Semaphores for optical telegraphy

Historians tell us that fires were used for conveying simple military messages at least three thousand years ago. There is little doubt that relays of fire beacons were used by the Ancient Greeks to send simple messages around 500 BC. Earlier use of such beacons is plausible, but distinguishing fact from fiction can be tricky. The news of the capture of Troy is said to have been conveyed rapidly to Argos by a sequence of burning beacons. However, the first account of this message that can be dated reliably is a play by Aeschylus written about 458 BC, which was several hundred years after the event. The shortest distance from Troy to Argos is around 400 km, which includes some extensive patches of sea. Telescopes had not yet been invented. Furthermore, during

Figure 8.3. Beacon indicating safe route to Salcombe harbour. A sailor on the required narrow course towards the harbour sees the beacon as a white light. If the boat is too far east, the beacon appears green (here represented by the light grey arc); if too far west, it appears red (here represented by the dark grey arc). The beacon can be seen from more than 10 km, which is outside the area of sea shown here.

Figure 8.3. Beacon indicating safe route to Salcombe harbour. A sailor on the required narrow course towards the harbour sees the beacon as a white light. If the boat is too far east, the beacon appears green (here represented by the light grey arc); if too far west, it appears red (here represented by the dark grey arc). The beacon can be seen from more than 10 km, which is outside the area of sea shown here.

Odysseus's lengthy journey home after the Trojan War he is reputed to have encountered a giant with one eye in the centre of his forehead. Perhaps this indicates that descriptions of optical arrangements around the time of the Trojan War should be taken with a grain of salt.

However, it is clear from the writings of Polybius in the 2nd century BC that some serious thought about optical signalling had occurred by the time the Roman Empire was developing. (Polyb-ius is usually pronounced with the stress on the second syllable, as in polygamy.) Although Polybius was Greek, he spent many years in Rome and wrote extensive accounts of the city's history. Within these he described a method of coding that allowed texts of military messages to be conveyed by displaying visible signals.

The technique involved the simultaneous display of two groups of identical symbols, such as flags. Each group consisted of between one and five symbols, so that the two groups could provide twenty-five possible combinations. The Greek alphabet contained only twenty-four letters, which allowed a unique combination to represent each letter. It seems that little use was made of this early form of semaphore, probably because the observer needed to be fairly close to identify the symbols displayed.

For thousands of years the use of light signals for sending messages over long distances was limited by the performance of the human eye. By the 13th century it was known that a glass lens in front of an eye could correct for long sight, but it was not until the beginning of the 17th century that it was discovered that two different lenses mounted at opposite ends of a tube could make distant objects appear larger and dim objects more visible. Such slow progress in optics is not surprising because the expounding of scientific theories in Europe was apt to lead to arrest, torture and death, even at the start of the 17th century. However, in the 1660s Louis XIV gave one of his sunny smiles to the Academie des Sciences in Paris, and in the same decade the Royal Society in London obtained its charter from King Charles II. One of the first officers of the Royal Society was Robert Hooke, who made important contributions in many fields of science. He described the use of telescopes in optical communications, and suggested that messages could be transmitted between London and Paris in a matter of minutes via a series of signal towers. However, it was not until a century later that political and military circumstances led to the establishment of long distance telegraphy using light to convey the information.

The pioneer of this method was a Frenchman called Claude Chappe. His career as a priest had been interrupted by the outbreak of the French Revolution in 1789, but the upheaval had some beneficial effects. The new form of government was less reactionary and more willing to consider new ideas. For example the metric system was proposed by the Academie des Sciences in 1793 and approved by the politicians in 1795. Furthermore Ignace Chappe, Claude's elder brother, had manoeuvred himself into a position with significant political clout. Perhaps the most decisive factor in getting support from the government was that France was at war with most of its neighbours simultaneously,

Figure 8.4. Semaphore structure of Chappe's mechanical optical telegraph. The fixed vertical post supported a pivoted beam with a pivoted indicator arm at each end. The beam was set at one of the four angles shown in the left drawing. Each indicator arm was set at one of the seven angles shown in the right drawing. The eighth angle for which the arm was completely hidden by the beam was not used. The grey drawings show two of the 196 (4x7x7) possible combinations of the three angles.

