Creating the Twentieth Century

by Vaclav Smil

  Edison filed the basic patent in April 1877, and soon afterward came other improved microphone designs by David Hughes and Emile Berliner. The next step was to insert carbon granules between the diaphragm and a metal backplate, and this design, patented by Henry Hunnings in 1879, was improved first by Edison in 1886 by packing the granules in a small button container and then, in 1892, by Anthony White, who interposed the granules between polished carbon diaphragms, with the front one of these placed against a diaphragm made of mica (Garcke 1911b). Known as the solid-back transmitter, this device was used basically unchanged until the mid-1920s, and its improved versions were installed in all telephones until the 1970s, when they were displaced by dynamic transducers.

  The Bell Telephone Co. was formed in July 1877, and a year later, with 10,000 phones in service, Theodore Vail (1845–1920) became its new general manager; he eventually guided the corporation to become a giant monopoly (Jewett 1944). Edison’s carbon transmitter became a centerpiece of a collection of other telephone patents (including Gray’s receiver) that Western Union used in November 1877 to set up its own telephone business. But it left the field entirely to Bell’s company by October 1879, when it acknowledged the priority of Bell’s patent and when Gray received a consolation payment of $100,000. Meanwhile, Bell continued to elaborate and adjust his basic design. Almost exactly a year after filing the patent, he was able to make a call between Boston and North Conway in New Hampshire over a distance of nearly 230 km (Anonymous 1877), and by October 1877 he had a simplified device in a more compact portable form, with an elongated cylinder forming a handle with a flaring mouthpiece (figure 5.13).

  In a society saturated with telephones, it is hard to imagine that the invention was not embraced as rapidly as possible—but there were many technical, financial, and organizational challenges to overcome (Casson 1910). Getting and staying connected were major hurdles. Only when the new service borrowed the established telegraph concept of a central switching office could the numbers of subscribers grow: the first manual switchboard that connected many phones through a single exchange was opened in January 1878 in New Haven, Connecticut, and a decade later more than 150,000 subscribers were served by nearly 750 main exchanges. In 1888 came the introduction of central battery supply to energize all telephones in an exchange instead of relying on subscriber’s own batteries or hand cranking. This important principle assures that modern telephone systems—still powered from a large central battery system backed up by emergency generators—remain open even during electricity outages.

  Although better transmitters and receivers soon obviated the need for shouting, and signal amplifiers (first in 1892) extended the length of useful telephone lines, technical problems and poor connections were common. Calling distances continued to lengthen: in 1877, the first regular line from Boston to Somerville was less than 10 km; by 1881 the link between Boston and Providence spanned 60 km, and by 1892 a call could be made from New York to Chicago. But the first very expensive transcontinental call (which took more than 20 minutes to arrange) could not be made until 1915, when New York was connected with San Francisco. Total number of American telephones rose from 48,000 in 1880 (or one phone for about every 10,000 people) to 134,000 by 1885 and to 1.35 million (or one phone for every 56 people) by 1900 (Garcke 1911b).

  FIGURE 5.13. Bell’s improved 1877 telephone and some of his demonstrations of the new device. Reproduced from Scientific American, October 6, 1877.

  Trials with direct dialing began as early as 1891, when an unlikely pioneer was a Kansas City undertaker Almon Strowger (U.S. Patent 447,918) who suspected that operators were switching his business to competitors. A much better automatic system introduced in 1901 produced “a peculiar humming sound” when the station called up was busy, and it also had an early version of rotary dial “provided at its circumference, opposite each figure, with an aperture into which the finger may be inserted” (Anonymous 1901:85). But none of these early automatic switching devices diffused widely before WWI, and it took even longer for the universal adoption of dial tone that was pioneered by Siemens in 1908. The first four decades of telephone service thus required large numbers of switchboard operators to run local, and later long-distance, service, and an effective solution of this major precondition of telephone’s diffusion relied on employing women.

