24
$\begingroup$

If I were to count my father, my grandfather, my great-grandfather, and so on up till, say chimps, or the most common ancestor, or whatever that suits the more accurate answer, how many humans would there have been in my direct lineage?

And would it be almost the same for every human being currently living?

Improve this question
$\endgroup$
5
  • 1
    $\begingroup$ Closely related question: Initial population when i count backwards? $\endgroup$
    – fileunderwater
    Sep 22, 2014 at 11:32
  • 1
    $\begingroup$ Note that when pedigrees collapse, there are not necessarily the same number of generations on each leg. So even for an individual there isn't a single number which is "the number of generations back to the last human/chimp common ancestor" since the number depends which route you take, although you could for example take the minimum. Consider for example Prince William (Windsor) whose parents were 7th cousins once removed (users.uniserve.com/~canyon/ancestor.htm) as well as other relationships. This isn't specifically aristocratic inbreeding, they're just better at keeping records :-) $\endgroup$
    – Steve Jessop
    Sep 22, 2014 at 12:05
  • $\begingroup$ @SteveJessop: I think when you only consider male ancestors, there's only one route, which bypasses that issue $\endgroup$
    – Mooing Duck
    Sep 22, 2014 at 17:51
  • $\begingroup$ @MooingDuck Whether you consider female ancestors or not is irrelevant if you are looking for a common ancestor, because the amount of time between each generation differs at each link in the chain. Hence the "once removed" part can occur where these differences added up to a whole generation. $\endgroup$
    – Michael
    Sep 23, 2014 at 5:28
  • $\begingroup$ @Michael: Ah right. I was pointing out that each person only has a single chain, thus pedigree collapse and "depends on which route you take" are incorrect. However, you raise a good point that not all people's chains are the same length. $\endgroup$
    – Mooing Duck
    Sep 23, 2014 at 16:29

4 Answers 4

Reset to default
26
$\begingroup$

A quick back-of-the-envelope answer to the number of generations that have passed since the estimated human-chimp split would be to divide the the split, approximately 7 million years ago (Langergraber et al. 2012), by the human generation time. The human generation time can be tricky to estimate, but 20 years is often used. However, the average number is likely to be higher. Research has shown that the great apes (chimps, gorilla, orangutan) have generation times comparatble to humans, in the range of 18-29 years (Langergraber et al. 2012).

Using 7 million years and 20 years yields an estimated 350000 ancestral generations for each living human. A more conservative estimate, using an average generation time of 28, would result in 250000 generations. However, some have argued that the human-chimp split is closer to 13 million years old, which would mean that approximately 650000 generations have passed (using a generation time of 20 years).

The exact number of ancestral generations for each human will naturally differ a bit, and some populations might have higher or lower numbers on average due to chance events or historical reasons (colonizations patterns etc). However, due to the law of large numbers my guess would be that discrepancies are likely to have averaged out. In any case, the current estimates of the human-chimp split and average historical generation times are so uncertain, so that they will swamp any other effects when trying to calculate the number of ancestoral generations.

However, this is only answering the number of ancestral generations. The number of ancestors in your full pedigree is something completely different. Since every ancestor has 2 parents, the number of ancestors will grow exponentially. Theoretically, the full pedigree of ancestors can be calculated using:

$$N_\text{ancestors} = \sum_{i=1}^t 2^i \hspace{1em} \text{or} \hspace{1em} 2^{t+1}-2$$

where t is the number of generations. However, this will yield an unreasonably large number of ancestors (~$2.3*10^{105}$ over just 350 generations), since it assumes that all of your ancestors are unrelated. Basically, you run into something called pedigree collapse, which means that your pedigree will have many overlaps (due to inbreeding, overlapping generations etc), so all of your ancestors cannot be seen as unrelated. In practice, this means that the number of ancestors in successive generations will stop doubling for each generation you go back, and will eventually start shrinking, which will drastically reduce the number of unique ancestors. For more on this, see answers to the question Initial population when i count backwards?. That question is basically asking about the same issue, but less specifically. There might be studies that have tried to calculate the average number of ancestors to living humans, but I haven't seen any. Therefore, I cannot say anything more specific about the average number of unique ancestors for a living human individual.

