January 30, 2009
Justifying universal and existence statements
My post on how we should implement the Year of Reason by asking religious people why they believe in god provoked quite a spirited back-and-forth in the comments section.
In the post, I said that there was no substantive reason that religious people could give in response to the question "Why do you believe in god?" and I categorized the likely things they would say under the headings Argument From Personal Incredulity, Argument From Wishful Thinking, and Argument From Vague Feelings. I endorsed Sigmund Freud's assertion that religion was a form of mass delusion since so many people believed in something for which there was no credible evidence whatsoever.
Commenter Lucas took exception to my post and critiqued it saying that I should not make statements such as that "There is no sense in believing in something for which there is no evidence" without substantiating them. He said that he himself did believe in god because he had "studied the evidence" and found it to be "extremely convincing".
This is where the discussion in the comments took an interesting and somewhat surprising turn. I asked for an example of the convincing evidence that he had that did not fit under the three umbrella headings that I had given. But Lucas resolutely refused to do so, saying that that was a detour, and that I was using it as an excuse to avoid addressing his challenge to my lack of substantiation.
This post seeks to clarify what seems to be a basic misunderstanding between us about where the burden of proof lies and what kinds of statements need evidence in support of them and what kinds of statements are justified by the absence of evidence against them.
To begin, let me repeat what I said in an even earlier post.
As mathematician John Allen Paulos argues in his book Irreligion: A mathematician explains why the arguments for god just don’t add up (2008), basic logic requires that existence claims and universal claims be treated differently.
Existence claims can be proved but not disproved. "No matter how absurd the existence claim (there exists a dog who speaks English out of its rear end), we cannot look everywhere and check everything in order to assert with absolute confidence that there's no entity having the property." (Paulos, p. 42) But all the person making the existence claim needs to do to prove it is to produce just one specimen. So the burden of proof is on the person making the existence claim, and in the absence of such proof, it is perfectly logical to deny the validity of the claim.
On the other hand, universal claims can be disproved but not proved. For example, the claim that all swans are white can be disproved by producing just one black swan. But no one can prove the universal claim since we can never say we have checked each and every swan. So the burden of proof is on the person denying the universal claim and in the absence of such proof, it is perfectly logical to assume the validity of the universal claim.
My statement that "There is no evidence for god" is a universal statement whose justification depends on the lack of a counter-example, and so according to the rules of logic, the burden of proof is on the person who challenges it to provide that counter-example. Equivalently, the statement by someone that he or she has convincing evidence for the existence of god that does not fall under the three categories that I provided is an existence statement, and again the burden of proof is on that person to provide an example, not on me to show that he or she has no such evidence.
But if we accept to Lucas's rules of logic, it seems like I cannot even make a claim such as that "there does not exist any dog that can speak English out of its rear end" unless I can provide citations from peer-reviewed journals that assert that the authors have checked every dog (or at least an extensive number of them) and found this statement to be true.
But of course that is absurd. The reason we can confidently make such a statement and expect them to be believed even in the absence of controlled studies is because we apply the commonly accepted rules of logic, not Lucas's rules. I have never personally encountered a dog that can speak out of its rear end and base my statement on the confidence that if anyone in the world had such a dog, it would be an event of such enormous significance that it would be publicized widely and known by everyone. So the absence of a counter-example is, by itself, sufficient to justify the statement.
Of course, someone could claim that I should still not say this because there may be a dog somewhere that can speak out of its rear end but that the owner is keeping it secret and that I do not know for sure that this is not the case. But no one would credit such a statement until the dog is actually produced. This is because the statement that such a dog exists is an existence statement, and the burden of proof is on that person to provide the evidence. It reminds of the claim by the Raelians in December 2002 that they had cloned a human being and would produce the baby later. While this generated a blizzard of publicity, when no baby was forthcoming, people rightly concluded that the whole thing was a hoax.
The point is that there are many statements that all of us can and do routinely make that are perfectly justifiable and accepted as such even if they are generalized from our personal experience of just a few cases, provided the negation of such statements would be extraordinary. These rules of logic are so commonplace and so basic that people may not even consciously realize that they are using them.
So I can confidently say that no cows have seven legs, although I have personally seen only a few cows, noticed that none of them had seven legs, but have not done an exhaustive literature search to see if anyone else had found one. This is because my statement that there does not exist a cow with seven legs is a universal statement. Someone who says I am wrong has to produce such a cow.
This is why I can make the statement that the stated reasons for the beliefs of religious people (except for people like Pat Robertson who have a direct line to god) will fall into the three categories: Argument From Personal Incredulity, Argument From Wishful Thinking, or Argument From Vague Feelings. My statement is a universal statement, based on all the reasons that have been given to me over a long time discussing these issues with thoughtful people, and similar to the ones about the absence of seven-legged cows or rear-end talking dogs. Its validity has to be challenged by providing a counter-example.
So getting back to Lucas's concerns, I am not even asking that any counter-evidence he produces be convincing because what is convincing to one person may not be convincing to another. All I am asking is that he produce any evidence at all that does not fall under those three categories because I am really curious what form such evidence would take, the same way I would be really curious to see what a rear-end talking dog would look like.
It is his choice whether he wants to provide such evidence.
POST SCRIPT: Why science and religion can never be reconciled
Jerry Coyne, a professor in the Department of Ecology and Evolution at the University of Chicago, has written a terrific review of two new books by scientists trying to reconcile science with religion: Saving Darwin: How to be a Christian and Believe in Evolution by Karl W. Giberson and Only A Theory: Evolution and the Battle for America's Soul by Kenneth R. Miller.
The review, titled Seeing and Believing: The never-ending attempt to reconcile science and religion, and why it is doomed to fail, contains arguments and conclusions that will be familiar to regular readers of this blog, but it is all in one place and very well-written, well worth reading.