Saturday, March 6, 2010

Boston Phoenix on Miller's Accommodationism

I found this interesting article in the Boston Phoenix.

Political philosopher James Madison is, perhaps, the confederacy's most important Federalist.

He has opposed counting slaves as full citizens for purposes of representation in Congress and the Electoral College but not for taxation. He has spilled considerable ink in defense of his position in opposition to full representation based on slave populations from slave states.

But lately, there has been a curious turn in the tale. Madison has come under heavy attack from the negro's fiercest defenders: the Abolitionists, a collection of sharp-elbowed intellectuals who have filled the air with provocative broadsides against slavery.

A flush-faced Gouverneur Morris shook his finger at Madison during a tense discussion at the Continental Congress, proclaiming "domestic slavery" to be in "defiance of the most sacred laws of humanity."

In fact, of course, this is satire. The Boston Phoenix article attempts to place biologist Ken Miller between two extremes of thought on evolution and God. One extreme is the creationists, and the other is the atheists. Miller's position is known as Accommodationism, the view that religion and evolution are compatible. And the point of my little (historically poorly-informed) satire is that positioning Ken Miller between two extremes in no way makes his accommodationist position on religion and science correct any more than the 3/5 compromise was right simply because there were opponents on both sides of the issue. And no ad hominems against Miller's critics can change that fact.

The author, Scharfenberg, describes Miller and his attempt to show science and Christianity to be compatible in laudatory terms, even describing a mentor who returned him to Catholicism in his college days as his "redeemer". This is not exactly neutral, but perhaps my worry about his bias is unreasonable and the criticisms of the "New Atheists" are convincing. Sadly, that does not appear to be the case.

But the cell biologist [Miller] also makes explicitly scientific arguments: maintaining, for instance, that quantum indeterminacy — the ultimately unpredictable outcome of physical events — could allow God to intervene in subtle, undetectable ways.

This sort of sly intervention, he argues, is vital to the Creator's project: if God were to re-grow limbs for amputees, for instance — if God were to perform the sort of miracles demanded by atheists as proof of his existence — the consequences would be disastrous.

Apparently this is supposed to be scientific evidence for the existence of God, but it's not clear how it could be. Here's my best guess for how this can be evidence for God's existence based on predictions we would make if God existed.

1. If God existed, God would not want human beings ever to have evidence that God exists.
2. If God's actions were detectable by humans, then there would be evidence that God existed.
3. Therefore if God existed, God would only intervene in ways that were completely undetectable by humans.
4. It is impossible for there to be something that has no effect on the world. (I'm guessing here; Jaegwon Kim calls this "Alexander's Dictum".)
5. The only way for God to intervene in the world without being detected is in probabilistic quantum events.
6. Quantum events are governed only by probability.
7. Therefore the purely probabilistic nature of quantum events is what we would expect if God existed.
8. Therefore the probabilistic nature of quantum mechanics is evidence that God exists.

If this is scientific evidence, then I can equally prove that Bigfoot exists by appealing to the utter lack of evidence for his existence. If there were a Bigfoot, he would want humans never to find him. Therefore if Bigfoot exists, we would not have evidence for his existence. So the lack of evidence of Bigfoot is the best evidence for his existence.

This argument is obviously nonsensical excuse-making constructed after it became apparent that there is no evidence for the existence of God. Clearly, if there were evidence, Miller and every other theist would leap on it like flies on s*#t. It's part of the standard Christian belief system that there are miracles of various sorts (by saints, by appeal to saints after their deaths, by Jesus, by God) that confirm the existence of God. That's why the Templeton Foundation funds studies on the effect of prayer on recovery from illness. Because they want to find evidence that God exists in an increased probability of survival for those for whom people pray compared to those for whom none pray.

And that's the more reasonable position. If God were an all-good, all-knowing, all-powerful being who would send people to hell for not believing in him (or her), then he would have a moral responsibility to provide adequate evidence for us to believe in him, he would have the ability to provide that evidence and he would know how to do so. So, the existence of a perfect God would imply evidence for His/Her existence, not that there would be no evidence for His/Her existence.

