Following the speculative, fictional post about clans and marriages at reference 1, Bing turned up the short book, paper rather, at reference 2, which is very much to the point: marriage and relationships in native Australian tribes. With my copy being derived from a copy held by the Anthropological part of the University of California, but downloaded from reference 3, on the other side of the world.
The author, John Hopkins, was a doctor in his day job, but was also something of an anthropologist and a sometime ordinary fellow of the Royal Anthropological Institute of reference 5.
Part of the interest here is the way in which section membership complements and interacts with the relationship vocabulary, the two combining to give elaborate rules about how two people may interact – or not – with each other. Section membership is simple in the sense that everyone belongs to a section – while relationships are relative: one man’s cousin might be another man’s grandfather. One might say that sections are one dimensional while relationships are two dimensional – and it may be that the former are more manageable and so more useful in an illiterate society.
The figure above is adapted from Hopkins’s type two (of three) on his page 7 and is said to describe the arrangements which underlie the complicated marriage arrangements of many of the (much studied) Aboriginal tribes of Australia. Type 1 has two sections, type 2 four and type 3 eight, with all three types being organised on comparable lines. I do not address what might be going on now, rather what was said to have been going on more than a hundred years ago.
In the case of type 2, we have a more or less closed population which is divided into four sections, called here Alpha, Bravo, Charlie and Delta. An endogamous population, although real populations are unlikely to be completely closed and are likely to have customs and rules about the assimilation (or not) of foreigners.
I note in passing that there is more or less no property in this population and certainly no land. Accumulation of property and its passing from one person to another is not the issue it is in the settled, agricultural population described at reference 13 – from which I associated to the old lefty slogan about property being theft.
Section membership is either passed through the father, that is to say patriarchal, or through the mother, that is to say matriarchal. Without loss of generality, for the moment we suppose matriarchal. Some might say a unilateral rather than a bilateral population.
The first marriage rule is that an Alpha person can only marry a Bravo person of the opposite sex and vice-versa. And that a Charlie person can only marry a Delta person of the opposite sex and vice-versa.
In the case that an Alpha female marries a Bravo male, we have Alpha dominance and their children, both male and female, are all Charlie. In the other case, we have Bravo dominance and their children are all Delta.
In the case that a Charlie female marries a Delta male, we have Charlie dominance and their children, both male and female, are all Alpha. In the other case, we have Delta dominance, the children are all Beta.
There is a strong preference for a man to marry one of his cross-cousins, that is to say either his mother’s brothers’ daughters or one of his fathers’ sisters’ daughters. An arrangement which, in some tribes, extends to one’s maternal uncles being more important in one’s life than one’s father.
Reciprocal marriages by two men of paired sections marrying each-others’ sisters are common. One gift is reciprocated by another, matching gift.
An arrangement which our rule allows, while disqualifying parallel cousins.
It also disqualifies parents and children, and others of those generations. Indeed, within a small group at least, the sections stratify the group by age to this extent.
That said, this does not of itself disqualify one from marrying someone of one’s grandparents’ or grandchildren’s generations. But a lot of the time the age difference will be such that one would not think of doing such a thing.
Unfortunately, wide variation in birth intervals complicate matters, as does the widespread custom for Aboriginal men to marry women who are ten to fifteen years younger than themselves. This last being addressed by reference 6.
Hopkins allows the degenerate case in which sections Alpha and Charlie are conflated and sections Bravo and Delta are conflated, his type 1. Here the cross-cousin preference still works in that if the man is Alpha, this makes his mother an Alpha, his mother’s brothers Alphas and his mother’s brothers’ daughters, that is to say his potential wives, Bravos. Or, his father a Bravo, his father’s sisters Bravos and his fathers’ sisters’ daughters, that is to say the other group of potential wives, Bravos.
He also allows the extended case where we have more than two pairs of sections. This possibility can be seen in the ladders which follows. Roughly speaking, the more sections, the more information about ancestry is coded into one’s section membership. On the other hand, having lots of sections starts to strain what is practical in a world which was illiterate, without written records.
The ladder
The ladder is another way of looking at this; a ladder on its side in the snap above with time and generations going from left to right. We now switch from matriarchal to patriarchal.
