Graunt’s other striking observation was that, year in and year out, more boys are born than girls—about one-thirteenth more—and here, disturbed by something that seemed to contradict the basic coin-toss assumptions of nativity, he went beyond the data to propose a reason:So that though more men die violent deaths than women, that is, more are slain in wars, killed by mischance, drowned at sea and die by the hand of justice; moreover more men go to the colonies and travel in foreign parts than women; and lastly, more remain unmarried than of women as fellows of colleges, and apprentices above eighteen, etc. yet the said thirteenth part difference bringeth the business but to such a pass, that every woman may have a husband, without the allowance of polygamy.
In other words, God in his mercy regulates the birth rate so that Christians need not live like Mohammedans. This argument from Providence persisted for more than a hundred years. It marked an interesting departure from previous ideas of personal experience of the divine: here was an example of God’s work that could be revealed only by the collection and analysis of mass fact.
Graunt’s more fortunate friend, Sir William Petty, showed what power could follow from a judicious use of data. Shipped for a cabin boy at the age of fourteen, he had been abandoned in Caen after breaking his leg, but soon attracted local help because he could speak Latin and Greek. By the age of twenty-nine, Petty had become a professor of anatomy (famous for reviving “half-hanged” Nan Green) and had patented a letter-copying machine.
When Cromwell’s government was carving up a conquered Ireland, Petty went as an expert at surveying—indeed, he was so expert that he returned with an estate of 50,000 acres. It was Petty who realized how valuable calculations like Graunt’s could be to the realm: mortality tables could at last reconcile the relative value of an income paid over a lifetime with a cash sum now, or rent on property with purchase price. At a time when most of the kingdom’s wealth was fixed in land, this was an essential matter. Petty proposed to Charles II the establishment of a central statistical office that would collect and analyze these vital facts, rationalizing taxes to give the realm a secure income without overburdening its taxpayers. The easy-going monarch chuckled, nodded . . . and no more came of it.
Seventeenth-century governments, almost constantly at war, needed to raise large sums of money quickly, and selling annuities (lifetime income paid in exchange for a single capital sum) seemed an attractive gamble. An annuity buyer is, effectively, betting against his own early death; so a canny government, if possessed of the facts, could offer longer odds than the mortality figures justified, relying on the instinctive belief that everyone dies at an average age—except me. The mathematical apparatus of old-age welfare, of Social Security and private pensions, actually began as an attempt to secure a house edge for the State.
Information about the public, if kept secret, offers private advantage; so social calculation fell into twin wells of concealment: the inner councils of life-insurance companies and the ministries of anxious kingdoms. Throughout the eighteenth century, population and mortality were considered State secrets. The dominant political theory was mercantilism, a form of exalted miserliness that taught that the country with the most gold and most people at the end of the game wins. A sensible monarch would therefore no more reveal his country’s population than a poker player would invite opponents to study his cards.
The constant problem was to find a dependable source of raw data: for many years, most scientific assessments of human mortality were based on the experience of one Prussian city, Breslau (now Wrocław), where the Protestant pastors had been bullied into compiling accurate and complete information. Prussia (a country that in this period was spreading across Europe with the stealthy rapidity of a bacterial colony in a Petri dish) encouraged census-taking and internal interpretation, but prevented publication of its results. It was the Prussian professors working on this data who first came up with the term “statistics”—but by this they meant the State’s numbers: the vital signs of the realm’s health. Prussia knew and numbered every barn and chicken-coop in its territory, but, like a quartermaster’s report, this was privileged information.
There were enthusiastic amateur census takers, though. In the 1740s, the Prussian pastor Süssmilch built on Graunt’s work: assembling immense amounts of information on births, deaths, and marriages throughout Germany, he found, first that God was clearly punishing sinful city-dwellers with higher death rates; and second, that glimmerings of some mechanism in society could be made out once the mass of facts became sufficiently large. For example, there appeared to be a fluctuating relationship between population and land. More available land meant peasants could marry and set up house earlier, meaning more children, meaning more future peasants, meaning less available land. As economic theory, this may appear very basic—Adam Smith developed far more interesting ideas out of his own head—but the point was that Süssmilch inferred it from facts, not from Reason.
Facts had become interesting—not just to government ministers, but to all Germans. There were weekly publications in many towns of whatever lists and numbers contributors had happened to pick up or tabulate. Johann Bernoulli (another one), traveling through Prussian territory, described a princely collection of Old Master paintings simply by their dimensions.
It was a time ready to see itself as the Statistical Age—in as confident and as vague a sense as the Atomic, Jet, or Information Ages would be. Mass, Mechanism, and Number were replacing Nature, Reason, and Proportion as the received ideas of the time. When Quetelet sketched the potential power of his moral physics, the effect was like removing the cork from a shaken bottle of champagne. Mental effervescence fizzed across the continent.
