Presidential address delivered before the American Historical Association at Chicago on December 28, 1944. Published in American Historical Review 50, no. 2 (January 1945): 213-27.

Between Slavery and Freedom

On April 18, 1864, at Baltimore, Abraham Lincoln spoke of the vagueness in the customary use of the word liberty.

The world has never had a good definition of liberty, and the American people, just now, are much in need of one. We all declare for liberty; but in using the same word we do not all mean the same thing. With some the word liberty may mean for each man to do as he pleases with himself, and the product of his labor; while with others the same word may mean for some men to do as they please with other men, and the product of other men’s labor. Here are two, not only different, but incompatible things, called by the same name, liberty.

This observation of President Lincoln still holds good. There are few words more vague in their cy to move out of the nome in which they worked. If they should cross over from one nome to another they were subject to arrest by the concessionaire who held the contract to produce the oil in that shop or by the higher nome officials. These employees in the oil monopoly were free men, working under a contract with the government; but their mobility, both occupational and spatial, was restricted to the nome in which each one of them lived.

In the case of still another free group, and in this case one numerically much larger, the Ptolemies again set up legal limitations upon the right of movement of their subjects. These were the royal peasants who worked, under contract, the agricultural lands of the Ptolemaic god-kings. In their contracts the peasants swore upon oath that they would remain each year upon the farm lands which they worked from the time of the plowing and sowing through the time of delivery in kind of their rent. Out of a document dated 107 B.C. the following translation of the peasants’ oath is taken: “I will be visible daily for you and for the agents of the Queen, remaining in the places [designated] for those engaged in working the soil, without recourse to temple refuge or to any secular protector.” The obligation to work, that is, the binding of the peasant to his occupation as state farmer, is here a consequence of, and inseparable from, his physical fixation at a given place.

In still other ways the situation in Greco-Roman Egypt reflects the fact that the limitation of the privilege of free movement was a vital principle in the process of adjustment of the relations between the Egyptian state and its subjects. In those Egyptian marriage contracts, for instance, which are specifically Greek in character, the wife often agrees not to absent herself from the home by day or by night without her husband’s consent, a provision which appears to be wholly economic in its motivation.

The recluses, called katochoi, in the precinct of the Serapis temple were persons who isolated themselves from the world by devoting to the god their right of free movement. Voluntarily they restricted themselves to residence in the sacred precinct until they should receive again from the god their option of going whither they pleased. Significantly, the word for their release by the god from their local fixation is the precise technical term which is used in the Delphic manumissions when a freedman under bondage services is liberated from the infringements, to which he had agreed, upon his economic activities and his privilege of free movement. In both cases the Greek word for the release is apolysis.

With every people among whom arise the problems of liberty and encroachments upon liberty, the free choices and the negations of these free choices which distinguished freedom from nonfreedom must necessarily differ. In Roman slavery a surprising alteration presents itself in the concept and the discussions of enslavement from those which the Greeks had developed. Among the Romans one fails to find any trace of emphasis upon freedom of movement as a part of the idea of liberty. This Roman failure to assert the importance of the concept of mobility is strikingly conveyed in the definition of liberty which emanates from the legalist Florentinus, writing somewhere in the period 140-220 A.D. “Liberty,” said Florentinus, “is the natural capacity of doing what each person pleases unless he is prohibited from so doing either by force or by law.” Here the Greek idea of unhampered mobility as one of the ingredients of freedom is completely lacking.

In his Elementa Philosophica de Cive Thomas Hobbes averred that slavery, in all of its forms, might be expressed entirely in terms of freedom of movement. If, then, the privilege of unobstructed movement seemed so important a factor in the substance of human liberty to the Delphic priests and to Thomas Hobbes, why does it fail to appear in Roman legislation and Roman discussions of slavery? The answer seems to lie in the fundamental position of the familia in Roman society. This structure of the great household, including the wife, all the children, the clients, and slaves, lay under the supreme authority of the pater familias. The Roman legalists came to define this power, as it was exercised over his wife and the blood members of the family, as patria potestas, over the slaves of the great household as dominica potestas, or owner’s authority. When a slave was manumitted, as freedman he moved into a position similar to that of a client in the family organization. His former dominus became his patronus. The pater familias had controlled the work services and the movements of his slaves. Also the clients of his household were indirectly affected in their freedom to move by the custom of salutation of the patronus at his place of residence at fixed intervals. Since the freedman now rose out of the level of his former servile domination into the range of the patronal domination of the clients it was an easy shift of the control over his right of movement from the old dominica potestas to the patronal authority which the head of the household maintained over his clientage.

