How many descendants does a pre-1700 ancestor have? 67,108,864

+57 votes
18.0k views

67,108,864.

This essay is especially addressed to new people to consider, before they rush to certify as pre-1700, because they mostly want to add that 1620s ancestor they have in their family member's research.

The point I wish to make here is that the chances are extremely HUGE that somebody has already created that profile, because there are a few living genealogists among those 67 million direct descendants.

Here is how the math works. Assume a typical generation is 25 years. From 1650 to 1980, there have been 13 generations.

I start with a presumption that each ancestor has four children. Of course it varies with time, some descendants have much more children, some have less or none. So four seems a reasonable presumptive average for each of them.

So the ancestor born 1625 has four children. Each of them has four children. That makes 16 grandchildren for the 1625 ancestor. In actuality, he has 20 descendants, 4 children + 16.grandchildren. But I am going to ignore that complication, just to be even more conservative. So the number I come up, with 67 million, is really just the number of living descendants today, and just in the 13th descendant generation, born 1980.

The way the math works to get there is 4 to the power of 13, which is 67,108,864. If you include those children's parents as well, born 1955, then add another 16,777,216 to that. That parent generation of descendants is 4 to the power of 12.

These days people don't have so many children as four each. But before the 20th century, they commonly had a dozen or more. So I think the 67 million is a reasonable, conservative guesstimate.

But even if you don't follow the precise math, or would nitpick it (easy to do, please feel free) please come away with an understanding of the key point I wish to make -- which is that when you are adding an ancestor in the 1600s, or especially earlier, you are presuming to personally take on the responsibility of ancestral identification for tens of millions of living and past people who are directly descended from that ancestor. So are you really prepared for that responsibility? Are you the only genealogist who has presumed to do so? And are you the best single one genealogist among them all, who is really prepared to accurately, or firstly, do so?

in The Tree House by Steven Mix G2G6 Mach 4 (47.3k points)
retagged by Robin Lee
Gabe I don't fully understand your question.  Are you asking how many descendants could be produced from 3,000 ancestors in 150 years?

Regarding slave-ownership by blacks in the south before the Civil War: most southern states had prohibitions against free slaves remaining in the state after obtaining their freedom.  Most were required to leave the state in 30 days.  Slaves that were free before the law were grandfathered into the law and allowed to remain.  So, if you received your freedom and you had other free family in the state, it was common that your relatives became your "owners" and in this way, you were allowed to remain close to family.
With potentially 67 million being descended from one person then the entire US population would be descended from less than 5 people in the 1650's.  That is if the population of the US is used for 2018.  It would be less than 4 if the 1980 population was used.  Needless to say, that 67 million is an extremely high number, even with tree collapse.  I see you point though.  However there are profiles of individuals that lack sufficient verifiable sources and information, or wrong information, to make a determination if they are the same person you are looking for or not.
Gabe, Let's assume that a slave owning couple and each of their descendants have four offspring who survive into adulthood and have offspring of their own.  Let's also assume that the average generation is 30 years, and that there is no pedigree collapse.  Our original couple starts with no offspring in 1860; in 1890 they have 4 offspring.  Thirty years later, in 1920, those four offspring have each had four offspring of their own for a total of 16 grandchildren of the original couple.  This multiplication by 4 continues every 30 years until the generation living in 2010 consists of 1024 descendants of the original couple.  If we consider your original estimate of 3000 blacks who owned slaves to consist only of the Male slave owners, and we assume that there is no marriage between any of their offspring, then the total number of offspring would be 3,072,000.  If the original figure of 3000 owners counted both husband and wife then the number of offspring is half of the 3 million figure.  If you also consider intermarriage between the offspring of the original 3000 owners, then the total number of offspring in 2010 is reduced by a factor of whatever the intermarriage rate is.  If the intermarriage rate is such that we have to reduce the generational growth rate to only 2 offspring per couple, then the number of offspring for the 3000 owners in 2010 would be 96,000.  I guess the actual figure is somewhere between 96,000 and 3,072,000.

Steve... LOVE the math... Isn't math wonderful?   

10 years ago I started my tree with two parents.. and they each had two parents and I now have almost 400,000 on my tree.   While I will admit I merged trees with my fiancé's tree, my own tree HAS increased quite a lot, and while not EVERYONE is related to me because there are many on his tree, as a whole, probably 2/3 of our 400,000 is connected to me and I haven't added ALL the brothers and sisters to those on the list.    

I find that some folks working on their trees and get back to when there were 11 and 12 children on a list, that maybe 2 are important to them, so they only enter information on those two.. the other 9 or 10 get left off.  <sigh>   So I have a LOT of work and research to do.   

And YES.. your 67 Million count sounds reasonable as well as conservative, without doing the calculations.    

BRAVO !!

Rebecca  

Hi Steven, The 67 million figure is way off.  If you go back only a few more generations from 1625, say to 1525, the number of descendants is in the 10s of billions! more people than are alive on earth!

So what's the problem? Well one thing that you are not accounting for is inter-marriage between descendants. Your math assumes that each descendant of person x never has children with another descendant of person x. When two descendants have children together, you are counting those kids twice.

Of course distant cousins marry all the time, especially in times and places where migration was less common.  I don't know how you would account for that in the math, but I think its a big factor.

Steven was clear: "Here is how the math works. Assume a typical generation is 25 years. From 1650 to 1980, there have been 13 generations." ("So I think the 67 million is a reasonable, conservative guesstimate.") I have been on WikiTree as a research coordinator now for some 7 years, and the average generation count for that period on WikiTree is precisely 13 generations. The point he was trying to make was also very clear: "which is that when you are adding an ancestor in the 1600s, or especially earlier, you are presuming to personally take on the responsibility of ancestral identification for tens of millions of living and past people who are directly descended from that ancestor. So are you really prepared for that responsibility? Are you the only genealogist who has presumed to do so? And are you the best single one genealogist among them all, who is really prepared to accurately, or firstly, do so?" So your comment about his guesstimate being off actually supports the point his making.

Absolutely off! A portion of my family tree from my mom's side comes from Nantucket and the families there intermarried. Even on my dad's side, he has early French-Candian ancestry and many of them intermarried. Not to mention when you start coming into the 1900's, you start seeing family sizes decrease. Also it would be nice to know how many descendants would be alive today, although it would be impossible to even guess and come close. His guess of 67,000,000 probably should be 47,000,000 and of those only a small number would be alive today.
It turned out to be a very interesting essay! I am now even more interested in the issue of birth record and control. If you have any interesting materials on these topics, I really want to see them from you!
I love it when I am reminded of a conversation like this!

@ Daniel, as a starting point have a look at:

https://en.wikipedia.org/wiki/Demographic_transition

edit I think  that the new link in your post above goes to an essay mill. Far better to explore a subject yourself.

18 Answers

+3 votes
 
Best answer

This is an interesting theory that relies on many assumptions, the probability of which is either unknown or at least not specified. For example the assumption of a generation being 25 years. To do a correct scientific analysis on this you would need to state the standard deviation of this age and carry the error bars forward into the calculation of each generation.

Secondly, the assumption of 4 children per family also  requires a statement of the standard deviation. The error bars associated with this count also must be carried forward into any calculation of subsequent generations. 

