The project aims to provide a contribution to the epistemology of mathematics, in particular on the role of symbols. Based on a previous study on the emergence of symbolic algebra during the sixteenth century we compare Renaissance algebra with algebraic practice in the Arab world, China, Japan and India. Our working hypothesis is that the symbolism in European mathematics lead to the idea of a mathesis universalis, a universal symbolic language which functioned as a normative model for acquiring truth and certainty. Symbolism thus provided Western science a competitive methdological avantage on Eastern traditions.