> The universe exists apart from being evoked by the human imagination, while mathematical objects do not exist before and apart from being evoked by human imagination.
Smolin says this in his conclusion. But when people talk about the unreasonable effectiveness of mathematics (at least when I've heard it), this is what they're talking about - not that mathematical objects exist in some nonphysical platonic space, but that they exist in our heads as a game we play - a formal axiomatic system. The question is, why does our formal game, which we think is mostly abstract, suddenly and surprisingly turn out to work so well to model the physical universe? (We don't find that chess works as a model, for example.)
Smolin answers that, sort of. He says that since the basics of mathematics are in nature, it's reasonable that as math progresses, it will continue to correspond to nature. But it seems to me quite a stretch to say that, because the natural numbers correspond to the existence of countable things in nature, and natural objects take up space, therefore pseudo-Riemannian manifolds will correspond to general relativity. To say that is reasonable, it seems to me, requires making a mysticism around nature.