Sir Karl Popper was one of the 20th century's most influential thinkers. His contribution to the philosophy of science was to turn upside down ideas of how a theory might become established. Instead of coming up with a hypothesis, whether by some inductive scheme or by some creative insight, and then seeking to find confirmatory evidence, Popper maintained that having come up with a hypothesis, one should seek evidence which refutes it. The best hypotheses were bold, and at least superficially, easy to refute because they made many testable claims. A bold hypothesis which survived these tests was well corroborated. Popper's scheme seemed both simple and straight forward. The crucial test seemed to be logically conclusive. It lead to a clear demarcation between science and non-science. Yet it was not without its difficulties.

There are statements which on the surface look like the kind of thing science allows us to say. But while they are empirically provable the are not falsifiable. Does this make them non-scientific in the same way that claiming Peter Pan can fly is non-scientific? Such a statement would be "bacteria exist". This cannot be disproved by observation. But does this mean it is not a true statement? Another problem concerns statements or theories which can be falsified. But how conclusive is falsification? A theory has to be combined with observations to put it into action, to test it. Take the example theory "all swans are white". Suppose you test this theory by going to the local pond and observing large birds that look like swans and there you see one that is black. You might reason as follows:

Theory: All swans are white
Observation: there is a black swan in the middle of the pond
Conclusion: ?

In actual fact you have a number of options at this point:
The theory is falsified - a new theory is needed
The swan in the middle of the pond is white, but it's covered in oil so it appears to be black
The large bird in the middle of the pond is not a swan

Because theories, to be tested, have to be combined with observations, when the observations conflict with the theory, very often the observations are discarded rather than the theory!

Use back key or return to index