So in this example, we suppose (a quite unrealistic) patrilineal society in which each marriage produces one child, with an equal chance of that child being male or female. Unrealistic, but it does serve to show the way it works.
In the first generation, we have an Alpha male, Alpha-1. He produces a Charlie son, Charlie-2. That son produces another Alpha son, Alpha-3. But Alpha-3 fails and he produces a daughter who has to marry a Delta male, Delta-4. They produce a son, Bravo-5. But he fails too and produces a daughter who has to marry a Charlie male, Charlie-6. And so on. We have a random walk along the ladder, with the ‘blood line’ circulating through the four sections.
While in this unusual case, the male line is preserved, unsullied, right along the ladder. Odds of near 32:1 against.
Age breakdown
An example prompted by my time in the Berlin of the Hohenzollerns noticed at reference 9.
[lifted Wikipedia]
An example which shows that, even when one restricts oneself to the male descent of Frederick I, Count of (Hohen)Zollern, Burgrave of Nurenburg – and probably not all of that – the page can get pretty crowded. Hohenzollern being the name of a mountain in the Swabian Alps with a castle on top.
The upper double headed arrow shows a pair of second cousins, of the same genealogical generation, but whose chronological generation has drifted apart. The lower one shows the daughter of the one on the right marrying the great grandson of the one on the left – in which one needs to take care with the interpretation of the broken and stepped black line. Probably of the same chronological generation, but certainly not of the same genealogical generation. Possibly, a second cousin twice removed.
Demonstrating that a marriage rule based on just genealogical generation might sometimes be inconvenient.
Going too far
The chart just above, about the Hohenzollern family, is just about manageable; one can work out what is going on without too much trouble. While Ingold, at reference 12 contains a chart drawn from reference 13 which, to my mind, tries to do too much. Ingold likens it to a diagram of a circuit board
[Lifted from a pdf photocopy of reference 12. With the original to be found as a fold-out at the back of reference 13, a book about kinship and land tenure in a Sri Lankan village, already noticed at reference 14]
Here, in a context where land and house tenure is of interest, we have most of the population of a small village, men women and children, dead and alive. The dead being included because they inform the alive.
It took me a while to work out the conventions here, particularly those for identifiers for individuals, but for present purpose it probably suffices to know that white is dead, black is alive, triangles are males and circles are females. Letters are about compounds, with the (fenced) compounds containing one or more houses or households.
My present point is that this chart, while a useful reference document, is a bit much most of the time, with the author including much simpler extracts in the text proper. And, by extension, keeping track of all this in an illiterate society is a bit much. Something is needed which is a bit simpler.
Noting that the illiterate is given some help in the Hohenzollern chart by the inclusion of identifying heraldic shields – shields which once served a similar secondary purpose in battles, when they were for real, rather than just for decoration or illustration.
Relationships
At reference 2, it is argued that the Aboriginal classes do a lot of the work that relationships like mother-in-law, aunt and uncle do in our world. And they do more work the more classes you have. Nevertheless, class does need to be supplemented by generation and age-related rules, at least for ego’s immediate family.
Or perhaps complemented would be a better word, as Aboriginal societies – as described, for example, at reference 11 – have quite elaborate vocabularies for describing relationships, vocabularies which carve the family world up rather differently from ourselves. Some of this can be seen in the snap above. Complete with the complication that while class membership is absolute, everybody sees class membership in the same way, relationships are relative, relative to the person usually called ‘ego’ by this generation of anthropologists. Which, presumably, makes it all much harder to keep track of – particularly so given the absence of written records or even diagrams.
An Excel worksheet
For any given population one might arrange relationship information as a (sparse) square array – or almost as a square array as it soon proved helpful to know the sex and age of the people involved. Maybe the marital status of demographers will come. And most societies go in for personal names in addition to section names.
And if one added a column for section one could look at the correlation between relationships and section.
Within the square part of the worksheet one has rules. So, for example, husband and wife are paired, father and son are paired. One might allow a man to have several wives but a woman to have at most one husband. All expressed in terms of a function R(x, y). In terms of the snap above ‘R(#001,#002)=son’ says that #001 is the son of #002. We allow the value ‘null’ for the common case that there is no named relationship between two people. We also have ‘R(a, b)=null’ if and only if ‘R(b, a)=null’.
By convention, we allow ‘ego’ for the diagonal entries, where we might otherwise have had ‘null’.
