******
****
Rather than depend on the bottom line in the reported deaths in the city of London as did Samuel Pepys, John Graunt began a systematic search of the records to produce a better understanding of the number of victims of the plague which swept the city in 1600 and 1650. He recognized that it was necessary to characterize the population so that it would be possible to state with surety the casualties of the disease.
When Graunt began his study of the quantitative "facts" about the population of London he was greatly aided by the English peccant for documentation of the most trivial facts. [As example, The Economic and Social History of An English Village, a book by Norman and Ethel Gras, which delved into the records of Crawley and brought to life the peoples of this village tracing back to 643 when it was chartered.]. Graunt had some fifty years of records to work with as from 1603 the City kept a weekly tally of the goings on in the churches the documentation of both births (christenings) and deaths (burials). Public records existed as well but there was no thread tying together the seemingly disconnected event.
The weekly bills of mortality, as example those reported in Samuel Pepys' Diary, could do little more than report the collection of corpses and the areas of the city which appeared to be more or less under siege. While the reports were being read to determine if it would be safe to return and when within the passing of the season was there the greatest danger, little thought was made as to the accuracy of the reports or the timeliness of the statistics. Herein lay one of the greatest errors in judgement as the weekly tolls rose and fell with no apparent cause or effect, often doubling or tripling only to fall back to a much lower level.
John Graunt's desire to know about the state of the plague and events preceding, required organizing the collected data in a manner so that the events could be related. He was fortunate in that there was two plague periods that descended upon England within his lifetime. How then could he compare these two events and perhaps wrench information from them, looking to the records and then drawing conclusions. This then was the origin of STATISTICS as we know it.
Setting out carefully the problem, being sure that the information to be collected could be qualified and quantified, drawing inferences and then proving the theory by testing the theory under a similar set of circumstances. Had Graunt only a single set of data that covered only one time period, it would have been impossible to have conclusively proved his theory of the cause and result of a disease that was little understood by the best physicks of the day.
While much is made of the determining of the actual number of dead from different causes by Graunt's examination, a far more important and less understood finding emerged. Only in the twentieth century has the relationship been properly examined and causes of death explained.. In examining the records of death, Graunt discovered that there were two seemingly related events that occurred in the disease which was called by various names such as, Black Death, Plague, Bubonic Plague, Glandular Plague, Hemorrhagic Plague, Pneumonic Plague.
This required the recording of the initial deaths from the plague and the characteristics of the dead by location within the City. Seemingly the disease first affected those at the docks and they developed the buboes associated with the disease. This first stage of the plague was no doubt the result of the importation of rats carrying the disease and their hoards of fleas which infested them. Workers at the docks and their families were stricken and died the horrible death associated with the enlarging of lymphatic tissues, the (buboes) bursting and spewing forth the poison, before they died. And the disease spread, but slowly. Graunt observed there was a period of perhaps as much as two years before the second stage of the disease became apparent. This was the pneumonic plague associated with "bad air" or miasmic caused by the inspiration of germs from those affected by the disease. While the victims died a painful death, it was not the gory and odious death of those who went before. Thus without naming the two seemingly unrelated events, Graunt had discovered the disease was caused by the same microorganism, Yersinia pestis (Pasteurella pestis if you prefer), infecting the body by different routes.
Thus, with Graunt's examination of the records of the plague of the turn of the century and then using the data of the current period, he was able to confirm the sequence of events that led to emergence of the epidemic. He threw out the epistemological reasoning (that based on knowledge of the past) and introduced the use of statistics to accept the new knowledge and prove that that knowledge was sound.
Here is quoted from Ian Hacking's book, the Emergence of Probability: "The swarms of mice that occasionally overran some of the towns of Central Europe, thousands dying frothing in the streets, were indeed a probable sign of plague to come. However, it is of no moment which signs seem sensible to us, and which absurd. Here we have a very clearly stated conception of partial prognostication, which is thereby possessed of probability, rather than certainty, and whose probability arises from frequency, from what happens "almost always' or else �often.' This represents Graunt's first statistical relationship.
