On medieval manuscript production in the Latin West, explorations with a global database

Abstract

This study is the first to present detailed information on the production and uses of manuscripts in eleven geographical areas, together comprising the Latin West. We estimated the absolute production (in numbers of manuscripts copied per century) in Europe between 500 and 1500. In total some eleven million books were copied in this medieval period, of which some seven million were still extant in 1500. There are signs that book production in the Middle Ages can be seen an indicator of the average local output per century. In the Latin West production of medieval manuscripts increased on average with nearly 1% per year. Over a millennium this then led to a more than 300- to 400-fold growth in output. Variables possibly explaining the size of the medieval book production such as numbers of monasteries, local Church organisation, urbanisation rates and universities are presented here for the fist time too; and on a similar geographical and time scale, in order to enable a detailed quantitative analysis. Our large data set allowed us to assess which of the various explanatory variables contributed most to European book production over time. With a specific statistical regression model we could show that in the early medieval period the production of books in the Latin West was mainly monastic. In the second half of the Middle Ages urban civilisation, for which universities were an important indicator, drove the local output of manuscripts. We separately quantified the losses of manuscripts in percentages per century from hundreds of medieval libraries and their current survival. For this we heavily leaned on data from the United Kingdom, supplemented with data from the Continent. The average loss rates of medieval manuscripts we calculated were some –25% to –40% per century. With these loss rates we then calculated the fraction that survived. With that fraction and the distributions of still existing medieval manuscripts from the database and other sources we finally estimated how many books had been copied in the Middle Ages. Though our whole approach is wrought with uncertainties, the considerable noise in our data is substantially less than the overall growth in medieval output, leading to a more or less robust pattern of production rates. We could not find reliable loss rates for books copied in any of the other regional clusters, and therefore did not estimate the book production outside the Latin West.