Prologue

THIS book, although designed to be read independently, is in one sense a continuation of an earlier work, Syntony and Spark: The Origins of Radio, published in 1976. That book dealt with the very earliest phase of radio technology, when the scientific work of James Clerk Maxwell and Heinrich Hertz was being transformed into a technology of communication by men like Oliver Lodge and Guglielmo Marconi and the first attempts were being made to base commercial enterprises on that technology. The present volume picks up the story in the closing decades of the nineteenth century and carries it through the 1920s, when the advent of popular broadcasting transformed radio from a means of point-to-point communication, competing with the wired telegraph, into the agency of mass communication it is today. I discuss the origins of broadcasting only briefly in this book. My interest is in the origins of the technology that made broadcasting possible. This was the technology of the continuous wave.

In the earliest days of "signalling without wires" the only known method of generating radio waves was by means of sparks. An induction coil, or sometimes a bank of capacitors, was used to place a high voltage across a spark gap; when a spark jumped the gap it created an electromagnetic disturbance that could be detected at a distance. A series of sparks following each other in rapid succession gave rise to a chain of such disturbances — a radio wave, in short — that could be interrupted to form the dots and dashes of the Morse code and thereby convey information. Such waves travelled at a constant velocity: the speed of light. Each wave had a specific wavelength — the distance between succeeding peaks or troughs — usually measured in meters; and therefore, given the constant velocity, it had a specific frequency (the number of cycles per second). Each wave, that is to say, had a particular "place" on the electromagnetic spectrum, defined by its wavelength or frequency. If it was to be detected, the apparatus used for receiving had to be capable of responding to waves of that frequency — that is, it had to find and react to a signal at that "place" and, if possible, reject all others. Today we do this by a process we call tuning. In the earliest days of radio it was more common to speak of "syntony." Receiving and transmitting circuits were said to be in syntony when they resonated at the same frequency.

Syntony and spark were the characteristics that gave unity to that first phase of radio history. Technological development consisted of devising more effective spark transmitters, receivers that could detect and respond to spark-generated waves, and syntonic circuits that made it possible for transmitters and receivers to "find" each other in the radio spectrum. Important elements in this process were the development of antennas that could radiate and pick up signals efficiently, and the trial-and-error discovery of which wavelengths were most suitable for transmission over long distances.

The radio wave generated by a spark transmitter was a wave of a particular type. Each spark discharge generated a series of oscillations that diminished rapidly in amplitude as its energy was radiated into space and absorbed by the internal resistance of the components. A common simile, and an appropriate one, was to compare the antenna to a bell struck by a clapper. The bell, when struck, would sound a note, radiating energy in the form of sound waves. But the strength of the note would diminish more or less rapidly, as the vibrations of the bell diminished in amplitude. If the vibrations were "damped," as for instance if one placed a hand on the bell's surface, the sound would die away very quickly. So it was with a spark discharge: it had a degree of damping, depending on the internal resistance of the circuit and the rate at which it radiated energy into space. The radio wave generated by a succession of spark discharges consisted of a series of these damped oscillations. In that sense, a spark transmitter, although it might radiate continuously, did not generate a true continuous wave. (See Fig. 1.1)

It can be shown mathematically, by a technique known as Fourier analysis, that a damped oscillation such as that generated by a spark transmitter (or indeed any other complex waveform) can be decomposed into a large number of other oscillations, each with a frequency and wavelength of its own. These constituent oscillations are sine waves, in which the signal changes in an exactly prescribed way through a full cycle, going first positive, then negative, following the sine function in trigonometry. (See Fig. 1.2) This is no mere mathematical transformation: if such a train of damped oscillations were radiated from an antenna, its constituent sine waves would appear on the electromagnetic spectrum and (unless filtered out by tuned circuits) affect any receiver with enough sensitivity to detect them. From this fact certain important practical implications followed. A spark radio transmitter generated not one radio wave, but a very large number of them. Its signal was not at a single "place" on the electromagnetic spectrum but at a very large number of places. A true unmodulated continuous sine wave, in contrast, if one could have been generated, would have had one frequency only; it would have appeared at one place in the spectrum and only at that place.

