d. Finally, the general trustworthiness of the writer or writers must

be taken into account.

2º. If we are told that there are no slaves, it is very probable that slavery really does not exist, for slavery is a phenomenon rather easy to observe, and the ethnographers are generally inclined to use the word in a too wide rather than in a too restricted sense. There is no need here for the extreme scepticism with which we have to receive an assertion of any tribe having no religion [67]. This does not prevent, of course, that if we find the existence of slavery denied by a writer who is generally badly informed, we may reject such an assertion.

3º. The greatest difficulties are presented by those cases in which no mention is made of slavery. Here the utmost caution has to be observed.

a. If it clearly appears from the description, that there are people considered to be the property of others, without the word “slavery” being used, the conclusion is evident.

b. Perhaps some facts are mentioned which make the existence of slavery highly improbable. We shall see that the main source of slavery is captivity in war; so, if it is stated that no captives are made, or that the lot of the captives is something else than slavery, the non-existence of slavery is probable. But even then it is not quite certain: there may still be slaves, acquired by other means. Further: if it is stated that there are no social classes, or if the classes are enumerated and slaves are not among them, there is a strong presumption that slavery does not exist. Yet here too there is no certainty; for slaves, among savage tribes, have not always the aspect of a social class. The description of the division of labour between the sexes may also suggest to us the non-existence of slavery. When we are informed that the men do some kinds of work and the women some other kinds, we are inclined to think: if there were slaves, their special work would be mentioned too. But this argument is most dangerous; for the slaves very often have no special kind of work allotted to them.

We see that none of these criteria prove quite reliable. Yet, taken together, they give a high degree of probability. And it is not even necessary that all of them can be ascertained. If an ethnographer, known to be trustworthy, gives us an elaborate description, pretending to be a picture of the whole social life of the tribe he describes, it were a wonder if he had entirely omitted slavery, while it existed; the more so as this phenomenon is not so difficult to recognize. The same argument obtains a fortiori, when several such descriptions of the same tribe exist.

4º. In doubtful cases we may take into account the state of the group to which a particular tribe belongs. It may be that in the general descriptions of a group no mention is made of slavery; that, further, all tribes belonging to this group of which we are well informed prove to have no slaves. If, then, the information we get concerning a particular tribe belonging to the same group, is not complete enough to rely upon, there is a strong presumption that this tribe will be in the same state as the rest of the group, i.e. that it does not keep slaves. Under the same conditions we may suppose that a tribe belonging to a slave-keeping group keeps slaves. The term “group” has to be taken here in a sociological, not in an anthropological or linguistic sense, and its application must be somewhat restricted. It must consist of tribes, that live in somewhat similar conditions and the institutions of which closely resemble each other; e.g. Australia (the continent) is a group in this sense, North America is not. It is almost superfluous to add, that this group-argument may be used only to strengthen existing, but insufficient, evidence.

We confine ourselves here to the phenomena of savage life; therefore we shall exclude the semi-civilized peoples. An exact distinction, however, between these two classes of peoples has not yet been made [68]; so we are fully aware of the possibility of mistakes, made here in this respect. As we were not able to apply any exact criterion, we have more or less followed our general impressions, paying most attention to the development of political institutions. So for instance we have excluded the Kabyls of Northern Africa, because their detailed legal system, as described by Hanoteau and Letourneux [69], proves that politically they have passed beyond the stage of savagery. And a developed political organization cannot exist without profoundly marking such an institution as slavery.

We shall find that several savage tribes have to a considerable extent been influenced by civilized or semi-civilized nations. In such cases the question arises: have we to deal here with phenomena of unadulterated savage life? This question is important and deserves full attention. We shall see that many savage tribes in their true aboriginal state have been acquainted with slavery, whereas many others when first discovered did not practise it; so neither the existence nor the non-existence of slavery must necessarily be due to foreign interference; either may be aboriginal, and must be supposed to be so wherever there is not a strong presumption to the contrary [70]. But there are also many cases in which foreign influence has undoubtedly been at work. We must, then, make a distinction. If the influence of the civilized or semi-civilized nation has led to a, so to speak, normal development, i.e. a development, which lies within the limits of primitive culture, we have to deal with a phenomenon of savage life. For instance, commercial intercourse with a nation of higher civilization has brought about an accumulation of wealth and a social differentiation, which render the existence of slavery possible. In such case we may safely speak of slavery as practised by savages, for it is the effect of trade in general, not of intercourse with a civilized nation as such, that we have to deal with here. If, on the contrary, there has been an abnormal development, i.e. a development which does not take place in countries inhabited by savages only, the present condition of the savage tribe has no interest to us. For instance, measures have been taken by a civilized nation on purpose to abolish slavery. Then we, for our purpose, must consider the savage tribe concerned as keeping slaves, and pay attention only to the descriptions of its institutions as they were before the abolition.

The ensuing paragraphs will show which savage tribes keep slaves and which do not keep any. The groups into which we have divided the several tribes are mainly geographical, not intended to answer any anthropological purpose. As long as the meaning of the term “race” is so unstable as at present, we think it better not to operate with it. Our groups are nearly the same as those given by Schurtz in his Völkerkunde.

At the end of each paragraph its result will be mentioned. The “positive cases” are the tribes which probably keep slaves, the “negative cases” those which probably do not keep slaves. Under the head of “no conclusion” we have given the cases in which the probability that slavery exists is nearly as great, or as little, as the inverse probability. The tribes, the names of which are printed in italics, are those which afford “clear cases”, i.e. where the probability nearly amounts to certainty.

At the end of the chapter a brief recapitulation will show the occurrence of slavery among savages in the several parts of the globe.

When, in the following paragraphs, we say: “Such a tribe keeps slaves,” or: “Such a tribe does not keep slaves,” this does not imply that the same state of things still prevails. The tribe we speak of may have died out; or, where slavery existed, it may have been abolished. When we know that such a thing has taken place, we shall use the past tense. But in many cases the only information we have got concerning some tribe dates from many years back; and we do not know what has become of this tribe in the meantime. Then, not to be obliged always to use such tedious formulas as: “In Mr. X.’s time slavery existed among such a tribe,” we shall simply say: “slavery exists.” Whether it still exists is certainly of much interest to a philanthropist; but to us, for the purposes we have set ourselves in the present volume, it is quite immaterial.

§ 2. North America.

1. Eskimos.

Rink’s account applies especially to Western Greenland at the time of the first European settlement, but may be taken as a general description of the Eskimos [71]. According to him the family in the restricted sense comprehended foster-children, widows, helpless persons adopted as relatives and more or less treated as servants. They were regarded as subordinate members of the family but never subjected to any corporal punishment [72]. He then describes their social organization, but makes no mention of slavery [73]. The question remains whether these helpless persons are to be called slaves. This does not very clearly appear; but, happily, we have more detailed accounts of the several Eskimo tribes.

