Structures and structuralism in contemporary philosophy of mathematics (Q1840995)

The authors study various forms of mathematical structuralism and mention a few of the problems associated with them, without taking a critical stance on them. The paper refers to the undisputed importance of what the authors call a structuralist methodology, exemplified by the emphasis on classes of structures like groups, rings, topological spaces, etc. However, they take it too much for granted that this extends to other parts of mathematics. In particular, contrary to what the authors seem to imply, the system of natural numbers and the system of real numbers, both capable of being characterized by categorical second-order axioms, are not regarded in the same way by the overwhelming majority of mathematicians who do number theory, real analysis, complex analysis, ordinary and partial differential equations, probability and statistics, etc. Structuralism has been extended to the natural numbers and real numbers only in recent times by philosophers of mathematics, stimulated by \textit{P. Benacerraf}'s 1965 paper, ``What numbers could not be'' [Philos. Review 74, 47-73 (1965); reprinted in: P. Banacerraf et al. (eds.), Philosophy of mathematics. Selected readings. 2nd ed. (Cambridge Univ. Press), 272-294 (1983; Zbl 0548.03002)], which pointed to the existence of multiple (isomorphic) models of the axioms for the natural numbers and the real numbers and gave rise to skepticism about our knowledge concerning those numbers. The fact that, for at least the two hundred years before Dedekind, Cantor, and Zermelo, mathematicians treated the natural numbers and real numbers as unique well-defined objects often seems to be ignored (except possibly by a few adherents of pattern structuralism, a variant treated by the authors on pp. 363-371). The authors mention in several places that set theory, regarded by the authors and most mathematicians as the foundation of modern mathematics, is exempt from structuralism, that is, it is not assumed that there are isomorphic, equally genuine, universes of sets. This entails that structuralism is of limited significance for the philosophy of mathematics. As far as the authors' aim of describing the main variants of structuralism is concerned, that has been accomplished with great breadth and clarity.