On Being Wrong

On Being Wrong

by Adam Frank for McGraw-Hill

Science is not simply a system of knowledge. It is more than textbooks and lab notes and telescopes and research papers. If that were all there was to science then its ability to inspire and push us forward as a culture would be pretty limited. Yes, the super-cool images beamed down to Earth from the Hubble Space Telescope give us a grander view of who we are and our place in the grander scheme of things but over its 2500 year history science has done more than simply give us cool results. It has, in its best form, asked us to ask more of ourselves. It asks us to act better, to do better. I found this myself this summer in the process of making a scientific mistake with a capitol "M".

The whole point of being a scientist is to create new knowledge. You do research on some topic that interests you and then you write up your results in the form of a paper. You submit the paper to a well-respected scientific journal and the journal sends it to one or two "referees" who read the paper and make comments. The comments can take many forms ranging from nasty - "Your ideas stink, it can't work like this, do not publish this paper"- to helpful - "Excellent idea can you include more details in section 3?" Either way, through this refereeing process, the community polices itself attempting to make sure that material that gets into the journals is correct or at least passes the rigor of clean reasoning.

This summer I submitted a paper on how stars like the sun end their lives. There are tremendous clouds of gas moving away from dying solar-type stars and they take on some of the most extraordinary forms. My premise was that magnetic fields could help blow material away from the star and help shape it into the beautiful sculptures we see on the sky. I was very excited about the paper but when I received the anonymous referee's report I was shocked. It said - "I can't reproduce equation 30". When I first got the response I though that I had simply not made things clear enough but after checking and rechecking and rechecking again I found to my horror that equation 30 did not follow from the other equations. Even though I had proof read the paper and my math many times I had missed something fundamental. The paper and its results contained a significant error and this is where the ethics of science come in.

In politics or business these days it seems that if you make a mistake you blame someone else. Find a scapegoat, come up with an excuse, make it somebody's else's problem. Science won't let us off the hook that way. Our job as scientists (and here is where science can and has inspired the culture as a whole) is to simply be honest. We need to look at the world with great care and then report on what we have seen. We do not report what we want to see, only exactly how we have seen in it. The ethic of science is one of deep integrity and honesty. In my case I had made a mistake. That was it. That was all. There was nothing to do but write back and say "You can't derive equation 30 because it its underivable. I withdraw the paper. I apologize".

My example is simple but looking wider you can see that when science addresses issues that impact the whole of society: health, the environment, and global warming to name a few, that ethic becomes paramount. We have to take the world as it is, not how we want it to be. We need to watch, be honest, admit mistakes and use what we know to guide and manage our impacts on society and the planet.

I have since gone back and re-derived my results and seems that while the particulars may change the basic premise is still correct. Magnetic fields can blow material away from dying stars. I am pretty happy about that but what is more important is that I got a real lesson in the ethics and responsibilities of science.

Points to Ponder

1. Think of examples were the scientific ethics come into conflict with society. When does sticking close to the data cause friction with politics and policies

Websites on Dying Stars

http://oposite.stsci.edu/pubinfo/pr/97/pn/
http://www.noao.edu/jacoby/pn_gallery.html