Crantz, in his description of the Greenlanders, gives many particulars about their servants. Mothers of illegitimate children are despised; sometimes a childless person buys her children. When a married couple have no children or no full-grown children, the husband adopts one or two orphan boys, who help him in his work and must provide the family with the necessaries of life. The same is done by the wife with daughters of others or with a widow. Although the adopted youths are employed as servants, they are free from any compulsion, and are regarded as the future lords of the house. The adopted maid-servant or daughter can leave when she likes. A man will never beat his man-servant, and were he to touch a maid-servant, he would incur great disgrace [74]. We see that the condition of these servants is not slavery. The boy is the future lord of the house, the girl may leave when she pleases. Servants are only required to occupy the place that in a normal household is taken up by the children. Labour is not asked for. “If a man dies without leaving behind relatives, or full-grown sons, nobody cares for the children, unless one wants a maid-servant.” A widow must try to get a lodging, in which she does not always succeed [75]. It is clear that to these Greenlanders slaves would not be of any use. Nansen, too, makes no mention of slavery [76].

Boas, describing the Central Eskimos, states that among them too children are adopted and regarded by the adoptive parents as their own children; so “an elder adopted son has a preference over a younger son born of the marriage,” viz. as to the right of inheritance. The following statement still more directly bears on our subject: “Sometimes men are adopted who may almost be considered servants. Particularly bachelors without any relations, cripples who are not able to provide for themselves, or men who have lost their sledges and dogs are found in this position. They fulfil minor occupations, mend the hunting-implements, fit out the sledges, feed the dogs, etc.; sometimes however, they join the hunters. They follow the master of the house when he removes from one place to another, make journeys in order to do his commissions, and so on. The position, however, is a voluntary one, and therefore these men are not less esteemed than the self-dependent providers” [77]. The last sentence clearly shows that these servants are not slaves. And as in no other place does Boas make any mention of slaves, it is certain that slavery does not exist.

Ribbach gives some notes on the Eskimos of Labrador. There is nothing on slavery in these notes. Describing their dwellings he says: “The principal family has of course the best place; the servants, widows or orphans, if there are such, have to content themselves with a place near the door, where the cold is most severe” [78]. This agrees so much with the foregoing descriptions, that we may suppose that the same state of things prevails here.

As little mention is made of slavery in the descriptions of some other Central Eskimo tribes, as the Frobisher Bay and Field Bay Eskimos [79], the Kinipetu Eskimos [80], the Tchiglit Eskimos [81], the Eskimos of the Ungava district [82].

Bancroft says of the Western Eskimos (or Eskimos of Alaska): “Slavery in any form is unknown among them”. Elliott makes no mention of slavery [83].

The describers of the Eskimos of Point Barrow, too, have not a word about slavery. Adoption is practised to a great extent [84]. So the same state of things probably prevails here as among the other Eskimos.

Some other tribes, belonging to the Eskimo group in the wider sense [85], may also be treated here.

Amongst the Aleuts, according to Bancroft, the chief “is exempt from work, is allowed a servant to row his boat, but in other respects possesses no power” [86]. No more is added bearing on our subject. Petroff, however, gives a detailed account of slavery among them. The slaves were prisoners of war and their descendants. The master could punish the slave with death, could sell and liberate him. Any attempt to escape was severely punished. [87]. So the Aleuts had slaves.

The Athka Aleuts, according to Petroff, had also slaves [88].

Among the Koniagas or Southern Eskimos “slavery” says Bancroft “existed to a limited extent.” This is affirmed by Holmberg [89].

2. Nootka group.

The Tlinkits formerly carried on slavery to a great extent. This is proved by the detailed accounts of several writers [90].

The same applies to the Haidas [91].

Krause, in a short note, speaks of a female slave of a Tsimshian chief. Niblack states that the Tsimshians acted as middlemen in the slave trade. Boas describes the legends of the Tsimshians; in these legends slaves and their occupations are frequently spoken of. Kane, in his census of Indian tribes states that among the Tsimshians there were slaves. [92]. Hence we may infer that slavery formerly existed among them.

Boas, speaking of the Kwakiutl Indians, writes: “All the tribes of the Pacific Coast are divided into a nobility, common people and slaves. The last of these may be left out of consideration, as they do not form part and parcel of the clan, but are captives made in war, or purchases, and may change ownership as any other piece of property” [93]. From this statement it appears that the Kwakiutl kept slaves.

Krause says of the Bilballas; “The chief possessed numerous wives and many slaves; also were these Indians much given to slave-stealing and the slave trade” [94]. As the Bilballas are reckoned by Bancroft among the Haidas, amongst whom, according to him, “slavery is universal”, we may suppose, that slaves were kept by them for their own use, not for export only.

The many details, given in the works of Sproat and Brown, prove that slavery existed among the Ahts of Vancouver Island [95].

The Nootkas, among whom, according to Bancroft, “slavery is practised by all the tribes”, seem to be the same as the Ahts [96].

Bancroft informs us that “slaves are held by all the tribes” about Puget Sound, and gives several details of their slave system [97].

Slavery also existed among the Fish Indians of British Columbia [98].

3. Tinneh group.

According to Kane, who was well acquainted with this group, “slavery in its most cruel form exists among the Indians of the whole coast, from California to Behring’s Straits, the stronger tribes making slaves of all the others they can conquer. In the interior, where there is but little warfare, slavery does not exist.” Niblack, however, states that slavery existed among the interior Tinneh but they “had no hereditary slaves, getting their supply from the coast” [99].

Of the Kutchins or Loucheux Jones says: “Slavery is practised among them. Any poor creature who has no friends is made a slave” [100]. Hardisty gives more details; he tells us: “As a rule slavery does not exist, but the orphan and the friendless are kept in servitude and treated so harshly as to be really little better than slaves, until such time as they get big enough and bold enough to assert their independence, when they are allowed to shift for themselves” [101]. Such ill-treated children, who when full-grown are “allowed to shift for themselves”, certainly are not slaves. Therefore we may safely infer, that slavery did not exist here, the more so, as the other authors [102] make no mention of slaves.

Mackenzie, describing the Chepewyans or Athabascas, states that “they are constantly at war with the Eskimos, and kill as many of them as they can, as it is a principle with them never to make prisoners” [103]. Neither in his notes on the Chepewyans in general, nor in those on some single tribes belonging to the Chepewyan family, as the Slave and Dog-Rib Indians, Hare Indians, Beaver and Rocky-Mountain Indians, does our informant make any mention of slavery. Nor is there a word to be found about slaves in Russell’s and Bancroft’s accounts. Hearne speaks of Northern and Southern Indians, divisions of the Chepewyans. Among the Southern Indians a wife sometimes begs of her husband, who is going to war, to bring a female slave with him for her to kill. The chief Matonabbee was the son of a Northern Indian man and a Southern Indian slave [104]. Hearne does not speak of male slaves. So we may suppose that slavery proper does not exist.

On the Tacullies Bancroft remarks: “Slavery is common with them, all who can afford it keeping slaves. They use them as beasts of burden, and treat them most inhumanely” [105].

Of the Atnas on Copper River, a division of the Kenai, Bancroft says: “Those who can afford it, keep slaves, buying them from the Koltschanes” [106].

Mrs. Allison informs us that among the Similkameem Indians of British Columbia “slaves taken in war were well treated, but always had one eye blemished to mark them” [107].

