Ancient Society. Lewis H. Morgan 1877

A Note
Mr. J. F. McLennan’s “Primitive Marriage.”

As these pages are passing through the press, I have obtained an enlarged edition of the above-named work. It is a reprint of the original, with several Essays appended; and is now styled “Studies in Ancient History Comprising a Reprint of Primitive Marriage.”

In one of these Essays, entitled “The Classificatory System of Relationships,” Mr. McLennan devotes one section (41 pages) to an attempted refutation of my explanation of the origin of the classificatory system; and another (36. pages) to an explanation of his own of the origin of the same system. The hypothesis first referred to is contained in my work on the “Systems of Consanguinity and Affinity of the Human Family” (pp. 479- 486). The facts and their explanation are the same, substantially, as those presented in preceding chapters of this volume (Chaps. II, and III, Part III). “Primitive Marriage” was first published in 1865, and “Systems of Consanguinity,” etc., in 1871

Having collected the facts which established the existence of the classificatory system of consanguinity, I ventured to submit with the Tables, an hypothesis explanatory of its origin. That hypotheses are useful, and often indispensable to the attainment of truth, will not be questioned. The validity, of the solution presented in that work, and repeated in this, will depend upon its sufficiency in explaining all the facts of the case. Until it is superseded by one better entitled to acceptance on this ground, its position in my work is legitimate, and in accordance with the method of scientific inquiry.

Mr. McLennan has criticised this hypothesis with great freedom. His conclusion is stated generally as follows (Studies, etc. p, 371): “The space I have devoted to the consideration of the solution may seem disproportioned to its importance; but issuing from the press of the Smithsonian Institution, and its preparation having been aided by the United States Government, Mr. Morgan’s work has been very generally quoted as a work of authority, and it seemed worth while to take the trouble necessary to show its utterly unscientific character.” Not the hypothesis alone, but the entire work is covered; by the charge.

That work contains 187 pages of “Tables of Consanguinity and Affinity,” exhibiting the systems of 139 tribes and nations of mankind representing four-fifths, numerically, of the entire human family. It is singular that the bare facts of consanguinity and affinity expressed by terms of relationship, even when placed in tabular form, should possess an “utterly unscientific character.” The body of the work is taken up with the dry details of these several systems. There remains a final chapter, consisting of 43 out of 590 pages, devoted to a comparison of these several systems of consanguinity, in which this solution or hypothesis appears. It was the first discussion of a large mass of new material, and had, Mr. McLennan’s charge been limited to this chapter, there would have been little need of a discussion here. But he has directed his main attack against the Tables; denying that the systems they exhibit are systems of consanguinity and affinity, thus going to the bottom of the subject.[1]

Mr. McLennan’s position finds an explanation in the fact, that as systems of consanguinity and affinity they antagonize and refute the principal opinions and the principal theories pro- pounded in “Primitive Marriage.” The author of “Primitive Marriage” would be expected to stand by his preconceived opinions.

As systems of consanguinity, for example: (1) They show that Mr. McLennan’s new terms, “Exogamy and Endogamy” are of questionable utility — that as used in “Primitive Marriage” their positions are reversed, and that “endogamy” has very little application to the facts treated in that work while “exogamy” is simply a rule of a gens, and should be stated as such. (2) They refute Mr. McLennan’s phrase, “kinship through females only,” by showing that kinship through males was recognized as constantly as kinship through females by the same people. (3) They show that the Nair and Tibetan polyandry could never have been general in the tribes of mankind. (4) They deny both the necessity and the extent of “wife stealing” as propounded in “Primitive Marriage.”

An examination of the grounds, upon which Mr. McLennan’s charge is made, will show not only the failure of his criticisms but the insufficiency of the theories on which these criticisms are based. Such an examination leads to results disastrous to his entire work, as will be made evident by the discussion of the following propositions, namely:

I. That the principal terms and theories employed in “Primitive Marriage” have no value in Ethnology.

II That Mr. McLennan’s hypothesis to account for the origin of the classificatory system of relationship does not account far its origin.

