Dissemination of Technical Knowledge in the Middle Ages and the Early Modern Era: New Approaches and Methodological Issues

The diffusion of technology as an object of historical study originated partly in the years after World War II as a critique of the West's self-confidence embodied in developed countries' solutions to poverty. Such solutions, usually defined in terms of resolving technical "backwardness," promulgated a simplistic and imperialistic notion of transfer that suggested quick action, linearity in time and space, and unilateral political decisions. Two prominent scholars, Paul Bairoch and Nathan Rosenberg, argued against this portrayal of technical diffusion.¹ In their view, history showed that transfers of technology were uncertain and complex processes. Hence the interest of historians in technical dissemination derived from a profound skepticism concerning the diffusionist notion of transfer developed in the context of expanding Western economic power.

Did this critique imply that the circulation of knowledge could not occur, that it was a kind of illusion? Although Rosenberg stressed the high rate of failure, even between countries of similar technical cultures, he did not claim that the circulation of knowledge could not occur. Rather, he pointed to methodological issues and intellectual tools for historians that would help them better understand the complexity of the process. He denied the heuristic values of leadership and backwardness, asserting that "economic growth has never been a process of mere replication." Instead, it depended entirely on the "special environments of individual countries"; diversity was therefore the cornerstone of his analysis of development paths. Rosenberg also stressed learning and the significance of mobile, skilled personnel for transmitting "noncodified knowledge." Since his seminal work first appeared, numerous case studies have lengthened this list of questions. In the spirit of Rosenberg's investigations, this article examines recent methodological innovations in order to understand the specificity of technical dissemination in the long term.

Our principal argument is that although technology transfers might have taken place across long distances, the macroeconomic scale is nevertheless inappropriate for their study. Over many centuries, skills become embedded within specific communities because of their needs and constraints, their habits and symbols, and their territories. As Peter Mathias put it, techniques "were only the visible tip of a submerged mass of relationships."² Therefore the transmission of techniques relied upon a whole set of resources, material and immaterial.³ Fundamentally, "techniques" are answers to specific needs and expectations. Their applications are not universal; they belong to a world of diversity, contingency, and heterogeneity. This has several implications.

First, intermediaries and host communities were not neutral or passive; instead, they always adapted and translated the techniques they conveyed or received. The creation of hybrids—"creative imitation"—was intrinsic to dissemination, as each locality followed its own path. This challenges any notion of a universal pattern of growth.⁴ Techniques that originated from different places and belonged to distinct generations frequently overlapped. Major questions here concern the identification of techniques—including the precise skills, materials, and processes involved—in reference to the diverse communities using them. Is it possible to follow one technique across multiple territories? What resources are available to the historian to identify techniques that were so rarely recorded and codified?⁵ What part did products play as conduits for technical knowledge? Are their appellations related to geographical origins or the names of their makers, or do identifying marks on the products themselves provide clues? The same sorts of questions can be applied to the routes. If failures were common, diversions were also frequent: madder dyeing, for example, which originated in India, had to be reimported several times through different routes between the sixteenth and eighteenth centuries before it was actually learned and adopted in the West. Mythical origin narratives often conceal the absence of linearity by attributing a certain technique to one place or one people—or a certain technology transfer to one group or even...