IX THE REGULATION OF NUMBERS AMONG PRIMITIVE RACES

1. We have now to ask what light these facts throw upon the quantitative aspect of the problem. Remembering that among all the races concerning which facts have been given there exists that form of primitive social organization the nature of which has been referred to, we may first examine briefly the theory of population as it applies to society in which there is co-operation; for the existence of this primitive form of social organization implies co-operation. We may then go on to apply what we learn from this review of the theory of population to the facts, so far as they concern the races of the first and second groups. We may next ask how far we can apply what we learn from primitive races to prehistoric races up to the opening of the third period, and finally we may inquire how it is to be supposed that the transition took place from the conditions under which the pre-human ancestor, lived to those under which the earliest societies, of which we can indirectly gain any knowledge, must be supposed to have lived.

Malthus was the first writer to set out a theory in detail and to support it with evidence.^[1032] Of the origin of his book some account has been given in the first chapter. In this book Malthus, according to his own account, attempted to show three things—that population was limited by the means of subsistence, that it almost always increases when the means of subsistence increase, and that there are three checks upon increase—vice, misery, and moral restraint. By ‘vice’ and ‘misery’ he meant disease, war, poverty, and so on. By ‘moral restraint’ he meant restraint from sexual intercourse. This last check was not mentioned in the first edition of the Essay; it was introduced for the first time in the second edition.

It is important to observe the nature of the argument put forth 198by Malthus. (The whole theory is one of the comparative rapidity of the increase of population and of the increase of food.) Population, he said, when unchecked, increases in a geometrical ratio, food only in an arithmetical ratio. Therefore there must always be checks at work limiting population. He made a survey of the social conditions in different countries, and pointed to the evidence of the existence of various forms of vice and misery; where moral restraint was practised, there was less vice and misery; where no moral restraint was practised, vice and misery reached their greatest prevalence. Furthermore the checks taken together must always be effective; it was, according to Malthus, merely a question of what kind of checks should be in operation. It was desirable in his view to increase ‘moral restraint’ in order to decrease ‘vice’ and ‘misery’. It also follows that, if the checks were always effective, there could be no such thing as over-population. The conception of over-population, properly speaking, did not enter into the theory at all. It belongs to the later theory, which is based upon the productiveness of industry—an idea which finds no place in his book.

In the later editions Malthus made certain reservations about the impossibility of subsistence increasing faster than in an arithmetical ratio. Facts incompatible with the theory regarding the increase of population in America had come to light, and he admitted the possibility of the increase of subsistence in a geometrical ratio in new countries under certain conditions. He maintained, however, that in general subsistence could not increase faster than in an arithmetical ratio, and that his theory was therefore in essential features still correct. The ratios were at the basis of his theory, and sum up the whole essence of the argument. It has frequently been said, however, that Malthus did not attach much importance to the ratios. Professor Nicholson, for example, says that he used them ‘not strictly—but as the basis for a simile’.^[1033] But Professor Cannan has shown that there is no foundation whatever for this view, and quotes a passage from Malthus exhibiting the importance which the latter attributed to this part of his theory.^[1034]

That subsistence can only increase in an arithmetical ratio, or in other words that the periodical additions to the average annual 199produce cannot be increased, has long ago been disproved by statistics. Such figures as caused Malthus to make his reservations about new countries have been forthcoming for old countries also. In his own lifetime the census showed the falsity of the arithmetical ratio for so ‘old’ a country as England. It is not necessary to go into the facts, which are well known and beyond dispute. It is only the result of the proof of the fallacy contained in the arithmetical ratio part of the theory that concerns us. The result was that the whole argument collapsed, founded as it was on the comparative rapidity of the increase of population and of food. As Professor Cannan says: ‘The Essay on the Principle of Population falls to the ground as an argument, and remains only a chaos of facts collected to illustrate the effect of laws which do not exist. Beyond the arithmetical ratio theory, there is nothing whatever in the Essay to show why subsistence for man should not increase as fast as an “unchecked population”. “With every mouth God sends a pair of hands,” so why should not the larger population be able to maintain itself as well as the former?’^[1035]

2. The answer to this question was in process of being discovered during the lifetime of Malthus. In the early years of the century attention was drawn to the high price of corn, to the position of agriculture, and especially to the fact that less rich land was being brought under cultivation. From the discussion which ensued, and to which Malthus contributed, there arose the idea of decreasing returns to agriculture. It has been pointed out that it must always have been recognized in practice that it did not ‘pay’ to employ more than a certain amount of labour on a given area of land. In these years, however, for the first time the principle which underlay this fact was made clear. West and Ricardo were chiefly responsible for bringing the matter to light. The development of the theory need not occupy us. It may be formulated shortly here. ‘Whether we consider an acre of land or a whole country, after a certain point is reached, the return to a given amount of labour and capital will diminish. It will do so, however, only under the supposition that the arts of agriculture, using the phrase in the broadest sense, remain stationary.’^[1036] The law is not limited to agriculture. It is applicable to all industry. Whenever some agent in production, upon 200which an industry is absolutely dependent, is strictly limited, after a time equal increments of capital and labour will not produce equal returns. In other words the returns to the same doses of capital and labour will diminish. This, however, will only occur when, as stated above, the supply of some agent in production is restricted; otherwise when an industry enlarges there will, as a rule, be an increasing return. The more labour, apart from diminishing returns, the better.

When speaking of agriculture it was said that diminishing returns only supervene when there are no improvements in the arts of agriculture. This is true in general of all other industries. Any increase in the arts of production will in general admit of more labour and capital being profitably employed; if there are no further improvements, there will after a time come a point when diminishing returns will again supervene. The following out of the working of these laws in detail would call for a long and intricate discussion. Since the problem of quantity is only touched upon in its broadest outlines, such a discussion is not necessary here. The broad fact which emerges, and which alone is relevant here, is that, since the laws in general are applicable to all industry, there will be, taking into account on the one hand the known arts of production and on the other hand the habits and so on of any people at any one time in any given area, a certain density of population which will be the most desirable from the point of view of return per head of population. There will, in fact, under any given circumstances always be an optimum number; if the population fails to reach that number or if it exceeds it, the return per head will not be so large as it would be if it attained that number. This conception is of such importance that its formulation in other words by a modern economist may be quoted. ‘At any given time, or, which comes to the same thing, knowledge and circumstances remaining the same, there is what may be called a point of maximum return, when the amount of labour is such that both an increase and a decrease in it would diminish proportionate returns.... If we suppose all the difficulties about the measurements of the returns to all industries taken together to be somehow overcome, we can see that at any given time, or knowledge and circumstances remaining the same, just as there is a point of maximum return in each industry, so there must be in all industries taken together. If 201the population is not large enough to bring all returns up to this point, returns will be less than they might be, and the remedy is increase of population; if, on the other hand, population is so great that the point has been passed, returns are again less than they might be, and the remedy is decrease of population.’^[1037]

This idea of an optimum number is one which can be developed in great detail. It is only necessary here to notice that when, as in the higher economic stages, the arts of production on the one hand are improving and the habits and so on of any people are on the other hand constantly altering, the most desirable density is in consequence frequently changing. In the lower stages, when progress in skill is slow and social conditions more or less stationary, the optimum number may remain about the same over long periods of time. Further, with regard to progress in skill, inasmuch as the productivity of labour is enhanced by every improvement in the arts of production, the result of such progress will be that the return per head will, as a rule, increase. Such progress will thus, as a rule, allow of a larger population which will have a larger income per head.

This idea of an optimum density of population is wholly different to that put forward by Malthus. To him the problem was one of the relative increase of population and of food; with us it is one of the density of population and of the productiveness of industry. To Malthus the position was much the same in all ages; in his view population, except under unusual circumstances, had in any country at any given time always increased up to the limit of subsistence, and was in process of being checked—chiefly by vice and misery. In the modern view increase in skill has brought to an increasingly dense population a larger income per head. The chief cause of the largest possible income not on all occasions being reached is that at times the density of population increases beyond the optimum number for the given conditions, though at other times the failure of population to reach the desirable level may produce the same result.

