GIGO

The Business of Proof

While it is believed powdered shark skeleton can treat arthritis, rheumatism, the skin disease psoriasis and cancer, no one has conclusively proved what the active ingredient is.

Is it just me, or does a statement like that make you wince too? Am I being picky about wording, or is there really a major leap in logic from the first clause to the second? Am I right to be concerned that this was in a recent press release from one of our Crown Research Institutes?

Perhaps I am being unduly sensitive, but I believe that scientific research has to be rigorous in its approach and uncompromising in its questioning. I have always winced whenever the phrase "we're going to prove that..." is used -- it has unfortunate implications for the way in which the research is being approached, the underlying assumptions that are being made. In effect, it defines the results well before any data is collected.

I recently advised a science fair hopeful that they should be very cautious in setting out to "prove" something, particularly in structuring their project around topics in which they had a personal belief. One of the things I always look for when judging science fair exhibits is evidence that the entrant has thought about why their results differed from what they expected or in what way they might have differed. I'd rather see a couple of sentences acknowledging this aspect of scientific investigation than whole handbooks of carefully drawn straight line graphs.

One could argue that this business of proof -- and related discussions concerning falsifiability, methodological analysis, etc -- is getting into too sophisticated a philosophical area for secondary school students to cope with. But therein we err, because by encouraging them into the idea that scientific investigations set out to prove things, we set science itself up for a fall the minute they realise that much of our "proof" is equivocal.

I can understand, in these days of directed project-based funding, that our scientific institutions are probably just as keen as the neophyte science fair exhibitor to "prove" a pet theory, particularly if it has lucrative commercial potential. But I still wince.

Vicki Hyde is the editor of New Zealand Science Monthly.