Monday, March 23, 2009

"I was raised religious," . . . .

Many people present and justify their religious views by appealing to their backgrounds, the way they were raised, or by their religious affiliation. "Well, I'm a Catholic, and we believe." or "I was raised Lutheran, and we believe....." or "My family has been Buddhist for centuries, and the views we hold are . . . . . "

There's a mistake lurking here. Imagine if someone said, "Well, I was raised as a serial killer, and we believe that more pain is better." "I come from a long line of pedophiles, and we have always done . . . . " and so on. My point is that merely pointing out that one was raised in some fashion doesn't give you any justification whatsoever for its being a reasonable, just, sensible, or moral thing to do. Whether or not the belief is reasonable is an entirely separate question from how you were raised. The comment stems from a fundamental confusion between the causes of belief and the justifications of belief. Analyzing your own belief as an effect of some external causes makes you a helpless machine--you can no more help what you are, in this sense, than a dog can change its breed.

The dangerous side, of course, is that many people feel that appealing to their family or cultural background like this is all the justification they could ever need for believing whatever they believe. It's as if the fact that you were raised that way effectively eliminates any further discussion of whether one should actually believe it. Even if you were caused to believe by your environment, finding that belief in yourself as a result doesn't entitle you to say or think that the claim is true. Only epistemic justification will do that for you.

I think part of this trend arises from our reluctance to criticize religious beliefs and practices and from our concerns to be respectful and honor the rights of individuals. That's fair. But in fact, we'll all benefit more and a person will be respected more if you take the belief seriously and try to understand why it is true (or false) or reasonable (or unreasonable), and not be satisfied merely with "I was just raised that way. . . ." We shouldn't let that go in ourselves or in others. These matters are too important.

Thursday, March 19, 2009

Belief Persistence Despite Discredited Evidence

Humans have a pronounced tendency to believe when they shouldn't, disbelieve despite counter-evidence, and sustain beliefs that are unreasonable. One of the ways that this is manifest is as belief persistence. A number of studies, including this one:

Anderson, Lepper, and Ross, “Perseverance of Social Theories: The Role of Explanation in the Persistence of Discredited Information,” Journal of Personality and Social Psychology, 1980, Vol. 39, No.6, 1037-1049

have shown that people will go on believing something even after their reasons have been shown to be faulty.

Anderson, Lepper, and Ross presented subjects with some evidence and asked them to theorize about it. Then they told the subjects that the evidence they had been given was completely fabricated and false. But when tested afterwards about the truth of their theory based on that evidence, the subjects still tended to believe. In philosopher speak: For any S who forms belief P on the basis of evidence E, belief in P will tend to persist even after evidence E has been completely discredited to S.

The implications here for religious belief are obvious. Spend a lot of time entertaining fairy tales about Adam and Eve, Noah, and people rising from the dead in Sunday school, and even if you discover that the evidence for those beliefs is suspect, the belief has a way of persisting. Once it's in there, it's very hard to get it out. There's a copy of the study here:

Persistence of Belief After Discredited Evidence

Wednesday, March 11, 2009

Proving Atheism and Bayes’ Theorem

There’s a significant difference between the actual probability that some event will or has occurred and an individual’s subjective assessment its probability. People frequently assign high probabilities, for example, to events that are exceedingly unlikely. As a person ages, for instance, she will estimate the likelihood that she will be the victim of crime as steadily higher even though the real odds of being a victim of crime are inversely correlated with age. The demographic that is at the highest risk, young men, estimates their risk as the lower than other groups. Bayesian’s have given us a way to treat these subjective estimates. They are a person’s prior probabilities, because they are the person’s estimates of probabilities prior to any real objective testing.

The incredible thing about Bayes’ theorem is that it allows us to account for the two different starting points in probabilistic terms and it has the facility for us to engage in an analysis that can resolve the issue. Two people can start with very different prior probabilities, but as long as they are committed to revising those estimates in the light of all the relevant information, they will converge on the correct answer. If we make enough observations or figure in enough of the right consideration that are relevant to the prior estimate of probability an answer about the real probability of x will become clear.