These are not the only assumptions that represent data that vary. Error bars on multiple variables, when carried forward into the calculation for the next generation interact with each other to grow the error bars on each subsequent generation. Then if you add in historical events such as plagues, floods, fires, wars, etc. you will need to make an assumption about how these events tended to decrease the population, and what effect this decrease had on the number of descendants in the affected generations.

My point is that with so many generations in between then and now, so many variables with unknown or at least unstated error bars, the figure of 67M descendants could be right for one set of assumptions.

With so many assumptions it is not possible to get an accurate count using only these data in the absence of an estimate of the probability that each assumption holds going forward into the next generation.

By the time you recursively carry forward a calculation with realistic error bars on each variable, the error bars on the calculated number may be so huge as to render the results be meaningless.

Given that this is the case, any conclusions drawn on these results also have the same degree of validity as the results on which they are based. Here again, more assumptions are required to carry forward the results into the conclusions.

This is not to discourage anyone from thinking about some of the points made in this essay. My point is that for some families, the essay will be more applicable than in others. 

Birth, death and marriage records may not be available in all parts of the world. Records are destroyed in fires and storms. Some information is falsified to cover up dark and bloody family secrets. Hand-written records are transcribed incorrectly and lead to erroneous data in profiles.

We can't wait for perfect source data, approved by government agencies in "official records" to enter into profiles what we know to be true from family documents, such as the genealogy pages in family Bibles. Whereas not all family documents are 100% accurate, generally the inaccuracies will be about exact years of birth and death, not whether or not a person existed.

Given that the assumptions in this essay do not apply equally to all of my ancestors, I would answer the questions in the end of the essay as follows.

1. So are you really prepared for that responsibility? Answer: If not me than whom? Who else knows what I know or is willing to work on it? I wish that cousins would know more than I and would be willing to come forward. I don't like being the "family expert" because there is so much that I don't know.

2. Are you the only genealogist who has presumed to do so? Answer: No, but I am the only one who seems to be curious enough to ask the questions about certain generations in my family and the only one with sufficient time to spend on it. I wish more cousins more often would help to confirm or refute my work, or at least add to it, but everyone seems to have too much to do.

We can't be too afraid to make mistakes because this will discourage genealogists from doing anything that might be imperfect. If you make a mistake, just correct it and go on. (And don't forget to thank the kind WikiTreer who helped you find the error.)

3. And are you the best single one genealogist among them all, who is really prepared to accurately, or firstly, do so? Answer: This is an absurd question designed to discourage anyone new from trying to document what they know. We have no way to determine if we are the "best" genealogist because we can't be sure that there is no one better. Even if we could know who is best, do we really want him or her to be the only one to work on the family tree? All we can do is to identify someone who is better at it than we are. That will be the genealogist to whom we pose questions when we get confused. Then we must ask whether or not this better expert has the time to put into researching a particular branch of the family that interests us.

When appropriate, I will happily and thankfully defer to those with more experience and better knowledge that what I have. However, there is no way that they can know everything that I know.

by Marion Ceruti G2G6 Pilot (351k points)
selected by Danielle Liard
Okay, so I will try to explain it again.

My point is that most people, by far, propagate way too much utter garbage about the most ancient ancestors. They do so because they overestimate their own relative competence, as well as their own unique descendancy, by far, by many magnitudes of error.

What prompted me to pose the idea about a need for consideration of one's own relative expertise in the matter is because the biggest mistake I ever made before I started examining the root trees on here was to first presume that the ancient genealogy on this site would be uploaded and managed by only the greater experts among us, as most people would be inclined to defer to such experts, out of a sense of proportion and humility.

While what I found instead was unfathomable mountains of garbage about imaginary ancients, propagated by everybody who ever stumbled across any bogus family tree on the Internet which may or may not have included somebody who kind of sort of resembled what they think were their ancestors. Probably had the same name.
Steven, WikiTree started with people uploading GEDComs from other sites, lots and lots of that.  There wasn't really any quality control at first.  WikiTree has evolved over time, there are pre-1700 and pre-1500 certifications, and requests for sources etc etc etc.  And one should not generalize on whether or not somebody is competent to manage profiles dating way back when.  That becomes accusative in a way, and is frowned upon.
Hello Steven, I agree with you about the mountains of garbage and also about the wrong assumptions people make about who has descended from whom. For example, before I knew anything about genealogy, I had been taught, in what turned out to be a family myth, that our Ceruti family had descended from the famous violin makers with the same name. I had to go to Italy to find out that this is not the case, so that family myth was exploded.

In another case, our family always displayed the armorial bearings of John Pakenham Stillwell because we have Stillwell ancestry.

https://www.wikitree.com/wiki/Stillwel-2

For years the Stillwell descendants in my family had big meetings in New York where we all had cards that displayed this coat of arms. It took a trip to the College of Arms in London to find out that this particular coat of arms was never granted to any of our ancestors. This is a case in point about mistakes based on the same family name.

Your points about the proliferation of family trees on the internet are well taken. Many private family trees have errors of some kind. What seems to be the most accurate is eye-witness information and cousins who can provide photos and more recent documents to back up their assertions about immediate family. The farther back we go the more we must rely on good sources. This is why it is becoming increasingly difficult to become pre-1500 certified.
Danielle, I posted this essay way back in 2017, before those qualifying badges you mention ever existed. They came into being precisely because of the issues I raised here so long ago. My essay was a mostly futile attempt to encourage people to use restraint rather than common hubris, when adding material to the site about the most ancient of their wannabe shared ancestors.

But as you note, avoiding making any comments that could come across as things like "accusatory" is the prime directive on this site, rather than being most careful about the most shared areas of genealogy, which frankly just gets a lot of lip service, even still. Despite its good points, WikiTree has always been extremely flawed in design in multiple ways, and this egalitarian ideal about contributions is just one of those ways.

I have always hoped to help improve the substance of the site, including improving the way people think about their approach to the site when I find that they largely are making colossal messes. But the pushback I get and the resistance to actually improving that substance is partly why I just barely work with it anymore. So it goes. It could be worse. I still work on my bits, while trying to avoid getting caught up in trying to fix too many of the ongoing disaster areas anymore.

And to be clear, I am not claiming to be any kind of expert, or better than other people about it. I have contributed my own fair share of sloppy stuff here, give or take. But my essay was specifically directed at the MOST ANCIENT of genealogy roots. Those are the MASSIVE problem areas, because they are the MOST shared among common descendants, just because of the math. That is the sole part I wanted people to reconsider here, about their own part in it. Because people usually just don't think in terms of the exponential math consequences in what they are doing in monkeying around in those early generations.
+45 votes
One estimate for descendants of ALL the passengers of the Mayflower is 35,000,000. See discussion at https://www.wikitree.com/g2g/231659/35-million-descendants-from-the-mayflower-in-1620?show=231659#q231659 and I think that may be 4x to 10x too high, so 67,000,000 from one person is probably way too high. Also clearly a lot of variability in number of descendants from person in 1600 to another. Could be anywhere from 0, to a few, to millions.

