The question of the difference between ancient slavery and medieval
serfdom may seem to be a rather minor point to be discussing, but it is
one that has absorbed the attention of generations of Marxist-influenced
historians. Karl Marx held that society evolved in accordance with
economic law. His colleague, Friederich Engels, proposed that that
evolution proceeds through four major stages, each characterized by its
principle mode of production. The first stage is ancient slavery, followed
by medieval serfdom, modern wage exploitation, and future communism. Lenin
elaborated the scheme a bit, dividing the era of communism itself into
four stages: the revolt of the working class and their seizure of the
means of production, the dictatorship of the working class and their
abolition of class, the withering away of a state that no longer has a
repressive function to perform, and final communism.|
The change from classical slavery to medieval serfdom is, therefore and in the precise sense of the term, an epoch-making event. It was, supposedly, the last thing Marc Bloch talked about before being machined gunned by the Gestapo in Lyon in 1944. As an aside, I note that the man who ordered the killing was able to evade capture for many years, assisted in this by the government of the United States. Be that as it may, the subject is well worth talking about.
The distinction between the ancient slave and the medieval serf in law and custom may seem a fine one, but was significant. The man (and of course there were women slaves) who was enslaved in ancient times was considered to have died; all that was his passed to his master, including the power of life and death. The slave who resisted his master for any reason could be killed, or killed for no reason at all if the master wished to do so.
The serf, by contrast, was a free man except for the obligations he owed to his lord and the rights his lord claimed over him. Both servile obligations and noble rights could be very extensive, but since the serf was a living creature with a soul, they could not be unlimited. The master could not deny his serf the amenities of the Church, work him on holy days, or demand actions of him that were immoral. As a living creature, the serf had the rights accorded him by natural law. He could resist a lord attempting to take his life or one attempting to withhold the necessities of life from him and his.
The distinction was just as significant in practice. The ancient laboring slaves who formed the vast majority of the slave class, even if the literature of the times deals far more with the servant class, were segregated by class and lived in prison-like barracks on the villa. They were under the tight control of a slave-driver who punished any sign of rebelliousness quickly and harshly. They were worked in gangs and possessed nothing to call their own.
Even though the word "serf" comes from the Latin "servus," and means "slave," the situation of medieval serfs was quite different from that of the slave of Classical times. There were two kinds of serf: those who were bound to the soil and those who were bound to the lord. Servants were drawn from the latter class, but the insecurity of their tenure probably made the condition of being bound to the soil preferable. The serf usually had a separate hut with an attached garden and lived with his family. His marriage was a holy union, and married couples were not supposed to be separated. The serf had duties assigned to him by the steward of the manor and was responsible for the tilling of demesne land and the provisioning of the manor house. He received, in return, food and clothing for himself and his family, and often had time to supplement his rations by gardening and, especially during the enforced idleness of the winter, could produce things which he was often allowed to keep for himself or sell. Although the life of the medieval serf was very hard, it was probably preferable to that of the ancient slave.
The transition from slavery to serfdom was apparently slow and fitful, but the critical point was when a slave owner allowed his slaves to live in family groups in separate houses, a stage in which they were called "houseled slaves." Most historians regard this as a step toward greater freedom and they may be right. It is also significant that it represents a diminished fear of slave revolts on the part of the slave-owner. Marxist historians, who have tried to find widespread slave revolts in such phenomena as the bagaudae and have claimed that ancient slavery ended with revolts just as did medieval serfdom, have ignored this basic factor. The truth is that we do not know why the slave- owners should have been less fearful or why they should have allowed greater freedom, although there are many theories about the latter.
Finlay and others say blithely that the slave-owners found that people work harder when they are free and are laboring for their own good and that of their families, but we don't really know that that is true. The slave-masters of the Roman Empire were heirs to a tradition of labor management at least a couple of thousand years old. One would think that, if free labor were actually more productive than that of slave gangs, that fact would have been discovered much earlier. Roman slave-masters were not ignorant of labor economics. The problem is that we do not have data on how productive truly exploited labor can be since the treatment of slaves in every slave system for which there is relatively precise data was mitigated by some degree of humanitarian sentiment.
We could find out, of course, but this brings up a complex ethical problem. The Nazi SS operated labor camps during the Second World War in which the laborers were simply worked to death. The SS kept meticulous records of how much labor could be expected for how long and for how much investment in food and care. They even experimented, apparently, with how much production the production of individual laborers could be increased by beating one of their number to death on the job. Those records exist and could provide an answer to the question of whether free labor is more productive than slave labor. But can historians morally use data compiled in such a way? Physicians and psychologists face much the same dilemma. The Nazis practiced atrocious experiments on humans in their death camps and the Japanese tested the effectiveness of various types of poison gas and deadly biological agents by infecting entire villages of Chinese (and some American prisoners of war), and the records of these atrocities are both available. They are also unique, because civilized people cannot conduct experiments in which killing people is an essential part of the process. But if historians or physicians used such data, would we not become partners of those who compiled the records?
Let us simply say that we do not really know that free workers are more efficient than slaves and that we do not know why the slave-masters of the late empire should have become less fearful of slave rebellions. If we do not know these things, we do not know what prompted the transformation of the ancient slave system into the medieval system of serfdom. It is an important issue, however, and deserves more rigorous and objective study than it has received up to this point.
Lynn Harry Nelson
Emeritus Professor of
The University of Kansas