Note that to put it this way is precisely not to argue that Cubist space was "based on" Riemann's; that would amount to suggesting, in a perfectly un-Einsteinian fashion, that Cubism accepted Riemannian space as its apriori and then went on to represent it illusionistically.