[lbo-talk] URPE Summer Conference -- Aug 15-18 -- REGISTER NOW! ORGANIZE A PANEL!

Miles Jackson cqmv at pdx.edu
Sat Jul 12 11:59:17 PDT 2008


Jim Farmelant wrote:


> In practice most theories are contradicted by at
> least some of the data that's out there. Defenders
> of a given scientific theory will then invoke ad hoc
> hyotheses to explain, or in some cases to explain
> away, the contradictions between the theory and
> the data. There are no hard and fast rules for
> determining when a theory can be said to be
> "consistently contradicted by data," just as
> there are no hard and fast rules for determining
> whether a theory is a "good scientific theory."

In practice, this is typically not a significant problem for scientists.

If a given hypothesis is disconfirmed repeatedly by various researchers at different locations using various measurement strategies, the theory will be drawn into question, and those who dogmatically cling to the theory will be marginalized. (Except in Economics, I guess!)


> Actually, it is quite relevant to the decisions made by
> scientists as to whether or not to adhere to a given
> theory. Writers like Philipp Frank, Thomas Kuhn,
> and Paul Feyerabend, have all pointed out that
> very often scientists will stick to a theory, even
> if it has been extensively contradicted by data,
> in lieu of there being another alternative theory
> to which they can switch their allegience to.
> Kuhn provided some examples of this in his
> *The Structure of Scientific Revolutions*.

I love Kuhn and Feyerabend, but I read this differently. K and F are not like Popper or the logical positivists; they are not providing a normative account of science. When they say that scientists cling to old theories until a new one comes along, they're not saying that's how science should work; they're just noting that there are cases of scientists doing this. I agree that some scientists tend to cling to theories that are repeatedly disconfirmed by data. However, on pragmatic grounds, I think the principle of empirical disconfirmation is much more helpful than the tendency to cling to a familiar theory. (And not just in science: imagine if the Bush administration had rigorously and thoughtfully applied the principle of empirical disconfirmation to the decision to invade Iraq!)

Miles



More information about the lbo-talk mailing list