Improve this answer
$\endgroup$
7
  • 1
    $\begingroup$ This estimation is the number of generations, not ancestors. Ancestors should double after every generation, since a human has a father and a mother. Thus, the number of ancestors in 350000 generations should be 350000^2 = 122500 million humans. (Sorry, deleted previous comment for it being not really precise. My mistake there) $\endgroup$
    – ederbf
    Sep 22, 2014 at 9:27
  • 1
    $\begingroup$ I have a question about this answer. I know the OP specified males, but... (if it were only male lineage) wouldn't it look more like a tree with the single trunk pointing to him, since he has two grandfathers, and they each have two grandfathers, so 2 + 4 + 8 + 16 + 32 + 64... for x generations of grandfathers? $\endgroup$
    – anongoodnurse
    Sep 22, 2014 at 9:28
  • $\begingroup$ @ederbf As I've added, I read the Q as asking for the number of "direct" ancestors. I agree that the full pedigree is something completely different, and have added a section on this, with a link to a related Q asking specifically about the full family tree. $\endgroup$
    – fileunderwater
    Sep 22, 2014 at 9:37
  • 3
    $\begingroup$ @ederbf Doubeling in every generation is 2^350000, not 350000^2. Exponetial, not polynomial. So the actual number would have been about 10^115000 - Except for the pedigree collapse that fileunderwater describes. $\endgroup$
    – Taemyr
    Sep 22, 2014 at 10:31
  • $\begingroup$ @Taemyr, That is true. I rushed too much writing, editing and such... $\endgroup$
    – ederbf
    Sep 22, 2014 at 10:38
4
$\begingroup$

Everyone on earth shares a single most common recent ancestor around 3500 years ago (source), and given the vast depth of time between this ancestor and the human/chimp split it seems reasonable to assume that the number of ancestors derived via this route vastly outweighs all other ancestors so it makes sense to conclude that the difference in number of ancestors between people is a small proportion of the total number.

As for estimating how large that number is, I'm really not sure how you'd do it.

Improve this answer
$\endgroup$
7
  • 4
    $\begingroup$ Are you claiming that two people from the native populations of Australia and South America share a common ancestor 3500 years ago? $\endgroup$
    – Joe
    Sep 22, 2014 at 17:51
  • 3
    $\begingroup$ @JackAidley: en.wikipedia.org/wiki/Most_recent_common_ancestor "the human MRCA may have lived 2,000 to 4,000 years ago. This estimate is based on a non-genetic, mathematical model that assumes random mating and does not take into account important aspects of human population substructure such as assortative mating and historical geographical constraints on interbreeding." $\endgroup$
    – Mooing Duck
    Sep 22, 2014 at 17:55
  • $\begingroup$ @MooingDuck: Perhaps you'd like to read the link I gave. $\endgroup$
    – Jack Aidley
    Sep 23, 2014 at 8:16
  • 1
    $\begingroup$ @Joe: It's a model. Any model will be, by it's nature, limited. The limitations are discussed in the paper, as is there potential impacts on the resulting number. $\endgroup$
    – Jack Aidley
    Sep 23, 2014 at 10:43
  • 2
    $\begingroup$ @Joe: You misread the abstract. "the random mating model ignores essential aspects like...geographically separated groups. Here we show that recent common ancestors also emerge from two models incorporating substantial population substructure." nature.com/nature/journal/v431/n7008/fig_tab/… shows that they do indeed take into account geographic seperation and migration. $\endgroup$
    – Mooing Duck
    Sep 23, 2014 at 16:35
2
$\begingroup$

Another factor not mentioned is that the number of generations will be a range even when accurately calculated for one individual. If, for example, my dad's dad's dad's etc generation was 25 years on average and my mum's mum's mum's etc generation was 20 years then you don't have to go back very far to have for them to be different. In the example 100 years ago in my dad's side is 4 generations and on my mum's side it's 5 generations.

Improve this answer
$\endgroup$
0
$\begingroup$

We could try to cut the answer out:

  • the Population Reference Bureau estimates that about 107 billion people have ever lived
  • today's population is about 7.6 billion
  • only 0-14 of them which are in your direct linage (parents up to grand-grand parents) are alive
  • world population of year 1500, when exponential growth started, is estimated to about 425 million (https://www.wolframalpha.com/input/?i=world+population+1500). This is about 25 generations ago. Almost all of them are not your ancestors (by integration you should get a number of about 20 billion people which are NOT your ancestors, while about 500.000 are)
  • world population 8000 b.c. is estimated to about 5 million, and 10000 b.c. to about 1 million.
  • majority of people before 10000 b.c. might be your ancestors. This could well be more than 10 billion people (much insecurity and definition problems here)

.. which means, we "only" need an estimation of 10000 b.c. to 1500 a.c. After playing around with Excel, I believe, it should be in the ballpark of 1 billion.

hence my best estimate would be 11 billion (10 billion before 10000 b.c. and 1 billion after).

Improve this answer
$\endgroup$
3
  • $\begingroup$ mixed people should have more ancestors. My bet is that Children of mixed Mexicans and Chinese have most ancestors. $\endgroup$
    – Ralf Anders
    Apr 29, 2019 at 2:54
  • $\begingroup$ at some point back in history, more than 2 children on average (per parent couple) go into your ancestor tree. This is why additions shrink further back in time. $\endgroup$
    – Ralf Anders
    Apr 29, 2019 at 2:55
  • $\begingroup$ billions of people died throughout mankind history without having siblings living today. $\endgroup$
    – Ralf Anders
    Apr 29, 2019 at 2:57

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .