Miscellaneous desk
< January 20	<< Dec \| January \| Feb >>	January 22 >

Welcome to the Wikipedia Miscellaneous Reference Desk Archives
The page you are currently viewing is an archive page. While you can leave answers for any questions shown below, please ask new questions on one of the current reference desk pages.

January 21

Unknown causal link

This may sound like science and/or mathematics, but I'm not sure that it is really.

Suppose I posit that there is a mutually causal link between two variables X and Y: both affect each other. I make no claim about what the relationship might be, and I do not claim that there are no other factors influencing the two variables, that is, I do not expect a value of X to have an associated value of Y.

So, am I positing anything? Is what I posit entirely without meaning?--Leon (talk) 14:12, 21 January 2017 (UTC)[reply]

See Causality. You as the proposer are the one to explain whether by "a mutually caused link" you mean a dependence that is predictable, quantifiable, repeatably demostrable in one/both directions, reversible, instantaneous/delayed,... because if you make no such claim, the "caused link" may be no more than a Questionable cause. It's easy to be misled by Post hoc ergo propter hoc e.g. "The rooster crows immediately before sunrise; therefore the rooster causes the sun to rise." and by Superstition. Blooteuth (talk) 15:04, 21 January 2017 (UTC)[reply]

Okay, I'll say that the link must be, principally, quantifiable, but could be a function of infinitely many other variables. Predictable and repeatedly demonstrable: not sure. Reversible: not sure what you mean. Timing: not necessarily instantaneous.--Leon (talk) 15:13, 21 January 2017 (UTC)[reply]

Sorry, I disbelieve in functions of "infinitely many variables". See Reversibility disambiguated. Delayed Action at a distance would raise questions of what is being propagated, how fast and through what medium. Blooteuth (talk) 16:06, 21 January 2017 (UTC)[reply]

What you are positing certainly has meaning. I would express it as

X=f(Y,Z,W)

Y=g(X,Z,V)

You have bidirectional causality, jointly confounding factors Z, and individual ~~confounding~~ factors W and V. Loraof (talk) 17:34, 21 January 2017 (UTC) Actually W and V are not called confounding factors, just additional causes. Loraof (talk) 19:19, 21 January 2017 (UTC)[reply]

Our article causality claims that causes must precede effects, implying that there could be no such thing as bidirectional causality. But usage depends on the discipline; for example, in economics it is common to allow instantaneous causality, as when the current quantity of a good demanded depends on the current price. Loraof (talk) 19:54, 21 January 2017 (UTC)[reply]

It would only allow for no simultaneous bidirectional causality. It is trivial to see bidirectional causality exists in the real world; feedback loops. Fgf10 (talk) 21:36, 21 January 2017 (UTC)[reply]