|
Rayiner mentions the Federal Rules of Evidence in his comment. Also writing from a lawyer's perspective, I'll mention the elaboration of rules of evidence that has developed both in formal rules like the federal rules and in centuries of common-law tradition in deciding cases. Presentation of evidence at trial is about letting the fact-finder (sometimes a jury, sometimes a judge) hear from both sides in an adversary proceeding about what happened, and why what happened matters. Some forms of evidence are "prejudicial"--that is, they are excluded from even being mentioned at trial because they trigger known human cognitive biases. Other forms of evidence are "irrelevant"--even if they are completely truthful, they bear no legal or logical relationship to how the case will be decided based on the other evidence at trial. Unlike Internet arguments, court trials also exclude some forms of highly relevant evidence to protect other social goals, for example by reducing incentives to police misconduct or violations of privacy. In law, there is a well developed body of precedent and rule on evaluating the credibility of witnesses. To show evidence that a witness has reason to lie (not just that the witness is directly contradicted by someone else's testimony) is highly relevant in most trials. If you are a lawyer who has ever read a case transcript, you will have no trouble believing that some witnesses testify in court under oath and penalty of perjury, with cross-examination by the other side, and yet still lie. This happens every day. To put some balance into a system in which any witness might have an incentive to lie, the rules of evidence at trial allow introduction of evidence that the witness is, for example, someone who has been convicted of "crimes involving dishonesty," or has a financial interest in a factual issue under dispute, or so on. Some statements about a witness are plainly irrelevant, even under that principle. It will not do to say "The witness was convicted of jaywalking," because jaywalking is not regarded as a crime that shows a tendency to be untruthful. Another aspect of the law of evidence that has heavily influenced my thinking about what people know and how they know it, and thus influenced my behavior in online discussion, is "basis of knowledge" considerations. (I sometimes explicitly mention my basis of knowledge when commenting in some threads. Here my basis of knowledge is having studied the law of evidence with a professor who literally wrote the book on the subject http://www.evidencecasebook.com/author2.asp when I was in law school.) If a witness testifies, "I saw him shoot the victim," and you can show that it was dark on that occasion, and the witness has bad myopia and wasn't wearing his eyeglasses, you are a long way toward undermining the witness's testimony on completely legitimate grounds. I am often puzzled why many Hacker News readers are so credulous about stories when it is plain that the person writing the story has no basis of knowledge adequate for making the extraordinary claims submitted with the story. (University press office hype about the future implications of preliminary research findings frequently has this weakness, http://www.phdcomics.com/comics.php?f=1174 as is well known in the scientific community but still too little known here on HN, much less Reddit.) When a Hacker News commenter suggests that someone has either bias (perhaps as an investor in a company) or insufficient basis of knowledge (perhaps as a rookie investigator before peer review and replication has happened), altogether too many HN replies will accuse the commenter of an "ad hominem" argument. They don't notice that arguments about bias and about lack of sufficient knowledge are even completely legitimate aids to truth-finding on controversial issues. Yes, don't rely on a list of fallacies (at least, not on the list alone) to guide your thinking, but also on deep thinking about the structure of the argument and adequacy and impartiality of the data. (All of the above said, I like the late Carl Sagan's baloney detection kit on the whole, and think applying it to more Hacker News threads would raise the level of discussion here. See also Paul Graham's essay "How to Disagree" http://paulgraham.com/disagree.html for more suggestions about how to disagree thoughtfully and productively.) |