One life at random from everyone who was ever born

Everyone Who Ever Lived

History remembers kings, generals and inventors. Almost no one who ever lived was any of those. This page draws one human being at random from all the births there have ever been, weighted by how many people were actually born in each age and place, and tells you what that life most likely looked like.

or press Space
Deal a hundred lives the statistics, made visible

A hundred draws from the current pool, with the birth-rate assumption you have set. Every line is a link to that life.

Compare with your own life where your birth ranks among everyone ever

Enter where and when you were born. The page places the conditions of that birth against the births of every human who ever lived, under the current birth-rate assumption.

Under the hood how the draw works, and what it assumes
The draw

Every draw is a random birth. A year is chosen with probability proportional to the number of babies born that year: the world population path (a founding population of about ten thousand people 190,000 years before the present, two million by 50,000 BCE, then the HYDE 3.3 reconstruction from 10,000 BCE and the UN series after 1950) multiplied by a crude birth rate. A region is then chosen with probability proportional to its population in that year times its own birth rate, and sex, survival, family, work, name, faith, food and roof follow from the region's conditions in that year. The same link always reproduces the same life.

The assumption that moves everything

Nobody knows the birth rate of the deep past. The Population Reference Bureau assumes 80 births per thousand people per year before year 1 and 60 until 1750, which is how it reaches its famous 117 billion humans ever born. Those rates would mean a life expectancy of 12 to 17 years in a stationary population. A moderate schedule of about 40 per thousand, consistent with the life expectancy of 25 to 30 that skeletons and forager studies suggest, gives fewer humans in total and a larger share of them alive recently. The slider runs between the two.

moderatePRB

Moving the slider keeps the seed of the current life but re-dates the person: the same quantile passes through a different curve, so a 1400 birth under the PRB rates becomes roughly a 1550 birth under the moderate ones.

Known limits
  • The population path uses HYDE 3.3 from 10,000 BCE, whose estimates for antiquity and the Middle Ages are below the ones the PRB used, so even with the PRB birth rates the total here is about 105 billion rather than 117.
  • Regional shares before 1000 BCE are hand-set from dispersal dates and McEvedy and Jones, because HYDE allocates prehistoric people to the Americas long before anyone lived there.
  • Survival comes from a Gompertz-Makeham life table fitted so that its mean equals the region's life expectancy at birth in that year (Riley 2005, UN), with child mortality derived from life expectancy where no measurement exists. Causes of death are illustrative draws, not reconstructions.
  • Names are common attested names of the region and period, drawn as "a name typical of your people"; peoples whose names were never written down are left unnamed. Religion, food and dwellings are typical of the region and era, not of a household.
  • Quality-of-life indicators are typical values for a region and era, not for a year. Occupation shares are rough, and women's work is counted as work.
  • Every life is computed in this file; nothing about it is sent anywhere. The site keeps two anonymous totals, visits and lives drawn, and cookieless page-view statistics. Open the console and run AHE.selfCheck() to compare 20,000 draws against the model.
Sources