On the Nature of Science
written by Karen Street for the
workshop, Friends Process:
Responding to Climate Change
co-led by Gretchen Reinhardt and Karen Street at Friends General
Most natural scientists share Karl Popper’s picture of a discipline
where ideas cannot be proved, but they can be disproved. As ideas are
challenged time and again, as they survive the challenges, they gain
credibility. Potential challenges are often proposed along with the
ideas. With the introduction of the Standard Model of Fundamental
Particles and Interactions (quarks and stuff), scientists identified
one kind of particle as a possible weakness in the model, and testing
began there. Many in science consider ideas which are not falsifiable
(string theory, as of today) as falling outside science.
In order to be falsifiable, a model must make predictions. A simple
example: new satellites, and not just the ones Newton could see, would
follow the same laws of motion. Explanations of dark matter predict the
size and mass of particles.
Science is unique in my experience in punishing fraud (chances are you
lose your funding forever). Making mistakes in your field is allowed,
if you confess to it and are polite (in public anyway), but if you say
things in your own field that have been shown to be false, you pretty
much lose the respect of the science community. Linus Pauling, often
presented as an example of what happens when a great man speaks outside
his field of expertise, shows that scientists, like everyone, are
allowed to talk trash about Vitamin C and the dangers of atmospheric
testing of nuclear weapons if those are not their fields of research.
Factual errors disappear relatively quickly from scientific
conversations, most rapidly within the field, but relatively soon among
other scientists as well. I remember one FB friend, a prominent science
writer, taking down a posting he learned was wrong; that behavior is so
unusual among non-scientists, I remember it.
For our time together, we will use the narrowest definition of science,
that it is about the physical world and can be tested in the physical
world. As a result, references to science in this handout exclude much
of economics and other social sciences and psychology.
Consensus in popular use: Kerry wants to order Chinese and Bob wants
Italian and they reach consensus on hamburgers. Scientific consensus is
different. It might help to consider scientific consensus a purer form
of Friends Unity: 100% agreement is not required, but challenges ARE—no
important idea is considered part of scientific consensus unless it has
Sometimes people with non-mainstream ideas (CO2 might not be to blame
for global warming; T-cells have nothing to do with AIDS) complain that
scientists aren’t as open-minded as they are. In fact, science is very
receptive to ideas that challenge conventional thinking; these ideas
are more likely to get Nobel Prizes. This leads to publishing bias, the
tendency to publish what challenges rather than confirms and thus gives
a distorted idea of what scientists are finding in the lab. Scientists
recognize this as a problem, but since they are most interested in what
challenges their thinking, it goes to a lower priority.
Scientific consensus is reached when scientists lose interest in
exploring an idea, because it has been challenged unsuccessfully in so
many ways, so many times, that the idea looks pretty solid, especially
basic ideas like evolution and the effect of greenhouse gases.
Consensus generally is not used to describe what everyone believes
unless those beliefs have been challenged and survived (eg, the belief
was once common that ulcers are never caused by bacteria, but in the
absence of challenge, it would not be considered consensus). Scientists
no longer explore whether evolution occurs, but there are many aspects
of evolution considered exciting and open. Many parts of climate change
science discussed often among the public (Is Earth warming? Is CO2
definitely a factor? Could it be the sun?) are not interesting
questions to scientists, as they were settled long ago. If information
does emerge that indicates that the consensus is wrong, the consensus
is reconsidered, eg, Newton’s laws do not work outside the universe we
see of medium-sized masses and low speeds.
On the other hand, there are fields where expert guesses may coalesce
around certain values, yet no consensus exists. Mainstream predictions
for sea level rise this century range from 80 cm to 2 meters (2.5 to
6.5 feet) but there is no consensus because there are too many
questions and challenges. The contribution from ice sheet melt, which
may (or may not) contribute the largest portion of sea level rise this
century, was omitted from the 2007 Intergovernmental Panel on Climate
Change report for that reason.
