Facebook? Unscientific? Whatever next?

The Facebook meme shown below is designed to poke a little fun at the silly stuff that passes for informed comment on FB and it’s welcome to the extent that we can all do with a good laugh. But I also see something in it that is worth teasing out into the daylight, because I suspect it is guilty of the very thing it is poking fun at – a pseudoscientific attitude posing as scientific rationalism.

I have three reasons for saying this. The first is that science is about the investigation of anomalies – stones that fall from the sky, ores that fog a photographic plate, radio signals from space. It follows that in many, perhaps most cases, scientific discoveries begin life as outliers – isolated observations whose very existence is doubted by orthodoxy. So to dismiss observations on the grounds of their singularity is to misunderstand the scientific process.

Of course, there are also good examples of scientists working in a collegiate way to arrive at a sound consensus view, especially in big science such as nuclear physics or radio astronomy. But even here, the infant science was very much a producer of initial outliers – Fermi building the first reactor in the Harvard squash court and the young Bernard Lovell trying to convince his colleagues that the planets are talking to us by radio.

My second reason for criticizing the FB meme is that the existence of a scientific consensus – while often useful – can be misleading when making comparisons with isolated observations. When scientists perform an experiment, they must necessarily correct their measurements for error. In doing so they must decide in which direction to correct their figures. In most cases, they naturally decide to correct them in the direction of the most recent past experiments.

The head of Metrology at the National Physical Laboratory (Metrology is the science of measurement) has given this process a name – “intellectual phase locking”.  A cynic might call it groupthink or more simply, the herd mentality in action.

A good example of this process is the speed of light. Successive measurements were made over decades which the physics community interpreted as asymptotically approaching a true measured value. In the end they gave up and defined the speed of light by international agreement. The most important fundamental constant in physics was decided by a committee, not by experiment.

The trouble with science by consensus in this way is that it is open to being swayed by advocacy groups. A good example of this is the recent disclosure that the sugar industry paid Harvard scientists in the 1960s to put the finger on fat as the main cause of heart disease, not sugar. Some think that a similar process (science by advocacy group, not necessarily by payment) is happening today with respect to climate change, fracking, and other controversial subjects.

My third reason for being suspicious of this apparently harmless bit of fun is that I have been unable to think of a single specific example of it in action (and those who have shared it on FB have been unable to provide such an example.)

This may sound like a trivial objection but it has what I think are important implications. It means that some highly intelligent, well educated people, who pride themselves on their analytical minds, are willing to accept something as self-evidently true without any corroborative evidence at all on the grounds that they are being “scientific” and others are not. And this raises the question, how is it – exactly –   that we come to believe something is true in science?

What made intelligent, informed people share this meme even when they knew of no examples from their own experience? What made them think they were being “scientific”, when actually they were being superstitious?

This is, in my view, one of the most important unaswered question in science, and this meme is not the answer.

Share this post

submit to reddit

Leave a Reply

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

scroll to top
%d bloggers like this: