Part 2 — CHAPTER IV. AGRICULTURAL TRIBES.
Reading edition
§ 1. Numbers of positive and negative cases in the three agricultural groups.
The list given in § 3 of the first chapter of this second Part contains the following numbers of agricultural tribes with and without slaves, classified according to the division into three groups we have made in § 2 of the same chapter:
Second agricultural group.
a2 a2 + c a2 + t a3 + c + t Total.
———— ———— ———— ————— ————
Pos. Neg. Pos. Neg. Pos. Neg. Pos. Neg. Pos. Neg.
N. America 11 11
S. America 1 2 1 2
Melanesia 1 3 1 3
Polynesia 1 6 1 6
Micronesia 5 5
Malay Arch. 29 1 3 7 2 41 1
Indo-Chin. Penins. 5 5
India 4 4 1 5 4
Bantu tribes 18 6 7 3 12 1 38 9
Soudan Negroes 5 4 4 2 7 4 20 6
Hamitic peoples 2 1 2 4 1
66 43 17 5 26 0 7 0 116 48
Third agricultural group.
a2 a2 + c a2 + t a3 + c + t Total.
———— ———— ———— ————— ————
Pos. Neg. Pos. Neg. Pos. Neg. Pos. Neg. Pos. Neg.
N. America 1 1
Malay Arch. 9 1 9 1
Indo-Chin. Peninsula 1 1
Caucasus 1 1
Bantu tribes 1 1 5 6 1
Soudan Negroes 1 1
Hamitic peoples 2 1 2 1
10 1 3 3 6 0 1 0 20 4
§ 2. Development of agriculture and development of slavery.
What do these numbers teach us?
In the first place we see that many (170) agricultural tribes keep slaves. Hence it appears that slavery is by no means incompatible with agriculture. But there are also many (95) agricultural tribes without slaves, so the existence of agriculture among savage tribes does not necessarily lead to the keeping of slaves.
In the second place it appears that the more agriculture is developed, the more frequent slavery becomes. Looking at those agricultural tribes among which subsistence does not depend to any considerable extent on cattle-breeding or trade (a1, a2, a3), we find that in the first group there are 26 positive and 37 negative cases, i.e. 41,3 per cent. of these tribes keep slaves. In the second group the corresponding numbers are 66 positive cases, 43 negative cases, and 60,6 per cent.; in the third group 10 positive cases, 1 negative case, and 90,9 per cent. We see that in the second group slavery is more frequent than in the first, whereas in the third group it is almost universal. It has, however, to be taken into consideration, that the great majority (9 out of 10) of the slave-keeping tribes belonging to the third group live in the Malay Archipelago, and 5 out of these 9 are divisions of the Battas. We may not, therefore, attach much importance to the numbers relating to the third group; for they may be strongly influenced by local circumstances. Taking the second and third group together we find 76 positive and 44 negative cases, i.e. 63,3 per cent. keep slaves, which percentage is considerably higher than that of the first group.
We do not claim mathematical exactness for these numbers. But at any rate we may say that they sufficiently prove, that slavery is considerably more frequent among truly agricultural tribes, which subsist chiefly by agriculture, than among incipient agriculturists, who still depend on hunting or fishing for a large portion of their food.
The total numbers lead to the same conclusion. Looking at these we find in the first group 34 positive and 43 negative cases, i.e. 44,2 per cent. keep slaves. For the second group the corresponding numbers are 116 pos. cases, 48 neg. cases and 70,7 per cent; for the third group 20 pos. cases, 4 neg. cases and 83,3 per cent.; for the second and third group taken together 136 pos. cases, 52 neg. cases and 72,3 per cent.
The best specimens of this type are found in South America.
We have seen (in chap. II of this second Part) that hunters hardly ever keep slaves; and as the “hunting agriculturists” so much resemble true hunters, it is easy to understand why among the majority of them slavery does not exist. Slaves cannot be employed for hunting, and the women can easily perform the small amount of cultivation wanted by these tribes. Moreover, the men who, as warriors, are able to procure slaves, are not likely to take them for the sole benefit of the women. And where the men are always hunting, the women would have to supervise the slaves and keep them in order, which is not easy for them.
That this lowest stage of agriculture is not favourable to the growth of slavery, is confirmed by what Mr. A. C. Kruyt had the kindness to write us regarding Central Celebes. There is one native tribe, the Topebato, that formerly did not keep slaves, the probable cause being that they had remained hunters longer than the other tribes in the neighbourhood. In a legend, in which it is told that the gods made gifts to the different tribes characterizing their manner of life, the Topebato get a dog allotted to them for the chase, and though now they till the soil, they are still passionate hunters. They now buy a few slaves and so in the course of time a slave class will originate among them.
Yet there are a considerable number of positive cases in our first group (26 out of 63). We will, of course, make due allowance for mistakes; there may be several tribes contained in our first group which on closer scrutiny would prove to be true agriculturists and not Jägerbauern. But we cannot think but that among these 26 tribes there are many, which have been justly placed in the first group. The existence of slavery among them will have to be accounted for by secondary causes, internal and external, such as we have found in the foregoing chapters and of which we shall perhaps find some more in the continuation of this chapter.
We must, however, bear in mind that our first group does not only contain “hunting”, but also “fishing agriculturists”; and we know that fishers are more likely to keep slaves than hunters. This may perhaps account for the existence of slavery among some of these tribes. But our numbers give us no hint in this direction. We find, indeed, that 11 out of 26 positive cases are afforded by tribes inhabiting the islands of the Malay Archipelago; but among these there are some Dyak tribes living in the interior of Borneo. Moreover, South America affords 6 positive, and the Melanesian islands 12 negative cases. We may suppose that whatever effect this factor has is neutralized by the intervention of other circumstances.
§ 3. Capital and labour among agricultural tribes.
We have seen that among the purely agricultural tribes of the second and third stages there are altogether 76 positive and 44 negative cases, i.e. more than three-fifths of them keep slaves. These numbers make it probable, that the existence of slavery among agricultural tribes has to be accounted for by general causes, that agricultural life as such is favourable to the growth of slavery. We have to inquire now, whether this supposition is justified by an application of the general principle laid down in § 6 of the second chapter, i.e. whether among agricultural tribes subsistence is dependent on capital, and if not, whether it is easy or difficult to procure.
Moreover, subsistence is fairly easy to procure. Agriculture, where it is carried on in such simple manner as among most savages, does not require much skill or application. As compared with hunting, seafaring and manufactures, it is rather dull work, requiring patience rather than strength or skill. It is one of the occupations about which there is no excitement, and which in many primitive societies are performed by the women. Hunting requires personal qualities, and a good hunter is held in high esteem; but we have not found it stated in a single instance, that a man’s influence or power depends on his ability in agriculture.
