Re: Moorad's assumed time line

From: Jim Armstrong <>
Date: Thu May 20 2004 - 22:48:13 EDT

I am inclined to think that something essential has been lost in some of
the discussions here about the nature of science. I suppose I state the
obvious here, but it seems to me that the core of science is the
process, the four familar steps. There is nothing intrinsically
mathematical about the scientific process, though it gets more specific
in accuracy (and confidence level) when it lends itself to mathematical
statement. Physics falls most clearly into that latter category,
involving quantitative instrumentation for much of what it does
experimentally, and wonderfully concise and accurate mathematical
description. However, the basic processes of science do not require
mathematics to come to some conclusions with reasonable confidence. The
level of confidence obviously is a variable that is more subjective as
one moves away from good crisp descriptive mathematical models.

Part of what makes the social sciences [I'm willing to characterize most
of them as science because most anything meaningful is concluded on the
basis of those four steps of the scientific method] softer with respect
to variations from central tendency and levels of confidence is the
underlying human element, statistical variation that derives from human
choice and a typically huge (perhaps indeterminate) numbers of
influential variables. I don't think that is a basis for sniffing at
their status as sciences as much as it is perhaps a call for more
respect on the basis of some important, even life-changing insights
achieved along the way; and perhaps even to better acknowledge their
success despite the difficult nature of the statistics-"endowed" terrain
and what will undoubtedly prove to be the relatively immature state of
the data-deriving "instruments".

I have a sense that we lose something fundamental and important when we
move the definitional focus of scientific practice away from the simple
4-step cycle to concentrate instead on the accuracies of physical sensor
derived data and mathematical models, just one particular subset of the
ways to "do science".

...or so it seemeth to me. JimA
Received on Thu May 20 22:48:29 2004

This archive was generated by hypermail 2.1.8 : Thu May 20 2004 - 22:48:29 EDT