4. Algonquin group.

The authors we have consulted on the Algonquins in general [108] make no mention whatever of slaves.

Loskiel, describing the Lenape or Delawares, states that captured boys and girls were received into their families, and employed as servants; sometimes, however, they were sold to Europeans. If such prisoners behaved well, they had nothing to complain of and were not overworked. If they ran away and were recaptured, they were generally killed. But the adult male prisoners, viz. those of them who were not killed, were adopted by families, instead of those who had been killed in war or had died in some other way, and from this moment were looked upon as members of the tribe to which they now belonged [109]. As these men became members of the tribe, it is not probable that the captive children were made slaves; we may safely suppose that as long as they were young they had to perform menial work, but when adult were on a level with the members of the tribe. And as neither Loskiel, in any other passage, nor Brinton refers to slavery, slaves were very probably not to be found among the Delawares. [110].

In Le Jeune’s account of the Montagnais no mention is made of slaves. Prisoners of war were cruelly put to death [111].

The Ojibways or Chippeways, according to Keating, killed the captive warriors and old women; the marriageable women became slaves and were very cruelly treated by the women of the victorious tribe; the children were adopted and treated fairly well [112]. Jones’s account is somewhat different. Most often all enemies were killed. Sometimes they made a few prisoners, who were adopted by those who had lost a relative; then the adopted prisoner became a relative or slave; if not adopted he was burned alive. The relatives of a murderer sometimes paid large indemnities to those of the victim; the murderer had then to work for them in order to pay off the debt; he was reduced to a kind of servitude [113]. In these accounts slaves and servitude are mentioned. The servitude of the murderer very probably was not slavery. He had to work: but it is not stated that he was made a slave, i.e. the property of an individual person. The prisoners who became “relatives or slaves” were adopted; therefore they were not slaves in the proper sense of the word. And as for the female slaves Keating speaks of, we know that a slave system without male slaves is not slavery proper. We may suppose, that these female captives became an inferior kind of wives, to whom the women of the tribe were unkind through a very natural jealousy. Kohl, in his elaborate description of the Ojibways, makes no mention of slavery. Their wars, he states, did not bring them any profit [114]. According to Carver, “all that are captivated by both parties, are either put to death, adopted, or made slaves of.” “That part of the prisoners, which are considered as slaves, are generally distributed among the chiefs, who frequently make presents of some of them to the European governors of the out-posts, or to the superintendents or commissaries of Indian affairs. I have been informed that it was the Jesuits and French missionaries that first occasioned the introduction of these unhappy captives into the settlements and who by so doing taught the Indians that they were valuable” [115]. From all the foregoing we may infer that slavery was not an indigenous institution among the Ojibways.

This inference is strengthened by what Tanner tells us of the Ottawas, an Ojibway tribe. He was adopted by an Ottawa woman, but was not at first on a level with the other children. The first few years she made him do various kinds of manual labour: he had to cut wood, fetch water and do other kinds of work, which were not generally required from children of his age. Yet when grown-up he was on a level with the Indians into whose tribe he was admitted, and married an Indian girl [116].

Before passing to the Ottawas, Tanner had been a captive amongst the Shahnees. He was very cruelly and ignominiously treated. Yet he was not a slave, for he had been adopted by a married couple on the grave of their youngest son, whose place he was to fill [117]. As this agrees with the general customs of this group, in which there are no slaves, we may suppose that here also slavery was unknown.

The Potawatomi also very probably had no slaves; for none of their describers make any mention of slavery [118].

Amongst the Crees or Knisteneaux, according to McLean, adoption of aliens was practised. A missionary, who had unintentionally killed a Cree boy, offered himself in his stead, and was adopted. Kane speaks of “the universal custom of Indian mothers eagerly seeking another child, although it may be of an enemy, to replace one of her own, whom she may have lost.... This child is always treated with as great, if not greater, kindness than the rest.” The existence of this custom, together with the fact that none of their describers makes mention of slaves, renders it most probable that they did not keep slaves [119].

The Cheyennes very seldom captured adult males; when they did, they generally put them to death. Children were adopted and treated like their own children; women became the wives of their captors [120]. Slavery is not mentioned.

The Blackfeet nation consisted of four tribes: Piegans, Blackfeet, Bloods, and Gros Ventres. We are told that once when at war against the Crows, the Gros Ventres “rushed upon them and killed the whole number” [121]. Grinnell, speaking of a Piegan chief, says: “He told his men not to kill the captured women. They also captured ... many children. The chief selected a wife for himself from among these women.” As a rule they spare none of their enemies, killing alike men, women and children. Sometimes they spare a captive for his bravery or from dread of sorcery; he is then provided with food and dismissed to his home [122]. These particulars being given, and no mention made of slavery by any of our informants [123], we may safely infer that slavery did not exist among these tribes.

Among the Abenakies, according to Maurault, prisoners of war were either tortured to death or adopted into the tribe. Hence we may infer that slavery, of which this writer makes no mention, was unknown among them [124].

Hoffman, in his description of the Menomini Indians, referring to Grignon, says that he does not know whether they had captive slaves; but certainly they had purchased slaves. Our informant saw 6 male and 8 female slaves, most of whom had been enslaved when young. The female slaves had been sold for 100 dollars each. The slaves were called Pawnees, though some of them belonged to other tribes [125]. This statement sufficiently proves, that in the time of this description the Menomini had slaves. But in Hoffman’s time they were already very much under the influence of European civilization. Whether at the more remote period from which most of our information on the Algonquin tribes dates slavery existed among them, we do not know.

5. Iroquois group.

The Iroquois had no slaves. This is stated by Morgan, who was intimately acquainted with them. “Slavery”, says Morgan, “which in the Upper Status of barbarism became the fate of the captive, was unknown among tribes in the Lower Status in the aboriginal period.” And the Iroquois are his typical instance of this “Lower Status”: “When discovered the Iroquois were in the Lower Status of barbarism.” Captives were either put to death or adopted [126]. Charlevoix states, that “most of their captives are condemned to death, or to a state of abject slavery in which they were never certain of their lives” [127]. But he gives no more particulars about this slave state, nor do our other informants [128]. On the contrary, Lafitau informs us, that the condition of prisoners, whose life is rather hard amongst the Algonquin tribes, amongst the Iroquois and Hurons is very easy [129]. The descriptions given by the authors of the fate of captives justify Morgan’s statement: they were either killed or adopted [130]; and though Lafitau calls the prisoners “esclaves”, their state, as he describes it, is not at all like that of slaves. So we may safely infer, that slavery did not exist among them, and that Charlevoix’s above quoted statement is erroneous.

Among the Hurons or Wyandots, according to Powell, the captives were either killed or adopted [131]. Lafitau’s and Charlevoix’s accounts of the fate of captives among the Iroquois apply also to the Hurons. So it is probable that they had no slaves.

