I don't quite agree. Complex analysis is more than powerful enough to recover the unit complex number corresponding to a specific angle without referencing any

I don't quite agree. Complex analysis is more than powerful enough to recover the unit complex number corresponding to a specific angle without referencing any kind of trigonometry. Of course you'll have reconstructed the sine and cosine functions that way, but you don't need to use a single trigonometric function or theorem to get there.