A reader asks about burden of proof in philosophy. I really ought to have a worked-out theory on this, but I don't. Here are some very tentative remarks.
1. In the law it is clear where the burden of proof lies: on the plaintiff in a civil case and on the prosecutor in a criminal case. The party bringing the charge must show that the accused is guilty; the accused does not have to show that he is innocent. One is presumed innocent until proven guilty. To be presumed innocent is of course not to be innocent. It is simply false that one is innocent of a crime unless or until proven guilty. And to be found innocent/guilty is not to be innocent/guilty. O. J. Simpson, for example, was found innocent of a double homicide. But I have no doubt in my mind that he was guilty. I don't mean that autobiographically as a report on my mental state; I mean the S.O.B. really was guilty. Agree with me on this or not, you must agree that someone found innocent can be guilty and someone found guilty can be innocent.
We should distinguish between burden of proof and standards of proof. In the criminal law, the probative standard for guilt is 'beyond a reasonable doubt,' while in civil cases the standard is less demanding: 'preponderance of the evidence.'
2. In philosophy it is not often clear where the burden of proof lies, nor what our probative standards ought to be. (What the hell did you expect?) 'Proof' can be used in a very strict way to refer to a valid deductive argument with objectively self-evident premises. But this is not what 'proof' means in 'burden of proof.' It means something like: burden of argument or burden of persuasion. It means that some claims need to be argued for, and some don't. Or perhaps: there is a (perhaps defeasible) presumption in favor of some claims but not in favor of their negations.
For example, I would say there is a defeasible presumption in favor of the claim that drinking coffee in moderate amounts carries no health risk for most people. So the burden of proof would be on a researcher who claims that coffee-drinking causes pancreatic cancer. And because the evidence that coffee-drinking is harmless is so strong, the probative bar the researcher must clear is correspondingly high. The researcher needs to give strong evidence for his claim; the rest of us don't need to do anything.
Now consider the Holocaust denier, the 9/11 'truther,' the Obama 'birther,' and the Osama-was-killed-in 2001 kook. Clearly, the burden lies on them to make their respective cases, and good luck to them. The appropriate thing to say to those of this stripe is "Put up or shut up." That 9/11 was an 'inside job' is a claim of such low antecedent probability that the case for it must be correspondingly strong.
A more philosophical example is provided by my present dispute with Peter Lupu about the modal principle that states that if proposition p is necessary, and p entails proposition q, then q is necessary. He thinks he has found a counterexample to this principle. Where does the onus probandi lie, and why? It seem clear to me that the burden lies on Peter since he is controverting a well-known principle of elementary modal propositional logic. (See. e.g., K. Konyndyk, Introductory Modal Logic, U. of Notre Dame Press, 1986, p. 32.) The burden does not lie on me since I am invoking a well-established, uncontroversial principle.
Can we generalize from this example and say that whenever one controverts something well-established and long-accepted one assumes the burden of proof? I doubt it. Galileo defied Aristotle and the Church when he made certain empirically-based claims about the moon. He claimed that the moon was not a perfect sphere. As the story goes, the Church authorties refused to look through his telescope. But it is at least arguable that the onus probandi rested on the authorities since they were flying in the face of sense perception.
But I hesitate to say that whenever one's case is based on sense perception one can shirk the burden of proof.
3. I doubt that there is any criterion that allows us to sort claims that need proof or argument from those that don't. Or can you think of one? Some maintain that whenever a person make a claim to the effect that X exists, then the burden of proof is on him. Well, it is in some cases, but surely not in all. If you claim that extraterrestrial intelligent beings exist, then the burden is on you. But if you claim that there are Saguaro cacti in Arizona, then the burden of proof is not on you but on the one who denies it.
Others seem to think that whenever one makes an affirmative claim one assumes a burden of proof. Not so. 'That hillside is studded with Saguaros' said to my hiking companion needs no proof. I shoulder no probative burden when I make a commonplace observation such as that.
4. Burden of proof and the ad ignorantiam 'fallacy.' Gun instructors sometimes say that every gun is loaded. That is plainly false as is stands, but a wise saying nonetheless if interpreted to mean: every gun is to be presumed loaded until proven unloaded. So if person A claims to person B that a certain gun is unloaded, the burden of proof is on him to show that it is unloaded; person B does not bear the burden of proving that it is loaded. Indeed it seems that B would be within his epistemic rights were he to claim that his ignorance of whether or not the gun is loaded is good evidence of its being loaded. But this is an appeal to ignorance. It has not been shown that ~p; therefore p gives us the form of the ad ignoratiam 'fallacy.' But in this case the appeal to ignorance seem nonfallacious. Safety considerations dictate a defeasible presumption in favor of every gun's being loaded, a presumption that shifts the onus probandi onto the one who maintains the opposite.
The situation is similar to that in a court of law. The defendant is presumed innocent until proven guilty, so the burden of proof rests on either the state or the plaintiff. In a criminal case the probative bar is set high: the accused has to be shown guilty beyond a reasonable doubt. Here too there is a legitimate appeal to ignorance: it has not been shown that the defendant is guilty beyond a reasonable doubt; therefore, he is not guilty.
There are 'safety' considerations in both the gun example and the law example. It is because we want to be on the safe side -- and not get shot -- that we presume every gun to be loaded. And it is because we want to be on the safe side -- and not sentence an innocent person -- that we presume the accused to be innocent until proven guilty.
But now what about God? Don't safety considerations apply here as well? If God exists, then our ultimate happiness depends on getting into right relation with him. So why can't one make a legitimate appeal to ignorance here? Now of course from the fact that no one has proven that God does not exist, it does not follow that God exists. That is an invalid deductive argument. That would be a truly fallacious instance of ad ignorantiam. But it is also invalid to infer than a gun is loaded because it hasn't been proven to be loaded, or that a man is innocent because he hasn't been proven to be guilty. It just doesn't follow in any of these cases. And yet we reasonably consider the gun loaded and we reasonably find the accused to be innocent. And so why can't we reasonably presume God to exist on the basis of the fact that he hasn't been shown not to exist? If the burden of proof rests on the one who claims that gun is unloaded, why doesn't the burden of proof rest on the one who claims that God is nonexistent? We don't want to get shot, but we also don't want to lose our ultimate beatitude -- if ultimate beatitude there be.
You can't say that that the burden of proof rests on the theist because he is making a positive claim; for there are positive claims that need no proof. And you can't say that the burden of proof rests on the theist becuase he is making an existential claim; for there are existential claims -- I gave an example above -- that need no proof. Nor can you say that the burden rests on the theist because he is controverting the widely-accepted; the consensus gentium is that God exists.
But I suppose you could reasonably say that the burden rests on the theist since he is making a claim that goes well beyond what is empirically verifiable.
Recent Comments