To Lois: Asymmetry in Causation
Wed, 6 Mar 1996 23:17:33 -0500
Content-ID: <>
Content-type: text/plain

Attach file.
Content-ID: <>
Content-type: text/plain;
Content-Transfer-Encoding: quoted-printable

I will post this response- to your request for =

clarification - publically, in case anyone else =

has the same questions. Your questions really =

get to the heart of the issues.
The actual numbers I put in the grid were
not meant to simulate any type of synthesis.
They were chosen at random and should not be =

interpreted as anything but an illustration of how we =

might set up a grid to study formal causes. With a =

real grid, the correlations that exist would be an =

empirical matter. The illustration is not a real grid.
Causation is not correlation. Some correlations are =

noncausal.They reflect incidental clumping of =

characteristics. It would be syncretistic to equate =

correlation with cause. =

Lets say education is PART of wisdom and =

competence, but only part. Wisdom would also =

require being careful and perhaps other constructs. =

Similarly, Competence is only partly made up of =

education. Talent, motivation, and other constructs =

may go into making up competence. Wisdom and =

competence may be constructed from education ( I am =

not saying this so in reality- just a hypothetical =

example), as well as being constructed from other =

constructs. Thus wisdom and competence would =

be more complex than education. =


Wisdom, would, if this model were true, logically =

imply education. Education would not logically =

imply wisdom- even if it has some statistical power =

to predict wisdom. There may be some foolish but =

educated people. Competence would also logically =

entail education. Competent people are educated. =

But education would not logically imply competence. =

Some educated people are incompetent. This =

asymmetry of implication is the core of the issue and =

is what distinquishes corresponding regressions from =

correlations and other correlation based methods like =

Wisdom and competence would not logically entail one =

another, even if their existences are correlated, =

incidentally. Wisdom could exist without competence, =

competence without wisdom. They are not subsets of =

one another, even though they may both separately
include education as one of their own subsets. They =

do not cause one another just because they share a
common subset. =

Thus wisdom and competence would not simply be a =

euphemism for education, since they are more than =

education and do not equal one another.. =

In the real world we do not know what causal model =

the person is using. By using corresponding regressions, =

we can discover what model they are implicitly using
(I think). =

Bill =