III. That Mr. McLennan’s objections to the hypothesis presented in “Systems of Consanguinity,” etc., are of no force. These propositions will be considered in the order named;

I. That the principal terms and theories employed in “Primitive Marriage” have no value in Ethnology.

When this work appeared it was received with favour by ethnologists, because as a speculative treatise it, touched a number of questions upon which they had long been working. A careful reading, however, disclosed deficiencies in definitions, unwarranted assumptions, crude speculations and erroneous conclusions. Mr. Herbert Spencer in his “Principles of Sociology” (Advance Sheets, Popular Science Monthly, Jan., 1877, p;.272), has pointed out a number of them, At the same time he rejects the larger part of Mr. McLennan’s theories respecting “Female Infanticide,” “Wife Stealing,” and “Exogamy and Endogamy.” What he leaves of this work, beyond its collocation of certain ethnological facts, it is difficult to find.

It will be sufficient under this head to consider three points.

l. Mr. McLennan’s use of the terms “Exogamy” and “Endogamy.”

“Exogamy” and “endogamy” — terms of his own coinage — imply, respectively, an obligation to “marry out,” and an obligation to “marry in,” a particular group of persons.

These terms are applied so loosely and so imprecisely by Mr. McLennan to the organized groups made known to him by the authors he cites, that both his terms and his conclusions are of little value. It is a fundamental difficulty with “Primitive Marriage” that the gens and the tribe, or the groups they represent, are not distinguished from each other as members of an organic series, so that it might be known of which group “exogamy” or “endogamy” is asserted. One of eight gentes of a tribe, for example, may be “exogamous” with respect to itself, and “endogamous” with respect to the seven remaining gentes. Moreover, these terms, in such a case, if correctly applied, are misleading. Mr. McLennan seems to be presenting two great principles, representing distinct conditions of society which have influenced human affairs. In point of fact, while “endogamy” has very little application to conditions of society treated in “Primitive Marriage,” “exogamy” has reference to a rule or law of a gens — an institution — and as such the unit of organization of a social system. It is the gens that has influenced human affairs, and which is the primary fact. We are at once concerned to know its functions and attributes with the rights, privileges and obligations of its members. Of these material circumstances Mr. McLennan makes on account, nor does he seem to have had the slightest conception of the gens as a governing institution of ancient society. Two of its rules are the following: (1) Intermarriage in the gens is prohibited. This is Mr. McLennan’s “exogamy” — restricted as it always is to a gens, but stated by him without any reference to a gens. (2) In the archaic form of the gens descent is limited to the female line, which is Mr. McLennan’s “kinship through females only,” and which is also stated by him without any reference to a gens.

Let us follow this matter further. Seven definitions of tribal system, and of tribe are given (Studies, etc. 113-115).

“Exogamy Pure. — 1. Tribal (or family) system.- Tribes separate. A11 the members of each tribe of the same blood or feigning themselves to be so. Marriage prohibited between the members of the tribe.

“2. Tribal system.- Tribe a congeries of family groups, falling into divisions clans, thums, etc. No connubium between members of same division: connubium between all the divisions.

3. Tribal system.- Tribe a congeries of family groups.

* * * No connubium between persons whose family name points them out as being of the same stock.

“4. Tribal system.- Tribe in divisions. No connubium between members of the same divisions-: connubium between some of the divisions; only partial connubium between others. * * *

“5. Tribal system.- Tribe into divisions. No connubium between persons of the same stock: connubium between each division and some other. No connubium between some of the divisions. Caste.

“Endogamy Pure. — 6. Tribal (or family) system.- Tribes separate. A11 the members of each tribe of the same blood, or feigning themselves to be so. Connubium between members of the tribe: marriage without the tribe forbidden and punished.

“7. Tribal system indistinct. — Seven dentitions of the tribal system ought to define the group called a tribe, with sufficient distinctness to be recognized. The first definition, however, is a puzzle. There are several tribes in a tribal system, but no term for the aggregate of tribes. They are not supposed to form a united body. How the separate tribes fall into a tribal system or are held together does not appear. All the members of each tribe are of the same blood, or pretend to be, and therefore cannot intermarry. This might answer for a description of a gens; but the gens is never found alone, separate from other gentes. There are several gentes intermingled by marriage in every tribe composed: of gentes. But Mr. McLennan could not have used tribe here as equivalent to gens, nor as a congeries of family groups. As separate bodies of consanguinei held together in a tribal system, the bodies undefined and the system unexplained, we are offered something altogether new. Definition is much the same. It is not probable that a tribe answering to either of these definitions ever existed in any part of the earth; for it is neither a gens, nor a tribe composed of gentes, nor a nation formed by the coalescence of tribes.