I have mixed feelings about your message. On the one hand, we don't want inexperienced people wrecking good established work, but if the profile is not there yet or has not been well developed, we want to encourage people (even if they don't think they are the foremost expert on the person) to take a first stab at it since the profile can always be improved later by others. You don't have to be the best, you just have to be willing to help and respectful of other's work. We currently have tons of profiles from the 1600s that are not well developed. What we need most is people willing to take the initiative and put in the effort to improve them. Discouraging people from working on a profile unless they think they are the world's foremost expert on the person is probably not what we want to do .
by Chase Ashley G2G6 Pilot (308k points)
Points taken. But even if my *conservative* guesstimate is too high by a factor of 100, then a person adding a brand new 1620s profile is taking on the responsibility for 670,000 living descendants born in the 1980s generation.

That still seems to me to be a HUGE responsibility to presume to take on. In what other endeavor, or field of work, would anybody presume to speak for almost a million peers?

So I simply want people to ponder it a bit, before they add or alter the most ancient profiles, based merely on a tree that they found somewhere from somebody at some time.

"That still seems to me to be a HUGE responsibility to presume to take on. In what other endeavor, or field of work, would anybody presume to speak for almost a million peer?"

But that's the glory of a wiki. It's not a huge responsibility because if I mess up, the millions of others will fix it.

"So I simply want people to ponder it a bit, before they add or alter the most ancient profiles, based merely on a tree that they found somewhere from somebody at some time."

Now THAT I can totally agree with. Add quality to the tree, don't spam it with junk genealogy that you haven't researched.

Well, if only there were a million fixers, and a just a few data dumpers. :)

It is precisely because of my having spent endless hours fixing and merging, again and again, which has prompted me to write the essay. So if I can get people to pause, and to work a bit smarter, especially when they first join and immediately start to duplicate ancient ancestors who already exist, I will be happier.
Another interesting point about the Mayflower, is that, probably, if you are descended from one of them, you are descended from ALL of them who had children. That is because they lived in a cohesive community, and later generations intermarried. So they share descendants down the line.

So basically, if the conclusion is that *a* Mayflower passenger has 35 million descendants, then my 67 million figure is really not so far off the mark.

Of course, each generation that you add or subtract to the calculation, will multiply or divide the total by the average number of children per generation. So it greatly matters how many generations we are precisely talking about.

It also greatly matters how many average children in each generation goes into the calculation. For instance, I originally wanted to state five children. But I came up with a calculation of over a billion descendants! So clearly, that number five is way too high for an average family size, in American descent.
Interesting answer Steven. I don't seem to be decended from any of the Mayflower passengers. This confused me because every Magna Carta Surety Baron is one of my Great Grandfathers.  Also, Deacon Samuel Chapin, Edmund Rice, Protestant martyr William Tyndall's father, John Plantagenet-King John I of England, Queen Eleanor of Aquitaine, Elizabeth Woodville York, - are all my great grandparents. I am the great grand daughter of dozens of Puritans who came to New England during the Great Migration (1620-1640). Therefore I have been confused by my lack of ancestors on the Mayflower.  Your explanation that they were a closed group who intermarried, is a helpful explanation

Interesting discussion. It caused me to recognize that my small handful of Mayflower ancestors are exceptions to that generalization about Mayflower passengers forming a closed community. Specifically, my four Mayflower ancestors include three who would not have been strongly inclined to marry within the Mayflower passenger community:

  • Two children of passengers who didn't accompany their parents on the Mayflower, but emigrated several years later, after their parents were dead (having died during the first winter): Matthew Fuller and John Rogers "of Duxbury"
  • A man who traveled on the Mayflower when he was a teenager, not in the company of his parents, and who wasn't old enough to marry until after a number of other immigrants had arrived: Henry Samson

The fourth is Peter Browne. His first wife arrived on the Fortune in 1621, and was thus presumably was part of the Mayflower community, but when he married his second wife in 1630 there were plenty of non-Mayflower women to choose from, and my ancestry is traced (somewhat tentatively) to him and his second wife.

Interesting!

Christine, I can see that your Gateway Ancestor Edward Kemp arrived in the 1630s. I think what is going on is that the Mayflower was just one of many voyages which quickly populated the colony.

So your particular branches must have spread out, before they could interact much with the small handful of Mayflower passengers. That would make sense, if in the first decade or so 99% of the population consisted of other arrivals and descendants. I have some like that in my line.

As I understand it, the actual Mayflower families simply became swamped by the larger pool of new settlers. So the chance of any particular one of them marrying into a Mayflower family became relatively very small.

New Netherland, on the other hand, is a different story. It was only about 1/10 the size of the Puritans. So whereas Puritans were made up mostly of succeeding waves of new immigrants year after year, the New Netherland families were largely direct descendants of each other, who married into each other's lines again and again and again. So I am descended from a bunch of them, and repeatedly.
Yes Steven. Some examples would be Thomas Harvey b.1617 who arrived with his brother William Harvey in Plymouth Plantation 1636 and John Whittlesey a founder of Old Saybrook  who bypassed Massachusetts Colony totally and settled in Connecticut Colony.
Hi cousin!
I think it's fine if inexperienced people create profiles, as long as they're OK with seeing them corrected. The main problem with wikitree profiles is that once they're created, some people develop this sense that they must be correct just by virtue of being here, and they fight tooth and nail to keep you from "destroying" them, even if they're full of unsourced and often ridiculously unlikely claims.

In a post, Andover ordeal, I give an example of a couple that lost all but two of their kids to throat distemper. Then, only one of those two had children. He had four kids of which only one lived and had progeny. Hence, we can point to that node as having descendants. Not only do we need to consider losses like that, we have inter-marriage (1st cousins marrying was not that uncommon) pruning the tree. 

But, let's go way back, to William Marshal. He has descendants only through his daughters. I like to refer to him as he kept us from all speaking French, sort of. 

There was a mathematically framed attempt at this type of estimate which will get better, however the assumptions would be quite well documented. Then, it would be a fuzzy number, no doubt.  

BTW, the U.S. Census? Has it ever succeeded in doing a complete count? No. Same sort of dynamics apply. Meaning, we probably need an estimate. It might be limited to a few generations. Perhaps, going against a known tree would help hone the idea. 

Mayflower descendants have really looked at their trees. We could take the five generations, perhaps. How full is that? One blog post (How long is a generation?) at the NEHGS' new blog tried to show the range of one generation (it covered over a century). 

No doubt, we'll find all sorts of interesting analysis work with better data and improving computation. 

Everyone has touched on most sides of this but I can say most of my work on here has been reloading and catching new uploads of profiles that need to be merged... Moreover the information is exactly the same as previously established profiles that it doesn't make sense how nothing was flagged during the creation process- thus where frustration occurs- are the suggestions ignored? Is there a large flaw somewhere that hasn't been discovered? I enjoyed the post and the ensuing conversations. Cheers!
Living Buckner, I've run across that situation several times.  I added verified sources to profiles and the "Owner" made it clear it was HERS, SHE created it and didn't want ANYONE doing anything to it.  I explained it was also my ancestor and the information was correct.  She didn't care.  She removed what I had added and said she didn't appreciate when someone added to HER work.  I even tried to explain Wiki is a joint effort.  Some people think they know it all and forget they had to start somewhere too.  Condescending attitudes do not make for a happy Wiki environment.  :-)
+18 votes
I belong to an international organization where most of us descend from the same couple in 1600s.  Michel Andres and his wife Appollonia Krieger  https://www.wikitree.com/wiki/Andres-222   They had 12 children.  I descend from their daughter Christine.  Over 400 people in this organization descend from this couple who we call our Mythic Couple.  Not means that I have over 400 cousins who are into genealogy but when I entered him he was the first one entered...   Go figure!
by Laura Bozzay G2G6 Pilot (824k points)
Ah, so I think the difference there being that it is in ancient Germany, and then into France, as I quickly glance at your descendant tree.