6. Choctaw-Muskoghe group and neighbouring tribes.

Adair, speaking of the Katahba, Cherokee, Muskoghe, Choctaw and Chickasaw Indians, states that they burned their prisoners. Only if a prisoner succeeded in escaping to the house of the high-priest or some other place of refuge, he was not burned; but what his fate was in such a case we are not told. Young prisoners were not killed; but it is not stated what became of them. If warriors had offended a neighbouring tribe, and the chiefs wished to prevent war, they sacrificed either one of the offenders belonging to a weak family or some unfortunate prisoner, who had been incorporated into a declining tribe [132]. The last sentence seems to show, that the custom of adopting prisoners prevailed here too. At any rate, no mention is made of slaves.

Rochefort remarks that among the Apalaches (who, according to Roosevelt, included the Cherokees, Chickasaws, Choctaws, Creeks and Seminoles [133]) an enemy who surrendered during the fight, was taken to the conqueror’s home with his wife and children, held in an honourable freedom and treated with as much leniency and care as their own servants [134]. Whether such persons were slaves does not clearly appear from this statement. But Adair’s record tends to prove that slavery did not exist, at least as far as the three former divisions of the Apalaches are concerned.

Loskiel relates that a prisoner was once condemned to death by the Cherokees. He had already been tied to the stake, when a Cherokee woman arrived. She brought a basketful of commodities, which she deposed at the feet of the man to whom the prisoner belonged, and bade him leave this prisoner to her, a childless widow, who wanted to adopt him as a son. This was done [135].

Bartram tells us that the Creeks formerly tortured their captives to death [136].

The Seminoles, too, according to Roosevelt, used to burn their prisoners [137].

From all the foregoing we may safely infer that slavery did not exist in the Apalache group.

Natchez warriors delivered their captives to the relations of those who had fallen in battle. The captives were always burned [138].

Bossu speaks of slaves among the Attakapas; but it does not appear from his notes whether they made slaves for their own use or for sale abroad [139].

Strachey describes the inhabitants of Virginia (several tribes). He makes no mention of slaves. It is stated in his account, that children and foreigners were sacrificed [140]; if there had been slaves, these probably would in the first place have been the victims. One of the objects of their wars was to capture women and children. Before the commencement of the battle it was announced that the conquered “upon their submission or comyng in, though two daies after, should live, but their wives and childrene should be prize for the conquerors” [141]. Another ancient writer gives a different description of the fate of their conquered enemies: “when they gain a victory, they spare neither men, nor women, nor children, in order to render revenge impossible” [142]. We cannot arrive at a definite conclusion here.

7. Dacotah group.

Owen Dorsey, describing the Dacotahs or Sioux, says: “There are no slaves among the Siouan tribes” [143]. This assertion is strengthened by the other authors [144] making no mention of slaves. Mrs. E. G. Eastman tells us, that captive women and children were well treated and restored on the conclusion of peace; but often they preferred to remain with their new husbands and adopted parents. Copway and Neill also speak of captive children being adopted [145].

Mathews states, that the Hidatsas generally adopt the children captured in war, and treat them like their own. When grown-up they sometimes return to their own tribe, but most often remain where they are [146].

Owen Dorsey informs us that among the Omahas “Slavery was not known”. “Captives were not slain by the Omahas and Ponkas. When peace was declared, the captives were sent home, if they wished to go. If not they could remain where they were, and were treated as if they were members of the tribe; but they were not adopted by any one” [147]. This positive statement is not weakened by James’s assertion about captive women becoming slaves [148].

Hunter states that among the Osage and Kansas Indians prisoners were adopted into the conquering tribe, as husbands, wives and children [149].

Of the Assiniboins we are told: “Chiefs never receive a gift, considering it a degradation to accept anything but what their own prowess or superior qualities of manhood acquire for them. Their hearts are so good and strong that they scorn to take anything, and self-denial and the power to resist temptation to luxury or easily acquired property is a boast with them” [150]. Where even the chiefs rely only on their own prowess, the existence of slavery is improbable.

Lewis and Clark, describing the Mandans, speak of prisoners living among them. One of their chiefs had been taken prisoner and adopted by them, “and he now enjoys great consideration among the tribe.” In another place they tell us of a woman, who was sold as a slave to a Mandan chief, who brought her up and afterwards married her [151]. The evidence is not sufficient to decide, whether their prisoners were held as slaves or were adopted into the tribe.

8. Oregon group.

Gibbs describes the tribes of Western Washington and North-Western Oregon in general. The principal of these tribes are the Chinooks and the tribes about Puget Sound. “Slavery,” says Gibbs, “is thoroughly interwoven with the social policy of the Indians of the coast section of Oregon and Washington Territory. East of the Cascades, though it exists, it is not so common.... Southward it ceases, so far as my observation has gone, with the Siskiou Mountains, which divide Oregon from California” [152].

“By the Flatheads,” says Bancroft, “captives are generally killed by their sufferings.” McLean, speaking of their wars, remarks: “When one party lost more than the other, compensation was made in slaves or some other kind of property” [153]. This statement does not, however, prove that slavery existed among them; these “slaves” might be members of the tribe, delivered up either to be killed or adopted. So we are left in doubt as to the existence of slavery [154].

The Chinooks had slaves. Bancroft says: “Slavery, common to all the coast families, is also practised by the Chinooks; ... the slaves are obliged to perform all the drudgery for their masters, and their children must remain in their parents’ condition, their round heads serving as a distinguishing mark from freemen”. Kane also gives many particulars about their slave system [155]. Equally Swan, describing the Chinooks and neighbouring tribes, makes mention of slavery as practised by them [156]. Lewis and Clark speak of a war, in which the Killamucks took several prisoners. “These, as far as we could perceive, were treated very well, and though nominally slaves, yet were adopted into the families of their masters, and the young ones placed on the same footing with the children of the purchaser”. This short note is not sufficient for us to arrive at any definite conclusion, the less so, as these writers themselves declare that they had not the opportunity of making a close study of the tribes of the Pacific Coast [157].

“The Shushwaps,” Bancroft remarks, “are said to have no slaves” [158]. Among the Okanagans, a division of the Shushwaps, according to Ross, “there are but few slaves ... and these few are adopted as children, and treated in all respects as members of the family” [159]. From this it would seem that slavery proper does not exist.

Another division of the Shushwaps are the Atnahs on Fraser River (not to be confused with the Atnas on Copper River). Mackenzie describes a division of Indians, whom he does not mention under a separate name; but they seem to be akin to the Atnahs. “The Atnah and Chin tribe,” says Mackenzie, “as far as I can judge from the very little I saw of that people, bear the nearest resemblance to them.” On these Indians he remarks: “The strangers who live among these people are kept by them in a state of awe and subjection” [160]. These strangers perhaps are slaves; but the lack of further details prevents our arriving at any positive conclusion.

Bancroft, after describing the manner in which some tribes put their prisoners to death, adds: “Among the Sahaptins some survive and are made slaves.... The Nez Percé system is a little less cruel in order to save the life for future slavery” [161]. So the Sahaptins or Nez Percés seem to have kept slaves, though we should wish for some more particulars that would exclude all doubt.

Powers states, that female slaves are more numerous among the Shastika than among the Californians [162]. This short note is the only evidence we have been able to collect on the subject.

Kane makes mention of slavery as practised by some other tribes, about which we could not collect further information (perhaps they are subdivisions of the tribes already enumerated) viz. the Macaws, Babines or Big-lips, Nasquallies and Kye-uses [163].

