In 2000, the Post-Autistic Economics Newsletter began covering the resistance of French students against the “uncontrolled use of mathematics” in their discipline. The students spoke of a “need to liberate economics from its autistic obsession with formal models that have no obvious empirical reference.” (Post-Autistic Economics Newsletter, issue 3, 27 Nov 2000) In an earlier newsletter they had elaborated, too often the lectures leave no place for reflection. Out of all the approaches to economic questions that exist, generally only one is presented to us. This approach is supposed to explain everything by means of a purely axiomatic process, as if this were THE economic truth. We do not accept this dogmatism. We want a pluralism of approaches, adapted to the complexity of the objects and to the uncertainty surrounding most of the big questions in economics (unemployment, inequalities, the place of financial markets, the advantages and disadvantages of free-trade, globalization, economic development, etc.) (Post-Autistic Economics Newsletter, issue 2, 3 Oct 2000). The students did not object to economics being a science; they just wanted to make it empirically relevant. This search for an empirically relevant science of economics has been a hard slog, for if the effort to show the autistic nature of orthodox nomothetic neoclassical economics has been relatively easy, the search for a praxis relevant alternative has not. I suggest that in looking for empirical relevancy primarily through expansion into the social sciences, post-autistic economists have been looking in the wrong place. For empirical relevance they need to focus on the relationship between economics and the shop floor.