Hacking quoted from Graunt's observations, "The contagion of the plagues depends more upon the disposition of the air (miasma) -than upon the effluvia (rupturing of buboes) from the bodies of men. Which also we prove by the sudden jumping which the plague hath made, leaping in one week from 118 to 927, and back again from 993 to 258, and from thence again the very next week to 852."
In examining his collected data Graunt was able to argue that the plagues of the two different periods had about the same intensity. Old granny, remembering the bad times, is often wrong and he was able by relating births and deaths show that loss from disease was essentially the same percentage. (He expressed the data as ratios, not percentages.)
Again quoting Hacking, "
One signal inference of Graunt's is the first reasoned estimate of the population of London. We know the number of births from the Bills. We have a rough idea of the fertility of women. Hence we can infer the number of women of child-bearing age. From this we form a shrewd guess at the total number of families. We also guess the mean size of a family. and thereby estimate the whole population. Of course the method is crude, and aside from its internal defects, there are other sources of error. Graunt tentatively allows for the effect of the plague not only through its ravages but also because of the exodus from the city in time of plague by all those who can afford to escape the corrupt air by moving to the country. Graunt checks his estimate of the population by two other methods of inference. One, briefly described, is a straightforward sampling of three parishes. The other is based on inhabited area and a guess at the density of habitation. His ingenious arithmetic refutes a view current at the time, that London could boast two million souls, but carries little conviction for Graunt's own estimate of 384,000. The first method, which had been confirmed by the other two, was exceptionally fruitful, for unlike survey sampling it could be applied to the past as well as to the present and so Graunt could plot the astonishing growth of the city and also prove that much of the increase was due to immigration, not procreation. He could also show that despite the horrors of plague, the decrease in population of the worst epidemic was always made good within two years." Pretty heady stuff considering the information he was working with.
".... no holding back Graunt's inventive mind. The course of various diseases across the decades, the number of inhabitants, the ratio of males to females, the proportion of people dying at several ages, the number of men fit to bear arms. the emigration from city to country in times of fever, the influence of the plague upon birth rates, and the projected growth of London: all these subjects are examined with gusto."
In these days of data dredging, where attempts are made to show relationships hidden within unrelated numbers to prove societal issues, it is well to remember that Graunt perhaps was the father of this infamous branch of statistics as well. Certainly it is well to remember that numbers don't lie, but statisticians do.
Graunt's work certainly makes a strong argument for him being remembered in the same ranks as Pascal, Thomas Lodge (used swarming of mice as evidence of pestilence to come), Aarhus (Danish Bishop who advised fleeing from the sores and corrupt air), Petty Logic and others..
But there is more. Graunt dabbled in concerns of state as well. He was an early advocate of welfare to the poor. Here according to Hacking is how his system was proposed to work:
"... Graunt recommends a guaranteed annual wage. He reasons as follows:
(i) London is teeming with beggars.
(ii) Hardly anyone dies of starvation.
(iii) Therefore the national wealth already feeds them.
(iv) They should not be put to work, for their produce will be shoddy and the Dutch (who at Ypres already subsidize idlers) will gain British trade.
(v) So, at no extra cost to the Nation we should feed them and keep them from defiling our thoroughfares by begging.
Graunt's book, Nature and Political Observations came out in January 1662, some months before the passing of the fundamental statute of English poor law, the law concerning 'Settlement and Removal'. The disastrous experience of British 'workhouses' confirmed Graunt's gloomy foreboding. What is notable is not Graunt's thesis. It was much in the air at the time and had been advocated at least thirty years earlier. Only his mode of argument is new.
It had little effect, whereas Paris, without benefit of statistics, in the single year 1654 managed to confine 1% of the population to hospitals for the poor, maimed, and mad."
So as we remember John Graunt (1620-1674) as a statistician, perhaps it's better to wonder what it must have been that caused the spawning of so many giants of literature and science in that period.
****
Joe Wortham's Home Page , About Joe Wortham , Directory
Questions? Comments? [email protected]