“You can’t prove it!” is the sneering response that I get from many believers who hear or read my arguments. The root of the criticism, I take it, is that they will reject my conclusions until I can produce some sort of exhaustive, deductive proof that provides perfect certainty. Perhaps then and only then would the critic be willing to take the argument seriously. But somehow I doubt it. For the dedicated believer, there is nothing, even in principle that could actually serve as counter evidence. No proof, no matter how compelling would be accepted. Perhaps sensing this, they feel confident in saying, “You can’t prove it.” because that is more of an avowal of their dedication to their view, than a broad epistemological analysis of the nature of proof in this realm.

We’ve known since Quine’s famous arguments concerning a priori knowledge, and other developments in the 20th century, however, that there’s really nothing that can be proven in the sense that is being asked for. What counts as a successful proof always depends upon other principles, and which principles are the most reasonable to adopt overall is a complicated, global affair. Ultimately, the theoretical descriptions we give to the world, including our standards of deductive and inductive proof are justified by their capacity to address our epistemological needs. And proving one claim always relies upon other claims that can, with some ingenuity, be called into question. Very little seems to be beyond some form of doubt, so for the determined skeptic, some sort of worries can always be engineered. Of course, that skeptic has much worse problems to deal with than the mere fact that we cannot prove God’s existence or non-existence to his satisfaction. His worries cut equally across all putative knowledge claims, whether they are trivial or momentous. So the atheist need not respond to the skeptic as if the skeptic’s challenges are the atheist’s alone to bear.

The pragmatist and naturalist who appreciates the state of play post-Quine should adopt some interesting and broad epistemological principles. The scientific method wherein we propose hypotheses and then actively seek out disconfirming evidence to test them is the best game in town for pressing forward with the knowledge project. It works. And we just don’t have anything that’s any better. We can’t trust the magical books from antiquity, we can’t trust superstition, or religious intuitions, and we can’t trust tradition or authoritative institutions to be reliable guides to the truth. The only thing we can do is adopt hypotheses that have been submitted to the tribunal of aggressive scientific testing as provisionally justified and true, and keep plugging away. That approach has, by far, been the most successful, and fruitful in the past. Indeed, only that approach has succeeded. And the success has been amazing.

In the big picture, then, all justifications are in a word inductive. The best we can ever do is acquire probabilistic conclusions about the world. How should we understand the various arguments that I have made here in those terms? If one is serious about attending to the evidence at all, then as we saw with Bayes’ Theorem, whatever prior probabilities you have concerning some issue, you should continuously fold new information into those considerations and revise those prior probabilities to achieve the most inclusive and well-justified synthesis you can. We can understand why a child believe in Santa for a time, and the prospect of flying reindeer might not seem that outrageous to them considering the wide range of magical things that inhabit the rest of their world. But at some point when enough information is available to them various parts of the story shouldn’t add up anymore. It should get increasingly difficult to reconcile the Santa view with the rest of what a child learns about the world. The probability that the Santa story is true should diminish to the point of being unreasonable.

Something similar has happened over time with Christianity’s magical claims from the Iron Age and their lack of fit with the modern world. Of course there are still those that believe. But as scientific literacy spreads and as the scientific enterprise penetrates deeper into the mysteries of the world, the cognitive dissonance grows. It becomes necessary for belief to retreat into a more and more apologetically complicated labyrinth. Alvin Plantinga has argued that no findings in evolutionary biology can be at all logically incompatible with theism. But one wonders what sort of theism is left from the one’s that started the movements that have such a stranglehold on the consciousness of their followers today. Could the first believers of Christianity have possibly imagined that quarks exist, or that the life had been evolving for billions of years on Earth? If they were being candid and honest, would they have insisted that none of what we have discovered in science casts any doubt on their God stories? If science’s story about the history of life is true, and we have every reason to think that it is, then what work exactly is there left for God to do? What need do we still have for the notion of God?

So what I am arguing here and in many of the previous posts is that when we take all of the various relevant considerations into account—the chain of custody (see Resurrection? Probably Not), religious zealots controlling the story for centuries (See The Fox Guarding the Henhouse), the Mark Bottleneck, the 500 Dead Gods, The Perfect Word of God, Abducted by Aliens, the 300 Year Gap, Grave robbers or Magic?, You Don’t Really Expect Us To Believe That, Do You?, and so on—it is not reasonable to believe that Jesus was resurrected on the basis of the information that we have. In Bayesian terms, whatever sort of prior probabilities you may have attached to events like resurrections, miraculously written books, and visits to Earth by divine beings, the arguments that I have given should give you reasons to revise those probabilities unless those arguments are unconvincing for some good reason, or you simply refuse to align your beliefs with the evidence. Of course, I could be wrong too. And if I’m going to be reasonable and consistent about this Bayesian approach, then I would need to revise my prior estimates in the light of compelling new evidence. If I am wrong, then I need you to show me exactly where in the arguments that happens and why. I want to know if I’ve got it wrong.

