Part IThe Story
In August 1909, a wrecking crew began taking apart a house on the northeast corner of Fifth Avenue and Seventy-eighth Street. The house was not yet twenty-six years old. Henry H. Cook, the railroad magnate who built it in 1883, had worked without a deadline or a fixed budget. Whenever he came upon a feature he disliked, he told the contractors to tear it out and try again, and by that method of expensive second thoughts a sandstone-and-marble residence became one of the most lavish dwellings on the avenue. Cook had bought the whole block from Marcellus Hartley in 1880 for $500,000. He sold lots on it only for private houses, never for anything tall, so that the block would stay a block of mansions.
The new owner paid $1.25 million for the Cook house in January 1909. He hired C. P. H. Gilbert to alter it and borrowed $700,000 to pay for the work. Then, sometime that summer, he changed his mind about the alterations and about the house itself. A fireplace that had cost $15,000 sold for $300. Paneling valued at $55 a section went for $3. The demolition contractor, who knew what he was looking at, gave the building a eulogy no architect would have minded:
— The demolition contractor on the Henry H. Cook mansion, 1909The best-built house ever torn down in New York City.
The buyer was James Buchanan Duke, fifty-two years old, of Durham, North Carolina, called Buck by nearly everyone who knew him. He was president of the American Tobacco Company. At that moment the United States government was suing to destroy it. In its real-estate column that August, The New York Times asked a question that was half admiring and half sly: "Is it possible that twenty-five years hence people will be talking about the 'old' Duke house?"
The question was reasonable, and it was aimed at the right man. Duke's fortune rested on a single repeated act: take something that already worked and replace it with something that worked faster, cheaper, and in larger quantities. He did it to the hand-rolled cigarette, to the independent tobacco manufacturer, and to his British competitors. He did it to the falling water of the Catawba River, and here he did it to a perfectly good mansion. No one who watched him for thirty years could have said what he would refuse to tear down. The surprise comes at the end. The last thing he built was designed to be impossible to tear down. He built it in his father's name, and he resisted putting his own name on it.
By the Numbers
The Duke Machine
133Cigarettes per minute from one Bonsack machine, a skilled hand roller's hourly output
$800,000Duke's advertising spend in 1889, against net profits of less than $400,000
~250Companies absorbed by American Tobacco within two decades of its 1890 founding
$25M → $316MGrowth in American Tobacco's equity under Duke's control
300+Cotton mills and industrial customers powered by the Dukes' hydroelectric system within two decades
$40M + $67MThe Duke Endowment at its December 1924 founding, plus the addition from his estate
The Bull in the Road
Durham in the late 1870s already had a famous tobacco brand, and it was not the Dukes'. Bull Durham, the shredded, pouched smoking tobacco also made in Durham, dominated the trade that W. Duke, Sons & Company was trying to enter. A family firm with ordinary resources could have kept grinding against that incumbent for another twenty years and probably lost. The decision attributed to the younger Duke in 1879 was to stop. He would not compete in that category. He would leave it for a product that barely existed as a mass commodity: the cigarette.
Washington Duke, the father, was born in 1820. He was a tobacco manufacturer who later became a Methodist benefactor, and his life fits the outline of the self-made Southerner almost too neatly. He married twice. With his second wife, Artelia Roney, he had Benjamin in 1855, James in 1856, and a daughter, Mary. An older half-brother, Brodie, came from the first marriage. The sons took over the firm in the 1880s. Benjamin, a year older than Buck, was the steadier and more sociable of the two, and he became the brother who ran things, sat on boards, and lived in the grand houses before James did. Buck was the one who saw the angles.
The choice of the cigarette deserves a moment, because it was not obvious. The hand-rolled cigarette of the 1870s was a labor product. Rooms of skilled rollers turned out a few thousand a day each, and the expense of the product was mostly the expense of fingers. A cigarette maker was a wage-bill business with tobacco attached. Leaving Bull Durham's category meant entering one where the only advantage anyone held was the size of its workforce. It was a category waiting for a machine, though in 1879 nobody could have said when one would arrive.
Robert Sobel titled his chapter on Duke "Opportunism Is the Spur," and the label fits almost too comfortably. Opportunism suggests luck noticed. What the record shows is narrower and harder. Duke was willing to leave the field where his family had grown up, admit the incumbent was too strong, and move to ground where nobody yet held an advantage. The ego cost of that kind of exit is real. Most second-generation heirs stay in the trade their fathers taught them.
One Hundred Thirty-Three a Minute
In 1882, a young inventor named James Albert Bonsack produced a cigarette-rolling machine. The surviving accounts tell us almost nothing about him as a person. They tell us what he made and what it did. The machine produced more than 133 cigarettes a minute, roughly what a skilled hand roller turned out in an hour. It cut the cost of rolling by about half, and it cut each cigarette to an exact, repeatable length. That last feature, which a modern operator would take for the whole point, was the part the market initially disliked. Uniform cigarettes looked machine-made, and in the early 1880s machine-made carried a stigma. Allen & Ginter of Richmond, one of the most prominent cigarette makers in the country, rejected the machine almost immediately.
The rest of the industry's early skepticism was not stupid. The Bonsack machine broke down, the product looked wrong to customers who knew hand-rolled work, and the incumbents' brands depended on exactly the craftsmanship the machine erased. Their caution was rational, and that is why it was dangerous to them. The innovation that kills you usually arrives looking like a downgrade.
The first Bonsack machine went into the Duke plant in Durham on April 30, 1884. The equipment mattered less than the deal around it. Duke agreed to produce all his cigarettes on his two rented Bonsack machines. In exchange, the Bonsack Machine Company cut his royalty from 30 cents per thousand cigarettes to 20 cents. The contract was secret. Duke also hired one of Bonsack's own mechanics, and as a result his machines broke down less often than his competitors' machines did.