Definitions 2nd, 3rd, 4th, and 5th are somewhat more intelligible. They show in each case a tribe composed of gentes, or divisions based upon kin. But it is a gentile rather than a tribal system. As marriage is allowed between the clans, thums or divisions of the same tribe, “exogamy” cannot be asserted of the tribe in either case. The clan, thum, or division is “exogamous,” with respect to itself, but “endogamous” with respect to the other clans, thums, or divisions. Particular restrictions are stated to exist in some instances.

When Mr. McLennan applies the terms “exogamy” or “endogamy” to a tribe, how is it to be known whether it is one of several separate tribes in a tribal system, whatever this map mean or a tribe defined as a congeries of family groups? On the next page (116) he remarks: “The separate endogamous tribes are nearly as numerous, and they are in some respects as rude, as the separate exogamous tribes.” If he uses tribe as a congeries of family groups, which is a tribe composed of gentes, then “exogamy” cannot be asserted of the tribe. There is not the slightest probability that “exogamy” ever existed in a tribe composed of gentes in any part of the earth. Wherever the gentile organization has been found intermarriage, in the gens is forbidden. It gives what Mr. McLennan calls “exogamy.” But, as an equally general rule, intermarriage between the members of a gens and the members of all the other gentes of the same tribe is permitted. The gens is “exogamous,” and the tribe is essentially “endogamous.” In these cases, if in no others, it was material to know the group covered by the word tribe. Take another illustration (p. 42): “If it can be shown, firstly, that exogamous tribes exist, or have existed; and, secondly, that in ruder times the relations of separate tribes were uniformly, or almost uniformly, hostile, we have found a set of circumstances in which men could get wives only by capturing them.” Here we find the initial point of Mr. McLennan’s theory of wife stealing. To make the “set of circumstances” (namely, hostile, and therefore independent tribes), tribe as used here must refer to the larger group, a tribe composed of gentes. For the members of the several gentes of a tribe are intermingled by marriage in every family throughout the area occupied by the tribe. A11 the gentes must be hostile or none. If the term is applied to the smaller group, the gens, then the gens is “exogamous,” and the tribe, in the given case, is seven-eighths “endogamous” and what becomes of the “set of circumstances” necessitating wife-stealing?

The principal cases cited in “Primitive Marriage” to prove “exogamy” are the Khonds, Kalmucks, Circassians, Yurak Samoyeds, certain tribes of India and Australia, and certain Indian tribes of America, the Iroquois among the number (p.p. 75-100). The American tribes are generally composed of gentes. A man cannot marry a woman of the same gens with himself; but he may marry a woman of any other gens of his own tribe. For example, a man of the Wolf gens of the Seneca tribe of the Iroquois is prohibited from marrying a woman of the same gens, not only in the Seneca tribe, but also in either of the five remaining Iroquois tribes. Here we have Mr. McLennan’s “exogamy,” but restricted, as it always is, to the gens of the individual. But a man may marry a woman in either of the seven remaining Seneca gentes. Here we have “endogamy” in the tribe, practiced by the members of each gens in the seven remaining Seneca gentes. Both practices exist side by side at the same time, in the same tribe, and have so existed from time immemorial The same fact is true of the American Indian tribes in general. They are cited, nevertheless, by Mr. McLennan, as examples of “exogamous tribes”; and thus enter into the basis of his theories.

With respect to “endogamy,” Mr. McLennan would probably refrain from using it in the above case: firstly, because “exogamy” and “endogamy” fail here to represent two opposite principles as they exist in his imagination; and, secondly, because there is, in reality, but one fact to be indicated, namely, that intermarriage in the gens is prohibited. American Indians generally can marry in their own or in a foreign tribe as they please, hut not in their gens. Mr. McLennan was able to cite one fair case of “endogamy,” that of the Mantchu Tartars (p. 116), “who prohibited marriage between persons whose family names are different.” A few other similar cases have been found among existing tribes.

