NP Witt List (was something Pynchon-related at some point, maybe)

[andrew at cee.hw.ac.uk]

I wrote:
> > And in the end this is what determines truth, sound data.
> SNIP
> > There is room for debate
> > (ill-advised or not) in the face of any data and clearly such debate
> > cannot be based on the data but on other considerations. 

Mark Smith asked:
> You know, I'm pretty sure I'll be sorry that I opened this can of worms,
> but I gotta ask what you mean here.  Do you mean "clearly such debate
> must be based on something other than data, since the data is there for
> all to see, if they want to, and yet different conclusions are drawn
> from the same data?"

No, I mean when it comes to resolving any question as to the
significance of data clearly the data itself cannot imply a particular
(one might say `natural') significance. That's a petitio principii.
The data needs to be judged by a criterion which is formulated
independent of the actual readings or measurements obtained. Selecting
such a criterion is *not a scientific judgement*.

The usual test of what scientific data means is to see how it accords
with the predictions of one or other (usually mathematical) model. But
note that the choice of this criterion makes no reference to the data
per se. Note also that there is always lots of room for debate over
what it means for figures to accord with the model. You can change the
model, you can change the measure of fit, you can even reject the
measurements. And you can do this even if you consistently obtain the
same results again and again. Why? Because it is up to you/us/
scientists to decide whether you expect things to keep on repeating in
the same way or to be different next time. The model and the previous
readings don't guarantee that readings will always come out as
predicted.

Most of us believe mostly the same things most of the time but that
consensus arises not because the data mandates it but because we
(communally) adopt systems of judgement which lead us all to have the
same expectations of given situations. When that commonality is
rejected due to differing priorities and needs we get people like
Creationists.

> . . . Evolutionists make no claim for perpetual truth.  Good science
> is in a constant state of flux. By and large science operates in good
> faith, or at least that is the principle.  The fact that the principle
> has been corrupted by research grants and funding considerations is
> tragic, but is no reason to abandon the notion of the scientific method
> altogether.  Remember Gould's definition of "fact": "In science, 'fact'
> can only mean 'confirmed to such a degree that it would be perverse to
> withhold *provisional* assent.'

Exactly. But who defines what is perverse? The thing that really bugs
good scientists is the dogma of Creationism, the claim to absolute
truth. The most important thing about most scientific models is that
they are left subject to revision. When prediction and measurement
accord all is well. But when they start to diverge, say as more
experiments are performed or `better' equipment provides more accurate
measurementsor even because the results just start to come in
differently and we realise that things were not as we thought, in
these circumstances the good scientist is willing to question and
either refine or replace wholesale both model and measure of fit
instead of just denying the data. Making that judgement is an art but
it's the sort of art which most Creationists are sadly unacquainted
with. Sufficient to the day are the realities thereof.


Andrew Dinn
-----------
How do you know but ev'ry bird that cuts the airy way
Is an immense world of pleasure clos'd by your senses five