From: Jed Rothwell
James Bowery wrote:
Quite aside from the fact that "correlation doesn't imply
causation",
Actually it does, as David Hume pointed out. In natural
science, that is pretty much all you have to go on in many cases.
This is an interesting point which includes LENR in several ways in the
sense of thermal gain being correlated with a nuclear reaction. In fact, no
one is correct and no one is wrong when we talk about timing and
coincidence.Turns out ... wiki-the-wonderful has an actual entry entitled "Correlation does not imply causation" which is a phrase used in statistics, which was purposely designed to over-emphasize the point that a correlation in time between two variables does not necessarily (or always) imply that one causes the other. Unfortunately this phrase is statistically false in itself - in that much of the time there is indeed this exact kind of temporal connection. Therefore, as Jed suggests, more often than not - correlation does imply causation. But the intent of phrasing of this bit of logic, having being cast in the negative - is to avoid the "generalization fallacy"... which is very similar to the racism fallacy. Also in law, we have what is known as "proximate cause," which is a recognition that timing can be important, but there is also the recognition that "coincidence" can invariably influence proper judgment of causation - in a negative sense. "Many statistical tests calculate correlation between variables. A few go further and calculate the likelihood of a true causal relationship; examples are the Granger causality test and convergent cross-mapping." "The counter assumption, that correlation proves causation, is considered to be a questionable cause logical fallacy in that two events occurring together are taken to have a cause-and-effect relationship. This fallacy is also known as cum hoc ergo propter hoc, Latin for "with this, therefore because of this." A similar fallacy: that an event that follows another was necessarily a consequence of the first event, is sometimes described as post hoc ergo propter hoc (Latin for "after this, therefore because of this"). IOW - in a few instances, a recurring correlation of the parameter of time will falsely imply causation; and "recurring coincidence" is a valid counter-argument to this implication... yet often, "timing" especially when repeatable, does indeed influence causation in a fundamental sense. Therefore, the bottom line is that there can be repercussions which are best avoided by foregoing this correlation/causation assumption - just as we forego assumptions deriving from race, gender and so on... In science we should seek real proof of causation - and "correlation in time" is NOT real proof... even when it can suffice as good evidence "most of the time". Jones
<<attachment: winmail.dat>>
