The objective of this article is to identify the argument that must have motivated rational, free and equal representatives of peoples to agree upon the European Union, supposing the EU did not exist and the representatives do not know which people they represent. The problem is that the variant of social contract theory that is thought of as the most sophisticated today—John Rawls’ “political liberalism”—is not readily applicable to the European Union. This is so because Rawls’ views about regional unions and federations conflict with his definition of a people. This conflict blocks a straightforward application of his political liberal approach to the European Union. The social contract argument for the EU can only be reconstructed if this problem is dealt with first. The article has three paragraphs. In the first I analyze in more detail the conflict. In the second paragraph I argue that the best way to respond to the problem is not to start with Rawls’ notion of a people, but—given that the EU is a legitimate political order—with the EU as the outcome of a social contract between peoples. The social contract argument is put in reverse, as it were and the question becomes: what argument must have motivated rational, free and equal representatives of what kind of peoples to agree upon the European Union? In the final paragraph I argue why the notion of a people which in this way may be said to underlie the EU need not be incompatible with political liberalism.