(FORTUNE Magazine) – Like any of today's chief executives, John Jacob Astor did his best to stick to a strict daily schedule. As the leading merchant of the early 19th century, he arrived at the office each morning to tend to his fur and China trade. Then, his affairs complete, he left work at 2 P.M. At that point came dinner, one glass of beer, and exactly three games of checkers. A leisurely ride around Manhattan on his horse followed and, in the evening, perhaps a musicale hosted at his home. "With little personal exertion," two historians have observed, "he became the richest man in America."

Needless to say, the CEO's daybook has changed a bit since. But what about the job itself? At a time when corporate scandals have people wondering about the future of the chief executive, we decided to do a little digging into the past and ask, How did we get here? Can we draw a line from Astor to Dennis Kozlowski?

What's clear is that today's disgraced chieftains didn't just happen. They evolved. In the century-plus that people have been leading great corporations, the CEO has been a protean figure, at times dominant, at times dominated, growing and shrinking in public stature--and undergoing several wholesale transformations. While these shifts were a result of larger social and economic forces, they were also driven by CEOs themselves--or rather by an especially influential handful. What follows, then, is not a list of the greatest CEOs ever. It's the story of the men who helped invent, and reinvent, the idea of the CEO.

Technically, not all of them were even chief executive officers. The title didn't gain wide usage until the 1970s--"president" had been the standard--and it wasn't until the late 1980s that one could safely use the acronym CEO without someone's asking what it stood for. Astor, truth be told, didn't even have many subordinates, since the businesses of his day were staffed by no more than a few clerks. It was only in the second half of the 1800s, with the rise of large hierarchies and a quickening in the pace of business, that the Erie Railroad drew up the first organizational chart. Atop such charts was a box with the word "President." And of the first generation of bosses to fill that box, none was quite so ingenious--or insane--as John H. Patterson.

The tyrant

Patterson built the National Cash Register Co. from 1884 to 1921 using a three-step approach to employee development. First he would break a man's self-esteem. Then he would rebuild it from scratch. Then he would fire him. One NCR executive learned of his dismissal by finding his desk and chair aflame on the company lawn. Charles Kettering, one of the great American inventors, got the ax no fewer than five times, once for nearly falling off his horse at a corporate equestrian event. "When a man gets indispensable, let's fire him," was Patterson's dictum, and the doomed executive could expect little in the way of explanation. "There are just two things," Patterson would say. "Everything you say is wrong. Everything you do is wrong."

As a member of the founding generation of many of America's great corporations--a generation that included such giants as John D. Rockefeller and Andrew Carnegie--Patterson embodied the kind of arbitrary one-man rule that was possible when owners and managers were one and the same. A devotee of crank theories, he once appointed a face reader to the company's board to issue reports on his executives. He banned "harmful" foods--bread and butter, tea and coffee, salt and pepper--from company premises while decreeing the consumption of malted milk and shredded wheat. He had NCR employees weighed and measured every six months, and for a while believed in chewing each bite of one's food 32 times. He poured the contents of executives' desks into the trash so that they could "start clean." He attributed occult significance to the number five.

Some of his innovations were less lunatic. Patterson was the first to introduce such methods as the designated sales territory, the "canned" sales pitch, the paid suggestion system, direct mail, and annual sales conventions where top sellers joined a Hundred Point Club. He even started a training school near the company's Dayton campus where he staged elaborate playlets of sales situations and lectured in his inimitable style. "I have seen him knock a table down and smash it, tear off his collar, take a pitcher of water and smash it against the floor," one executive reported. According to a magazine article cited by a later NCR chief, one-sixth of the country's top executives from 1910 to 1930 had been schooled--and canned--by Patterson.

Though his methods helped pave the way for the modern corporation, Patterson himself never modernized. In 1913 he and 29 NCR officials were convicted of various antitrust violations, including the use of "knockout men" to intimidate store owners from buying from competitors, and the selling of machines bearing rivals' logos that were designed to break. Sentenced to a year in jail, Patterson "needed either a reversal on appeal or an act of God," Mark Bernstein wrote in his book on turn-of-the-century Dayton, Grand Eccentrics. "He got the latter."

On the Tuesday after Easter, Patterson surveyed the rain-swollen Great Miami River and concluded that a terrible flood was going to engulf Dayton. By the time the levees broke that morning, NCR personnel were already building 275 rowboats and baking 2,000 loaves of bread, helping rescue hundreds from a disaster that ultimately claimed 300 lives. Patterson was a national hero. When the court ordered a retrial that never took place, he celebrated characteristically, firing all his co-defendants within months. One of them--whose motto, "Think," Patterson admired--left vowing "to build a larger and more successful company than Patterson's." And that, at IBM, is just what Thomas J. Watson did.