Though the problem of diminishing returns was being discussed as the successive editions of the Essay appeared, the idea was 202not applied by Malthus to the population question. This was undertaken by later writers. The development of the modern theory need not detain us; it is of interest, however, to observe that J. S. Mill, whose influence over contemporary thought was so great, never shook off the deep impression made upon him early in life by the Malthusian theory. He exhibited a remarkable dread of over-population, and in his Principles of Political Economy seems to have regarded diminishing returns as so often arising that only rarely could an increase in population be advantageous. He appears in fact to have thought that after a certain stage all further increase in population was harmful, and to have looked upon ‘the degree of industry which is required for the maximum productiveness of industry as something fixed once and for all’.^[1038] Mill’s view, therefore, though founded upon the productiveness of industry, differs from the modern view to which it has given place. It has since been clearly established that there is no maximum density desirable from the point of view of productiveness; so long as skill increases, other things being equal, so long will the desirable density increase. The influence of Mill’s writings has had the effect of keeping alive in England longer than elsewhere the pessimistic view of the problem which the teaching of Malthus had originally given rise to.^[1039] So strong were Mill’s opinions that he regarded the limitation of the family by virtuous restraint as one of the most desirable of social reforms. He was led in the following passage to express views that are strongly in contrast with those of the many who now deplore the decline of the birth-rate among the upper social classes of to-day. ‘Little improvement’, he said, ‘can be expected in morality until the producing of large families is regarded with the same feelings as drunkenness or any other physical excess. But while the aristocracy and clergy are foremost to give the example of this kind of incontinence, what can be expected of the poor?’^[1040]

3. This conception of an optimum number holds good wherever there is social co-operation between groups of men living within definite areas. As we have seen a primitive form of social organization exists among all these races. This implies in itself a certain 203degree of co-operation. It remains to show that these races are all, without exception, divided into groups which are strictly limited to definite areas—contrary to the still common notion that they wander where they please—and we may further note that the inhabitants of these areas co-operate to a greater or less degree in the search for food, and that there is a social obligation upon each man to do his share. We may review the evidence, beginning with the first point, which is of such importance that it may be treated at some length.

It is both misleading and dangerous to apply terms carrying modern legal conceptions to these races; unfortunately, any terms that may be used are to some extent biased. We have, however, somehow to describe the results of investigation into the customs of these races, and it has been found that among all these races, without exception, groups of men are recognized as, if not owning, then as enjoying the usufruct of certain very clearly defined areas. According to Ling Roth it is not clear whether the Tasmanian tribes had any definite hunting grounds or not.^[1041] Bonwick confidently asserts that it was so,^[1042] and Wheeler thinks that the conditions in Tasmania were probably the same as in Australia.^[1043] From Australia we have abundant evidence; the facts recorded from different parts of the continent vary considerably, and we must either believe that customs differed from place to place, which is not at all unlikely, or that many observers are mistaken, which, considering the positive nature of their assertions, is not very probable. Wheeler has reviewed the evidence and sums up his conclusions as follows: ‘Our information shows the existence in some, at any rate, of the areas of Australia of what must be held to be private ownership in land, but it does not follow that the whole of the tribal territory was so allotted. The unit would generally be the individual family, but there are a few indications that the ownership might even be vested in single persons within the family, other than the head. The clearest mentions of individual or of family ownership seem to come from the south-eastern area, where the physiographic conditions are most varied, and where, in consequence, fishing rights become important. But it seems likewise that the rights of families or of individuals, as also those of local 204groups, were, in general, subject to tribal over-rights, though we have no clear information on this point.’^[1044]

There is no doubt, in the first place, that the tribes were everywhere restricted to clearly defined territories.^[1045] Doubt only arises with regard to the smaller groups. The usual mode of living is in small groups of one to three families, which usually form a portion of a local group, but which may exceptionally be actual local groups. Only in the more fertile districts is the number of families living in contact greater.^[1046] These local groups, into which the tribes are divided, seem generally to have their clearly defined areas within the tribal territory. Thus Brown, speaking of the Kariera tribe of Western Australia, says that ‘the country of a local group, with all its products, animal, vegetable, and mineral, belongs to the members of the group in common. Any member of the group has the right to hunt over the country of his group at all times. He may not, however, hunt over the country of any other local group without the permission of its owners.’^[1047]

In some parts no further subdivision of the land was recognized. ‘I could not find’, says Brown of Western Australia, ‘any evidence of the individual ownership of any part of the soil or of any of its products. The whole of the territory of the group and everything in it seem to belong equally to all the members of the group.’^[1048] In other parts we hear of family ownership; Stanbridge says of Victoria that the tribal land ‘has been from time immemorable parcelled out among its families and transmitted by direct descent to the present generation’.^[1049] Further, some accounts speak of individual ownership. The natives of King George’s Sound ‘who live together have the exclusive right of fishing and hunting upon the neighbouring grounds, which are, in fact, divided into individual properties, the quantity of land owned by each individual being very considerable’.^[1050] Eyre speaks of the parcelling out of territory among the individual 205members of the tribe. ‘Every male has some portion of land, of which he can always point out the exact boundaries. These properties are subdivided by a father among his sons during his own lifetime, and descend in almost hereditary succession.’^[1051] As further examples of the ideas held by the Australians on the subject of property, it may be noticed that Smyth describes individual property in trees which passes by inheritance,^[1052] while Lumholtz says that ‘if a native finds a hive of honey in a tree, but has not an immediate opportunity of chopping it out, he can safely leave it till some other day; the discoverer owns it, and nobody else will touch it, if he has either given an account of it or marked the tree, as is the custom in some parts of Western Queensland’.^[1053]

Wherever we turn we find similar evidence of the recognition of distinct areas over which groups have more or less exclusive rights.^[1054] Every tribe certainly has its own clearly defined territory; it seems further very probable that in most places the tribal territory is divided among local groups, if indeed the subdivision does not go farther. The Bushmen were formerly divided into tribes occupying ‘well defined tracts of country, which they looked upon as their own ancestral hunting ground’.^[1055] Their respect for property is shown by the fact that, when a man found a bees’ nest, he put his mark upon it, and thenceforward it became ‘the sacred property of the finder’.^[1056] Klutschak describes distinct territories for the Eskimo tribes, who confine their seasonal migrations to these definite areas.^[1057] Subdivision of land does not seem to go farther than subdivision among the villages; the inhabitants of a village had the right to refuse to allow any strangers to settle permanently in the neighbourhood. ‘If a new family wished to settle at an inhabited place, the newcomers had to await the consent of the people already settled there, which was given by certain signs of civility or welcome, the strangers having meanwhile put their boat ashore, but not yet begun bringing up their goods. If these signs were not given they pushed the boat off again and went on to look for another place.’^[1058] It is so well known that the American Indians recognized clearly defined tribal boundaries that we need not attempt to 206illustrate the fact by means of references.^[1059] It is perhaps not so well known that there are also many indications of family and even of individual property in land. In the first place we often hear of the right of the inhabitants of a village to clearly defined tracts. ‘Each tribe had its village sites and contiguous hunting and fishing grounds; as long as the people lived on these sites and regularly went to their hunting grounds, they could claim them against all intruders.’^[1060] Of the Carrier Tribes, Harmon says that ‘the people of every village have a certain extent of country, which they consider their own, and in which they may hunt and fish; but they may not transcend these bounds, without purchasing the privilege from those who claim the land’.^[1061] Of the Western Tinneh, Hill Tout records that the heads of the clan own the hunting grounds, ‘the limits of which were always very clearly defined’.^[1062] Among the more nomadic Eastern Tinneh we hear that bands used the same hunting grounds; these areas, however, were not regarded as belonging exclusively to them.^[1063] Among the Salish of the Interior, ‘all hunting, fishing, root and berry grounds were common property and shared in by all alike’; whereas among those of the coast the food grounds were the property of the septs and local groups.^[1064] From other accounts it appears that family and individual ownership were recognized in some places. This applies especially with regard to the fishing stations of the Pacific Coast tribes, where ‘varying lengths of shore were held as private fishing rights by heads of families, and these rights were passed from father to son and were always respected’.^[1065] Swanton gives an especially interesting account of the practices of the Haidahs of Queen Charlotte Islands. ‘Each Haidah family had its own creek or creeks, or portion of a creek, where its smoke houses stood. Some of the smaller creeks are said to have no owners; and, on the other hand, some families are said to have had no land. In the latter cases they were 207obliged to wait until another family was through before picking berries, and had to pay for the privilege. Any family might pick berries on the land belonging to another after the owners had finished picking, if it obtained the consent of the latter and paid a certain price.’^[1066] Referring to the Sitkin Indians, Elliot says that ‘the coastline, and especially the margins of streams and rivers, are duly divided up among the different families. These tracts are regarded as strictly private property.’^[1067] Krause further gives an account of what can only be called individual private property in land among the Thlinkeets.^[1068] The remarkable system in vogue among the Veddahs deserves notice. ‘The whole Veddah country was divided into small hunting regions, of which each family possessed one. The arrangements were most elaborate; the size of the tracts varied in accordance with the goodness of the land, and each included a portion of hill country to which each family thus had access during the rainy season without trespassing upon the ground of other families.’^[1069] Cooper has very fully reviewed the evidence for the Fuegians. It has been stated that communism existed among them. This is not correct. ‘While all Fuegians are nomads,’ he says, ‘yet a Yaghan, for instance, is chary of poaching on Alacalufan or Onan territory. Even within recognized tribal territory the existence of more or less definitely marked off family hunting grounds is explicitly attested for the Onas by Professor Furlong and Dr. Dabbene and implicitly by Dr. Gallardo.’^[1070]

