As long as dimension n = 3 and so [dimension of R^n] n = n(n-1)/2 [dimension of so(n)]. Not to disregard these low-dimensional isomorphisms; there’s quite a few of them (they run out at about n = 10 IIRC), each one different from the next, and most are important in one way or another. But expressing momenta and angular momenta in those terms gives the impression that their existence is due to the (numerous and sometimes very complex) particulars of three-dimensional space, which it isn’t.