9. Californians.

Of the Northern Californians Bancroft tells us: “Although I find no description of an actual system of slavery existing among them, yet there is no doubt that they have slaves. Illegitimate children are the life-slaves of some male relative of the mother, and upon them the drudgery falls; they are only allowed to marry one in their own station, and their sole hope of emancipation lies in a slow accumulation of allicochick (shell-money), with which they can buy their freedom” [164].

Powers gives some more particulars about two North Californian tribes. Among the Karoks it is thought ignominious for a man to have connection with a female slave. When the purchase-money for the wife has not been paid, the children are looked upon as bastards; they live as outcasts and marry none but persons of their own condition. Among the Hupas a similar system prevails. A bastard is much despised; when old enough he is taken from his mother and becomes the property of one of her male relatives; he is not a slave, and yet has no share in the privileges of the family. The produce of his labour belongs to his master; he may marry only a person of his own condition, and is treated with ignominy. What he wins by gaming is his own; when this amounts to 15 or 20 dollars, he is free. His children are of the same rank [165].

Although these bastards present a close resemblance to slaves, Powers explicitly says that they are not slaves. Probably they are only a despised class; for social status, among these tribes, depends largely upon wealth. The chief “obtains his position from his wealth, and usually manages to transmit his effects, and with them his honours, to his posterity” [166]. “The ruling passion of the savage seems to be love of wealth; having it he is respected, without it he is despised” [167]. We may therefore suppose, that these bastards are despised because penniless, and as soon as they possess 15 or 20 dollars, respected for their wealth. And as we “find no description of an actual system of slavery existing among them,” slavery probably does not exist [168].

Gatschet, describing the Klamath Indians, makes mention of slaves. Once they attacked the Pit River Indians, “killed the men, abducted the women and children to their homes, or sold them into slavery at the international bartering place at The Dalles.” According to Judge E. Steele “they had been selling to whites and others Indian children of their own and other tribes, and also squaws, the latter mainly for the purpose of prostitution” [169]. Whether all slaves were sold abroad, or any slaves were kept by them, does not appear. According to Bancroft “Mr. Drew asserts that the Klamath children of slave parents, who, it may be, prevent the profitable prostitution or sale of the mother, are killed without compunction” [170]. Altogether the notes given by our informants are insufficient for us to decide, whether slavery really exists here.

On the Central Californians Bancroft remarks: “Slavery in any form is rare, and hereditary bondage unknown.” “They do not appear to have kept or sold prisoners as slaves, but to have either exchanged or killed them” [171]. Here “rare” is perhaps a synonym for “absent”; at any rate the existence of slavery here is doubtful.

As for the Southern Californians, according to Bancroft, “Hugo Reid affirms of the natives of Los Angeles County that all prisoners, after being tormented in the most cruel manner, are invariably put to death.... Female prisoners are either sold or retained as slaves” [172]. From Boscana’s narrative also it would appear, that there were formerly slaves among them. “No quarter” he says “was ever given, and consequently, no prisoners were ever made among the men, excepting of such as were killed, or mortally wounded. These were immediately decapitated.... The women and children taken prisoners, were either disposed of by sale or detained by the captain as slaves.... The women and children were never released,—ever remaining as slaves to their enemies” [173]. But as no more details are added and as Boscana describes a state of things which in his time had already ceased to exist, we are not quite sure whether slaves were really kept by these Indians.

The Nishinam, according to Powers, killed their male prisoners. Women, after being flogged, were married; but sometimes they were also killed [174]. So it seems that they had no slaves.

10. New Mexicans.

On the Shoshones and Utahs we are not very well informed. “An act which passed the legislature of Utah in 1852 ... set forth that from time immemorial slavery has been a customary traffic among the Indians.” But we are not told who bought and who sold the slaves. It is only stated that the Utahs sold their wives and children into slavery to the Navajos [175]. It is not probable that the Shoshones and Utahs themselves had slaves; for Bancroft states that prisoners of war were killed, or in some cases dismissed unhurt, and Meline tells us, that the Utes and Pueblos almost invariably sold their prisoners to the Mexicans for slaves [176].

Bancroft, describing the Apache family (including Apaches, Comanches, Navajos, Mojaves, etc.), says: “All the natives of this family hold captives as slaves” [177]. But his account of the Comanches does not quite agree with this general statement: “Prisoners belong to the captors and the males are usually killed, but women are reserved and become the wives or servants of their owners, while children of both sexes are adopted into the tribe” [178]. According to Schoolcraft, “prisoners of war belong to the captors and may be sold or released at their will”. Captive children are adopted and afterwards are on a level with the members of the tribe [179]. Gregg speaks of prisoners being enslaved and ill-treated by the conquerors. But whether he means to say that they remained slaves is not clear [180]. Cessac also speaks of slaves. “If, among the captives of the deceased, one was particularly loved by his master, he is sacrificed and buried with him.” “If a favourite slave is sacrificed, it is to give the master a fellow-traveller.” No more particulars about these slaves are given. In their wars against the Mexicans they spare none but the children, whom they treat as their own. “These captives forget their origin and later on, when full-grown, become an integral part of the tribe” [181]. It is not clear whether the slaves Cessac speaks of are identical with these adopted children; he would not have used then the term “slave” in its proper sense. Ten Kate, a careful observer, states that a number of Mexican captives, altogether about fifty, live among the Comanches and Kioways; they have almost entirely adopted the manners and customs of the Indians and are regarded by the latter as members of their tribes [182]. Another author tells us of a Mexican boy and girl, taken prisoners by the Comanches. The boy was afterwards sold to the Cherokees, the girl was married against her will by a Comanche. Another captive Mexican woman was married to a Comanche chief [183]. Comparing these several statements, we think it probable, though not quite certain, that the Comanches did not keep slaves, but adopted their prisoners.

Ten Kate’s above-quoted statement applies also to the Kioways. Möllhausen met with two young Mexican prisoners among them, a man and a woman. The young man declared, that he did not want to exchange his present abode for another. The woman, though married to a chief and mother of a young chief, expressed the wish to return to her own country; but the chief would not let her and her child go [184]. We may suppose, that the same state of things prevailed here as among the Comanches.

Of the Apaches proper Bancroft (besides his above-quoted general statement) says: “They treat their prisoners cruelly; scalping them, or burning them at the stake; yet, ruled as they are by greediness, they are always ready to exchange them for horses, blankets, beads, or other property. When hotly pursued, they murder their male prisoners, preserving only the females and children, and the captured cattle” [185]. This is not very suggestive of slavery; and Bancroft’s general statement about the Apache family appears rather strange. Schoolcraft tells us: “These [the chiefs] can have any number of wives they choose; but one only is the favourite. She is admitted to his confidence, and superintends his household affairs; all the other wives are slaves to her; next come his peons, or slaves, and his wife’s slaves, and the servants of his concubines; then the young men or warriors, most generally composed of the youth who have deserted other tribes on account of crimes, and have fled to the protection of the chief of this tribe.... Then come the herdsmen, and so on” [186]. These “slaves”, ranking even above the warriors, very probably were not slaves in the true sense of the word. According to Bourke, the Mexican captives, living among the Apaches, were treated very kindly and often rose to positions of great influence. It does not appear that these captives were kept in a slave-like state. Fremont and Emory say: “Women, when captured, are taken as wives by those who capture them, but they are treated by the Indian wives of the capturers as slaves”. It is evident that we have not to do here with slaves in the true sense. Taking into consideration all the foregoing statements, we may suppose that slavery did not exist among the Apaches [187].