If the organizations, for example, of the Yurak Samoyeds of Siberia (82) the Magars of Nepaul (83), the Munnieporees, Koupooees, Mows, Muram and Murring tribes of India (87), were examined upon the original evidence, it is highly probable that they would be found exactly analogous to the Iroquois tribes; the “divisions” and “thums” being gentes. Latham speaking of the Yurak or Kasovo group of the Samoyeds, quotes from Klaproth, as follows: “This division of the kinsmanship is so rigidly observed that no Samoyed takes a wife from the kinsmanship to which he himself belongs. On the contrary he seeks her in one of the other two."[2] The same author, speaking of the Magars, remarks: “There are twelve thums. All individuals belonging to the same thum are supposed to be descended from the same male ancestor; descent from the same great mother being by no means necessary. So husband and wife must belong to different thums. With one and the same there is no marriage. Do you wish for a wife? If so, look to the thum of your neighbour; at any rate look beyond your own. This is the first time I have had occasion to mention this practice. It will not be the last: on the contrary, the principle it suggests is so common as to be almost universal."[3] The Murring and other tribes of India are in divisions, with the same rule in respect to marriage. In these cases it is probable that we have tribes composed of gentes, with intermarriage in the gens prohibited. Each gens is “exogamous” with respect to itself, and “endogamous” with respect to the remaining gentes of the tribe. They are cited by Mr. McLennan, nevertheless, as examples of “exogamous” tribes. The principal Australian tribes are known to be organized in gentes, with intermarriage in the gens prohibited. Here again the gens is “exogamous” and the tribe “endogamous.”

Where the gens is “exogamous” with respect to itself, and “endogamous” with respect to the remaining gentes of the same tribe of what use is this pair of terms to mark what is but a single fact — the prohibition of intermarriage in the gens? “Exogamy” and “endogamy” are ok no value as a pair of terms, pretending as they do to represent or express opposite conditions of society. They have no application in American ethnology, and probably none in Asiatic or European. “Exogamy,” standing alone and applied to the small group (the gens), of which “exogamous” tribes in America, but a plenty of “exogamous” only it can be asserted, might be tolerated. There are no “exogamous” tribes in America, but a plenty of “exogamous” gentes; and when the gens is found, we are concerned with its rules and these should always be stated as rules of a gens. Mr. McLennan found, the clan, thum, division, “exogamous,” and the aggregate of clans, thums, divisions, “endogamous”; but he says nothing about the “endogamy.” Neither does he say the clan, division or thum is “exogamous,” but that the tribe is “exogamous.” We might suppose he intended to use tribe as equivalent to clan, thum, and division; but we are met with the difficulty that he defines a “tribe [as] a congeries of family groups, falling into divisions, clans, thums, etc.” (114), and immediately (116) he remarks that “the separate endogamous tribes are nearly as numerous, and they are in some respects as rude, as the separate exogamous tribes.” If we take his principal definitions, it can be said without fear of contradiction that Mr. McLennan has not produced a single case of an “exogamous” tribe in his volume.

There is another objection to this pair of terms. They are set over against each other to indicate opposite and dissimilar conditions of society. Which of the two is the ruder, and which the more advanced? Abundant cautions are here thrown out by Mr. McLennan. “They may represent a progression. from exogamy to endogamy, or from endogamy to exogamy” (115); “they may be equally archaic” (116); and “they are in some respects” equally rude (116); but before the discussion ends, “endogamy” rises to the superior position, and stands over toward civilization, while “exogamy” falls back in the direction of savagery. It became convenient in Mr. McLennan’s speculations for “exogamy” to introduce heterogeneity, which “endogamy” is employed to expel, and bring in homogeneity; so that “endogamy” finally gets the better of “exogamy” as an influence for progress.