The vast majority of genealogy and WikiTree in particular is dominated by the United States and its precursor ancestor immigrants. So perhaps my essay should be qualified to address American genealogists tracing their American roots.
Michel has descendants who were in Colonial America.  Big family dispersed literally around the globe...   I think at last count we covered about 5 to 6 languages in that group...  I agree that the farther you go back you can expect the tree to grow exponentially.
+18 votes
As already mentioned, your figure is rather on the high side.  In addition to the fact that family sizes in the last few decades has greatly decreased, at least in the Western world, even in the early days not every child lived to be married or was able to have children.  Now if you look at the number of children your ancestors had you might notice mostly large families but that's partly a statistical anomaly.  Say a random sampling of 5 families came up with 0, 0, 1, 4, 12.  That would be a little over 3.5 children per family, but if you asked the children how many children were in their family it would average out to over 9 since most of them were in the larger families.
by Dave Dardinger G2G6 Pilot (437k points)
Well, I did invite people to nitpick my math. :)

But I hope the larger point of the essay does not get lost, that the descendant populations of 1600s ancestors (not mortality infants) is extremely large. And so people should always use great caution and consider deeply, before adding any such ancestor as a new profile.

As far as the modern era, birth control as a major population limiter really only dates to the mid-1960s. Before that, my parents had four kids, their parents had four kids and six kids, their parents had 8, 8, 6, 6, more or less off the top of my head. And so that is all just within the 20th century.

Going back into the 19th century, I can see that pattern continue just like that, on every branch, just from we have researched and recorded.

But as you go farther back, branches get lost to any particular researcher. So we don't see the common 6 and 8 kids anymore. We just see the one particular ancestor sibling that we happen to be tracing our descent from. So the other siblings exists, and had children, but we just don't know yet who they are. Their descendants may know who they are, but they do not know that we are cousins through the common parents.

When you look at any particular descendant tree, the vast majority of genealogy lineages remain incomplete like this, because we just don't know yet the common deep ancestors among the many thousands of living descendants, once we go back a couple centuries. Each of us only knows which ancestors trace along the line to our own descent, for the most part, with maybe a sibling or two that have also been connected.
I see, on Wikitree, a LOT of profiles with only one child.  Then I find the will and see 12 children listed.

Remembering back to my history at Uni - the colonists had very large families until the population swelled, the number of children per family waned, the colonists pushed west and the family sizes boomed - and continued booming until the 1960's.
I think that one reason for this is because to make a profile, at least ideally, we try to enter as much accurate data as possible. Sometimes we refrain from creating a profile for siblings of ancestors because we don't (yet) have precise data, or the data from various sources are in a state of conflict. Another problem with entering profiles for all the siblings is that the existence of some children has been covered up for various reasons. Their data will not be well attested. In extreme cases, data proving their existence can be found only in the US Census data where respondents are discouraged legally from reporting lies about children in their household. In theory, we would all like to see completed family trees with all siblings represented with their own profiles, but the reality is that we don't always have the data we need, sometimes by design.

Getting back to Steven's original post, WikiTree has a good method to determine whether a profile already exists to prevent duplicates. New genealogists should familiarize themselves with the great tools we have available and make sure to include complete citations of sources so others can assess their reliability. Approximately right is not good enough and the farther back in time your ancestor lived, the more important this is.
+18 votes
But it is VERY unevenly distributed.

I have about 300 pre-1700 profiles (of which about 30 are  pre-1500)

Less than 10 of them are "shared" with other WikeTree profiles.
by Janet Gunn G2G6 Pilot (157k points)
When you refer to sharing, you apparently are talking about sharing with other profile managers. That's not necessarily a good measure of the number of descendants. Many of us are not managers (nor even on the trusted list) of the profile of every ancestor who has a WikiTree profile.
I was responding to this statement, which implies that there is another WikiTree-user  descendant who can do a  better job. And that you should not create, or work on, pre-1700 profiles unless you are a "super-genealogist"

Your calculation seems to ignore several factors

1- typically  half of the children die before reaching reproductive age

2- typically not all of the children marry

3-Typically some of the married children have no children

Yes, these get balanced out by the one surviving child who has 12 children, but i still think you are overestimating

I don't remember the source, but I remember reading a study of British titles.  It took each title when it was originally granted, and tracked how many generations until it died out (no known male -line descendants of the original person).

The key number was 6 generations. I do not remember if that was the average number of generations before it died out, or if it was the median,  "50% died out by the 6th generation".

Obviously if you include female descendants, it will go further.  But still there are LOTS of lines that die out completely.
Janet, I calculated as low as 4 surviving and procreating kids each, to intentionally account for the three things you list. From what I see, 6-8 kids per family is rather typical, before the 1960s birth control, and certainly in the prior centuries.

So 4 surviving kids who each procreate is maybe a 40% mortality rate in each past generation, which seems reasonable to me.
I am looking at a table of annual US population growth. About 2% growth around the year 1900.

So take a village of 100 people. At 2% annual growth, over a generation of 25 years, the village will become 150 people, in simple math. They add 2 new people per year, over and above the deaths. So each year, the village has four births, and two deaths. The net result is two new people added to the population each year.

So take a typical family in the village, at that growth rate. A family of six. Two grandparents, two parents, two kids. After 25 years, they are a family of nine, a 50% growth, matching the average growth for the whole village. To get to nine, the two grandparents and one infant have died. But they also added four new children who lived to adulthood, and two new infants born to the older children. That math is 6 - 2 - 1 + 4 + 2 = 9.

So they have five children surviving to adulthood. In my scenario, it takes only four of those five to procreate in numbers of generating four surviving children each, to make 16 grandchildren. So, before marriage, one of those five adult children dies of disease or in war, and one other has only one surviving adult child of their own, who goes on to procreate, while the remaining three original children of the five have five survivor children each.

So in the process of this one generation, there are indeed 16 grandchildren of the original parents, just as my math guesstimate would predict. And all that is based in the actual population growth numbers, so ignoring the relatively minor effect of immigration, for simplicity.