In an above-quoted passage Bancroft states that the Utahs frequently sell their wives and children as slaves to the Navajos. According to Bent, the Navajos, “have in their possession many prisoners, men, women and children, ... whom they hold and treat as slaves” [188]. But these statements are not sufficient for us to go upon; these prisoners may be adopted, or intended to be sold, as well as kept as slaves.

Miss Olive Oatman, who had been detained among the Mohaves or Mojaves, says: “They invented modes and seemed to create necessities of labour that they might gratify themselves by taxing us to the utmost, and even took unwarranted delight in whipping us on beyond our strength. And all their requests and exactions were couched in the most insulting and taunting language and manner, as it then seemed, and as they had the frankness soon to confess, to fume their hate against the race to whom we belonged. Often under the frown and lash were we compelled to labour for whole days upon an allowance amply sufficient to starve a common dandy civilized idler” [189]. Though such prisoners are held in a slave-like state, yet evidently the object of the masters in imposing disagreeable work upon them is not to get useful labourers, but to “fume their hate”. This account may warn us against attaching too much value to statements about slavery among similar tribes, especially where the “slaves” are whites. For such tribes as the Apaches, who are always ready to exchange their prisoners for some property, will be very apt to take prisoners, especially whites, who are likely to offer a better ransom than Indians. In such cases the prisoners are not killed; for by killing them the Indians would lose their ransom; but they may safely, as in Miss Oatman’s case, be treated as slaves by way of vengeance. But where these are the only slaves existing, a regular slave-system does not prevail. As for the Mohaves, no more particulars being given, we do not know whether they have slaves.

In Bancroft’s account of the Pueblo tribes no mention is made of slavery. On the Pimas he informs us: “If prisoners are taken, the males are crucified or otherwise cruelly put to death, and the women and children sold as soon as possible” [190].

In Parker Winship’s article it is quoted from Mendoza’s letter, that the Cibola Pueblo “keep those whom they capture in war as slaves” [191]. This being the only reference made to slavery, we are unable to decide whether it really existed.

Ten Kate in his detailed account of the Zuñi (a Pueblo tribe) makes no mention of slavery; so they probably have no slaves [192].

Bancroft, describing the Lower Californians, has nothing about slavery. Although their battles are described at some length, no mention is made of captives; probably they took no prisoners [193]. We may therefore safely infer, that slavery did not exist among them.

Result. Aleuts, [194] Positive cases: Athka Aleuts, Koniagas, Tlinkits, Haidas, Tsimshian, Kwakiutl, Bilballas, Ahts, Tribes about Puget Sound, Fish Indians, Tacullies, Atnas on Copper River, Similkameem, Chinooks, Atnahs on Fraser River and allied tribes, Sahaptins or Nez Percés, Southern Californians, Klamaths, Navajos, Cibola Pueblos. Negative cases: Greenlanders, Central Eskimos, Eskimos of Labrador, Frobisher Bay and Field Bay Eskimos, Kinipetu Eskimos, Tchiglit Eskimos, Eskimos of the Ungava District, Western Eskimos or Eskimos of Alaska, Eskimos of Point Barrow, Kutchins or Loucheux, Chepewyans or Athabascas, Lenape or Delawares, Montagnais, Ojibways or Chippeways, Ottawas, Shahnees, Potawatomi, Crees or Knisteneaux, Cheyennes, Blackfeet nation, Abenakies, Iroquois, Hurons or Wyandots, Katahbas, Cherokees, Muskoghe, Choctaws, Chickasaws, Creeks, Seminoles, Natchez, Dacotahs or Sioux, Hidatsas, Omahas, Osages, Kansas Indians, Assiniboins, Hupas, Apaches, Pimas, Zuñi, Lower Californians, Okanagans, Karoks, Central Californians, Nishinam, Shoshones, Utahs, Comanches, Kioways. No conclusion: Menomini, Attakapas, Inhabitants of Virginia, Mandans, Flatheads, Killamucks, Shastika, Mohaves.

We may add here a short account of Negro-slavery among the Indians.

According to the census of 1860, several Indian tribes had Negro-slaves. Our informant enumerates the Choctaws, Cherokees, Creeks and Chickasaws. Slavery was carried on to a great extent; some owners had from 50 to 200 slaves [195]. We may remember that all these tribes originally had no slaves.

The Creeks already in Bartram’s time (1789) had slaves. He tells us of a chief who kept 15 Negroes; they were slaves until they married Indian women, and then acquired the privileges of the tribe. Schoolcraft informs us that “if an Indian should murder a Negro, the law is satisfied with the value of the Negro being paid to the owner” [196].

The Seminoles also had Negro-slaves, according to Roosevelt and Gregg [197]. But Maccauley is not quite certain about it. He observed a few Negroes living with them. It had been said that they were slaves; but our informant is not of that opinion [198]. Maccauley’s account, however, dates from a later period than the other statements.

The Shahnees in Gregg’s time also kept a few Negro slaves [199].

But these facts do not represent phenomena of unadulterated savage life. These Indian tribes had already undergone great changes by contact with the whites. Moreover, the Negroes kept by them as slaves were in a very peculiar condition, living in a foreign continent amongst foreign races. So we are justified in omitting these cases from our list of slave-keeping Indian tribes.

§ 3. Central and South America.

About the treatment of prisoners by the wild tribes of North Mexico Bancroft remarks: “Seldom is sex or age spared, and when prisoners are taken, they are handed over to the women for torture, who treat them most inhumanly, heaping upon them every insult devisable, besides searing their flesh with burning brands, and finally burning them at the stake, or sacrificing them in some equally cruel manner. Many cook and eat the flesh of their captives, reserving the bones as trophies” [200]. These particulars given, and no mention being made of slaves, slavery probably does not exist among them.

Among the wild tribes of Central Mexico “the heads of the slain were placed on poles and paraded through their villages in token of victory, the inhabitants meanwhile dancing round them. Young children were sometimes spared, and reared to fight in the ranks of their conquerors, and in order to brutalize their youthful minds and eradicate all feelings of affection toward their own kindred, the youthful captives were given to drink the brains and blood of their murdered parents” [201]. Whether these children became slaves is not quite clear; we should think not, as they were “reared to fight in the ranks of their conquerors”; but this may also be the case with slaves [202]. The lack of further particulars prevents our arriving at a positive conclusion.

Bancroft’s notes on the wild tribes of South Mexico are very scanty. They sacrificed their prisoners. The Mayas had female slaves [203].