4. Passing now to the races of the second group it is everywhere found that tribal territories are recognized, and it has been shown that, compared with hunting and fishing races, there is an increase in the number of cases in which land is held by smaller groups, if not by individuals, and a corresponding decrease in the number of cases in which communal ownership is recognized.^[1071] The details of the variation in land tenure are not relevant, and a few descriptions of the conditions among different tribes will serve as examples 208of the manner in which groups of men among these races drew their means of subsistence from clearly defined areas. ‘Among the Navahos a section of territory was parcelled out and held as clan land and, as descent in the tribe was traced through the mother, was spoken of by members of the clan as “my mother’s land”. Upon such tracts the woman worked raising maize, &c., and the product was recognized as their property.’^[1072] Of the North American Indians generally we are told that ‘occupancy gradually established a claim or right to possess the tract from which a tribe or individual derived food. This occupancy was the only land tenure recognized by the Indian; he never of himself reached the conception of the land as merchantable.... As long as a person planted a certain tract the claim was not disputed, but if its cultivation were neglected, any one who chose might take it. Among the Zuni, according to Cushing, if a man, either before or after marriage, takes up a field of unappropriated land, it belongs strictly to him, but is spoken of as the property of his clan, or on his death it may be cultivated by any member of that clan, though preferably by near relatives, but not by his wife or children, who must be of another clan.’^[1073] So, too, among the Omahas maize was cultivated in patches of one-half to three acres in size; property in these patches was recognized so long as cultivation was continued. Afterwards any one could take them.^[1074] In Mexico there was an elaborate system of land tenure.^[1075] Conditions in South America were very similar to those in North America. Spix and Martius speaking of Brazil say: ‘The savages consider the lands they have put under cultivation to be in some measure the property of their tribe.... One or several families unite to clear a part of the virgin forest and plant maize, manioc, cotton, or bananas.... The same ground is cultivated every year, because it would be too difficult every year to clear new portions of the forest.... A field cultivated for several years is considered to belong to a family, and the neighbours recognize these rights.’^[1076]

The rights over land are very similar in Africa. Bartle Frere says that ‘it is clear, from the accounts of early Dutch and other travellers in South Africa, that every Hottentot tribe had its 209territory, into which strangers might not intrude for pasture or hunting, without the leave of the whole tribe; each kraal had its pasture lands distinct, over which the people of that section of the tribe moved their mat huts, as the need of their herds, for grass or water, rendered advisable. As each kraal had more or less of a family constitution, it is difficult to say how far the pasture lands were held in common, or as the property of the individual occupant.’^[1077] Of the Ewe-speaking people, Ellis says that among most tribes there is no private property in land but that the family in occupation of any tract cannot be disturbed.^[1078] Of the Yoruba-speaking people the same author says that land, belonging to the community collectively, is vested in the chief, who distributes among households and families as required. No man can be depossessed of land once allotted to him, and the usufruct descends to his children, but the land cannot be sold.^[1079] ‘It is doubtful’, says Talbot, ‘whether any part of the Ekoi country remains at present unowned.’^[1080] According to Scott Elliot the rights of property in land of the different families in Sierra Leone are carefully maintained, especially in times of famine.^[1081] In Northern Nigeria a ‘man is free to mark out a farm on any unoccupied land’, and has a claim to it as long as he cultivates it.^[1082] Among the Bangala of the Upper Congo ‘the boundaries of a town are well defined, and the lands belonging to a town are well known to all other towns in the neighbourhood. If an animal is killed on ground owned by a town other than that to which the huntsman belongs, he has to send a portion—generally the head—to the chief of the town which claims the ground.’^[1083] Further, ‘every woman had her farm, which was her exclusive property, and not even a fellow wife had any rights over it’.^[1084] The hunting grounds of the Bushongo are well defined; the usufruct of the soil belongs to individuals, and trees are private property.^[1085] Among the Mangbetu ‘every tribe has an accurate knowledge of the boundary of the territory over which it can hunt, move about, and establish its villages’.^[1086] Hobley states that among the Akamba there is individual property in land which passes on the death of the father to the sons.^[1087] The 210limits of the properties are very well defined and well known among the Ababua; cultivation renders the title good.^[1088] The Baganda ‘lived in their gardens or plantations. These gardens were often joined one to the other, and a number of people lived in a community, often forming four or five miles of continuous garden with families living each on their own plot.’^[1089] Even the fallow land among the Akikuyu was ‘all in private ownership’, and could not be brought into cultivation without the owner’s consent.^[1090] In British Central Africa a man making a garden is ‘perfectly free to choose so long as the ground is not in cultivation, or has not been bespoken by some one else; and, once marked, no one can interfere with it’.^[1091] Of the same region we are told that ‘these negroes have clear ideas of property. The waste land is usually considered to belong to the chief, but plantations and enclosures belong personally to the individual who originally made them.... Natives have clear ideas of large or small estates, or of their kingdom; and in the case of the former they are marked by the planting of certain trees of thick growth, while of course streams and mountains are recognized as boundaries and natural limits of territories.’^[1092] Of the Thonga tribe in Portuguese East Africa we hear that ‘each man has his own field which he tills’.^[1093] Speaking of the Bantu races south of the Zambezi, Theal says that ‘the chief apportioned to each head of a family sufficient ground for a garden according to his needs, and it remained in that individual’s possession as long as it was cultivated’.^[1094] So Conder says of Bechuanaland, ‘the land belongs to the chief. He divides it among his head men and they in turn among their people. There is no division of grazing land. The mealie fields are practically the property of the cultivator so long as they are tilled. I found each patch to belong to an individual, and to be divided generally by untilled land from the next patch.’^[1095] Among the Ovaherero the whole land is common property, but the rights of tribes and also of individuals to particular spots are recognized so long as they are occupied. Whoever first appropriates a spring and the surrounding pastures can maintain his right.^[1096]

211The conditions in Oceania are in their main elements, which alone are relevant, similar to those in America and Africa. In the Pelew Islands every family had its plot of land, which was considered as private property so long as it was occupied and cultivated.^[1097] In New Zealand ‘land was held primarily by tribal rights; and within this tribal right each free warrior of the tribe had particular rights over some portion’.^[1098] In Sarawak ‘each tribe had its limits, which have been handed down from father to son for ages, so that every old man of a tribe knows the exact extent of its district’.^[1099] In British New Guinea we hear of ‘properly regulated and well-defined property rights, certainly not bounded by surveyed lines commonly used to indicate European land claims, but marked and known by natural features such as surface conditions afford’.^[1100] It may also be noticed that rights over certain areas are recognized by the peoples of Northern Asia, such as the Tunguses and the Yakuts.^[1101]

5. Passing to the second and third points we find that within these areas there is evidence even among the most primitive hunters of co-operation in the search for food and of strict rules for the division of the available food. Coupled with this we find that there is a social obligation which is strongly enforced for every man to do his share. Thus, where hunting and fishing can only be carried out by joint parties, strict rules exist for the division of the catch among those who have taken part and among their dependents. This may be illustrated by a few examples. Howitt gives many details of the rules in force in Australia with regard to the division of food. These rules not only differ from place to place but also for the various forms of game. Thus among the Kurnai ‘a wombat is cooked, then cut open and skinned. The skin is cut into strips and divided with parts of the animal thus. The head to the person who killed the animal. His father the right rib; mother the left ribs and backbone, which, with some of the skin, she gives to her parents. Her husband’s parents receive some of the skin. The elder brother gets the right shoulder, the younger the left. The elder sister gets the right hind leg, the younger the left hind leg, and the rump and the liver are sent to the young men in the camp.’^[1102] ‘If a man only killed enough game or procured enough food for 212himself, his wife and his children, then he need not divide with others; but if he found that his father had no food, he would give them what he had and go out and look for more.’^[1103] ‘All the males in the Chepara tribe are expected to provide food, if not sick. If a man is lazy and stays in camp, he is jeered at and insulted by the others.’^[1104]

The principal features of the Australian customs as described above are to be found among all hunting and fishing races, though such elaborate rules as are usual in Australia are somewhat uncommon elsewhere. The Bushmen may have had their rules for the division of game; all we are told is to the effect that they shared food. ‘When one feasted they all partook; and when one hungered they all equally suffered.’^[1105] We have evidence both of the sharing of food and of the dividing up of game among the Eskimos. Details of the latter are given by Nansen;^[1106] with regard to the former we have frequent references to the division of food within the villages among all who need it.^[1107] Of the Eskimos we are also told that ‘it might be considered a law that every man, as far as he was able to do it, should practise the trade of a hunter on the sea, until he was either disabled by old age or had a son to succeed him. This duty neglected, he brought upon himself the reprehension not only of the other members of his own family, but also of the wider community.’^[1108] Rules for the division of food were almost universal among the Indians; when a hunter of the Hare tribe kills an animal, he is only allowed the tongue and ribs; the rest is distributed according to a system.^[1109] So, too, among the Lillooet a regular partition of the game took place, of which one of the features was ‘that the persons who had the game had no preference over others’.^[1110] Such phrases as ‘studied equity in the distribution of necessaries’,^[1111] which is applied to the Seri Indians, indicate the existence of similar rules.^[1112]