  Growth of manual telephone exchanges provided many women from middle-class families with their first entry into the labor force. But even modest wages of telephone operators could not make the service cheap. By 1900, more than 20 years after the technique was invented, basic monthly telephone charge in major U.S. cities was between a quarter and a third of the average monthly wage. This meant that the overwhelming majority of customers were businesses and that, for most Americans, telegraph remained the dominant mode of intercity communication well into the first decades of the 20th century. And while Europe of the last two decades of the 19th century pioneered such critical inventions as steam turbines and internal combustion engines, its adoption of telephones lagged far behind the U.S. pace.

  The United States and Europe diverged early in their approach to the problems of capital investment, affordable rates, and reliable service. In the United States, Bell’s monopoly lasted until 1894 and allowed the company to charge high prices and make high profits; the subsequent increase in the number of independent telephone companies was only temporary, and virtually all of them were connected to Bell’s long-distance service, which was monopolized by the company’s new parent organization, American Telephone and Telegraph. Bell’s strategy of high long-distance charges used to subsidize low-cost local services lasted for most of the 20th century until the deregulation and breakup of the company on January 1, 1984.

  In the United Kingdom, small companies initially operated independently under the supervision of the British Post Office, and as a result the already noted contrast between the speed of diffusion of electric lighting in the United Kingdom and the United States applied to telephones as well. In Samuelson’s (1896:128) words, “the telephone has been coldly received in England” and compared to the United States the country “has lagged behind… to an extent which is almost ludicrous.” After the Post Office, which had the long-distance monopoly from 1896, took over all services in 1912, the United Kingdom had one phone for about 60 people, as did Germany, while the U.S. rate fell to just below 10 people/telephone; French ratio was still more than 150, and the Russian one just dipped below 1,000.

  The large investments of the monopolistic companies in installed equipment, which they expected to last for decades, was by far the most important reason for the slow progress of telephonic innovation during most of the 20th century. The first one-piece black set incorporating transmitter and receiver in the same unit was introduced only in the late 1920s. Subsequently, the telephone device stagnated—with the exception of pushbutton dialing introduced in 1963—until the beginning of a rapid conversion to electronic telephony begun in the late 1970s (Luff 1978). Only then was the carbon microphone replaced by a dynamic transducer (identical device acting as a receiver), and small integrated circuits eliminated bulky bell ringer, transformer, and oscillator. Phones got smaller, lighter, and cheaper, and soon the entire innovation cycle was repeated with wireless cellular devices.

  Recorded Sound

  The pedigree of Edison’s phonograph can be traced to both telegraph and telephone. While trying to improve his telegraph transmitter, Edison discovered that when the recorded telegraph tape was played at a high speed, the machine produced a noise that resembled spoken words (Josephson 1959). Could he then record and replay a telephone message by attaching a needle to a diaphragm of a receiver and producing a pricked paper tape? Soon he tried a tinfoil cylinder and was surprised as he heard himself reciting “Mary had a little lamb.” As this was the first device able to record and to reproduce sound, it attracted a great deal of public attention. In 1878 Edison took the phonograph on the tour that included a demonstration to U.S. President Ruther
ford Hayes, and set up a new company to sell the machine that he wanted to see in every American home (figure 5.14).

  During the early 1880s, when most of his attention was given to the development of new electric system, Edison looked for a substance to replace the foil as the recording medium—and so were other inventors. In 1885, Chichester Bell and Charles Tainter (1854–1940) got the patent for the Graphophone, a phonograph with wax-coated cylinders incised with vertical grooves, and by 1887 Edison patented his improved phonograph, which used wax cylinders as well as a battery-powered electric motor (figure 5.14). Both devices gave rise to a new recording industry, but neither was a great commercial success. Phonograph was marketed as a multifunctional device: as a recorder of family’s vocal mementos, a music box, a dictation machine for businesses, an audio text for the blind, and later also in a miniature form inside Edison’s talking dolls.

