Response to Dr Brian N. Hall’s Articles on British Wireless in the First World War

I. Introduction

Dr Brian N. Hall’s recent article in the Journal of Military History ¹ cavalierly dismisses a recent book ² on First World War British wireless communications as offering ‘very little in the way of fresh information’. Pointing to the authors’ presumed lack of scientific and technical understanding, he further states that

one recent study [the book] has argued that it was the ‘significant institutional bias’ of the British High Command against wireless communication that prevented it being employed both earlier and to its fullest extent. Such a conclusion not only underestimates the scientific and technical limitations of the wireless sets of the era for their successful employment on the Western Front but, more specifically, fails to take into account the British Army’s global communications experience. ³

The authors of the book do understand the ‘scientific and technical limitations of the wireless sets of the era’ only too well. The authors have academic and professional backgrounds in both electrical engineering and military history. The book’s point is that it would have been possible to overcome these limitations earlier in the First World War, but the British high command failed to do so. In contrast, as soon as it became a belligerent the US Army equipped its forces with modern wireless communications, as the Royal Navy and the United States Navy had done just prior to the war. This argument will be further developed in the body of this article.

The Journal of Military History article referenced another article by Dr Hall in War in History which similarly disparaged the book. With regard to the costly first day of the battle of the Somme, he stated in that article, ‘Contrary to the claims of one recent work [the referenced book], it was a situation that wireless could do little to alleviate.’ ⁴ This comment reveals little comprehension of what modern wireless communication could have provided or of the process of technical development, as will be explained below.

The book indeed focused on the institutional failure of the British army to adopt modern wireless communications earlier in the war. The key phrase is modern, i.e. vacuum tube and continuous wave (CW), communications. The British army fought the war with a spark wireless communications system that was obsolescent after 1916. As pointed out in the book, the British High Command (British Expeditionary Force as well as London) did not fully appreciate the need to implement a clear, consistent policy on communications technology and development until 1917. ⁵ Dr Hall disagrees with the book’s institutional failure assertion. This article was written to answer Dr Hall, of course, but also to marshal new evidence to buttress the book’s conclusion that the British army failed to modernize its communication system when it could have. The book and this article do not argue that the British could have implemented modern wireless earlier than 1916. The technology to do so was not available before then, but was maturing rapidly. As shown in the book and described in this article, the Royal Flying Corps (RFC) built and demonstrated an airborne radiotelephone (using CW and vacuum tubes) to the British high command in February 1916. This achievement was not fully appreciated and forcefully exploited. As shown in the notes to Table 1 below, an airborne radiotelephone was not deployed until 1918. The first radiotelegraph sets using the new technology were not delivered to the ground forces until late 1917.

OPEN IN VIEWER

Table 1.

Quantities of continuous wave radio equipment types produced and delivered.

	Americans	British
Aircraft radio telephone set	9,659 (SCR-68, 59) a	621 (Mk II) b
Ground radio telephone set	527 (SCR-67) a	0 (527 in 1919) c
Ground radio telegraph receiving set	8,052 (SCR-77, 78) d	2,949 (Mk I, II, III) e
Ground radio telegraph transmitting set	2,637 (SCR-77, 78) d	3,152 (Mk I, II, III) e
Total number of sets	20,875	6,722

a

Col. G.W. Mixter and Lieut. H.H. Emmons, United States Army Aircraft Production Facts Compiled at the Request of the Assistant Secretary of War (Washington, DC, 1919), p. 91.

b

Louis Meulstee, Wireless for the Warrior, Compendium 1, Spark to Larkspur (Groenlo, Holland, 2009). This was not delivered until 1918 (ibid., p. 101).

c

Ibid., p. 99.

d

Benedict Crowell, Assistant Secretary of War, America’s Munitions (Washington, DC, 1919), p. 583.

e

Meulstee, Wireless for the Warrior, pp. 43–7. Mark I (170 produced) and Mark II (129 produced) of this set were capable of transmitting and receiving. The Mark III transmitter and receiver were produced as separate sets. Thus 299 sets were added to the totals for both the Mark III transmitter (2,853 produced) and receiver (2,650 produced).

This article does not deviate from the book’s position. Factors are mentioned either in the book or this article, or in both, that soften the book’s harsh conclusion, such as technical difficulties with vacuum tube development and production, fear of enemy interception, lack of recognition that poor communications led to high casualties, and the unanticipated scale of industrialized warfare between well-armed and technologically developed nations. Both this article and the book state that these factors offer a partial explanation for the failure, but do not justify the long delay.

The authors’ issue with Dr Hall is not with the quality or depth of his research into the deployment of wireless by the British army. They do not doubt his amply documented conclusion in both of his articles that the British army Signal Service made the best use of the wireless equipment issued to it. The book also supports and amplifies this point. In the book, however, the authors state that the high command of the army did not have the vision nor did it direct the application of the resources needed to equip the field forces with the best wireless technology available at the time. That it could have done so is convincingly demonstrated by the contemporaneous success in the same endeavour achieved by the US Army, even when allowances are made for the different manufacturing capabilities available in Britain at that time. The numbers are telling. By November 1918, in 18 months of war, the Americans produced and were delivering to their forces 20,875 modern CW radiotelephone and radiotelegraph sets. In the last 30 months of war, despite a demonstration of this technology to the high command early in 1916, the British produced and delivered less than a third of that number, 6,722 modern sets. Table 1 summarizes the quantities and types of sets delivered.

The article provides a brief technical history of wireless and its military application, describes the slow British army development of modern CW wireless, and summarizes the Royal Navy’s and the United States Navy’s success with wireless. Then it presents the contrasting rapid US Army development and deployment of CW wireless and discusses the American general who led that effort. It also discusses the challenges the new wireless sets presented to British and American manufacturers. Finally, it presents a conclusion which affirms the judgement of the referenced book and a rebuttal of Dr Hall, and questions the scholarship that ignored evidence which contradicted his argument.

II. Background

The reason Dr Hall’s statements cannot be allowed to stand unchallenged is that, perhaps lacking a technical background, he does not recognize that between 1910 and 1920 a major change occurred in wireless technology, from spark to continuous wave. This was ‘a major paradigm shift in radio technology – a shift that created the technical base for what we recognize as radio today’. ⁶ A technical paradigm shift is the term Edward Constant, writing about the replacement of the aircraft piston engine by the turbojet engine, used for a discontinuous change in technology. ⁷ The point of the book is that despite the lessons of war and ready access to the new CW technology by 1916, the British failed to exploit it, and did not employ CW technology on a major scale until mid- to late 1918. To underline this argument this article demonstrates that the Americans, once they became a belligerent, and in the same time frame, did exploit the new technology and produced large quantities of modern wireless sets for military communications.

To assist the reader in understanding this article’s conclusions, a brief history of the development of wireless communications and its application to the conditions of the First World War follows. Further technical details are provided in the appendix to the referenced book. ⁸

In 1873 the Scottish physicist James Clerk Maxwell published four equations which combined all that was then known about electricity and magnetism. Mathematical manipulations of Maxwell’s equations produce the wave equation, which describes the propagation of electromagnetic waves. Therefore in 1873 Maxwell predicted that electromagnetic waves moving at the velocity of light could be produced by time-varying electric currents in conductors such as wires or surfaces, a remarkable result.

Experimental verification of Maxwell’s equations did not occur until 1887. The German physicist Heinrich Hertz designed an experimental apparatus which would generate an electromagnetic wave and a device which would detect it. He generated the wave by creating a spark between two conducting hemispheres and initially detected the wave by seeing a spark in the gap between the two ends of an interrupted loop of wire placed in the vicinity of the two hemispheres. ⁹

A young Italian, Guglielmo Marconi, and others saw the potential of using electromagnetic waves to communicate. Following Hertz, he used a spark between two conductors to generate the waves. He recognized, however, that the waves would be generated more efficiently if the energy from the spark were fed into vertical or horizontal wires of varying lengths and of varying configurations, called antennas. He and others also felt that a detector used later by Hertz, a glass tube filled with fine iron filings, which would align and conduct in the presence of a magnetic field, needed improvement. After trial and error by him and many others, notably Edouard Branly and Oliver Lodge, an improved detector, called a coherer, was developed. The spark transmitter and the coherer were the basis of the wireless telegraphy (W/T) system Marconi first demonstrated to the British government in 1896 and eventually developed into a hugely successful commercial enterprise. Marconi’s W/T system, with engineering improvements – such as the use of a transformer to couple spark energy to the antenna (which fortuitously enabled crude tuning of the noisy broadband spark transmission, a phenomenon called ‘syntony’, originally discovered by Oliver Lodge ¹⁰ ), and magnetic or crystal detectors to replace the coherer – dominated world wireless communications for approximately 20 years, from 1900 to 1920. (A key element of Marconi’s dominance lay in his company’s ownership of British patent no. 7777, issued on 26 April 1900, which allowed the company to exploit the syntony property. ¹¹ )

Most belligerent armies in the First World War used spark technology for almost the entire war. The communications system was capable of W/T only, using Morse code, no voice. The system employed inefficient spark transmitters and receivers which created a spark by interrupting the flow of current in a coil, just as an ignition coil works in a car. The sets were heavy and bulky, because spark transmission is inefficient, radiating only a small fraction of the power supplied to the transmitter, though later on the sets were somewhat reduced in weight and size. (Inefficiency meant that large and heavy batteries were needed. Batteries were the heaviest components of the sets and limited their portability.) Most importantly the system was subject to the crippling problem of interference. Signals transmitted by spark transmitters are damped sine waves, which are spread over many frequencies, causing severe mutual interference between adjacent transmitters. This is a fundamental and inescapable problem for spark transmission. It is the reason spark transmitters have been illegal throughout the world since the late 1920s.

In a military environment, mutual interference limits the number of available communication channels, a major drawback. During the war, with spark technology, multiple transmitter/receiver pairs interfered with each other if they were less than 3,000 yards apart, despite using syntony to reduce interference. ¹² This was a severe restriction in an environment that requires significant communication capacity. For example, a division on the Western Front occupying 1,000–3,000 yards of front (not atypical) had only one or two interference-free wireless communications channels available for the entire division, if that.

The solution was the use of CW transmission, which supports a large number of non-interfering channels on multiple frequencies (think of an AM or FM radio). This was known to radio engineers well before 1914, but could not be implemented until a reliable CW source, light and compact enough to be portable, was available.