Conditions are similar among the races of the second group. In Fiji ‘public opinion took care that no man in the community 213shirked his work’.^[1113] In New Ireland, ‘should a man neglect his family, a mode of punishment very similar to one practised by schoolboys among civilized races is adopted. A double row of men, women, and children—the whole population of the village—armed with stiff birches is formed; and at a signal from the chief the delinquent is obliged to run a certain number of times through the lines and receive a general castigation from the rods of the villagers’.^[1114] Of the Pawumwa Haseman says ‘all work together.... If any one refuses to assist in planting, the chief forces him to work. I saw one Indian with a long scar on the side of his head and neck, the result of punishment for laziness.’^[1115]

6. It is thus clear that within any group in any primitive race, the members of which co-operate together to obtain their food from a definite area to which they are confined, the principle of the optimum number holds good. There is, that is to say, taking into account the abundance of game, the fertility of the land, the skilled methods in use, and all other factors, a density of population which, if attained, will enable the greatest possible average income per head to be earned; if the density is greater or if it is less than this desirable density, the average income will be less than it might have been. Obviously it must be a very great advantage for any group to approximate to this desirable density. There are three possibilities open to any group. The desirable number may be approached, it may be exceeded up to the point where men can only just exist, or it may not be reached. Extreme departures from the optimum number must be very disadvantageous; if numbers increase until they are limited by starvation only, then no benefit arises from the use of any skilled methods that may be known. Under such circumstances all inventions in the methods of hunting, fishing, and cultivating the ground profit nothing. Social conditions must also inevitably be unstable where starvation alone limits numbers.

This being so, how are numbers regulated? We may observe to begin with that there is a number of factors at work among all these races which incidentally limit increase. This they may do either by decreasing fertility or by increasing elimination. To the former class belong pre-puberty intercourse and prolonged lactation, to the latter war and lack of care of children. Two 214characteristics of these factors are noteworthy. The effect they have upon the limitation of increase is incidental; it is the chance accompaniment of the practice of certain customs or of certain habits. In the second place the working of any single factor, so far as it reduces fertility or increases elimination, is fairly regular; in any primitive race at any given time such habits as prolonged lactation and early intercourse, when they occur, cause a certain definite reduction of fertility; and such customs as those of war and those connected with the upbringing of children cause a certain definite amount of elimination. The nature of the factors present and the degree of their incidence varies considerably from race to race, but it follows from what has been said above that in any primitive race over a considerable length of time the amount by which fertility is decreased and the amount of elimination which is caused remain fairly constant.

There is another class of factors the primary and not the incidental function of which it is either to reduce fertility or to cause elimination. These factors are prolonged abstention from intercourse, abortion, and infanticide. The view put forward here is that normally in every primitive race one or more of these customs are in use, and that the degree to which they are practised is such that there is an approach to the optimum number. With regard to this view we may first deal with the question as to the prevalence of these customs. We may then refer to the evidence as to the nature and origin of these practices. We shall then be in a position to ask how it may be supposed that they are so practised as to bring about that amount of restriction of increase which will result in an approach to the optimum number being made.

7. The evidence regarding the practice of these customs—prolonged abstention from intercourse, abortion, and infanticide—has already been given in the last two chapters. This evidence is also summed up in the Appendix to which the reader is invited to turn. No attempt has been made to conduct an exhaustive inquiry; but it is claimed that there is ample evidence of the widespread prevalence of one or more of these practices. There is no indication of the correlation of any one practice with any one economic stage. As far as the evidence goes, any practice may be in use in any economic stage. To the meaning of this we shall refer later. It is further claimed that, when the influence of contact with Europeans is taken into account, the evidence of the existence 215of these customs is still more impressive; for the observations upon which we have to rely have for the greater part been made after these races had been in contact with Europeans, and such contact is followed by a diminution of these practices.

One of the first changes brought about by this contact is connected with the introduction of diseases previously unknown. These diseases are often peculiarly fatal, causing a very high death-rate, and it is clear that, unless the practices of abstention from intercourse, abortion, and infanticide were largely abandoned, the race would perish. There is no difficulty in understanding how these practices would actually be abandoned soon after the introduction of disease. The proximate causes of these practices are largely, as we shall see, the difficulty of transporting many young children and the undesirability of having more than one child during the period of lactation. If disease began to carry off many children, the immediate causes for these practices would largely or entirely disappear and so would the customs themselves.^[1116]

Apart from the introduction of disease, the effect of contact is in other ways to reduce the extent of these practices. Contact has often resulted in warfare between aboriginal races and European settlers—as in Tasmania, Australia, and America, or as between the Bushmen and the Boers. The largely increased death-rate would, as in the case of disease, be followed by a diminution of the extent of these practices. Further, under other conditions, as for instance in Polynesia, the efforts of missionaries have long been directed to putting a stop to these customs.^[1117] We may also note that there is often a bias on the part of observers to under-estimate the extent of the employment of infanticide and abortion. Many observers are attracted by the races who come under their notice and seem to think that these practices are incompatible with the kindly nature or pleasant disposition of the people they describe—that in fact the attribution to them of such customs as 216a normal feature of existence is a kind of libel—and if they report them at all, they persuade themselves into thinking that they are infrequent and abnormal. We have only to remember the objection taken to the interpretation of the finds in neolithic graves in England as evidences of infanticide—an objection based on the dislike of believing that our ancestors could have been guilty of such a habit. Or, again, we may remember that Tacitus, when desirous of holding up the Germanic tribes as an example to the Romans of his day, declared that they never committed infanticide—the implication being again that infanticide is to be regarded as a degenerate and unworthy custom.^[1118]

Nevertheless, in spite of all such tendencies working for the rapid disappearance of these customs, and in spite of the bias against believing in their existence, there is, as the Appendix shows, ample evidence that one or more of these practices are recorded for almost every people.

8. We may now consider the evidence regarding the origin and nature of these practices. We shall find that abortion and infanticide arise owing to the difficulty of providing for more than one child at a time. We shall also find that these customs are practised as normal features of social life and in such a manner as to keep the number of children at a fairly constant average figure. There is frequently a belief that a particular number of children is the right number of children. Further, it will be seen that the number of children to be preserved is a matter for consideration in which the wishes of not only the parents but also of the relations and of the community in general have to be taken into account, and that the practices are enforced by social pressure.

A large amount of evidence has been given to the effect that the number of children is always small—the number, that is to say, after these practices have taken effect. This is in itself evidence, it may be noticed, that these practices do result in reducing the number of children to a small constant average. Some additional evidence may now be noted.

According to Curr infanticide in Australia ‘resulted principally from the difficulty, if not the impossibility, of transporting several children of tender age from place to place on their frequent marches’,^[1119] and later he observes that ‘infanticide was common 217among the Bangerang.... They themselves gave as their reason for it the impossibility of the woman carrying more than one infant in their constant wanderings.’^[1120] ‘Every child which was born before the one which preceded it could walk was destroyed, because the mother was regarded as incapable of carrying two,’^[1121] says another observer of the Narrinyeri tribe. Considerable information is given by Howitt for various Australian tribes. ‘Infanticide is practised by the Mining to some extent, the mode of killing being by starvation. After a few days of short commons the child becomes peevish and troublesome, and in consequence more neglected, being placed by itself away from the camp and fires, and it is said to be afflicted with Mupurn (magic). When death ends its sufferings Mupurn is the cause. The reason they give for this practice is that if their numbers increase too rapidly there would not be enough food for everybody. Yet they are very fond of their offspring, and very indulgent to those they keep, rarely striking them, and a mother would give all the food she had to her children, going hungry herself.’^[1122] ‘In the Tongaranka tribe the practice of infanticide was common, because a baby was frequently too much trouble to look after, and it was often the mother who killed it. But it was not done until the family consisted of three or four; but after that too much work in hunting had to be done to keep the family in food.... In the Mukjarawaint tribe the children belonged to the grandparents, though the parents had care of them. If, for instance, a boy was born and then a girl, the father’s parents might take them, or the mother’s parents, and so also with another couple of children. If, then, another child was born and one of the grandparents took it, it would be kept. If not, it was killed, there being too many children. The grandparents had to decide whether a child was to be kept alive or not. If not, then either the grandfather or father killed it, by striking it against the mother’s knee, and then knocking it on the head.... According to Buckley, if a family increased too rapidly in the Wadthawing tribe, as, for instance, when a woman had a child within twelve months of the previous one, there was a consultation in the tribe as to whether it should live or not.... Infanticide in the Kurnai tribe arose through the difficulty of carrying a baby when there were other children, 218especially when the next youngest was not able to walk.’^[1123] Of the aborigines of the River Darling we are told that ‘it seems to have been the custom to kill many of the children directly after birth, to save trouble and privation in time of drought, when long distances must be travelled in the search for food and water’,^[1124] and the same proximate cause of infanticide is mentioned by Beveridge.^[1125] Of the women in the Port Lincoln district it is said that they put forward as the cause of infanticide the fact that they ‘cannot suckle and carry two children together’;^[1126] similar reasons are said to be given by the women of Central Australia.^[1127] Speaking of the natives of Port Darwin Foelsche says that the reason for infanticide is that ‘too many children encumber the parents in travelling about for food’.^[1128] It is also worthy of note in connexion with Howitt’s statement about there being at times a dread of over-population that, according to Spencer and Gillen, this is not the case at any rate in Central Australia;^[1129] according to Curr, however, ‘in many tribes there is a great fear of a want of food arising from over-population’.^[1130] Among the tribes of Port Lincoln ‘the number of children reared by each family is ... very limited, rarely exceeding four’, and ‘if a mother has children in rapid succession ... the young infant is killed’.^[1131] In the Dieyerie tribe about 30 per cent. of the children are destroyed at birth.^[1132] In the neighbourhood of Port Darwin children are killed ‘where a woman has more than three or four’.^[1133] In Central Australia ‘each mother only rears, upon an average, two children’.^[1134] Among the Northern tribes of Central Australia the number of children is kept down by infanticide to two or three.^[1135]

Similar evidence is forthcoming for other races of this group. Thus of the Puelches Guinnard says that ‘among these almost primitive creatures, children are not nearly so numerous as might be imagined, for the existence of the new born infant is submitted to the judgment of the father and mother, who decide on its life or 219death’.^[1136] Charlevoix says of the Abipones that ‘they seldom rear but one of each sex, murdering the rest as fast as they come into the world, till the eldest are strong enough to walk alone. They think to justify this cruelty by saying that, as they are almost constantly travelling from one place to another, it is impossible for them to take care of more infants than two at a time; one to be carried by the father, the other by the mother.’^[1137]

Turning to races of the second group in Funafuti every mother was allowed to keep alternate children, but the second, fourth, and so on had to be destroyed.^[1138] ‘A Tikopian family is usually limited to four children, any excess of this number being killed by burying them alive in the house or just outside it; occasionally five or six may be kept alive but never more. If the first four children are girls one or more of these may be killed in the hope that the succeeding children may be boys, in which case the lives of the boys would be spared.’^[1139] In New Guinea, in the later years of marriage, abortion is used in order to lessen the number of children, since a large family would mean too much work for the parents.^[1140] In the Kingsmill Islands ‘a woman seldom has more than two children, and never more than three; when she discovers herself to be enceinte for the third or fourth time, the foetus is discharged by a midwife’.^[1141] In the Sandwich Islands, ‘however numerous the children among the lower orders, parents seldom rear more than two or three, and many spare only one; all the others are destroyed, sometimes shortly after birth, generally during the first year of their age’.^[1142] ‘They consider three children a burden,’ says the same author in another passage, ‘and are unwilling to cultivate a little more ground, or undertake the small additional labour necessary to the support of their offspring during the helpless period of infancy and childhood.’^[1143] ‘No married pair (among the Gilbert Islanders) are allowed by their law to have nor bear more than four children, that is only four children get the chance of life. The woman has the right to rear or to endeavour to rear one child. It rests with the husband to decide how many children shall live, and this depends on how much land there is to divide.’^[1144] In Fiji ‘infanticide is more prevalent among 220the poor classes than the rich’.^[1145] Speaking of the Murray Islands Hunt says that ‘after a certain number had been born, all succeeding children were destroyed, lest the food-supply should become deficient’.^[1146] Codrington says of the Melanesians that ‘abortion and infanticide were very common. If a woman did not want the trouble of bringing up a child, desired to appear young, was afraid the husband might think the birth before its time, or wished to spite her husband, she would find some one to procure abortion.... Infanticide was more prevalent in some islands than others.... The old women of the village generally determined whether a new-born child should live; if not promising in appearance or likely to be troublesome, it was made away with.’^[1147] Among the Mafulu of British New Guinea a woman must not give birth to a child unless she can give a pig to a village feast, and consequently children are often destroyed either by abortion or infanticide, both of which are common.^[1148] Of the Western Islands in the Torres Straits we hear that ‘few women rear more than three children, and besides most of those born before marriage are doomed to be killed immediately after birth, unless the father—which is seldom the case—is desirous of saving the child.... Even of other infants some, especially females, are made away with in a similar manner when the mother is disinclined to support it.’^[1149] Of the Eastern Islands it is said that ‘after a certain number had been born, all succeeding children were destroyed, lest the food-supply should become insufficient’.^[1150]

Rengger, who remarks upon the small number of children among the Guaranis, traces it to the regular practice of abortion after a certain number had been born.^[1151] ‘Infanticide is quite common among the Lenguas, an interval of seven or eight years being always observable between children of the same family. Not only are babies, which are born in the interval, immediately killed, but abortion is also practised.’^[1152] Among the Creek Indians ‘to destroy a new-born infant is not uncommon in families that are grown so numerous as to be supported with difficulty’.^[1153] Of the Cheyennes it is said that ‘it has been the custom that a woman should not have a second child until her first is ten years old. 221When that age is reached, the man is likely to go with his wife and child to some large dance or public gathering, and there ... to announce publicly that now this child is going to have a little brother or sister.’^[1154] There is sometimes evidence, for instance among the Sioux^[1155] and the Brazilian tribes,^[1156] that abortion is committed after consultation with the husband. Among the Pima Indians ‘sometimes a mother nursed a child until it was six or seven years old, and if she became pregnant in the meantime she induced abortion’.^[1157] In Fiji there was no cohabitation until the child was two years old. ‘This separation ... was in bygone times invariably enforced’ and abortion was practised when there had been enough children.^[1158] Speaking also of Fiji Seeman says that ‘relatives of a woman take it as a public insult if a child should be born before the customary three or four years have elapsed, and they consider themselves in duty bound to avenge it in an equally public manner’.^[1159] Similarly Gutmann says of the Wadschagga that it is considered most disgraceful if a woman, who is still suckling a two years child, again becomes a mother, and that in consequence abortion is frequent.^[1160] In German New Guinea, according to Krieger, only three children are as a rule brought up owing to a fear of scarcity of food.^[1161] In Radeck ‘every mother is allowed to bring up only three children; every fourth and succeeding one she is obliged to bury alive herself’.^[1162] Infanticide was ordered by law in Vaitapu and not more than two children were allowed in a family.^[1163] Of the Roro-speaking tribes of New Guinea we are told that ‘formerly it was not customary for a woman to have children until her garden was bearing well’.^[1164] In the New Hebrides ‘infanticide was sadly prevalent. As the burden of plantation and other work devolved on the woman, she thought she could not attend to more than two or three children, and that the rest must be buried as soon as born.’^[1165] In the New Britain group ‘after marriage children are not borne by the woman for a period of from two to four years. I am informed that this is the result of a popular dislike to speedily becoming a mother on the part of the women, 222who use various means to procure abortion and use them successfully.... A considerable period elapses between the birth of one child and the birth of another. The general term is about three years. One child is always well out of hand before another appears.’^[1166] We may also note that, according to Rivers, infanticide is frequent among the Todas and is practised not when food runs short but as a regular custom.^[1167]

It will be remembered that, when setting out in the last two chapters the evidence for the extent of abortion and of infanticide, these practices were in many cases stated to be committed when there was a certain number of children thus corroborating the evidence given to this effect above.

9. We are now in a position to discuss the manner in which numbers are regulated among these races, leaving until later the discussion as to how the position among these races was derived from that among species in a state of nature. Everywhere groups of men are, as we have seen, confined to definite areas. Among the few things which the men composing these groups do know with accuracy are the limits within which their food must be obtained. Further they co-operate in the search for food. It follows, therefore, that within any such area there is—taking all the relevant facts into account—an optimum number. The advantages to any group of approaching this number are immense; a wide departure from this number can only be socially disastrous. We found that within any group there is a number of factors—some of which reduce fertility and others of which increase elimination—and that the average amount of restriction of increase resulting from the action of these factors was fairly constant. We also found that there were certain other factors—prolonged abstention from intercourse, abortion, and infanticide—which are everywhere present and considerably restrict increase. If, as we shall see later, there is reason to think that some approach to the optimum is everywhere attained, it is clear that the former factors cannot of themselves alone sufficiently restrict increase. It is therefore to the latter factors, the primary function of which it is to restrict increase, that we must look when we seek for the mechanism by which numbers are brought near the desirable level.

It is clear how these factors originate. Among more or less nomadic peoples abortion and infanticide are practised because 223of the difficulty of transporting and of suckling more than one child at a time. Abstention from intercourse arises as a taboo. The problem we have to face is how these practices could come to be of the necessary intensity. Now men and groups of men are naturally selected on account of the customs they practise just as they are selected on account of their mental and physical characters. Those groups practising the most advantageous customs will have an advantage in the constant struggle between adjacent groups over those that practise less advantageous customs. Few customs can be more advantageous than those which limit the number of a group to the desirable number, and there is no difficulty in understanding how—once any of these three customs had originated—it would by a process of natural selection come to be so practised that it would produce an approximation to the desirable number. There would grow up an idea that it was the right thing to bring up a certain limited number of children, and the limitation of the family would be enforced by convention.

Though, however, adjustment is understandable as the result of a natural selection of customs, the evidence shows that there is even among the most primitive races at times at least some deliberation as to whether a child shall be allowed to live. In the more advanced races there is increasing evidence of deliberation. It cannot be supposed that in deliberation of this kind there is any grasp of the true position regarding the importance of the optimum number, but it may be supposed that under these circumstances the actual position at the moment as to whether there are too many or too few in the group does weigh when taking the decision. To all members of such a group, confined as they are within the knowledge of them all to a limited area, the disadvantages of too many mouths must be obvious. Therefore even among the more primitive races there may be some semi-conscious adjustment of numbers by means of one of these methods. However this may be, it is clear that, even if there is no semi-conscious deliberation among the lower groups, there is to some extent an automatic adjustment to the needs of the moment. Suppose disease or severe weather, for instance, to produce a higher rate of infant mortality, then abortion and infanticide, inasmuch as they are practised because of the difficulty of transporting more than one child, will be less practised.

10. Leaving aside for the moment the evidence that an approximation 224to the desirable number is in this manner attained, we may note that in order that the system should be effective, something more is necessary. The conception of an optimum number involves the idea of a standard of living. The attainment of the optimum number indicates that the highest standard, which is possible taking all the circumstances into account, has been reached. In order that the standard of living may be maintained, it is not merely enough that numbers should be restricted; the younger generation must become proficient in the skilled methods which make this standard possible of attainment, and in particular it is important that the young men should not marry unless they are both energetic and skilful—unless, that is to say, they are both willing to keep up and are capable of keeping up the standard of living previously attained. There is abundant evidence to the effect that pressure is exerted upon the younger generation. It is commonly recognized that among primitive races the girls marry at or soon after puberty. It is not so often recognized that the young men not infrequently do not marry until some years later. The inefficient and the physically incapable sometimes do not marry at all. In such facts we may see evidence of the pressure exerted by social conditions and conventions. We find also that not only are young men carefully instructed in the skilled methods, but that the parents of the bride anxiously inquire as to the bridegroom’s energy and capability of supporting a family. The conditions regarding marriage have been much studied, and in marriage by service and marriage by purchase we can observe the pressure which forces a young man to show himself proficient in the skilled methods which are in use among his people. Obligations on the would-be bridegroom to work and save up the purchase money or to serve his future parents-in-law make it necessary for the young man to learn skilled methods and to exhibit energy and competence before marriage, whereby it is in any case rendered unlikely that new families will be set up which will adopt a lower standard of living than that of the previous generation. At the evidence for these facts we may now glance.

In Australia betrothals generally take place in infancy and marriage follows later; these betrothals are usually arranged as exchanges, but the system shows some features of marriage by purchase.^[1168] Considerable postponement of marriage is often 225observed, especially among the less skilled members of the community; thirty is mentioned as a not uncommon age for marriage.^[1169] Among the Eskimos there is a very strong feeling that a boy must show himself proficient in the difficult arts of hunting and fishing as practised by them. A young man seldom marries until he is over twenty, though often betrothed when an infant.^[1170] Before that age he cannot learn all the methods of hunting game, managing a kayak, and so on. Marriage is considered impossible because he would not be in a position to maintain his family, and it is the necessity of showing that he is in this position before marriage is allowed which is remarked upon by all observers of Eskimo life.^[1171] The usual form of marriage in America is by purchase.^[1172] As a rule, the amount to be paid is relatively large, and the boy must either spend some time in getting it together or else obtain it from his father. At other times marriage is by service. Thus among the Kenai the bridegroom performs a year’s service,^[1173] and among the Haidahs boys were often betrothed at an early age and went to live with the girl’s family and worked for them until marriage.^[1174] A Seri bridegroom leaves his own family and enters that of the bride for a year; ‘he must display and exercise skill in turtle fishing, strength in the chase, subtlety in warfare and all other physical qualities of competent manhood’.^[1175] Among the Jakun a husband ‘is expected to provide a hut, cooking pots and other necessary articles such as will suffice to enable house-keeping to be started with reasonable comfort’.^[1176] Among the Fuegians ‘as soon as a youth is able to maintain a wife by his exertions in fishing and bird catching, he obtains the consent of her relations, and does some piece of work, such as helping to make a canoe, or preparing sealskins’.^[1177]

226In this connexion it may be noticed that we often hear of the scorn poured upon the weak and unsuccessful; it appears to be the case that such men sometimes never marry.^[1178] Among the Kutchins ‘poor men, whose abilities as hunters are small, and who have been unable to accumulate herds, remain bachelors’.^[1179] One of the most difficult things an Eskimo has to learn is how to catch seals. ‘The poor wretch’ that cannot do so ‘is despised to the last degree, and is obliged to subsist on women’s diet, such as scolpings, which he can fish for on the ice, mussels, periwinkles, dried herrings, &c.’^[1180] With regard to postponement of marriage, from twenty-three to twenty-five is the usual age of marriage for men among the Thompson Indians,^[1181] from twenty-one to twenty-five for men among the Lillooets,^[1182] and twenty-five among the Abipones.^[1183]

With regard to the races of the second group, among the Topebatos men marry when about eighteen. They have both to present gifts to, and to do work for, their future parents-in-law.^[1184] The bridegroom also works for his future parents-in-law among the Bontoc Igorot.^[1185] In Fiji ‘young men of the lower orders married rather late in life for a primitive people, rarely, it appears, before the age of twenty-five’.^[1186] Later the same author says that ‘marriages were often delayed for years when the bridegroom’s family were too poor to acquire property commensurate with their pride’.^[1187] In the Maldive Islands ‘although a man is allowed four wives at a time, it is only on condition of his being able to support them’.^[1188] In the Caroline Islands ‘a suitor serves for his wife in the house of his father-in-law elect as Jacob did with Laban and frequently has his pains for nothing’.^[1189] Marriage by purchase is common in New Guinea and often results in a postponement of marriage while the purchase money is being collected.^[1190] Speaking of the New Britain Group, Danks says ‘some I have met who have never 227married, but the cause lay in their inability to raise the shell money with which to purchase a wife’.^[1191] In Sumatra marriage by purchase is said to constitute a certain hindrance to marriage, in spite of which, however, there are few celibates.^[1192] Among the Negritos of Zambala the amount of the purchase money is large and ‘there is no doubt that the gifts made represent almost all the wealth of which a young man and family can boast’.^[1193] In the Western Islands of the Torres Straits men marry when between twenty and twenty-five years old,^[1194] marriage being by purchase.^[1195] In certain cases where head hunting is practised prowess in this art must be shown before marriage.^[1196] In certain parts of Borneo ‘it is a rule among all the tribes that no youth can regularly wear a mandan, or be married or associate with the opposite sex, till he has been on one or more head hunting expeditions’.^[1197]

Marriage by purchase is often found in Africa. The necessity of collecting the purchase money frequently involves some postponement of marriage. Formerly among the Thonga a man generally married when about twenty-five years old,^[1198] but there was some variation in the age owing to the varying difficulty experienced in getting together the necessary cattle.^[1199] Sooner or later nearly all the men among this tribe get married;^[1200] and this is true of all Bantu peoples. ‘The kind of individual called a bachelor does not abound among the Bantu. The wretched, the invalids, the weak-minded only, are deprived of the legal marriage which for the black man is and remains the one object in life.’^[1201] Sometimes in South Africa ‘a young man too poor to acquire a wife by the transfer of cattle would make an arrangement with the father of the girl to live with her and serve him’.^[1202] Among the Baronga young men do not marry for several years after puberty.^[1203] Werner’s description of marriage among the Zulus is of particular interest. ‘The price paid by the Zulus (under the name of Lobola) and others cannot properly be called purchase, being rather in the nature of a settlement or guarantee that the suitor is able to support a wife; it is held by her family in trust for her and her children.’^[1204] The same author, speaking generally of British Central 228Africa, says that ‘young men may have to wait for some years, owing to lack of means or other reasons. In the country under the Angoni chiefs they are called on ... to “serve their time”, herding the chief’s cattle, and later, perhaps, going to war.’^[1205] The men of the Akikuyu are described as ‘not marrying very young’.^[1206] Roscoe enumerates the various articles required to make up a bride’s price among the Baganda, and adds that ‘they were difficult to obtain and represented a large sum to a poor person, so that it took him a long time to collect them; a man frequently spent twelve months begging among his relatives and friends the amount asked; for though as a rule he had secured some of the things before he went to ask for the lady, there would still be a balance to find’.^[1207] Among the Wapagoro regular sexual intercourse is practised by children. At puberty they are separated and the boy must then begin to collect the purchase money, and not until he has finished may intercourse be resumed.^[1208] Plas notes postponement of marriage among the Kuku owing to the amount of the purchase money.^[1209] An interesting account of the Akamba is given by Dundas. ‘During his lifetime a man so divides his stock that he allots a portion to each of his wives. On his death the portion of each wife goes to her son or sons.... If the cattle left are not numerous enough to buy a wife for each son they are left with the eldest son until the increase suffices for the purchase of a wife for him. When the increase is again large enough it is given to the second son to buy a wife, and so on until each has a wife.’^[1210] According to Johnston, ‘the Masai warrior is not allowed by the elders of his tribe to marry until he has reached about thirty years of age and has accumulated a fair amount of property or else so distinguished himself by his bravery as to merit an early retirement’.^[1211] Among the Bangala, if a man is accepted by a girl he has to pay the ‘bespoke’ money, after which ‘the girl is reserved for him until such time as he can pay the whole or larger part of the marriage money’, and, while he is collecting this money, he will build a house if he does not already own one.^[1212] Further, ‘a man can marry as many women as he can afford to pay marriage money for, but to each he must give a house’,^[1213] and 229as a result of the high marriage price, if his family cannot help him, a man ‘cannot save enough to procure a wife until he is thirty or even older’.^[1214] Cureau states that in the Congo basin a bride’s relatives are much interested in estimating a suitor’s capabilities of supporting a family and of, in general, doing well.^[1215] Among the Ekoi people, ‘by a native custom if a man wishes to marry an Ekoi maiden he must serve her people for some considerable time, usually from two to three years. His work mostly consists in helping to clear bush for next season’s farms, but other services may be required of him. During this time he is expected to make presents to the relations of his future wife.’^[1216]

