The example he gives about geometry is actually quite interesting. It is one of the early highlights of a deep question: is this knowledge, geometry in this case, learned/learnable or is it, somehow, innate? Do we learn this from scratch or do we have innate pre-existing cognitive structures that are “configured” by experience? If the latter, what does “learning” mean? It’s definitely not what we usually mean. If the former, we meet Hume and Kant and have to show how we arrived at space and geometry ex nihilo.

If learning is essentially based on “configuring” innate structures, you can IMO state it is eternal or uncovered or whatever poetic vehicle you desire. I’d say give these pre-modern guys a break.

These are issues being discussed way into the modern era starting (again) with the likes of Hume and Kant and no easy solutions are available. This is not a solved problem.