Interview: Skeptic Michael Shermer Answers (Part II)

6. Labels

[rickyjames] “Skeptic” is a shorthand label that is synonymous to many
with “atheist”. How do you define the difference between various labels applied to
those who think rationally, and do you think it is desirable for rational adherents
to adopt a new label such as “bright” that has been suggested by some?

[Shermer] A skeptic is a scientist who applies the methods of science to all
questions, who strives to remain open-minded, and who attempts to avoid the confirmation
bias where one seeks confirmation in the data for what one already believes. Skepticism
is not a position one automatically takes (“I’m skeptical”) to claims; rather, it is a
method, an approach to claims (“I’m curious,” “I would like to find an explanation”).
In the last survey we conducted of members of the Skeptics Society over a quarter claimed
belief in some form of higher power (God). Thus, skeptic is not synonymous with atheist.
The label atheist is very specific: one can only be an atheist with regard to the claim
of God’s existence (as opposed to a “theist”). Skepticism is much broader. One can be a
skeptic with regard to any claim, not thus theistic claims. Finally, I have written along
essay on “The Big ‘Bright’ Brouhaha” for Skeptic magazine, posted at
www.skeptic.com, which give the longer answer to the
question about the “bright” label. I very much like the concept of unifying our disparate
groups, and I was, in fact, the first to sign up to be a bright when the concept was first
introduced at the Atheist Alliance International conference in Clearwater, Florida
(my name literally appears first on the sign up sheet, just ahead of James Randi
and Richard Dawkins). However, when it became apparent to me that the vast majority of
people most likely to fall under the “bright” rubric would not use the label, I became,
well, skeptical that it would succeed. For the time being I continue calling myself a skeptic.

7. Thought Control in Democratic Societies

[Drog] Any fan of Noam Chomsky would be highly skeptical that the
U.S. government (and perhaps most other governments in the world) would ever mandate
the teaching of critical thinking in the school systems, because it would be so much
more difficult to control the population if they all thought for themselves rather
than buying into whatever viewpoint the media presents to them. What is your view
on this? Do you think Dr. Chomsky is correct in his assessment of how populations
cannot be disciplined by force and thus must be subjected to more subtle forms of
ideological control via the media? Do you believe that critical thinking should be
taught in schools? Do you believe that it eventually will be?

[Shermer] Critical thinking not only should be taught in schools, it already
is (albeit not as effectively as it might be). In fact, we have invested considerable
time and resources at the Skeptics Society to this very task, developing the Baloney
Detection Kit and the Baloney Detection Books for students and teachers. The idea that
the government (the proverbial “they” in such discussions, as in “they would never allow…”)
would orchestrate a campaign to prevent the teaching of critical thinking because then
such skills might be turned on the government itself, is utter nonsense and conspiratorial
claptrap. Anyone who thinks that the government has successfully quieted or squelched
political criticism hasn’t read a book or newspaper, listened to the radio, or watched
television in the past couple of decades.

8. Education

[Sweetwind] Way back in high school geometry I remember a fellow student
asking (with some annoyance) what good it was to learn how to do a proof. The teacher
replied that it is so you learn to think logically, and I thought that was a good answer,
and sufficient justification to keep it in the curriculum. But it wasn’t until I went
to college that I learned about things like
confirmation bias
and the
fundamental attribution error,
and knowing about these tricks our minds tend to play on us is also important for
being able to think clearly. Dr. Shermer, what would you like to see added to the
standard secondary (or even primary) school curriculum to help our children learn
to think logically?

[Shermer] The very thing we are already working on: the Baloney Detection Kit
for teachers and the Baloney Detection Books for teachers and students. Skepticism is
critical thinking, and more…. Skepticism also includes science and the scientific method,
the best tools ever devised for finding causality and for understanding how the world works.

9. The “right” amount of skepticism?

[Drog] How do we ensure that we maintain a healthy amount of skepticism
towards radical new ideas while at the same time ensuring that we remain open-minded
enough to not dismiss them without due consideration? I suspect that the answer is to
simply subject all new ideas to the same level of scientific scrutiny. But in reality,
is this always the case? Sometimes an area of research may be “stained” by the claims
of pseudoscientists, frauds and charlatans, thus making it difficult for any actual
scientific study to ever be taken seriously (e.g. parapsychology, Atlantis, UFO’s, etc.).
Other times, a sound new scientific theory may be at odds with an existing theory in
which many scientists have a heavily vested interest.

An example of the latter that comes to mind is
The Great Sphinx Debate,
in which
geologists Schoch and West claimed that the Sphinx shows classic patterns of erosion
by rainfall, thus dating it to within 5000 and 7000 B.C.E.–in stark contrast to the
accepted archaeological dating of 2520-2494 B.C.E. The ensuing debate between geologists
and egyptologists, who claimed that the geologists’ dating would topple the entire
framework that they had painstakingly built up over many years and is not supported
by any other archaeological evidence, became extremely heated, indicating (to me)
that scientists are human too and are not nearly so detached and objective as we
would like to believe.

Another example is the work of geophysicist J. Marvin Herndon (interviewed
here
on SciScoop last April), who has spent ten years trying to convince the scientific community
to even listen to his theory that a nuclear reactor resides at the Earth’s core, let alone
debate his evidence.