However, the appeal to quantum mechanics is basically a diversion. The fact that physical law is at bottom only probabilistic does not mean if God exists, those statistical laws would not be violated if there were divine intervention. Physicists find that the probabilities for events are constant across large numbers of such events. Even though an individual event may not be determined, but only have its probability fixed only statistically, there are still laws that govern events on a large scale, and if God intervened in quantum events, then it would be impossible for there to be such stochastic laws.

Presumably Miller thinks that God intervenes in quantum events in such a way that the probabilities on a large scale remain the same as they would be without his intervention, but that's not what one would expect from a perfectly good Christian God. Suppose, for example, that cancer is caused by probabilistic quantum events. Since a perfect Christian God would prefer moral Christian believers to others, God would reward their belief and good behavior and would punish disbelief or poor moral behavior. God, then, would prevent or cure the cancers of good Christian believers. This would fit the idea of God intervening only in the probabilistically governed quantum events. But the probabilistic nature of such events does not mean that there would never be violations of those probabilities when there is divine intervention. In this case, one would suppose that good Christians never got cancer, or always recovered from it, whereas non-theists or non-Christians would suffer from them disproportionately. Clearly that would violate a probabilistic law that, say, everyone in circumstance C has an equal probability of getting cancer, or everyone with cancer has an equal probability of recovery. Indeed, we do not find that Christians are much more likely to recover or avoid cancer in the first place. (There have been some studies of the efficacy of prayer or religious social organizations. These mostly show that social organizations, and maybe belief in some set of beliefs or code of living, matters, but they don't show specifically that belief in God is the only way to gain these benefits. Further, they should support benefits for only one specific religion and not all of them, but that is not the case as far as I know.)

Miller, on the other hand, says that we should not expect that Christians would avoid cancer or generally be better off than non-Christians. If this were the case, then we could detect God by such tests, and that would violate his undetectability requirement. (Again, many theists try to find such evidence, but one can only say the evidence is as yet unconvincing. That, I maintain, is why Miller claims that there should not be evidence for God's existence were God to exist.) I'll come back to this argument later, but now my point is that an event's being only probabilistically determined (or probable rather than necessary) is not enough to show that God's interventions in such cases would be undetectable. If God were perfect, and still intervened only when there were statistical probabilities that would allow for such intervention to be undetectable for a single instance. But, at least intuitively, we would expect God to intervene so that His/Her intervention was detectable on large scales. So, the fact that fundamental laws are subject to statistical laws does not entail that one could never detect violations of such statistical laws.

Probabilistic laws are not necessary for undetectability either. God could intervene in otherwise necessary natural laws as long as no one is testing the laws themselves. God's intervention in deterministic laws would be undetectable as long as God made sure no one was trying to detect it whenever God intervened. Perhaps I am too hopeful in thinking Miller, or accommodationists in general, would be offended by the idea of a God who violated our natural laws but only did so when God was sure no one was looking. But if God intervenes in probabilistic laws, that's still an intervention in laws of nature and creates exactly the same problems that might follow for intervention in deterministic laws.

It's obvious that Miller wants the veneer of scientific respectability to be gained by appeal to scientific laws, when it would be clearly excuse-making if we imagined that God might intervene whenever we were not looking. But this "scientific" appeal is irrelevant to his argument; his argument is still transparent excuse-making, just with a thin veneer of quantum nonsense.

But what about Miller's response to the objection that God should intervene for moral reasons. Miller thinks God wants us to believe in him freely, and providing evidence for his existence would undermine that free choice. Here's Miller:

"Suppose that it was common knowledge that if you were a righteous person and of great faith and prayed deeply, all of a sudden, your limb would grow back," he says. "That would reduce God to a kind of supranatural force . . . and by pushing the button labeled 'prayer,' you could accomplish anything you wanted. What would that do to moral independence?"

This is absurd. First, Christians often claim that God (or belief in God) does work wonders, either in one's personal life or in the form of literal miracles. Suddenly, given that there is no evidence of miracles, we conclude that God wouldn't want there to be miracles. If there were miracles, that would be evidence of God's existence. But it turns out the lack of miracles is also evidence for God's existence. Clearly, this is unfalsifiable rationalization for a lack of evidence. Second, why wouldn't God want people to pray to him in order to have their limbs healed? Jesus is supposed to have healed people frequently, and he was perfectly capable of determining whether the people he was healing were sincere and in need and whether they merely wanted to use him. Why couldn't God do as good a job telling the difference between sincere and insincere prayers? The IRS is capable of telling when we're making a real claim on our tax returns, why is God so less knowledgeable than the IRS? Our parents, when we are young, will not let us get away with the mere pretense of an apology, why couldn't God do the same?