The administrator

Patterson could afford to be a tyrant because he owned nearly all of NCR's stock. But by the early 20th century, founder-entrepreneurs like him were already passing from the scene, replaced by what business historian Alfred Chandler has called a "new subspecies of economic man--the salaried manager." This was a different breed of boss: not an owner (by 1938 the average big-company CEO held just 0.3% of his company's shares) but a professional hired hand, accountable to a widely dispersed group of shareholders. One man, above all, came to define this new role.

Alfred P. Sloan Jr. was Patterson's temperamental antipode. Tall, ghostly thin, and reserved to the point of remoteness, he took the wheel of General Motors at a time, 1923, when the company had descended into chaos under founder William Durant. "He could create," Sloan later wrote, "but not administer." Sloan's first step was to create a corporate office that would coordinate but not run the company's tangled skein of divisions. This multidivisional structure--head-smackingly obvious today, but a major breakthrough at the time--would allow Sloan to spend his time setting company policy while, he noted in a memo, "limit[ing] as far as practical the number of executives reporting directly to the President." (The boss was, after all, a busy guy.) GM's various car lines, which had been competing intramurally, were meanwhile aligned into a finely graduated hierarchy: "Chevrolet for the hoi polloi," this magazine put it, "Pontiac ... for the poor but proud, Oldsmobile for the comfortable but discreet, Buick for the striving, Cadillac for the rich."

Binding it all together was a complex system of committees--and Sloan's relentlessly rational approach to decision-making. A 1938 FORTUNE article described his style:

"Picture him, then, arriving at his office shortly after 9:30 of a Monday morning, striding nervously over the taupe carpets of the broad, paneled entrance halls.... At 10:00 there is a meeting of the Policy Committee.... In all this committee work Mr. Sloan displays an almost inhuman detachment from personalities, a human and infectious enthusiasm for the facts. Never, in committee or out, does he give an order in the ordinary sense, saying, 'I want you to do this.' Rather he reviews the data and then sells an idea, pointing out, 'Here is what could be done.' Brought to consider the facts in open discussion, all men, he feels, are on an equal footing. Management is no longer a matter of taking orders, but of taking counsel."

The same could not have been said of Sloan's chief rival. Henry Ford had continued to run his company in the personal-fiefdom mode, taking it private in 1919, granting wide powers to a thuggish ex-boxer, and literally vandalizing the company's record-keeping department. "There is no bent of mind more dangerous than that which is sometimes described as the 'genius for organization,'" Ford growled in his autobiography, adding the bizarre boast: "And so the Ford factories and enterprises have no organization, no specific duties attaching to any position, no line of succession or of authority." Ford paid a steep price for his principled primitivism, business historian Tedlow has noted. GM would post higher profits than Ford every year from 1925 to 1985.

The faceless CEO

Sloan's model of committees, councils, and controls was so widely copied by other corporations that for a time the CEO seemed to disappear altogether. The 1940s and 1950s produced a generation of dutiful corporate servants whose names were no better known than their chauffeurs'. For a time both General Motors and General Electric were run by men named Charles Wilson, while another Wilson, Sloan Wilson, captured the anonymity of it all in his book The Man in the Gray Flannel Suit.

The migration of power from the individual to the System--what the sociologist Max Weber called the "routinization of charisma"--was, in some sense, complete. But a new, more exciting type of CEO was on the horizon. It was the dawn of the age of the conglomerateur.

The numbers machine

When Harold S. Geneen took the helm of International Telephone & Telegraph in 1959, it was a middle-rung communications concern with operations in Latin America. A decade later ITT was a 350,000-employee colossus with holdings as diverse as Avis Rent-A-Car, Sheraton Hotels, Continental Baking, Aetna Finance, Hartford Insurance, and homebuilder Levitt & Sons. In all, Geneen acquired 350 businesses in 80 countries--often, it is said, after inspecting a company's books for no more than ten or 20 minutes.

A new kind of giant now stalked the earth--and with it a new kind of CEO. While Patterson had been a salesman and Sloan a manufacturing man, Geneen had begun his career as an accountant. That put him in the vanguard of a trend: In 1939 just 7% of CEOs came from financial backgrounds; by 1969 the figure had climbed to 20%. (A decade later it was 31%.) This new generation of polymaths would push Sloan's managerial revolution to its extreme. With the right financial tools, they believed, smart managers could run any assortment of businesses, whether it was (in ITT's case) making lawn products, publishing books, manufacturing auto parts, or selling Twinkies. The product wasn't the important thing. It was all about the numbers.