Before the war it became clear that the three-element vacuum tube or triode was the answer. CW radio requires a single-frequency tunable oscillator, a modulator (to add information) and an amplifier. The vacuum tube performs all three of these functions in a light, compact package. Dr Ambrose Fleming, of Marconi, patented a two-element vacuum tube, or diode, in 1904, calling it a ‘valve’. ¹³ (Britain continued to use this nomenclature for all vacuum tubes in the twentieth century.) Building on Fleming’s diode, Lee De Forest patented a three-element triode in 1906, calling it the ‘audion’. ¹⁴ (America did not continue to use this name.) Using the triode, Edwin Howard Armstrong (who received a commission in the US Army during the war) invented and patented the oscillator, modulator, and regenerative amplifier in 1912. ¹⁵ By 1916 vacuum tubes were in mass production for all of the belligerents, but were primarily used for detection and interception of enemy spark signals, since they were more sensitive than magnetic detectors or other devices. This capability also made them useful as detectors for friendly spark or CW communications systems, but widespread employment of them for CW communications depended on the development of reliable high-power transmission tubes, as elaborated upon in the next section.

III. British Army Development of CW Wireless

In 1915 a CW airborne radiotelephone was developed by a group led by Major Charles Edmond Prince, a senior Marconi engineer seconded to the RFC. ¹⁶ By early 1916 the radiotelephone was operational and was demonstrated to Lord Kitchener in February of that year. ¹⁷

Wireless was recognized early as the best means to communicate with aircraft. ¹⁸ In addition to Prince, Major Robert Orme, Captain Harold J. Round, and other leading Marconi engineers were commissioned in the RFC. They were directed to build a radiotelephone for aircraft. ¹⁹ In the summer of 1915 Prince and the RFC engineers built a vacuum tube airborne transmitter that enabled wireless speech to be received from an aircraft. The first practical set evolved was a transmitter capable of employing either speech or CW telegraphy. ²⁰ In February 1916 wireless telephony from an aircraft to the ground was demonstrated to Lord Kitchener at a distance of 40 miles. ²¹

In modern terminology Prince and his group had accomplished ‘proof of concept’, which means that they had proved that it was possible for an airborne transmitter to communicate speech to the ground and had developed a ‘prototype’ or first device capable of doing so. The next step in a development process would have been to produce a number of similar devices to accomplish this same end and submit them for testing under a variety of conditions. It would have been appropriate to test them in the field, as well as in more controlled environments. The results of this testing would have enabled the engineers to improve the initial design, such as enhancing speech intelligibility, extending the range of transmission, reducing power requirements, increasing durability and ruggedness for use in the field, and making the transmitter easier to operate.

At the same time that the airborne transmitter was being evaluated, the possibility of using it as part of a ground communication system could also have been explored. Since it was capable of transmitting either voice or Morse code, developing a dual-use transmitter could have been considered. Also, for the ground application, a transmitter-power versus antenna-length design trade-off would have been carefully evaluated. The military advantages of a shorter antenna are less visibility to the enemy and less vulnerability to enemy artillery.

Another advantage of submitting the transmitter and receiver to the field for initial testing would have been that the field units could employ them under actual military conditions, and determine what the strengths and weaknesses of the units were for military operations. In so doing the field units would also be training themselves in the use of wireless telephones and CW telegraph sets, determining how best to use the new technology. An added benefit would have been that the field units could have employed these pre-production units at the Somme (as was suggested in the book, which prompted Dr Hall’s negative comment).

Once this period of controlled and field testing was completed, the engineers working on the project would have absorbed the considerable amount of data collected and would have proceeded to refine the design and produce a truly ‘militarized’ radio suitable for installation in observation aircraft and for use on the ground. By the end of 1916 Prince and his team of engineers could have designed and placed into mass production a reliable lightweight radiotelephone and CW telegraph set suitable for use not only in the air, but on the ground as well. ²²

This is not a far-fetched sequence of events or time frame for development in wartime. Such a process was followed by the British in 1915 in the design and production of tanks ²³ by visionaries in the British army such as Colonel Ernest Swinton. ²⁴ Swinton persuaded influential individuals such as Maurice Hankey and Winston Churchill to back this idea. ²⁵ A demonstration in June 1915 showed that such a vehicle could cut through barbed wire and advance through no man’s land. Swinton oversaw the development of the first combat tank in January 1916. In other words, the tank proceeded from proof of concept to the delivery of a prototype in six months. The Mark I tank, as this prototype was called, ²⁶ was placed in mass production and entered combat in September 1916 at Delville Wood on the Somme. A second-generation production tank, the Mark IV, appeared early in 1917. ²⁷

The development and application of CW wireless sets did not proceed on this fast track, or indeed on a slow track. Cost was not a deterrent. The costs associated with the development of more capable wireless would have been a fraction of the resources spent on the development of the tank (or combat aircraft). The referenced book offered a number of suggestions as to why this happened. ²⁸ Much recent work, quoted by Dr Hall ²⁹ and the referenced book, ³⁰ has identified poor communications as a key cause of battlefield failure and high casualties. It was not so clear in 1915. It is still difficult to understand why the army did not recognize at the time that poor communications were the cause of so many battlefield failures and therefore make every effort to correct the problem.

From today’s perspective, CW wireless was the obvious answer, but the army’s own experience with spark wireless had been disappointing – barely portable sets, limited communications capacity because of mutual interference, telegraphy only, no voice. The junior officer engineers of the RFC had an understanding of what CW was capable of providing, but, as pointed out in the book, the Royal Engineers did not. ³¹ The ‘institutional bias’ identified in the book refers to the inability of the army to recognize that CW wireless would substantially improve communications. No one in a position of authority understood the promise of CW and directed that resources be applied to develop it. Improving the difficult communications between commanders and subordinates, between infantry units, between infantry and artillery, and between aircraft and infantry or artillery was not a high priority at the time. The invention that met an obvious need (such as the tank) was backed more readily than an equally promising development that required a degree of subtle understanding.