If I’m not, then even if your prior probabilities had led you to think that God’s existence was a virtual certainty, at some point after folding in enough of the relevant information and revising accordingly, you’ll reach the tipping point and what you used to believe will cease to be plausible in the light of current considerations. And that’s as good a proof of anything that anybody ever gets.

Monday, March 2, 2009

Disbelieving the Believers

By and large, belief in God is a socially transmitted phenomena. People today believe because their parents, family, and friends believe. A believer’s source of information about God is acquired through other people. In the larger picture, Christians, Muslims, and Jews who believe now had their doctrines communicated to them from other believers. A belief of a Christian today is built upon the belief of Christians 2,000 years ago who first began to propagate the stories of Jesus. Some of these stories were later codified in a collection that was deemed to be canonical in the Bible, with the stories that weren’t approved being rejected and even destroyed. Nevertheless, believers today take the beliefs of some of the original followers as evidence.

The reasonableness of a person’s belief today, therefore, will be dependent upon the reasonableness of the earlier followers’ beliefs, at least to the extent to which the modern believer bases their belief on the early believers. Modern Christians base their belief heavily on the authenticity and reliability of the early followers’ beliefs. The Gospels, as well as the rest of the Bible, are the primary source of information that the modern Christian has concerning the reality of Jesus, the resurrection, God, and so on. If Jones’ belief depends upon Smith’s, then there are number of important questions that Jones’ must address if she hopes to be reasonable and justified. “I believe it because Smith said that it is true, and Smith is a reliable source of information,” would be a perfectly common and reasonable inference. That is, Jones shouldn’t accept the claim unless she is satisfied that Smith’s belief is reasonable, and that Smith’s standards of justification meet certain minimal standards.

That means the real question about the origin of the modern Christians’ belief is about the reliability of the first Christians’ beliefs.

Consider important epistemic differences between us and them, and their impact on the question of their reliability.

How disposed is a person, in general, to accept or reject claims about supernatural entities, forces, or events? Let’s call this their Supernatural Belief Threshold (SBT). If they have a low SBT, then they are more readily disposed to believe that supernatural claims are true. If their SBT is high, then they do not have a high receptivity to supernatural claims. Of course, many people will accept one supernatural claim but not another. And many supernatural claims are exclusive because accepting one requires or is associated with rejecting some others. Nevertheless, we can roughly separate people into those who would be more or less willing to accept the general notion that there are forces, events, or agents that exist and that occur that are beyond the mere physical world. A person who accepts that Jesus is the son of God, that God exists, and that the influence of the Holy Spirit is present in the lives of many people finds more supernatural propositions to be plausible than someone is an atheist or who denies those claims and claims like them. Even today, there are many people who have a low SBT and as a result accept claims that most of us would reject. Devout believers routinely go to faith healing revival meetings and witness what appear to be miraculous healings. They leave convinced that a real supernatural event has occurred. But a few questions and some simple investigation reveals that the cases are most likely the result of enthusiasm, mistakes, or, lamentably, outright fraud. Faith healers often provide complimentary wheelchairs to people coming to the meeting who are able to walk, and then in front of thousands they magically command that the person’s ability to walk be restored. When that person gets up and walks away from the wheelchair, thousands of people who don’t know about the deceit are suitably impressed.

What effect would it have for a person to have a low SBT? Their error rate with regard to supernatural claims would be higher than it would be otherwise. They would conclude that miracles were more common than they really are, for example. Suppose there were some supernatural propositions that were true and that were well supported by the available evidence. A person with a low SBT would be quite likely, we will assume, to accept and believe those claims. And if they did and then communicated those claims to you, they would be communicating something true to you. But if there were supernatural ideas circulating about that were false or unfounded, this same person would be more likely to believe them too, so he would be more likely to communicate those mistaken ideas to you too. A person with a high SBT would be the opposite sort of source. Presumably, she would find fewer false or unfounded supernatural claims to be reasonable. And if there were true and reasonable ones available, she would be less likely to believe those and to become a reliable source to you about them. She will mislead you in some cases, but in the direction of having fewer supernatural beliefs than are well-supported. The person with a low SBT would mislead you also, but in the direction of accepting more of those claims than are true and well-supported. The former is too reluctant about such claims, the latter is too promiscuous.