Very similar institutions are found in America. Among the Nandowensis, ‘when one of their young men has fixed upon a young woman he approves of, he discovers his passion to her parents, who give him an invitation to come and live with them in their tent. He assiduously accepts the offer, and by so doing engages to reside in it for a whole year, in the character of a married servant. During this time he hunts, and brings all the game he kills to the family, by which means the father has an opportunity of seeing whether he is able to provide for the support of his daughter and the children that might be the consequence of their union.’^[1217] Among the Ojebway Indians marriage is also by service and the future father-in-law is described as being anxious that the suitor should be a good hunter.^[1218] Of the Natchez it is said to be ‘rare for young men to marry before they be five-and-twenty. Till they arrive at that age they are looked upon as too weak, without understanding and experience.’^[1219] According to Dorsey, men of the Omaha tribe did not marry formerly until between twenty-five and thirty.^[1220] Among the Attakapas, ‘if a savage desires to marry a girl whose father is still living, he approaches the latter; the latter then inquires if he is a brave warrior, a good hunter’, if he can make weapons, and so on.^[1221] In Mexico men married about twenty.^[1222] In British Guiana marriage is by service, and the bridegroom must show that he is capable of 230taking a man’s part; within a certain time he must clear a piece of land of a given area.^[1223] Another author gives further details: ‘Before he is allowed to choose at all he must prove that he is a man and can do a man’s work.’ There are various tests. Among others ‘he clears a space in the forest to be planted with cassaba, and brings in as much game and fish as possible, to show that he is able to support himself and others’.^[1224] In Peru a man had to be twenty-four years old before he could marry.^[1225] Von Martius states that the system of marriage by purchase in Brazil is to be regarded as a symbol that the bridegroom can support a family.^[1226] Men among the Guanas marry when over twenty.^[1227]

11. The evidence so far adduced shows that the mechanism whereby numbers may be kept near to the desirable level is everywhere present. When inquiring into the nature of the mechanism, we found certain indications that it was effective. The regularity with which certain customs are practised, the small average number of children, and other facts afford strong but not conclusive evidence that an approach is normally made to the optimum number. We may now ask what further evidence there is as to the position in this respect.

Conclusive evidence is not available. It is only when we are able, as among some of the races in the third group, accurately to measure the average income over a period of years during which numbers are changing that we can arrive at a definite result. With regard to other races we have to adopt methods which, though less precise, nevertheless afford important evidence. We may inquire into the general conditions of life and ask whether there are indications of the approach to the highest standard of living within reach, or whether living is more usually reduced to the bare level of subsistence. We may ask whether famine and starvation frequently occur, what the average physical condition of the people is, whether old age is often reached, whether, in short, the conditions are such as we should expect to find them when the optimum number was approached, or as we should expect to find them when the numbers reached the maximum which could just be kept alive.

The conception of savage life which formerly prevailed was that 231primitive races were always in a condition of semi-starvation.^[1228] In this connexion it has to be remembered that the conception of the optimum number has regard to all the conditions, and that among these races, taking the degree of skill, social customs, and so on into account, there would be, even if the optimum number was attained, times of scarcity, if not of famine. The existence of times of scarcity, therefore, is no evidence that, so far as numbers are concerned, these races have not attained the best conditions possible for them. It is undoubtedly the fact that they are physiologically adapted to withstand periods of scarcity in a manner that civilized men are not adapted, and it is interesting to observe that, according to the results of certain experiments, occasional periods of semi-starvation are far less harmful than is continual under-feeding.^[1229]

Descriptions of the Australians suggest the picture of anything but an emaciated people in a condition of semi-starvation. In one place Spencer and Gillen speak of them as well nourished;^[1230] in another place they describe a typical Arunta as ‘by no means poor in physique; in fact he might often serve a sculptor for a model, and, when walking behind a native, you are continually struck with his proportions and beautiful carriage’.^[1231] These authors go on to say, however, that ‘there are times when he is hard pressed and during a long continuance of drought his life is not a happy one’.^[1232] So, too, Schürmann says of the tribes of Port Lincoln: ‘the male sex exhibits a great deal of unstudied natural grace in their deportment, their walk is perfectly erect and free, motions of the body easy and gestures natural under all circumstances.’^[1233] Further, we are told that ‘their food, if of indifferent quality, was at least wholesome and readily procurable, six hours a day abundantly sufficing for that purpose, so that hunger was little known’,^[1234] and that ‘in most of the districts with an indigenous population game is so abundant compared to the number of inhabitants, as to enable every one to procure for himself and his family as many pounds of meat a day as his heart desired’.^[1235] Perhaps this is too optimistic a view; there are certainly many references to lean times when food is difficult to procure. ‘In few 232parts of Australia’, says Thomas, ‘can the native count on anything like regular supplies of food’,^[1236] and we hear of those powers of enduring hunger and thirst so common among these races and so incomprehensible to Europeans.^[1237]

Burchell is enthusiastic about the Bushmen; ‘as we rode onwards, I could not cease admiring the beautiful symmetrical form of our Bushman guide, who walked and sometimes ran before us, with a gait the most easy and free that I ever beheld. All the limbs, unshackled by clothing, moved with a grace never perhaps seen in Europe. The contemplation of his well-proportioned, although small and delicate figure, his upright manly port, his firm and bold step, and the consciousness of liberty which beamed in his countenance afforded us indescribable pleasure.’^[1238] Nevertheless the Bushmen had ‘periods of fasting’^[1239] and were ‘often destitute of food for several successive days during seasons when both roots and game were scarce’.^[1240] They were also ‘capable of remaining a long time without food, and could then devour immense quantities of meat without any ill effect’.^[1241]

In most countries a lean season occurs periodically once or more every year. Such lean seasons are more exacting in some regions than others. The Australians manage to subsist for the most part without storing up food;^[1242] in other countries the storing of food becomes a necessity.^[1243] We often find that where food has to be stored up the stock runs low towards the end of the lean season. The Eskimos have to depend for many months upon the food they have preserved, and if for any reason, as for instance the late formation of the floe, their calculations are upset, they may be placed in a very trying situation.^[1244] Of them we hear stories, as of other races, regarding their powers of withstanding hunger; ‘a man who has eaten nothing for three days, at least nothing but sea grass, can manage his little kayak or canoe in the most furious waves’.^[1245]

233Of the Kutchins of the Peel River we are told that they are ‘an athletic and fine-looking race, considerably above the average stature, most of them being upwards of six feet in height and remarkably well proportioned’.^[1246] ‘The Indians north of the Columbia are, for the most part, good-looking, robust men, some of them having fine, symmetrical forms. They have been represented as diminutive; with crooked legs and uncouth features. This is not correct; but, as a general rule, the direct reverse is the truth.’^[1247] The Montagnais are described as ‘tall, strong, erect, well-proportioned and agile’;^[1248] on the other hand, we hear that they are liable to shortages of food, are able to go without food for three days together at a time, and are guilty of great excesses of eating and drinking when food is plentiful.^[1249] Hardisty says that the Loucheux can always obtain food except under very unfavourable circumstances.^[1250] ‘It sometimes happens that the Ahts are in straits for want of food, when the fish do not appear until late in the spring,’^[1251] but ‘they can bear the want of food a long time without becoming exhausted’.^[1252] The account of the Californians given by the acute Jesuit missionary Baegert is of particular interest. ‘Notwithstanding the barrenness of the country, a Californian hardly ever dies of hunger, except perhaps now and then an individual that falls sick in the wilderness and a great distance from the mission, for those who are in good health trouble themselves very little about such patients, even if these should happen to be their husbands, wives or other relations, and a little child that has lost its mother or both parents is also occasionally in danger of starving to death.... The food of the Californians is certainly of a mean quality, yet it keeps them in a healthy condition, and they become strong and grow old in spite of their poor diet.’^[1253] ‘Californians can endure hunger easier and much longer than other people; whereas they will eat enormously if a chance is given.’^[1254] Three days without food appeals differently to an observer in a mission station on the one hand and to a prisoner 234among the Indians on the other. ‘So protracted was the hunger we experienced,’ says Cabeça de Vaca, ‘that many times I was three days without eating anything. The natives also endured as much, and it appeared to me a thing impossible that life could be so prolonged.’ The same author goes on to say, however, that ‘they are a merry people considering the hunger that they suffer; notwithstanding they never cease to dance nor to observe their festivities and ceremonies. To them the happiest part of the year is the season of eating prickly pears; for then they have hunger no longer, and pass all the time in dancing and they eat day and night.... It occurred to us many times while we were among this people, and there was no food, to be three or four days without eating, when they, to revive our spirits, would say to us not to be sad, that soon there would be pears and we should enjoy plenty, and drink of the juice, and that our bellies would be very big, and we should be content and joyful, having no hunger.’^[1255]

It is submitted that the conditions indicated by the type of evidence of which examples have been given above are not compatible with the state of existence on the bare means of subsistence.^[1256] With this evidence there should be considered the facts given in the sixth chapter regarding the good health and advanced age to which these races generally attain.

