Won't you let me live?

July 21, 2006 1:05 PM

Starting with Sumer, approximately how many human beings have died?

Also, at what point will there be more people alive than have ever died?

posted by milarepa to Grab Bag (11 answers total)

It turns out that it's the nature of exponential population growth that at any given time the number of people alive exceeds the cumulative number who have ever died -- if the exponential growth rate is high enough.

For much of human history, after we mastered agriculture, it has been. Which means that for most of the last 5000 years there have been more humans alive than have ever died.

[Of course, that depends on what you consider a "human being". Australopithecus?]
posted by Steven C. Den Beste at 1:11 PM on July 21, 2006

Check out Carl Haub's article How Many People Have Ever Lived on Earth? from Population Today. Also John Durand's Historical Estimates of World Population: An Evaluation from Population and Development Review.
posted by j.edwards at 1:13 PM on July 21, 2006

Carl Haub's conclusion:

So, our estimate here is that about 5.8 percent of all people ever born are alive today. That's actually a fairly large percentage when you think about it.

Of course, that's starting from the beginning, not from Sumer. On the other hand, I'm not sure what Sumer has to do with anything. And in any case, it's certainly not true that "for most of the last 5000 years there have been more humans alive than have ever died."
posted by languagehat at 1:26 PM on July 21, 2006

Response by poster: Thanks for the links.

I chose Sumer as it is a relatively definite starting point (compaired to say that start of argriculture) and one with some sort of meaning, at least to me.
posted by milarepa at 1:37 PM on July 21, 2006

There's a book by McEvedy and Jones titled "Atlas of World Population History" that would be useful for finding more information about specific times and places.
posted by j.edwards at 1:50 PM on July 21, 2006

One interesting thing, to me, from the Haub article is the estimates of infant mortality. For the majority of human history, infant mortality has been around 50% or even higher. That means his 5.8% of all people who have ever lived takes into account a staggering number of people who never even developed fully-fledged consciousness, because the poor little buggers died so soon after they were born.

To my mind, that throws his figure out of whack with what I would "feel" that number should be. I'd rather see something like, how many of the people that lived past age 2 [or some cut-off] are alive today. But that seems harder to do, and I'm not going to do it. I think the number would still be less than 10%, but it's staggering to think how many billions have drawn so few breaths.
posted by teece at 1:51 PM on July 21, 2006

This has come up before...
posted by Xelf at 3:17 PM on July 21, 2006

Here is a more mathematical derivation for those so inclined. It also gives a result of 5% to 10% depending on your choice of starting date and estimate of average lifespan.

The idea that the number of people alive exceeds the number who have ever lived doesn't make a lot of sense even if you assume exponential growth. What that would mean is that the area under a slice of the curve at the far right exceeds the area under the rest of the curve. That obviously couldn't be the case unless you assume that today people have extremely long lifetimes (the slice is very large) or that the birth rate has suddenly increased ( the exponential function has gotten much steeper).
posted by JackFlash at 3:23 PM on July 21, 2006

The idea that the number of people alive exceeds the number who have ever lived ... obviously couldn't be the case unless ... the exponential function has gotten much steeper.

you do know what an exponential is, right? the point is that it gets much steeper.

steven c den beste rightly pointed this out at the beginning of the thread - consider the classic exponential-growth case of a single bacterium, that at the end of its life, divides and becomes 2 cells, and then 4, and so on..

step 1: current living = 1, past living = 0
step 2: current living = 2, past living = 1
step 3: current living = 4, past living = 1 + 2 = 3
step 4: current living = 8, past living = 1 + 2 + 4 = 7

and so on. so for a growth constant of 2, from the very beginning the current population is always greater than all those that came before it. i'm trying to figure out what the conditions on lifespan and mortality are that this be true.
posted by sergeant sandwich at 6:08 PM on July 21, 2006

Sergeant, I think you proved my point. In your example, the only way you get eight bacteria at step 4 is if none of the bacteria ever die -- they have infinite lifespan. The number of current living is the same as the number of past living.

Likewise, for the human population, the only way the area of the slice on the right side of the curve can exceed the area on the left side of the curve is if the slice on the right is unusually large. This implies unreasonably long lifespans. In your example the slice of living people is the entire curve. The other possibility is that that the rate of growth increases dramatically, that is, there is a kink in the exponential growth curve.

By the way, people don't divide like bacteria and all of the bacteria don't divide at the same instant. For humans a reasonable reproduction rate is about 20 to 80 births per thousand people per year.
posted by JackFlash at 7:11 PM on July 21, 2006

Another way to look at it is that while the birth rate is exponential, the death rate is also exponential, but at just a slightly lower rate. For every generation of people that are alive, there is an almost as large number of people that die, and this number of dead people accumulates for each generation in the past. Most estimates are between 90 and 110 billion dead people, at least 15 times as many as are now alive.
posted by JackFlash at 11:46 PM on July 21, 2006