  The two main problems with the phonograph were the recording mode and the recording medium. A vertical (hill-and-dale) cut needed a mechanism that would prevent the stylus from jumping out of a groove, and the wax cylinder was obviously too soft and too fragile to make permanent recording; it broke easily with repeated handling, it needed a storage box, and it could not be reproduced cheaply in large quantities. But Edison—displaying his penchant for lost causes that led him later to invest years, and large sums of money, in fruitless efforts to beneficiate iron ore and mass produce concrete houses—kept improving the home phonograph for decades after its introduction, and the Edison company continued to make recorded cylinders until 1919. By that time the phonograph was a true antique as the era of gramophone was in full swing (literally, given the popularity of new jazz recordings).

  FIGURE 5.14. Edison with his first perfected hand-cranked tinfoil phonograph (reproduced from Scientific American, July 25, 1896) and his improved wax-cylinder, electric motor-powered phonograph (inset; reproduced from Scientific American, December 24, 1887).

  Origins of the gramophone can be traced directly to the failings of the phonograph: vertical-cut grooves, soft recording surface, and difficult mass production of recorded sound. All of these were successfully resolved in a relatively short time by Emile Berliner (1851–1929), a German immigrant who after a succession of jobs in New York became a cleanup man in the laboratory of Constantine Fahlberg (1850–1910), the discoverer of saccharine. There he became interested in experimenting with electricity, and in 1876 he invented a simple, loose-contact telephone transmitter (a type of microphone), which he patented in 1877 and which earned him a job with the Bell Telephone Co. of Boston (LOC 2002). Six years later Berliner returned set up a small research laboratory in his house in Washington, D.C., and just four years later he patented (U.S. Patent 372,786) a new system for recording sound with a lateral stylus and playing the records on a hand-cranked machine. Its first public presentation took place at the Franklin Institute on May 16, 1888 (Berliner 1888).

  Berliner’s solution was clearly anticipated by ideas about the “process of recording and reproducing audible phenomena” that Charles Cros (1842–1888) submitted to the French Academy of Sciences in 1877—but of whose existence Berliner, and even the U.S. patent examiners, were ignorant (Berliner 1888). Cros’s key idea, based in turn on Scott’s phonautograph, was to attach a light stylus to a vibrating membrane and use it to produce undulating tracings on a rotating disk—and then, by means of a photoengraving process, to convert the undulatory spiral into relief or intaglio lines and to produce a playable record. Nothing practical came out of Cros’s suggestions, but when Berliner eventually learned about them he very generously acknowledged their priority.

  Berliner’s first recordings were done by a stylus tracing a fine undulatory line in a thin layer of fatty ink, which he prepared by mixing one part of paraffin oil with 20 parts of gasoline and drying the deposit after the gasoline evaporated (Berliner 1888). A stylus attached to a mica diaphragm recorded the vibrations by moving laterally; the grooves were etched by acid, and a 15-cm zinc record, with 2-minute capacity, was placed on a hand-cranked turntable to be played by a steel stylus at 30 rpm. Multiple reproduction was done by electroplating the master zinc record and using the metal negative to stamp out the desired number of positive copies. After trying many different substances, Berliner chose celluloid, and soon after he switched to hardened rubber, which could withstand better the pressure exerted on steel needles by the cumbersome combination of the tone arm and horn.

  Berliner (1888:18) correctly anticipated that the records (he called them phonautograms) would provide both substantial royalties for performers and a great deal of enjoyment to their collectors:

  Prominent singers, speakers or performers may derive an income from royalties on the sale of their phonautograms, and valuable plates may be printed and registered to protect against unauthorized publication. Collections of phonautograms may become very valuable, and whole evenings will be spent at home going through a long list of interesting performances.

  Berliner’s U.S. Gramophone Co. sold 1,000 machines in 1894 and 25,000 15-cm hard rubber records. In 1895 Berliner patented (U.S. Patent 534,543) the recording on a horizontal disk (figure 5.15); in 1896 he discovered that shellac (an organic compound prepared from a gummy secretion of Asian scale insect Coccus lacca) from the Duranoid Co. was superior to hard rubber and switched to it in 1897, and the material remained dominant until the late 1940s, when it was replaced by vinyl.

