Descartes made a very strong distinction between the kind of truth you look for in philosophy--the truth of absolute certainly beyond a doubt--and the kind of truth that you have to use as a basis for action, which is more probabalistic, based on the best evidence you have at the moment. Both are important, but don't get them confused. That's the big distinction I see here. I want moral truth on the level of absolute certainty--this is, in fact, the very heart of my own justification for the life of the mystic (shameless plug for another of my own essays, I know, but really, it's all based on the necessity of finding some absolute basis for morality, and the impossibility of doing so through reason).
At the same time, there is the practical level that you have to act on. From all I can tell, right now, well-fed people are better than starving people. That unjustified assumption gives enough basis for making decisions.
Ok, all philosophic discussions have to run up against the questions of epistemology: "How do you know what you know?" Absolutely everything gets harder to talk about once you recognize the fact that you have incomplete and incorrect knowledge. In college my twin brother and I took a seminar class called "Personal and Disciplinary Approaches to Truth," in which a bunch of different professors from various departments discussed how their particular field (science, engineering, design, etc.) thought about truth. For our final exam we had to write papers about what we had learned from the class, and then in the last class meeting the instructors asked us to pick out the one most important sentence from our papers to share with the class. I don't remember what I wrote, but I do remember my brother's: "You could be wrong."
The problem with epistemology, of course, is that it has no end. There are an infinite number of explanations that can be made to explain our observed phenomena. Maybe the world was made by an omnipotent God. Maybe it was created by a Flying Spaghetti Monster. Out of the millions of potential models for the world, which one do we pick to guide our actions?
Epistemology can be used to forestall action forever. We are all familiar with government agencies and officials who, when pressured to implement a policy they don't like, will declare that "further studies are needed." I call it the "epistemological punt" – "We don't know enough yet to know if it's the right thing to do, so let's do nothing." Cowards and couch-potatoes delight in the realm of epistemology, since it gives a never-ending reason to put off doing absolutely anything, forever, for no particular reason. Intuitively, most of us realize that "how-do-you-know-that-you-know-that-you-know-that-you-know" is a trap, a recipe for inaction not unlike death. We need some other mechanism or standard for operating in the realm of partial or conditional truth.
One can, however, make the exact opposite mistake about epistemology. "Oh, since we can't know anything for sure, all models are equally valid." I call this the "epistemological push" – "you can't really prove anything, so I'm going to go right on believing that homeopathy really works." While the Punters would have you do nothing in the absence of sure knowledge, the Pushers believe that anything goes. This form was mastered by the Postmodernists, who declared nothing was true and therefore anything was valid and therefore we have to believe anything they say. Pushing can be used in moderation, too, to justify whatever we happen to uncritically believe, because "that's just the way I see it."
So, getting back to our original discussion about the moral demands of helping the poor . . . Kenny seems to think that I'm Punting. I started questioning the logic and assumptions that he used in his assertions, essentially raising epistemological doubts to the degree that I could say, "You don't know that for sure, therefore your case is flawed and I don't have an infinity moral responsibility to alleviate poverty." And, for my money, I think Kenny is Pushing: he's making claims without sufficient proof, and claiming that the necessity of immediate action is sufficient to overcome any deficits in our absolute knowledge.