scientists are wrong (a lot)

I've been following a developing discussion about what characterises scientists (and what happens when some of them go to the 'dark side' of pseudoscience…) and I've just found a posting on Respectful Insolence that I want to share with you. (Follow this link to read the whole thing. And many thanks to Orac for writing it.) The proposition here is that scientists are frequently wrong…


ERV shows great insight in pointing out that scientists are wrong all the time. Indeed, science can almost be defined as a system or a method of self-correction that brings us closer to models of how nature works. An absolutely essential part of science, therefore, is that we scientists must test our hypotheses and try to falsify them. When we attempt to do so, there are generally one of three results:

  1. The hypothesis is not falsified.
  2. The hypothesis is falsified.
  3. The results are not sufficiently clear to falsify or support the hypothesis.

When a hypothesis is not falsified, generally scientists will either try to find new ways of falsifying it until they are satisfied that it takes all reasonable challenges. Alternatively, they will build on it and refine it based on their experiments, after which they try to falsify the new versions of the hypothesis until they succeed. If the initial hypothesis is falsified, scientists generally will move on to a new hypothesis. True, they may not do so quickly or easily; after all, scientists are human too and just as prone to becoming emotionally attached to their favorite ideas and hypotheses, but move on they generally do–eventually. Of course, result #3 is the most common result; the answer is not always immediately clear. Indeed, this uncertainty may persist for years, if not decades, before some scientific questions are resolved. That's what true scientific controversies, and, once again because scientists are human, they can be quite rancorous, on rare occasions even escalating to the point of scientists yelling "bullshit!" at each other at seminars and scientific meetings. (Such meetings can actually be kind of fun.) Over time, however, evidence will accumulate, and experimental results will start pointing towards an answer. Sometimes a dramatic result, a stroke of genius, like Eintstein's Theory of Relativity or Charles Darwin's Theory of Evolution by Natural Selection, will appear like a bolt out of the blue and change everything. However it happens, once a hypothesis is roundly falsified, though, scientists will indeed, even if reluctantly, admit that the hypothesis was incorrect, form a consensus, and then move on to test other hypotheses.

…Perhaps the best illustration of this attitude among scientists was a tale told by Richard Dawkins in The Root of All Evil? about an elderly and esteemed scientist who had held to a certain hypothesis for many years. One day a visiting professor from America came to give a talk and presented evidence that conclusively refuted this professor's favorite hypothesis. Afterwards, according to Dawkins, the old man strode to the podium, shook the speakers hand, and thanked him profusely, and said, "I have been wrong these fifteen years." In response, the audience applauded uproariously. Whether this story is apocryphal or not or whether it's grown with retelling over time, it is nonetheless the ideal towards which science strives. Scientists are supposed to be willing to give up cherished hypotheses if that's what evidence and experimental results show. Of course, the difficulty in doing so tends to be proportional to both the length of time the hypothesis has been cherished and the intensity of attachment.

Leave a Reply

Your email address will not be published. Required fields are marked *