We see that the general economic state of truly agricultural tribes may account for the existence of slavery among so many of these tribes. We shall now inquire what secondary causes there are at work among agricultural tribes, and what effect they have. But we shall have to speak first of a great factor in economic life, which we have not met with before.
§ 4. Land and population.
The general principle laid down in the last paragraph is that in primitive agricultural societies capital is of little use and subsistence easy to acquire; therefore every able-bodied man can, by taking a piece of land into cultivation, provide for himself. Hence it follows that nobody voluntarily serves another; he who wants a labourer must subject him, and this subjection will often assume the character of slavery.
But this general rule requires an important qualification. Hitherto we have supposed, that there is much more fertile land than is required to be cultivated for the support of the actual population. Such, indeed, is the case among most savages; but it is not always so. And where it is not so, our general rule does not obtain. When all land fit for cultivation has been appropriated, a man, though able-bodied and willing to work, if he owns no land, cannot earn his subsistence independently of a landlord; he has to apply to the owners of the land for employment as a tenant or servant. In such case free labourers are available; therefore slaves are not wanted.
In this and the ensuing paragraphs we shall endeavour to prove by facts the hypothesis arrived at here by a deductive reasoning, which we may express thus: where all land fit for cultivation has been appropriated, slavery is not likely to exist.
The same view is held by some theoretical writers.
Loria also holds that slavery requires an abundant supply of ground; but his arguments are quite different from Cairnes’. His reasoning is as follows.
As long as there is land not yet appropriated, which a man destitute of capital can take into cultivation, capitalistic property cannot exist; for nobody is inclined to work for a capitalist, when he can work for his own profit on land that costs him nothing. If, then, the capitalist wants by any means to get a profit, he must violently suppress the free land to which the labourer owes his force and liberty. And as long as the population is scarce and therefore all land cannot possibly be appropriated, the only means of suppressing the free land is by subjugating the labourer. This subjugation assumes at first the form of slavery; afterwards, when the decreasing fertility of the soil has to be made up for by a greater fertility of labour, slavery gives place to serfdom, which is milder and makes labour more productive.
When the population increases, and all land that can be cultivated by labour without the aid of capital had been appropriated, quite another state of things prevails. The labourer has now no other resource but to sell his labour to the capitalist for such wages as the latter likes to give; he is compelled to yield to the capitalist the greater part of the produce of his labour. Now the latter need no longer use violence to get his profit; for it falls to him by the automatic operation of the social system. Yet, even then the capitalistic régime is not absolutely certain to arise, for there is still land not yet appropriated, that can be cultivated with the aid of capital. If, therefore, the labourers could save a portion of their earnings and thus accumulate capital, they would be able to take this land into cultivation and so make themselves independent of their employers. This consideration induces the capitalist to keep wages so low that they cannot exceed the immediate wants of the labourers, which he brings about by various artificial means.
When, finally, a further increase of population makes the total appropriation of the land possible, the mere appropriation of it by the capitalist class renders the labourers for ever subjected. The capitalist need no longer have recourse to artificial methods of reducing wages; the system operates automatically. The capitalists have only to retain the land for themselves; they will then secure a perpetual revenue at the cost of the labouring class.
Much more fully has the true reason, why in densely peopled countries there is little use for slaves, been recognized by Wakefield in his book on the art of colonization. With him, the theory is not based upon a general conception of society, but upon the facts of the colonial history of his own time.
Wakefield, then, complains that in Australia and other colonies manufactures cannot thrive; the reason for this is, according to him, that there are no labourers to be had; for there is so much free land that every newly-arrived labourer becomes a landowner rather than work for wages. Therefore there are many colonies which would keep slaves if the home government let them. This leads the writer to an investigation of the circumstances which induce men to keep slaves.
“They are not moral, but economical circumstances; they relate not to vice and virtue, but to production. They are the circumstances, in which one man finds it difficult or impossible to get other men to work under his direction for wages. They are the circumstances … which stand in the way of combination and constancy of labour, and which all civilized nations, in a certain stage of their advance from barbarism, have endeavoured to counteract, and have in some measure counteracted, by means of some kind of slavery. Hitherto in this world, labour has never been employed on any considerable scale, with constancy and in combination, except by one or other of two means; either by hiring, or by slavery of some kind. What the principle of association may do in the production of wealth, and for the labouring classes, without either slavery or hiring, remains to be seen; but at present we cannot rely upon it.…
“Slavery is evidently a make-shift for hiring; a proceeding to which recourse is had, only when hiring is impossible or difficult … it is adopted because at the time and under the circumstances there is no other way of getting labourers to work with constancy and in combination. What, then, are the circumstances under which this happens?
There is one more reason why slaves are of little use in those countries where all land has been appropriated. When there is free land, a man can, by increasing the number of his slaves, to any extent augment his revenue: every slave will take a new patch of land into cultivation; the more slaves a man owns, the more land he will have in tillage. But when the supply of land is limited, each landowner can employ only a definite number of labourers. As soon as there are hands enough to cultivate his grounds, an increase in the number of labourers soon becomes unprofitable. What we have said of pastoral tribes obtains here too: it may be that slaves are wanted, but when they are procured the point will soon be reached at which a further increase in their number yields no longer any profit. Therefore, when all land has been appropriated, even though it be equally divided between the members of the community and so a labouring class be wanting, there is little use for slaves.
It must be understood that we speak here of self-dependent agricultural countries. Where manufactures and the trade with foreign parts are highly developed, economic life becomes much more complicated and presents quite another character.
What we want to prove is that in such self-dependent agricultural countries, when all arable land has been appropriated, slavery is not likely to exist.
It is not always clearly stated whether all land has been appropriated. Then we shall have recourse to some criteria from which we may infer whether such be the case.
The appropriation of the land does not imply that all land is actually being cultivated. There may be land actually out of tillage and yet valued by the owner. But when it is stated that all land is being cultivated, it must all have been appropriated. This will therefore be our second criterion.
There is another criterion that proves that all land has not yet been appropriated. When we are told that clearing a piece of land is a modus acquirendi of landed property, there must still be free land.
We shall not, in order to prove our hypothesis, examine the regulations of landed property among all agricultural savage tribes, but confine ourselves to one geographical group, in which the phenomena we have spoken of in this paragraph most strikingly present themselves. This group is Oceania, comprehending Polynesia, Micronesia, and Melanesia. We shall, however, leave out of regard New Guinea, one of the largest islands of the world. The rest of Oceania consists chiefly of small islands.
Slavery in Oceania (with the exception of New Guinea) has never prevailed to any great extent. In the second chapter of Part I it has been shown that slavery, so far as we can know, existed only in the N. W. Solomon Islands, on the Gazelle Peninsula of Neu Pommern, and in New Zealand.
We shall try to account for this fact by showing that on most of the Oceanic islands all land had been appropriated, which led to a state of things inconsistent with slavery as a social system.
