The Internet was born in a distinctly American telecommunications environment — the United States treated telegraph and telephone providers very differently than the rest of the world — and there is good reason to believe that this environment played a formative role in its development, shaping the character of the Internet to come. Let us, then, take a good look at how this came to be. To do so we, must go back to the birth of the American telegraph.
The American Anomaly
In 1843, Samuel Morse and his allies convinced Congress to spend $30,000 to build a telegraph line from Washington D.C. to Baltimore. They believed it was only the first link in a government-sponsored chain of telegraph lines that would span the continent. In a letter to the House of Representatives, Morse proposed that the government buy the full rights to his telegraph patents, and then charter private companies to build out the individual pieces of the network, while retaining certain government lines for official communications. In this manner, Morse wrote, “it would not be long ere the whole surface of this country would be channelled for those nerves which are to diffuse, with the speed of thought, a knowledge of all that is occurring throughout the land, making, in fact, one neighborhood of the whole country.”1
It seemed to him that such a vital communications system naturally fell within the public interest, and thus the sphere of government. Facilitating communications among the several states via a postal service was one of the few powers of the federal government specifically enumerated in the Constitution. His motives were not altogether informed by public-feeling, though. Government control provided Morse and his backers with a clear endgame for their venture – a single windfall payout from the government. In 1845, Cave Johnson, Postmaster General under James Polk, indicated his backing for a public telegraph system such as Morse had proposed: “The use of an instrument so powerful for good or evil cannot, with safety to the people, be left in the hands of private individuals,” he wrote.2 But that was as far as it went. The rest of Polk’s Democratic administration wanted no part in a public telegraph, nor did the Democratic Congress. The party frowned upon Whig schemes for government spending on “internal improvements,” which they considered an invitation to favoritism, venality, and corruption.
Since the government would not act, Morse collaborator Amos Kendall began developing a telegraph network with private backers instead. But the Morse patent did not suffice to supply its owners with a monopoly on telegraphy. Within a decade, dozens of competitors had sprung up, either licensing an alternative telegraph technology (primarily Royal House’s printing telegraph) or simply wildcatting, on highly questionable legal footing. Lawsuits flew back and forth, paper fortunes rose and fell, and ailing companies collapsed or were purchased with watered stock by eager rivals. Out of this scrum, by the late 1860s, rose one dominant player: Western Union.
Fearful whispers of “monopoly” began to spread. The telegraph had already become essential to several aspects of American life: finance, railroads, and the press. Never before had a single private organization loomed so large in American life. The proposal for government control of the telegraph gained new life. Over the decade following the Civil War, the postal committees within Congress mooted a variety of plans for somehow bringing the telegraph within the orbit of the Post Office. Three basic variations arose: 1) the Post Office could sponsor one more more new competitor to Western Union, granting them special access to post offices and postal right-of-ways in return for honoring certain limits on their rates. 2) The Post Office could run its own telegraph service in competition with Western Union and other private enterprises. 3) The government could simply nationalize the entire telegraph system, placing it entirely under the control of the Post Office.
These plans for some kind of postal telegraph found a handful of staunch supporters within Congress, such as Alexander Ramsey, chair of the Senate Post Office Committee. But much of the energy behind the campaign came from outside lobbyists, most notably Gardiner Hubbard, who had experience with public utilities as the organizer of the city water and gas lighting systems in Cambridge, Massachussetts3. Hubbard and his allies argued that a public system would provide the same kind of generally useful distribution of intelligence then provided by paper mail, driving down rates. Surely, they argued, it would serve the public more ably than Western Union’s system, which primarily targeted business elites. Western Union, of course, countered that the prices for a telegram were determined by its costs, and that a public system with artificially low rates would suffer financial ruin and serve no one.
In any case, the postal telegraph never had sufficient backing to face trial in battle on the floor of either chamber of Congress. Instead every proposed bill suffered the inglorious indignity of quiet suffocation in committee. The specter of monopoly never had sufficient power to overcome the fear of government overreach and abuse. The Democrats secured control of Congress once more in 1874, the spirit of national reconstruction in the immediate aftermath of civil war dimmed, and so did the already feeble prospects for a postal telegraph. The idea of putting the telegraph (and later telephone) under state control was revived from time to time over the succeeding decades, but other than a brief (and nominal) government takeover of the telephone as a wartime exigency in 1918, nothing came of it.
This hands-off approach to the telegraph and telephone was a global anomaly. In France the telegraph was nationalized before it was electrified. In 1837, when a private company tried to set up an optical telegraph (using signal towers) alongside the existing state-controlled system, the French parliament promulgated a law that forbade the development of telegraphs not authorized by the government. Britain, on the other hand, allowed private telegraphy to develop unimpeded for several decades. But public discontent at the resulting duopoly led to the imposition of state control in 1868. Throughout Europe, governments put telegraphy and telephony under the authority of the national post office, just as Hubbard and his allies had proposed.