We can also think about the level of ignorance (I) that a person have regarding the issues surrounding x. If someone is largely ignorant of the important background information concerning profession football, for instance, like I am, then that ignorance should be factored into whether or not you should accept the claims that I make about it. If my ignorance about some topic is high, then my authority and my believability or my reliability about it is low. So my ignorance about it should diminish the confidence you have about one of those claims being true, all other things being equal.

We have independent evidence that there is a close connection between belief in God and education level. Numerous studies have demonstrated that as education level increases, belief in God drops off. (http://www.harrisinteractive.com/harris_poll/index.asp?pid=359) This study also shows that as a person’s education level increases, their belief in survival of the soul, miracles, heaven, the resurrection, the virgin birth, hell, the devil, ghosts, astrology, and reincarnation drop off dramatically. Gallup Polls have consistently found similar results: http://www.gallup.com/poll/109108/Belief-God-Far-Lower-Western-US.aspx

We can see an important parallel here. These studies show that currently across different levels of education, religiousness, superstition, and supernaturalism are positively correlated with ignorance. When people have more education, they are less likely to believe. Now consider the difference between your education level, and the general level of knowledge that the average American with a K-12 education has and the level of ignorance of a simple fisherman or a beggar living in the first century in Palestine. Almost all of the information that you take for granted, the technology, and the methods for acquiring information were unavailable to them. A tiny fraction of the population would have been literate. Their mathematical abilities would have been worse than today’s average 3rd grader. They did not know that the Earth moves, or what the Sun was. They did not know that the Earth was a sphere. They did not know what caused disease, or pregnancy, or death. It is difficult to exaggerate the extent of the difference between the things that you know as a matter of obvious common sense and what they knew about the world. If religiousness, superstition, and supernaturalism rise as education goes down, then they must have been rampant among the people who had contact with Jesus (if he was real at all.)

Let’s consider one more epistemic variable that affects the reliability of a source. Abstractly, we could think of the general level of skepticism, doubtfulness, or disposition towards critical scrutiny that a person has. If a person habitually reflects on the evidence carefully, makes a conscious and careful effort to gather the broadest body of relevant evidence, and actively seeks out disconfirming grounds for a claim, that, all other things being equal, is favorable with regard to their trustworthiness as a source of information. If a person whose skepticism (S) is high becomes satisfied that X is true, then you could be more confident that it is true, all other things being equal, than you would be if your source for the same claim was someone who is generally gullible, uncritical, and who does not reflect or seek out disconfirming evidence.

The people who are the sources of information about Jesus and his alleged resurrection--the authors of the Gospels, the people who told them the stories, the people who originally heard these stories and then propagated them--would have had a low supernatural belief threshold, at least concerning Jesus stories. They were also ignorant of the broad body of information that we have today concerning religious tendencies, religious group dynamics, psychology, alternative explanations for paranormal beliefs. They were also ignorant of the 2,000 years of examples of allegedly supernatural events that turned out to be easily explainable in natural terms. In that 2,000 years, we have learned a staggering amount about how human psychology works, errors in reasoning, problems in eye-witness reports, gullibility, mistakes, social-religious phenomena, and so on.

The people who were sources of information about Jesus and his alleged resurrection would have been much less skeptical overall than many people who are good sources of information now are. They would not have been trained or practiced or even familiar with the notions of disconfirming evidence, alternative explanations, bias, and justification. Many of these concepts as we are familiar with them in modern scientific contexts are only a few centuries or a few decades old. The multiplication rule in probability was not understood until the 1800s. The scientific principles of seeking out evidence that could disconfirm a hypothesis, or of double-blind, objective investigation protocols have only recently begun to be understood. People relaying stories about the resurrection of Jesus in 35 AD would not have known, or employed these methods.

Again the implication of lacking skepticism is that such a person would be an unreliable source of information. Failing to invoke doubt or to seek out alternative, natural explanations for allegedly supernatural events would result in their accepting and relaying more of those stories than are true or well-supported. These sources of information will be prone to mislead us towards gullibility. If we were to accept their claims, we might acquire some true, well-justified views, but we are also more likely to end up believing more that are not. Someone who is highly skeptical, however, will be a better source in that she will be less likely to accept claims that are false or unjustified, all other things being equal. The downside will be that she will also be less likely to communicate true claims to you.

So the people who would have relayed stories about Jesus’ resurrection in the early years would have been prone to accept supernatural beliefs, they would have been quite ignorant of many of the relevant facts, and they would have lacked the skepticism. These three traits would have contributed to high error rates in their communicating information about important religious matters like the resurrection of Jesus. In general, they would have been highly prone to accept such claims even when they were not true or supported by the evidence. How unreliable? Consider the problem this way: knowing what you know now, would have trusted them for medical advice? information about nature? guidance about how to grow plants? information about the weather? Are there any topics except the most obvious and rudimentary where you would accept them as authoritative sources?

The Romans who were contemporaries to Jesus would have had a wide range of religious and supernatural views. They believed in a wide range of omens, and spiritual phenomena. They accepted the existence of a number of gods. Why wouldn’t you accept their claims about those matters? Probably because their SBT was low, their ignorance was high, and their skepticism was low. You don’t deem them to be good sources of information about such matters. There are countless people today who are much better educated and who have a much better body of background information who make supernatural, miracle, and magical claims on a regular basis, yet you do not believe them. We are surrounded by smart and skeptical people making supernatural claims that we reject as suspicious, yet we accept the most outrageous claims from utterly unreliable people in the first century.

Here is the irony of the problem here. Suppose that Jesus really was a divine being and he really was resurrected from the dead. And suppose that the people who alleged to witness the events surrounding that resurrection did see them and it happened more or less as it has been relayed to us. Even if it all happened and they got it right, their overall unreliability about such matters is so high, you shouldn’t accept what they claim is true simply because it is coming from them. Since they are the only source of information that we have, and since they would have been such untrustworthy sources, we have no choice except to suspend judgment about what they say. Imagine that a NASA astronomer with a Ph.D. from MIT tells me that Mars is the fourth planet from the sun. And suppose that an illiterate tribesman from an isolated village in the jungles of Borneo were to tell me the same thing. And suppose that I didn’t have any other information about the matter that would allow me to corroborate their claims independently. The astronomer would be a reliable source of information who would justify me, but the tribesmen would not, even though what he is saying is true. We must, especially in matters where testimony is the main or only source of information, consider the source before we accept what he or she says. A group of illiterate, uneducated Iron Age religious zealots who had become deeply engrossed and invested in a religious movement are not reliable sources of information about extraordinary, miraculous events that are alleged to have happened to their religious leader. They are the people who you should trust the least for information of this sort. As a result, it is unreasonable to believe that Jesus was resurrected on the basis of the information that you have. And if that’s unreasonable, then it’s a mistake to be a Christian, particularly when you are aware of the poor state of its epistemic foundations.

The common error when modern Christians think about the early Christians, I think, is to assume to a large extent that they were like us concerning relevant epistemic criteria. If you are going to believe because they believe, then you have to adopt this stance. You can’t take their word for it, and simultaneously acknowledge that they were irrational, unjustified, or uninformed about the matter. If I think of them as being more or less like me with regard to skepticism, rationality, and information, then it would make sense for me to believe what they believe. But this projecting ignores the facts: the early Christians were from a radically different, ancient culture that did not have any of the scientific, educational, or historical advantages that we take for granted. Their background, their propensity toward supernaturalism, and their ignorance would have made them radically different, and radically worse, epistemic agents than us. And those differences make them utterly unreliable as sources of information about Jesus.

In 1911, some Californians discovered a man name Ishi who was the last living member of an isolated tribe of Yana indians who lived in the hills near Lassen. Anthropologists were fascinated with the case because he was one of the closest examples ever found of contact between a group of people who were virtually living in the stone age with people living in the modern era. Ishi achieved some level of assimilation and enjoyed some celebrity until his death from tuberculosis in 1916. Ishi was an expert archer, and he was accomplished making stone arrowheads and shooting a simple bow. But Ishi also believed a mystical Land of the Dead where the souls of the Yana had a shadowy existence. Link You might plausibly accept Ishi as a reliable source of information about making and shooting a bow and arrow, but no reasonable person would accept his views about the Land of the Dead merely on the grounds that he said that it was real. To accept the early Christians claims about Jesus, God, and the afterlife would be a comparable mistake.