Log in

No account? Create an account

Previous Entry | Next Entry

Interesting Links for 25-11-2016


( 15 comments — Leave a comment )
Nov. 25th, 2016 11:10 am (UTC)
I find the genealogy article quite unconvincing, because it requires the assumption that for your 7938 ancestors in 1587, the chance of each of them being descended from Edward III to be independent. Given the degree of class stratification of British society, that seems an obviously false assumption.
Nov. 25th, 2016 01:26 pm (UTC)
It really doesn't take much permeability of the class structure / social mobility for that objection to crumble, though (and social mobility has varied greatly over the centuries). The current royal family's claim to the British throne comes from the relationship between Edward III's son John of Gaunt with a commoner, and the retrospective legitimation of their children born out of wedlock. And even more often, younger sons or daughters drop off the genealogical lists into obscurity. The last heir to the British throne works as a therapist in Rostock, in the former East Germany.

Consider also the case of Mary Garritt, the wife of Thomas Webb, a surveyor in Stow-on-the-Wold in the mid-18th century. Her daughter Frances (1775-1862) married Thomas Salisbury, landlord of Marshfield House in Yorkshire. Their daughter Anne (1806-1881) married another gentry type, Edwyn Burnaby of Baggrave Hall in Leicestershire. Their daughter Caroline (1832-1918) married a widowed clergyman who was the grandson of a duke. Their daughter Nina (1862-1938) managed to bag an earl as her husband. Her daughter Elizabeth (1900-2002) did rather better than a mere earl. Her daughter, another Elizabeth, was born in 1926 and is still alive; you will find her depicted on certain useful everyday objects, ie money. But her direct female line ancestry can be traced back only six generations before it is lost in the aspirant lower middle class of Gloucestershire.
Nov. 25th, 2016 01:51 pm (UTC)
It really doesn't take much permeability of the class structure / social mobility for that objection to crumble, though

I'm really not sure that's true - obviously there is some permeability, but it's far from clear to me that that's enough to lead to an even probability distribution. My hunch is that the conclusion that the probability of a random person in the UK descending from Edward III is quite high is correct, but hunches about probability are notoriously badly grounded in reality, and it would take more stats knowledge than I have (and definitely more than is demonstrated in the linked article) to verify that hunch.
Nov. 26th, 2016 12:10 pm (UTC)
You should be able to falsify the hypothesis that classes are permeable, by showing a distinct genetic difference between classes. I should warn you that such a difference has been looked for before, and the researchers looking failed. My Google fu isn't up to finding a simple discussion of whether class difference can be detected in population genetics (most of what I find is essentialist fools asserting that individual genetics affect class, which is backward anyway, and which they then fail to demonstrate)

So instead let's try a reductio ad absurdum argument instead. You would agree that if class is impermeable, then the children of royals are royal and not non-royal. We agree that Edward III had many descendants; this has been counted, not just estimated. Therefore we should agree that there are many royals. But there are not many royals, so we have a paradox. Therefore classes cannot be impermeable. The descendants of Edward III must be in other classes. And indeed when we look at these counted descendants, most have migrated to another class than royalty.

We could try an evolutionary argument. If genes were different between classes, and classes propagated at different rates (rich more, or poor more, it doesn't matter) then quickly everyone would have the more successful genetics. We'd either be one unitary population with different classes, or one class, having driven the other classes into extinction. But that contradicts our attempt to find persistent genetic differences between classes. Paradox again.

Thirdly, look what landed aristocracy did with their sons. One got the estate, the others got to be soldiers, clergymen, or professionals, an immediate and inevitable demotion in class, in just one generation. It seems unlikely it would stop there. Fourthly, look at our own families: class mobility abounds, and that's just officially, counting marriages, not counting rape and adultery. All the evidence points to class being about who gets the privileges; the genes themselves get everywhere.
Nov. 26th, 2016 05:49 pm (UTC)
I think you're arguing against a point that no-one's making. I've already acknowledged that class boundaries aren't impermeable. That doesn't mean that they have no effect whatsoever on the distribution of ancestry. (Although now I think about it, geography probably has much more impact - for how many of the generations between Edward III and the present day was it remotely common for people to marry outside their localish community?)

My point isn't that E3 has no working-class ancestors in the Outer Hebrides, but that the distribution of probabilities is complicated and certainly not completely independent, and the original argument is based on bad maths. There may well be good maths out there that supports his conclusion, but he has not used it.
Nov. 26th, 2016 09:30 pm (UTC)
That doesn't mean that they have no effect whatsoever on the distribution of ancestry

Yes it does. Mathematics is quite clear on that. And you can do the same analysis on geography.
Nov. 26th, 2016 10:49 pm (UTC)
Do you have any sources for that? My limited google-fu is leading me to information about the genealogy of mathematicians, which whilst potentially quite interesting, isn't particularly helpful on the mathematics of genealogy.
Nov. 27th, 2016 12:54 pm (UTC)
"The Royal We" Steve Olson The Atlantic

Joseph T Chang Department of Statistics Yale
"Recent common ancestors of all present-day individuals" [PDF]
This paper rightly reminds you it's about the mathematics of heredity, not genetics, and admits it assumes no barriers

Douglas L.T.Rohde Steve Olson & Joseph T. Chang
"Modelling the recent common ancestry of all living humans" [PDF]

None of these are slam-dunks for the position you're doubting, which is that merely moderate mating barriers confer no "immunity" to the "disease" of being Edward III's descendant. But remember, this isn't about genetic proportions; the rule is a one-drop rule, Everyone who has the Edward III virus passes it down to all descendants.