At some time in the later Republic Roman legislation fixed upon a term called obsequium, reverent obedience, to characterize the correct attitude of the freedman toward his former owner. It also fixed in precise terms what the bondage services were to be which were required of all freedmen. These were called operae. As compared with the contractual agreement between the former slave-owner and his present freedman in the Greek paramoné manumission the Roman statutory fixation of these services was rigid. Both the reverent obedience and the labor services were to endure throughout the life expectancy of the freedman. It is this Roman rigidity which explains, since it created them, a number of fundamental differences between Roman slavery and manumission and their results from those developed among the Greeks. The Greek slave, for example, if owned by two masters, could be freed in the part owned by one of them and retained in enslavement in that part belonging to the other. Half slave, half free. Roman law refused to recognize partial freedom. Part slave, full slave, was an accepted dictum of the Roman law.

The Romans, whether this explanation be right or wrong, paid not the slightest attention to mobility. Slaves could not have it. Freedmen and clients simply did not have it. Despite this complete neglect of mobility in their theory, in the highly co-ordinated system of the empire the Romans in their ruling practice came, in the end, to the point of binding the great mass of their free subjects to their jobs, and thereby to the places of their domicile, through statutory restrictions upon that very feature which they failed to recognize. This fact has long since been presented in his brilliant studies of the Roman colonate, published in his younger manhood by a former president of our Association, Professor Rostovtzeff. The peasants of the empire were eventually nailed down at the place of their origo, their nativity. Then came, in its turn, the binding of the collegiae, the craftsmen groups. By long custom these skilled craftsmen had tended to associate, naturally enough, by local groups belonging to the same town or city. The first step was to bind them to the towns in which they worked, to the places of their legal residence. In an imperial constitution one reads: “To the members of the workingmen’s corporations it is not permitted to live beyond the boundaries of their city.” The right to move away was gone from the craft workers.

Thus the fiscal system of the Roman Empire bound its farmers and handicraftsmen to their work by restricting their mobility to the places of their domicile. Thereafter it fixed the well-to-do classes also to the locale of their responsibility, which was that of collecting the taxes in their districts and of paying these in to the state officials. The social structure thus created was not a slave system. It seems more aptly to fall within the scope of “involuntary servitude” as that term is used in the thirteenth amendment to our Constitution. Without question it widened enormously the expanse of the social and economic area between slavery and freedom and confined therein a much greater proportion of the subject population than ancient slavery had ever done.

From 154 A.D. we have a decree of the Roman prefect of Egypt regarding persons who had left their native districts during a peasant revolt in order to escape the burdens of compulsory government services. The prefect ordered such persons to return to their homes and not to wander about in parts of Egypt foreign to them, like hearthless and homeless men, leading lives of wretched brigandage. If they should not return, as directed, they were to be arrested and sent to him for punishment. This is a characteristic example of the way in which the empire, in its eastern sectors, put its clamps upon the mobility of its free, non-Roman subjects.

The Greek term for the coloni of the eastern part of the empire was enapographoi georgoi, meaning only “registered farmers.” Their registration, and the localization consequent upon it, was one which attached the peasant to the village of his registration. The old Greek term of the indentured freedman’s condition–paramoné–continues in use in the east regarding these farmers. We have a large body of correspondence between the great estate owners in Egypt, the Apion-Strategus family of the fifth and sixth centuries, with their managers. In these documents the managers sometimes report upon new tenant farmers and give guarantees for their permanence upon the estate. The outstanding clause is invariably the same. The manager engages that the peasant whom he has registered upon the estate will “remain without intermission and spend his time upon his holding with his family and his wife and herds and all his possessions.” Sometimes there is this addition: “And never will he leave the same nor go apart into another place.” Mention of the services due from him as workman of the estate is rare, always brief in statement, and obviously secondary to the elimination of his right of movement.

Considered also from the point of view of the later imperial constitutions the term commonly applied to these peasants, ascripti glebae, and even more its customary translation, “bound to the soil,” is none too exact. A decree of Valentinianus, Theodosius, and Arcadius, datable between 384 and 389 A.D., puts it thus: “A law passed by our ancestors detains the coloni by a kind of eternal right so that they are not permitted to depart from those places whose fruits nurture them nor to desert those places which they have once undertaken to cultivate.” Again, under Honorius and Theodosius an imperial rescript declares: “We have ordered them so to adhere to the soil that they ought not to be moved away from it for even a moment.” Another rescript says: “Granted that they seem, in status, to be free men, nevertheless they are thought to be slaves of the ground for which they have been born and they have not the capacity of departing whither they wish or of changing their places.” In all of these imperial decisions it is the loss of self-mobility of the coloni which receives emphasis in the legislation.

