Re: Religious scientists

Matt Silberstein wrote:
On Tue, 14 Mar 2006 10:49:43 +1000, in talk.origins , John Wilkins
<john@xxxxxxxxxxxxx> in <dv53v4$15a2$1@xxxxxxxxxxxxxxxxxxxx> wrote:

Matt Silberstein wrote:
On Mon, 13 Mar 2006 09:52:20 +1000, in talk.origins , John Wilkins
<john@xxxxxxxxxxxxx> in <dv2c6n$2bn1$5@xxxxxxxxxxxxxxxxxxxx> wrote:

Matt Silberstein wrote:
On Fri, 10 Mar 2006 00:10:46 +1000, in talk.origins , John Wilkins
<john@xxxxxxxxxxxxx> in <dupd11$22ro$1@xxxxxxxxxxxxxxxxxxxx> wrote:

Matt Silberstein wrote:
On Thu, 09 Mar 2006 11:47:30 +1000, in talk.origins , John Wilkins
<john@xxxxxxxxxxxxx> in <duo1fg$17k9$1@xxxxxxxxxxxxxxxxxxxx> wrote:

[snip]

By science I mean the investigation that which is *measurable*. Questions of
naturalness and ontology can safely be left to one side so long as you are
able to make accurate, precise and repeatable measurements of phenomena.
Anything else is either not yet scientifically investigated, or is not
scientifically investigable by definition.
I tentatively disagree: science is the study of that which is regular.
That is, that which (seems to) has* a nature. If Zeus (tends to) gets
angry if you miss his worship, then it is scientific to study his
nature.
How do you ever know Zeus is angry, regularly or not, if you can't measure
that? You need to at least be able to observe the fact or phenomenon first
before you can start to study Zeus. Measurability implies that things are
steady enough to be measurable, so in that sense measurability implies a
"nature". [But gods are notoriously arbitrary and wilful (it's in their nature
:-) ). Zeus may take exception to your measurements and start to behave like a
natural phenomenon like the hydrological cycle for a while just to piss you
off. How would you know?]
I agree that measurability is a major marker for nature. I was arguing
that nature is what we are studying, measurability is the mechanism we
use.
And I am arguing that "nature" is *defined* as "that which we can measure".
No, it is not. Nature is that which something does because of what it
is. Or, if you prefer, the "that" that leads to things acting as they
do. By observation we have discovered that there is some regularity in
that nature. This regularity in nature is what we care about since
prediction is the goal. And measurement, we have determined, is the
best way to start discovering these regularities. Much after the fact
we might do some hand waving elimination of other factors and assert
that the measured (or the measurement even) is all that exists.
Consider this (to the tune of "Losing my religion" by REM):

There are an indefinite number of ways of representing the world, assuming the
world is real (which I do). Which of these is going to serve as "the"
regularities? I can generate any number of hypotheses (trivially, by adding
Intelligent Designers cardinally). What is it we "discover"?

I think you are confusing my understanding of, my modeling of, the
nature of something with its nature.

I claim that we discover only that a hypothesis is empirically adequate, and
more so that whatever other hypotheses are on the board (we cannot compare
only *possible* hypotheses, since, as I said, there are too many of them). We
end up with the best local hypothesis, and on that basis we derive an ontology
(we claim atoms as "real" because atomic theory is the best alternative - but
how does the "real" predicate add anything?).

It sure simplifies my modeling to act as though those things have some
nature.
That said, I have a slightly different take. Science is the process of
predicting future observations (which are closely related to
measurements). It is the future observation that we care about, not
the cause per se. We often speak of testing as the way we judge
models, but that is the tail wagging the dog. We want the predictions,
the models are our way of producing them.
I disagree as a universal claim about science. Some science wants and expects
predictions. Some wants it but doesn't expect it, and some neither wants nor
expects it (e.g., archeology).
As unarrogantly as I can I disagree. First, archeology certainly does
make predictions. Second, I assert that the predictions are what we
want, science is the best way to get them, and if some people in
science have drifted from that no big deal. I don't demand that each
and every scientists do nothing at all but produce predictions. The
process requires a wide range of tasks and requirement creep certainly
plays a role. Prediction enables our three major goals: how to not
get killed, how to get some food/water, and how to get laid.
Archeology makes retrodictions. Or, if you like, it predicts that there *will
have been* some data or phenomenon, if and when we discover it. But if you are
trying to predict the future behavior of societies, that's sociology.

I think you are playing a game now. Retrodiction differs from
prediction only if we fully accept the existence of a real natural
world outside ourselves. Ignore that and anything *I* have not seen is
grist for a prediction.

Time series for data are internal to the representations. Sure, you can
*justify* them in terms of a real world, but they don't need to be.

And so far as your three major goals, you don't need science, or even
accuracy. We are notoriously bad at predicting outcomes (we tend to be
overcareful, jumping at sticks that might be snakes, and failing to
distinguish between poisonous spiders and the really cute orb spiders that
make wonderful webs). But we survive, get food and get laid sufficiently to
persist (although teenaged boys would disagree on the last two counts).

I don't assert that we need science for this, I assert that science
is, by definition, the best method for making these predictions.

Fine. That doesn't detract from my claim that nature is a construct. It merey
means that science is best for our cognitive goals.

Understand that I am a realist (worse, a scientific realist - I think that the
ontology of our best theories is very likely to be natural). But the notion of
there being a "nature" apart from our theories strikes me as quasireligious.
"Nature" is a conceptual apparatus, data are the only link with the
nontheoretical world.

People apotheose nature. It is something we first describe and then ascribe
independence to our descriptions of. But we then tend to project onto that
descriptive construct properties that really exist in our conceptions, not in
the world.

*Science*, on the other hand, is more rigid. It demands statistical rigor,
empirical correlation, and precision.

Because those do the job well.

The cognitive job, yes.

I think science is an evolutionary
process on its own, relatively independent of the exigencies of hairless
upright apes.

Only because making better predictions have, so far, served us well in
that prediction business. Or do you want to assert that it has just
drifted into place?

No, I claim that the prediction business is the result of selection for good
ways of making them and testing them.

[snip]




--
John S. Wilkins, Postdoctoral Research Fellow, Biohumanities Project
University of Queensland - Blog: evolvethought.blogspot.com
Who are you going to believe? Me, or your own eyes?

.

Relevant Pages

  • Re: The Dimension of Time
    ... symmetry by saying that it is by nature unobservable, ... realm of religious faith and not science any longer. ... of the exotic predictions or difficult questions of existing theories. ... (and therefore without the expected level of energy input). ...
    (sci.physics.relativity)
  • Re: Religious scientists
    ... that which has* a nature. ... You need to at least be able to observe the fact or phenomenon first ... Measurability implies that things are ... We want the predictions, ...
    (talk.origins)
  • Re: Religious scientists
    ... that which has* a nature. ... Measurability implies that things are ... We want the predictions, ... I disagree as a universal claim about science. ...
    (talk.origins)
  • Re: The Dimension of Time
    ... >>> relative time simply predicts that time is not navigable, ... which really does little to define science. ... they make different predictions of measurable outcomes in a common set ... It doesn't matter that the theory is self consistent and correctly predicts and explains experimental results (ie that nature works in the way the model predicts). ...
    (sci.physics.relativity)
  • Re: In AD, the particle always exits a single slit
    ... predictions) that says the aether does not exist. ... accurate predictions, and that one does not need an aether. ... is religious faith about what you think nature is. ... theory, then you need to do it like science does it, not the way you ...
    (sci.physics.relativity)