We were discussing the possibility of heritability and psychological traits, in a class today, and the professor made the point that "if people have an implicit theory that, for instance, height and intelligence are linked, and if this theory persists, psychologists will observe a correlation between heritability and intelligence."

He went on to say (and I agree with this), "this linkage, though 'real' isn't because there isn't likely a gene (or genes) that controls cognitive ability and height, but rather, tall people, if assumed to be smarter, will be given opportunities that will enable them to develop into more intelligent people. As a result of this association, (since height is highly heritable), psychologists will observe a heritable intelligence link."

Basically that correlations don't prove causation, with the added little bonus of "teleologies will reify themselves."That sounds great!" I say. And it does.

So I, being the uppity chap I am, offer the following in some form (again, this is all hindsight, and I've been trying to clarify it all day, so it's probably more clear now than it was in the original):

"If there is a widely held belief about a construct, reified by repeated social behavior (performative acts, if you will), then our definition of that construct (and our ability to understand/measure/analyize it) is defined (at least in part) by that belief about the construct."

The response was something to the effect of, "Thats true, but even if our measurement technique was perfect, and it isn't, we'd still observe differences, because intelligence--or whatever trait--will actually change because of the social factors." (Again, not doing justice here; he really is very articulate.)

And then the class was over.

This represents, I think, why I feel out of place in psychology. There is an assumption that psychometrically valid constructs and factors, exist a priori. Therefore the problems with psychological investigation, are a product of the measurement and design, when in fact there's no there, there. That we think if we just work hard enough and are meticulous enough we'll be able to finally understand the essence of construct x, y or z.

Here's a food example: It's like studying french-fryness while encountering only one variety of french fries, without realizing that *either_ other kinds of french fries *or potatoes exist_ and that frenchy fry cutting factories exist; and then deciding that the only way to really learn about french-fryness is to encounter as many different iterations of that only one variety of french fries in as many ways as possible.

Because that would work so well.

Cheers, Sam