Bancroft informs us that “one principal object of war among the ancient nations of Honduras was to make slaves; but the Mosquito Coast was free from this scourge, according to all accounts.” “When prisoners were taken they were usually held as slaves, after having the nose cut off.... The coast people ... usually kill their prisoners.” Wickham, who gives a detailed account of the Woolwa or Soumoo of the Mosquito Coast, makes no mention of slavery [204]. So the inhabitants of Honduras had slaves, whereas those of the Mosquito Coast had not.

Slavery, according to Bancroft, was in force among the inhabitants of the Isthmus of Panama and Costa Rica, with the exception of the Caribs. “The prisoner is the slave of the captor; he is branded on the face and one of his front teeth knocked out. The Caribs however used to kill and eat their prisoners”. Gabb, who gives several particulars about the tribes of Costa Rica, makes no mention of slavery. According to Pokalowsky, the Indians of Coctu in Costa Rica, when conquered by the Spaniards in the 16th century, had slaves. “They cut off the heads of their prisoners and preserve them as trophies; the boys and girls of the enemies are enslaved or sacrificed to their gods. If a master dies, his slaves are killed and buried with him; this custom prevails here to a greater extent than in any other part of India” [205].

Bancroft’s statement about the Caribs of the Isthmus is confirmed by the fact that Pinart, who has largely drawn upon ancient Spanish literature, makes no mention of slavery [206].

Rochefort speaks of slavery existing among the Caribs of the Antilles. They believed, that the bravest warriors of their nation after death would live in happy islands, and have their enemies, the Arawaks, for slaves, whereas the cowards would be the slaves of the Arawaks. In their wars with the Spaniards they did not kill the Negro slaves, but took them with them and made them work. Sometimes slaves were killed after their master’s death, to serve him in the other world. Male prisoners were killed and eaten after a few days. Captive women became slaves; their children were reared with those of the Caribs. Female prisoners were sometimes taken for wives; then the children were free, but the mothers remained slaves. In the isle of Saint Vincent there were in Rochefort’s time English boys and girls, captured when very young; they had quite forgotten their parents, and would not even return with them, so accustomed were they to the mode of life of the Caribs, who treated them very kindly, just as if they were of their own nation. De la Borde makes no mention of slaves [207]. It seems to us very doubtful whether slavery really existed here. Rochefort’s statement that captive children were reared with those of the conquerors is more suggestive of adoption of captives than of slavery, and the enslavement of Negroes by the Indians is something foreign to the aboriginal state of things, as has been said before. So we cannot arrive at a definite conclusion.

The Continental Caribs, according to Gumilla, killed all their prisoners, except the young women and children, whom they sold [208]. So slavery probably did not exist among them.

Ling Roth, in his article on Hispaniola or Hayti (inhabited by Arawaks), makes no mention of slavery; but this does not prove much, as his sources of information (early Spanish literature) were very incomplete. For instance, he has not been able to find anything bearing on the division of labour between the sexes [209].

The several describers of the Indians of Guiana [210] make no mention of slavery. The tribes most fully described are the Arawaks, Warraus, Macusi and Roucouyennes. Martius however states that the Arawaks have slaves, who work in their houses and on the fields [211]. So we are not certain about the Arawaks; but we may safely suppose that among the three other tribes slavery does not exist.

The Saliva of Columbia, according to Gumilla, made war in order to acquire slaves to till their lands [212].

In Sievers’, Reclus’ and Simons’ descriptions of the Goajiro no mention is made of slaves [213]. According to De Brettes, however, “slavery exists; but the slave is a member of the family, though looked upon as an inferior being that may be killed if he refuses to obey”. A few more details are added about these slaves [214]. Sievers, reviewing De Brettes’ articles, remarks that this author is generally not very trustworthy, but that the ethnographical parts are the best of his work. Speaking of a photograph of Goajiro slaves given by De Brettes, be adds: “If there can be any question at all of slavery among them” [215]. On the same page, however, he translates De Brettes’ ethnographical account of the Goajiro, in which it is stated that slavery exists, without any commentary. Considering all this, we cannot arrive at any accurate conclusion.

De Lery, speaking of the ancient Tupinambas, describes at considerable length the fate of their captives, who were killed and eaten; even the child of a captive and a woman of their own tribe was not allowed to live. Though an expert hunter or fisher, and a woman well able to work, were preserved somewhat longer than the rest, all were invariably killed after a few months. Nowhere in De Lery’s book does it appear, that they made slaves by capture or by any other means [216]. According to another ancient writer, however, they kept prisoners as slaves. The slaves were kindly treated, allowed to marry free women, but finally killed and eaten. They had to catch fish and game and to bring it to their master. Without the master’s consent they were not allowed to work for others. If they tried to escape and were caught, they were killed. A slave, who died a natural death, was not buried but thrown away in the bush [217]. All this is very suggestive of slavery. But the fact, that D’Evreux got his information through an interpreter, prevents us from decidedly concluding, against the testimony of our other informants, that the Tupinambas kept slaves.

Martius remarks about the Indians of Brazil in general: “Many of these tribes keep slaves.... Captivity in war is the only cause by which one loses his freedom, especially if a male; for the husband may sell his wife and children; but this is of rare occurrence” [218].

The Apiacas (a group of the Central Tupis), according to the same author, in their wars kill all adult prisoners, male and female, and eat them. Children they take with them and rear them with their own; they make them work in the plantations; but when about twelve or fourteen years old, these children are killed and eaten [219]. Though these children may be kept in a somewhat slave-like state, a tribe that kills its slaves when full-grown is not properly to be called a slave-keeping tribe.

The same author informs us, that the Mundrucus and Mauhés have slaves [220].

Of the Miranhas we are told that they enslave their prisoners; but usually these prisoners are intended to be sold to the whites. It does not appear whether any of them are kept for the Miranhas’ own use [221].

Keane, von Tschudi and Ehrenreich make no mention of slavery among the Botocudos. According to Zu Wied “the conqueror persecutes the vanquished, and but seldom makes captives, at least among the Botocudos; but on the Belmonte there are said to be seen some who were used as slaves for all kinds of work” [222]. We do not know what this last second-hand information of Zu Wied’s is worth; but we are justified in inferring that the Botocudos in general (except those on Rio Belmonte) have no slaves.

Azara states that in his time (he travelled in South America from 1781 to 1801) the Guaycurû had nearly died out, only one man being left [223]. But according to Boggiani Guaycurû is a general name for the tribes that inhabit the Gran Chaco [224], so this statement of Azara’s seems to apply to a small division of the Guaycurû only. Southey and Martius give some particulars about the slave system of the Guaycurû [225]; but Colini, who has taken great pains to ascertain the identity of these tribes, quotes these descriptions as referring to the Mbayás. Of the ancient Guaycurû he says: “In their combats they gave no quarter to the adult males; but they spared the lives of the youths, whom they educated after their customs and gave in marriage to their daughters, so as to augment the number of their tribe. Full-grown women were sold to the neighbouring nations, who made them slaves” [226]. The only captives whom they kept among them, the youths, were not slaves; so slavery probably did not exist among them.

Two tribes inhabiting the Gran Chaco and so belonging to the Guaycurû in Boggiani’s sense, are described by Thouar. Of the Chiriguanos he says: “The prisoners are the property of their captors and must serve the mistress of the hut.” In his description of the Tobas he makes no mention of slavery [227]. Thouar, however, does not seem to be very well informed [228].

