never will anything but for the best, then the existence of suffering on our planet — human suffering and natural suffering, of other animals, for example — is something that needs desperately to be explained. And we’ve had over the course of 400 or 500 years of wrestling with this problem the Free Will defense, and the mystery-mongering […] defense, and nobody has managed to provide a satisfactory explanation. And I insist that the problem is logical. And Dr. Craig needs to tell us exactly how an omnipotent god, and an entirely benevolent god, had to have the Holocaust, in order to produce the good outcome, whatever it might be, that he intends for our ultimate providence. […] In all honesty, if Dr. Craig could provide me with any way of a logical, coherent account that could reconcile the evident fact of the horrors of human and infrahuman life on this planet over the last 3.5 billion years, with the existence of a benevolent, omnipotent agent, then I will turn Christian.

Assessment: Misleading

Rosenberg’s argument here is perplexing. His actual points are perfectly fine — as inductive, probabilistic arguments. Many of Craig’s own arguments are probabilistic. But Rosenberg repeatedly uses the word ‘logical’, which Craig takes to refer to ‘the logical problem of evil’, the attempt to deductively prove the impossibility of God’s coexisting with evil. Either Rosenberg is misrepresenting the force of his own arguments, or there’s a serious communication gap between him and Craig.

If Rosenberg is happy to settle for induction, then that would explain why he repeatedly demands a theistic explanation for atrocities like the Bubonic Plague and the Holocaust. It doesn’t make any sense to demand explanations for logical contradictions like square circles; we can simply note that they’re impossible and move on. But it does make sense to demand explanations if you just think that God is overwhelmingly unlikely, rather than impossible.

This miscommunication is doubly unfortunate because it leads Rosenberg and Craig to talk past each other in terms of the burden of proof: Rosenberg repeatedly demands that Craig explain how a good God could have allowed evil, while Craig repeatedly demands that Rosenberg prove the impossibility of there being some good reason we haven’t yet figured out. When what’s being disputed is unclear, the burden of proof will be correspondingly unclear.

That said, there might be some interesting deductive arguments against the coexistence of evil with certain concepts of a benevolent God. For instance, here’s one I came up with:

1. God is perfect. Among other things, this means that God is perfectly benevolent and perfectly knowledgeable.
2. God is the sole creator of our universe.
3. If God is perfect, then in a situation in which only God existed, there would be no shortcomings.
4. If a situation has no shortcomings, then it cannot be improved upon.
5. So God’s creation of our universe could not have been an improvement. (from 1, 2, 3, 4)
6. A perfectly benevolent being will not knowingly bring about a situation that risks producing evil, if doing so could not improve upon the prior situation.
7. Creating our universe risked producing evil.
8. So God is not perfectly benevolent. (from 1, 5, 6, 7)
9. Contradiction. (from 1, 8)

This argument is valid, but some of its premises may be counter-intuitive. In particular, some people may want to insist that God’s creation of free agents was an improvement upon the status quo; but it’s hard to articulate how that could be so without watering down 1. Alternatively, some may want to insist that benevolence is not about improving scenarios (denying 6). But this just doesn’t seem right. Benevolence may not only consist in improving reality, but that’s surely at least one important factor; all else being equal, it’s better for the world to be better, to have a higher good-to-evil ratio! And, again, given God’s perfection, it’s hard to articulate what advantage could outweigh the colossal suffering (or risk-of-suffering) God engendered.

But I digress. I just wanted to illustrate what a deductive argument from evil might look like. Rosenberg himself doesn’t clearly formulate one.

________________________________________________________

Christus statue of Jesus in the Church of Jesus Christ of Latter-day Saints Temple Square.

11. Naturalism

Craig: “But if God does not exist, then I think metaphysical naturalism is true. Metaphysical naturalism doesn’t follow from epistemological naturalism, but it does follow from atheism. The most plausible form of atheism is, I think, metaphysical naturalism. But there are all those absurd consequences that result from that that I describe.