In Egypt of the sixth and seventh centuries, just preceding the Mohammedan conquest, Greek leases of the peasants dealing with the farms of the great landowners, in the majority of the preserved documents, are extended either for the life expectancy of the tenants, as upon the Apion-Strategus big estates, or for an undefined period terminable only at the will of the big estate owner. In such contracts the colonus had no voice about changing his tenantry. He moved out when thrown out.

The restrictions upon individual freedom of movement which seemed so important to the priests of Apollo reappear in the canons of the church respecting manumissions made by members of its priesthood. In 633 A.D. the Council of Toledo ordained that priests might liberate a part of the slaves that they had acquired, “in this wise, that as free persons they remain under patronage of the Church with their peculia and their descendants, carrying out the useful services enjoined upon them in full measure as far as they can do so.” Manre–“to remain”–is the verb found here, the Latin form of the Greek paramenein which appeared in the Delphic manumission formula and in the later Greek developments of the half-free peasants and handicrafters. The compulsory labors in the church canon are limited to the capacity of the freedman to carry them out, again reproducing an idea present in the Delphic formula of the second century B.C.

In symbolic form the acquiring of mobility by a new freedman appears in a Lombard manumission ceremony of the ninth century of our era. The slave who was to be freed was taken by his owner to a crossroad. Thence he was permitted to go upon whichever of the four roads he might choose. This is also found, in explicit expression, in the Lombard documents granting liberty to slaves: “Let him have license and power to walk from the crossroads and live where he may wish.” The legal terms of this statement again seem to have been taken over–through Byzantine law, presumably–from the ancient Greek formulas. In Bavarian law the slave, if a Roman, is to proceed through opened gates and depart in whatever direction he may desire to walk. In the Germanic law of Henry I the right is stated thus: “He who manumits his slave … is to establish for him roads that are free, assign open gates and place in his hands a lance and a sword or whatever are the arms of freemen.”

The restrictions upon movement applied to the villeins and serfs in the medieval manorial organizations of France and England have long since become textbook clichés. Rarely could the villein leave the plot which he cultivated, at least if he wished to take with him his livestock and other belongings. Throughout the Middle Ages, through the period of declining serfdom into modern times, and spatially from Russia to the British Isles, a similar general pattern reappears, though widely diverging in its details. Fixation of labor by contractual or statutory infringements upon its mobility seems to reappear as a constant of the methods of labor control. Under the Russian and Uigur system of the kabalas from the thirteenth to the seventeenth centuries, debtors bonded themselves or their dependents, in case of default in payment, to serve their creditors in lieu of interest. The engagement entered into is that the bonded person “lives in the family” of the creditors. Again, we have a status of nonfreedom of persons who were fixed in their lives between slavery and freedom and tied by restriction upon choice of habitation, as in the Greek paramoné of freedmen.

In early Germanic society, in Poland under the Wisluga system, in Hungary until the right of free movement was restored to the agricultural laborers in 1838, we find the same general method employed for the immobilization of labor, though with manifold differentiations. In all of the groups whose status lay between a fairly complete liberty of action and total enslavement two phenomena are recurrent. The first is that all of the groups had fixed rights which were fully accepted by social custom and legally acknowledged. The second lay in the fact that all of them were either restricted in, or completely deprived of, any right respecting choice of domicile. In all of these manifestations loss of the right to move freely is the common denominator of their partial enslavement.

It may easily happen that restrictions may be placed upon the privilege of voluntary mobility for reasons which lie quite aside from any motivation of labor control and that these may produce results of primary importance, though quite without design, in that precise domain. This occurred, for example, in England in the seventeenth century in the application of the parish poor laws. For fiscal reasons poverty-stricken laborers were not admitted by one parish from another, lest they add to the burden of the local parish relief. As a result of this the poor could not move from their home parishes. Restricted in his right of movement the ordinary laborer was dependent upon the limited opportunities of local employment in his own parish, and his liberties were really controlled by the officials of that parish.

