According to David Armitage and Jo Guldi, digitized sources and quantification almost naturally lead to the sort of longue durée history that they seek to promote. This article questions that assertion on the basis of the long tradition of quantitative history, open to exchanges with the social sciences and revived, not annihilated, by microhistory. The digitization of numerous historical sources does not call for less caution in our analyses—quite the contrary, as it creates new biases. More importantly, it does not solve the crucial question of controlled anachronism, that is, the need for carefully constructed categories in any quantification based on the longue durée. The article also addresses the implications of choosing the longue durée as the exclusive basis for reflections on historical processes and causality. Is the longue durée purely a scale for description? If not, can it escape a simplistic vision, a monocausal path dependency? If we are to avoid such pitfalls, the wider debates within all the social sciences on time-scales and causality must be taken into account.