Doug is here referring to "Granger causation". Basically and shorn of statistical language, A "Granger-causes" B if and only if:
1) A precedes B in time 2) A provides information about B which is not available any other way.
Clive Granger (who won the Nobel Prize for Economics for, among other achievements, proving that if a linear combination of N nonstationary series is stationary, there must be some flow of Granger-causation between them) has remarked on occasion that he has often asked people what it is that is involved in "proper" causation over and above "mere" Granger-causation and never got a satisfactory response. I'm not saying I agree with him, just that the guy who invented the Granger causality test puts somewhat more philosophical oomph behind it than the majority of users.
Granger is one of only three Welsh people to have won the Nobel Prize; the other two (the physicist Brian Josephson and Bertran Russell) also have slightly heterodox ideas about causality.
best dd
___________________________________________________________ Yahoo! Messenger - want a free and easy way to contact your friends online? http://uk.messenger.yahoo.com