This is the reason why, today, spark transmitters are universally outlawed. The radiofrequency spectrum is, to be sure, a unique resource, in that it can never be used up. But it can be overused; it can be overcrowded. Congestion of the radio spectrum creates a form of pollution in which transmitters interfere with each other and receivers are unable to select the signal that conveys the desired information from among interfering signals that are essentially noise. The danger always exists of a new "tragedy of the commons," in which overuse of a resource freely available to all creates a situation in which it is available to no one. To prevent such a situation, governments and international supervisory agencies resort to frequency allocation — essentially, the rationing of scarce space on the spectrum.

A spark transmitter is inevitably a dirty transmitter. It pollutes the spectrum by contaminating frequencies far removed from those nominally being used to carry the message. Its undesired effects can be minimized by reducing the degree of damping, which is to say by approximating more and more closely to a continuous wave. But there is a point, with spark, beyond which amelioration cannot go. Furthermore, spark transmissions make selective tuning much harder to achieve. The spark gap itself is a high-resistance element; its presence unavoidably lowers what engineers call the "Q" (quality) of the transmitting circuit and introduces a large damping coefficient. A spark-generated wave, therefore, is necessarily "broad." Quite apart from the harmonics it generates, it occupies an undesirably large space on the spectrum.

In the closing years of the nineteenth century and the first years of the twentieth, a few radio experimenters and scientists arrived at the conviction that spark radio transmission would have to be abandoned. They were, at first, a very small minority and they had difficulty making their case. On the one hand, the full potentials of spark transmitters had by no means been exhausted. Each year more powerful and more sophisticated spark transmitters appeared on the scene. Higher spark rates and lower damping coefficients approximated ever more closely to a continuous wave. Spark was a familiar and proven technology and there seemed no need to abandon it. On the other hand, devices and circuits that could generate true continuous waves were not at hand. A few experiments had been made with alternators, similar in principle to those that generated alternating current electricity for homes and factories, but they were low-frequency devices and delivered little output power. Some interesting work had been done with oscillating arcs that could be made to generate sound waves. Oscillating triode vacuum tubes, later to be the almost universal method of generating continuous waves, were unknown. A continuous wave transmitter that could generate substantial amounts of power at radio frequencies did not exist in 1900. In the circumstances, to believe that continuous wave radio could and should replace spark called for an act of faith.

Historians of technology have learned to recognize situations of this type and to attach special importance to them. Edward Constant, for example, writing about the introduction of the turbojet engine, points out that the men who, in the 1930s, built the first turbojets were not responding to any current failure of the conventional piston engine, which at that time had by no means reached the limits of its development. They were responding, rather, to insights that left them convinced that the conventional aircraft propulsion system would inevitably run into difficulties at some point in the future, and that a new and radically different system could and should be built. Such a situation sets the stage for discontinuous change, for what Constant, borrowing a term from Thomas Kuhn, calls a shift in technological paradigms. In Kuhn's model, designed originally to explain major discontinuities in the history of scientific theory, attempts to develop a new paradigm begin not when the explanatory power of its predecessor is fully exhausted but when anomalies begin to accumulate: phenomena that accepted theory cannot explain, or that it manages to explain only by successive ad hoc adjustments and extensions. Similarly, for Constant, attempts to develop a new technological paradigm begin not when conventionally accepted practice has failed in any absolute sense, but when a minority has become convinced that at some point in the future it will fail. At the time, however, there is still much potential for development in the conventional system; the anomaly is presumptive, not presently existing. It is visible to some but not to others. And those who see it and become convinced of its reality require, as part of their motivation, a kind of dedicated determination that to outsiders often seems unreasonable. To more commonsensical people, what they seek is probably unattainable and there is no real need for it anyway.

For Constant, the source of the insights that convince some individuals of the existence of a presumptive anomaly in technological practice is always science — in his case, advances in aerodynamics, the branch of physics that deals with gas flows. Presumptive anomaly occurs when scientific insight, or assumptions derived from science, indicate either that, under some future conditions, the conventional technological system will fail or will function badly, or that a radically different system will do a much better job or do something entirely novel. Science, according to Constant, provides the rational component that balances the nonrational "fanaticism" of the technological innovators — the "provocateurs," as he calls them.