One of Mr. McLennan’s mistakes was his reversal of the positions of these terms. what he calls “endogamy” precedes “exogamy” in the order of human progress, and belongs to the lowest condition of mankind. Ascending to the time when the Malayan system. of consanguinity was formed, and which preceded the gens, we find consanguine groups in the marriage relation. The system of consanguinity indicates both the fact and the character of the groups and exhibits “endogamy” in its pristine force. Advancing from this state of things, the first check upon “endogamy” is found in the punaluan group, which sought to exclude own brothers and sisters from the marriage relation, while it retained in that relation first, second, and more remote cousins, still under the name of brothers and sisters. The same thing precisely is found in the Australian organization upon sex. Next in the order of time the gens appeared, with descent in the female line, and with intermarriage in the gens prohibited. It brought in Mr. McLennan’s “exogamy.” From this time onward “endogamy” may be dismissed as an influence upon human affairs. According to Mr. McLennan, “exogamy” fell into decay in advancing communities; and when descent was changed to the’ male line it disappeared in the Grecian and Roman tribes. (p. 220.) So far from this being the case, what he calls “exogamy” commenced in savagery with the gens, continued through barbarism, and remained into civilization. It existed as completely in the gentes of the Greeks and Romans in the time of Solon and of Servius Tullius as it now exists in the gentes of the Iroquois. “Exogamy” and endogamy” have been so thoroughly tainted by the manner of their use in “Primitive Marriage,” that the best disposition which can now be made of them is to lay them aside.

2. Mr. McLennan’s phrase: “The system of kinship through females only.”

“Primitive Marriage” is deeply coloured with this phrase. It asserts that this kinship, where it prevailed, was the only kinship recognized; and thus has an error written on its face. The Turanian, Ganowanian and Malayan systems of consanguinity show plainly and conclusively that kinship through males was recognized as constantly as kinship through females. A man had brothers and sisters, grandfathers and grandmothers, grandsons and granddaughters, traced through males as well as through females. The maternity of children was ascertain- able with certainty, while their paternity was not; but they did not reject kinship through males because of uncertainty, but gave the benefit of the doubt to a number of persons — probable fathers being placed in the category of real fathers, probable brothers in that of real brothers, and probable sons in that of real sons.

After the gens appeared, kinship through females had an increased importance, because it now signified gentile kin, as distinguished from non-gentile kin. This was the kinship, in a majority of cases, made known to Mr. McLennan by the authors he cites. The children of the female members of the gens remained within it, while the children of its male members were excluded. Every member of the gens traced: his or her descent through females exclusively when descent was in the female line and through males exclusively when descent was in the male line. Its members were an organized body of consanguinei bearing a common gentile name. They were bound together by affinities of blood, and by the further bond of mutual rights, privileges, and obligation. Gentile kin became, in both cases, superior to other kin; not because no other kin was recognized, but because it conferred the rights and privileges of a gens. Mr. McLennan’s failure to discover this difference indicates an insufficient investigation of the subject he was treating. With descent in the female line, a man had grandfathers and grandmothers, mothers, brothers and sisters, uncles, nephews and nieces, and grandsons and granddaughters in his gens; come own and some collateral; while he had the same out of his gens with the exception of uncles; and in addition, fathers, aunts, sons and daughters, and cousins. A woman had the same relatives in the gens as a man, and sons and daughters, in addition, while she had the same relatives out of the gens as a man. Whether in or out of the gens, a brother was recognized as a brother, a father as a father, a son as a son, and the same term was applied in either case without discrimination between them. Descent in the female line, which is all that “kinship through females only” can possibly indicate, is thus seen to be a rule of a gens and nothing more. It ought to be stated as such, because the gens is the primary fact, and gentile kinship is one of its attributes. Prior to the gentile organization, kinship through females was undoubtedly superior to kinship through males, and was doubtless the principal basis upon which the lower tribal groups were organized. But the body of facts treated in “Primitive Marriage” have little or no relation to that condition of mankind which existed prior to the gentile system.

3. There is no evidence of the general prevalence of the Nair and Tibetan polyandry.

These forms of polyandry are used. in Mr. McLennan’s speculations as though universal in practice. He employs them in his attempted explanation of the origin of the classificatory system of relationship. The Nair polyandry is where several unrelated persons have one wife in common (p. 146). It is called the rudest form. The Tibetan polyandry is where several brothers have one wife in common. He then makes a rapid flight through the tribes of mankind to show the general prevalence of one or the other of these forms of polyandry, and fails entirely to show their prevalence. It does not seem to have occurred to Mr. McLennan that these forms of polyandry are exceptional, and that they could not have been general even in the Neilgherry Hills or in Tibet. If an average of three men had one wife in common (twelve husbands to one wife was the Nair limit, P. 147), and this was general through a tribe, two-thirds of the marriageable females would, be without husbands. It may safely be asserted that such a state of things never existed generally in the tribes of mankind; and without better evidence it cannot be credited in the Neilgherry Hills or in Tibet. The facts in respect to the Nair polyandry are not fully known. “A Nair may be one in several combinations of husbands; that is, he may have any number of wives” (p. 148). This, however, would not help the unmarried females to husbands, although it would increase the number of husbands of one wife. Female infanticide cannot be sufficiently exaggerated to raise into general prevalence these forms of polyandry. Neither can it be said with truth that they have exercised a general influence upon human affairs.

