The solution to the Torddjen example is that accounting is a social construct; the underlying mathematics is not. Similarly, the fact that the 1:1.618 (or 0.618:1) ratio is the basis of Greek architecture such as the Parthenon is social, the Fibonacci sequence is not, the mathematical rules which Fibonacci used are not.

Now that I think about, maybe Fibonacci was on the borderline. The application of the Golden Mean when depicting perspective in art is likely subjective to humans, but a CS professor once told me that perspective is learned; someone who grew up in a dense jungle where the visibility was limited to a few metres would have no sense of visual perspective. To what extent is our art and culture derived from biological traits common to all humans? Or to what extent is art about triggering pre-existing, default patterns in the brain, and to what extent is it about appealing to patterns developed in the mind through living within a society?

In summary, math itself is not a social construct, but applications of it may be.