Today, let's bet on science. The University of
Houston's College of Engineering presents this
series about the machines that make our
civilization run, and the people whose ingenuity
An article in the August
25th, 1998, New York Times strikes a chord.
It's about how so many scientists place bets on the
outcome of experiments, or on yet-to-be-understood
physical behavior. Scientists make such bets with
one another all the time.
In 1684, architect Christopher Wren issued a bet.
He offered a book worth 40 shillings to anyone who
could derive Kepler's laws from the inverse square
rule which dictated the influence of the sun's
gravity on planets. Newton snapped back that he'd
already done it, but he didn't publish the result
in the Principia until three years later.
While we have no evidence Wren ever paid up, the
Principia does call him one of the great
geometers of all time.
Almost a century before, Kepler bet a colleague it
would take him only a week to predict Mars's orbit
around the Sun from earlier observations. (It
actually took him five years.) Modern scientists
like Feynman and Hawking often bet on outcomes.
The stakes tend to be fairly modest -- typically 10
to 500 dollars, dinner for four, or six bottles of
champagne. These are token bets. They don't reflect
strong convictions. They're bets that people can
afford to lose, and which they often do lose. In
many cases they even hope they'll lose. And
so the bet really appears to be a metaphor for
These betting scientists leave me reviewing my own
lifetime of constructing experimentally-driven
theories of heat flow. I've made many hypotheses
about why things behave as they do, and I've been
wrong more often than right. I take great pride in
a few of my theoretical models that've proven to be
right. While I haven't bet with colleagues, I've
placed a kind of bet every time I set out to do a
new experiment. I've bet my time, energy, and
All this betting is driven, not by conviction, but
by hope. Constant betting reminds us that
scientists aren't experts. If they were -- if they
dwelt in domains of established fact -- then
betting would be fruitless, and science would be no
Scientists have used a stockpile of rigorous
strategies to keep from being deceived by their
hopes. The task of science is to narrow down what
is true out of the welter of things that
might be true. In that, it's not so
different from a horse race.
The science we teach is another matter. We teach
what's been reduced to near certainty by the
scientific process. It's a little like teaching the
record of Kentucky Derby winners, then asking why
students don't feel the excitement of the horse
race. Horse racing is exciting only when we
don't yet know which horse will win.
All that lays a peculiar demand on those of us who
try to teach science. Can we give finished
science the taste and feel of science in the
making? Can we make science back into something
that has students on edge about what might be the
truth of things?
I'm John Lienhard, at the University of Houston,
where we're interested in the way inventive minds