The Malayan, Turanian and Ganowanian systems of consanguinity and affinity, however, bring to light forms of polygyny and polyandry which have influenced human affairs, because they were as universal in prevalence as these systems were, when they respectively came into existence. In the Malayan system, we find evidence of consanguine groups founded upon brother and sister marriages, but including collateral brothers and sisters in the group. Here the men lived in polygyny, and the women in polyandry. In the Turanian and Ganowanian system we find evidence of a more advanced group — the punaluan in two forms. One was founded on the brotherhood of the husbands, and the other on the sisterhood of the wives; own brothers and sisters being now excluded from the marriage relation. In each group the men were polygynous, and the women polyandrous. Both practices are found in the same group, and both are essential to an explanation of their system of consanguinity. The last-named system of consanguinity and affinity presupposes punaluan marriage in the group. This and the Malayan exhibit the forms of polygyny and polyandry with which ethnography is concerned; while the Nair and Tibetan forms of polyandry are not only insufficient to explain the systems, but are of no general importance. These systems of consanguinity and affinity, as they stand in the Tables, have committed such havoc with the theories and opinions advanced in “Primitive Marriage” that I am constrained to ascribe to this fact Mr. McLennan’s assault upon my hypothesis explanatory of their origin; and his attempt to substitute another, denying them to be systems of consanguinity and affinity.

II. That Mr. McLennan’s hypothesis to account for the origin of the classificatory system does not account for its origin. Mr. McLennan sets out with the statement (p. 372) that “the phenomena presented in all the forms [of the classificatory system) are ultimately referable to the marriage law; and that accordingly its origin must be so also.” This is the basis of my explanation; it is but partially that of his own.

The marriage-law, under which he attempts to explain the origin of the Malayan system, is that found in the Nair polyandry; and the marriage-law under which he attempts to explain the origin of the Turanian and Ganowanian system is that indicated by the Tibetan polyandry. But he has neither the Nair nor Tibetan system of consanguinity and affinity, with which to explain or to test his hypothesis. He starts, then, without any material from Nair or Tibetan sources, and with forms of marriage-law that never existed among the tribes and nations possessing the classificatory system of relationship. We thus find at the outset that the explanation in question is a mere random speculation.

Mr. McLennan denies that the systems in the Tables (Consanguinity, pp: 298-382; 523-567) are systems of consanguinity and affinity. On the contrary, he asserts that together they are “a system of modes of addressing persons.” He is not unequivocal in his denial, but the purport of his language is to that effect. In my work of Consanguinity I pointed out the fact that the American Indians in familiar intercourse and in formal salutation addressed each other by the exact relationship in which they stood to each other, and never by the personal name; and that the same usage prevailed in South India and in China. They use the system in salutation because it is a system of consanguinity and affinity — a reason paramount. Mr. McLennan wishes us to believe that these all- embracing systems were simply conventional, and formed to enable persons to address each other in salutation, and for no other purpose. It is a happy way of disposing of these systems, and, of throwing away the most remarkable record in existence respecting the early condition of mankind.

Mr. McLennan imagines there must have been a system of consanguinity somewhere entirely independent of the system of addresses; “for it seems reasonable to believe,” he remarks (p. 373), “that the system of blood-ties and the system of addresses would begin to grow up together, and for some little time would have a common history.” A system of blood-ties is a system of consanguinity. Where, then, is the lost system? Mr. McLennan neither produces it nor shows its existence. But I find he uses the systems in the Tables as systems of consanguinity and affinity, so far as they serve his hypothesis, without taking the trouble to modify the assertion that they are simply “modes of addressing persons.”

