Initially it seems surprising that Edgar Zilsel's work has found as little response among philosophers as it has. After all, his contributions to the Vienna Circle's debates about probability and protocol statements were published in Erkenntnis. Already his doctoral dissertation dealt with a central problem of modern philosophy of science—the status of statistical laws in physics—and revealed a remarkably knowledgeable mathematician, physicist and philosopher. Yet the way in which Zilsel raised the issues, namely via Leibniz, Spinoza and Kant, was not easy to accept for many of the later logical empiricists. Zilsel stuck with what in his dissertation he had called the "problem of application" and held that it needed to be solved even once the framework of logical empiricism had been accepted. By contrast, Richard von Mises and Otto Neurath considered it a pseudo-problem. Zilsel's views are difficult to categorise and nowadays even difficult to understand. Just as Mises and Neurath were puzzled by the problem of application, so contemporary readers are likely to be puzzled by Zilsel's search for "historical laws". What were they supposed to be and why did Zilsel think it so important to discover them?