Flanders and Swann jokingly say, "Always be sincere, whether you mean it or not." But God could tell whether we meant it when we begged for succor. Deception would be useless; God could enforce sincerity in our prayers if they were to be answered.

More importantly, you cannot undermine someone's freedom by withholding information. People are more free when they have sufficient evidence and less free, or even unfree, when they lack it. Suppose a doctor offered a patient two treatments for her illness. The patient requests information about the success rates of the two treatments, their possible side effects and perhaps their costs. The doctor could not refuse to provide this information on the grounds that such information would undermine her freedom of choice. More information, ceteris paribus, makes us more free.

Can we be morally independent if God always intervenes when we need help? Perhaps moral independence is overrated, but Miller would need to establish that moral independence would be impossible if God answered (sincere) prayers. Is moral independence worth all the suffering that both natural and moral evil cause? This question is especially relevant since natural evils, such as earthquakes, tornadoes, tsunamis, hurricanes, and volcanic eruptions, are caused by God's indirect action, and thus God is the one harming us (since human intervention cannot prevent or change these events).

But I do not think this is necessarily the case. We could be morally independent if God created us to be the sorts of people who always did the morally right thing in every circumstance. If God is all-knowing and all-powerful, then God should be able to create us this way. And punishing people in order to create moral independence is reprehensible. Next time you're on trial for murder (and how often that happens!), try using the moral independence argument in court and see how far it gets you. It's just not morally acceptable to kill people in order to make other people morally independent. If a parent killed one of his children to make the other learn to care for herself, we would rightly lock that parent away and never let him see his children again.

The final argument Miller gives is a practical argument that supporters of evolution should claim that religion and science are compatible since, if the general public comes to see them as incompatible, this will undermine public support for science.

Scharfenberg summarizes the point this way:

Some of that [opposition] may be rooted in religious devotion. But the real motivation is more practical, Ecklund (Elaine Howard Ecklund, a sociology professor at Rice University) says: scientists know that aggressively anti-religious views could threaten public support for scientific research.

It may well be that people will cease to support science, but if the evidence is that they are incompatible, and these scientists are aware of this, then it's irresponsible for them to claim otherwise. In the end, I think, if science and religion were shown to be incompatible (still not a position I have taken in this post) would win out. Science is simply too beneficial for people to reject it in favor of a religion that has never produced any tangible benefits. Presumably Miller actually thinks religion and science are compatible, but these unnamed other scientists appear to be another matter. If they do not think religion and science are compatible, then they are deliberately deceiving the American people if they claim otherwise. Perhaps these scientists are agnostic about the compatibility of science and religion. If so, then they might allow people to maintain their religious beliefs. But even then the responsible position would be for these organizations to endorse agnosticism about the compatibility of religion rather than explicitly endorsing accommodationism as they now, apparently, do.

It is impossible to evaluate the arguments of these unnamed scientists since we are not presented with them, but if this is the entire argument, clearly the better alternative is to educate people about the incompatibility of science and religion so they can rationally decide which to choose. It is always better to stand by the evidence and try to change public opinion rather than avoid confrontation and go along with public opinion in the hopes that things will get better on their own. There are too many examples of regressive social views that were overturned, and public opinion changed, only because people had the courage to argue, protest and put their lives and careers on the line to support the truth. Without any further argument that accommodationism is actually correct, scientists and philosophers have a responsibility not to support it, and even, if the arguments support this, to argue against it.

Update (3/27): Revised for clarity.


  1. I'm starting to think you have an agenda.

    What gets me about this piece is you quote the article about Miller and his claim that "quantum indeterminacy ... could allow God to intervene in subtle, undetectable ways" and then conclude "Apparently this is supposed to be scientific evidence for the existence of God." But a mere half a dozen sentences after the section you quote it says: "Miller insists ... he is not saying that indeterminacy is proof of God's existence, but rather that it allows for God's existence." Why spend so much effort refuting a claim that has not even been made?