IV. The Royal Navy

Unlike the British army, the Royal Navy adopted wireless early and exploited it successfully. Like aircraft, there was literally no other way for ships out of sight of each other to communicate, or for a ship at sea to communicate with land. The Royal Navy began its investigations of wireless at the same time as Marconi, but independent of him. Its leading proponent was Henry Bradwardine Jackson, a man of extraordinary brilliance and vision, who later became an admiral of the fleet and was first sea lord for the first two years of the First World War. In 1887 he observed that Hertzian waves (as electromagnetic waves were then called) might be used to help defend against enemy torpedo boats. In 1895, when he was named as commanding officer of HMS Defiance, he had an opportunity to acquire apparatus to conduct experiments with Hertzian waves. With this apparatus, in August 1896, he succeeded in transmitting and receiving Morse code signals at a range of 50 m, the maximum available from the ship. At the end of August he also met Marconi. He learned that ‘the principles on which Signor Marconi’s apparatus were constructed were similar to those employed by Defiance, but [were] more fully developed and the instruments themselves were much more sensitive’. ³² Jackson also informed Marconi that he had been working on similar apparatus but put Marconi’s mind at ease by telling him that he had no intention of patenting his work. ³³

Jackson became a firm champion of Marconi’s invention within the navy. After trials of Marconi’s equipment were conducted on Salisbury Plain, Jackson sent Marconi a general specification for a ship-borne set. ³⁴ After receiving equipment developed according to his specifications, Jackson organized a series of trials using torpedo boats and HMS Defiance. Successful transmission of signals over a range of 3 miles was achieved. Jackson also determined that the wireless apparatus was rugged enough for sea duty and did not require any more attention than other electrical devices on board. He did, however, identify the problem of interference, noting that it was impossible for two ships to transmit simultaneously.

By 1910 technical improvements in wireless such as the quenched spark ³⁵ and Poulsen arc ³⁶ transmitters and Marconi’s crystal detector ³⁷ were installed in naval vessels as they became available. The Royal Navy also exercised the wireless system extensively during the annual naval manoeuvres, learning how to exploit the new communication capability and to identify its shortcomings. By keeping up with technical progress and by constantly exercising its capability, at the start of the First World War the Royal Navy had a wireless communication system whose equipment, organization, and training were superior to those of any other nation.

V. The United States Navy

The US Navy also understood the great potential of wireless communications, but was slower to implement it than the Royal Navy. As described in L.S. Howeth’s comprehensive history of naval communications, ³⁸ the US Navy attempted to purchase Marconi (spark) equipment in 1899, ³⁹ but was unable to do so because Marconi would only lease the equipment, a provision unacceptable to the Navy. In 1902 the Navy purchased (spark) wireless equipment from the German firm Allgemeine Elektricitäts-Gesellschaft (Slaby-Arco) for testing. ⁴⁰ Successful tests of this equipment led to the purchase of additional equipment from the same firm, despite vigorous protests from American manufacturers, in particular the scientist and inventor Reginald A. Fessenden of the National Electric Signaling Co. ⁴¹

In the first decade of the twentieth century the American radio industry was in turmoil, and the Navy encountered much difficulty in equipping its ships and shore stations with reliable (mostly spark) equipment at reasonable cost. Besides Fessenden and his company the Navy had to deal with fractious individuals and competing (sometimes unscrupulous) companies, such as the American Marconi Co., Lee De Forest and his American De Forest Wireless Telegraph Company, and John S. Stone and his Stone Telephone & Telegraph Co. ⁴² In 1907, for example, the Navy installed De Forest radiotelephone equipment on ships of the ‘Great White Fleet’ prior to their departure on their ‘Around the World Cruise’. The installations had to be dismantled ‘because they were being used improperly and because they interfered with normal radio communications’. ⁴³

The Navy appreciated the benefit of CW communication early and was very interested in the Poulsen arc, a source of continuous waves. In 1912 the Navy’s procurement of this equipment enjoyed better success than the above procurements in their dealings with Cyril Elwell and the Federal Telegraph Company, which held the US patents for the Poulsen arc. The company’s engineering improvements to the system increased its power, provided frequency stability, and enhanced its reliability. The unit was the ‘Navy’s darling of the World War I period. The Federal Telegraph Co. soon became a Navy-subsidized firm and was purchased by the Government during World War I.’ ⁴⁴ In its procurement of this equipment, the Navy showed it understood the benefit of CW wave communications. Justifying its selection of the arc-based system over competing systems, it noted that:

There being two fundamentally different systems of radio involved, namely, the spark system or system of damped oscillations and the system of continuous undamped oscillations, sometimes called the Poulsen system, there was a matter of choice. All the other so-called systems bearing proper names such as Marconi, Shoemaker, Fessenden, De Forest, Telefunken, etc., were but variations of the spark system and did not represent real differences so far as fundamental classification was concerned. ⁴⁵

When the country entered the First World War:

