Re: There is no such thing as absolute falsification

From: AllYou! (idaman_at_conversent.net)
Date: 02/02/05 

Date: Wed, 2 Feb 2005 14:40:31 -0500

<reany@asu.edu> wrote in message
news:1107371859.780067.271450@c13g2000cwb.googlegroups.com...
>
> AllYou! wrote:
> > <reany@asu.edu> wrote in message
> > news:1107323654.015590.226920@g14g2000cwa.googlegroups.com...
> >
> > [snip]
> >
> > > I'll rephrase: The set of all empirical knowledge does not
> logically
> > > dictate the use of any particular model.
> >
> > Empirical:
> > 1.. Relying on or derived from observation or experiment:
> > 2.. Verifiable or provable by means of observation or experiment:
> >
> >
> >
> > > And since people are free to
> > > hold to any theory or statement come what may (Quine), people are
> free
> > > to invent any theory they want and claim that is it right.
> >
> > But you've never given any credible justification for the claim that
> *people are free to
> > hold to any theory or statement come what may * unless you're
> claiming that they can do so
> > despite the invalidation of the theory.
>
> Well, I have on many occasions: People can always offer excuses for why
> their theory or statement or model that looks false is really not
> false.

Some people kill babies, but what some people do is not the point. The credibility of a
theory is based upon the extent to which the predictions of the theory have been
demonstrated through experimentation.

Look, our model of nature is not some solid, unchanging, rigid structure. This seems to
come as a shock only to you, but our model of nature is fuzzy and constantly in a state of
flux. Some aspects of it are much more solid than others, and reasonable people can
disagree about the solidity of various aspects of it.

But all of this model building consists of matching predictions to experiments, and
experiments require observations, and observations consists of matching sensory input with
previously acquired sensory input (experiences). The free inventing part comes into play
in creating the theories in the first place. We are free to invent any theory we wish,
but we are not free to make conclusions about the credibility of the theory despite the
results of observations.

> > In that case, why should that theory ever be
> > considered in science?
>
> Theory is the central concept of science. It is an explanation in the
> form of a deductive system. It seems that the underlying question you
> have is how I could be right and science work at all. I'll answer that
> question. Science works, not because scientists know what the true
> theories are, but because men and women of good conscience over all
> make good pragmatic decisions about when to abandon a theory or model.

And men and women of good conscience over all have no choice but to make these decisions
consistant with that which they've observed. Good conscienc and freedom to make *** up
are mutually exclusive. They are bound by what they observe.

> However, sometimes it comes back to life after they do. Are we going to
> approach the philosophy of physics from the pragmatic viewpoint? (Say
> Yes for a change!)

We need to approach physics in terms of a goal, and the goal is to make sense of all of
the stimuli to which we're exposed. We're simply trying to figure out that which affects
us.

> > You claim that science is improperly taught in school, so would you
> teach the uninitiated
> > that any theory is valid?
>
> Of course I would, since to be a valid theory is to be a theory that
> appears to work well (empirically), and there are hundreds of them to
> point to.

No, I said *any theory*. Is any theory valid, or just the ones that work well? At what
point does a theory go from working OK to working well?

> > If so, then what is the sense of the term *empirical theory*?
> > Of what use is a theory which is claimed to be *right* if it's
> empirically wrong.
>
> What if you believe (STRONGLY) that the theory was judged empirically
> wrong for the *wrong* reasons, such as false negatives or prejudice or
> out and out fraud?

What of it? That has nothing to do with my question.

> > > That seems
> > > like a truism to me.
> >
> > I'm sure you feel that way.
> >
> > > What's important in the philosophy of science is
> > > that there is no proof that any theory constructed that way is
> actually
> > > false.
> >
> > What's important to science is that proof has no meaning in science.
> There are only
> > pairings of predictions to observations.
>
> And what if the observations that supposedly "falsify" are false
> negatives. How ya gonna know?

*Know* has no place in science. All we have are more observations and the consistency of
patterns. Those will determine how credible any given theory is, and how solid that part
of our model is.

> > > It's true that if all initial empirical testing of such a theory
> > > is conterindicative, that looks bad for the health of the theory,
> but
> > > it is never definitive. There is no such thing as absolute
> > > falsification.
> >
> > And who claims there is? That does not mean that there are not
> degrees of credibility.
> >
> > > >
> > > > > What I said is that there is no logical connection between
> > > > > the models we invent and reality. What I mean by that is
> > > > > that there is nothing in our empirical knowledge that forces
> > > > > us to adopt (logically deduce) any particular model of
> > > > > anything.
> > > >
> > > > We can be forced to discard models.
> > >
> > > Never!
> >
> > How do you know?
> Just refuse to do so, and offer up an reason why.

But how do you know that we can never be forced to discard a model?

Quantcast