A frequent argument from outside science is that scientists use
consensus instead of data. There is no consensus without data. Others
argue that consensus requires 100% of scientists (or even the public)
to agree. There were still scientists disagreeing with a sun-centered
solar system more than a century after Galileo, but Newton, born the
year that Galileo died, assumed the Earth went around the sun in his
work. Heliocentrism was established as fact among scientists by the
time Newton became aware of the issue, yet at his death, scientists
could still be found clinging to the old view. See Steven Sherwood’s
article in Physics Today, Science controversies past and present,
parallels between how long it takes to get 100% of scientists to accept
climate change and other major ideas in science.
Media coverage of medicine, social science and psychology often reports
first results breathlessly, long before there is consensus, making it
sound like scientists are always changing their minds.
Scientists Communicate Results
Scientists consider peer-reviewed (reviewed first by equals, other
scientists in the field) publication necessary but not sufficient to be
included in the scientific discussion. If you don’t submit your work to
peer review, you aren’t asking anyone in science to consider your
ideas. After initial peer review, some papers are given closer scrutiny
by experts in the field. During that second level of review, some
fields of medicine and the life sciences are notorious for how much
published work is found to have inadequate information and analysis to
support conclusions. Further research is done on some ideas (repeating
the tests or doing different tests).
Government agencies such as NASA and Nuclear Regulatory Commission are
counted on to provide higher levels of review, and rare disagreements
among agencies are covered in Science and Nature, two top general
journals. Even higher levels of review are provided by National Academy
of Sciences, International Atomic Energy Agency, Intergovernmental
Panel on Climate Change, and the like. And they are subject to
complaints from the science community published in magazines like
Science. Scientists devoted pages to finding every error in a 900-page
report from IPCC, including the use of one “very likely” instead of the
more accurate “likely”. Especially at the highest level, but often from
the first article, scientists routinely make clear what is known, and
how well, what is not known, and promising paths to expand on or
challenge the thinking.
Yale Forum on Climate Change posted an analysis
of all English-language
climate contrarian books published through 2010, 108 in all. Their
points about zombie arguments (arguments that don’t die no matter that
they are refuted), unqualified
authors and such apply to a number of discussions about the solutions
to climate change as well.
When we bring up scientific ideas in discussions, consider where they
fit in this continuum.
Scientists find it more difficult to communicate with the public than
among themselves for obvious reasons. The subject is hard, and some
words have very different meanings in science and the public discourse.
Eg, “uncertainty” in science refers to a range of estimates, and
“positive feedback” is not good. (See the table
in Physics Today and the
excellent source Climate
Communication.) Or when International Energy
Agency says that keeping temperature increases below 2°C is
challenging, what does the public hear?
In addition to the inherent conservatism of scientists, who don’t want
to say what they don’t actually know, scientists in general, and IPCC
in particular, almost always err on the side of “no drama”, as Naomi
Oreskes has pointed out. This “no drama” approach has led to scientific
estimates of the rate of climate change over the last couple of decades
that more often underestimates, rather than overestimates, the rate of
Scientists Are Always Changing Their Minds!”
Asimov received a letter from a non-scientist explaining that
scientists often think they get things right, but succeeding centuries
show them they were wrong. His response, “The basic trouble, you see,
is that people think that "right" and "wrong" are absolute; that
everything that isn't perfectly and completely right is totally and
equally wrong.” Instead scientists see knowledge as being refined.
Asimov’s example is Earth shifting from flat to a sphere (350 BCE) to
not quite a sphere because it rotated (Newton) to… , while the
non-scientist described a process by which Earth is a sphere this
century and a doughnut next.
“What actually happens is that once scientists get hold of a good
concept they gradually refine and extend it with greater and greater
subtlety as their instruments of measurement improve. Theories are not
so much wrong as incomplete.
“This can be pointed out in many cases other than just the shape of the
earth. Even when a new theory seems to represent a revolution, it
usually arises out of small refinements. If something more than a small
refinement were needed, then the old theory would never have endured.”