A simplified chart of Latin kinship terms

At Language Log, a post asks whether many English speakers use the expression “brothers-in-law” to refer to men whose relationship is that their wives are sisters and “sisters-in-law” to refer to women whose relationship is that their husbands are brothers.  So would it be idiomatic to say that my wife and my brother’s wife  are one another’s sisters-in-law?  Commenters on that post have mentioned the poverty of English vocabulary in kinship terms as compared to other languages.  One linked to a Wiktionary article about the expression “co-mother-in-law,” an article which ends with sixteen examples of languages which do have words in widespread use that mean “the mother of one spouse, in relation to the parents of the other spouse.” 

For a long time it’s struck me as strange that English has so few kinship terms.   About 14 years ago, I was in graduate school and I read an article that was then already rather old, “What does Latin tell us about the Romans?” by Carl R. Trahman.  If you have access to JSTOR, here’s a link to Trahman’s article; if you don’t, you can go to the nearest research library, look up volume 67, number three of The Classical Journal (February/ March 1972,) and turn to pages 240-250.  Here’s one thing Latin told Trahman about the Romans:

Perhaps the most telling evidence, in the case of the Romans, that the vocabulary of a language will lead to an understanding of its users lies in the terminology of Latin for family relationships.  In English we are content to speak of “in-laws,” of “cousins once removed,” of “uncles on the father’s side.”  Latin has specific words for all of these and for dozens more such affinities.  Your great-great-grandmother is your abavia; your uncle on your father’s side is patruus, but on your mother’s side is avunculus.  Your mother-in-law is socrus.  The hated step-mother is noverca and the stepson privignus.  Does your husband have a sister?  The word is glos.  Does he have a brother?  The word is levir.  In such matters, the Romans truly had a word for it.  They actually possessed a word to denote the relationship of two women married to two brothers: they were ianitrices.  Now what is the significance of such precision?  It indicates the immense importance of the family in Roman life.  If we had no other testimony of this feeling for family, which can hardly be overstated, this amazingly rich terminology would be more than enough.  It is interesting that two of the phrases used for our word prejudice, for which as I have said Latin had no proper word, are iudicia iam facta domo (Cicero) and domo adlata opinio (Seneca.)  They suggest family councils at which policy was determined and the stand to be taken by the gens on public issues yet to be debated.  (page 244)

Writing in the early 1970s, Trahman devoted a fair bit of space to the Sapir-Whorf hypothesis, an idea that the structural limitations of a given language are in some way commensurate with the range of thoughts available to the speakers of that language.   In its most extreme form, the Sapir-Whorf hypothesis can be identified with the view that a people whose language lacks a word for a given concept must therefore lack that concept.   Trahman himself clearly does not go to that extreme.  In the passage quoted above, Trahman identifies the concept expressed by Latin phrases like iudicia iam facta domo and domo adlata opinio with the concept that we express by the single word prejudice.  So he believes that they had the concept, even though they lacked a specific word for it.  Trahman does seem to suggest that something like the Sapir-Whorf hypothesis was already familiar to the ancients.  He quotes the Roman poet Ennius who said that because he spoke Latin, Greek, and Oscan he had three hearts; Trahman elaborates, “The word he used was cor, which in his day meant not only ‘heart’ but ‘mind and soul’ as well.  So [Ennius had] it all, and it could not be improved upon.” 

So we can’t say that English speakers have a poorer set of concepts for family relationships than did Latin speakers just because we have so much poorer a vocabulary through which to express those concepts.  What we can say is that the Romans probably talked about those relationships more often than we do.  This isn’t surprising.  Most people in the developed world today live in nuclear family households and see members of their extended families only occasionally, so it isn’t especially likely on any given day that you will have to explain that someone is your spouse’s sibling’s sibling’s spouse.  If it takes several words and repeated case-endings to identify that person, you probably won’t lose much time over the course of a long life.  But in the ancient world it was more usual for several generations of a family to live under the same roof, grandparents and their siblings, parents and their siblings, one’s own siblings and their spouses and children and grandchildren, one’s own spouse and children and grandchildren, etc etc, and to spend all day working side by side with other members of that population.  So of course you would need single words that could express those relationships quickly and easily.  Not only might it become tiresome to have to speak a lot of words every time you had to clarify a family relationship, but it would certainly be taxing to have to listen to a lot of convoluted phrases connecting one kinship title to another.  If you tell me that some person is your spouse’s sibling’s sibling’s spouse, I’m likely to come away thinking that the person is your sibling’s spouse’s sibling’s spouse unless I concentrate.  If we have a word like ianitrix in common, we can relax. 

So why does it strike me as strange that we have fewer kinship terms in English than the Romans had in Latin?  For one thing, because English has such a huge vocabulary overall; for another, because it doesn’t seem English was particularly rich in kinship terms even when most English speakers lived in extended family groups.  But most of all because there are a number of relationships that really are quite important to English speakers that have no simple names in English.  For example, I would think it was safe to say that most grandparents would agree that they have something important in common with their grandchildren’s other grandparents.  Yet they have no single word to express that relationship.  And a Google search for “grandparents of the same child” brings up just two hits, as I write this.  “Co-grandparent” produces hits for laws concerning grandparents in the state of Colorado (postal abbreviation CO,) for “The Grandparent Company,” and for a number of uses of “co-grandparent” meaning something like  “honorary grandparent.” 

At about the same time Trahman was writing his article, Archie C. Bush published an article in the journal Ethnology under the title “Latin kinship extensions: An interpretation of the data.”  Here’s a JSTOR link; the citation is Ethnology, volume 10, number 4, (Oct 1971) pp 409-432.  Bush opens with a list of Latin names for 110 family relationships, sorted into six grades of consanguinity.  The system of grades derives from Roman law; a text attributed to the jurist Julius Paulus listed 448 family relationships. 

Strikingly, there is no Latin word for “grandparent of the same child” on Bush’s list or in Paulus, nor can I come up with such a word in any of the dictionaries to which I have ready access at the moment.  This is really amazing.  Most marriages in the ancient world were arranged by the couple’s parents in order to build a kinship relation between one household and another household.  In that sense, one could say that the basis of marriage in those days was the hope of the parents of the bride and groom that they would be bound together as grandparents of the same child.  One could hardly imagine a more highly valued relationship.  Yet it was a relationship with no name of its own.