"The plural of anecdote is not data." -- [source in dispute]Science aspires to be an all-encompassing way of looking at the world. Some scientists are Christians or Buddhists, but for many, science is their religion. Many people who are not professional scientists have also taken the scientific worldview to heart, as an overarching paradigm of how the world works that leaves little wiggle room for an omnipotent deity to hurl his Olympian thunderbolts. We don't believe in miracles.
I say "we" because I am a scientist, and I number myself in this group. Science is the foundation of my worldview, but...
But does the community of professional scientists authentically embody the scientific worldview?
But has science, as it is practiced today, retreteated from the aspiration to explain all that can be explained?
But has our common notion of the scientific worldview become identified with a kind of mechanistic reductionism, even as quantum mechanics -- which is the deepest and most successful scientific theory in history -- is holistic at its core?Modern science is a career for a select few million people around the world. To an extent we don't like to recognize, our research projects are driven by business considerations. Can we get funding to do this? Does our lab have the resources to address this question? Can this question be encapsulated in a project with an endpoint clear enough to make a suitable dissertation topic for a grad student?
At the center of my concerns is the computer revolution. Computers have made possible some kinds of science that were not possible as recently as fifty years ago. We routinely sift through vast amounts of data to find an outlier, and we imbue it with meaning. A hundred years ago, physicists focused their attention on the small subset of simple equations with analytic solutions. When I went to school in the 1960s, there were entire courses on the tricks that could be used to solve differential equations in a long list of special cases. Now we solve systems of equations numerically and plot the results in a few minutes, not even bothering to check whether any of these tricks are applicable. Even in pure mathematics, computers are performing proofs that involve checking out more cases and more bookkeeping and more symbolic manipulation than any mortal human could perform without succumbing to boredom and its consequence, error.
The danger now is that the tools have begun to direct the science. We collect data not because we think it will help to answer a question of vital interest, but because we can. We have stopped asking the questions that cannot be addressed by collecting more data.
The greatest loss, in my opinion, is that we have dismissed whole classes of observations as "anecdotal evidence" and refused to take their message to heart. Among these stories and one-off observations, there are many that call our fundamental assumptions into question, and scientists, like most humans, become uncomfortable when it appears that their fundamental assumptions may need to change. We are committed to our research agendas and don't like distractions. "Damn the torpedos -- full speed ahead." becomes "Please don't confuse me with the facts." We don't want to look down to notice that the reasoning on which our science is based has cracks in the foundation.
We have become reluctant to ask the kinds of questions that computers cannot help to answer. Too many scientists have developed a contempt for what they call "anecdotal evidence" -- the compelling stories that are the driving force behind our curiosity. The believe that sets of numerical data are the only kind of observations that science should consider. What would they have made of the one-off observation of Michelson and Morley in 1887 that gave Einstein the idea for relativity?
Humans set our roots in stories, and these usually take their force precisely from their unique, irreproducible nature. My first kiss. A July snowstorm during my honeymoon in the Alps. Trying to calm the tears of my younger daughter by the side of a pool, oblivious to the fact my older daughter lay unconscious at the pool bottom. A 1979 scientific meeting at which I was taken under the wing of a Sufi master...
Many scientists and more administrators have come to believe that "if it can't be reproduced, science can't study it." Indeed, if a surprising new result is reported in a journal, other labs will try to reproduce the experiment, and if their results differ, the new result is dismissed as a mistake. Many journal editors have the idea that it is more conservative to avoid printing something that turns out to be wrong than to allow open discussion of speculative new science. Hence, if a submitted manuscript goes against what they believe to be true, they will refuse even it the space in the journal (and the opportunity for discussion that this provides) until the result has been replicated by more than one lab.
You can't do science if you're afraid to be wrong. The business model of maximizing prospects for success is fundamentally incompatible with the conduct of science.
All this is insidious because it looks from the outside as though science is thriving. It's not just more and more articles in more and more journals. Technological breakthroughs are coming along at a pace faster than society can accommodate them. The number of things we can do now that we couldn't do twenty years ago is truly dizzying.
But this success at the top blinds us to a void at the bottom. There have been no fundamental new discoveries in science since I was a child. The first half of the 20th century brought us relativity, quantum mechanics, the new synthesis of Darwinian evolution with Mendelian genetics, the expanding universe, the double helix, the incompleteness theorem of Godel, the 3 degree microwave background and the Big Bang. The last of these was 1964. Has there been any comparable discovery in the last fifty years?
The biggest danger is that we take this lack of fundamental new discoveries as evidence that our basic understanding is now correct, that we have discovered the large principles that govern life and the universe, and it remains for us now to build on this solid foundation and fill in the details. This attitude could spell the death of science.
So what exactly am I talking about? Where is the glaring evidence that science is ignoring, to its peril? Here is my list of 10 areas of reported observation that will change the face of biology once they are addressed: