Looking Forward

Prediction and Uncertainty in Modern America

Jamie L. Pietruska

• Publisher: University of Chicago Press
• Year: 2017
• Print ISBN: 9780226475004
• eBook ISBN: 9780226509150
• Topic: History
• Subtopic: North American History
• Index: History

In the decades after the Civil War, the world experienced monumental changes in industry, trade, and governance. As Americans faced this uncertain future, public debate sprang up over the accuracy and value of predictions, asking whether it was possible to look into the future with any degree of certainty. In Looking Forward, Jamie L. Pietruska uncovers a culture of prediction in the modern era, where forecasts became commonplace as crop forecasters, "weather prophets," business forecasters, utopian novelists, and fortune-tellers produced and sold their visions of the future. Private and government forecasters competed for authority—as well as for an audience—and a single prediction could make or break a forecaster's reputation. 

Pietruska argues that this late nineteenth-century quest for future certainty had an especially ironic consequence: it led Americans to accept uncertainty as an inescapable part of both forecasting and twentieth-century economic and cultural life. Drawing together histories of science, technology, capitalism, environment, and culture, Looking Forward explores how forecasts functioned as new forms of knowledge and risk management tools that sometimes mitigated, but at other times exacerbated, the very uncertainties they were designed to conquer. Ultimately Pietruska shows how Americans came to understand the future itself as predictable, yet still uncertain.

Chapter 1

Cotton Guesses

In the fall of 1899, the crowd on the floor of the New York Cotton Exchange reportedly burst into laughter at a poem satirizing the yield forecasts of Henry M. Neill, the decade’s most renowned and most vilified “cotton prophet.”1 “Strange,” the poet observed, “that this city farmer, regardless rain or shine . . . makes crop that’s always ‘fine.’” When the Brazos River floods, Neill predicts that cotton pickers wearing “diving suits” will harvest “an extra million bales” in Texas. Then, when drought hits the rest of the South, Neill foretells, with a wink, a “monster crop” that will mean “the price must further drop.” At the end of the poem, Neill insists—just as the real Neill had in August 1899—that the year’s yield will surpass eleven million bales and possibly even twelve. The poem, written by Louise Prudden Hunt, wife of prominent Georgia grower Benjamin W. Hunt, concludes with this lament: “Oh, Mr. Neill, this cotton (so queer the south ne’er knew), / These phantom fleecy millions no planter picked but you.”2 Imaginary cotton bales were no poetic fancy but rather the center of a decade-long struggle among southern growers, commercial forecasters, and the federal government to produce the most accurate and authoritative cotton statistics. Indeed, the poem’s swipe at Neill’s perennially bearish estimates encouraged, according to the Arkansas Democrat, “those who were fighting the south’s battle against Mr. Neill,” a battle for control of the market in which predictions determined prices.3

The cotton exchange poem also reveals the economic and epistemic crises of cotton forecasting in the 1890s, a decade in which Henry Neill was widely acknowledged as the primary market mover. A recognized cotton authority in Britain during the American Civil War, Neill achieved “cotton prophet” status on both sides of the Atlantic for his uncannily accurate predictions of the three largest yields in history in the 1890s. But in the context of plunging agricultural commodity prices and economic depression, Neill’s forecasts came to symbolize the ruination of the cotton economy to southern growers, who organized a campaign of statistical resistance designed to break Neill’s power over prices in what the New-York Tribune termed a “battle of crop estimates.”4

The battle unfolded in a series of contests, which this chapter traces, over claims to statistical authority, objectivity, and predictive certainty in cotton forecasting. Yet despite such claims, the statistics produced by Neill, the US Department of Agriculture, and growers’ associations did not combine to stabilize prices, curtail speculation, or rationalize markets, but rather had the opposite effect of perpetuating uncertainty and heightening concerns over market volatility.5 The outcomes of these statistical battles led the USDA to reconceptualize crop forecasting as no longer a late nineteenth-century quest for statistical certainty that sought to eliminate judgment. Rather, by the early twentieth century, it was seen as a more flexible mode of statistical judgment that acknowledged the persistence of unpredictability, thereby rationalizing uncertainty into modern agricultural commodity markets. “Rationalizing uncertainty” has a dual meaning, referring to the late nineteenth-century attempt to impose statistical order and bureaucratic rationality on the uncertainty of future cotton yields as well as to the early twentieth-century redefinition of uncertainty as an inescapable feature of crop forecasting. Of course, the implications of uncertainty depended on one’s position in the “worldwide web of cotton,” as incomplete figures on shipments and yield bedeviled brokers and spinners; inaccuracies in crop reports frustrated statisticians, lawmakers, and growers; and price fluctuations thwarted growers and rewarded speculators.6