Assessment: Implausible

By ‘theism’ Craig seems to mean the belief in a necessary, uncaused, simple, immaterial person who existed outside of spacetime, freely created the universe, and is identical to goodness. But he also seems to treat ‘atheism’ here as just the negation of theism; it’s any view on which theism is false. But then there are numerous monotheisms and polytheisms that qualify as ‘atheistic’ in the relevant sense, since they deny at least one of the properties Craig ascribes to God (e.g., simplicity, or necessity, or benevolence).

There’s also some ambiguity in Craig’s claim that “metaphysical naturalism[…] does follow from atheism“. Polytheistic doctrines surely do not count as naturalisms. For that matter, we intuit that werewolves, sorcery, and astrological influences are ‘supernatural;’ they violate metaphysical naturalism. But we don’t have to believe in Craig’s deity to consistently believe in magic. Either Craig is committing a false dilemma fallacy with respect to theism and atheism, or he’s committing a false dilemma fallacy with respect to naturalism and non-naturalism.

So let’s reconstruct a more charitable version of the argument. I don’t think Craig means to say that metaphysical non-naturalism logically entails his version of theism. Rather, he takes it as self-evident that naturalism is false — because he (a) equates naturalism with physicalism, and (b) assumes that human thought, perception, and language cannot possibly be physical. The former, (a), is very nonstandard, and constitutes a third false dilemma. But let’s grant it for the moment. Craig’s argument then is, I think, that the truth of (b) does not entail theism, but rather that theism is the only serious contender for a satisfactory explanation of (b). Craig’s issue with atheism, then, is that it denies the best explanation for the data; and he thinks this is only intellectually sustainable if one also denies the (unphysical) data themselves.

As such, these are the points Craig needs to focus on in order to make his case:

(1) Show that seemingly non-physical things, like thoughts and words, cannot be explained by or analyzed into physical processes (e.g., brain computations). This gets rid of reductive physicalism.
(2) Establish that eliminative treatments of thoughts and words are not only counter-intuitive or silly-sounding, but actually false. This gets rid of eliminative physicalism, including Rosenberg’s view.
(3) Establish that all possible (or plausible) metaphysical naturalisms must be physicalistic. Given 1 and 2, this gets rid of naturalism.
(4) Establish that all possible (or plausible) metaphysically non-naturalist views must appeal to Craig’s version of the God hypothesis.

And in the course of the above, Craig must not merely establish that his version of theism is the best (i.e., least terrible) explanation, but that it’s probably right.

That may sound like a lot, but it’s only fair that Craig start to seriously fill in the details in his view, given how many arguments he typically demands that his debate opponents make!

________________________________________________________

12. Biblical Language

Rosenberg: “And all of [the New Testament scholars] tell us that it was written by people who were illiterate. […] And of course the Aramaic in which they [the Gospels] were written was completely lost, and all the extant New Testaments are in Greek.

Assessment: False

The Bible was written by illiterate people? A miracle!

OK, I think this is a scrambled version of what’s supposed to be the claim that because the Gospel writers were literate, they couldn’t have been the (mostly illiterate) apostles. But this argument is a bit superfluous, since the Gospels themselves make no claim to be written by apostles.

The second claim is also wrong. As Craig points out, the New Testament was originally written in Greek. This does suggest a cultural divide between the New Testament writers and the early Aramaic/Hebrew-speaking followers of Jesus, but such a divide doesn’t require that the texts be mistranslated.

________________________________________________________

13. Quantum Indeterminacy

Rosenberg: “Now, if every event has to has a cause, if everything that comes into existence has to have a cause of its coming into existence, then there’s got to be some difference between the two atoms in virtue of which one of them emitted an alpha particle and the other didn’t. But quantum mechanics tells us, and all the experimental evidence which confirms it to twelve decimal places tells us, there is no difference. End of story. There is an event without a cause. […]

This is not an issue about the interpretation of quantum mechanics. I happen to think that among the interpretations of quantum mechanics, some of the deterministic ones are more plausible than others. This is a matter of experimental physics. This is a matter of a fact about the nature of reality. And it also seems to me clear that insofar as we have here good evidence that things can happen with no cause at all, it follows that therefore the universe can come into existence with no cause at all. And, indeed, that’s what the best guesses of contemporary physical theorists is.