And Geneen had a supercomputer's capacity for absorbing them. He claimed to digest every monthly report submitted by his 250 division managers. His readings for the annual business plan ran eight feet high. When traveling, it was said, he was accompanied by as many as 14 briefcases. "The drudgery of the numbers," he preached, "will set you free." Then there were the meetings. By Geneen's own reckoning, the company devoted some 200 days a year to "meetings at various organizational levels," the most important of which took place every month in Brussels. There, for four days straight, sometimes 14 hours at a stretch, 120 ITT executives gathered around a green horseshoe table with a large screen displaying statistics. Each executive's presentation was subject to a withering cross-examination by Geneen, who wanted to see not only the numbers, but also the expression of the man presenting them.

For a long span, Geneen made it all work. Earnings grew for 58 consecutive quarters. FORTUNE called him the "man widely regarded as the world's greatest business manager." His many imitators included Gulf & Western's Charles Bluhdorn and W.R. Grace's J. Peter Grace. Yet by the time of his retirement in 1977, the "Geneen machine" was losing steam. ITT had been hamstrung by antitrust suits and discredited by its effort to finance the toppling of Chilean President Salvador Allende. But the slow disintegration of ITT and the other conglomerates had a deeper cause: Managing by the numbers, as opposed to understanding what lay underneath them, couldn't hold together such a crazy-quilt empire.

Oddly, shortly before his death in 1997, Geneen published a book that criticized the modern merger craze. "If you mix beef broth, lemon juice, and flour, you don't get magic," he wrote, "you get a mess."

The statesmen

The decade of the 1970s ranked among the worst for American business. But it remained one of the coziest for American businessmen. Stocks were moribund. Foreign competitors were carving up U.S. firms' market shares. And hints were emerging that American corporations were, in the words of business historian Harold Livesay, "bloated house cats rather than ... the Darwinian tigers they imagined themselves to be." But no one, as yet, was blaming the CEOs themselves.

Instead, CEOs spent a great deal of time blaming outside factors: inflation, the oil shortage, unions, and especially government regulations. What emerged was a group of CEO-statesmen who were better known for their public-policy pronouncements than for their records as enterprise builders. Men like GE's Reginald Jones, DuPont's Irving Shapiro, Chase Manhattan's David Rockefeller, and GM's Thomas Murphy "became almost as familiar around the capital as the Marine Band," noted Time. They lobbied government through the Business Roundtable. They dined with world leaders. They provided a respected and dignified face for corporate America. And they retired together.

At the beginning of 1981, all four men stepped down nearly simultaneously, marking the sudden passing of both the World War II generation of CEOs and the stability of their era. For the world they had known--a world of gentleman's capitalism--was about to come to an end.

The neutron bomb

The man who helped blow up that world was, ironically, Reginald Jones's handpicked successor. When John F. Welch assumed command of GE in April 1981, he finally screamed what no one had wanted to hear: that corporate America's problems were the fault of corporate America itself. After a half-century of unchecked power, managers had become more accountable to one another than to shareholders or customers (indeed, GE was described as a company "with its face to the CEO and its ass to the customer"), resulting in bureaucracy, bloat, and bad products. Breaking from tradition, Welch refused to play the statesman in Washington or even to sit on other corporate boards. Instead, all his energies would go into spinning GE on its head.

The outlines of Welch's revolution are by now well known. He wiped out whole layers of management, jettisoned underperforming units, introduced tough performance measures for employees, and junked the venerable "blue books" that for years had told GE managers what to do and how. Most significant, he redefined the CEO's central purpose in life. Before, GE had focused on growing revenues, even though a bigger company didn't necessarily mean a more valuable one, while its CEOs talked about balancing the interests of employees, shareholders, and society as a whole. Now Welch set a new yardstick for success: to increase GE's stock price, pure and simple.

In this Welch succeeded spectacularly--GE's stock rose a dizzying 5,021% during his 20-year reign, making him the most admired and copied CEO of his day. Yet perhaps the most notable aspect of Welch's shakeup was that he did it before anyone asked him to. The corporate raiders had yet to arrive on the scene. Shareholder revolts were not yet a threat. But they would be soon. And the CEOs who were still operating as passive custodians of their firms would end up wearing targets.

The year of reckoning was 1992. That October, in a shot heard round the corporate world, GM's board unseated CEO Robert Stempel and many of his lieutenants after a $4.45 billion corporate loss. "He [is] the first GM chief executive in more than 70 years to lose control of his board," gasped the Wall Street Journal at the time. Other heads followed: James Robinson at American Express; John Akers at IBM; Paul Lego at Westinghouse; Kay Whitmore at Eastman Kodak. Behind the spate of firings and resignations was a new force on the capitalist scene: big institutional investors. Quietly, almost imperceptibly, their holdings had grown from 15% of major U.S. stocks in 1955 to more than 50% in 1990. Now they had the muscle to challenge or dethrone chief executives who didn't deliver for shareholders. For the first time CEOs had a boss.