Histories of nineteenth-century commodity exchange have depicted the savvy and unscrupulous speculator as a persistent source of market volatility and the object of moral critiques of gambling.7 But these accounts overlook the work of the crop forecaster as a market-manipulating statistical middleman who, like the speculator, sometimes profited at the expense of rural producers.8 Cotton forecasters clearly influenced markets—sometimes intentionally, sometimes not. And speculators wielded not objective, static economic data but rather dynamic and at times manipulated statistics that were produced by a combination of government statisticians, commercial interests, and farmers themselves.

Similarly, knowledge infrastructures are a vital but often invisible part of cotton’s “global network of land, labor, transport, manufacture, and sale.”9 In the late nineteenth century, government and commercial crop estimates in cotton-producing nations circulated around the globe as the quadruplex telegraph, undersea cable, and stock ticker enabled nearly instantaneous transmission of market news and prices.10 In national and global markets, knowledge of local crop conditions had minimal value, since the price a farmer could expect at harvest was determined not in relation to his neighbor’s crop but rather to the crop’s aggregate condition and yield.11 In this context, agricultural statistics became increasingly significant as a mechanism that enabled the valuation of an individual farmer’s crop relative to crops a county, a region, or half a world away.12 As the USDA’s chief statistician John Hyde explained at the turn of the century, agricultural statistics produced a new form of knowledge that was indispensable to commodity markets:

The aggregation[s] of individual facts . . . are usually too large to be perceptible to the senses. Men know by sight a bushel of wheat or potatoes, but practically they can know millions of bushels only by representative figures. It would, therefore, appear that statistics are as necessary to our larger comprehension as are vision, touch, and other senses to our so-called personal knowledge of things.13

However, the “larger comprehension” of the objective aggregate was not, in fact, that far removed from the subjective judgment that yielded “personal knowledge.” Both comprised the knowledge infrastructure that, in addition to relaying figures on millions of cotton bales and prices of cotton in cents per pound, actively constructed market information that could exacerbate uncertainty in the world of cotton exchange.

“The uncertainty of any further supplies from the South”

The poetic satire on the floor of the New York Cotton Exchange in 1899 was not the first time that Henry Neill was accused of conjuring cotton bales. The outbreak of the American Civil War signaled, alongside the irreparable political crisis over slavery and its expansion, a crisis of predictability in cotton estimating in Britain. By the eve of the war, cotton textile production dominated global manufacturing, and the labor of enslaved men and women in the American South produced the overwhelming majority of cotton used by spinners throughout Europe. Neill Bros., the firm owned and operated by brothers Henry and William Neill, predicted no easy or quick resolution to the war and could not anticipate how much cotton the American South would produce, if it could be exported, and at what price.14 A combination of Confederate export bans and Union blockades shut off the flow of cotton to Europe, causing the “cotton famine” that plagued Lancashire and other European textile production centers with mill shutdowns, soaring unemployment, and food riots.15 Henry Neill himself was reportedly caught blockade running aboard the Cumbria, a steamer headed from Liverpool to New Orleans that was captured off the coast of Charleston in 1862.16 The cargo seized included “heavy cloth, saltpeter, [and] sixty cases of Enfield rifles,” but Neill was likely more concerned with bringing cotton (or family members) back from the port of New Orleans than he was with delivering weapons to the Confederacy, considering that he was born into a family of abolitionists.17 Neill was released and sailed back to Britain, where his firm continued to issue circulars addressing the shortages and rising prices caused by “the uncertainty of any further supplies from the South for some time to come.”18

The war rendered estimates of American yield “mere guesses,” as the Neills acknowledged in a December 1863 circular.19 After the war ended, Neill Bros., whom Frederick Law Olmsted described as “the most pains-taking collectors of information about the cotton crop in the country,” dispatched three reporters to survey the cotton states and then assembled their observations into a cautious estimate of 1.5 to 1.8 million bales for the September 1866 to September 1867 season.20 The value of yield estimates, tentative and otherwise, soared in the aftermath of the war as a global network of cotton production and trade sprung up, linking growers and merchants from India, Egypt, and Brazil into the structures of global capitalism. As merchants and mill owners across the world sought “a secure and predictable supply of inexpensive cotton,” private forecasters like the Neills competed with the USDA to make the supply of American cotton ever more predictable.21

In 1866, Boston cotton manufacturer Edward Atkinson accused Neill Bros. of inflating their yield estimates for the South. Atkinson and the Neills feuded in the pages of the Boston Daily Advertiser and in private correspondence after Atkinson charged the Neills with issuing “diametrically opposite” instructions in October 1865. They advised the public, who was told to anticipate 2 to 3 million bales on hand (not the 1.5 million bales generally expected) to wait; at the same time, they instructed private subscribers, who were given a less optimistic estimate, to buy. Atkinson, convinced that depressed cotton prices would inhibit “the restoration of order in the South,” wrote indignantly to the Neills that their public circular was “so written as to depress the market, the other to stimulate it.” The Liverpool Mercury and the Boston Daily Advertiser printed extended comparisons of the Neills’ private and public circulars from October 1865, which Atkinson pointed to as evidence for his “distrusting” of Neill’s numbers.22

The Neills did not deny having issued two separate circulars and indeed justified their existence. “The value of our information to our friends,” the Neills reportedly wrote to their clients, “will depend upon its being confined to a limited number, and we have therefore concluded not to print it in our circular, but to reserve it for those for whom we are receiving orders.”23 Moreover, the Neills made no pretenses of statistical objectivity. In a letter to the New York Times, William Neill reminded readers “that Messrs. Neill Brothers & Co. are chiefly agents for English cotton buyers. . . . It will therefore be well to consider our position very carefully.”24

The feud between Atkinson and the Neills ended in little more than finger pointing. Both complained of being unfairly targeted, and the Neills insisted that their circulars had been selectively excerpted and their meaning altered in the press. Atkinson apologized to the Neills for basing his judgment on ostensibly edited letters published in the Liverpool Mercury but privately doubted that their content had been misrepresented.25 But allegations of self-interested cotton statistics would continue throughout the late nineteenth century, and Atkinson’s complaint that Neill’s cotton figures received “undue prominence” from the Boston Daily Advertiser would be echoed by southern newspapers, especially the Atlanta Constitution, which lamented Neill’s singular authority in the world of cotton forecasting.26

“Uncertainty is the mother of speculation”

Long before the Atkinson-Neill dispute over objectivity in cotton circulars, various organizations and individuals sought to collect disinterested agricultural statistics that could be used to make predictive judgments in commodity markets. In the first two decades of the nineteenth century, the Massachusetts Society for Promoting Agriculture and Virginia’s Albermarle Agricultural Society, motivated by imperatives of scientific agriculture and the logic of improvement, undertook quantitative assessments of local farming practices. State-sponsored agricultural surveys, in New York in 1825 and in Massachusetts from 1837 to 1841, extended these efforts, and large-scale, systematic crop reporting emerged after the Panic of 1837, when the US Census, the US Patent Office, and the Maryland Agricultural Society launched experimental crop surveys between the late 1830s and the mid-1850s, which were complemented by occasional newspaper accounts from correspondents reporting on what they had seen and heard of the upcoming harvest. American Agriculturalist editor Orange Judd, who had published readers’ descriptions of crop conditions in the mid-1850s, aggregated his readers’ monthly reports of acreage and estimated yield from May to September 1862 in what amounted to the first published forecasts of the coming harvest.27 Judd enlisted over 1,500 correspondents from over twenty states in a collective effort against the statistical uncertainties of agricultural commodity markets and speculators’ intentionally misleading forecasts. But Judd, who had always believed that the federal government should oversee the collection of agricultural statistics, ended his project in September 1862 and yielded to the USDA, which had been established in May of that year.28

Immediately upon its founding, the USDA began to create a knowledge infrastructure to produce objective and authoritative agricultural statistics. The first commissioner of the USDA, Pennsylvania’s Isaac Newton, recognized that data furnished by the decennial census, already two years old by the time it was published, rendered yield per acre, total acreage, and soil condition “questions of the highest magnitude but of which no one could speak with any certainty.”29 In 1863, the Division of Statistics was formed within the USDA and began issuing monthly reports on acreage under cultivation and...

Table of contents

• Cover
• Title Page
• Copyright Page
• Dedication
• Contents
• Acknowledgments
• Introduction: Crisis of Certainty
• 1  ·  Cotton Guesses
• 2  ·  The Daily “Probabilities”
• 3  ·  Weather Prophecies
• 4  ·  Economies of the Future
• 5  ·  Promises of Love and Money
• Epilogue: Specters of Uncertainty
• Archival Collections
• Primary Source Databases
• Index
• Footnotes