The improvements which had been made in the United States to the Poulsen arc transmitter made available a fairly satisfactory continuous wave transmitter. Although at the beginning of this period it was of relatively low power it was capable, with heterodyne ⁴⁶ reception, of reliably covering far longer distances than could be accomplished by the old spark transmitter despite all the efforts to improve its capabilities. ⁴⁷

Once the United States entered the war, the Navy had a requirement to provide communications for aircraft, in particular, flying boats needed to patrol for submarines, search for mines, find vessels in distress, and provide convoy escort. ⁴⁸ This requirement was different from that of a land-based aircraft system. Typically the aircraft were larger, and the radio sets were larger and had longer range. In addition a specially trained operator was needed because of the many duties he was required to perform. ‘The problem faced was the need to build large quantities of equipment which would be simple and inexpensive to manufacture, small, light weight and possessing extreme ruggedness and simplicity of control.’ ⁴⁹ Because of its weight and power requirements, the Poulsen arc was unsuitable for aircraft, so the Navy had to use vacuum tube equipment. An example of the equipment produced was the set designed for the large flying boat, the SE 1100, a 200 W transmitter designed and manufactured by General Electric (GE). ⁵⁰ Installed, including all components and a receiver, this equipment weighed approximately 210 lb. CW interrupted by a key provided telegraph transmission. The CG 1130 had a telephone range of 200 miles and a telegraph range of approximately twice that distance.

VI. The Successful US Army Development of CW Wireless

In contrast to the British army, upon entering the war the US Army enthusiastically exploited CW technology, as described in the US Army Signal Corps Official History, ⁵¹ a 1976 article, ⁵² a 1974 dissertation, ⁵³ and a newly published biography. ⁵⁴ As shown by these works, even without the lessons of war, the Americans quickly developed a family of radiotelephones and radiotelegraphs based on CW technology for aircraft and ground use.

The development of the airborne radiotelephone was the key to American success in the development of US Army tactical communications, for both ground and air applications. For three years the Europeans had employed vacuum tubes to intercept enemy W/T communications, and much later a few of their own W/T wireless sets, but had not applied the vacuum tube to radiotelephony. ⁵⁵ The British did not appreciate the significance of Prince’s work, and did not share it with their allies until August 1918. ⁵⁶ The French had tube circuits in abundance but no radiotelephone or any plans for voice communications by radio. Worse, the French had only to insert a microphone into the input circuit of their W/T vacuum tube sets to achieve radiotelephony. ⁵⁷

American interest in airborne radiotelephony had begun in 1910 at Belmont Park in New York and continued in further experiments at San Diego. ⁵⁸ The vision of voice command of large formations of aircraft was advanced by the Chief Signal Officer (CSO) George Owen Squier and other technical officers, despite the scepticism of many flying officers. ⁵⁹ Squier had the authority, technical background, and access to resources, including close ties to the premier American communications-engineering organizations and their most capable engineers, to succeed in the project.

Shortly after the American declaration of war Squier commissioned a number of radio engineers directly from the radio industry to make up a radio section within the Chief Signal Office. He also invited the French and the British to send scientific missions. The French delegation, composed of prominent electrical engineers, arrived with samples of their most recent signalling equipment. On 22 May 1917 Squier convened an important conference in his office to begin the development of radiotelephones for aviation. Attendees included Allied representatives, Captain Charles C. Culver, the officer who had conducted the San Diego experiments, and representatives from American industry. ⁶⁰

Among the many difficult problems involved were the development of an efficient antenna which would not hinder manoeuvring, a microphone system sensitive to voice in the presence of engine and wind noise, a soundproof helmet, light and rugged apparatus requiring minimum adjustment to operate, a wind-driven power supply, and small, rugged vacuum tubes suitable for mass production. Of all these the vacuum tube presented the greatest challenge. These would be essential for radiotelephone sets for aircraft as well as every other piece of US Army Signal Corps communication equipment. They would be required in quantities far larger than ever before attempted. Vacuum tube production would have to be increased from the current 300–400 tubes per week to over 40,000 per week. Two standard types of tubes, receiving and transmitting, were to be developed and employed in a variety of circuit designs, operating with each other, and suited for performing several circuit functions. Western Electric (WE) undertook the main responsibility for developing a prototype of the aircraft radiotelephone. In June 1917 the first tests of workable airborne radiotelephone were conducted at Langley Field, Virginia. By early July successful voice communications took place between two planes transmitting over a distance of about 3 miles, ground to plane, and plane to ground. By October a set was produced which was considered to be rugged enough for combat use.

After the tests of experimental apparatus, the equipment had to be standardized and mass-produced. By December 1917 the receiving vacuum tube, named the VT-1, was placed into mass production. Despite manufacturing tolerances of 3/100 of an inch and the need for a high vacuum, it was rugged and easily replaceable, since all VT-1s had identical electrical parameters. The development of a sending tube presented greater difficulties because of the requirement for high operating power. The first sending tubes, called the VT-2, produced high-frequency signals as high as 5 W. Further development produced the VT-18, which could produce an output of 30 W. To mass-produce these tubes, physical sizes, shapes, testing equipment, and procedures had to be standardized. Powering the radiotelephone was the last challenge. The solution was to design a dynamo attached to the fuselage of the aircraft and driven by a propeller placed in the airstream. The Signal Corps official history states that, ‘Despite the conscientious efforts by government and industry, the limited duration of America’s involvement in the war left little time for the development and application of new technology … While some aerial radiotelephone apparatus arrived in France by the fall of 1918, it did not see use in combat.’ ⁶¹ It is certainly true that the full weight of American radiotelephone production had not been felt at the front nor deployed to France at the time of the armistice, but some sets were being flown on front-line aircraft. Initial production models of the SCR-68 (see below) installed on DH-4s tested at Orly airfield suffered from the usual teething problems associated with new equipment, such as breakage in shipment, poor isolation of send/receive circuits, bad contacts, and cold solder joints. So great, however, was the need for the early introduction of airborne radio that SCR-68s were installed on combat aircraft of the 92nd Aero Squadron for further testing. ⁶²