In the following paragraphs we shall inquire what our ethnographers have to say about landed property in Oceania and the extent to which the land had been appropriated.
§ 5. Land tenure in Polynesia.
In this and the following two paragraphs we shall not mention all particulars of land tenure given by the ethnographers, but only those which may enable us to decide whether all land had been appropriated.
We see that on most Polynesian islands all land has, or had been appropriated. Regarding Rotuma, Tahiti and Easter Island this is explicitly stated; and the same must be the case in Tonga, Hawaii, Rarotonga, the Marquesas Islands, Mangarewa, part of the Tokelau group, and the Abgarris, Marqueen and Tasman groups, where classes destitute of land are found. In Manahiki property in land was strongly developed; but whether there was still free land is not clear. In Samoa there was still unappropriated bush land, though this was already “sometimes claimed by those who own the land on its borders.” In New Zealand clearing was still a modus acquirendi, which proves that all land had not yet been appropriated. Equally on Savage Island titles can be acquired by cultivation.
§ 6. Land tenure in Micronesia.
The conclusion is that on most of the islands of Micronesia all land has been appropriated, most often by the upper classes to the exclusion of the lower. In Pelau, though a vast amount of land is actually out of tillage, the regulation of landed property related by Kubary proves that all land is held as property. Here, as well as in Rotuma and Tahiti, we have to deal with the effects of the depopulation that has taken place in Oceania. In Mortlock there seems to be free land; but Kubary’s account is not very clear.
§ 7. Land tenure in Melanesia.
From the foregoing it appears that in many parts of Melanesia clearing is a modus acquirendi, viz. in the Solomon Islands, Northern New Hebrides, Banks Islands, Santa Cruz Group, New Caledonia, Gazelle Peninsula of Neu Pommern and Nissan Islands. Yet the rights of landowners are recognized everywhere in these islands. Here, as Ricardo would say, land of the second degree of fertility has already been taken into cultivation, and so rent has commenced on that of the first; but there is still free land. In Aneityum, too, there seems to be land not yet appropriated. In Fiji people destitute of land are found. Among the Western tribes of Torres Straits all arable land is divided up into properties, as Haddon tells us; but whether the rest of the land is still free is not quite clear. With regard to the Murray Islands we cannot arrive at any definite conclusion.
Generally speaking we may conclude that in Polynesia and Micronesia all land has been appropriated, whereas in the Melanesian Islands free land still exists.
We see further that not only the arable land is held as property, but often also the fruit-trees, lakes and streams, the shore and the lagoon as far as the reef. On most Polynesian and Micronesian islands whatever portion of land or water can yield any profit has been appropriated.
§ 8. Landlords, tenants and labourers in Oceania.
Another state of things would not be inconceivable. It were quite possible that every inhabitant had appropriated a portion of the land, nothing of it remaining unclaimed. Yet it is easy to understand that, when all land has become individual property, a class of people destitute of land is likely soon to arise. In large families the portions falling to each of the children will often become too small to live upon. And where it is customary to buy and sell land, there may be improvident people who squander the land that was to afford them subsistence. But the principal cause probably is the arbitrary conduct of the chiefs and other men of power who appropriate the land of their enemies, and even, under some pretext, that belonging to their own subjects.
It appears that this conquering of land does not always create a class destitute of land; sometimes the inhabitants have only to pay a tribute. But where individuals belonging to the victorious tribe receive portions of the conquered land allotted to them, as in New Zealand and the Kingsmill Islands, or where, as in Samoa, the land becomes the private property of the conquering chief, the original owners consequently are deprived of their property.
It also occurs that within the tribe the land is taken away from its owner.
We can now perfectly understand why people destitute of land are found in so many of these islands. And as most often not only the arable land, but fruit-trees, lakes, streams, and the sea adjoining the land are individual property, these people are entirely at the mercy of the landowners. We shall see that they have to perform the drudgery for the landlords, and are sometimes heavily oppressed.
Gardiner, in his description of Rotuma, makes no mention of social classes.
Regarding the condition of the common people in Micronesia we have already mentioned many particulars in § 7 of the second chapter of Part I in inquiring whether they were to be regarded as slaves, and in § 6 of this chapter in order to prove that all land had been appropriated. We shall briefly repeat here what bears on their condition and the work imposed upon them, adding such details as have not yet been mentioned.
On Nauru, the lower classes (sometimes called “serfs” or “slaves” by the authors) are in the service of the chiefs and nobles.
In Ebon, the common people live on land allotted to them by the chief who can take it from them at his pleasure. Every week they have, each of them, to provide the chief with a fixed quantity of food.
In Mortlock, according to Kubary, social ranks do not exist.
In Yap, the lowest class (whom Gräffe wrongly calls slaves) are obliged daily to provide the upper classes with agricultural produce, and whenever the chiefs require it to aid in constructing houses and canoes. Whatever they possess, even to their wives and daughters, may at any time be required by the upper classes. Yet all labour is not exclusively incumbent on them. They are only bound to definite taxes, viz. to a tribute of victuals, and of mats and other materials for housebuilding; and their “slave-state” consists rather in a low and dependent condition than in being taxed with labour.
In Pelau, according to Kubary, there are no social classes; but the wants of the chiefs are generally provided for by the work of dependent relatives, who are treated as adopted children and may at any time leave their employers. Semper, however, speaks of a despised working class.
On the Kingsmill Islands, according to Parkinson, there are two subjected classes. One is the class of the te torre, who live as vassals on the lands of the great landholders; they get a small piece of land for their own use; they must provide their lord with men when at war, and bring him the number of cocoanuts he desires, and what he needs for his household. The lowest class are the te bei or kaungo. They have no property, no land to live upon; they live with the great landholders by whom they are maintained; they on their part must work for their lords, i.e. fish, prepare food, etc. The lord, by giving them a piece of land, can raise them to the class of the te torre. These two classes have no voice in government matters; they follow their lord without grumbling; his will is their will; an offence against the lord is regarded by them as a personal offence, and avenged as such. Generally no one marries outside his class. In ordinary life there is no difference between master and vassal; they drink, dance, and play together; they wear the same kind of dress.
We shall inquire now what is the condition of the lower classes in Melanesia.
This conclusion is strengthened by consulting some other writers.
Hunt, in his description of the Murray Islands, makes no mention of social ranks.
This depopulation may perhaps also account for the discrepancy between Semper’s and Kubary’s accounts of Pelau. Semper spoke of a despised working class; but Kubary, who wrote several years after, stated that there were no social ranks. It is quite possible that in the meantime the population had been so greatly diminished, that every one could obtain possession of a piece of land.
There are some more details on record proving that labourers were not wanted.
We see that common labourers are little wanted in Oceania. Some kinds of workmen, however, are much in request in some of these islands.