In this book, although we shall use Kuhn's overworked term sparingly, we shall indeed be dealing with a major paradigm shift in radio technology — a shift that created the technical base for what we recognize as radio today and that was in its time as radical as the shift from piston engine to turbojet of which Constant writes. But we shall not take it for granted that what powered that shift were new insights derived from science, if by science is meant a body of articulated theory and a set of repeatable experimental observations. To assume that information generated by science is the only possible source for the detection of presumptive anomalies comes close to assuming that advances in science are the only possible source of major technological change. And that in turn comes near to defining technology as applied science — an identification that few scholars today are willing to make.

What role science played in the technological shift that is the central concern of this book is a question to be asked, not an assumption to be made. The men we shall be dealing with were not without scientific training; they had a keen sense of scientific literature and scientific personnel as resources on which they could draw; and they understood the importance of clothing their judgments in the prestigious language of science. But this is not to say that their commitment to the continuous wave was founded on scientific insight, nor that their perception of the presumptive anomaly facing spark technology was deduced from any body of scientific theory. On the contrary, if we are to judge by their own statements, their dissatisfaction with spark was based on more pragmatic grounds. The test of performance was whether or not radio communication could be reliably maintained over considerable distances. By this standard, spark was· in their judgment a poor prospect, primarily because, with its broad signal and multiple harmonics, a spark transmitter dissipated its power. Concentrate the available energy on a single frequency and your chances of achieving distance, of cutting through interference, fading, and atmospherics, were likely to be very much better. That kind of thinking called for no sophisticated knowledge of Fourier analysis. It was a strictly practical matter. Could spark do the job that its apologists claimed? Advocates of the continuous wave believed that it could not. There had to be a better way.

And there had to be a better way of detecting radio signals than by the device — the coherer — that typically accompanied spark transmission. Even the best of coherers was a temperamental device, hard to keep in adjustment; the need for "tapping-back" to restore sensitivity after a signal had been received, meant slow transmission speeds; and, since the coherer responded to voltage impulses, it was incapable of discriminating between signals and atmospheric noise. Escaping from spark technology called, therefore, not only for developing transmitters capable of generating continuous waves but also for finding receivers capable of detecting them. This was to require new circuits as well as new devices.

Those who decided to abandon spark and find a better way were responding not to new insights derived from science but to their sense that spark was a technological dead end and that continued reliance on spark would jeopardize radio's economic viability. The issue was whether radio could find for itself an economic niche in which it could grow and develop. Was this likely as long as spark reigned supreme? Some thought not. The presumptive anomaly that these individuals saw on the horizon appeared, not where technology and science met, but at the hazy boundary where radio stopped being a matter for visionary experimenters and started to become a hardheaded business capable of gaining and holding a commercial market. The criteria that the new technology would have to meet were economic criteria: could capital invested in radio earn the going rate of return? The expectation that spark radio would fail this market test was the rational ground for turning to the continuous wave.

Breaking away from spark, however, was not easy. Spark was the technology through which radio had come into existence; it had provided the only dramatic successes that the new means of communication could claim; and it was the technique to which the major operating organization of the day — the Marconi Company — seemed irrevocably committed. What was involved in the shift to the continuous wave was not just an incremental improvement in radio technique; it was a change in the way you thought about radio, the way you conceptualized it and visualized it. If you took it seriously it had some of the elements of religious conversion: it affected everything you did in the field thereafter. This had economic implications. To insist, as for example Reginald Fessenden did, that spark radio was following a fundamentally wrong track piled a new uncertainty on top of serious economic uncertainties already present. That did not make it easier to find and keep financial backers. And it had personal implications. Those who advocated such unconventional and visionary ways of thought and action paid a price in terms of the personalities they developed and the style of life they followed.

One element in the vision that these individuals followed was wireless telephony: the transmission by radio of voice and music. Without this element — that is, if radio communications had continued to be thought of exclusively as the dots and dashes of Morse code telegraphy — it is questionable whether continuous wave radio would have seemed a gamble worth taking. Without exception, the early devotees of continuous wave radio had in mind the transmission of the human voice and not merely a marginal improvement in radio telegraphy. This had implications for the standards of performance that continuous wave radio was expected to meet. A slight competitive advantage over spark telegraphy was not enough to attract venture capital: the lure was wireless telephony — initially not for broadcast entertainment but to provide the kind of point-to-point communication that the American Telephone and Telegraph Company provided, only without the fixed costs of a wired network.