April 21, 1 88 1 J 



NA TURE 



585 



them. His coadjutor, Mr. Howitt, though he has had 

 some interesting information to give about the Kurnai 

 tribes of Gippsland, has had the same chief object ; so 

 that the worlc the two have produced is much more a 

 polemic on behalf of Mr. Morgan than a record of new 

 Australian facts. We must begin, then, by statmg what 

 Mr. Morgan's theories are (so far as the work before us 

 is concerned with them), and indicating, and estimating 

 the value of, the evidence on which they rest. 



Mr. Morgan, having collected a great mass of facts 

 concerning the terms in use between relations and con- 

 nections throughout the world, and having found that 

 those terms were, broadly speaking, divided into three 

 orders, proceeded to spell out of the two earlier orders 

 (the third consists of the modern terms of consanguinity 

 and affinity) the whole of the early history of marriage 

 and of the family. In what he has called the Malayan 

 system of relationships, parent and child, grand-parent 

 and grandchild, and brother and sister (or rather elder 

 brother, younger brother, elder sister, younger sister, for 

 there are no words for brother and sister) are the only 

 terms in use ; and one or other of these terms is used in 

 addressing a person, according as the person addressed 

 is of the speaker's generation or of the generation above, 

 or of that below it. They are the terms always used 

 when persons address one another, there being among 

 those who use the system an invincible objection to the 

 mention of their personal names. Mr. Morgan assumed 

 that those terms were expressive of consanguinity and 

 affinity ; and conjectured that when first used they accu- 

 rately described the relationships at the time existing, 

 "as near as the parentage of children could be known." 

 -And it appeared to him that if there were a body of men 

 and a body of women in the same tribe who all regarded 

 each other as brothers and sisters, and all the men 

 married all the women in a group, there would exist a 

 marriage and family system which would e.xplain the 

 Malayan terms — the relationships arising out of which, 

 so far as they were ascertainable, "as near as the 

 parentage of children could be known," those terms 

 would accurately express. 



Accordingly, he framed the hypothesis that the first 

 stage of marriage was the marriage in a group of men 

 and women of the same blood calling themselves brothers 

 and sisters. The family founded upon this kind of 

 marriage he has named the consanguine family, and he 

 regards it as the earliest form of the family. He does 

 not say that such a system of marriage, or such a family 

 system as he has supposed, has been found at any time 

 anywhere ; what he says is that this supposition of his 

 explains the origin of the Malayan terms, and that 

 nothing else can explain them. But does it explain 

 them? It is at once obvious that there is one term, and 

 that the most important of all, the use of which Mr. 

 Morgan's hypothesis does not account for. Paternity 

 may be doubtful — and if it were thought of at all in a 

 group such as Mr. Morgan has conceived of, any man of 

 the group might have as good a right as any other to be 

 called father of any chill born within it. But there can 

 be no doubt about a man's relationship to his mother. 

 In the case of mother and child the parentage is known 

 with certainty, and therefore, on Mr. Morgan's hypothesis, 

 a man should in the Malayan system have had only one 

 mother. Now that system applies the term for mother 

 to many women besides the actual mother — mother's 

 sisters, father's sisters, uncle's wives, and so on, if not 

 indeed to all women of the mother's generation. Here 

 then the hypothesis breaks down ; and the point at 

 which we find it breaking down is really the only point 

 at which it can be tested. The relationship between 

 mother and child, too, which is confused or ignored in 

 the Malayan system, is the pne relationship or which it 

 can be said with confidence that no system really founded 

 on relationship could fail to recognise it. The explana- 



tion ]\Ir. Morgan offers is that in the Malayan system the 

 relationship of stepmother " is not discriminated,'' and 

 there being no name for stepmother, stepmothers had to be 

 called mothers, because " otherwise they would fall with- 

 out the system." And he has what may be called a 

 subsidiary hypothesis to account for there having been 

 no discrimination between stepmother and mother. It is 

 that the affiliation of children to the groups of men and 

 of women to which they belonged would be so strong 

 "that the distinction between relationships by blood 

 and by affinity would not be recognised in every case." 

 The fact of motherhood would be made little of, that is 

 — there would be no discrimination between stepmother 

 and mother— because the whole group would be, by a 

 child, regarded as its mother. But this is equivalent to 

 saying that, from the nature of the case, it was not to be 

 expected that note should be taken of the relationships that 

 could be known ; and that is to abandon the hypothesis— 

 as well as to deny us all chance of judging whether it is 

 a good or a bad one. Possibly explanations of the 

 failure of his hypothesis, such as Mr. Morgan suggests, 

 might have some weight were he accounting tor the 

 Malayan terms as terms of address ; but he takes them 

 to denote actual relationships "as near as the parentage 

 of children could be known." And no explanations can 

 get over the fact that the Malayan terms are equally 

 extensive in their application where, in the consanguine 

 family, parentage would be known with certainty, and 

 where it would not be known at all. The consanguine 

 family is clearly a bad hypothesis. It might be thought 

 it would hardly seem to anybody a plausible one ; but 

 Mr. Morgan always speaks of it as if it were among the 

 best vouched of historical facts ; and we are bound to 

 say that ?ilr. Howitt beUeves in it as imphcitly as Mr. 

 Morgan. 



To show the hypothesis of the consanguine family to 

 be unstateable is to undermine Mr. Morgan's whole 

 history of marriage and of the family. But Mr. Morgan 

 has propounded a hypothesis as to the second form of 

 marriage and the second form of the family, and as it is 

 at this point that Mr. Fison (who does not quite believe 

 in the consaiguine fainily) lends him his advocacy, it is 

 indispensable that we should give some account of it. 

 Punaluan marriage, upon which was founded the punaluan 

 family, was introduced by some reformatory movement, 

 according to Mr. Morgan, to put a stop to the evils 

 attendant upon brother and sister marriages. It existed 

 in two forms. In one form of it a group of men, brothers 

 or reputed brothers, had in common their wives who 

 were not their sisters and ; ot the sisters of each other ; 

 in the other form, a group of women, sisters or reputed 

 sisters, lived in common with husbands who were not 

 their brothers and not the brothers of each other. 

 Punaluan marriage has not been obser\'ed at any time 

 anywhere any more than the cons.nguine family; but 

 Mr. Morgan believes that, in 'both its forms, it has 

 existed everywhere, and probably during many ages. 

 A correspondent wrote to Mr. Morgan stating that m 

 the Sandwich Islands men whose wives were sisters and 

 women whose husbands were brothers called each other 

 punalua, which meant dear friend or intimate companion. 

 And possibly drawing his bow at a venture, "the rela- 

 tionship," he said, " is rather amphibious. It arose from 

 the fact that two or more brothers with their wives, or 

 two or more sisters with their husbands, were inclined to 

 possess each other in common." Whether conjecture or 

 fact, this amounts to very little ; but it was this which 

 gave Mr. Morgan the suggestion of punaluan marriage. 

 For proof of his hypothesis he again relied upon thetertirs 

 he had collected— and at first upon its fitness to explain 

 those same Malayan terms which, as we have seen, have 

 more than enough to do to bear the weight of the con- 

 sanguine family. In his latest work (" Ancien. -Society ) 

 he holds it to be proved by a nomenclature considerably 



