London’s Death Table Listed 22 People Killed by Fright. A Haberdasher Turned It Into Life Insurance Math.

“Frighted, 22.”

That is a real line from a London death table. Twenty two people, across twenty years of counting, listed as killed by fright.

The same table records “Grief, 279” and “Overlaid, and Starved, 529.” The printed bill for the plague year of 1665 carries “Teeth and Worms, 2,614” and “Rising of the Lights, 397,” on a page whose bottom line reads “In all, 97,306.”

These were the Bills of Mortality, the weekly death lists for London’s parishes. When someone died, the parish searchers, whom John Graunt described as “antient Matrons, sworn to their office,” went to the house, looked at the body and named what had killed it.

No doctor signed those verdicts. Graunt wrote that a cup of ale and a doubled fee, two groats instead of one, could change what a searcher reported.

The bills were printed every Thursday and sold by subscription at four shillings a year, so that wealthy Londoners could watch the plague numbers and decide when to leave town. From the last week of 1603 that was, by Graunt’s own account, the entire point of them.

Then a shopkeeper read the back issues.

The haberdasher with a side project

John Graunt sold small wares. Buttons, thread, ribbon, pins, the trade London called haberdashery.

He held no medical degree and no university post. What he had was a pile of old bills nobody else had bothered to add up.

In 1662 he published a slim book called Natural and Political Observations Made upon the Bills of Mortality. Much of it is tables, including a casualty table that runs twenty years and 229,250 burials deep.

He knew the source was garbage in places. He wrote about it plainly, and worked around it anyway.

Then he asked a question the bills had never been read for. Not who died. Whether the numbers repeated.

They repeated

They did. Year over year, most causes took close to the same share of the total, which Graunt wrote up as a “constant proportion unto the whole number of Burials.”

It is easy to miss how strange that was in 1662. Any single death in London looked like an accident of fate.

Stack enough of them and the accident turns into a proportion. The city was killing its people at a rate.

A rate can carry a number. A number can carry a price.

Graunt also used the bills to settle an argument about how many people lived in London. The men of experience he kept company with in the City talked “seldom under Millions.”

His answer was about 384,000. He got there four separate ways, from births, from deaths per household, from the militia rolls and from a 1658 map of the city, and the four routes agreed.

What a table of survivors is worth

The piece that carried furthest was his life table. A life table is a chart of how many of a group are still alive at each age.

Graunt started with 100 “quick conceptions,” his own starting unit. By his count 36 of them were dead before the age of six, 64 were alive at six, 40 at sixteen, 10 at forty six, and one was left at seventy six.

The Royal Statistical Society’s 2020 tribute calls him the first person to build what would now be called a life table. That shape is the product. Everything a life insurer sells sits on top of it.

A life policy is a promise to pay when a named person dies. The company on the other side of that promise does not need to know when any one person will die. It never has.

It needs to know how many out of ten thousand will die this year. Graunt showed that the second question has an answer even when the first one does not.

Thirty one years later the astronomer Edmond Halley finished the job. In 1693 he took birth and death records from Breslau, a city with unusually complete registers, and published a table in the Royal Society’s Philosophical Transactions.

Halley then did the arithmetic Graunt had left open. His paper’s own title promises “an attempt to ascertain the price of annuities upon lives,” an annuity being a contract that pays a set sum every year until the buyer dies.

That is a price. Built from a death table. It is the ancestor of the number on a term life quote today.

The King let him in over the objections

Graunt was a tradesman, and the Royal Society was not a tradesman’s room. The Society’s own historian, Thomas Sprat, recorded that Charles II recommended him and gave “this particular charge to his Society, that if they found any more such tradesmen, they should be sure to admit them all, without any more ado.”

He was elected on February 26, 1662, the month his book appeared.

The rest went badly. The Great Fire of 1666 burned his home and his business, he never recovered the money, and he died on April 18, 1674, aged 53. His widow applied to the Drapers’ Company for a pension on the grounds of poverty.

The tables outlived the shop by more than three and a half centuries.

What actually got inherited

Modern mortality tables carry serious names and long committee histories. They are still doing the one thing Graunt did in 1662.

They take a mass of individual deaths, most of them meaningless on their own, and turn them into a rate that holds still long enough to be quoted.

The garbage inputs got inherited too, in a sense. Graunt worked with causes recorded by parish searchers with no medical training, who guessed. He knew it, wrote it down, and worked around the error instead of pretending it away.

The tables were never clean. They were honest about not being clean, and steady enough to bet on regardless.

There is one more thing the 1662 book does that its descendants mostly stopped doing. It prints the ugly categories in plain language.

Frighted, Grief, Teeth, Rising of the Lights. A modern table would call all of that something with four syllables and a code number, and nobody at dinner would repeat it.

The 1662 Observations is in the public domain, and the tables are still legible in facsimile at the first link below.