As much as any other discrimination upon them, restrictions upon the right of movement of Negro freedmen in the United States tended to immobilize them in the area between liberty and slavery. In the New England states they remained under the necessity of showing a pass if they wished to go beyond the bounds of the town in which they lived or to appear upon the streets after nine o’clock at night. In some states of the South in the early nineteenth century the conditions imposed upon freedman movement were peculiar, so far as I know, to the American scene. In several states legislative compulsion was put upon the freed Negro requiring that he move. In North Carolina the manumitted Negro was forced, so far as the state laws were concerned, to emigrate from the area of the state’s jurisdiction within ninety days after manumission. In Virginia it was within a year. In Tennessee, Alabama, and Florida the time permitted for emigration was not specified in the respective state codes. Application of these laws was practically nullified, it is true, by resort to the right of petition to the county court or to the state legislature. With the support of character recommendations by white persons such petitions for permission to stay were customarily granted, and the colored freedman was permitted to remain in that state in which he had lived as slave.

In some states immigration of free Negroes from other states was forbidden. It was probably because of this elimination of the slaves’ ability to move and the restrictions placed upon the Negro freedman’s right of mobility that Mr. Justice Joseph P. Bradley, in his opinion written in 1883 upon civil rights cases, was led to recognize the loss of mobility as one of the six “inseparable incidents of the institution of slavery.”

In the pursuit of some compelling ideal in times of peace, and more drastically under conditions of warfare, when unusual powers of compulsion are granted to governments, far-reaching restrictions are placed upon individual mobility of all the population and accepted almost without complaint. Although these restrictions must often, in wartimes, be decreed without consultation with those against whom they apply, nevertheless in democracies they are to be regarded, in principle, as imposed by general consent. Under powers of compulsion acceded to by their subjects the governments of the United States and Great Britain have, for the period of the war, placed important checks upon the free movement of their populations. In Britain mobilization and transfer of its labor potential by government, both spatially and occupationally, has been carried out upon a wide scale. In the United States the limitations upon mobility imposed upon our people have not been onerous, applying chiefly to the free right of travel of the general public.

Quite aside from the imperious compulsions of the war effort, in the last two decades infringements upon the mobility of employees have been appearing in the United States which have projected the problem of the right of free movement into a new sphere. The locale of this projection lies entirely outside the boundaries of slavery and quite beyond the limits of anything which can correctly be classed as helotage. It may serve, however, to emphasize the historical continuance of restrictions upon spatial mobility as a method which is constantly and easily employed in the field of the relations between labor, whether free or enslaved, and those who employ labor. The new application appears in the adoption by some of the labor unions in the United States, under agreement with the employing companies, of the seniority rule. Under this type of agreement seniority in the plant or, in other cases, in a department of the plant, determines the question of the layoff and of rehiring upon the job when business revives. One who accepts a position in another shop loses the seniority which he had in his previous job and plant. At the first depression affecting the trade in which he is engaged he may be laid off along with any junior in the new plant. The desire for security in the job dictates the result that the seniority rule tends to operate against the union member’s desire to move from his present job or to change his domicile to a new locality in search of better conditions of living and of employment.

The sovereign states signatory to the Slavery Convention adopted by the assembly of the League of Nations in 1926 pledged themselves to abolish slavery “progressively and as soon as possible” and “to prevent compulsory and enforced labor from developing into conditions analogous to slavery.” By these words they recognized the continued existence of actual slavery and that conditions of labor compulsion were at hand which lie just beyond the boundaries of enslavement–and further that a little change might transform these into slave conditions. “Between free men and slaves stand the Helots and the Penestae.” According to the intent and application of this phrase of Julius Pollux, slavery and even helotage, which lies just beyond the periphery of enslavement, have disappeared in most parts of our modern world in the sense that they are no longer condoned or accepted by advanced world opinion. The vocabulary of the slave system, nevertheless, remains in constant use. The ancient words are the same; but their content is different. Except where actual private slavery or semiservile conditions, deriving from labor compulsion, still persist the word enslavement denotes another bundle of restrictions upon personal liberty. Fragile they may be; but they still are shackles–the necessary shackles which bind people together in the unavoidable process of adjustment between human beings who must work together in their social relations. The necessity of adjustment is the constant in these relations. In the application of this social constant, infringements upon freedom of movement, as applied to individuals or to social groups, are a first recourse of the empowered party in the combination. They may be imposed by external compulsions or they may be self-applied. In either case the free option of movement, as an instrumentality in the process, becomes a vital factor in the stabilizing of the relations between work-giver and worker. As such an instrumentality the priests of Apollo at Delphi isolated it over two thousand years ago and clearly defined the importance of the option of movement as one of the four essentials of human freedom.

William Linn Westermann was professor of history at Columbia University.