Our starting assumption is that only one relationship term applies to any one cell – which may not be appropriate for some societies. This would not be true if, for example, we wanted all of sibling, brother and sister. Or if we wanted all of elder brother, younger brother and brother.
One might go further, defining relationships in terms of a small number of basic functions and relations and then using some kind of an inference engine to either validate or populate the worksheet from lower grade input material. One could probably do everything using two functions – S(x) to give the sex of person ‘x’ and A(x) to give the age of a person – one relation P(x, y) which said whether or not ‘x’ was a parent of ‘y’ and some axioms, axioms which, for example stopped a person having two fathers and stopped loops, that is to say a person being an ancestor of him or herself. One might achieve this last by a rule which said that a parent had to be older than its child.
While such a worksheet contains a lot of the information that one might need, I do not recall ever seeing such a thing before. So maybe not such a convenient way of marshalling one’s information. Would Leach have used Excel on his laptop in his hut at Pul Eliya (reference 13), had that been an option in the middle of the 20th century?
Conclusions
An excursion into what can be done with relatively simple section rules – aka clan, aka totem – by comparison with the complications of full-on genealogy. With rules of this general sort said to have been common in Australia at the end of the 19th century,
I started with a rule which does the same sort of thing as that invented at reference 1 – but this one has more currency in the real world, at least in the recent past. There is some interesting material how two real worlds might interact at reference 8 – and I dare say one could dig up something similar for India when it was part of our Empire. Maybe also for the status of First American marriages in today’s USA. I associate to the complications arising in the white lives resumed of white ladies who spent serious time, perhaps years, in First American families. Some of this is touched on at reference 15 and in the further references therein. Noting in passing that a copy of ‘Six Weeks in the Sioux Tepees – Sarah Wakefield – 1863’ is something else to be found at the Internet Archive.
I came away with a sense that these complex social systems evolved to meet two conflicting objectives. One wanted the cohesion – the peace and prosperity – which one got from binding the group together with marriage (endogamy), while maintaining genetic diversity by mixing (exogamy) – which last can be glossed as avoiding incest – however this might be defined. A similar challenge to that faced by animal breeders today – say the breeders of dogs, horses and cattle, with all three groups going in for elaborate records of ancestry. Some of this is to be found at reference 7.
The existence of the four stable, long-life sections in my opening example, Alpha, Bravo, Charlie and Delta, provide some permanence in the very uncertain world of the Australian Aboriginals. Useful pegs on which all kinds of other stuff can be hung.
With a secondary objective being to give the men something to talk about, to argue about and to strut about at their gatherings or in their lodges.
Contrariwise, the basic relations of marriage, despite their apparent simplicity, generate a complexity of relationships which defy comprehensive yet digestible presentation in two dimensions, in a simple diagram, never mind capture in oral recitation and ritual. The simplicity and elegance of the opening diagram has been lost.
References
Reference 1: https://psmv6.blogspot.com/2026/03/clans-and-marriages.html.
Reference 2: The family chain: Marriage and relationships of native Australian tribes – John Hopkins – 1914.
Reference 3: https://aiatsis.gov.au/. The Australian Institute of Aboriginal and Torres Strait Islander Studies.
Reference 4: https://historywiki.therai.org.uk/index.php?title=John_Hopkins. A website which can be slow to respond.
Reference 5: https://therai.org.uk/.
Reference 6: Kinship, marriage and age in aboriginal Australia – Woodrow W. Denham – 2012.
Reference 7: Coefficients of Inbreeding and Relationship – Wright, Sewall – 1922.
Reference 9: https://psmv6.blogspot.com/2026/04/berlin.html.
Reference 10: https://en.wikipedia.org/wiki/House_of_Hohenzollern. Which shows that it is not just Australians who get excited by this sort of thing.
Reference 11: The native tribes of central Australia – Spencer, Gillen – 1899.
Reference 12: Lines: A Brief History – Ingold, T – 2007.
Reference 13: Pul Eliya: A village in Ceylon. A study of land tenure and kinship – Leach E R – 1961.
Reference 14: https://psmv6.blogspot.com/2026/04/legal-bytes.html.
Reference 15: https://psmv5.blogspot.com/2023/05/life-in-minnesota.html.
No comments:
Post a Comment