Outside Europe and North America, much of the world was still under the control of colonial powers, and thus had no say in the development and regulation of telegraphy. Where independent governments existed, however, they generally built state-run telegraph systems on the European model.[^undersea] These systems generally lacked the capital to expand at anything like the rate of their American and European counterparts. Brazil’s state telegraph company, for example, under the auspices of the Ministry of Agriculture, Commerce and Public Works, had only 1300 miles of line 1869, whereas the U.S., with a similar area and only four times as many people, had 80,000 miles by 1866.4.
[^undersea] Supplemented by international undersea telegraph cable companies backed by Anglo-American capital.
The New Deal
Why did the United States follow such a idiosyncratic path? One might give some weight to the American “spoils system” of party patronage, which lasted until the last years of the nineteenth century. The entire government bureaucracy, down to local postmasters, consisted of political appointments which could be used to reward loyal allies. Both parties were loathe to create large new sources of patronage power for their opponents, and such would certainly be the case if the telegraph came under the control of the federal government. But the simplest explanation is the traditional American suspicion of a powerful central government – the same basic reason why the structure of American healthcare, education, and other public services appear equally anomalous compared to its peers.
Given the increasing importance of electrical communication to national life and national security, however, the United States proved unable to preserve a completely hands-off stance towards its development. Instead, over the first few decades of the twentieth century, it developed a hybrid system in which private telecommunications was checked by two forces: on one side, a bureaucratic body to continually monitor the rates of the communications companies to ensure that they did not use a monopoly position to extract excessive profits; on the other, the blunt threat that misbehavior would mean dismemberment by anti-trust law. These two forces could act at cross-purposes, as we will see: the theory of rate regulation accepted monopoly power as a given, even “natural” in some circumstances, where duplication of service would be wasteful. The regulators generally tried to mitigate the negative consequences of monopoly by controlling prices. Antitrust prosecution, however, sought to destroy monopoly at its root, and force a competitive market into existence.
The concept of rate regulation originated with the railroads, and was embodied, at the federal level, in the Interstate Commerce Commission (ICC), created by act of Congress in 1887. The primary impetus for the bill came from small businesses and independent farmers. They generally had no choice in which railroad they used to get their products to market, and claimed that the railroad companies exploited this fact to squeeze them for every penny they were worth, while simultaneously granting sweetheart deals to big corporations. The five-member ICC was given authority to oversee the services and rates of the railroad and curb any such abuses of monopoly power, in particular forbidding the railroads from giving special rates to favored companies.5 The 1910 Mann-Elkins Act extended the authority of the ICC over telegraph and telephone. However the ICC, focused on transportation issues, never took much interest in this new responsibility, and more or less ignored it.
Simultaneously, however, the federal government developed an entirely different tool for dealing with monopoly power: trust-busting. The Sherman Act of 1890 empowered the Attorney General to pursue in court any business “combination” found to be in “restraint of trade” – that is, impeding competition through monopoly power. The law was invoked to dissolve several major corporations in the succeeding two decades, include the Supreme Court’s decision in 1911 to dissect Standard Oil into 34 constituent pieces.6
By this time, the telephone, and its dominant provider, AT&T, had eclipsed the telegraph and Western Union in import and power. So much so that in 1909 AT&T was able to acquire a controlling stake in Western Union. Theodore Vail became joint president of both companies, and began the process of weaving them into a coherent whole. Vail firmly believed that the public interest would best be served by a benevolent telecommunications monopoly, and promoted a new company slogan in that spirit: “One Policy, One System, Universal Service.” Bell was ripe for the attention of the trust-busters.
The accession to power of the Woodrow Wilson administration, in 1913, provided an opportune moment for his cabinet of progressives to wield the truncheon of antitrust action threateningly. Postmaster General Sidney Burleson favored full postalization of the telephone, along the European model, but, as usual, this idea got little traction. Instead, Attorney General George Wickersham let it be known that he considered AT&T’s continued acquisition of independent telephone companies to be in violation of the Sherman Act. Rather than go to court, Vail and his deputy, Nathan Kingsbury, brokered an agreement known to history as the “Kingsbury Commitment”, whereby AT&T agreed to three provisions:
- To stop acquiring independent companies,
- To divest its interests in Western Union,
- To allow independent telephone companies to connect into the AT&T long distance network.
