The dark brilliance of John Graunt: “Father of modern statistics”

The father of modern statistics is widely acknowledged to be John Graunt, who was born in London in 1620. According to John Aubrey, who counted him as “my honoured and worthy friend,” Graunt “was bred-up (as the fashion then was) in the Puritan way.” He went into the family haberdashery business, where he showed that he had “an excellent working head.” The business thrived to such an extent that by the age of twenty-one he was running his own shop, specializing in buttons and lace collars. At the same time he was also appointed a freeman of the Drapers’ Company, the guild to which the haberdashers belonged.

As if this were not enough, Graunt also “rose early each morning to study before Shop-time.” The field which he chose to study was not only original but quite outside the scope of his work. In this, Graunt’s life bears a curious resemblance to that of his contemporary, Dutch pioneer Antonie van Leeuwenhoek. But where Van Leeuwenhoek pioneered microscopy, Graunt pioneered statistics. Before starting work each morning he began copying down details from the City of London bills of mortality (i.e. death registers). It has been suggested that his original motive for doing this was professional — that he was researching the extent, age and composition of his living clientele — though this seems unlikely. Either way, Graunt’s interests soon extended far beyond such limits.

The collection of bills of mortality was a comparatively recent phenomenon. As the population of England became increasingly urbanized, it was found necessary to collect details of these citizens for the purposes of tax and military service. The first regular bills of mortality for the City of London were collected in 1603, the year of Queen Elizabeth I’s death, when there was an outbreak of plague. Weekly lists were made of the number of people who had died, with the figures categorized according to various causes of death. Paradoxically, listing these causes of death serves to bring the city vividly to life. Take, for instance, a typical week’s list:

Aged: 13
Drown’d: 8 (…four at St Katherine’s Tower) 

Evil: 3
Feaver: 60
French Pox: 6
Griping in the guts: 134
Impostume [abcess]: 6
Stopping in the stomach: 9
Teeth: 38
Wind: 3
Worms: 2

Other weeks include such causes as ‘Murthered and shot’, ‘Found dead in streets’, ‘kild by several accidents’, ‘Stone & strangury’.

Graunt would publish his Natural and Political Observations on the London Bills of Mortality in 1662, when he was forty-two. It contained his analysis of the Bills of Mortality using ‘shop arithmetic’, which is far more inventive than it sounds. The bills of mortality also included the numbers of all children christened. Graunt took these christening figures, along with the mortality figures, to calculate the rate of mortality of infants before the age of six. First he chose the categories of death that he knew only included babies or very young children — such as ‘Stilborn’, ‘Overlaid and starved at Nurse’, ’Infants’ — and added these together. Then he used his own observations and intuition, and calculated accordingly. For instance, he reckoned that half the deaths from smallpox and measles occurred in children under the age of six; similarly, for one third of any plague deaths. From this he ‘inferred’ that 36 per cent of the overall deaths in London were of children under six. And making use of the number of baptisms he was then able to calculate the rate of infant mortality. Armed with such figures he went on to calculate “that of 100 persons born, 36 die before age 6 and 7 survive to 70.”

Graunt may have used only shop arithmetic, but not for nothing does he emerge as the first statistician. He was also able to estimate the size of an army which could be raised in London “based on the assumption that there are 107 males for each 100 females” (Graunt was the first to note this discrepancy).

More tellingly, Graunt was also well aware of the inaccuracies present in the figures he was dealing with. Few other than Descartes have begun with such healthy skepticism regarding their basic data. For example, Graunt observed that the figures for ‘French pox’ (syphilis) were invariably too low. This was because families wished to be spared the disgrace, so that “onely hated persons, and such, whose very Noses were eaten off, were reported.” On the other hand, those who were recorded as dying of the more respectable ‘old age’ were inevitably exaggerated.

On publication, Graunt’s Observations attracted the attention it deserved. Within a month it had been read by Charles II himself, who recommended that Graunt should be made a Fellow of the Royal Society. There was a little quibbling about letting a mere shopkeeper enter such exalted ranks, but the king would hear nothing of such objections: “If they find any more such Tradesmen, they should be sure to admit them all, without any more ado.”

Alas, Graunt’s fall from grace was equally swift. When the Great Fire of London broke out in 1666, he was accused of having ‘some hand’ in this. At the time he was an officer of a water company, but there is no certain evidence of any negligence on his behalf, nor of his encouraging the fire, despite the widespread incompetence of the firefighters.

The fact is, Graunt had converted to Catholicism – at the very moment when public opinion was turning against ‘Papists’. Furthermore, Graunt’s haberdashery premises were burned to the ground in the fire, leaving him facing bankruptcy. He would die in 1674, aged fifty-three, ‘descended into poverty’. Not even his work would be spared — his ‘friend’, the unscrupulous William Petty, would put it about that he had sponsored Graunt’s work, providing the original ideas which lifted his activities from mere fact-collecting to original statistical analysis.