To Count the Absent (or the Problem of the Total Number of Genoese Merchants in Byzantium)

Abstract

The author develops a mathematical-statistical approach to the problem of estimating the size of Genoese medieval population in Byzantium. The data source is notarial acts covering commercial partnerships, freightage, wills, purchase and sale of houses, goods and people drawn up in the Genoese colony of Constantinople at the end of the 13th century. The will form has a fairly uniform structure. In addition to the mandatory record of names of the contracting parties and witnesses of the transaction, it may also register names of the third parties. Thus, these data on the clientele of Genoese notaries represent a dataset which may indirectly indicate the size of the entire trading Genoese community of Byzantium. This approach is based on a constructed formalized model that describes the behavior of merchants when visiting and concluding a transaction attested by a notary. This makes it possible to pass in a natural way from the initial to the statistical problem of estimating the size of a finite aggregate and use this mathematical theory for its calculation. In this case, the author applies the approach associated with the use of the maximum likelihood function that is a novelty. The resulting formula allows (with a certain degree of probability) one to estimate the required size of the Genoese population. It is interesting that this estimate, on the whole, coincides with the result of A.L. Ponomarev obtained earlier for the same problem using Zipf's empirical law.