Re: people, places, things, and ideas

Kragen Sitaker (
Fri, 8 Jan 1999 20:25:19 -0500 (EST)

On Fri, 8 Jan 1999, Robert S. Thau wrote:
> Kragen Sitaker writes:
> > On Fri, 8 Jan 1999, Robert S. Thau wrote:
> > > Economists claim that what they're doing is a science. However, what
> > > you're saying here is that whatever they do fails to meet the basic
> > > criterion of falsifiability.
> >
> > Well, not any particular specific theory, necessarily, but general
> > theories like people wanting to get maximum benefit, yes. The same
> > thing could be said of essentially all pre-quantitative theories,
> > including most psychological theories.
> Huh? It is *certainly* possible to have a non-quantitative theory
> (why the "pre-"?) which is falsifiable. Lamarckism comes to mind.

If a theory predicts that a particular cause will produce two opposite
effects that tend to cancel, any experiment trying to measure these
effects can determine which of the two tends to dominate. But if the
theory is not quantitative, the results of the experiment can always be
explained by the theory.

If a theory predicts that a particular cause will produce a particular
effect, and experiments cannot find it, then if the theory is
quantitative, it is possible to determine whether the predicted effect
should be larger than the experimental error. But if the theory is not
quantitative, it is always possible for the predicted results to be
smaller than the experimental error.

If a theory predicts that a particular cause should have no effect, and
experiments find a definite effect, then, well, I guess the theory is
pretty much shot. So I was wrong there. :)

Lamarckism has not been experimentally falsified, but Darwinism has
quite a bit of experimental evidence, as well as observed mechanisms
that would explain how it works, which Lamarckism lacks. So nobody
gives Lamarck any credence these days.

When I say "has not been experimentally falsified", I mean that no
experiment has been done that conclusively demonstrates that Lamarckism
is incorrect. I think no such experiment is possible -- although I
could be wrong.

> And in any case, economists generally tend to go to the other extreme,
> focusing on quantitative metrics of economic activity, to the point of
> often neglecting important matters simply because they *can't* be
> easily quantified.

>From the little that I understand, theories of economics include
qualitative relationships, along with some quantitative relationships.
(Idealized theories of economics include quantitative relationships
covering everything, but do not predict the real world.) The
qualitative relationships can be jiggered in your model to produce just
about any result you want.

I probably ought to take some economics courses before I continue this
thread, though :)

> > Psychology is, IMHO, at about the same stage of development that
> > medical science was in in 1800 or 1850. And medical science is still
> > in its infancy.
> Maybe so, but we were talking about economics,

Sorry -- I was meandering. Psychology is just another example of a
field that has the same problem. (Although clearly a solid theory of
psychology is necessary to firmly found a complete theory of

<>       Kragen Sitaker     <>
A good conversation and even lengthy and heated conversations are probably
some of the most important pointful things I can think of.  They are the
antithesis of pointlessness!  -- Matt O'Connor <>