So, how do we make decisions, especially moral decisions, in the absence of absolute knowledge?
There is one field of human knowledge that spends almost all its time on this question – the law. In civil and criminal cases, judges and juries are routinely asked to suss out the truth in the face of conflicting claims, and make weighty decisions on the basis of their findings. The court system is designed to be as fair and true as possible while still being pragmatic. What can the law tell us about practical epistemology?
For one, courts use different standards of evidence for different kinds of cases. Thanks to Law & Order and other such courtroom dramas, we're all familiar with the standard of evidence for murder trials: the defendant must be found guilty beyond all reasonable doubt in order to be convicted, and twelve different people must all agree. Lesser standards prevail in most other cases. Many legal principles use the "reasonable man" test: what would any reasonable person conclude, given certain evidence? The important point here is that our standards for evidence vary with the case, and extraordinary measures (e.g. executing someone) must be justified with extraordinary proof (i.e. beyond all reasonable doubt).
The other important concept outlined in the law is burden of proof. The law recognizes that epistemological confusion can justify anything or nothing, depending on your inclination, so it determines ahead of time what course should be taken when insurmountable doubts arise. In nearly all criminal proceedings in the United States, the burden of proof is on the prosecution: if reasonable doubts about a particular case exist, criminal law presumes the person is innocent. (Note that in lots of other non-legal proceedings, the burden of proof is quite different; when a salesman is trying to convince us to buy a used car, the burden of proof is upon the salesman to prove the car is not a lemon, and even the slightest suspicion of problems can be sufficient grounds not to buy the car.) The important point here is that we need to decide what's the reasonable "default" to take with certain kinds of cases.
Does any of this help with our ethical questions of helping the poor? The standard of evidence required is going to vary with just how big a claim you make about our moral responsibilities to alleviate poverty. If you say everyone should give a little bit to help the poor, then you can probably get away with the simple "fed people are better than starving people" argument. If you are going to claim (as Kenny did) that everyone has an infinite duty to help the poor, to the extent that they should give away everything until they live in poverty themselves . . . well, that's going to demand a pretty high standard of evidence. It almost certainly flunks the "reasonable man" test; a grand jury would probably approve of pressing murder charges against a man who stands by and watches a girl drown in a pond, but would never dream of prosecuting someone for not giving to Oxfam. The prevailing "reasonable" position is that giving to the poor is good, but superogatory. And, I believe, the burden of proof is upon Kenny to prove that it's otherwise.
Again, none of this proves that Kenny is wrong about helping the poor. Principles of law are often not the same as principles of ethics, nor should they be. I just want to make it clear that I'm not hiding behind an epistemological fog. Reasonable standards of practical evidence do not demand the ethical standard that Kenny is proposing, at least with the rationale he has presented so far.
Dude. All I'm claiming is that helping a child grow up healthy and productive, instead of suffering permanent damage from malnutrition or disease, is more important than a mocha cappuccino. While I have to grant that this is not defensible in any ultimately logical sense, it hardly seems like an extraordinary claim that requires some extraordinary proof.