I thumbed through a copy of Carl Sagan's The Varieties of Scientific Experience in the bookstore yesterday (it's also on my Amazon wish list).

I pretty quickly flipped to the back of the book, where they had included some transcriptions of questions that people had asked Sagan when he'd delivered the speeches in the book. There's lots of good stuff there, but I particularly liked this exchange:

Questioner: How do you recognize the truth when it is upon us?

Carl Sagan: A simple question: How can we recognize the truth? It is, of course, difficult. But there are a few simple rules. The truth ought to be logically consistent. It should not contradict itself; that is, there are some logical criteria. It ought to be consistent with what else we know. That is an additional way in which miracles run into trouble. We know a great many things--a tiny fraction, to be sure, of the universe, a pitifully tiny fraction. But nevertheless some things we know with quite high reliability. So where we are asking about truth, we ought to be sure that it's not inconsistent with what else we know. We should also pay attention to how badly we want to believe a given contention. The more badly we want to believe it, the more skeptical we have to be. It involves a kind of courageous self-discipline. Nobody says it's easy. I think those three principles at least will winnow out a fair amount of chaff. It doesn't guarantee that what remains is true, but at least it will significantly diminish the field of discourse.

Ah, it's too bad Sagan is dead. He was one of the clearest voices of reason.

To summarize...three general guidelines from distinguishing truth from hooey:

• Check for internal consistency
  
• Check for external consistency (does it jive with the best available knowledge?)
  
• Be as unbiased as possible
  

This is sort of a truncated version of Sagan's Baloney Detection Kit from A Demon-Haunted World, and it's a nice summary.