Figure 8.4. Semaphore structure of Chappe's mechanical optical telegraph. The fixed vertical post supported a pivoted beam with a pivoted indicator arm at each end. The beam was set at one of the four angles shown in the left drawing. Each indicator arm was set at one of the seven angles shown in the right drawing. The eighth angle for which the arm was completely hidden by the beam was not used. The grey drawings show two of the 196 (4x7x7) possible combinations of the three angles.

which increased the appeal of any proposed new technology offering a military benefit.

The first mechanical and optical telegraph link was brought into use in 1794, connecting Paris and Lille, which lie about 200 km apart. Along the route at intervals of around 12 km were relay stations, each consisting of a tower supporting a semaphore structure, illustrated in figure 8.4. The term semaphore implies that the information was represented by the angles of adjustable components. In Chappe's system there were three separately adjustable angles that could be reset in less than one minute and observed from the next relay station with the aid of a telescope. The fixed vertical post carried a centrally pivoted beam, sometimes described as the regulator. At each end of this beam was an indicator arm, pivoted at one end to allow its angle to be set independently. Ropes and pulleys were used to control the angles. To reduce errors in recognition, the beam was set at one of only four angles and each of the indicators at one of seven. The total number of possible combinations of the three angles was 4 7 7, which equals 196. This is much larger than the number required for the 26 letters of the alphabet and the ten numerals, so most combinations could represent whole words, thereby compressing the message and allowing it to be sent more quickly.

Some simple arithmetic permits the performance to be compared with more modern telecommunications systems. Because 27 equals 128 and 28 is 256, each of the 196 distinct settings of the semaphore is equivalent to between seven and eight bits in modern IT jargon. If each successive semaphore position is displayed for one minute, the rate could be described as about 7.5 bits per minute or 0.125 bits per second. (As we shall see in chapter 10, nowadays even 1010 bits per second transmitted through an optical fibre does not win any prizes.) A compressed message consisting of twenty successive combinations would take 20 minutes to send to the next relay station 12 km away. Although light travels at almost 300000 km/s, this complete message effectively travels at only 36 km/h. (This would be the maximum speed at which this message could move from end to end if the entire message needed to be received at one relay station before it was sent to the next.) Although this is slow, it does allow error correction at each relay station. If a small fraction of the combinations is misread, the errors may be obvious and correctable before onward transmission. Without such correction the errors could accumulate as the message travelled on, possibly resulting in gobbledygook for the ultimate recipient. On the other hand there would be no point in delaying the onward transmission if the operators at each relay station were unable to make corrections. Often the messages were enciphered to make them incomprehensible at intermediate stages, because the relay stations could easily be observed by spies. So without waiting for the entire message to arrive, each combination could be sent on from one relay station only two minutes after its arrival. Moving 12 km in two minutes corresponds to a speed of 360 km/h. This is somewhat faster than conveying the message on the TGV, the modern French high-speed train that takes about one hour on the journey between Paris and Lille.

The speed of messages on this mechanical optical telegraph was so much higher than that of any available alternative that an extensive network of links was developed during the next thirty years, in spite of severe turbulence in the French political scene. The second and third links were completed in 1798, connecting Paris to Brest in the west and to Strasbourg in the east. The link to

Lille was extended to Brussels in 1803 and reached Amsterdam in 1811. A transalpine link reached Milan in 1809. (Although these cities are not part of France today, they were under French control in the early years of the 19th century.) By 1823 there were links more than 650 km long to Bayonne on the Atlantic coast near Spain and Toulon on the Mediterranean coast. The French telegraph system employed more than three thousand people in the middle of the 19th century.