We see that those trades which require no particular abilities are most despised here, whereas skilled labour is highly honoured and performed even by the higher classes.
In some more cases it is stated that skilled workmen are better paid and more highly valued than unskilled.
Skilled labour is thus highly valued in some of these islands. The skilled workmen, so far from being slaves, are held in high esteem; but those who have no peculiar accomplishments are obliged to perform the rudest and most despised work. This applies especially to the agricultural labourers. These are entirely dependent on the landowners, and there are more of them than can be profitably employed. They much resemble the proletarians of modern European countries.
Their state would even be worse than it actually is, were it not that they are useful in another, non-economic way: they strengthen their employers’ force in warfare.
Concluding, we may remark that the facts observed in Oceania fully justify our theory, that slavery is inconsistent with a state of society in which all land is held as property.
§ 9. Transition from serfdom to freedom in Western Europe.
But we must first remind the reader of what we have already said in the Introduction. We confine ourselves in this book to an investigation of the facts of savage life. The study of these facts leads us to conclusions, some of which (and among these the conclusion we are now dealing with) have a wider bearing and can further our understanding of the history of civilized nations. But the scope of the present volume does not allow us to make any special investigation of this history and inquire whether the same causes can be seen at work here that have been found to shape the social life of savage tribes. Accordingly, we shall not try to prove that the appropriation of the soil has really been the main cause of the disappearance of slavery and serfdom in Western Europe. We only intend to show that the matter can be viewed from this side. We wish to claim attention for this important factor, that is commonly overlooked, and thus clear the way for future research. This and the next two paragraphs have to be regarded as a digression, standing apart from the main body of our book.
The character of these paragraphs may justify us in limiting our remarks to two countries, the economic history of which has of late years been the subject of thorough study by eminent writers, viz. England and Germany, and using only a small part of the literature existing on this matter. We have, however, taken care to consult none but first-rate authorities.
We have spoken of the disappearance of slavery and serfdom. In connection with this two remarks have to be made.
It is of some interest to emphasize what we have said concerning the difference between serfs and free tenants. A free tenant, whatever be the conditions of his tenure, can always remove from the land and cannot be compelled to cultivate it. A serf, whatever be the extent of his obligations, is bound to the soil; if he escapes the lord can bring him back and set him to work again. The right of emigrating (German “Freizügigkeit”) is the true mark of freedom. And therefore, as soon as the obligations which were personal have become territorial, i.e. as soon as the services and payments which formerly were exacted from definite persons, are exacted from the cultivators of definite pieces of land as such, nobody being any longer obliged to become or remain the cultivator of any definite piece of land,—serfdom has ceased to exist, even though the services and payments have remained exactly the same.
Now we cannot wonder that the lawyers and landlords of the Middle Ages had no very clear ideas about serfdom and freedom. But modern writers on economic history should have the true distinction always before their minds. Some of them, however, we think fail in this respect.
Now, though none of our writers on economic history explicitly say that these two things, the commutation of labour dues for money and the transition from serfdom to freedom, are identical, we think some of them do not sufficiently keep in view the difference existing between the two. Thus only can we account for the prevalence of a theory which seems to be the current mode of explaining the fall of serfdom and rise of freedom.
This theory has been introduced by a German writer, Professor Hildebrand. He distinguishes three stages of economic development: natural economy, money economy, and credit economy.
In the system of natural economy goods are exchanged directly for goods; when money economy prevails use is made of a means of exchange, money; and when credit economy has been developed goods are exchanged for a promise in the future to give back the same or a like value, i.e. on credit. Every nation begins with natural economy, for the use of money as a means of exchange supposes an abundance of labour or products of labour which enables people to procure the precious metals. As long as natural economy prevails capital does not exist: the soil and human labour are the only productive agencies. There are, therefore, two classes of people only; labourers and landowners. Sometimes every landowner is at the same time a labourer; in such cases democracy prevails. But it often occurs that labourers and landowners form separate classes. These are then mutually dependent on each other; for the labourer wants a landlord to give him employment and so enable him to earn his subsistence, and the landlord wants labourers to cultivate his lands. This interdependence effects that the relations existing between the two classes assume a durable character. The labour contracts are made to last for the life of the labourer or even become hereditary. The labourer is bound to the soil and forbidden to leave the manor.
As soon as money economy exists, capital arises and takes its place as the third factor of production. The owning classes comprise now both capitalists and landlords. The labourer has no longer to apply to the landlord for employment, but can leave him and work for the capitalist; he is therefore no longer astricted to the soil. The wages he receives from the capitalist are paid in the form of money, and so the labourer is much freer than before, for this money can be turned to various purposes. Moreover, capital (as opposed to land) can be augmented to any extent, and this enables the labourer to become a capitalist himself.
The position of the labourers who remain on the land also undergoes a change. The landlord who brings the agricultural produce to the market can pay wages in money and therefore hire able and dismiss incompetent labourers. His lands, worked with free labourers who serve for wages, yield him a far greater income than formerly when they were cultivated by serfs. Moreover, the fixed labour dues of the serfs do not answer the purposes of an improved economy. It is thus the interest of the landlord to put an end to his fixed and hereditary contracts and loosen the ties with which natural economy had bound the agricultural labourer. The dues in kind and services are commuted for money payments. The labourer, who was a serf, becomes now either a free peasant or a free servant and day-wage worker who is no longer astricted to the soil, but can leave his employer whenever he likes and seek such work as most agrees with his capacity and inclination.
This is Hildebrand’s theory, so far as the condition of the labouring classes is concerned. Natural economy, according to him, leads to serfdom, money economy leads to freedom.
We think this theory is erroneous.
But even if we speak of self-dependent and commercial communities, we cannot admit that in the former labour must necessarily be servile and in the latter free.
First. How can natural economy, i.e. the absence of commerce, lead to serfdom? Hildebrand says: landlords and labourers are mutually dependent on each other, and so their relations assume a durable character, and the labourer is astricted to the soil. We think his meaning is the following. In self-sufficing communities the fluidity of labour which exists in modern society, is wanting. In such countries there can be famine in one district, whilst in a contiguous district there is plenty of food; similarly labour can be scarce in one place whilst it is abundant in a neighbouring place. Therefore a landlord cannot afford to let his labourers leave the manor; for as there is little intercourse between the different villages and districts it is difficult for him to procure other labourers. It is thus most convenient for him to bind his labourers to the soil and forbid them to leave him.
