Data on wages are essential to an analysis of wartime material conditions. But can these statistics be used without falling into the trap described by Sen as the construction of empty taxonomies conventionally known as ‘living standards’? To be sure, wages and incomes form part of the notion of ‘capabilities’ and ‘functionings’ adumbrated by Sen. But his work on the Bengal famine of 1943 added an additional and essential dimension to the problem of ‘capabilities’: the dimension of inflation, which in Berlin in 1916-18 as in Bengal in 1943, soared out of control. Under conditions of spiralling inflation, the very notion of ‘living standards’ is problematic. But even when less severe price movements occurred, the simple correlation between wages and prices is remote from Sen's concept of ‘capabilities’, since real wages are not equivalents of the degree to which urban populations responded individually or collectively to rapid economic fluctuations.

This is especially the case in wartime. How do we quantify uncertainty, introduced violently with a price spiral that appeared to be able to spin out of sight? What point of reference could contemporaries have used? The inflation of the Napoleonic wars was robust, though over decades, not months, and besides, that experience had passed into history long before 1914. In the Berlin case, we have the further difficulty of the black market, the ubiquity of which casts serious doubts on the utility of official price series.