At this 2% annual growth rate, the rule of 72 says that population will double every 36 years. So we have, very roughly, 10 population doublings since 1656 as follows:

1656 - 300,000 base number, more or less

1692 -  600,000

1728 - 1.2 million

1764 - 2.5 million

1800 - 5 million

1836 -10 million

1872 - 20 million

1908 - 41 million

1944 -82 million

1980 - 165 million

2016 - 330 million
This doesn't take into account any new arrivals in the area. IIRC there has been considerable immigration into the US during the last 200 years.
Yeah, I would say that's consistent with my experience. Before 1700, I occasionally hit a tree that someone else is working on, but mostly I'm off in the wilderness by myself, and that's even when I'm actively trying to hook up to lines other people have documented. Usually, I make links to other trees a high priority, but even then they don't happen all that often. There are a few deep trees that have been extensively covered in secondary sources (whether they're right or not is another story), and I've noticed a lot of people tend to follow those back and conclude that Wikitree has a pretty thorough coverage because that particular tree is covered well, but when you're doing original research, it's a different story.
+15 votes

Just a small example.   Daniel Doty, the GG grandson of Edward Doty, of the Mayflower is one of my ancestors. 

https://www.wikitree.com/wiki/Doty-757

I've decided to start tracing some of his descendants and so far, he as 12 children and 42 grandchildren and I have more to fill in. 

I know I won't be able to fill them all of his descendants in, nor will I try, but I can at least get a couple of generations in.   

 

by Craig Albrechtson G2G6 Pilot (102k points)
+23 votes
If any of us were really intimidated by the thought that we are accepting some sort of awesome responsibility by creating a profile for an ancestor with many descendants, we would certainly stop long before we got to the 1600s and millions of descendants.  Fact is, of 67M descendants (or whatever number), about 66.99M don't care.  Ther ones who do care are free to ignore or reject your work, or replicate it in their own way, or jump in and try to help you improve it.  Usually, it seems, they don't do any of those things, they just give you a little pat on the hind quarters and say "I'm glad somebody is doing this.  Keep up the good work."
by Dennis Barton G2G6 Pilot (549k points)
Perhaps your comment makes sense for recent generations, Dennis. But the point of my essay is to inform mostly the brand new people who join, and who then immediately start to add duplication of those deepest ancestors.

Because they have not the least idea of the generational exponential growth, so they carry that mindset that they are the only ones, back to the deepest ancestors which have long since already been done. So I want people to consider it first.
You may not so far of the truth, Steven. My documented parenteel starts with a Nicolaas Johannes (van Weert) born in approx. 1590 (he had 7 children). Assuming 13 generations, 4 children of which 2 males, one might expect 8192 live van Weert males currently living in the Netherlands. In fact  there are about 1000. It does not take a great deal of tweaking the assumptions to to match that number. I personally think the number of procreating children should be assumed lower than 4, farmers wanted only one successor so as not to have to split the farm. Second sons, if any, went into very low offspring occupations such as the church, army, VOC, etc. Other adjustments would involve the fact that there were several clusters of van Weert's in addition to Nicolaas Johannes. That should be balanced by population loss due to diseases and successive invasions by Spanish, French and German armies. (I am surprised the Civil War was is not mentioned in the US estimations).
Excellent point about the losses in the American Civil War. We lost about a half million men, I think it was, out of a population of perhaps 15 million, taking numbers from my chart above.
wrt

"Perhaps your comment makes sense for recent generations, Dennis. But the point of my essay is to inform mostly the brand new people who join, and who then immediately start to add duplication of those deepest ancestors."

I completely agree about adding duplicates of early ancestors.

But that does not apply to adding new ancestor profiles.

For instance, on WikiTree there are 63 Battiscombe profiles, going back to 1165.

6 are orphaned (the earliest ones)

2 (post 1850) have other managers

I created/manage the rest.  If I had not created them, nobody else would have.  There is no connection yet with any other user's profiles until the 19th century.

Yes, I mostly have entered my direct line of descent, but I intend to fill out the tree. And any other Battiscombe descendant can connect to the profiles that are there

I do not understand the suggestion that BECAUSE they potentially have many descendants, I should NOT create them.
Janet, I am highly skeptical of many pre-1500 profiles, especially of ones that do not exist in the tree yet. I have seen too many bogus lineages.

So it just raises a ton of questions about whether those people even existed, by those names and dates. And who has enough expert competence to prove it?

My personal belief is that BECAUSE the profiles represent the ancestors of millions of people, there is an implied responsibility that goes with it. A person who propagates it should really know that they are posting fact, and not fantasy. And be able to back it up, on an expert level. Not just repeat some tree that they got from somebody from somewhere.

The more fantasy that gets published on WikiTree, the more it will be disregarded as a whole by people who do care about the difference.
FWIW, I actually agree with these sentiments, and sort of agree with the actual statements.  But Steven, you know, the site already has a pre-1700 self-certification and a pre-1500 certification badge that's required before you can play in those arenas.  What else would you do, or what standards would you apply, or how would you propose to administer genealogy activities correctly in that era?  How would you measure their accuracy?  Or would you just cut it all off at a more recent point when the research starts to become harder and the results less certain?  And your note sort of begs the question, how do you know the lineages you have seen are bogus?
I can assure you that none of my profiles (pre or post 1500) come form a "tree that I got form someone".  They are all backed up with sources and references.  And where I have had to make assumptions (as to whether we are dealing with one person, or two people with the same name) I spell that out too.

One thing that may make a difference is that all of my ancestors are in Great Britain. I am my own "Immigrant ancestor".  It seems that a lot of the problems crop up with people who are trying to trace the British origins of early immigrants to North America.

That doesn't make me  super  genealogist, just a competent one.
And there are already restrictions to exclude the incompetent from editing pre 1700 profiles.  Most of the really bad ones were created before we had the pre-1700 and pre-1500 certifications.
With you tracing lineage in Britain, Janet, I can see where the lineage might not have yet been entered. But as for sources, what sources are good enough for pre-1500? A lot of lineages have been replicated because people followed a source that has been discredited as complete fiction. That sort of thing is rampant in pre-1500, because some person once paid money to somebody to *find* the connections they wanted to have.

Do you have the original parchment documents? Can you decipher the ancient handwriting? Is there little damage from water, and sunlight fading, and insects, not so severe that some quality data transcription can truly be made from what still exists on the parchment? Are you experienced in interpreting the convoluted sentence structure that commonly existed in documents centuries ago? Are the variable phonetic spellings on the document close enough to modern words that the meaning of each phrase is really clear?

So the farther removed you are in expertise from each of those conditions, the more that possible errors are introduced, sometimes from wishful thinking. I know of one such document, or summary of such a document, in my lineage. And interpreting the thing is almost impossible to my eyes. I *think* I have figured out what it says. But I am not clear enough about it to publicly create the ancestors. And that only goes as deep as the 1400s.

So in dealing in this ancient era, when people say they have *sources*, what they actually have is a third-hand interpretation of something that somebody had once transcribed, and maybe amended, re-interpreted, maybe lost the trail of who did what with the info along the way. Unless your source is a professional genealogy journal item by a pro who has actually examined the original document and with the competence to read it properly.

Well, Dennis, the thing that prompted me to write the essay is because a new member created a duplicate 1650s profile, among several, and I set a merge pending.

Then the same person a few days later, added and atttached new duplicate parents to that pending merge profile. I mean, the merge pending was already linked right there. But the person failed to even look at the comparison, and see that original parents were already attached.

Keep in mind that this person had to pass the pre-1700 test in order to do all these inexplicable actions, after passing. And this sort of thing is not unique. It happens often. Even with identical spellings.

