Engineer and Economist
The phrase “French ingenieurs economistes” usually designates mathematicians such as Antoine-Augustin Cournot as well as engineers like Dupuit, but their approaches are often different.
The approach of the mathematicians is most of the time abstract: Cournot’s ambition was to develop a mathematical theory of wealth. Dupuit instead, like most engineers, started from concrete problems, for example, how to know whether a public investment - a bridge, a road - would be useful enough to justify its construction? Must it be financed by a toll and, if so, at which level should it be fixed?In order to find a solution to these problems, Dupuit used economic theory. He reasons in exactly the same way when dealing with the determining tolls or the distribution of water (Chatzis 2009: 668). He thought that theory alone can explain natural phenomena and the engineer or the economist must rely on it, and not only on experience, when trying to solve concrete problems.
French liberals discussed for a long time the nature and object of political economy. Henri Baudrillart (1863: 250) maintained that political economy is a moral science and differed from physics or mathematics because a role has to be given to opinions. Dupuit disagreed. Political economy, being the science of society, is not dependent as regards its principles on opinions. One of its most outstanding features “is that it presents a set of principles that can be demonstrated rigorously and form a dense doctrine, on which every intelligent man is not susceptible to have an opinion” (Dupuit 1863 [2009]: 153).
Dupuit did not explicitly use mathematics in his economic articles, but he always stated that political economy would benefit from such a use. He compared the role of mathematics with that of machines: they allow a smaller amount of expenses, time and labour. They favour the exactness of the result while our mind always runs the risk of being lost in a complex chain of arguments. He takes Jean-Baptiste Say (1803 [2006]: 519) as an example, who was mistaken when he advanced that doubling the quantity of wine produced would half its price. Mathematics would have allowed him to state rigorously the relationship between the abundance of production and the price, and would have avoided his error and, still better, would have shown that, while the price is a decreasing function of the quantity produced, it also depends on many other data.