I agree with Alex about the way mathematics is independent of science. Einstein proposed that space is curved and hence non-Euclidian, but this didn't undermine Euclidian geometry, because that geometry is about an abstract space defined by the axioms of the system, not about physical space. So Euclidian geometry turns out not to apply to physical space, but it has not been refuted by physics. There is however another way in which science and mathematics are not independent. Mathematicians may choose which problems to work on with an eye to what kind of mathematics might be particularly useful in science, and even more frequently scientists choose which problems to tackle by reference to the mathematical tools that are available to them.

Theories are descriptions, and they come in two flavors: true and false. So the Theory of Evolution can be both a theory and true, which is just what a great number of scientists believe. When evolution by natural selection is called a theory, however, this is sometimes intended to emphasise that there is no proof that it is true. Now if by 'proof' we mean what pure mathematicians produce, then this is correct. There is no proof of the Theory of Evolution, and there is no proof of any other empirical theory either. Proof in this sense is not an option in science, because all theories go beyond the evidence upon which they are based. There can similarly be no proof that the sun will rise tomorrow. But the sense in which it is true that there is no proof of evolution is compatible with the claim that there is overwhelming evidence that it is true, which is again what a great number of scientists believe.

People who believe that science is a social construct do not deny that there is a world independent of ourselves, with real causes and real effects. What they emphasise is something consistent with this, namely that the theories that scientists invent to account for the world are strongly influenced by social context.