This may at first sight seem a reasonable explanation of the origin of serfdom. But on closer scrutiny it will be seen that this argument does not hold. When there is little intercourse, each landlord is dependent on the labourers of his own district; and there must be a great stability in the relations of the two classes. But this need not bring about an astriction of the labourers to the soil. The landlord cannot easily procure labourers from other districts; but it is even more difficult for the labourers to find employment in foreign parts; for such intercourse as there is, is kept by the ruling, not by the labouring classes. Therefore it is not necessary to bind the labourers to the soil; for they are already naturally dependent on the landlords of their own district. We think slavery and serfdom can only be accounted for by a general scarcity of labour. When labour is everywhere scarce a labourer who leaves his employer can everywhere find employment, whereas an employer cannot easily procure labourers; it is then the interest of the employer to prevent his labourers from leaving him. But the mere lack of intercourse limits the labourer in his choice of employment even more than it limits the employer in his choice of labourers.
Our conclusion is that, though in medieval Western Europe serfdom and natural economy existed at the same time, the former is not a necessary consequence of the latter.
Secondly. Does money economy, i.e. commerce, always lead to freedom? We know now that serfdom is not invariably connected with natural economy. Yet it might be that, wherever both natural economy and serfdom exist (as it was the case in the early Middle Ages) the rise of money economy always brought serfdom to a close.
The argument by which Hildebrand attempts to prove this is rather strange. The development of town life and manufactures, according to him, enables the labourers to find employment in manufactures; they are now no longer dependent on the landlords. The manufacturing capitalists pay them money-wages which they can spend in whatever way they like, and so they become more free than they were before.
We think this argument is quite insufficient. The labourers find employment in manufactures, says Hildebrand. But he had told us before that they were astricted to the soil. What enables them now to leave the landlords? Further: why do not the town labourers become slaves or serfs? Here Hildebrand’s reasoning is very strange. They receive money-wages which they can spend in whatever way they like. Now one who receives money with which he can buy all kinds of commodities is in a certain sense more free than one who, under a system of natural economy, receives bread and meat which he cannot sell. But this has nothing to do with the legal status of the labourer. A slave who receives pocket-money from his master is free to buy with it what he likes, yet he remains a slave.
But the condition of those of the labouring classes who remain on the land also undergoes a change, according to Hildebrand. The landlord can now sell the produce of his land for money, and this money enables him to hire free labourers.
We cannot see why this should be so, why the mere possibility of paying money-wages (all other circumstances having remained the same) should lead to free labour contracts. We should rather think that the afflux of labourers to the towns of which Hildebrand speaks would make agricultural labour scarce, and each landlord would be most anxious to retain those labourers that had not yet escaped to the towns; they would now, more than ever before, be astricted to the soil.
Hildebrand, however, thinks it will be the interest of the landlord to put an end to the hereditary tenures of his serfs and work his lands with free labourers. And he adds that the dues in kind and services are commuted for money payments. The cultivator who was a serf becomes now either a free landholder or a free labourer.
Whether Hildebrand means to say that the commutation of dues in kind and services for money payments is identical with the transition from serfdom to freedom, does not clearly appear.
We think that the regarding of this commutation as the main fact in the economic history of the later Middle Ages lies at the root of the evil and has given rise to this theory. Money economy, according to Hildebrand, leads to commutation, and commutation is the same as, or at any rate leads to, the disappearance of serfdom.
Hildebrand’s theory is: money economy led to commutation and commutation led to freedom. We have seen that the first half of this does not hold. What about the second half? Can the commutation have loosened the ties which bound the cultivator to the soil? We think not. For if the landlord could not let the villein who worked on the demesne leave the manor, because he was difficult to replace, he had exactly the same reason for keeping the villein who paid money astricted to the soil. The use of money and the rise of commerce had not augmented the number of agricultural labourers; they had even decreased, as many of them had gone to the towns. And it was even more difficult to replace the money-paying than the labouring villein; for the former had to be a fit person who could conduct his business well enough to be able at the end of the year to furnish the required sum, whereas any able-bodied man could perform agricultural labour under the supervision of the lord’s bailiff. Our conclusion is that money economy did not lead to commutation, and that commutation did not lead to freedom.
The disappearance of serfdom was thus accelerated by the measures of the Italian cities. How much of sentiment there was in these measures, and how much of self-interest, we do not know. But at any rate serfdom must already have been drawing to an end before the cities meddled with it. For a firmly established system that discharges an important economic function is not uprooted by mere sentiment. And so far as the self-interest of the citizens induced them to replace the serfs by free tenants, the latter system must have been economically more useful than the former, which was probably only kept up by the landlords because they were accustomed to it. Times had changed and the old system of cultivation had become obsolete; and the citizens of the towns, whom no personal relation bound to the serfs, expelled them and let the land to free tenants. Before their intervention there must already have been at work an internal cause, which effected that cultivation by serfs was no longer the most profitable mode of managing landed properties.
But however this may be, we are certain that the rise of money economy cannot have been the sole, or even the chief cause of the disappearance of servile labour. This is sufficiently shown by the fact that before the emancipation of the Negroes a system of servile labour on a large scale prevailed in the United States and the West Indies, i.e. in countries working for export.
Our conclusion is that the rise of money economy was not the cause of the disappearance of serfdom. We shall inquire now whether Wakefield’s theory, with which we agree, can further our understanding of the economic history of England and Germany.
§ 10. The rural classes of medieval England.
The first detailed account of the economic condition of the country is contained in Domesday Book, in which William of Normandy embodied the results of an inquiry into the state of the kingdom he had secured.
All this agrees with our theory. All land had not yet practically been appropriated; therefore people could not be got to cultivate the land for the benefit of the landlords, unless they were deprived of personal freedom.
Here again our theory holds. Population had increased, land became scarce, and the transition from serfdom to freedom commenced. If the population of England had continued increasing, most of the villeins would probably have become freeholders or copyholders, whereas the lands that the lords had kept in their own hands would have been leased. And poor people who had neither land of their own nor capital enough to become farmers would have served for wages.
The depopulation of this time caused a reappearance of free land, i.e. of land which had practically no value, and so agricultural labourers were scarcely to be had. Therefore the lords to some extent reattached the cultivators to the soil.
Here again we see the consequences of the abundance of land. The land alone could not fetch a reasonable price; stock and land had to be leased together.
As these leaseholders were taken from the labouring class, this measure, of course, still further diminished the supply of labour.
But far graver evils resulted from the appropriation by the lords of the commons and the land held by villeins or customary tenants.
Even when the cultivator had not left his tenement, the lord sometimes appropriated and “inclosed” it.
Our theory can thus be of much use in accounting for the changes which have taken place in the rural economy of England. As long as there was still free land, i.e. land which, though sometimes claimed by an owner, could not fetch any reasonable price, the cultivators were astricted to the soil; but as soon as the proprietors had got the practical command over the whole of the land, many of the villeins were evicted and replaced by leaseholders or tenants at will or became such themselves; and the remainder became copyholders, i.e. proprietors obliged to some services or payments without being personally unfree.
We shall inquire now whether in Germany too the appropriation of the whole of the land coincided with the transition from serfdom to freedom.