The Mbayás, according to Azara, in his time had two kinds of slaves, one composed of the Guanás, the other of Indian and Spanish prisoners of war. But the former were no real slave class. The Guanás “used to repair in troops to the Mbayás, to obey and serve them and till their lands without any payment. Hence the Mbayás always call them their slaves. This slavery is indeed very mild, as the Guaná voluntarily submits to it, and leaves off whenever he likes.” Such “slaves”, who lead a tribal life and come and go when they like, certainly are not slaves. The others however were real slaves. They procured the fuel, cooked the food, took care of the horses and tilled the land. When Azara once offered a present to a Mbayá, the latter would not take it himself, but ordered his slaves to receive it for him. Even the poorest Mbayá had three of four slaves. During the mourning-time women and slaves were not allowed to speak or eat any meat. One place in Azara’s book seems to show that they had no slaves: “They said they had received a divine command to wage war against all nations, kill the adult males and adopt the women and children, in order to augment their number.” But where the recorded tradition and the description of the actual state of things disagree, we hold that the latter is to be accepted [229]. Colini refers to Azara and Martius, and then adds: “Serra however asserts that among the Mbayás slavery proper (la schiavitù vera e propria) did not exist; the slaves might rather be called servants.” They fought together with the freemen and took part in the public council, even when it decided upon war and peace. They married free persons, but were themselves looked upon as slaves. On the master’s death, the sons or next relations, according to the rules of inheritance, became masters of the slaves; but these rights were only nominal. The slaves gradually merged into the tribe. Yet it was always considered degrading to be a descendant of a slave; those who had in their ancestry none but members of the tribe were very proud of it. Generally the best slave girls were married to their masters; the boys of greatest promise were treated as sons, whereas the others were set to do the ruder work [230]. This account, however, does not prove that the captives were not slaves. Some of the boys only were treated as sons; what were the rest if not slaves? And even slaves may to a certain degree be treated as sons. The slaves gradually coalesced with the tribe (though not entirely); but we are told that this change took place “through personal merits and intermarriage.” This shows that all captives were not on a level with the freemen; probably it was only the most deserving prisoners, and the offspring of slaves and freemen, who attended the public council and were on an equal footing with the main body of tribesmen. Slaves may be kindly treated and yet be slaves. Our opinion is, that we have here a schiavitù vera e propria; the more so, as the description given by Azara leads to the same conclusion.

The present Caduvei, according to Boggiani, are the same people as the ancient Mbayás. Very probably he is right here. Yet we have seen how much confusion there is in the application of the terms Guaycurû and Mbayás; so we are a little sceptical. Moreover, there is a great lapse of time between the early descriptions of the Mbayás and Boggiani’s travels, and during that time their state of culture has greatly changed; from nomadic hunters they have in the 19th century become settled agriculturists [231]. So we are justified in treating them separately. The Caduvei keep slaves. The slaves are well treated, but looked upon as an inferior race. The ruder kinds of work, and the tilling of the soil, fall to their share. As a rule they are kindly treated, without being allowed to forget their duties. The Caduvei exchange the slaves among them for horses, cattle and various commodities [232].

Pohl states, that the Canoeiros had captured a young man and treated him well [233]. Nothing more being added, we cannot make out whether slavery exists.

Of the tribes of Central Brasil, visited by Von den Steinen, the principal are the Bakaïri, Paressi, Bororo, Suya and Yuruna. In his description of the Bakaïri, Paressi and Bororo he makes no mention of slavery. If it existed, this careful observer would certainly have mentioned it. Neither is a word about slaves to be found in Hensel’s description of the Coroados, who are often identified with the Bororo [234].

Among the Suya Von den Steinen observed Indians of other tribes, who were kept as slaves. The presents, which the author gave to one of these slaves, had to be delivered to his master [235].

The same writer speaks of captives residing among the Yuruna; but his short remark on this subject cannot lead us to a safe conclusion as to the existence of slavery among them [236].

The Karayas on Rio Araguaya keep captive women in a somewhat slave-like state. Prisoners of war, adopted into the tribe, sometimes are made chiefs if they have distinguished themselves [237]. This is not very suggestive of slavery; but the details given are not sufficient for us to arrive at a clear conclusion.

On the Záparos we get this scanty information. They are always at war, killing many of the men, and stealing the women, children and chattels of the enemies, the children either for use as servants or for sale. A boy or girl stolen by them is commonly sold to traders [238]. Apparently the author himself is in doubt, whether any of these captives are kept as slaves.

Some savage tribes of Peru are treated of by Ordinaire. In his account of the Campas or Antis there is nothing bearing on slavery. He states that he met with a Lorenzo child living among the Campas; but it is not clear whether this was a slave; and the rest of his ample record makes the existence of slavery rather improbable [239].

About the Conibos and kindred tribes the same writer remarks, that among their wives there are some slaves captured from neighbouring tribes. But as he states, that all the fatiguing work is incumbent on women, it would seem, that there are no male slaves and therefore no slavery proper. Prisoners of war are killed at their feasts [240]. From a description of about a hundred years ago we learn, that these tribes kept prisoners as slaves. Several of these slaves were observed among the Panos; the masters treated them with as much affection as their own children and married them to their daughters. The conquerors married the captive women in order to augment the number of their tribe [241]. The details given are not sufficient to decide, whether the prisoners merged into the tribe or constituted a slave class.

Smyth and Lowe, speaking of the Sencis of Peru, remark: “They give no quarter, and take no prisoners in the battle.... The women and children are taken for slaves, and if there are any in infancy, or much advanced in age, they are killed as useless.” [242]. Whether the fate of the captive women and children was really slavery, is not clear from this short note.

The Guanas probably had no slaves. We are told that the head of the tribe “is obliged to work for his subsistence, as nobody serves him” [243].

As little does it appear that the Yuracarés and Mocéténès are slave-keeping tribes. The former live in families, and even in these subordination is unknown. The latter are not warlike [244]. It is not, however, a first-rate authority to whom we owe these particulars.

The Chiquitos, according to the same author, in their wild state attacked their neighbours, and made prisoners, to whom they gave their own daughters as wives [245]. Whether these prisoners were slaves is not clear; we should rather think not.

The Chapacuras were very peaceable, and but seldom attacked their neighbours [246]. Whether slavery existed among them we are not told.

The Moxos, in D’Orbigny’s time, had already long been civilized and christianized. What their political institutions were in their former wild state we do not know [247].

Muratori, speaking of the Indians of Paraguay and some neighbouring districts, states that they kill and eat their prisoners of war. Some tribes, however, he tells us, are more peaceful and take all pains to induce their prisoners to reside among them. Children of prisoners are sold by some of the tribes to other nations [248]. From all this it would seem that slavery did not exist. But we shall presently see, that the other information we have got does not entirely agree with Muratori’s general statement.

The principal native tribe of Paraguay were the Abipones, described at large by Dobrizhoffer. The prisoners they made were very leniently treated. They gave them the best of their food, and tended them when