Talk:Principle of induction

I deleted the following phrase from the end. It cast too bleak a shadow on the possibilty of human knowledge: Claims to know are therefore inherently unscientific.

Fundamental problems

I believe there are some fundamental problems with this article.

"Without it, experiments could not confirm a hypothesis" - we can never confirm a hypothesis, we can only 'not falsify it'. To go with Popper's classic example, I may go theorise that all swans are white. I may then go down to a lake and observe a number of swans that are indeed all white. But this does not confirm the general theory. We can logically say that all the swans I have observed are white, but it does not follow logically that ALL swans are therefore white. Science works upon falsification, not verification.

Prediction in science does not follow the inductive method, but rather the deductive method. Take for example the various models which climatologists produce to try and predict changes in the climate. These models are built using theories as to how the atmosphere, hydrosphere etc. function and interact. From these general models specific predictions are thus deduced, which must be true if the general theories behind it are true. In this case we are going from a general to a specific statement, which is deduction, not induction.

The argument about the probability of future event occurring increasing does not hold true either. This is not logical, it is simply a matter of psychology that we believe this to be the case.

Induction is used in formulating theories (though Popper argues otherwise), but it is not used in the testing stage. DWiggins 21:30, 29 October 2009 (EDT)