One of the many things that annoy me is the prevalence of fallacies about population issues. So to vent my spleen I’m planning a few notes on common population fallacies.
I’ll start with the idea that in the ‘old days’ (paleolithic, neolithic, medieval, pre-industrial, or whenever) everyone died young.
It’s quite true that in pre-20th century societies the average lifespan, or life expectancy at birth, was usually between 30 and 40, as compared to over 70 in most Western countries now. From this it is inferred that hardly anyone lived beyond the age of 45, that someone would be considered old at 50, and that real old age (70 or older) was extremely rare.
But I trust it is obvious to GNXP regulars that average (mean) lifespan tells us very little about the distribution of mortality. An average lifespan of 40 would be consistent with half the population living to 80. To say anything useful about the age structure of the population we need to know not just the mean but the distribution of lifespans.
Pre-20th century mortality patterns varied, but the main feature of all of them was that infant mortality was high. Typically over 15% of children would die in their first year, and between 30% and 50% would die before age 10. These high infant mortality rates had a dramatic effect on life expectancy at birth, dragging it down well below present levels.
But for those who survived the perils of childhood, life expectancy was not so bad. Admittedly, mortality between age 20 and 60 due to epidemic disease (smallpox, cholera, etc) and chronic infections (syphilis, TB), was higher than we would like, but from 20 up to about age 55 the risk of death in any given year would only be between 0.5% and 2% – worth saying your prayers to avoid, but not worth losing sleep over. There was a reasonable prospect of reaching old age.
These points can be illustrated by some figures calculated by Charles Babbage based on the experience of the Equitable Life Assurance Society in 18th century England. Of a given cohort at birth, based on this pattern, we would expect 65% to survive to age 10, 62% to survive to age 20, 57% to age 30, 51% to age 40, 45% to age 50, 35% to age 60, 25% to age 70, 14% to age 80, and 2% to age 90. This gives a median life expectancy at birth of about 41, and an average of about 39 (assuming that infant mortaliy is concentrated in the first year.)
This is lower at every age than we expect today, but even so, a quarter of all those born could expect to live to the age of 70, which is not bad going. If we consider the life expectancy of those who had survived childhood to reach age 15, about half would live to age 63, a quarter to age 78, and a tenth to age 85. About 3% would survive to age 90, which is rare but not extraordinarily rare.
For amusement I have also checked the lifespan of major C17 and C18 philosophers. To the nearest year this gives: Bacon 65, Berkeley 68, Condillac 65, Descartes 54, Diderot 71, Hobbes 92, Hume 65, Kant 80, Leibnitz 70, Locke 72, Malebranche 77, Priestley 71, Reid 86, Spinoza 44, Voltaire 83. Average lifespan 71.
Of course these aren’t representative samples. People who die young seldom become famous philosophers, and the customers of the Equitable were all wealthy enough to afford life insurance. Probably a cohort of peasants or coal miners would have fewer people living to real old age. But the figures do show that even with average life expectancy below 40, there can be lots of old people around. There may be some societies or occupations where the stereotype of ‘all dead by 50’ is accurate, but this needs to be proved from specific data, not a general assumption.