The idea of the simultaneous—many distant others doing things just at this moment—arrived with the railroads and their need for uniform time. In the factories, interchangeable parts not only made mass production possible, they changed the products—muskets, ship’s tackle, spinning frames—from hand-shaped objects made for the here and now into assemblies of components with potential use at any time or place. Capital was transforming from a solid—my gold in this bag—to the universal fluid of credit. Steam—amorphous, portable, tireless—led industry up from its deep river gorges and made all places equally suitable for a mill. Machine tools, mechanisms to make mechanism, brought in absolute numerical standards of flatness, pitch, and diameter to replace the millwright’s personal fit of hand, eye, and material.
Even the nature of numbers was changing: the decimal system tempted us to express proportion as a percentage, giving it the appearance of absolute value. No longer were things “in the relation of one to three” or “two shillings sixpence in the pound” or “and about thus far again”—they were 33 percent, 12.5 percent, 100 percent, precise, uniform subjects of the universal law.
Inspired by Quetelet, British scholars founded the Royal Statistical Society, but found themselves caught in a dilemma: was this new discipline a science or just a method that aided other sciences? With a characteristic wariness of Big Ideas, they decided on the latter, choosing as their emblem a sheaf of ripe wheat with the modest motto aliis exterendum: “let others thrash it out.” Their first questionnaire, On the Effect of Education on the Habits of the People, had as its first question “What is the effect of Education on the habits of the People?”—their technique, luckily, would soon improve.
Statistics were opening the minds of historians and philosophers to the possibility of understanding social mechanics. Alexis de Tocqueville wrote three books that still illuminate the essential distinctions of habit and expectation that separate French, English, and Americans—with not a page of statistics in any of them. When he first saw André-Michel Guerry’s essay on the moral statistics of France, accompanied by its beautifully complete returns for sanitation, suicides, and crime, he exclaimed that, were it not for the dishonor, he would willingly be condemned to prison for life if the sentence allowed reading such splendid tabulations.
At once the most extreme and the most ingenious exponent of this new view of history was Henry Thomas Buckle—a meteor that streaked across the skies of fame and is now seen no more. A sickly child, he was indulged in everything by a mother on whom he doted. By the time he reached adulthood, he had acquired fluency in seven languages; a library of 22,000 books; a wide if inconsistent range of knowledge; and two minor vices: cigars and chess.
Buckle had no fear of Big Ideas, and his own was unapologetically vast: that free will, God, and the power of the State were all fictions. The principal, indeed the sole, influences on the development of the human race were Climate, Food, Soil, and the General Aspect of Nature (this last was necessary to explain imagination, poetic feelings, and so on). The differences we might see between us, all the various racial or national distinctions, were straightforward consequences of these mechanical influences.
If humans are simply products of their environments, then, of course, we need to know everything about the environment to know humanity. This was the great use and value of statistics:They are based on collections of almost innumerable facts, extending over many countries, thrown into the clearest of all forms, the form of arithmetical tables; and . . . they have been put together by men who, being for the most part mere government officials, had no particular theory to maintain, and no interest in distorting the truth of the reports they were directed to make.
Wherever statistics showed uniformity, whether the matter was physical or moral, then social law was at work—and there was nothing that Church or Crown could do about it: The great enemy of civilization is the protective spirit; by which I mean the notion that society cannot prosper, unless the affairs of life are watched over and protected at nearly every turn by the state and the church; the state teaching men what they are to do, and the church teaching them what they are to believe.
Buckle’s History of Civilization in England appeared in two volumes—one a preamble and the other a trial run on Scotland and Spain (easiest to describe because, in Buckle’s terms, so little worthwhile had happened in either country). The third was to have covered Germany and the United States, before limbering up for the actual subject of the book’s title—but fate intervened. Buckle’s mother died in 1859, prompting him to rethink his previous denial of the immortality of the soul. Distraught, unable to work, he traveled to the Holy Land and succumbed to a fever at Damascus. They say that his deathbed delirium was one familiar to any author: “My book! I have not finished my book!”
He should not have worried: like Quetelet’s, Buckle’s thesis was all the more powerful for not having been fully elaborated. As open-ended speculation his ideas circled the earth. In America, the young Henry Adams felt sure a science of history had arisen. Strindberg based Master Olof on Buckle. Young Romanians looked to Buckle’s work as the pattern for development of their country. Dostoevsky told himself in his notebooks to read and reread Buckle; in every word of the Grand Inquisitor in The Brothers Karamazov you can hear the voice of that “protective spirit” Buckle described and hated.
Yet Dostoevsky, with characteristic contrariness, also presented an extreme form of resistance to the implications of a statistical world. The unnamed protagonist of Notes from Underground mocked the efforts of the nineteenth century to erect a Crystal Palace of certainty over everything: “to affirm, for instance, following Buckle, that through civilization mankind becomes more gentle and consequently less bloodthirsty . . . Logically it does seem to follow from his arguments. But man has such a propensity for systems and abstract deductions that he is ready to distort the truth intentionally, he is ready to deny the evidence of his senses only to justify his logic.” Dostoevsky’s own notebook makes the point more clearly: “How does it come about that all the statisticians and experts and lovers of humanity, when they enumerate the good things of life, always omit one particular one? One’s own free and unfettered volition, one’s own caprice, however wild, one’s own fancy, inflamed sometimes to the point of madness—that is the best and greatest good.”