So, in an age when scientific study is becoming big business, when one’s research grant
(and sometimes one’s very employment) depends on the frequency of one’s published papers and
books, when there are enormous pressures to find and publish evidence supporting one’s own
theories while ignoring any non-supporting evidence, how can we ensure that we are as
skeptical as we need to be but not more than is justified?

[Shermer] One of the most important tasks of skepticism is finding what I call
the exquisite balance: being open-minded enough to recognize and accept a viable radical
new idea, but not being so open-minded that your brains fall out. Unfortunately there is
no simple formula for the right amount of data and evidence that when a critical mass is
reached one should shift theoretical alliances. The data never just speak for themselves.
And in science there is a very social process of debate, discussion, and peer review and
analysis that rolls along until a consensus is reached or until it is clear that more data
and analysis are needed before a consensus can be reached. The goofier the idea, the more
data one needs and the more difficult it will be to convince the experts in the field.
Claims that science is dogmatically closed-minded to new ideas is absurd and gainsaid by
the history of science, which is choc-a-block full of revolutions and paradigm shifts.
Most radical new ideas are simply wrong and that is why they are not accepted. Occasionally,
a good new idea goes unnoticed or ignored or rejected for far too long (for a variety of
social and psychological factors), but nevertheless they do eventually triumph.

10. Specific Example

[rickyjames] What’s an appropriate “skeptic” attitude towards the frontier
mathematical fields of cellular automata / chaos / fractals? In particular, how does a
skeptic reconcile convincing visual evidence and patterns of apparent deeply fundamental
forces at work with biological processes that work in ways that are seemingly totally
unrelated to these mathematical ones?

In Part I of a
recent article
here on SciScoop I effectively used the “Wolfram”
argument in trying to convince people that math has a deep philosophical component.
I said look at this math, it produces things that look like ferns, wow, that means the
math is almost mythically special and has to be tied to some special attachment to the
Universe. Wolfram did the same thing with mollusk coloration patterns that look like
they’re generated with (mathematical) cellular automation processes he outlines. Gee,
both ferns and mollusks sure LOOK good as an example of Mystical Something, and they’re
certainly intriguing enough to merit continuing legitimate interest in the math, but as
a hard-headed scientist I still have nagging doubts and a sense of disbelief.

There seems to be no tangible physical process that biological cells can go
through during growth to match how cellular automations and especially fractals generate
similar-looking patterns. For example, fractals when they generate fern like shapes do
so by reeling off a series of points that are only unconnected dots until you get enough
of them plotted to visualize – THEN the fern like shape jumps out at you. It’s almost
like Impressionist art. How can there be a corresponding Impressionist biology? How
can a cell react to another cell that is so far away in space and time that it hasn’t
even formed yet in the area it needs to be in to complete the fern-shape?

[Shermer] Vol. 10, No. 2 of Skeptic has a cover story by David Naiditch
on Wolfram’s book and theory. Naiditch has written what I consider to be the best and
balanced critique of Wolfram to date. He too identifies the problem of correlating
mathematical products in a computer to biological products in nature (the mollusk
coloration patterns being a prime example). We have a pretty good understanding of the
links between genes and proteins, and how genotypes end up as phenotypes, and it does
not appear to be anything like what goes on in computer cellular automata. Still, to be
fair to Wolfram, his theory is still in its infancy so we should keep an open mind until
more scientists have an opportunity to test his theory.

11. Can We Know?

[Chronosphere] There are, arguably, some certainties in the universe
(maybe a different set for every human, but I will leave this aside for the moment).
Some people jump to conclude, after a determinated certainty has been discovered, that
the set of rules that we use to attempt a description or an explanation of that
certainty is also part of this “objective universe”. To my knowledge, most skeptics
adhere firmly to science, but, and here lies my question, what if the set of rules
(or laws) we think we are seeing are only “inside our heads” and in the end doesn’t
“describe” anything but our map? That’s my skepticism. I think the explanation is
just and only just a set of rules of language in relation to perception. But as time
passes our set of explanative rules changes, and there is no “final set of rules” to
define the universe, as we are always changing and evolving, both as individuals and
as societies. In this sense, science is just the ultimate tool, but I think it will
be not the last one, to try to have a complete set of rules to explain everything.
Nothing more and nothing else. So, it can’t be a foundation to real knowledge.

[Shermer] Science is socially embedded and culturally bound, and it is
produced by humans filled with psychological foibles. So to that extent science
will always be limited, but like democracy, it is the best system we have so we
should make the best use of it that we can. Some areas of science, however, are
much more obviously valid than others. If you want to send a spacecraft to Mars
you use astronomy not astrology. Why? Because it works. This is very clear cut
and simple. Either the spacecraft arrives or it does not. The biological and social
sciences are not so objective. Do more guns cause more crime or less? Are the
racial differences in I.Q. scores a result of genetics or environment (or both)?
Is global warming caused by human activity or natural forces (or both)? In such
areas we must be especially vigilant because of the numerous nonscientific forces
at work that make it more difficult for us to really know reality.

facebooktwittergoogle_plusredditpinterestlinkedintumblrmail