[To an ASTR 121 student, Jan. 1999]


...On the interpretation of science & how it is practiced.  Yes, this
was a quick discussion without any nuances.  By using the word "pure"
I don't mean to imply ethically pure or isolated from societal
pressures like careers, incomes, and status.  What I meant was simply:
"divorced from possible applications."  Astronomy is an excellent
example of "pure" science in that definition, since there is almost no
expectation that research in astronomy will produce technical
applications in the short run.

The people involved in science are, of course, subject to all the same
pressures, foibles, and character flaws as most people.  They make
lots of mistakes, exercise bad judgement, and have strongly subjective
biases (mostly in favor of their own ideas).  However, the difference
in terms of the results of science is that there is an EXTERNAL
STANDARD for determining which ideas have merit.  That is to compare
to empirical evidence, and anyone is entitled to make those
comparisons and draw their own conclusions.  

For every idea or interpretation which is advanced, there are always
several countervailing ideas (the number increases with the importance
of the issue being investigated).  A number of people will have
stakes, but different ones, in the outcome.  The result is a
continuing series of "shootouts," mostly in the written literature but
sometimes at conferences, between the different ideas.  And over time,
the idea which best explains the data usually clearly emerges.  The
whole process encourages scientists to be as self-critical as possible,
before others can parade their mistakes in public.  This is the
"self-correcting error mechanism" I talked about. 

Young scientists are particularly eager to find errors in their elders'
work, chinks in the armor, because that's how they make names for
themselves.  Established ideas are therefore tested continuously, and
important scientific ideas which have stood the test of time are
really pretty robust.  Yes, they may be found faulty on the basis of
new evidence in the future, but they are consistent with what we know
now.  

A good deal of federal funding on "basic" research is actually driven
largely by curiosity, but this is on the very practical grounds that
learning about the basic mechanisms of nature ultimately winds up
producing useful technologies.  That was a key lesson the government
learned in World War II, and that is still the basis of much of the
funding programs at NSF, NASA, NIH, and DOE.  Thoughtful people know
that research in quantum physics or nucleic acid structures will not
produce applications this year or possibly for 50 years or possibly
ever.  But without the basic research, one cannot move beyond the
current generation of technologies.  If you focus only on obvious
opportunities for applications, you will be stuck with the obvious.  A
good example of basic research producing (immediate) applications was
the discovery of X-rays.  Another was the discovery of the structure of
DNA.  That was made in 1953, but only in just the last few years has
it begun to result in applications---and those will completely
transform medicine within 25 years.  

I don't think you are expressing it, but there is a rather fashionable
extreme school of thought among philosophers of science which
concludes that there is no objectivity whatsoever in the results of
science---that all is subjective and socially/politically
constructed.  According to this view, there is no more substance to
the principles of science than to the 1988 Republican National
Platform.  As I suggested in my lecture, the easiest answer to this
criticism is that the applications derived from science overwhelmingly
demonstrate that science works.  If the lights come on, the airplane
flies, the antibiotic cures, then this "socially constructed" argument
goes down the drain.  I've often wondered what goes through the heads
of these philosophers of science when they discover that their toasters
actually produce toast.  

   Thanks for your very pertinent questions   --- Bob O'Connell