So my essay is a meager attempt to try to educate, to innoculate against such actions, maybe with a bit of shock value to get noticed.

But my personal wish is to enforce a universal moratorium against any pre-1700 profile for new members. Make them wait a month or three, no exceptions. That would at least give them time to poke around in the ancient profiles a bit first, to actually see what is already there. But I get resistance every time I suggest such a thing. So my essay is a different approach for me to try to solve this problem.

Most of my early sources come from HoP (History of Parliament) and VCH (Victoria County History), the authors of which ARE experts in interpreting early documents.

The other major source (which I have in transcription), the Tropenell Cartulary, attempts to discredit my ancestors to justify Tropenell's underhanded means of acquiring their lands.  No sugar coating there. Maybe because I studied (briefly, not my major) Old and Middle English in college, I do not have much difficulty interpreting it.  The bigger problem is interpreting the  internal inconsistencies.

Where the records exist, I actually find the 15th century genealogy easier/ more certain than the 18th century genealogy.

I am quite comfortable with your suggestion of a restriction for NEW members (with exceptions where appropriate)  But not for longer term members.
I too am extremely suspicious of any NEW pre 1600 profiles. I only trust existing proven pedigrees. Realistically, the chances of discovering some new unknown church document, land deed, Will, etc. - which uncovers the existance of new undiscovered ancestors - is so unlikely as unlikely as finding a newly discovered Hampton Court letter chest at a village antique shop. I consider it impossible.  Existing pre 1600 documents which have been saved and protected, have already been analyzed, and analyzed by experts in Middle English language style, law, ietc. My approach to my genealogy efforts is to trace my New World 20th, 19th, 18th, 17th century ancestors back to their "gateway" ancestor. Once I have linked in to a proven accepted Pedigree, i consider my work done. It would be presumptuous of me to dabble with and change Renaissance and Medieval pedigrees. I don't have the primary documents and i certainly don't have the expertise. I feel that it's important to maintain  humility on Wikitree and recognize one's limitations.
I think with more and more European cities / towns putting their histories and records on line we are finding access for the armchair genealogists to records from the 1600s that were not readily available before.  

I find this particularly true in France and German records today.  Most of those were not in either Ancestry or Family Search (might have been on microfilm in Family Search but had not made it online).  I have copious examples of this in my possession.  

France is actively working to get all records on line.  You have to be able to read French and Latin to make sense out of them but they are there.

Baden is doing the same in Germany.  I think Westphalia just did the same.
Christine may be correct when referring to British "sources", but my experience with Dutch pre-1600 sources follows Laura's comments. The Dutch digitalized their records after WWII, with most of them transcribed in modern Dutch. Both versions are easily available, as well as the original document, nearly always free of charge. Are these NEW?  To the archivist no, to me yes. The armchair genealogists, including me normally computer search birth, wedding, death documents by last name. Dutch pre-1600 sources are largely notarial and court documents, and require much more time to search. I extended my tree from 1590 back to a document of sale dated 1358 of a Jan van Weert and his wife Mette Jacobsdaughter selling a parcel of land to a nearby monastry (which kept the document all these years). I have not added these people to WikiTree.  Should I?
Gus I have pre-1700 and pre-1500 certification and have added many but not all of the people in my lines who fit these time periods.  I do try to find multiple sources.  Part of the problem is I may have a document that shows 3 people all sharing the same name... because these documents are for guilds and not genealogical they do not indicate relationships unless you luck into one that says so and so is the son of such and such.  We know they are all related because to be in the guild you had marry within the guild and these were very closely held family secrets.  We just don't always know if they are brothers, cousins, uncle and nephew, grandfather, son, grandchild.  What makes it worse is they may all have the same name.  So a lot of genealogy for this time period is guessing based on other known pieces of data like  (I am going to append a number 1, 2, 3 so you can follow this)  Jean 1 died in 1645 now we can guess he was likely born before 1625 if he was working as a master glass maker because that title typically took a minimum of 15 years.  Kids as young as 10 began journeyman status so the latest he would have been born was 1625 but it could be earlier.   Jean 2 took over the running of the glassworks in 1645 when Jean 1 died.  OK,  is he a master glass maker.. if so then it pushes his birth to 1625 too.. so are these brothers with names like Jean Paul and Jean Nicholas but the family call names of Paul and Nicholas did not make it onto the official document or is this father and son so it would then push the birth of Jean 1 back another 20 years or so?  Jean 3 has a birth record that shows a Jean (is it 1 or 2) as a father and a witness listed as a grandfather also listed as Jean.  So we now assume but have to explain the assumption here that Jean 1 is the father of Jean 2 and the grandfather of Jean 3.
Three cheers to the Dutch for dititizing their records. After reading some responses to my previous comment, I see that I have much to learn about ways to find new "old" ancestors.
As to old documents, handwritten originals, I've looked at lots of microfilms of them. These are all pre-1650, going back to 1400. Mostly land transactions.  Luckily all are in Latin, not Scots Gaelic! There are books, readily available in our library (admittedly the second largest University one in the world), which carefully describe the legal language of those days and also the "secretarial script" used, a sort of shorthand. I do, however, remember my middle and high school Caesar's Latin, which helps with the endings and grammar, which remain more or less the same to this day.

It wasn't really hard actually, to do the job. I did manage to prove the connection implied, but only implied, in a grant of (Scottish) arms to an 18th century cousin of mine. This line is Magna Carta project certified on Wikitree.

But this really isn't the point of Wikitree. Its NOT about original research, especially in that time period. The place to discuss that remains the old Usenet group soc.genealogy.medieval. Wikitree is about entering reasonably solid connections of existing lines.
Your point isn't lost on me, Steven Mix! I just think that the nerds are getting excited about the maths.

Let's think about it: if the project coordinators and members who own the most profiles didn't have all this useless work to do, they would have more time answering real questions and dealing with real issues, like family members who aren't given the right family ties. Steven happens to be one of the persons responsible for some of the profiles on which I worked (New Netherlanders), and a number of problems that I flagged where never corrected. It must be a couple of years, now!

This is a collaborative tree, not your own personal playground. Don't waste people's time! Take time to learn how to use the platform, and make sure that your research is legitimate before you share it with the world.
+14 votes
I have a different method, that generates an actual, not theoretical result.  I am the admin of the Clan Donald USA DNA Project.

The "Clan Founder" (which he sort of was) was named Somerled.  He died in the year 1164. We are tracing his PURE MALE LINE descendants, i.e., bearing his Y chromosome. He had no surname, the surname starting more or less with his great grandsons. But paper trails say that the great radiation of the surname McDonald began with his 3rd great grandson who was named John, born before 1324 d.1387. We have paper trails of superb quality and matching DNA from three of his sons.

In our project we have about 900 men named M(a)cDonald or McDaniel (the southern US form, from the Gaelic pronunciation as heard by the boat-dock registrars.)  182 of them are male line descendants of John, as proven by the very unique DNA. That's 20%.

I surveyed online statistics from Scotland, England, Ireland, Canada, the USA, Australia, and South Africa. I estimate that there are 230,000 men with matching surnames there.   That works out to about 46,000 pure male line descendants of one man, living in 1350. Its actually more as there are some named MacDowell and the occasional adopted name.

