Can an objective theory of welfare be developed at both the individual and collective levels? The question is far from straightforward, since moving from subjective feelings to a quantified representation and then aggregating them requires a series of assumptions whose arbitrariness must be examined. In a comedy sketch written for Pierre Palmade, actor Gérard Darmon uses humour to convey just how tricky it is to quantify satisfaction when he asks: "Would you rather be as hairy as a gorilla or have eyes on the sides of your head like a fish?" Comparing and ranking two situations is therefore anything but simple when welfare is at stake. And moving from the individual to the collective level is more problematic still.
We now know that objective individual preferences cannot be aggregated under reasonable conditions (see Théorie des jeux, Bibliothèque Tangente 46, 2013). This is the subject of the famous Condorcet paradox, given mathematical form by Arrow's impossibility theorem, which Kenneth Joseph Arrow (1921–2017) proved in his 1951 dissertation.
So to what extent can we seek to optimize welfare, whether individual or collective? The problem is all the more complex because any attempt at modelling will inevitably carry a political ideology.
A debatable principle
----------------------
At the turn of the 19th century, the ideas put forward by the Italian Vilfredo Pareto appeared original. To understand them, we need to set them alongside those developed by the French economist Léon Walras (1834–1910). Walras based his analysis on an economy in equilibrium, with perfect competition and complete freedom for economic agents, which, in his view, would guarantee maximum welfare at both the individual and collective levels. For Pareto, this model was "circular": since the individualistic principle of private property was one of its premises, the model could only conclude that this very principle produced the greatest collective satisfaction.
