Rosental’s starting point is that he wants to challenge a traditional picture put forward by French historians and social scientists, that France was a country where there was no, or at least, a very low mobility. People grew up in a village, and stayed there for the rest of their lives. According to Rosental this tradition has been strong and long-lived. Of course, it would be very interesting to know why this view has survived for such a long time: does it suit a certain perspective on French history, or what is it that explains this view? For me as a Swedish historian it is easy to identify with this picture, because it also existed in Sweden for a long time until it was challenged during the 1950s and 1960s, by among others, the geographer Torsten Hägerstrand. Historians and geographers were able to show that Swedish society was characterised by high mobility as early as the xvii^th century. This indicates that French migration research did not take into account the research outside France.

Rosental challenges the traditional French view of a society with hardly any mobility at all by using a micro-demographic method. He follows individuals and families over generations using marriage registers, and discovers a general pattern, if I understand him correctly, that, at the beginning of the xix^th century, people moved in rather limited circles and areas, between villages and small towns which were quite near each other. However, over time the mobility increased over wider distances, and at that time social mobi-lity increased as well. Traditional patterns of social mobility within traditional rural occupations, such as farmers and artisans, broke up and with an increasing distance new occupations were taken up, for example occupations that were the creation of a modern so-ciety, such as white-collar employment in the public sector.

Rosental’s systematic approach and method has many advantages. It is easy to follow these patterns when individual and family cases are presented. The details become clear when Rosental uses his micro-history model, because macro analysis often highlights differences on the individual and local level. His way of working also makes it easier to understand the process of mobility. By using individual and family data it is possible to see why these individuals and families made their journey through the French countryside into urban areas. At the same time Rosental’s method raises a major question. How much can we generalise from these examples? My question is based on the fact that his sub-sample of 97 families were selected from a larger sample of 3000 families chosen initially to represent a miniature of France around the year 1800. This 3000 families database, organised by Jacques Dupâquier and Kessler, and used by Rosental, took into consideration many different aspects which were relevant in 1800, but is it representative for the period 100 years later? Is the third generation of the database also a fair sample of France?

Another question that came into my mind when I read Rosental’s book had to do with my own research on the petty bourgeoisie. Rosental’s results demonstrate very well the development of handicrafts in the xix^th century. When he starts his study a common tradition within the handicrafts was that a son would follow in his father’s footsteps, or at least pick up an occupation within the handicrafts. This type of behaviour did not promote mobility over long distances, because it was not necessary to move far to get a job or to set up a small business. When industry started to challenge traditional handicrafts many people had to look for jobs elsewhere, and they had to move. The local community did not have room for them. Later on, at the end of the xix^th century, economic competition was so fierce within the handicrafts sector that the artisan family did not want its sons to take over the business. Instead they encouraged their sons, and also their daughters, to take jobs outside the small world of handicrafts, especially white-collar employment. It was regarded as safe, especially employment in the public sector, for example in the Post Office or the railways.

There is also a problem with Rosental’s classification of the different communities he studies. He has used population size as his measure. It might well be the best way to do it, but I think he should have tried to classify his communities according to the dominant trade instead. Such a classification might perhaps make it possible to say something more about the mobility pattern. It is very important to know if a person moves to an industrial town, dominated by heavy industry, but with the same population size as another town dominated by le petit commerce or to a town functioning as a local admi-nistrative centre. That type of classification might strengthen his analysis. I think that if we look at traditional pull and push factors in the migration process, such a distinction could lead to other conclusions about why people moved to one place instead of another.