§ 11. The rural classes of medieval Germany.
Land was thus abundant, slavery existed, and tenant farmers and free labourers were absent.
We see that in this period there was still much free land; slavery existed and serfdom was on the increase; leaseholders, tenants at will and free labourers were wanting. All this agrees with our theory.
At the same time a class of free tenants arose. Lamprecht remarks that while the value of land had considerably increased the tributes which the villeins had to pay had remained unchanged for several centuries. In an economic sense the landlords had been dispossessed of a large proportion of their property in the land. Therefore it was not their interest to let serfdom continue.
We cannot but think that the reason given by Lamprecht for the transition from servile to free tenure is true. For even when the original rent was not higher than the former customary payment plus the value of the labour dues, the possibility of raising the rent after each term remained.
We hear of free tenants in this period, but not yet of free labourers. This is exactly what our theory teaches us to expect. Land, in some parts of Germany, had already acquired a high value; such land must have been very remunerative, and so people were ready to pay a rent for its use, even though there was still land to be had gratis or at a nominal rent, but far from the market and therefore less profitable. But the country was not yet so densely peopled that there were men who could not secure the use of any piece of land; therefore a class of people dependent on wages did not yet exist.
All land was now held as property; consequently the land was more and more held by free tenants and worked with free labourers dependent on wages, whereas serfdom gradually died out.
In the 15th century, however, according to both Lamprecht and Inama-Sternegg, serfdom and even slavery reappeared.
Serfdom, in Southern and Western Germany, thus died out towards the end of the Middle Ages, at a time when population had become numerous and land scarce.
We think the above remarks on England and the older parts of Germany may suffice to show that our theory can throw some light on the agrarian history of Western Europe.
§ 12. Open and closed resources.
Savage tribes with closed resources are: 1º possibly some Australian hunters, if it is true that among them every inch of ground is held as property, 2º the Eskimos (fishers), who cannot get on without a boat, or a sledge and dogs, 3º all pastoral tribes, 4º the agricultural tribes inhabiting most of the Polynesian and Micronesian islands, the Fijians and perhaps a few agricultural tribes outside Oceania.
We shall not inquire whether the civilized nations of ancient and modern times have, or had, open or closed resources. We will only remark that in Western Europe resources, from open, have become closed. Yet they are not altogether closed, as long as there are still thinly peopled countries open for emigration. Whether the white races will still have room for expansion for a considerable time, we cannot know.
In the following paragraphs we shall speak of the effect of secondary causes among agricultural tribes. We shall not, however, enter into many details. The difficulty, in our branch of science, is always, that we have so few works of predecessors to rely upon. For instance, we shall speak of the influence of trade among agricultural tribes. Now, if any accurate researches had been made into the general effects of trade, we should be able to conclude that trade having been proved to have such general effects, it must have such an influence on slavery. But as such is not the case, we should be obliged, if we were thoroughly to investigate the subject, to inquire what are the general effects of trade. An equally close study of militarism, of the condition of women, etc., would be required. And as in this way our book would never come to an end, we shall content ourselves with giving a few outlines, which we hope may turn the attention of other ethnologists to the important problems which the ensuing paragraphs will contain.
§ 13. Condition of women.
As we have remarked before, the position held by the women of a tribe determines to some extent whether or not slaves are wanted. Where all the drudgery is performed, and can be performed, by the women, and the men do not want to relieve them of it, there is no great use for slave labour. But where women enjoy high consideration, the men are more likely to procure slaves who are to assist the women in their work.
We shall speak first of the latter fact, of slaves performing female labour. It is very often stated that slaves are employed for domestic labour. And as, in countries where slavery does not exist, domestic labour is nearly always incumbent on the women, slaves who perform such work alleviate the women’s task. Where slavery prevails to a great extent it even occurs that slave-owners, female as well as male, have scarcely anything to do, all work falling to the share of the slaves. The slave-owners, in such case, form the aristocracy; the slaves, and the poor freemen unable to purchase slaves, are the labouring classes. We may remind the reader of ancient Rome, where the domestic slaves, the familia urbana, performed all domestic services required by the rich, and of the women of the upper classes in Mohammedan countries, who spend their time in idleness in the harems.
We have seen that among some pastoral tribes domestic labour is the chief occupation of the slaves. We do not recollect having found any instance of a similar state of things among any agricultural tribe, and cannot think that such will anywhere be the case. A rich cattle-keeper can easily support, by the produce of his cattle, some domestic slaves who perform no productive labour. But among agricultural tribes it is otherwise: subsistence here depends almost entirely on labour; therefore slaves performing unproductive labour can only be kept if there are other slaves who till the soil and procure food for the family. It is not probable that the master will himself undertake the cultivation required to feed the slaves who assist his wife in her work; nor would the wife be glad to receive slaves from her husband, if she had to provide for them by working on the field. The Romans would not have kept a familia urbana, if there had not also been a familia rustica.
Hitherto we have spoken of unproductive female labour. But women, in primitive agricultural societies, often perform productive labour also: in many cases the tilling of the soil is incumbent on them. We may suppose that the introduction of slaves has often served to free the women from this task. We shall not, however, proceed to a closer examination of this point; for this would require a digression on the division of labour between the sexes, which falls beyond the scope of the present volume.
On the other hand, the place of slaves is often supplied by wives. We have seen that among the Australian hunters polygamy widely prevails and serves economic purposes. The same is the case among some agricultural tribes.
We see that these Melanesian wives supply the place of slaves. They are bought like slaves; they have to work for their owners like slaves; and their labour, like that of slaves, increases the wealth of their lords. Another point of resemblance is this. In slave countries it is generally the rich only who are able to procure slaves; poor freemen have to work for themselves. Here it is the rich who appropriate the women; and many of the poor have to remain single. Here, as in all countries where polygamy is practised, it is only the minority of the men who can live in polygamy; for everywhere the number of women is nearly equal to that of men. And as in Melanesia the rich, who otherwise would want slaves, have many wives to work for them, slaves are not required.
§ 14. Commerce.
In § 1 of this chapter we have found the following numbers of positive and negative cases among commercial agricultural tribes:
Positive. Negative.
a1 + t 8 3
a2 + t 26
a2 + c + t 7
a3 + t 6
a3 + c + t 1
Total 48 3
We use the term “commercial tribes” in a wide sense, as including not only those tribes among which many people subsist by trading, but those that exchange a considerable part of their produce for foreign merchandise. For instance, a tribe that grows corn for export is a commercial tribe in the sense we attach to the word.
We see that, with very few exceptions, all commercial agricultural tribes keep slaves. This proves that among agricultural tribes the development of trade greatly furthers the growth of slavery.