Because statistics was a philosophical stance before it was a numerical technique, the first objections to it were on similarly philosophical grounds. Many German schools of thought disliked the way Buckle’s ideas made lonely atoms of us all. Yes, there might be laws to history, but the proper subject of these laws was collective: our Culture, our Class, our Community, our Nation—our Race. You can begin to see the ways these ideas would play out in the twentieth century.
Novelists split into those who sought realism through true depiction of statistically revealed types—Balzac, for instance, appeared to be checking off a list of French subclasses with every novel—and those who stood against the implications of uniformity and determinism. Tolstoy’s claim that every unhappy family is unhappy differently is a rumble of protest against the presumption of statistics. Dickens loathed the statisticians, with a city boy’s conviction that every street and every tenement was unique. His attack on the tyranny of social law is most overt in Hard Times, where poor Cissy must learn her statistics in the unforgiving school of Mr. Gradgrind: “In this life, we want nothing but Facts, sir, nothing but Facts.” Gradgrind gets his comeuppance when his son Tom turns out a thief—but the boy had learned all too well the statistical view of fate: “So many people are employed in situations of trust; so many people, out of so many, will be dishonest. I have heard you talk, a hundred times, of its being a law. How can I help laws?”
Even some determinists disliked the notion that numbers constituted the most important facts. August Comte, the creator of Positivism, in all other respects a committed believer in social law, dismissed this quantification as impossible: he never forgave Quetelet (“some Belgian savant”) for taking his term “social physics” and applying it to this numerical travesty. In revenge, he invented the word “sociology,” which he thought far too ugly for anyone to steal.
Are we individuals or collectives? Is experience determined or free? Do its laws remain constant or change? Can we quantify life without losing its essence? It took much wrangling over the philosophical and moral implications of Quetelet’s work before people went back to look more closely at the numbers. When they did so, they found less certainty than had been advertised. Toward the end of the nineteenth century, Wilhelm Lexis considered the raw material again, comparing birth, suicide, and crime figures with their probabilistic equivalents. That is, he created a probabilistic model for a given statistic: an urn filled with balls, either marked (“boy,” “despair,” “murder”) or blank, in the ratio of their mean value from the observed figures. He then calculated the likelihood of getting the same values observed for a given year by drawing from the urn—in effect comparing real experience with the ideal that Quetelet had said lay behind all phenomena. For births, the model and the real fit perfectly—a demonstration that the proportion of male to female births really is determined by independent random events. Beyond that, though, only Danish suicides for 1861-1886 actually corresponded with the stable distribution of urn-drawing. Other phenomena were just too variable to be explained in these terms only.
Lexis created what he called the “index of dispersion,” Q, which compared the observed phenomena with their probabilistic model. Where Q = 1, they coincide: the real world is behaving like the flip of a coin—events are independent and random. Where Q is less than 1, the world is being driven by some underlying law; things are happening for a reason; Buckle rubs his hands. When, however, Q is greater than 1—as Lexis found it was for most social statistics—then fluctuation is king: at least a considerable subgroup is changing significantly but unpredictably. Society may look stable and determined—but that’s only because we are looking at too short a time-series.
Who, though, wants just to look at society? When children are dying of cholera, old women of cold; when families are huddling in rat-haunted rooms, accessible only by wading through filth? Who would not want to change society? The nineteenth century saw great conflicts of ideas, but it was also—pa
rticularly in Britain—a period of enormous practical energy. For the women and men who looked around them and felt a call to action, social numbers were not just an object of contemplation—they were a powerful tool for getting things done.
Take those Scottish chests: the original article in the Edinburgh Medical Journal was not about the wonderful uniformity of Scotland’s soldiers; it was about the worrying difference between the stout farm boys of Kircudbrightshire and the wizened mill hands and miners’ lads of Lanark—forty miles away. Nor were these the only social data collected in Scotland. In 1791, a Highland landlord, Sir John Sinclair, had begun the first national study ever undertaken: the Statistical Account of Scotland.
With genial high-handedness, Sinclair had stolen the word “statistics” from the Germans during a Continental tour in the 1780s. He felt he had a better use for it:By statistical is meant in Germany an inquiry for the purpose of ascertaining the political strength of a country, or questions concerning matters of state; whereas the idea I annexed to the term is an inquiry into the state of a country, for the purpose of ascertaining the quantum of happiness enjoyed by its inhabitants and the means of its future improvement.
“Quantum of happiness” was taken from Jeremy Bentham; like him, Sinclair looked for the basis of improving society in the examination of particular facts with an eye to putting them to use: he wanted to improve life for the Highlands—a region of extreme poverty, stoically borne—so he set about collecting his material.
Following the practice of the German mortality tables, he took local clergymen as his reporters—but he did not allow any respect for their cloth to stand in the way of his desire for complete information. He wrote his final letters of reminder in red, allowing their recipients to draw their own conclusions from “the Draconian color of his ink.”
Chances Are Page 15