That savage and barbarous tribes the world over, and through untold ages, should have been so solicitous concerning the proper mode of addressing relations as to have produced the Malayan, Turanian and Ganowanian systems, in their fullness and complexity, for that purpose and no other, and no other systems than these two — that in Asia, Africa, Polynesia, and America they should have agreed, for example, that a given person’s grandfather’s brother should be addressed as grandfather that brothers older than one’s self should be addressed as eider brothers, and those younger as younger brothers, merely to provide a conventional mode of addressing relatives — are coincidences so remarkable and for so small a reason, that it will be quite sufficient for the author of this brilliant conception to believe it. A system of modes of addressing persons would be ephemeral, because all conventional usages are ephemeral. They would, also, of necessity, be as diverse as the races of mankind. But a system of consanguinity is a very different thing. Its relationships spring from the family and the marriage-law, and possess even greater permanence than the family itself, which advances while the system remains unchanged. These relationships expressed the actual facts of the social condition when the system was formed, and have had a daily importance in the life of mankind. Their uniformity over immense areas of the- earth, and their preservation through immense periods of time, are consequences of their connection with the marriage- law.

When the Malayan system of consanguinity was formed, it may be supposed that a mother could perceive that her own son and daughter stood to her in certain relationships that could be expressed by suitable terms; that her own mother and her mother’s own mother stood to her in certain other relationships; that the other children of her own mother stood to her in still other relationships; and that the children of her own daughter stood to her in still others — all of which might be expressed by suitable terms. It would give the beginning of a system of consanguinity founded upon obvious blood-ties. It. would lay the foundation of the five categories of relations in the Malayan system, and without any reference to marriage-law.

When marriage in the group and the consanguine family came in, of both of which the Malayan system affords evidence, the system would spread over the group upon the basis of these primary conceptions. With the intermarriage of brothers and sisters, own and collateral, in a group, the resulting system of consanguinity and affinity would be Malayan. Any hypothesis explanatory of the origin of the Malayan system must fail if these facts are ignored. Such a form of marriage and of the family would create the Malayan system. It would be a system of consanguinity and affinity from the beginning and explainable only as such.

If these views are correct, it will not be necessary to consider in detail the points of Mr. McLennan’s hypothesis, which is too obscure for a philosophical discussion, and utterly incapable of affording an explanation of the origin of these systems.

III. That Mr. McLennan’s objections to the hypothesis presented in “Systems of Consanguinity,” etc., are of no force. The same misapprehension of the facts, and the same con- fusion of ideas which mark his last Essay, also appear in this. Be does not hold distinct the relationships by consanguinity and those by marriage, when both exist between the same persons; and he makes mistakes in the relationships of the systems also.

It will not be necessary to follow step by step Mr. McLennan’s criticisms upon this hypothesis, some of which are verbal, others of which are distorted, and none of which touch the essence of the questions involved. The first proposition he attempts to refute is stated by him as follows: “The Malayan system of relationships is a system of blood-relationships. Mr. Morgan assumes this, and says nothing of the obstacles to making the assumption” (p. 342). It is in part a system of blood- relationships, and in part of marriage-relationships. The fact is patent. The relationships of father and mother, brother and sister, elder or younger, son and daughter, uncle and aunt, nephew and niece and cousin, grandfather and mother, grandson and daughter; and also of brother-in-law and sister-in-law, son- in-law and daughter-in-law, besides others, are given in the Tables and were before Mr. McLennan. These systems speak for themselves, and could say nothing else but that they are systems of consanguinity and affinity. Does Mr. McLennan suppose that the tribes named had a system other or different from that presented in the Tables? If he did, he was bound to produce it, or to establish the fact of its existence. He does neither.

Two or three of his special points may be considered. “And indeed,” he remarks (p. 346), “if a man is called the son of a woman who did not bear him, his being so called clearly defies explanation on the principle of natural descents. The reputed relationship is not, in that case, the one actually existing as near as the parentage of individuals could be known; and accordingly Mr. Morgan’s proposition is not made out.” On the face of the statement the question involved is not one of parentage, but of marriage-relationship. A man calls his mother’s sister his mother, and she calls him her son, although she did not bear him. This is the case in the Malayan, Turanian and Ganowanian systems. Whether we have consanguine or punaluan marriages, a man’s mother’s sister is the wife of his reputed father. She is his step-mother as near as our system furnishes an analogue; and among ourselves a step-mother is called mother, and she calls her step-son, son. It defies explanation, it is true, as a blood-relationship, which it does not pretend to be, but as a marriage-relationship, which it pretends to be, this is the explanation. The reasoning of Mr. McLennan is equally specious and equally faulty in a number of cases.