The last point is an important one. In self-dependent agricultural countries the main use of slave labour consists in providing the master with food. If, then, a man keeps a large number of slaves who work for him, he is able to entertain his friends, or to keep a retinue of unproductive slaves or servants, whose wants are provided for by the work of the soil-tilling slaves. But where this is the sole profit one can derive from one’s slaves, an owner who keeps a considerable number of them does not want to make them work very hard; he often contents himself with receiving a tribute, and so the slaves become serfs. In this way the slave-owner gets less out of his slaves than would otherwise be the case; but he does not want more, and he need not now continually supervise their work. Slavery is not likely to exist on a large scale.
Of such “wholesale slavery”, as Bagehot terms it, we find a few instances among savage tribes.
It is clear that among these tribes slavery would not prevail to such a great extent, if the preparing and transporting of the articles of export did not require so much labour.
In these cases trade is the cause of “wholesale slavery”, not necessarily of slavery in general. “Retail slavery” may have existed among these tribes before they became so largely commercial. But, seeing that among them the extension of commercial relations has so greatly increased the use of slave labour, we may safely suppose that in several cases the development of trade has given rise to slavery among tribes which did not practise it before. This is also made probable by the list given at the beginning of this paragraph.
We shall not proceed to a closer investigation of this subject. We have already remarked that as yet we know very little about the general effects of trade and the place it occupies in social life among savages. And we must know more of this, before we can arrive at any accurate conclusion with regard to the influence of trade on the rise and growth of slavery.
When speaking of hunters and fishers, we have found that the influence of trade is more considerable where manufactured goods, than where raw products are exported. This will probably also apply to agricultural tribes, viz. if we take the term “manufactured goods” in a wide sense, as including agricultural produce. Raw products in our sense means articles which can be exported without any labour being previously applied to them, e.g. the various kinds of stone and earth exchanged by Eskimos and Australians. The articles exported by commercial agricultural tribes are nearly always manufactured goods in this wider sense.
It might be interesting to divide the commercial tribes (in the wider sense) into three categories, according as they export agricultural produce, manufactured goods in the common, restricted sense, or articles purchased abroad (articles of transit trade), the last category comprising the commercial tribes in the restricted sense, and inquire what are the social effects of commerce in each case. But such a subject wants separate treatment; we cannot deal with it here. We will only express our opinion, that the significance of trade and industry among savages is commonly underrated. Whether we are right here will appear when these points have been more closely studied than they are now.
In another paragraph we shall have to speak of a peculiar branch of trade, the trade in slaves.
§ 15. Slaves employed in warfare.
In several cases it is stated that slaves are employed in warfare. Leaving out of regard the cattle-breeding agricultural tribes, of which we have already spoken in chapter III, we find the following instances.
These are the only instances we have noticed, of purely agricultural tribes among which slaves perform a military function. There may be some more cases. Our informants do not always enter into many details; therefore, when they are silent on the subject, this does not always prove that slaves are not allowed to fight. Yet, considering the small number of instances we have been able to collect, we may safely conclude that among most agricultural tribes slaves do not share in military operations.
Yet the cause of this seeming incongruity is not difficult to detect. Pastoral tribes are always stronger, from a military point of view, than primitive agricultural tribes. In chapter III we have seen that the former often keep their agricultural neighbours in a state of subjection. Therefore it is much easier for them to employ their slaves in warfare than for hunting agriculturists; the latter, if they are to keep slaves at all, must take care to disarm them and so prevent them from doing harm.
The last passage shows that even in a state where the power of the government and the military art are highly developed it is not safe to employ slaves in warfare. They may actually be the ready tools of the aristocracy; but in the long run they will come to form a dangerous element in the state. Yet, as it may be momentarily convenient to an ambitious statesman to employ them, it will sometimes be done; whereas among hunting agriculturists the danger is so obvious that it is not even attempted.
As it is only among a few agricultural tribes, and these in the higher stages, that slaves perform a military function, we cannot think that this has been an important factor in the rise of slavery; and it has probably been nowhere the only motive for making slaves.
§ 16. Slaves kept as a luxury.
Sometimes we are told that it is considered an honour to possess many slaves. We shall give a few instances.
But, though a rich man may display his wealth by keeping a great number of slaves, we do not mean to say that among any agricultural tribe all slaves are kept as a mere luxury. This seems improbable. Among pastoral tribes, as we have seen before, it sometimes occurs. The owner of numerous herds may support a large body of slaves, who have scarcely anything to do. But among agricultural tribes it is quite otherwise. Subsistence here is largely dependent on labour; much labour is required to provide for the slaves, and the master will not choose to work for them himself. The slaves, therefore, must perform at least as much productive labour as is required to provide for their own wants; and there is no reason why the master should not make them work somewhat more, to receive a surplus; the more so, as he is thus enabled to display his wealth in the other manner mentioned by Leroy-Beaulieu, viz. by practising hospitality on a large scale.
It is, of course, perfectly possible, and indeed it often occurs, that some slaves are kept as a mere luxury, either doing nothing at all, or performing personal services. But then there must be other slaves who, by performing productive labour, provide for their master, their fellow-slaves, and themselves. There cannot be a class of unproductive labourers, without there being a class of productive labourers too. We can only imagine one case, in which all slaves might be kept as a mere luxury. It might be, that there were a class of productive labourers consisting of freemen, who provided for the slave-owners and their unproductive slaves. There might, for instance, be an aristocracy, levying taxes on the common people and keeping slaves as a luxury. Such is indeed the case among the pastoral Beni Amer. It might also occur among agricultural tribes; but we have not found any instance of it.
Only in two cases have we found it stated that slaves are not productive.
We have never found it explicitly stated that among any agricultural tribe slaves are not employed in agriculture.
Our conclusion is that luxury has not probably been among any agricultural tribe the only motive for keeping slaves. Yet many slaves are kept as a mere luxury, and consequently among some tribes slaves are far more numerous than they would be if all of them were engaged in productive pursuits.
§ 17. Other secondary internal causes.
We have seen that among the tribes of the Pacific Coast of North America the growth of slavery is furthered by their preserving food for the time of scarcity, whereas hunters who live from hand to mouth have less use for slave labour.
Hunting agriculturists much resemble true hunters: subsistence depends largely on the vicissitudes of the chase. But where a tribe lives principally on the produce of the soil, it is necessary to preserve the reaped fruits until the following harvest.
In the higher stages of culture the growth of militarism enables strong agricultural peoples to subject their weaker neighbours; and the growth of population prevents the latter from receding. But among agricultural savages subjugation of tribes is rarely found.
§ 18. External causes, especially the slave-trade. Recapitulation.
In the second and third chapters we have spoken of external causes. We have seen that for slavery to exist it is not sufficient that there should be some use for slave labour; it must also be possible to keep slaves; the freemen of the tribe must have a coercive power strong enough to keep the slaves subjected and prevent them from escaping. It has been shown that this coercive power is most strongly developed, where men have fixed habitations, live in large groups, and preserve food; and further, that the slave-trade greatly facilitates the keeping of slaves.