But after this moment of danger, the threat of anti-trust went quiescent for decades. The gentle star of rate regulation ascended, with its assumption of a natural monopoly in communications. By the early 1920s, the first condition was relaxed, and AT&T resumed the process of ingesting small independent telephone companies into its operating network. This attitude was enshrined in the 1934 act that formed the Federal Communications Commission (FCC), which replaced the ICC as the regulator of wireline communication rates. By this time the Bell System controlled 90% or more of America’s telephone business, by any measure: 84 million of the 88 millions miles of wire, 2.1 billion of the 2.3 billion monthly telephone calls, 990 million of the billion dollars in annual revenue.7 Yet the FCC’s primary stated purpose was not to revive competition, but “to make available so far as possible, to all the people of the United States, a rapid, efficient, nation-wide, and world-wide wire and radio communications service with adequate facilities at reasonable charges.”8 If a single entity could best provide such service, so be it.
Over the middle decades of the twentieth century, state and federal telecommunications regulators developed a multilayered cross-subsidy scheme to facilitate the development of a universal communications service. Regulatory commissions set rates based on the presumed value that each customer derived from the network, not from the cost of providing service to that customer. Thus business users (who relied on the telephone to conduct their affairs) paid higher prices than residential customers (for whom it was a social convenience). Customers in large urban markets with easy access to many other users paid higher prices than those in smaller towns, despite the greater efficiency of large exchanges. Long-distance users paid an outsize share of the cost of telephone capital, even as technology relentlessly drove down the cost of interstate transmission, and the wages of local exchange operators increased. This complex arrangement of belts and cogs to transmit costs from one place to another worked quite smoothly… as long as there existed one monolithic provider within whom the machinery could operate.
The New Technology
We are trained to see monopoly as a deadening force, which produces indolence and lassitude as a matter of course. We expect a monopoly to jealously guard its position and the status quo, not to serve as an engine of technological, economic, and cultural transformation. Yet it is hard to reconcile this view with AT&T in its heyday, which generated innovation after innovation, anticipating and facilitating the arrival of every new advance in communication.
In 1922, for example, AT&T set up a commerical radio broadcasting station on its building in downtown Manhattan, just one-and-half years after the first such large station went on the air, Westinghouse’s KDKA. The following year it used its long-distance network to re-transmit an address by President Warren Harding to local radio stations across the country. A few years later AT&T also gained a toehold in the film industry, after Bell Labs engineers developed a machine for coordinating motion pictures and recorded sound. Warner Brothers studio used this “Vitaphone” to produce the first Hollywood picture with a synchronized music track (Don Juan), followed by the first “talkie” (The Jazz Singer).
Walter Gifford, who became President of AT&T in 1925, decided to withdraw AT&T from ancillary ventures such as radio broadcasting and film, in part to avoid anti-trust scrutiny. Though the Justice Department had not threatened any action since the Kingsbury Commitment, It would not do to attract undue attention with actions that might be interpreted as an attempt to abuse the telephone monopoly for unfair advantage in other markets. So instead of transmitting its own broadcasts, AT&T, became the primary carrier of signals for the Radio Corporation of America and other radio networks, relaying programs from their studios in New York and other major cities to affiliate stations across the country.
Meanwhile, Radiotelephony service spanned the Atlantic in 1927, inaugurated with a banaly query by Gifford to his counterpart at the British Post Office: “How’s the weather over in London?” Not exactly “What hath God wrought”, but nonetheless it marked an important milestone, which made intercontinental conversations possible decades before the undersea telephone cable, though at great expense and with low fidelity.
But the most important developments, for the purposes of our story, were in high-capacity long-distance transmission. AT&T always wanted to attract more traffic into its long-distance network, which was its primary competitive advantage vis-a-vis the few remaining independents, and highly profitable to boot. And the easiest way to attract more business was to develop new technology to reduce transmission costs – generally by cramming more conversations in a single wire or cable. But, as we have already seen, demand for long-distance communication was expanding beyond traditional, person-to-person telegraph and telephone messages. The radio networks needed their own channels, and television, with far greater capacity requirements, loomed just over the horizon.
The most promising way to meet these new demands lay in the coaxial cable, consisting of concentric (i.e. co-axial, sharing an axis) metal cylinders. The properties of such a conductor, known then as a “concentric main,” were studied as far back as the nineteenth century by the giants of classical physics: Maxwell, Heaviside, Rayleigh, Kelvin, and Thomson. As a transmission line, it had great theoretical advantages, since it could carry wide signal bands and was entirely shielded by its own structure from cross-coupling or interference with other outside signals. When television began to develop in the late 1920s, no existing technology could effectively handle the megahertz or more of bandwidth needed to carry a high-quality broadcast. So Bell Labs engineers set out to turn the theoretical advantages of the cable into a working long-distance, wideband transmission line, including all the necessary ancillary equipment for generating, amplifying, receiving, and otherwise processing signals. In 1936 AT&T established a field trial, with FCC authorization, over 100 miles of cable from Manhattan to Philadelphia. After first testing the system with twenty-seven voice circuits, engineers successfully transmitted moving pictures by the end of 1937.