Then of course you have to try to extrapolate to the descendants not in the pure male line.
by James McDonald G2G6 Mach 1 (11.3k points)
edited by James McDonald
My first cousin on my mother's side is married to a McDonald male (South Africa) ...
+12 votes

You make a good point - that someone alive in 1700 would have a lot of descendants alive today - but your numbers are massively out. This is for all the reasons given above - in particular both endogamy which reduces the numbers but also the number of 4 children per person is inaccurate and I don't think 25 years per generation is accurate.

For the average number of children, you can work this out relatively easily by looking at the population of the country in question and the natural increase (ie excluding migration) with time. I suspect the number will be much less than 4 - depending on the country I would expect less than 3.

Regarding the time per generation, most people started having children about age 25 but then continued for 10-20 years. I find using three generations per century more accurate for my family.

Given that you have answered your question both in the title ("How many descendants does a pre-1700 ancestor have? 67,108,864") and in your description, could I ask you to amend these to reflect these challenges?

by Andrew Turvey G2G6 Mach 4 (43.3k points)
Andrew, in the example above, I showed that the growth rate of the American population has been historically 2% per year. That number came from an actual statistical source chart, and accounts for annual growth after both birth and deaths, with a relatively minor effect from immigration.

I showed how the rule of 72 means that the population doubles every 36 years at that 2% rate.

I also showed how a village of 100 then becomes a village of 150 in a quarter century. Or a family of 6 in the village becomes a family of 9 in that time. The number of 4 children having 4 children still seems fairly accurate to me.

But I will concede that the total number of descendants is coming out too high by the naked math. And I think it is mostly a factor of the number of generations in the calculation. Endogamy may happen a lot in early generations, which will reduce the number of generations in the math. And that is where the big reduction happens.

Also what actually happens is that the exponential effect is not smooth, over a long time. it will randomly get interrupted on each of the branches along the way, and then have to restart the exponential growth from a lower than expected base.

So, for example, if say endogamy reduces the effective number of generations to 12 instead of 13, the total count is reduced by 4, to 16 million.

I don't think I want to alter the premise, because I want to show how the math works. And the actual final number is not quite so important, whether it is tens of millions or tens of thousands .The point of the essay is simply to encourage people to pause and think about the ramifications of the exponential descendancy, before they create that 1620s profile.

In other words, it is not just another profile that they are adding. It is rather the root ancestor of another tens of thousands or of tens of millions of more descendants.
I would guess that a LARGE contributor to the _US_  growth of 2% per year is immigration rather than reproduction.

Checking back my own ancestry I find my Great grandparents where born ~100 years before me (all I have definite DOB for are within 10 years of that), so 3 generations /100 years does seem more reliable.

The 4 children pre generation also looks like an exaggerated figure. AFAIK none of my great grandparents had more than 12 grand children. It seems that 3 children per generation getting married is closer to the truth among the data I have available, even with some ancestors marrying at least 3 times. In several places ancestors from the 1800's have only 1 or 2 known grand children despite their large number of children.

Using more realistic values of 33 years per generation & 3 children pre generation your calculation gives 310 which is 59049. Still quite a sizeable figure but WAY down on the 67 million estimate.

+9 votes

Pedigree collapse

https://en.wikipedia.org/wiki/Pedigree_collapse

"Without pedigree collapse, a person's ancestor tree is a binary tree, formed by the person, the parents (2), the grandparents (4), great-grandparents (8), and so on. However, the number of individuals in such a tree grows exponentially and will eventually become impossibly high. For example, a single individual alive today would, over 30 generations going back to the High Middle Ages, have 2 30 {\displaystyle 2^{30}} 2^{{30}} or roughly a billion ancestors, more than the total world population at the time.[2]

This apparent paradox is explained by shared ancestors, referred to as pedigree collapse. Instead of consisting of all different individuals, a tree may have multiple places occupied by a single individual. This typically happens when the parents of an ancestor are related to each other (sometimes unbeknownst to themselves).[3][4] For example, the offspring of two first cousins has at most only six great-grandparents instead of the normal eight. This reduction in the number of ancestors is pedigree collapse. It collapses the ancestor tree into a directed acyclic graph.".

by
+8 votes
The Wall Street Journal had an article not long ago by a Statistician saying if each generation had two kids and mother/father, when you reach 133 generations, you will have ONE TRILLION ancestors.

He didnt think that most families had 6-12 kids (no birth control).
by Christopher StJohn G2G5 (5.4k points)
+9 votes

There is a mathematical point I want to make that I didn't see in the prior "nit-picking".  People have mentioned that some of the children will not live to reproduce.  The "zeroing" of a population can actually occur after several generations, and I think this is more common than many would expect.

In mathematical terms, you have given in your post an "exponential model" for the growth of a population, say, the descendants of John Smith born 1650.  The descendants of John Smith are a value that changes over time, and you could "model" it with all sorts of models.  The exponential model is wonderful because a mechanism explains how it arises (i.e., your explanation of how to arrive at 4^13 after 13 generations) and it is very simple, so easy to work with.  But any work like this is creating only a model of the world, not describing the world for certain.  It is a mathematical abstraction of what will happen, and while the real world is certain to differ from what your model says, you try to pick a model that will most closely describe what happens in the real world.  People have already explained how endogamy makes this model unrealistic after enough generations.

One of the most common ingredients to include that complicates the exponential model but also makes it more realistic is an "extinction threshold".  This is a threshold where if the population is low enough, and things go wrong, the population goes to zero. There would be some work to implement this (you'd need to move to a stochastic model, blah, blah, ... don't worry about that), but the point is, the long-term behavior of a population will end up doing two drastically different things:  growing or heading to 0 and then staying put.  It may take many generations to reach 0, but once it gets there, it is certain never, ever to grow.

My point is that the process of going to 0 can take more generations than one would expect.  I offer as an example Susan Jackson

https://www.wikitree.com/wiki/Jackson-27869

Susan was probably born around 1800.  She was one of six illegitimate children of James Jackson of Ballybay, a scion of a fabulously wealthy family from County Monaghan in Ireland in the 1700s.  They were politically active, and most of the family fled in the late 1790s and early 1800s.  

I am descended from Nothern Irish Jacksons, and a fourth cousin on my line found a letter written to her aunt from a woman named Lillian Virginia Koller in the early 1900s and sent it to me.  It was clear that it contained a transcription of the family bible of James Jackson of Ballybay, so I was very excited. No one seems to know this bible existed, or at least existed within memory.

With some help, I eventually traced Lillian's ancestry back to Susan.  Then I went looking for living descendants of Susan hoping to find one who may know about the bible, and so far as I can tell, there may be none, with the last descendants passing away around 1950.  You can see that she had quite a few descendants:

https://www.wikitree.com/genealogy/Jackson-Descendants-27869

(Actually, I didn't create profiles for them all because it got got depressing creating so many profiles for terminal lines.)

I have written to the Staten Island Historical Society Museum to see if the bible ended up in their collections.  They said that it is common for the last known scion of a house to pass away with a bible and that if no other relatives can be found, it may be bought at auction by a museum or historical society.