Assessment: False

Rosenberg is simply wrong here. The standard, early-20th-century interpretations of the data and formalisms of quantum mechanics were indeed indeterministic. But these ‘Objective Collapse’ interpretations have become increasingly unpopular, because they posit a fundamental discontinuity in the laws of nature, a sharp point where the laws of microphysics abruptly give way to the laws of macrophysics. This is not only inelegant, but empirically implausible, since we have yet to identify any well-defined criterion for circumstances in which collapse does or doesn’t occur. (For instance, some Collapse theorists suggest that wave functions collapse whenever a ‘measurement’ occurs. But what, in physics, counts as a ‘measurement’? There is no rigorous definition.)

As a result, alternative interpretations of quantum mechanics have become increasingly popular. And a primary distinction between the older and newer interpretations is that the newer ones are deterministicEverett-style (‘Many Worlds’) interpretations explain the apparent indeterminism anthropically, by suggesting that the observer somehow becomes cut off from an equally real but hidden portion of the wave function. And de-Broglie-style (‘Hidden Variables’) interpretations explain the apparent indeterminism by positing an unobservable difference between the initial state of the two systems, the precise position of the particle.

Both of these types of interpretations have their problems, and it will take a great deal of argument to compare their flaws and merits to those of the Collapse school. But the basic reason Rosenberg is mistaken isn’t that he favors Collapse over its rivals; it’s that he falsely asserts that Collapse is a fact, an observation, a truth of experience. It isn’t. It’s an unverified and unfalsified way of construing  the data. The claim that smoke detectors wouldn’t work if a deterministic model like Bohmian Mechanics were correct falls somewhere between the speculative and the absurd.

Eldred suggested to me that we fortify Rosenberg’s position with an argument that depends less on choice of interpretation, say, “If no experiment can determine whether events need causes [then] no experiment can determine whether the universe needs a cause.” I’m not sure this argument works either, but at least its premise is less speculative, given that the major interpretations of quantum mechanics are empirically equivalent. (EDIT: After talking with Rosenberg, I believe he prefers this version of the argument.)

Craig and Rosenberg both raise a lot of difficult issues, and some of them I haven’t even touched on — like the projects of naturalizing mathematics, morality, and meaning. But this should be plenty to sift through for the moment. If you have any questions, comments, or concerns, let me know! I welcome any opportunities to have my current beliefs upset and overturned.

___________________________________________________________________

Further reading:

How to be a god

What are gods?

In some ways, the question is more important for atheists than for theists. If I’m a theist, after all, I don’t need to understand what it takes to be a ‘god’ in general; I just need to know that my pick of the litter is a bona fide god. It’s the atheists who must speak in broad strokes of all gods, in order for their chosen self-label to even be contentful. These six criteria do the job of pinpointing gods quite well:

1. They are people, capable of thought and action.

2. As a rule, they are (or normally appear) roughly human-shaped and human-sized.

3. They can sometimes be weakened or killed, but by nature they are significantly more powerful and long-lived than a human.

4. As a rule, they are associated with some domain they have dominion or strong influence over, be it a place, a natural phenomenon, or an abstract category.

5. They are natural objects of worship.

6. They are in some way ‘magical‘ or ‘supernatural’.

Within a religion, the gods often form a natural grouping. So, as an added complication, otherwise similar beings may be construed as non-divine if they lack a certain ability or lineage that unifies a tradition’s paradigmatic deities.