Now by asserting barriers between class and continent, I hope you agree that there are no such barriers within the enclosures, so you should agree that the virus will quickly spread inside a class or continent. It's the question of whether a moderate barrier confers moderate protection from the plague. I have seen work that says merely moderate barriers confer next to zero protection, but I still can't find any. Chang admits he didn't model that.
Nov. 27th, 2016 01:15 pm (UTC)
Actually, it appears the Nature paper does have continental barriers:it models the world as a graph of 10 nodes, with countries and towns, and accepts a little, but not much, migration between each. I think that actually qualifies as the demonstration you were looking for. So, even with a whole world, and poor communications, the paper estimates the Most Recent Common Ancestor of everybody as about a thousand years ago.
Nov. 27th, 2016 11:06 am (UTC)
Can I divert this a little to the west and a few centuries back?

In 2006, Irish researchers found a particular Y-chromosome haplotype as having 16.9% prevalence in parts of northwestern Ireland, 4% in Ireland as a whole and 2% in New York; they speculated that this might reflect direct male line descent from the semi-mythical fifth-century ruler Niall Noígíallach. It certainly reflects male-line descent from a common male ancestor, and it's reasonable to suggest that that would have been a high-status individual.

If 4% of the Irish population shares a single male-line descent from a few centuries back, then it's pretty much certain that everyone of Irish background is descended from Niall Noígíallach (or whoever it may have been) one way or the other, in numerous different ways; the lines of ancestry must be saturated. 4%, oddly enough, was precisely the proportion of direct male-line descendants among the living descendants of Prince Albert and Queen Victoria when I wrote them up in 2010.

If you can accept that effectively all people of Irish descent are descendants of Niall Noígíallach, then the question is not, is there an equivalent for England, but who was he or she and when did they live?

(Of course, the descendancy from Niall Noígíallach is so over-saturated that the real most recent ancestor of most Irish people must have lived much more recently.)

Edited at 2016-11-27 11:18 am (UTC)
Nov. 27th, 2016 12:06 pm (UTC)
For the leaping of geographic barriers, I was looking for analyses of European descent from Genghis Khan, Sadly, Google hits are now saturated with discussion of Y-chromosome haplotype prevalences indicating descent from Genghis Khan in the male line only. This was disappointing, because it's much more stringent than what I'm looking for, which is any descent at all, including female line. As you say, the likelihood is that 10% male-line descent probably means 100% descent is inevitable, but I wanted the mathematical models, and Google has buried them.
Nov. 27th, 2016 12:53 am (UTC)
I'm agreeing with you on this one, I'm from a very rural Devon family and while there was some very low level changes with the very low level gentry it wasn't usual.

Plus, the article says cousin marriage is very unusual, both my own family tree (both grandmothers married their first cousin1) and the local gravestones in the two parishes the family are mostly from say distinctly otherwise.

It may be on aggregate that a 'typical' person is likely to be descended from royalty, but typical isn't the same as mostly and if you've pockets where cousin marriage and interbreeding within the same village are the norm (which is definitely the case up until relatively recently) then it reduces the odds, the idea that 99% can trace descent from a King makes far too many assumptions.

1: Having said all that I have no idea who my paternal biological grandfather was, Granny married the man I knew as Granddad and who adopted my Dad when he was about 4, and I've never had the guts to ask more. So I might be descended from anyone, I do however doubt that my cousins still living in the village are as likely to be, I've seen the parish records.
Nov. 27th, 2016 10:51 am (UTC)
You admit yourself that full knowledge of your own little pocket barely goes back two generations!

I look at it another way. Everyone who is dead either has living descendants, or doesn't. The farther back you go, the more the proportion of those without living descendants rises (to between a third and a half of those alive at any given time, according to various estimates).

All people of English descent are descended from some of the people of English descent who were living at any particular epoch. If you take the wider population of people of English descent, there will be a point in the past where almost all people of English descent are descended from almost all of the English people living at that time who have descendants living today.

So there is a sweet spot out there in unwritten history; the only question is, when was it?

Edward III lived 660 years ago, and we know that he has living descendants.

Allowing a very generous 30 years between generations, that's roughly 22 generations ago. So there are potentially 4,194,304 lines of ancestry back to the England of 1356. (I reckon the average gap between generations is closer to 26 years, in which case we are talking 25 generations, 33,554,432 lines of ancestry, to 1366.) Some of them will overlap of course - but all of them? Surely you'll agree that there are likely to have been at least a million distinct ancestors of yours alive in England in 1351?

The estimated population of England in 1351, immediately after the Black Death, is 2.6 million. Not all of those will have left descendants; almost all of those who did leave descendants are going to be sharing lines of ancestry to you with their parents and/or children. The number of available lines of descent from the 1351 population to the present day is probably also around a million.

So the argument is simply that, given the numbers, it would be pretty odd if you were not descended from Edward III, given that everyone of English descent alive today had English ancestors alive in 1351 who left descendants, and we know that Edward III was living in England at that time and had ancestors.

I admit that he is near the mathematical sweet spot I mentioned above. But I think he is safely on the sweet side of it, particularly since he had 28 grandchildren.
Nov. 25th, 2016 06:40 pm (UTC)
Say it ain't so
My passport's maroon.
No glass of ours was ever raised
To toast The King. :-)

Nov. 25th, 2016 10:52 pm (UTC)
I'm not convinced about the possible "hacking" for the US presidential election results either. I tried to put forward some possibilities why the claimed statistics might be misleading to some friends, but they were too caught up in the possibility of Clinton winning to listen.
( 15 comments — Leave a comment )

Latest Month

March 2018


Powered by LiveJournal.com
Designed by yoksel