    I'm not commenting to defend Miller. (Before reading this article I'd never heard of him. - And based on the article I don't know that I think that much of him.) Miller's claim that you need to look at quantum mechanics to allow for God's existence is silly. I think the idea that the article attributes to Gould (religion and science do not overlap) is essentially correct. It is unnecessary to defend religion from science; religion addresses questions that are outside the realm of science.

    I think Miller is correct is that it would be foolish to let science become the domain of one particular religion (Atheism). And it would be foolish for valid political reasons: many Americans (and many people in other parts of the world as well) reject Atheism and anything associated with Atheism. Atheism is not necessary for science.

  2. First point: I didn't notice the following passage. Thanks for pointing it out.

    But for the New Atheists, Miller's focus on quantum indeterminacy sounds a lot like a classic ID formulation — a "God of the gaps" argument suggesting that anything not explained by science is proof of God's existence; an argument that grows increasingly tenuous as science expands its explanatory power and believers shrink into smaller and ever more ornate gaps.

    Miller insists there is a difference: he is not saying that indeterminacy is proof of God's existence, but rather that it allows for God's existence. It is, naturally, a distinction that carries little weight with his opponents.

    This passage is not exactly consistent with the previous claim that Miller is providing scientific evidence for existence of God. Apparently, according to this claim, Miller is saying that quantum indeterminacy allows for God to intervene without detection, so the lack of evidence for God's existence does not disprove it. Now, as I argued, I do not think this is true. (I tried to clarify this point in the post before I read your comment.) But even if this were correct, showing that something was not convincing evidence against a claim is far different than showing that there is evidence for the claim. So, the author's description of Miller's view would simply be inaccurate. Perhaps that is simply a problem with the journalist's description rather than with Miller himself.

    On the other points, I did not argue against accommodationism, I just discounted the arguments for accommodationism presented here (such as they are). So, as far as this post goes, there is no guarantee that science entails, or even increases the likelihood of the truth of, atheism. That would take a much longer post.

    Does religion address other issues that science does not? I agree that this is so. The question is whether it addresses questions that are not more properly the province of some area of rational inquiry and whether it addresses them in a rational way.

    As to the claim that atheism is a religion, I think it's fairly obvious that this is not the case. I do not have a definition of religion (and see my post on demands for a theory to see why I think this is unnecessary), but there are at least clusters of properties that religions have in common, and atheism lacks (nearly?) all of them. Religions tend to have a set of dogma or doctrines that one adheres to by faith, that one cannot reject them based on evidence; there are characteristic activities--say, rituals--and social organizations one is a member of by virtue of membership in that religion. Atheism really has none of that. Atheists, by dint of being atheists, disbelieve in God, but that doesn't make atheism a religion. Plumbers believe in plumbing by virtue of their job as plumbers, but that does not make plumbing a religion. Some might be atheists without reason or based on no evidence, but that's not essential to atheism, and it is certainly possible to be an atheist for convincing reasons. (You may doubt this, but that's another post.) So, you'd need to do a lot to convince me that atheism was a religion; it's primarily a rejection of religion, not simply a different religion.

  3. I think it's fairly obvious that atheism is a religion. The fact that we disagree on this matter indicates to me that we probably have different ideas about what constitutes atheism, what constitutes religion, or both. So I'll rephrase the first sentence in the last paragraph of my previous comment in an attempt to be more clear:

    I think Miller is correct is that it would be foolish to let science become the domain of one particular unprovable set of beliefs that is orthogonal to the questions of science.

    Miller is correct: if people believe they must either abandon science or abandon God, then many (most?) will abandon science. Fortunately this is not necessary. The challenge is to convince those who have already abandoned science in favor of religion that they can come back to science without abandoning their religion.

  4. Denying the existence of something does not require proof. Do you need proof in order to deny the existence of Zeus? How could you prove a negative? Or is it rational to disbelieve in something until you have evidence for it? Virtually the only exception people make to this rule is God. What reason could there be for that?

    The lack of evidence for the existence of God is sufficient reason to deny that God exists.

    Secondly, if God is perfect, then God's existence is inconsistent with the existence of unnecessary evil. And there is unnecessary evil. So there must be no perfect God.

    I'm not sure what you demand as proof (obviously I cannot prove that you exist), but this is very strong evidence (even though no evidence is actually necessary).