P.S.  I have done a lot of work on identifying people in the 1774 Rhode Island Census, which quite often includes going back and looking for ancestors into the 1600s.  There are many such New England profiles that do not exist yet on Wikitree.

by Barry Smith G2G6 Pilot (287k points)
+7 votes
I would like to offer a possible solution where uncertified and newbies can "practice" but do no harm.   What if we have a free space page where new pre-1700 profile data is entered (not into a form like a real profile but following the basic informational structure). Then badged folks and project folks like French Roots, German Roots, etc.  get to look at it and discuss it in G2G before it is actually added to the tree.   Things like spurious sources can be discussed, conflicted information can be researched and debated.  This would give everyone who has a vested interest a chance to review, collaborate, and refine before someone is actually added to the tree.   And if someone is not ready for prime time they still exist for reference in the free space page.   

What do you think?
by Laura Bozzay G2G6 Pilot (824k points)
It would be wise to factor in the percentage of men who do not have offspring. Childhood mortality should also be factored in. Any data on how many women didn't have children or how many women remain spinsters subtract a certain amount could still have children. Each little thing would play a roll in increasing the accuracy of any said estimate. Bottlenecking always occurs to some extent in the majority of relationships. It is most common that couples will share common heritage especially the further back in time you go. This is effect is certainly amplified by the fact people on average are more likely to be attracted to someone with similar genes to themselves, that is to say more related than a randomly selected person from the same population.
+5 votes
Is this a math problem or a genealogy question. No matter what the number, it’s a lot of people. WT isn’t even close yet, but we’re working on it.
by Bob Keniston G2G6 Pilot (260k points)

Good aim, within reason, but let's make sure we don't make people twice, or make profiles for people who didn't exist. :)

In my honest opinion I do think that when we are balancing the aims of quality and quantity, it is better for Wikitree to aim at quality. Creating masses of profiles on a website like this is not really so difficult, and other websites probably do that better. 

We face the biggest challenges when it comes to quality, because we are merging the family trees of millions of people, and that inevitably leads to types of problems which can be quite difficult to find and fix.

+6 votes
Unfortunately, from the tone of the end of the article, I come away with the impression that you wish renders to feel they are doing all the work on every relative for any generated profile they manage. Or that no one is good enough to take on this role except those that are already doing it. I doubt either interpretation was intended, but worry it may discourage some.
by Doug McAdams G2G5 (5.7k points)
+8 votes

I'm currently working on my Norwegian 17th century ancestors. Apart from 5 millions living Norwegians and a miliion or two Americans with ancestors from Norway, I wonder where the other 60+ millions postulated descendants of theirs might be hiding.

I don't even think they are the ancestors of all currently living Norwegians. Some fraction of them, surely. But not all, by a long way.

The population of Norway in 1400 has been estimated to between 250,000 and 400,000. The 1801 census counted about 800,000 persons, that is roughly a doubling in 400 years. The annual growth from the lower 1400 estimate is .275 %, and from the higher .174 %. This indicates that the average family had just a tiny fraction above 2 children who went on to reproduce.

"I start with a presumption that each ancestor has four children."

If an ancestor had 4 children, does that mean that his wife also had 4 children? It doesn't make sense to talk of individuals who procreate, it always takes a couple.

It should also be taken into account that in preindustrial societies, endogamy was rampant.

Anyway, it is clear that a person who lived in the 17th century either has a large number of descendants, or none at all. But I doubt that any person born in the 1600s has more than at most a couple of millions living descendants. And maybe a few thousands who care.

by Leif Biberg Kristensen G2G6 Pilot (205k points)
edited by Leif Biberg Kristensen

Yes, you are right, but it is arguably a question of defining terms. It is a good point to clear up.

In this thread, the way I understand it, the numbers being simulated are not going to be the number of different ancestors. They are, to search for a term, the number of positions in a family tree in a specific generation.

In other words, as I think you also realize, the same people will tend to appear many times when you get this far back.

In fact, if you go back far enough you will reach a point where everyone with descendants, who was alive at that time, is an ancestor.

The presumption that "each ancestor has four children" is simply wrong. Or, at least, it's far too simplistic to be used as a model. If it, as you say, should be taken to represent a "number of positions" in a family tree, I regard that as a rather pointless exercise, somewhat analogous to turning the Sosa-Stradonitz system on its head. (Which naturally won't work, as a couple actually has an arbitrary number of children.) It's of course the number of unique descendants that really counts.

A couple of ancestors may have four children. Those four children may go forth and marry four other individuals, and then these four couples will produce 16 children. So far, so good. But hey, where do the spouses come from? Of course they come from the general population in the area, and already in the second generation of 16 grandchildren you may expect some intermarriage.

I have done an extensive study on the descendants of a peculiarly prolific pair of my own ancestors, Jon Jensen and Anne Christensdatter. I've still got a lot of them waiting to be entered into WikiTree, but there are enough to illustrate my point.

If you go to Anne's profile and click "Show descendants", there's a really long list. But if you scroll about halfway down, you will see the legend [N has already appeared on this list; see descendants above] on a fairly large number of them, actually on 77 of the 2,213 profiles. Those 77 lines represent whole branches. You may also notice how many of the descendants died young, in each generation about one third.

And, thank you, after 25 years of genealogy studies, I'm well aware of how an ancestor may appear multiple times in a pedigree. Jon and Anne appear only 4 times in mine, but I've got another ancestor, Tyge Andersen (c. 1580-c.1650), who appears 15 times in my pedigree. Which perhaps is to be expected, as he is at least one generation above Jon and Anne.

Edit: fixing typos

+5 votes

The math is off by an exponential factor.

For example, Zacharie Cloustier born around 1590 and married in 1616, who migrated with his wife to New France, is considered to be the ancestor of French origin who has the most descendants in North America.  He died in 1677, with a total of 6 known children for the couple.

The descendants have been added up for the year 1729, they numbered 2,090. (source on profile).

The numbers you give are exponential, which is an error.  Zacharie is my ancestor through 9 different paths from what the relationship finder tells me.  Endogamy abounds in my tree.

by Danielle Liard G2G6 Pilot (647k points)
+5 votes
The math is way off; the most descendants I'd expect a single 17th century couple to have is around 500K-1M, give or take, and even then this really only applies to the descendants of colonial immigrants to North America; Europeans who didn't emigrate have fewer descendants, because A) Europe was more urbanised (cities were population sinks, before the mid-19th century, because of disease and poor sanitation) and B) 200+ years of European wars truncated many lines who now have no descendants.
by C Handy G2G6 Pilot (208k points)

Related questions

+28 votes
0 answers
+7 votes
2 answers
212 views asked Jun 8, 2015 in Genealogy Help by Bea Wijma G2G6 Pilot (305k points)
+12 votes
2 answers
474 views asked Apr 26, 2015 in Genealogy Help by Bea Wijma G2G6 Pilot (305k points)
+4 votes
1 answer
181 views asked Mar 29, 2018 in WikiTree Tech by Peter Joseph Promper G2G2 (2.3k points)
+6 votes
3 answers
+5 votes
2 answers

WikiTree  ~  About  ~  Help Help  ~  Search Person Search  ~  Surname:

disclaimer - terms - copyright

...