Through Squier’s leadership a new tool of warfare had been developed. He had envisioned the need for command and control of massed aircraft, and through airborne radiotelephony had provided the means to provide it. As significant as this achievement was, it was overshadowed by what followed. Prior to the development of the airborne radiotelephone Squier had anticipated that the technology developed to support it would be equally applicable to ground warfare. For example, in setting out the requirement to produce a million vacuum tubes a year he was anticipating the development of a whole new family of tactical radios for the infantry, artillery, and tank combat arms. He had insisted on ruggedness, light weight, and standardization of parts (to ease repair and replacement) for the airborne radios; he knew these same qualities were what were needed for the demands of ground warfare. In this he was implementing what the British had failed to do when they had built a radiotelephone. Not only had they waited three years to install it in aircraft, but they had failed to appreciate until much later what the technology could provide for the ground forces. Squier understood that the new radios would enable infantry units to readily communicate with their commanders, with each other, with their supporting artillery, tanks, and aircraft. He had anticipated the communication needs of combined arms warfare before Heinz Guderian (who was a signals officer) conceived it and experimented in the Reichswehr’s manoeuvres in the 1920s. ⁶³

Squier made it clear that the technology of airborne telephony was the basis for all US tactical radio. The Annual Report of the CSO for 1919 ⁶⁴ stated:

The development of radio apparatus for telephonic communication from plane to plane and from plane to ground constitutes one of the most interesting and involved problems in the Radio Development Section. … There evolved from all this the following standard Signal Corps Radio (SCR) equipment: the SCR-57 and 57A, SCR-59, SCR-67 and 67A, SCR-68 and 68A, SCR-69, SCR-75, SCR-91 and SCR-109. ⁶⁵

With the assistance of Mr Carl Bobrow of the Smithsonian Institution’s staff, the authors were able to locate digitized versions of the original pamphlets produced by the Signal Corps for each of these radios. ⁶⁶ The pamphlets contain pictures of each radio, dimensions, circuit schematics, parts lists, front and rear panel connections, operating parameters such as power requirements, frequencies, and equipment used with it, such as earphones, microphones, power supplies, and antennas. These pamphlets provide a detailed understanding of how these radios functioned by themselves and as part of an integrated communications system. The whole structure – the design and manufacture of the radios to high standards (what are now called militarized standards), close attention to how they interfaced with each other, and detailed documentation designed to train personnel in their use and maintenance – reflects a sophisticated understanding of communication system design that would do credit to a twenty-first-century engineering organization.

As the 1919 report states, the most significant of the new radios, from which all other sets were derived, were the SCR-68, 68A (an improved version of the 68), and 59 Aircraft Radiotelephone Sets. ⁶⁷ Signal Corps Pamphlet 20 describes them as follows:

The SCR-68 and SCR-68A sets are airplane radio telephone transmitting and receiving sets designed primarily for inter-plane communication work between airplanes in squadron formation. These sets are used by the commanders of squadrons. The other planes of the squadron are usually equipped with the type SCR-59 receiving set. The set is also used for two-way communication with ground stations equipped with SCR-67 or SCR-67A sets. ⁶⁸

The SCR-78 Tank Radiotelegraph Set was designed as a very rugged set to be installed in a tank. Like the airborne radiotelephone, it had to operate on a platform that would subject it to intense vibration. The decision was made to install a radiotelegraph instead of a radiotelephone in the tank for several reasons: the noise level inside a tank is as high as, if not higher than, inside an aircraft, making it difficult to hear or speak; the security benefit is that a telegraph signal could be encrypted to hinder interception, while voice could not.

The SCR-77 ‘Loop’ CW Radiotelegraph Set was developed to meet the requirement for forward infantry units to communicate. The experience of the ‘Lost Battalion’, dependent on pigeons to call off friendly fire, exemplified the requirement. Again, a radiotelegraph instead of a radiotelephone was selected, for reasons of security.

Under Squier’s leadership, at the time of the armistice the US Army was ready to field CW airborne radiotelephone sets and several squadrons of aircraft and trained personnel to use them. ⁶⁹ Not only was Army aviation to be provided with modern communications and trained personnel, but the ground forces, infantry, artillery, and tank units, would also receive modern communications equipment derived from the sets designed for aircraft.

Admiral Jackson played a key role in championing wireless for the Royal Navy; Colonel Swinton and First Lord of the Admiralty Churchill were the champions for the tank. British army wireless did not have a similar champion who understood the technology, had the vision to see its importance, and possessed the power and position to carry it forward. The Americans did have such a champion, in the person of Major General Squier. It is appropriate to learn more about him.

