Wolpert's French Flag: what's the problem?

Two phrases attributed to Lewis Wolpert – ‘positional information’ and ‘The French Flag Model’ – have become so intertwined that they are now used almost interchangeably. Here, I argue that this represents an unfortunate oversimplification of Wolpert's ideas that arose gradually in the developmental biology community, some significant time after his key papers were published. In contrast to common belief, Wolpert did not use the phrase French Flag ‘Model’ but instead introduced the French Flag ‘Problem’. This famous metaphor was not a proposal of how patterning works, but rather an abstraction of the question to be addressed. More specifically, the French flag metaphor was an attempt to de-couple the problem from the multiple possible models that could solve it. In this spirit, Wolpert's first article on this topic also proposed (in addition to the well-known gradient model) an alternative solution to the French Flag Problem that was self-organising and had no gradients, and in which each cell ‘cannot compute where it is in the system’, i.e. there is no positional information. I discuss the history and evolution of these terms, and how they influence the way we study patterning.