Passing from the Malayan to the Turanian system, he remarks (p. 354), “It follows from this that a man’s son and his sister’s daughter, while reputed brother and sister, would have been free, when the “tribal organization” had been established, to intermarry, for they belonged to different tribes of descent.” From this he branches out in an argument of two or three pages to prove that “Mr. Morgan’s reason, then, is insufficient.” If Mr. McLennan had studied the Turanian or the Ganowanian system of consanguinity with very moderate attention, he would have found that a man’s son and his sister’s daughter are not “reputed brother and sister.” On the contrary, they are cousins. This is one of the most obvious as well as important differences between the Malayan and Turanian systems, and the one which expresses the difference between the consanguine family of the Malayan, and the punaluan family of he Turanian system.

The general reader will hardly take the trouble necessary to master the details of these systems. Unless he can follow the relationships with ease and freedom, a discussion of the system will be a source of perplexity rather than of pleasure. Mr. McLennan uses the terms of relationship freely, but without, in all cases, using them correctly.

In another place (p. 360), Mr. McLennan attributes to me a distinction between marriage and cohabitation which I have not made; and follows it with a rhetorical flourish quite equal to the best in “Primitive Marriage.”

Finally, Mr. McLennan plants himself upon two alleged mistakes which vitiate, in his opinion, my explanation of the origin of the classificatory system. “In attempting to explain the origin of the classificatory system, Mr. Morgan made two radical mistakes. His first mistake was, that he did not steadily contemplate the main peculiarity of the system — its classification of the connected persons; that he did not seek the origin of the system in the origin of the classification” (p. 360). What is the difference in this case, between the system and the classification? The two mean the same thing, and cannot by any possibility be made to mean anything different. To seek the origin of one is to seek the origin of the other.

“The second mistake, or rather I should say error, was to have so lightly assumed the system to be a system of “blood- ties” (p. 361). There is no error here since the persons named in the Tables are descended from common ancestors, or connected by marriage with some one or more of them. They are the same persons who are described in the Table showing the Aryan, Semitic, and Uralian systems (Consanguinity, pp. 79-127). In each and all of these systems they are bound to each other in fact by consanguinity and affinity. In the latter each relationship is specialized; in the former they are classified in categories; but in all alike the ultimate basis is the same, namely actual consanguinity and affinity. Marriage in the group in the former and marriage between single pairs in the latter, produced the difference between them. In the Malayan, Turanian and Ganowanian systems, there is a solid basis for the blood-relationships they exhibit in the common descent of the persons; and for the marriage-relationships we must look to the form of marriage they indicate. Examination and comparison show that two distinct forms of marriage are requisite to explain the Malayan and Turanian systems; whence the application, as tests of consanguine marriage in one case, and a punaluan marriage in the other.

While the terms of relationship are constantly used in salutation, it is because they are terms of’ relationship that they are so used. Mr. McLennan’s attempt to turn them into conventional modes of addressing persons is futile. Although he lays great stress upon this view he makes no use of them as “modes of address” in attempting to explain their origin. So far as he makes any use of them he employs them strictly as terms of consanguinity and affinity. It was as impossible that “a system of modes of addressing persons” should have grown up independently of the system of consanguinity and affinity (p. 373), as that language should have grown up independently of the ideas it represents and expresses. What could have given to these terms their significance as used in addressing relatives, hut the relationship whether of consanguinity or affinity which they expressed? The mere want of a mode of addressing persons could never have given such stupendous systems, identical in minute details over immense sections of the earth.

Upon the essential difference between Mr. McLennan’s explanation of the origin of the classificatory system, and: the one presented in this volume — whether it is a system of modes of addressing persons, or a system of consanguinity and affinity — I am quite content to submit the question to the judgment of the reader.


1. “The Tables,” however, are the “main results” of this investigation. In their importance and value they reach beyond any present use of their contents the writer may be able to indicate.” — “Systems of Consanguinity,” etc., Smithsonian Contributions to Knowledge, vol. xvii, p., 8.

2. “Descriptive Ethnology"” London ed., 1859, i, 475.

3. Ib., i, 80.