VII. Major General George Owen Squier

Major General George O. Squier held a PhD in electrical engineering (from Johns Hopkins University), one of the first awarded in the United States. An excellent description of Squier’s life and accomplishments is contained in Paul Wilson Clark’s unpublished dissertation. ⁷⁰ Charles Gross provided additional biographical material in his paper on Squier’s contribution to military aviation in the First World War. ⁷¹ The newly published biography of General Squier provides further details of his life and work. ⁷²

Squier was born on 21 March 1865 in Dryden, Michigan. He won a competitive appointment to the United States Military Academy in 1883. An excellent student, he graduated seventh in his class and was assigned to a coast artillery unit at Fort McHenry in Baltimore, Maryland. He took advantage of the opportunity of being in Baltimore by enrolling in a doctoral programme at Johns Hopkins University. He was awarded a PhD in electrical engineering in 1893, the first US Army officer to hold a doctorate. ⁷³

In 1898 Squier transferred to the Signal Corps and continued his research. In 1905 CSO General Adolphus W. Greely placed him in charge of establishing the new Army Signal School at Fort Leavenworth, Kansas. Squier also helped prepare the specifications for the first American military aircraft from the Wright brothers ⁷⁴ and himself set an endurance record on a 10-minute flight with Orville in 1908. ⁷⁵

Squier returned to communications-related research in 1909, establishing a new research laboratory at the National Bureau of Standards in Washington, DC. It was his idea to apply the techniques of radio communications to a wired circuit. Radio communications use carrier multiplexing, which separates communication channels by assigning them to different frequencies, thereby increasing the amount of information that can pass from transmitter to receiver without interference. Squier saw that the high frequencies generated by the recently developed Alexanderson alternator could also be used as carrier frequencies on wired circuits. ⁷⁶

In 1912 Squier was appointed American military attaché in London. It appears that General Leonard Wood, Army Chief of Staff personally intervened to send an officer with the background and education needed to understand the complexities of modern warfare. ⁷⁷ Shortly after his arrival in England, Squier was also asked to testify before a parliamentary committee which had been established to ‘consider the utilization for military services of wireless telegraphy and telephony, with specific reference to recent developments in the science, and to report’. ⁷⁸ Squier was not only highly qualified to testify on this subject but was probably the most qualified witness to appear. Unfortunately there is no record of his testimony available, but whatever he said had little influence on the overriding conclusion reached by the committee, that ‘the state of wireless in the Army was such that it would be better to abandon it altogether as an inefficient wireless service in war would be a constant source of doubt and danger’. ⁷⁹

Squier’s performance, acceptance by the British general staff, and technical ability more than justified the expectations of Wood and Walter Hines Page, the American ambassador to Britain. Squier worked closely with the leading British scientists to conduct experiments on radio phenomena. ⁸⁰ He also found time to invent new techniques for radio and submarine cable telegraphy. ⁸¹

With the outbreak of war, the importance of America as a friendly neutral and potential ally prompted the British government to extend special privileges to the American military attaché. Lord Kitchener, the Secretary of State for war, arranged for him to make a secret trip to the Western Front when no other attaché was allowed to do so. Squier was allowed to go anywhere, ask any questions, and interview anyone he wished. Overall he visited the front three times, more than any other attaché was allowed to visit, even those representing British allies.

In May 1916 Squier returned to Washington, DC, to take charge of the army’s troubled aviation programme. ⁸² He assumed his duties with characteristic energy and set about to remedy the situation by restoring morale and by establishing close ties with aviation organizations inside and outside the government. He was so successful in this endeavour that in February 1917 he was promoted from lieutenant colonel to brigadier general and appointed CSO of the Army. In this position, which he would hold until 1924, he initially oversaw both the Army’s aviation and communications missions – an extremely wide scope of responsibilities.

VIII. Manufacturing Issues

To this point the article has not addressed the vacuum tube manufacturing issues faced by Britain and the United States. In Britain, particularly, these issues were significant. At the beginning of the war, vacuum tubes were used only as experimental detectors in radio receivers and as repeaters in telephone circuits. During the war, tubes were needed to provide four essential functions for CW wireless: generating the carrier wave in transmitters, modulating the carrier wave with the audio signal, detecting the signal (both CW and spark) in receivers, and amplifying signals in both transmitters and receivers. Vacuum tubes were also required for local frequency generators (called oscillators) used in heterodyne receivers. The large number of tubes required and their complexity strained the manufacturing capabilities of all the belligerents.

French engineers designed an improved version of the De Forest audion known as the TM (Telegraph Militaire) tube. Mass production began in October 1915, and by war’s end more than a million had been made, most of them by Grammont and Compagnie Générale des Lampes, two incandescent-lamp manufacturers. ⁸³ Because the TM tube was both effective and robust, the British mass-produced several versions of it, beginning in 1916, manufactured by the British Thomson-Houston Company, the Edison Swan Electric Company (Ediswan), and the Osram-Robertson Lamp Works of the British General Electric Company (GEC). ⁸⁴ The British manufacturing plant, however, was severely strained by wartime demands, and ‘engineers and scientists were working under pressure to meet the incessant demands for more and better communications equipment’. ⁸⁵

The United States had a similar problem. Even though WE, the manufacturing arm of the American Telephone and Telegraph Company, had been mass-producing vacuum tubes for long-distance telephone communications before the war, it was unable to meet all wartime demands. GE had been mass-producing light bulbs for many years. Both light bulbs and vacuum tubes consisted of metal filaments, with external connections, in an evacuated or gas-filled glass enclosure. But vacuum tubes were more difficult to manufacture because their behaviour depended on the form and spacing of electrodes, the degree of vacuum, and the amount of residual gas in the bulb. The chemist and physicist (and later Nobel laureate) Irving Langmuir solved many of the problems posed, and other GE engineers worked out techniques for mass production of the tubes he designed. ⁸⁶