  FIGURE 5.15. Emile Berliner’s drawing of the apparatus he invented (U.S. Patent 534,543 filed in 1892 and granted in February 1895) to produce gramophone records (lower right) and use of the instrument for recording speech (lower left) and an early design of a hand-cranked reproducing gramophone (top; reproduced from the cover of Scientific American, May 16, 1896). With Berliner’s design, it was necessary that “the record tablet be covered with a thin film of alcohol, and for this purpose a thin stream of alcohol (stored in the vessel on the right) is directed upon the center of the tablet… from which the alcohol spreads in all directions by centrifugal force…”

  Berliner’s key recording feature, whereby the groove both vibrates laterally and propels the stylus, remained the standard audio technique until stereo LPs, introduced during the late 1950s, combined the lateral and vertical cut. Soon after gramophone’s introduction, Berliner began improving the machine’s basic design, by replacing hand cranking with either windup spring or electric motors and by recording an increasing variety of performances. Rising sales brought a number of illegal competitors, and in 1900 Berliner turned over his U.S. patents to Eldridge R. Johnson of the National Gramophone Co. in Camden, New Jersey, at that time the main supplier of playback machines (LOC 2002). Johnson formed a new business that eventually changed its name to the Victor Talking Machine Co. and, after its 1929 merger with RCA, became the world’s largest and best-known recording enterprise, RCA Victor. Berliner relocated his company to Montreal in 1900.

  During the second year of his Montreal operation, Berliner sold 2 million single-sided records: the other side featured what was to become one of the world’s most famous trademarks, Francis Barraud’s painting of his dog, Nipper, listening to a gramophone (His Master’s Voice, source of the company abbreviation HMV), which Berliner registered on July 10, 1900. Victor’s two major milestones were the first recording that sold a million copies (the first Red Seal record with Enrico Caruso, on an exclusive contract, singing Vesti la giubba from Leoncavallo’s I Pagliacci) in 1903, and the introduction of the Victrola (in 1906), the first entirely enclosed cabinet phonograph, which despite its high cost ($200) became a big seller for the next two decades.

  The first double-sided disks were produced in 1902, but even so the playtime remained limited, with both sides containing no more than 10 minutes of music. Recording complete symphonies or operas was thus still impractical, but it was done anyway: HMV’s first complete opera, Verdi’s Ernani, filled 40 single-sided disks in 1904. But segments of classical compositions and key parts of famous operas became c
ommon, opening up entirely new audio experiences. Prior to that, even musical aficionados might have heard a favorite composition only once or twice in a live performance or had to be content just studying its score. More important, recordings introduced great music to millions of listeners who would could not make it to a concert hall. On the pop side, the earliest hits included folk songs, band music, and beginning in 1912, the craze for ballroom dancing led to a proliferation of dance band records (but the first Dixieland Jazz record, which sold 1.5 million copies, came in 1917).

  Before leaving the subject of recorded sound, I must note the genesis and delayed commercialization of magnetic recordings. Valdemar Poulsen (1869–1942), a Danish telephone engineer, patented his system of sound recording and reproduction on steel piano wire in 1898, and a working model of his Telegraphone was one of the electrical highlights at the Paris Exposition in 1900 (Daniel, Mee, and Clark 1999). An improved design, able to record 30 minutes of sound, became available in 1903, but better established and cheaper Edisonian phonographs and dictaphones delayed commercial development of the device, and the first Magnetophon recordings were made only in 1935.

  Hertzian Waves

  There were no fundamental obstacles to prevent the discovery of high-frequency electromagnetic waves years, even decades, before Hertz did so. Already in 1864, James Clerk Maxwell (1831–1879; figure 5.16) had presented to the Royal Society of London a dynamical theory of the electromagnetic field (Maxwell 1865), and his ideas generated great deal of interest, and disbelief, particularly after they were published in a two-volume Treatise on Electricity and Magnetism (Maxwell 1873). Implications of Maxwell’s theory were clear: the existence of waves of varying length that would propagate at the speed of light, curl around sharp edges, and be absorbed and reflected by conductors; moreover, it was known how to calculate their length and how to produce them (Lodge 1894). But Maxwell himself did not make any attempts to test the wavelike nature of electromagnetic radiation, and it took more than two decades after the publication of his paper before somebody investigated their existence.