In 1931 Kurt Gödel published his incompleteness results, and some years later Church and Turing showed that the decision problem for certain systems of symbolic logic has a negative solution. However, already in 1921 the young logician Emil Post worked on similar problems which resulted in what he called an "anticipation" of these results. For several reasons though, he did not submit these results to a journal until 1941. An interest in the early work of Emil Post however should not only be motivated by these historical facts. It will be shown how Post’s specific approach and methods led him to his systems of tag. These forced him into the mathematician’s practice of testing specific cases and examples in order to be able to understand the properties of the systems. Starting from his Ph.D. dissertation and his work as a Proctor fellow at Princeton (1920–1921) it will be shown how this research on tag systems led Post from the idea that the whole of mathematics is decidable and complete to the opposite view.