Turning to races of the second group, in the Eastern Islands of the Torres Straits ‘nutritious food is generally very scarce at the end of the dry season’.^[1257] The Dyaks usually experience a season when it is difficult to procure food.^[1258] We are told that famine was unknown in Fiji,^[1259] but that from November to February there was sometimes a scarcity when the last yam crop had been consumed and the next crop had not ripened.^[1260] So too among the Baganda ‘no one ever went hungry while the old customs were observed’,^[1261] but there are at times lean seasons. ‘When food is abundant they have their three meals daily; when it is scarce they content themselves with two, and hope for the rain and a plentiful supply of fruit.’^[1262] Cureau, describing 235the races of the Congo Basin, gives a general description of the conditions which corresponds very closely with that just quoted for the Baganda.^[1263] In British Central Africa “the time of hunger” comes after the rains, when the last year’s corn is eaten, and the new is not yet ripe—about March’.^[1264] Of the Kagero of Northern Nigeria we are told that the ‘people are naturally more healthy and better conditioned in December, say, than in August’, because towards the end of the year there is plenty of food from the harvest whereas later in the wet season there is usually a scarcity.^[1265] So of the Mandja of the Northern Congo we are told that ‘generally by the month of October, the Mandja family has consumed its reserves of food; thereupon follows want for two months’.^[1266] In North America Catlin describes times of scarcity among the Mandans^[1267] and im Thurm describes similar conditions in Guiana.^[1268]

Of the fine development of physical form there is abundant evidence. The inhabitants of Timor Laut are ‘handsome-featured fellows, tall, erect, and with splendidly formed bodies’.^[1269] One of the earlier travellers, Kotzebue, was very much struck with the physical features of the inhabitants of the Sandwich Islands^[1270] and of Radeck.^[1271] So, too, in Africa the Akikuyu are ‘exceedingly strong, muscular, healthy and well set up’.^[1272] ‘The Damaras, generally speaking, are an exceedingly fine race of men. Indeed it is by no means unusual to meet with individuals six feet and some inches in height, and symmetrically well proportioned withal. Their features are, besides, good and regular; and many might serve as perfect models of the human figure.’^[1273] As in the case of the races of the first group, we are told that formerly these people lived to a great age. ‘Under natural conditions the Bantu were a longer lived people than the Europeans.’^[1274] Catlin was greatly impressed with the physical fitness of the North American Indians. He never saw ‘a more hardy and healthy race of men’, while the women were ‘exceedingly healthy and robust’.^[1275] According to Heriot, ‘the North Americans are in general robust, and of a healthful temperament, 236calculated to live to an advanced age’.^[1276] Du Pratz speaks highly of the physical characteristics of the Natchez and much similar evidence could be quoted.^[1277] Azara, for instance, says that the Payaguas lived to an advanced age^[1278] and also mentions especially the fine physical features of the Mbayas^[1279] and of the Charruas.^[1280] Of the Mandrucos Wallace says that ‘their figures are generally superb; and I have never felt so much pleasure in gazing at the finest statue, as at the living illustrations of the beauty of the human form. The development of the chest is such as I believe never exists in the best formed Europeans, exhibiting a splendid series of convex undulations, without a hollow in any part of it.’^[1281] Humboldt remarks of the Chaymas that both men and women ‘are very muscular, but fleshy and plump’.^[1282] So, too, Spix and Martius, speaking of Brazilian tribes, say that ‘the Indians are seldom sick and generally live to an advanced age’.^[1283]

Lastly it may be observed as evidence of some weight that, as between different races, the higher the skill and the greater the natural richness of the surroundings, the higher, so far as we can judge, is the return per head. Though the return cannot be measured, all that we know of the conditions of life point to this conclusion, whether we contrast the agricultural races in general with the hunting and fishing races, or whether we contrast such hunting and fishing races as those of the north-west coast of America with the Fuegians.

12. The conclusions derived from a study of the first two groups are therefore to the effect that an approximation to the desirable number is normally attained by the practice of one or more of the three customs mentioned. As a result, the small average size of the family is arrived at—the size being just about that which allows for the average mortality from various causes later in life, so that in the next generation there will be as nearly as possible the same number of adults as in the former generation. It may be that, as an occasional coincidence, just that amount of reduction of fertility and just that amount of elimination necessary to bring about approximation are effected by means of those factors which were described as having incidentally these results. But such a coincidence must be very rare and the 237evidence shows that the practice of one or more of the other class of factors is widespread, if not universal. Further, it has been shown that a more or less automatic adjustment is attained by means of variations in the intensity of the operation of these latter factors, whereas it is difficult to see how it could come about by variations in the former factors only, and, however unprogressive social organization and general conditions may be, some adjustment must be necessary from time to time.

It has been observed that there is no apparent connexion between the practice of any of these customs and the different economic stages, and it may be stated that an investigation undertaken to see if any connexion could be detected was without result. This conclusion is not surprising. The problem of how to control numbers is one which is presented to all races at all times. We have grouped races according to their economic status; but there is no reason to expect that under any one economic system a particular method of controlling population would be adopted more than any other method—no reason, for instance, why infanticide rather than abortion should be practised under one system rather than under another. As has been said, the problem is always present, and the method adopted must originally have depended upon some factor quite independent of the economic stage; abortion, for instance, might have been practised in one country because of the presence of some herb which experience showed to be effective, while in another country a taboo upon sexual intercourse for magical reasons might have been developed into a taboo during lactation.

This survey of primitive races is limited to the elucidation as to what is normally the position as regards the regulation of numbers. The evidence is not sufficiently detailed to allow us to judge whether in any one particular instance there was or was not a close approximation to the desirable number before contact with Europeans had changed their conditions. It is merely suggested that there was in general a tendency for such an approximation to occur and it may further be deduced that, since the desirable number remained the same over a great length of time, rendering the approximation, so to speak, easy of attainment, the adjustment to the optimum number normally came about. And in this connexion the limitation to clearly defined areas may be borne in mind. The very fact of the universal 238recognition and careful maintenance of these areas is an indication that such was the normal condition. Migration is abnormal, and this fact is apt to be forgotten for two reasons. First, when reviewing the course of history, migrations stand out as the chief, if not the only, known facts and the huge intervals of time between one movement and another are forgotten. Secondly, our knowledge of primitive races is largely derived from observations made when migration had been set on foot owing to contact with Europeans, as, for instance, in America. Migration may be a disturbing factor, upsetting the adjustment of numbers; but it is an abnormal condition and hence it is disregarded here.

And here, in answer to the objections that a greater prevalence of infanticide and so on is assumed than there is evidence for, we may, bearing in mind the many reasons why the evidence is deficient, reply that this assumption is not unreasonable. If such an assumption is not made, the position among these races is not comprehensible. It may be granted, for instance, that there is evidence of the practice of infanticide on a large scale among certain Australian tribes; but it may be pointed out that similar evidence is lacking in the case of other tribes. Allowing that a far more detailed examination of the evidence is desirable than there is space for here, especially with regard to the nature of the evidence and the date to which it refers, we may ask what it is supposed was going on among those tribes who did not, as it may be alleged, practise infanticide—abortion and abstention from intercourse being uncommon or unknown in Australia. The fecundity of all Australians is presumably very similar; it would be very remarkable indeed if it were not. The factors tending to lessen fertility and to produce elimination do not differ so very much from one tribe to another. It must follow that, if one tribe practised infanticide on a large scale and maintained its numbers nevertheless, another tribe which did not do so must have been rapidly increasing. This rapid increase is not compatible with the strict maintenance of well-defined territories and with all that we know of normal inter-tribal relations, and it is submitted that the assumption made here, namely that infanticide or some other custom was almost always practised, provides the only explanation of the position.^[1284]