The single principle I should teach the world is that there is really no such thing as science - science is merely systematised, institutionalised common sense. The infant - human or other animal - learns, from their experience of trial and error, what to eat, avoid, mate with, etc. That learning from the experience of trial and error is already science.

Doomed is the forager who is not a scientist - who does not learn to avoid the grains that have made him sick in the past, favour the terrains that have been plentiful, or crack the nut before sinking his teeth into it. Doomed, too, is the organism that is indifferent to the outcome of its experiments - the one that keeps sticking its nose back into the fire.

We humans have a second way of doing science, over and above individual trial-and-error, or experimental learning guided by the error-correcting feedback arising from the consequences of our actions. We also have language, and we can save one another a lot of risky and time-consuming experimentation, by telling one another what's what.

Of course, we can also lie or be in error. So our statements can be true or false. But the true statements - 'do not eat the mushrooms with the stripes, they will make you sick', or 'find a rock and smash the nut, and you will find something good to eat inside' - are our second way of doing science, by describing the outcomes of our experiments and by sharing them with others.

Stevan Harnad is editor of books including Sleep and Dreaming: Scientific Advances and Reconsiderations (buy this book from Amazon (UK) or Amazon (USA)). See his website.