by slomo » Sun Sep 17, 2006 1:15 am
First of all, correlation and causation are not the same thing. [Unless you're prepared to cite the relevant passages of Pearl (2000), or the articles upon which the book is based, I would stick with the heuristic that events can be correlated without being causally linked.] All sorts of unmeasured variables can link the two events: genetic (even if the buffalo are from different "lines", if the "lines" themselves are genetically correlated relative to other buffalo, which given the geographic proximity I would not rule out) or environmental (food, water, etc.). Though I'd doubt that such environmental factors are in play. Still, if the "lines" (however they're defined) have common ancestors even several generations back, I would not presume the events to be uncorrelated.<br><br>I don't want to get into a big argument about this. I should have kept silent, since I usually do when I see all kinds of mathematical nonsense on this board.<br><br>I'll agree that white buffalo are rare and known to be a sign of coming apocalypse among native Americans of certain tribes. I'm willing to leave it at that, without trying to attach exact probabilities to the event. <p></p><i></i>