GE and WE agreed to divide the task of supplying the Army and Navy, with GE manufacturing transmitting tubes and WE receiving tubes. ⁸⁷ WE developed a number of tubes for the US Signal Corps and the US Navy. For example, the 203A (designated the VT-1 by the Signal Corps and the CW-933 by the Navy) was a general-purpose tube, used as detector, amplifier, and oscillator. By the end of the war GE had supplied the military with some 200,000 tubes, and WE had supplied half a million. ⁸⁸

A concluding observation for this section and a fair assessment of British CW development is contained in a history of the US telephone system. The history includes a gracious comment on the British effort to develop radiotelephony, i.e. modern CW communications, during the war:

The War came to Europe before the vacuum art had been applied to telephony and as a consequence the vacuum tube received only limited use in the European nations and only for radiotelegraphy. In the United States it was possible to continue peacetime development for several years after the war began and as a consequence this country’s technicians were in a much better position to apply radiotelephony to the war effort. ⁸⁹

IX. The Role of a Champion

The change from spark to CW technology was a paradigm shift for wireless which occurred simultaneously with the waging of the First World War. Only one of the belligerent armies, the American, took full advantage of this to provide its forces with modern radios. The British army failed to do so, despite the pioneering work accomplished by the outstanding engineers they had commissioned from the Marconi Company, the premier wireless company in the world. Two years before the Americans used the development of the airborne radio to spur the development of a suite of modern radios for air and ground forces, Major Charles Prince and other RFC engineers invented an airborne radio. Unlike the Americans, the British army lacked the vision to exploit it and commit the resources necessary to modernize its wireless communications.

There are many reasons why the Americans succeeded where the British and others failed, but one appears to be dominant in this case. A champion performs a vital role for the exploitation of a new technology: Jackson was the champion for wireless in the Royal Navy and Swinton and Churchill were champions for the tank in the British army. Squier was the ideal champion for modern wireless in the US Army for a number of reasons. First, as CSO, he was ideally placed to determine a course of action and carry it out. Though he reported to the chief of staff of the Army and to the secretary of war, and had to obtain necessary funding from Congress, he was the key decision-maker, and gave the orders to move forward and ensured they were executed. Second, his education, his extensive experience as an electrical engineer, and his knowledge of the communications-engineering state of the art gave him a complete understanding of all the technical issues involved. Third, he was a career soldier who had observed the European war perhaps more closely than any other American officer, and was aware of the high casualties and poor coordination caused by inadequate communications. Fourth, he enjoyed the technical respect of all American communications organizations, academic, government, and industrial, and their leading engineers; when he asked for their help it was freely given. Finally, he had the vision to see both that military aviation’s future lay in the direction of command and control of formations of aircraft and that the Army’s future lay in the direction of combined arms warfare enabled by a modern integrated communications system.

X. Conclusion

The authors are aware that the conclusion stated in the book and referred to throughout this article is a harsh judgement. The conclusion that the British army did not modernize its communication system when it could have means that it fought the war with a wireless communications system which was obsolescent after 1916. The consequences were battlefield failure, elevated casualties, and prolongation of the war. Both the book and this article briefly address possible mitigating factors that offer a partial explanation for the army’s failure to modernize its communications. The authors suggest these factors need further attention. They include technical difficulties with vacuum tube development, ⁹⁰ fear of enemy interception, ⁹¹ lack of recognition that poor communications led to high casualties, ⁹² the unanticipated scale of modern warfare, ⁹³ lack of a champion, ⁹⁴ and institutional bias and inertia. ⁹⁵

XI. Rebuttal

The authors assert that a reasonable background in technology and engineering development is critical to understanding the issues raised in the book and in this paper. The eminent historian of technology Hugh G.J. Aitken quoted earlier (notes 6 and 12) has such a background. It is not clear that Dr Hall has. As a military historian he may not be familiar with the technology that led to the profound paradigm shift in wireless or how swiftly CW replaced spark. This may explain his statement that the book’s authors lack the understanding of ‘scientific and technical limitations of the wireless sets of the era’.

Dr Hall also may not be familiar with the engineering and production processes necessary to take an invention from prototype to full production. This may explain his comment (referring to the costly first day of the battle of the Somme) that ‘Contrary to the claims of one recent work [the referenced book], it was a situation that wireless could do little to alleviate.’ What the book stated was that early production CW radiotelephones might have helped reduce the immense loss of life. ⁹⁶ These models would have been available had the British followed up on the RFC’s radiotelephone in 1915–16, as the Americans did in 1917–18 after they had independently built and tested their own radiotelephone.

Finally and most seriously, Dr Hall’s dismissal of the book as offering ‘very little in the way of fresh information’ is wrong and demonstrates questionable scholarship. The book carefully documents Prince’s work, the demonstration of the technology to Kitchener, and how the technology could have been exploited to provide the British army with modern communications. Dr Hall referenced the book four times in his two articles. Yet instead of engaging with a viewpoint at variance with his own he chose to denigrate the book as an unoriginal study. Perhaps he did so because it contradicted his assertion that ‘On the whole, the British army was remarkably successful at recognizing the utility of wireless and exploiting its full military potential during the First World War.’ ⁹⁷