The slave-trade is of more interest to us here. It has been, shown that among the tribes of the Pacific Coast of North America a brisk trade in slaves was carried on, which must have greatly accelerated the growth of slavery; for it made the keeping of slaves much easier than it would have been if each of these tribes had had to procure its slaves by capture in war. It has also been remarked that among pastoral tribes slavery exists almost exclusively in those parts, where a trade in slaves with civilized or semi-civilized peoples is or was carried on, viz. in Arabia, the Caucasus, North-east and North-west Africa; whereas the pastoral tribes that live in outlying regions (Siberia, South Africa) with the only exception of the Ovaherero, do not keep slaves. We shall inquire now whether the same is the case with agricultural tribes, whether among them too slavery is of rare occurrence in those parts where the slave-trade has never been carried on.
In North America, at the time of its discovery, slavery did not exist among any agricultural tribe. Negro slavery, practised by a few of them in later times, was derived from the whites.
We see that in most parts of the world inhabited by slave-keeping agricultural savages, slaves are, or were, purchased or captured by civilized or semi-civilized peoples; whereas in those parts where the slave-trade has always been unknown, slavery has never prevailed to any considerable extent.
We shall try to account for this fact. But we must first reply to a question which the reader may perhaps have asked, viz. why we have not at the beginning of this second Part discussed the question as to whether slavery is ever a phenomenon of genuine savage life, or has always been due to intercourse with higher races. We have not done so, because it seems to us that this question is one of secondary interest. When we see that among a savage tribe slavery is an institution playing a great part in native life and slave labour is of much use, we must come to the conclusion that slavery is perfectly consistent with the economic and social state in which this tribe lives, whether intercourse with superior races gave rise to it or it already existed before any such intercourse had taken place. And the conclusions we have arrived at in this and the foregoing chapters, as to the various circumstances which further or prevent slavery, remain the same in both cases. Moreover, there are some savage tribes which at the time of their discovery kept slaves without, so far as we know, having had any intercourse with superior races: the New Zealanders and the fishing tribes of Kamchatka and the North Pacific Coast of North America.
Probably, however, the slave-trade has also had considerable influence. The mere fact that foreigners purchase or capture slaves from savage tribes cannot, of course, account for the existence of slavery among these savages; it may even result in the disappearance of slavery among them, if they find it more profitable to sell their slaves than retain them. But the slave-trade will often augment the opportunities of acquiring slaves. Slave-traders generally trade in other articles besides slaves; therefore they will often exchange some of the slaves they may have procured for something else. If some slaves are too weak to perform the journey to the coast, it is more profitable to sell them in the interior than to leave them behind. Moreover, the slave-traders often induce savages to make raids upon their neighbours and sell the captives they may secure; and if a savage chief has once learned in this way to kidnap slaves, he will perhaps continue to do so though there be not always a ready sale for them. In a word, the procuring of slaves becomes much easier.
Now we must bear in mind that an institution does not always exist in all countries where it would be economically useful. Not only is an institution which would further the public welfare often wanting, because the immediate interests of individuals are not concerned, but the individuals are not always aware even of their own immediate interests. Such may also be the case with slavery. It may be that, if slavery were introduced into some savage tribe, the tribe, or at least some members of the tribe, would profit by it, and that yet no slaves are made because there are not sufficiently strong motives for doing so, or because there are stronger motives working in the opposite direction. Tribes which have never kept slaves and are unacquainted with slave-keeping tribes, do not know the use of slavery; therefore they have to come to it gradually. Taking for granted that the first source of slavery, as seems very likely, was captivity in war, and that the captives, at first adopted, were afterwards differentiated from the main body of citizens, we can understand why slavery does not exist in all countries where the keeping of slaves would be profitable. It may be that it is the custom to eat the prisoners, or to sacrifice them, or to restore them when peace is concluded. It may also be that the adopted prisoners have from times immemorial been regarded as the equals of the tribesmen. In all these cases the new motive, the want of slaves, must be strong if it is to break through the established custom. The vis inertiae plays a great part in the history of mankind.
Returning to the slave-trade, we may remark that this entirely overturns the existing state of things. It enables the tribe to procure slaves who are not captives, and regarding whom therefore no custom has been established. Moreover, the tribe becomes acquainted with the institution of slavery, which it had not seen practised before. Now there is no longer any reason why the members of such a tribe should not purchase the slaves offered to them and set them to work. An external cause has sufficed to disturb the former equilibrium.
Our conclusion is that slavery existed among some savages who had never had any intercourse with superior races, but that this intercourse, especially where the slave-trade was carried on, has often greatly furthered the growth of slavery.
In our chapter on pastoral tribes we have remarked that a tribe living in the vicinity of inferior races is more likely to keep slaves than one surrounded by tribes of the same or a higher level of culture. Enslavement of lower races also frequently occurs among agriculturists; but the agriculturists of the higher races (e.g. Malays in the restricted sense, inhabitants of North Africa, etc.) are to be called barbarians rather than savages and have therefore been excluded from the survey of slavery given in our first Part.
Briefly recapitulating the conclusions we have arrived at with regard to agricultural tribes, we have to remark that the general character of their economic and social life is favourable to the existence of slavery: subsistence is easy to procure, and independent of capital, except where cattle are kept.
This applies especially to true agriculturists (agriculturists of the higher stages) as opposed to hunting agriculturists. The latter bear much resemblance to true hunters, who hardly ever keep slaves, accordingly slavery is not so frequent among them as among true agriculturists.
Our general rule, however, requires an important qualification. Where all land has been appropriated, a class of free labourers commonly exists, and slaves are not wanted.
We have not entered into a thorough investigation of the influence of secondary or additional causes. A closer study of this matter will perhaps yield important results. What we have found with regard to these secondary causes is the following.
Slaves often perform female labour; on the other hand female labour sometimes serves as a substitute for slave labour.
Commercial agricultural tribes are far more likely to keep slaves than agricultural tribes among which commerce holds a very subordinate place.
Slaves are sometimes kept for military purposes, or as a mere luxury. These two circumstances, however, though they often lead to the keeping of a larger number of slaves than would otherwise be required, have not probably ever given rise to slavery.
Subjection of tribes as such, which among some pastoral tribes serves as a substitute for slavery, hardly ever occurs among agricultural tribes, except in Oceania, where slavery is already absent for the general cause mentioned above, the non-existence of free land.
Some external causes we had found before: fixed habitations, living in large groups, preserving of food, and the neighbourhood of inferior races, call for no special notice here.
Intercourse with superior races, especially where these carry on the slave-trade, proves to be a factor greatly furthering the growth of slavery.
We shall not give here a list of the causes found up to the present. We think it more convenient to place such a list at the end of our general survey.