Topics Science

Question If scientific theories claim to provide ultimate truths about the nature of reality, then how can we explain new theories and revisions of existing theories? To take an extreme example - if we once believed that the world was flat, how can we justify our current belief in a spherical world? In this particular case, our current ideas can clearly be explained by the increases in technology and no one would seriously question the shape of our planet. However, in more complicated instances how can we be so sure that science offers anything more than a set of beliefs (i.e. in a similar way to religion or myths).

Accepted:
November 3, 2005

Comments

Later scientific theories often contradict the theories they replace. Does this mean that science is not in the truth business? There is a simple and much discussed argument for this conclusion. The argument is known as the 'pessimistic induction': all scientific theories more than say two hundred years old are now known to be false, so it is likely that all present and future theories will eventually be found to be false as well. So much for truth.

There are a number of responses those of us who do think science is in the truth business can give. We can quibble with the premise of the pessimistic induction: surely not all theories more that two hundred years old are now known to be false. For example, we still believe the theory that the blood circulates around the body, and that theory is much more than two hundred years old. But even if the premise is an exaggeration, lots of our best old theories are contradicted by what we now believe. For example, if Einstein is right, then Newton was wrong. Sure, Newton makes pretty good predictions when the velocities are low, but even then the theory is, strictly speaking, incorrect. (And probably not just strictly speaking: the very concept of mass is different for Einstein than it is for Newton.)

Another reply to the pessimistic induction is to stress that saying science is the truth business does not mean you have to say that current scientific theories are the truth, the whole truth, and nothing but the truth. It is enough if our theories are approximately true, and that the approximation is improving. So someone who thinks science is in the truth business can just admit that probably almost no theory was or will be completely true.

Here is one more possible response to the pessimistic induction. Does the fact that we have made mistakes in the past really make it more likely that we will make mistakes in the future? We have better instruments and more data now than we used to have. And anyway, we learn from our mistakes. So maybe the fact that we have constructed theories and then figured out where they go wrong actually makes it not less likely but more likely that future theories will be true!