I find this contract more revealing than anything Duke is recorded as saying, partly because he is recorded as saying so little. In one transaction he committed his whole production to an unproven technology, used that commitment to extract a permanent cost advantage, made sure his rivals did not know they were paying a third more for the same machine, and bought the one form of knowledge that could not be patented: the tacit skill of keeping the thing running. Under that arrangement every cigarette he sold cost less to make than the same cigarette made by anyone else. He could cut prices further than they could and still make money. By one account he licensed the machine in 1885. By 1890 he supplied about 40 percent of the American cigarette market.
Something less visible also happened in Durham in 1884. A Southern family firm, in a region still poor two decades after the war, became the most mechanized operation in its industry. The North had the capital, the cities, and the established brands, and the most advanced production line belonged to Durham.
Histories of Poor Boys
In 1888, W. Duke, Sons & Co. published a set of fifty miniature booklets, each the size of a cigarette card, and slipped them into packages. The series was called Histories of Poor Boys Who Became Rich and Other Famous People. One booklet told the story of Jay Gould. Another told the story of Edgar Allan Poe. A companion series of fifty was titled Histories of Generals, and one of its subjects was George Edward Pickett, whose name the South had attached to its most famous charge and failure.
A smoker in 1888 could buy a machine-made cigarette and get with it a tiny printed sermon on the American promise, or a tiny monument to the Confederacy. Duke's card series covered the rest of popular life too: musical instruments, fancy-dress ball costumes, holidays, gymnastic exercises, lighthouses, jokes, yacht colors, "Presidential possibilities," and the rulers, flags, and coats of arms of the world. The cards were collectible, which meant the customer came back. They were also a cheap encyclopedia and a pocket magazine, and they made the package worth keeping after the cigarettes were gone. The cigarette card, which American Tobacco would later run through brands like Mecca, Fatima, Recruit, Turkey Red, Old Judge, and Old Mill, eventually produced the T206 baseball set and the Honus Wagner card. That is a strange afterlife for a sales device: a promotional throw-in became one of the most valuable pieces of printed paper in the world.
Nobody seems to have recorded whether Duke recognized himself in the Poor Boys series. The Dukes were not poor by 1888. They had not been rich for long either, and the distance between Washington Duke's farm and the pages of those booklets was a single generation.
Duke's machine had a second-order problem. It made more cigarettes than the market was buying. In the 1880s the industry's growth rate was slowing just as his costs collapsed. Cheap supply needs demand, and demand had to be manufactured almost as deliberately as the product. In 1889, Duke spent $800,000 on advertising. That year his net profits were less than $400,000. He was spending twice what he earned to buy attention, and he was cutting prices while doing it.
The arithmetic only works if you know something your rivals do not, and Duke did. His cost per thousand was structurally lower. A price war that bled the others nearly to death merely made him thinner. His four major competitors — Allen & Ginter, the firm that had turned down the machine; W. S. Kimball & Company; Kinney Tobacco; and Goodwin & Company — had to lower their prices to stay in the market. Then, in 1890, they joined him.
The American Tobacco Company was founded on January 31, 1890. Its five constituent companies made 90 percent of the cigarettes produced in the United States that year, which was also the year the company listed on the New York Stock Exchange. The business historian Patrick Porter called it "one of the first great holding companies in American history." Congress passed the Sherman Antitrust Act the same year. Neither event was aimed at the other yet.
Without Regard for Previous Ownership
The newspapers called it the Tobacco Trust almost at once, and the name stuck because it described the thing accurately. In 1896 American Tobacco was one of the original twelve companies in the Dow Jones Industrial Average. Within two decades of its founding it absorbed about 250 companies, including the Lucky Strike company, and made 80 percent of the cigarettes, plug tobacco, smoking tobacco, and snuff produced in the United States. Under Duke's control, its equity grew from $25 million to $316 million.
The integration reached all the way into the factories. Inside the combination, one department ran a given process for the whole organization, producing brands that had belonged to other companies a few years earlier. "Plants had been assigned specific products without regard for previous ownership," a later account noted. The founding firms stopped existing as firms. Allen & Ginter's Richmond works, Kinney's, Goodwin's, and Duke's own Durham branch became organs of a single body, wired together by function instead of by history. As an operating model it was close to perfect. As a legal position it turned out to be fragile.
The trust had a curious relation to the land. American Tobacco focused on making and selling, and left the growing of tobacco to independent farmers. That looks like restraint until you see the other side of it. A company that buys most of a crop from farmers who have nowhere else to sell sets the farmers' price. Duke's aggressive cuts in the prices paid to growers led directly to the Black Patch Tobacco Wars of 1906 to 1908. The company that did not own the fields still governed them.
In 1910, Lewis Hine photographed child laborers working for American Tobacco in Wilmington, Delaware. The photograph sits in the record next to the cigarette cards with their fancy-dress costumes and lighthouses. The same enterprise produced both.
The trust also looked outward, to Great Britain, China, and Japan. At the start of the 1900s Duke went at the British market the way he had gone at the American one. The response was instructive. The British manufacturers, who had been divided, merged into the Imperial Tobacco Company of Great Britain and Ireland, and after two years of hard competition in Britain, Imperial carried the fight into the United States. American Tobacco looked for a settlement. Under the agreement, American kept the American trade and Imperial kept the British territories, and a jointly owned third company, the British-American Tobacco Company, took the rest of the world. BAT was founded in 1902. It is worth remembering.
Duke was also diversifying across products. He kept interests in plug, snuff, and smoking tobacco in case the cigarette market turned. That was sound hedging in business terms. In legal terms, the breadth of the consolidation — every form of tobacco, every major market — "ultimately made the Trust so vulnerable to regulation and judicial dissolution," in the historian Allan Brandt's account. The hedge against the market increased his exposure to the state.
Testimony from the Bedroom
Early in 1908, James B. Duke gave testimony in the federal antitrust case against his company. He gave it from his bedroom in a house that belonged to his brother.
The house was 1009 Fifth Avenue, a Beaux-Arts mansion with a copper mansard roof at the southeast corner of Eighty-second Street, across from the Metropolitan Museum of Art. The developers William W. and Thomas M. Hall had built it on speculation, and Benjamin Duke bought it in June 1901. Contemporary reports first said James was the buyer, a small confusion that suits the brothers' shared life. Benjamin was worth about $60 million at the time. For reasons no one has explained, he did not live in his own new house. He stayed at the Hoffman House hotel until 1907, and the mansion sat empty.
The brothers' private lives had moved fast in the meantime. In 1904 James married Lillian Fletcher McCredy. They divorced in 1906 and had no children. Late in 1906, at a party in Benjamin's house, he met Nanaline Holt Inman, a widow and cotton heiress. He married her eight months later, in July 1907, and in November he acquired 1009 Fifth Avenue from his brother. By the 1910 census the house held James, Nanaline, her mother Florence Holt, and nine servants. Benjamin had moved to the Plaza Hotel.
So the government deposed the most powerful man in American tobacco in a borrowed bedroom, and that is where its case against him began to become a record. The Department of Justice had filed suit in 1908, naming sixty-five companies and twenty-nine individuals. The Supreme Court heard argument from January 3 to 6, 1910, and ordered reargument, which took place from January 9 to 12, 1911. On May 29, 1911, the same month the Court ordered the Standard Oil Trust dissolved, Chief Justice Edward D. White delivered the opinion for eight justices. John Marshall Harlan concurred in part and dissented in part.
The holding was not subtle. The combination, the Court said, was in restraint of trade and an attempt to monopolize. The opinion went further than that and condemned the whole structure, every corporate and individual element of it, taken together and taken apart. (Its language appears in Part III.) The integration that had been the trust's strength now made the Court's order hard to carry out. Dissolution required eight months of negotiation over how to break up something that had been designed so it could not be divided. Trademarks had to be assigned, plants had to be sorted, and investors holding millions in securities had to be protected. No resulting manufacturer could be left with a monopoly on any single type of tobacco product.
The assets went to four companies: a reconstituted American Tobacco, R. J. Reynolds, Liggett & Myers, and P. Lorillard. One detail in the dissolution reads like an epigram. Liggett & Myers got Allen & Ginter's Richmond plant, the old works of the firm that had rejected the Bonsack machine. Liggett & Myers also got the Duke-Durham branch of the American Tobacco Company, the factory where Buck had installed that machine on April 30, 1884. The machine-taker and the machine-refuser ended up under the same new owner, and it was not Duke.
Duke kept one thing: the British-American Tobacco Company. The company he had made as a peace treaty with his only real rival became the only part of the tobacco empire still under his control.
Economists still argue about what 1911 accomplished. The monopoly became an oligopoly. The main effect of the breakup, by one reading, was a surge in advertising and promotion, as the successor companies fought each other with marketing instead of price. In 1938, Thurman Arnold's Antitrust Division held hearings before the Temporary National Economic Committee on whether the successors were again acting together, and found that three of the four were. The economic historian Leslie Hannah later argued in an essay, "The Whig Fable of American Tobacco," that most historians have misread the company altogether. The case is settled law. The meaning of the case is not.
What can be said with confidence is that Duke had built a machine so integrated that the state could only take it apart, not stop it. Advertising had been his weapon in 1889, and after 1911 it became the whole industry's weapon.
Falling Water
Walker Gill Wylie had a river and not enough money. In 1900, he and his brother financed a hydroelectric station at India Hook Shoals on the Catawba River near India Hook, South Carolina. That was the beginning of the Catawba Power Company. Wylie's ambition went past one station to a series of hydroelectric plants strung along the river, and the second plant needed capital that the first could not supply. He found that capital in Durham, by way of New York. He persuaded James B. Duke and Duke's partner, James Blaney, to put money into a new company, the Southern Power Company, founded in 1905. Accounts vary on dates and titles: one credits Duke with establishing the Catawba Power Company in 1904 after organizing the American Development Company to buy land and water rights along the Catawba. The shape of the story is the same in every version. A man with an idea and an underfunded dam met a man with a fortune and a theory.
The theory was about mills. In 1892 the Dukes opened their first textile firm in Durham, run by Benjamin. Textile mills need power, and power in the Carolina Piedmont was scarce, local, and unreliable. Electricity generated on the Catawba and carried by wire could run the Dukes' own factories. It could also run everyone else's.
Within two decades, the Duke power system supplied electricity to more than 300 cotton mills and other industrial companies and built a grid serving cities and towns across the Piedmont of both Carolinas. In 1917, Blaney founded the Wateree Power Company as a holding company for the utilities Duke and his associates had built or owned. In 1924 its name became Duke Power. In 1927, most of the subsidiaries — Southern Power, Catawba Power, Great Falls Power, Western Carolina Power — were folded into it. That same year a five-story headquarters, the Power Building, was completed at 440 South Church Street in Charlotte. It was demolished on February 24, 2007. The Buck Steam Station in Rowan County, built in 1926, carries his nickname. Lake James, a power reservoir in western North Carolina created in 1928, carries his given name. He did not live to see either one.
The parallel with tobacco is exact, and that is the most interesting thing about it. In tobacco, Duke had found a fragmented industry, introduced a technology that gave one producer a decisive cost advantage, and consolidated. In power, he found a fragmented region full of mills, each depending on whatever water ran past it, and offered a technology that made the region's energy a single system. The cigarette trust sold a product to millions of individual consumers. The power company sold an input to industrial customers who could not easily leave. The first got him sued by the federal government. The second became a regulated utility — a monopoly the state approved instead of prohibiting.
The company's later history is not Duke's to answer for, but it continues the same lineage of scale and friction. In 1973 Duke Power, through its Eastover Mining subsidiary, faced a thirteen-month strike at the Brookside mine in Harlan County, Kentucky, which ended after a foreman shot and killed a twenty-two-year-old miner named Lawrence D. Jones. In 2014 the Dan River coal-ash spill led to guilty pleas and $102 million in fines and restitution. The utility he co-founded now serves about 7.2 million customers over 104,000 square miles, through more than 250,000 miles of distribution line. In 2023 it reported $29.06 billion in revenue. Wylie's single station at India Hook Shoals sits somewhere at the root of all that, the way the Bonsack contract sits at the root of the cigarette business.
A House That Looks Smaller Than It Is
Horace Trumbauer had no formal architectural education. Duke hired him anyway in 1909, and nobody knows exactly why. Duke's peers on Fifth Avenue had mostly hired McKim, Mead & White, and Trumbauer had finished few buildings in Manhattan. He was a self-taught Philadelphian who handed the details of his designs to the trained architects in his office. The client and the architect had something in common: both had arrived at the top of their trades without credentials, and both knew how to take credit for systems run by other people.
For Duke, Trumbauer borrowed heavily from the Château Labottière, a townhouse built in Bordeaux in 1773 by the architect Étienne Laclotte, and scaled it up. The Fifth Avenue house was designed to look like a two-story building from the street. A third floor sat hidden behind the rooftop balustrade, and the service rooms were buried in the basement. Kitchen, laundry, and servants' dining room went downstairs. Twelve suites for servants went into the concealed attic. The public saw a limestone façade so fine it read as marble, long windows, high ceilings, and a two-story portico on Seventy-eighth Street topped by a pediment with carved figures. Inside were four great corner rooms — library, dining room, music room with its small performers' balcony, drawing room — set around an entrance hall with a polychrome tile floor and a U-shaped marble staircase. The household that made the house work was largely out of sight.
One of the architects in Trumbauer's office who worked out those details was Julian Abele, an African American designer, and the Duke mansion was among his early works.
The house cost about a million dollars in the end. The Dukes moved in during 1912 with fourteen servants. Nanaline was pregnant, and their only child, Doris, was born that November 22. In January 1914 the Times ran a full illustrated page on it and called it the "costliest home opened on Fifth Avenue within a year." The 1920 census found that every servant then in the house had been hired after 1915, except the cook, Mathilda Andrews, who was fifty. The staff kept turning over around her.
The irony is that Duke did not much like living there. The Seventy-eighth Street house was a secondary residence. His real attention went to Duke Farms in Hillsborough, New Jersey, where he worked with the landscape designers James Leal Greenleaf, of the Olmsted firm, and Horatio Buckenham to turn more than 2,000 acres of farmland and woodlots into a made landscape: two conservatories, nine lakes, thirty-five fountains, forty-five buildings, more than two miles of stone walls, and more than eighteen miles of road, with sculpture throughout. Nanaline spent a good deal of time in the city. After the First World War, James mostly did not. The family took Rough Point in Newport. In his last five years Duke lived largely at Lynnwood, a Colonial Revival house at 400 Hermitage Road in Charlotte, designed by Charles C. Hook, built in 1914 and greatly enlarged in 1919. Charlotte was the capital of the power company. The man who had conquered New York spent his last years back in the Carolinas, close to the river.
There are two kinds of Duke houses, and they rhyme with his two kinds of business. The New York houses were statements, either borrowed from his brother or built from a French model, with the servants hidden away. The Southern houses — the farm landscape in New Jersey and the enlarged house in Charlotte — were working places, organized around water, roads, and the things that grow or turn. He tore down Henry Cook's house to build a façade. He spent his own attention on systems.
Equal Footing
The man who most deserves credit for Duke University may be one whose name it does not carry.
John Franklin Crowell was a Yale-trained economist who became president of Trinity College in 1887. Trinity was then a small Methodist school in rural Randolph County. Its origins went back to 1838 and Brown's Schoolhouse, a subscription school organized by Methodists and Quakers, and it had changed names three times before adopting its motto, Eruditio et Religio. Crowell believed in the German university model, where research outranked recitation. He rewrote the curriculum and persuaded the trustees that the college should move to a city. In 1892 it opened in Durham.
The move was paid for in tobacco money. Julian S. Carr, already the college's largest benefactor, gave a sixty-two-acre tract called Blackwell Park, valued at $50,000, which became the original campus. Washington Duke gave money for the endowment and for construction. One account puts his first gift at $85,000 and another at $100,000. In 1896, 1899, and 1900 he gave three more gifts of $100,000 each. One stipulation came with this money, and he described it as the only condition he ever placed on a gift to the college:
— Washington Duke, condition attached to his gift to Trinity CollegeOpen its doors to women, placing them on an equal footing with men.
When that gift arrived in 1896, Trinity had four women enrolled, and three of them were daughters of faculty members. A women's dormitory went up in 1897 and was named for his daughter Mary. By 1904 there were fifty-four women. In 1903, Washington Duke wrote to the trustees withdrawing the condition, apparently satisfied that it had done its work.
The year 1903 brought a harder test. John Spencer Bassett, a popular professor, published an essay in the South Atlantic Quarterly titled "Stirring Up the Fires of Race Antipathy." Near its end he wrote that Booker T. Washington was "the greatest man, save General Lee, born in the South in a hundred years." Democratic Party leaders, newspapers, and much of the state demanded he be fired, and parents were urged to withdraw their sons. Bassett offered to resign. The trustees voted eighteen to seven to refuse his resignation, in a statement on academic liberty that is still quoted. In 1905, Theodore Roosevelt came to Trinity to praise them. The president who led the college through that fight was John C. Kilgo, inaugurated in 1894, who had done more than anyone to draw the Duke family into the college's life. In 1900, under Kilgo, Trinity had become the first white institution of higher learning in the South to invite Booker T. Washington to speak.
In 1910, William Preston Few became president. He wanted a Southern counterpart to Harvard and Yale, and he was patient enough to wait for one.
On December 11, 1924, James B. Duke signed the indenture creating the Duke Endowment, a $40 million trust. Its income was to go to specific places: Trinity College, to be built into a university; Davidson College; Furman University; Johnson C. Smith University; nonprofit hospitals and children's homes across the two Carolinas; rural Methodist churches in North Carolina; and retired Methodist preachers and their surviving families. This was not an open-ended pledge to "education." It was an allocation formula written in the same spirit as a utility's service territory map, with named recipients in a defined region and the shares fixed in advance.
Few insisted the university be renamed Duke. Duke at first resisted, on the reasonable ground that it would look self-serving. He accepted the name on the understanding that it was a memorial to his father and family. The tycoon who had put his initials on nothing in particular allowed a university to be named for Washington Duke.
Construction started at once. The original campus, now East Campus, was rebuilt in Georgian style between 1925 and 1927. A mile west, most of the Collegiate Gothic buildings of West Campus were finished by 1930. Duke Chapel, 210 feet tall and set at the campus's highest point, was completed in 1935. The architect who largely designed West Campus was Julian Abele, the same man who had worked out the details of the Fifth Avenue house a decade and a half earlier, again from inside Trumbauer's office.
That is where the American story turns hard to look at directly. In the 1920s the university Abele designed excluded Black students and kept Black people out of dining halls and dormitories. It stayed a whites-only institution, in students and faculty, until 1963, when five Black students enrolled: Mary Harris, Gene Kendall, Wilhelmina Reuben-Cooke, Cassandra Smith Rush, and Nathaniel White. The man who drew its quadrangles could not have studied in them.
In 1925, Duke added another request: a School of Medicine, a School of Nursing, and a hospital, with the aim of improving health care in the Carolinas. When the medical school accepted applications for its first class in 1929, three thousand people applied for seventy places, among them four women. Within four years of its founding the Association of American Medical Colleges ranked it among the top twenty medical schools in the country. Robert Durden's biography, Bold Entrepreneur: A Life of James B. Duke, describes in detail how he approached the endowment the way he approached everything else: as a structure to be engineered, not a gesture to be admired.
The money that built the hospital came from cigarettes. The record shows no evidence that anyone involved in 1924 saw that as a paradox, and it would be anachronistic to pretend they did. The paradox is ours, and it is permanent, which may be the most Duke-like thing about it.
Sold, American!
James B. Duke died at the house on Seventy-eighth Street on October 10, 1925, after several weeks of pneumonia. He was sixty-eight. His wife and daughter were at the bedside. The newspapers could not agree on what he was worth. The Times said his will disposed of $100 million. The Herald Tribune headline said $150 million and reported that his twelve-year-old daughter would get the bulk of it. About half of his estate went to the Duke Endowment, adding roughly $67 million to the trust he had founded ten months earlier. Much of the rest went to Doris, and the press named her "the richest girl in the world."
He was buried in the Memorial Chapel on the Durham campus beside his father and his brother. A statue of him stands in front of the West Campus quadrangle.
His will contained an odd clause. The houses were to be sold immediately, and Doris was to receive enough money to buy them back. Nanaline wanted to sell. Doris did not. In February 1927, in a "friendly" suit, the fourteen-year-old won the Fifth Avenue house outright, valued at $1.6 million, with furnishings worth $600,000, and with responsibility for a $615,000 mortgage. She also won Duke Farms, and she made few major changes to it except to turn her father's conservatory into Display Gardens in his memory. These were open to the public from 1964 until her foundation's trustees closed them in May 2008. As an adult she called the Seventy-eighth Street mansion "the rock pile." In 1929, as apartment towers rose around Fifth Avenue, Nanaline asked a court to cut the house's assessed value from $1.6 million to $970,000. Doris married James H. R. Cromwell there in February 1935. In January 1958, mother and daughter gave the house to New York University's Institute of Fine Arts. Robert Venturi's firm, on its first project, converted the basement service rooms into laboratories. The servants' attic became faculty offices, and the second-floor bedrooms became a library. In 1970 the city designated it a landmark.
The Times had asked whether, twenty-five years after 1909, people would be talking about the "old" Duke house. They were. They still are.
The cigarette business went on without him, and in the direction he had set. In 1925, the year Duke died, George Washington Hill's father died, and Hill became president of the American Tobacco Company. A year later Lucky Strike held a fifth of the American market. In 1927 Hill began aiming cigarette advertising at women, the first time anyone had done so, with help from Albert Lasker of Lord & Thomas and Edward Bernays. Lucky Strike rose to 38 percent of U.S. sales. Hill's salary during the Depression exceeded $2 million. On radio, and later on early television, American Tobacco's ads featured a fast-talking tobacco auctioneer, L. A. "Speed" Riggs, whose chant always ended with the same two words: Sold, American! The industry Duke had made into a mass market through printed cards and price had become a mass market through voice.
In 1969 the company reorganized under a holding company, American Brands. In the 1970s and 1980s American Brands bought businesses that had nothing to do with tobacco, and it later renamed itself Fortune Brands. American Tobacco left Durham in the late 1980s. In 1994, British American Tobacco, the joint venture Duke had founded in 1902 to divide the world with his British rival and the only piece of the empire he kept after 1911, bought American Tobacco, the company he had founded in 1890. The child bought the parent. The corporate entity was formally wound up on December 22, 1994, and Lucky Strike and Pall Mall went to BAT's American arm, Brown & Williamson. A decade later Brown & Williamson merged with R. J. Reynolds, one of the successor companies the Supreme Court had carved out of Duke's trust.
The Durham factory sat mostly empty until 2001, when Capitol Broadcasting bought it, and in 2004 it reopened as offices, shops, and restaurants. The brick buildings carry names from the old company — Fowler, Crowe, Strickland, Reed, Washington — and a waterfall now runs through the middle of the site. The railroad bed that once served the factories has become a twenty-two-mile walking trail named for the American Tobacco Company. Duke University now occupies much of the office space.
Two signs stand above it. One is a water tower, painted with the Lucky Strike name. The other is a neon sign of a bull, the Old Bull. Bull Durham was the rival a young man walked away from in 1879 because it was too strong to fight. Both signs light up at night over a campus where the university named for his father rents the rooms.
Part IIThe Playbook
Duke left behind almost no aphorisms, letters for posterity, or speeches. What he left were contracts, consolidations, dams, an indenture, and buildings. The principles below are read from those structures. Some are admirable, some are cautionary, and several are both. They are offered as patterns to study, not virtues to imitate without thinking, and the evidence for each comes from the story above.
The principles
- 1.Leave the category you cannot win.
- 2.Adopt the technology incumbents find embarrassing.
- 3.Buy exclusivity with commitment, not cash.
- 4.Put the aspiration in the package.
- 5.Spend the margin to buy the market.
- 6.Integrate until the seams disappear — and know what that costs.
- 7.Your ceiling is political, not economic.
- 8.Partition the world when you meet your equal.
- 9.Back the builder who has run out of money.
- 10.Move from product to infrastructure.
- 11.Write the gift like a charter.
- 12.Tear down the merely good; build the permanent.
D
The Decision Ledger
The moves that defined Duke's career, in sequence.
1879
Declines to fight Bull Durham in pouched smoking tobacco; moves the family firm into cigarettes.
1884
Installs the first Bonsack machine in Durham on April 30 under a secret contract: all production on Bonsack machines, royalty cut from $0.30 to $0.20 per thousand.
1888
Packages "Histories of Poor Boys Who Became Rich" booklets with cigarettes.
1889
Spends $800,000 on advertising against net profits under $400,000; forces rivals to cut prices.
1890
Founds the American Tobacco Company on January 31; five firms, 90% of U.S. cigarettes.
1902
After two years of war with Imperial Tobacco, divides the world and forms British-American Tobacco.
1905
Backs Walker Gill Wylie's Catawba hydro ambitions through the Southern Power Company.
1911
Supreme Court orders the trust dissolved on May 29; Duke keeps BAT.
1924
Signs the $40 million Duke Endowment indenture on December 11; Trinity becomes Duke University.
1925
Requests a medical school, nursing school, and hospital; dies October 10, leaving roughly $67 million more to the Endowment.
Principle 1
Leave the category you cannot win
Duke's first strategic act was a retreat. Bull Durham owned pouched smoking tobacco, and the Duke family firm was a smaller competitor in the same town. Founders often make the opposite choice. They believe that effort, pricing, or a slightly better product will eventually wear an incumbent down. Duke saw that the contest was structural and that his side had already lost it. His move into cigarettes was an exit before it was an entry.
The cigarette was attractive precisely because nobody yet held a lasting advantage in it. In 1879 the business was handwork with a tobacco supply attached, a category where the eventual winner had not been decided because the decisive technology had not arrived. Duke put himself where change would come before he knew what the change would be. When the Bonsack machine appeared three years later, he was already standing in the one market it would transform.
Leaving was disciplined, not cowardly. It required him to admit that his family's existing trade was a dead end for them, which is a harder thing for an heir to say than for an outsider. The more personal your attachment to a market, the more expensive this kind of honesty feels.
Tactic: List every market where a rival has an advantage you cannot copy within three years, and move your best people out of those markets into the one where no one has won yet.
Principle 2
Adopt the technology incumbents find embarrassing
Allen & Ginter turned down the Bonsack machine almost immediately. The reason was not that it failed. The reason was that it worked in a way that looked cheap. Machine-rolled cigarettes were uniform, and uniformity carried a stigma. To an established brand built on skilled hand work, the machine looked like a threat to quality, and in a narrow sense it was. In every sense that mattered commercially, it was an enormous improvement: 133 cigarettes a minute, half the rolling cost, and perfect consistency.
Incumbents are usually right about the flaws in a disruptive technology. They are wrong about how much those flaws matter. Their reasons for rejecting it — brand risk, customer complaints, breakdowns — are real costs for them because they have something to protect. Duke had less to protect, so the same flaws cost him less. The stigma that kept his rivals away made the machine cheaper and less contested for him.
There is a useful symmetry in how 1911 turned out. Allen & Ginter's Richmond plant and Duke's Durham branch both went to Liggett & Myers in the dissolution. The firm that refused the machine and the firm that embraced it ended under the same owner. In between, the one that said yes ran the industry for twenty years.
Tactic: When a respected competitor dismisses a new tool because it "cheapens" the product, run a small test of it in your own operation within the quarter, because their reason for staying away is likely to become your advantage.
Principle 3
Buy exclusivity with commitment, not cash
Duke's Bonsack contract is a lesson in bargaining from leverage that looks like weakness. He did not buy the machines. He rented two. He did not pay a premium. He paid less. He got that discount by offering the one thing the Bonsack company most needed: a customer who would run all of his production on its machines and prove the technology at commercial scale. The royalty fell from 30 cents per thousand to 20 cents, permanently lowering his costs below what every licensee paying full price could match.
He also hired one of Bonsack's mechanics, which was arguably the most valuable part of the deal. The machines broke down, and the expertise to keep them running could not be licensed or patented. It lived in a person, and Duke bought it. His uptime beat his competitors' uptime, which meant his effective capacity was higher than the machine count suggested.
He kept the arrangement secret. That is the part that should make modern readers uneasy and also pay attention. A cost advantage that rivals know about can be competed away through their own negotiations. A cost advantage they do not know about lets you set prices they cannot understand.
Tactic: In your next negotiation with a supplier of new technology, offer volume commitment and public proof of the product in exchange for preferential terms, and hire or retain the person who knows how to keep the system running.
Principle 4
Put the aspiration in the package
The 1888 booklets, Histories of Poor Boys Who Became Rich and Other Famous People, were marketing that sold a worldview along with the product. The cigarette was cheap and uniform, and the card or booklet inside was the thing that varied: Jay Gould one week, Edgar Allan Poe the next, Confederate generals, lighthouses, yacht colors, jokes. The product became a subscription to a small, cheap encyclopedia of aspiration and diversion.
This solved the commodity problem that Duke's own machine had created. Once every cigarette looked alike, the brand had to differentiate through something other than the cigarette. Collectability brought customers back. Story gave them a reason to like the company. The cards turned a habit into a hobby, and that is a stronger form of loyalty than price.
The darker side is also visible. The same packages sold a product with a cost the 1880s did not count. The playbook works whatever it is attached to, which is exactly why it needs judgment about what it is attached to.
Tactic: If your product is becoming interchangeable, add a collectible, serialized, or story-driven element to the experience that makes customers want the next one more than they want a cheaper substitute.
Principle 5
Spend the margin to buy the market
In 1889, Duke spent $800,000 on advertising while earning net profits of less than $400,000, and he cut prices at the same time. On a profit-and-loss view that was reckless. In terms of market structure it was a siege, and he could afford it because his costs were lower than anyone else's. Each dollar of price cuts hurt him less than it hurt Allen & Ginter, Kimball, Kinney, or Goodwin.
The goal was never to win that year. It was to make independence unbearable for his competitors. In 1890 they gave up and joined the American Tobacco Company, which then controlled 90 percent of U.S. cigarette production. A year of thin profits bought a generation of dominance.
The precondition is everything here. Without the Bonsack cost advantage, the 1889 campaign would have been suicide. Spending through your margin is a weapon only when you know, with evidence, that your opponents run out of money first.
$
When Spending Through Margin Works
The conditions present in 1889 — and their absence in most imitations.
| Condition | Duke in 1889 | Typical imitator |
|---|---|---|
| Structural cost advantage | Present Bonsack royalty at $0.20 vs. rivals' $0.30 | Absent Same suppliers, same costs |
| Rivals' visibility into your costs | None Contract was secret | High Pricing is public and benchmarked |
| Endgame | Defined Consolidation into one holding company | Vague "Market share" without exit |
| Regulatory exposure | Latent Sherman Act passed the same year | Immediate Modern antitrust scrutiny |
Tactic: Before you launch any price or spending war, write down your cost per unit and your best estimate of your strongest rival's, and fight only if the gap means they run out of cash at least twelve months before you do.
Principle 6
Integrate until the seams disappear — and know what that costs
Inside the American Tobacco Company, the old firms disappeared. One department ran a single process for the whole organization. "Plants had been assigned specific products without regard for previous ownership." This is what real post-merger integration looks like: every facility assigned by function, so that the combined enterprise is more efficient than its parts could ever be separately.
That same integration made the dissolution an eight-month negotiation over trademarks, plants, and securities. Once the state decided to act, the seamlessness could not shield the company from the breakup. It only made the breakup more painful and complicated. Integration raises efficiency and also raises the cost of separation, for owners, for courts, and for anyone trying to undo the structure later.
The operator's lesson is to integrate deliberately, knowing it is a one-way door. Integration is what makes a roll-up worth more than its pieces. It also means that if the roll-up is ever challenged, the pieces cannot simply be returned to their former owners.
Tactic: When combining acquisitions, decide in advance which units must stay separable — with their own books, brands, and leadership — so that a forced or voluntary divestiture does not destroy value.
Principle 7
Your ceiling is political, not economic
By 1910 American Tobacco had absorbed about 250 companies, controlled 80 percent of American tobacco products in every category, and grown its equity more than twelvefold. It had not run out of market. It had run out of political permission. The breadth that protected it against changes in consumer taste — cigarettes, plug, snuff, smoking tobacco, and expansion to Britain, China, and Japan — "ultimately made the Trust so vulnerable to regulation and judicial dissolution."
The Sherman Act was passed in 1890, the year Duke founded the company. For seventeen years the law and the trust grew side by side without colliding. Then came an indictment in 1907, a suit naming sixty-five companies and twenty-nine individuals in 1908, and dissolution in May 1911. Duke did nothing different in 1907 than in 1897. The country changed. A dominant position that keeps growing will eventually be judged by the sovereign, not the market.
Duke's later career suggests he learned this. His power business became a regulated utility, a monopoly that operated under state supervision instead of in defiance of it. The second empire was built in a form the government could live with.
Tactic: Once you have more than half of any market, assign someone senior to model how a regulator would see your position and design the business so it can survive their likely remedy.
Principle 8
Partition the world when you meet your equal
In Britain, Duke met something he had not met in America: a rival that could fight back. The divided British manufacturers merged into Imperial Tobacco, and after two years Imperial carried the war into the United States. Duke did not escalate to the end. He negotiated a division. America went to American Tobacco, the British territories went to Imperial, and the rest of the world went to a joint venture, British-American Tobacco.
The settlement was strategically excellent and personally decisive, though no one could have known that in 1902. When the Supreme Court broke up the American trust in 1911, BAT was the only part that stayed under Duke's control. The peace treaty outlasted the conquest. Ninety-two years later, BAT bought the company Duke founded.
Knowing when a war has no profitable end is a rare skill among aggressive operators. The fighting instinct that built the American monopoly would have destroyed value in a transatlantic contest between equals. Duke turned a war into a jointly owned company.
Tactic: If you and a competitor of similar strength are damaging each other across several markets, propose a split of territories plus a jointly owned vehicle for the contested ground before either of you runs out of money.
Principle 9
Back the builder who has run out of money
Walker Gill Wylie had what Duke lacked: a hydroelectric station already running at India Hook Shoals and a plan for a chain of plants along the Catawba. Wylie lacked what Duke had in abundance, which was capital. The Southern Power Company of 1905 joined the two. Wylie persuaded Duke and his partner James Blaney to invest, and that investment became the root of a utility that now serves millions.
The pattern appears more than once in Duke's career. The Bonsack machine came from someone else's invention. The power system came from someone else's dam. Trinity College came from someone else's institution, built over decades by Crowell, Kilgo, and Few. Duke rarely started from nothing. He recognized working prototypes owned by people who could not scale them and supplied the resources and system-building that turned a proof of concept into a regional or national structure.
The skill is not invention. It is noticing where an idea has already shown it works and where the only missing piece is the one you have. That is a far better investment profile than funding pure speculation.
Tactic: Keep a list of operators who have proven their concept but cannot finance its expansion, and approach them with capital plus a specific plan for scale instead of an open-ended check.
Principle 10
Move from product to infrastructure
Cigarettes are sold to individuals who can switch brands. Electricity is sold to mills that cannot easily switch suppliers once they are wired in. Duke's move from tobacco to power, starting with powering his family's own textile mill, was a move from a consumer product to infrastructure underneath an entire regional economy. Within two decades the system served more than 300 cotton mills and industrial customers and ran grids for Piedmont cities and towns.
Infrastructure behaves differently. It is capital-intensive and slow to build. Once built, it is very hard to dislodge, and it is regulated in a way that often protects the incumbent. The power company also solved a problem Duke had felt directly: the state allowed a utility monopoly in a way it had not allowed a tobacco monopoly. The name Duke Power, adopted in 1924, still exists inside one of the largest utilities in the United States.
The most durable businesses usually sit beneath their customers, not in front of them. Duke grasped that the economic history of the Carolinas would be written on spindles and looms, and he put himself under the looms.
Tactic: Identify the input your best customers cannot operate without and cannot easily switch, and ask whether you can own or build that layer instead of competing at the surface.
Principle 11
Write the gift like a charter
The Duke Endowment was not a general donation. It was an operating document. Its indenture named its beneficiaries: four colleges, nonprofit hospitals and children's homes in two states, rural Methodist churches in North Carolina, and retired pastors and their families. It defined a region. The original $40 million was later supplemented by roughly $67 million from his estate, so that about half of his fortune went into a structure he designed in full during his last year alive.
The form is as important as the size. Open-ended philanthropy drifts with the views of whoever administers it. A charter with named recipients and a defined region carries the founder's intentions forward. The Endowment's beneficiaries, the Duke University School of Medicine among them, have been building on that design for a century. Within four years of founding, the medical school ranked among the top twenty in the country.
Duke treated philanthropy as he treated business, as a system that could be engineered to operate without him. The endowment was his last consolidation, and it was the one no court could dissolve.
Tactic: If you are giving away a substantial sum, write the governing document with specific beneficiaries, geography, and allocation rules so that the gift carries out your intent after you are no longer around to explain it.
Principle 12
Tear down the merely good; build the permanent
Duke demolished the Cook mansion, "the best-built house ever torn down in New York City," because good was not the standard he wanted. The Times wondered whether, twenty-five years later, his replacement would be called the "old" Duke house. It was, and it survived. It was designed by a self-taught architect working through Julian Abele, given to NYU in 1958, and designated a city landmark in 1970. The thing he built in place of the thing he tore down lasted.
The pattern runs through his whole career: hand rolling replaced by machines, independent firms replaced by a holding company, scattered mill power replaced by a grid, a small Methodist college turned into a research university. In each case something adequate was cleared away to make room for something designed to last. The cost in each case was real. Farmers in the Black Patch paid, competitors paid, and the people the new institutions excluded paid. The man who tears down the merely good takes on a responsibility to build something that justifies the demolition.
Durden's Bold Entrepreneur: A Life of James B. Duke is the best long account of how those demolitions and constructions fit together across the whole of Duke's life.
Tactic: Before you replace a system that works, write down what the replacement must do that the old one cannot and how long it must last, and do not start the demolition until you can answer both questions.
Free playbook
Get The Business Model Playbook
58 business models, one visual page each: how the money flows, the metrics that matter, and who runs it. Free when you join the Faster Than Normal email.
Free. No spam. Unsubscribe anytime.
Part IIIQuotes and Maxims
In their words
Duke left few words on the record, and his opinions mostly survive in the form of contracts and buildings. The voices below are those that judged, praised, and preserved what he built.
— United States v. American Tobacco Co., 221 U.S. 106 (1911)In and of itself, as well as each and all of the elements composing it whether corporate or individual, whether considered collectively or separately [was] in restraint of trade and an attempt to monopolize, and a monopolization within the first and second sections of the Anti-Trust Act.
— Trinity College Board of Trustees, refusing John Spencer Bassett's resignation, 1903We cannot lend countenance to the degrading notion that professors in American colleges have not an equal liberty of thought and speech with all other Americans.
— Theodore Roosevelt, speaking at Trinity College, 1905You stand for Academic Freedom, for the right of private judgment, for a duty more incumbent upon the scholar than upon any other man, to tell the truth as he sees it, to claim for himself and to give to others the largest liberty in seeking after the truth.
— The New York Times, on Duke's demolition of the Cook mansion, August 1909Is it possible that twenty-five years hence people will be talking about the 'old' Duke house?
— New York City Landmarks Preservation Commission, designating the James B. Duke House, September 15, 1970One of the adornments of Fifth Avenue and one of the last reminders of the Age of Elegance.
Maxims
- Retreat is a strategy. Leaving a category you cannot win is the first step toward owning one you can.
- Stigma is a discount. The technology incumbents call embarrassing is often the cheapest advantage on the market.
- Commitment is currency. Offering a supplier your whole volume can buy terms that cash alone cannot.
- Hire the mechanic. The knowledge needed to keep a system running is worth more than the system's license.
- Sell the story with the product. When the product becomes uniform, the package has to carry the aspiration.
- Bleed only when they bleed faster. Spending through your margin is a weapon only when your costs are structurally lower.
- Integration is a one-way door. What makes a combination efficient also makes it painful to break apart.
- The state sets the ceiling. Past a certain size, your limit is political permission, not demand.
- Make peace with your equal. A jointly owned venture can outlast every conquest you make alone.
- Build what outlives you. Write the charter, name it for someone else, and design it so it can run without your explanation.