Borrowing from E.B. Taylor’s 1871 Primitive Culture, many anthropologists have distinguished ‘god’ worship (theolatry) from ‘spirit’ worship (animism). However, this division, predicated on the theory that cultures naturally ‘evolve’ from animism to polytheism to monotheism, has increasingly fallen out of favor. Cultures have often been labeled ‘animistic’ more because they were seen as ‘primitive’ than because they unambiguously saw everything as intelligent or alive. And even world-views with myriad all-pervading minds can easily blend into world-views with a single all-pervading super-mind. (Compare brahman in India.)

If we do wish to distinguish lesser spirits and monsters from gods, it will need to be because the former are weaker, or less authoritative, or lack human form, or just have a different lineage and name. However, none of these is sufficient on its own. The classification we use will always be arbitrary to some extent, and will depend on the structure of the overall belief system.

Not amused.

Gods interact with humans, and with each other, leading to myths and communicative or transactional rituals. Gods can become human, and humans can become gods, so the human/god distinction can become just as fraught as spirit/god one.

The requirement that gods be ‘natural objects of worship’ distinguishes them from magical beings that lack the power, authority, or virtue true reverence demands. For instance, powerful daemonic spirits may fail to be gods if they cannot or ought not be worshiped, but exist instead to be opposed, manipulated, or befriended. However, some traditions recognize evil gods, and most traditions fail to clearly distinguish religious worship from magical manipulation, so the line between the divine and the demonic is again fuzzy.

Linguistic and conceptual divides mean that lots of interpretive work is required to find a common classification for legendary beings across religious traditions. To see how this works in practice, I’ll present four examples, which I encourage you to treat with some skepticism.

Abrahamism: Early Judaism was henotheistic, believing in many gods but worshiping only one, Yahweh. It shifted from polytheism to monotheism as it came to see rival magical beings as increasingly unworthy even of foreign worship, identifying gods with evil spirits. At the same time, Yahweh’s heavenly court of gods became identified with Yahweh or with his angelic messengers. And the angels themselves, initially treated as manifestations or incarnations of Yahweh, lost their divine status and became lesser spirits. In Christianity, this process worked both ways, as Jesus acquired a status similar to the original angels’.

Buddhism: Buddhas — particularly in their ‘truth body’ (dharmakāya) — are often ascribed attributes similar to the modern Abrahamist God, whereas devas resemble the gods of Greek mythology. (Following the Hindu Paranas, a distinct group of gods, the asuras, were seen as lesser wicked spirits.) However, the word ‘god’ is restricted to the devas so as to highlight buddhas’ unusual features. Upper case ‘God’ is generally reserved for a creator deity like Brahmā or Īshvar. Since Buddhists believe the world has no beginning, it can be said that they deny ‘God’ (issara) but accept ‘gods’ (devas).

Zoroastrianism: Here, the Indian progression is reversed. The supreme god is an asura (Ahura Mazda), while the daēvas are wicked lesser gods, eventually downgraded to demons and ogres. Such developments are often unpredictable or historically contingent; in Germanic religions, asuras again became the ruling gods, the æsir.

Raëlism: This UFO religion, founded in 1974, has no creator gods, spirits, magic, or souls. It is physicalistic, atheistic, and emphasizes science over the supernatural.  The existence of such religions suggests that skeptical and antireligious movements shouldn’t narrowly focus on ‘atheism’. However, Raëlians do believe that we were intelligently designed by a powerful, benevolent alien race, the ‘Elohim’.

Simulation hypotheses posit even more extraordinary creators than Raëlism does. They suggest that we are in a Matrix-like virtual reality, meaning that our entire universe is the product of a transcendent designer. Yet we do not ordinarily think of powerful aliens or computer programmers as ‘gods’. What sets them apart will ultimately rest on the most important criterion I haven’t talked about here, the ‘magical’ or ‘supernatural’ element. Defining the ‘natural’ is a very difficult and messy task, far more problematic than any of the issues I’ve raised above. That story will have to wait for another post.