Thursday 5 November 2009

Is There Such A Thing As A Scientific Fact?

Of late there has been obsession that things should be based on scientific facts which, on the surface, sounds like a good idea. Drug policy should be based on the science, global worming is a scientific fact, the medicines our doctors give us have been subject to rigorous scientific evaluation etc and we will all be sure that we are heading to a bright new future.

But there is a problem, there is, no such thing as scientific fact. Surprised? You should be because it can change the way we view the pronouncements made by scientists.

Let me start at the beginning with something called Scientific Method. Basically it looks like this:

Hypothesis - what, after observation, do you think explains something you have noticed.
Experiment - device experiments to test whether you are right
Analysis - does the experiment prove you right or wrong (start again if it shows you were wrong)
Publish Results - a very important stage (called Peer Review) where others can see if they get the same results and conclusions.

If everyone agrees the outcome is sometimes called a scientific fact when all it actually says is 'The experiments carried out seem to confirm the hypothesis.' Notice that word 'seems' - there always remains the possibility that further experiments will come to different, sometimes contradictory, conclusions.

Let me give you 2 examples, one old and one current.

In 1687 Newton published his PhilosophiƦ Naturalis Principia Mathematica in which, among other things, he formulated methods of calculating the way forces act on bodies to cause motion. At the time these seemed so robust that they became know as Newton's Laws of Motion - laws because they were always right. However in the early 20th century Einstein came along with his Theories of Relativity which, once tested, showed that Newton's 'Laws' didn't always apply.

My second example is still under evaluation but it may turn out to be a very significant find. I, like so many others, was told (at school?) that you tell what the temperature was like in the past by measuring the amount trees grew (tree ring thickness in other words) and we all assumed this was correct. Some researchers in Scotland found they had a good opportunity to check this and their results have come as a very big surprise. It appears that there is virtually no relationship between tree growth and weather but there is one with cosmic ray intensity. If confirmed by peer review the implications for this are enormous; most of what we 'know' about the historical climate is based on tree ring measurements as these were 'known' to be a proxy for temperature but if there is no connection we have been measuring the wrong thing! If our historic data is wrong it throws great doubts on any forecast made using them whether to prove or disprove man made climate change.

So next time someone tells you that something is based on the science remember all it means is that the experiments have shown that that a theory might be totally right, partially right or even wrong (interpretation of data can be very subjective.)

5 comments:

jstainer said...

Interesting post! I have been fiddling with this very idea of scientific fact for a while as well.

What got me thinking about it was the old Newtonian vs new Quantum mechanics and how completely and utterly different those 2 things are, yet how sure everyone was of the first line of thinking.

Also, the fact that something must be testable to be a scientific hypothesis really makes things a little tricky if in fact there are real things that cannot be determined in that way.

Good stuff, keeps me thinking.

Stori Lundi said...

This is an interesting article on the true definition of "theory": http://www.uh.edu/engines/epi1531.htm

It's quite different than the colloquial use.

Still Breathing said...

Stori, sorry for the slow response - post on that coming sometime.

Interesting article which made me glad I'd remembered to use the word 'hypothesis' in my brief attempt at describing scientific method.

Monkey Courage said...

A friend of mine pointed this out to me. Frankly I expect more from an Englishman.

You seem to work under the assumption that scientists don't already understand the principles behind the scientific method. You also miss a crucial step in the description of the method (though you allude to it). Observation is a step in the Method. Then an hypothesis is formed and tested. If, through repeated experiments, the hypothesis is confirmed it becomes a Theory. A Theory is provisional and can be overturned at a moments notice. However, for a theory to be overturned doesn't require a complete paradigm shift. Newtonian Physics is still valid at sublight speeds. Einsteinian Physics is still valid at macro scales. Those Theories were less overturned than they were refined. Newton was on to something. But his study required more delicate measurements to get the complete picture. The same goes for Evolution or Climate Change. Something is occurring. There are correlations. Those correlations must be isolated and tested for causation.

Of course there can be bias. That is why we have peer review. We hope it keeps the scientists honest. And it does.

What amazes me is that the term "fact" is used in response to charges that so called skeptics make against the term "theory." A theory is a provisional fact. But pundits have purposefully obfuscated the meaning of theory as to be practically useless against a credulous layperson. And now that "fact" is being used to solidify what knowledge science have painstakingly garnered along comes you armchair philosophers explaining to the credulous once again that there are no scientific facts and that all knowledge is provisional. And you've conveniently withheld an explanation of the term "theory."

Well done, sir. You've uncovered a non-mystery and further confused the debate.

Still Breathing said...

Monkey Courage,

Sorry for the delay in reponding but I have only recently discovered that I could view comments without trawling back through all my posts. This also means it is some time since I wrote this but I do remember editing it down to try and make it less of a long, dry read. In doing so I only alluded to Observation, didn't try to define theory (I thought it was obvious but the OED describes it as "a supposition or a system of ideas intended to explain something, especially one based on general principles independent of the thing to be explained") and I also fell into the very trap that I was commenting about.

All I can say in my defense is that my son-in-law, who holds a Masters in Chemistry from Cambridge, thought I was basically correct.

I am surprised that you "expect more from an Englishman." but I can only assume you haven't met that many; most of us aren't that bright.

Thank you for your comment, I appreciate all contributions and make no claim about always being right - I just tell it as I see it.

God bless
Hugh