The celebrity

Commentators declared an end to the days of corporate loyalty. And in a way they were right. CEOs would never again enjoy the security they once did; according to one study, those hired in the early '90s were three times more likely to get the boot than those hired before 1980. Yet at the same time, CEOs would rise to a level of prominence never before dreamed of.

The era of the celebrity CEO had its beginnings in September 1979, with the arrival of Lee Iacocca at Chrysler Corp. Riding to the car company's rescue at a time when America's economic self-esteem was at a nadir, Iacocca captivated public attention like no CEO since the robber barons. His book Iacocca sold seven million copies. Never mind that Chrysler's turnaround required a huge government bailout or that its stock lagged behind the market 31% during the second half of Iacocca's tenure (or that Germans would own the company within six years of his departure). Here was a CEO cast in the role of national savior--an image that Iacocca, who appeared in more than 80 of his own commercials and toyed with running for President, did little to discourage. "Running Chrysler has been a bigger job than running the country," he once maintained. "I could handle the national economy in six months."

Fueled by a long bull market, mass media fascinated with personality, and business magazines that lionized the top performers (who, us?), the cult of the CEO gathered steam. According to a study by Harvard Business School professor Rakesh Khurana, the number of times a certain business weekly featured a CEO on its cover rose from just once in 1980 to 18 in 2000. Of the five CEOs whom Time has named Man of the Year, three of them--CNN founder Ted Turner, Intel's Andy Grove, and Amazon.com's Jeff Bezos--received the honor in the 1990s. (The other two were Walter Chrysler in 1928 and GM's Harlow Curtice in 1955.)

The media coverage helped foster an impression, accurate or not, that the fate of massively complex organizations hinged mainly on a Great Leader's personality. The result, Khurana writes in his book, Searching for a Corporate Savior, was that CEOs "were no longer defined as professional managers, but instead as leaders, whose ability to lead consisted in their personal characteristics or, more simply, their charisma." It was no longer enough to have a capable CEO. Wall Street wanted star power. As boards increasingly looked to the outside in search of it, executive mobility increased (in 1993 a record 31% of new CEOs at major companies were outsiders), while CEO pay, which had already gone through the roof during the 1980s, headed straight into outer space. Between 1981 and 2000, the compensation of America's ten highest-paid CEOs rose an astonishing 4,300%.

The man in the gray flannel suit had, improbably, become a rock star.

The destroyer

All these trends--the celebritization of the CEO, the huge executive paydays, the single-minded obsession with "shareholder value"--reached their absurd and ultimately abusive culmination in the person of Albert J. Dunlap. If Wall Street had been allowed to play Dr. Frankenstein and build a CEO, his biographer has noted, it probably would have come up with "Chainsaw Al." As head of Scott Paper in the mid-1990s, he boosted the company's stock price 225% by cutting 11,000 employees, R&D, plant improvements, company philanthropy, and nearly everything else within reach, setting up a quick sale to Kimberly-Clark and netting Dunlap $100 million for 18 months' work. "Most CEOs are ridiculously overpaid, but I deserved the $100 million," he reasoned in his bestselling autobiography, Mean Business. "I'm a superstar in my field, much like Michael Jordan in basketball."

On the announcement in 1996 that Dunlap had been hired to turn around troubled Sunbeam Corp., the company's stock jumped 49%. "This is like the Lakers signing Shaquille O'Neal," gushed Scott Graham, an analyst at Oppenheimer & Co. Once again Wall Street cheered Dunlap's slash-and-burn tactics--the more blood spilled, it seemed, the better. Only this time Dunlap's game caught up with him. After several media accounts called Sunbeam's numbers into question, the board of directors began suspecting the worst: that Dunlap's so-called turnaround was based on little more than short-term accounting gimmickry. The man known as "Rambo in pinstripes," who had once posed for a photo holding automatic weapons and wearing an ammunition belt, was fired and accused of engineering a massive accounting fraud. In 2002 he paid a $500 million settlement to the SEC and became the focus of a new investigation by the Justice Department. By then, bankrupt Sunbeam was trading at 2 cents a share.

In the name of creating shareholder value, Dunlap had done more than anyone to destroy it. Yet like many of the CEOs in this article, Dunlap was, in a way, ahead of his time. His downfall prefigured the recent wave of scandals that has seen executives led off in handcuffs and public confidence in CEOs fall to its current low. The end of this disturbing trend is still not in sight. But if anything, it ensures that the job of the CEO--after more than 100 years of evolution--is going to change once again.

FEEDBACK juseem@fortunemail.com