At the same time, however, another challenger was emerging as a potential high-bandwidth, long-distance communications medium, the radio relay. Radiotelephony, of the sort used in the 1927 transatlantic link, used a pair of radio broadcast signals create a two-way voice channel in the shortwave band. Tying up two entire radio receivers and transmitters and an entire frequency band for a single phone conversation was not economical for overland use. If a way could be found to multiplex many conversations on the same radio beam, however, the financial tables might be turned. Though each individual station would be somewhat expensive, one hundred or so might suffice to relay a signal across the entire United States.
Two frequency bands contended for use in such a system: ultra-high-frequency (UHF) radio (with wavelengths measured in inches or tens of inches) and microwave (measured in centimeters). The higher frequencies of microwave tempted with their greater potential bandwidth, but also posed a greater technical challenge. In the 1930s, responsible opinion at AT&T leaned towards the safer UHF option.
But microwave technology leapfrogged forward during the Second World War, due to the extensive use of those frequencies in radar equipment. Bell Labs proved the feasibility of microwave relay radio with the AN/TRC-69, a truck-borne system capable of carrying eight telephone circuits to and from another antenna within line-of-sight. This allowed military headquarters units to quickly re-establish voice communications as they relocated, without the need to wait for the laying of wire (and without the risk of a line being cut by accident or military action).
After the war, Harold T. Friis, a Danish-born lifer at Bell Labs, led the development of microwave relay communications. A 220 mile trial route from New York to Boston opened at the end of 1945. Beams leapt 30 miles at a time between towers on high ground – using the same basic relay principle as the optical telegraph, or, for that matter, a chain of beacon fires. Upriver to the Hudson Highlands, across a series of hills in Connecticut, over to Asnebumskit Mountain in western Massachussetts, and then down into Boston Harbor.
AT&T was not the only company interested in microwave relay, nor the only company to have gained wartime experience in the technology for manipulating microwave signals. Philco, General Electric, Raytheon, and the television broadcasters all built or planned their own experimental systems in the immediate post-war years: In fact Philco beat AT&T to the punch, lighting up its Washington to Philadelphia line in the spring of 1945.
For over thirty years, AT&T had avoided any real challenge to its business model by either the anti-trust or the regulatory forces within the federal government. In large part it was protected by the assumption that its services formed a natural monopoly – the assumption that to have multiple, competing, disjoint systems laying wires across the country and into every neighborhood would be terribly inefficient. The microwave was the first serious chink in that armor, allowing multiple parities to provide long-distance service without prodigious waste.
Microwave transmission drastically lowered the barrier-to-entry of would-be competitors. Since the technology required only a series of stations every thirty miles or so, there was no need to acquire thousands of miles of right-of-way to build a useful system, nor to build and maintain thousands of miles of wire or cable. Moreover, the capacity of microwave dwarfed that of traditional wire pairs, with each relay able to carry thousands of telephone conversations, or several TV broadcasts. The competitive advantage of AT&T’s existing wired long-distance network thus dwindled to relative insignificance.
However the FCC shielded AT&T from the implications of this for man years, with two decisions it handed down in the 1940s and 50s. First, it refused to issue anything but temporary, experimental licenses to non-common-carrier providers. (A common carrier being anyone offering service to the public at regulated rates, as opposed to, for example, a private internal network for a single enterprise). Thus to even enter the field risked revocation of one’s license at any time. The commissioners were very concerned about unleashing the same sort of problems that had faced broadcasting twenty years earlier, and led to the creation of the FCC in the first place: a cacophony of interference as a wide variety of transmitters polluted a limited frequency commons.
The second decision concerned interconnection. Recall that the Kingsbury Commitment required A&T to allow competing local telephone companies to connect into its long-distance network. Did the same requirement apply to competing private microwave relay systems? The FCC ruled that it only applied in areas where no adequate common-carrier system existed. Thus any competitor building a regional or local network also faced the prospect of sudden disconnection from the rest of the country as soon as AT&T decided to enter their geographical area. The only alternative to preserve connectivity was to build an entire competing national network of one’s own, a daunting prospect to undertake under an experimental license.
By the end of the 1950s, therefore, there was still only one major player in long-distance telecommunications – AT&T. Its microwave network was carrying 6,000 telephone circuits on each route, and reached into every state in the continental United States.10
The other shoe was not long in dropping. But the first landmark challenge to AT&T’s complete, end-to-end control of the telecommunications network came from an altogether different angle.
Gerald W. Brock, The Telecommunications Industry (1981) The telecommunications industry : the dynamics of market structure / Gerald W. Brock
John Brooks, Telephone: The First Hundred Years (1976)
M. D. Fagen, ed., History of Engineering and Science in the Bell System: Transmission Technology (1985)
Joshua D. Wolff, Western Union and the Creation of the American Corporate Order (2013)