It did not take long for the capabilities of the mechanical telegraph to become known in other countries. The second country to adopt mechanical telegraphy was Sweden. Abraham Edelcrantz devised and constructed a system with one intermediate station to convey messages between the royal palace in the centre of Stockholm and the royal country residence at Drottningholm, about twelve kilometres to the west of the city. A royal greetings message was sent by the first version of the telegraph on the birthday of King Gustav IV in November 1794, less than four months after the inauguration of the French link between Paris and Lille.

The Swedish mechanical telegraph differed from the French one in that it did not use a semaphore system. Instead each tower was equipped with an array of ten shutters, each of which could be set either open or closed. In modern ITjargon, each shutter represented one bit and ten bits were sent in each time slot. The ten shutters provided 210 or 1024 combinations. In 1796 Edelcrantz published a treatise on mechanical optical telegraphy that was translated into French and German. The Swedish telegraph system was extended for military purposes, and proved to be particularly valuable during the war with Russia in 1808 and 1809. The technique remained in use in Sweden longer than in any other country in Europe, surviving in the Stockholm archipelago until 1876 and around Gothenburg until 1881.

It might also be said that the French were responsible for the development of mechanical telegraphy in Britain. The Admiralty, based in London, decided that rapid communication with the coast was needed because peaceful co-existence with the French appeared increasingly improbable towards the end of the 18th century. Before the end of 1796 London had been connected to Deal and Portsmouth, each link being about 120 km long with relay stations at intervals of about 15 km. Each relay station normally had a crew of four, two men to observe the adjacent stations

Figure 8.5. Shutter structure for the original British mechanical optical telegraph. The 3 2 array of six shutters is shown with four shutters closed and two open. The shutters rotated on a horizontal axis, but the mechanism by which the operators below changed the positions is not shown. The pattern of dark and light here has been arbitrarily selected from 64 (26) possible arrangements.

Figure 8.5. Shutter structure for the original British mechanical optical telegraph. The 3 2 array of six shutters is shown with four shutters closed and two open. The shutters rotated on a horizontal axis, but the mechanism by which the operators below changed the positions is not shown. The pattern of dark and light here has been arbitrarily selected from 64 (26) possible arrangements.

with the aid of telescopes and two to operate the levers controlling the settings. In 1806 the London-Plymouth connection was completed and in 1808 London was linked to Great Yarmouth. The original British system resembled the Swedish one in the use of shutters, which were set either open or closed. They were mounted in a wooden frame about six metres high and could be rotated about a horizontal axis. As shown in figure 8.5, they were almost square in shape and arranged in a three-by-two array, which permits 26 or 64 possible combinations. With 36 combinations allocated to letters and numerals there remained 28 for use as shorthand for important and frequently used naval terms. A large model of a British shutter telegraph relay station is displayed at the Royal Signals Museum near Blandford Forum in Dorset, close to the site of one of the former relay stations between Plymouth and London.

When Napoleon relinquished power and was exiled to Elba in 1814, France was no longer considered a threat. In order to save money the British mechanical shutter telegraph links were closed down and dismantled. When Napoleon resumed political and military activities for a few months in 1815 the British telegraph system was in no fit state for conveying to London the news of Napoleon's defeat at the Battle of Waterloo. The information reached Paris much sooner. The ensuing embarrassment in Britain led to renewed enthusiasm for mechanical optical telegraphy. The new British system used neither the same technology nor the same routes as the old one. A substantial post with two semaphore arms replaced the array of shutters at each relay station. Each arm could be set in seven distinct positions at 45 intervals, the two vertical positions (up and down) being treated as indistinguishable. Because the two arms were at different heights they could not be mistaken for each other and so they provided 49 different combinations, equivalent to about 5.5 bits in IT jargon. The link between London and Portsmouth was in use from 1822 to 1847. The route selected lay a little to the northwest of the previous one and had more intermediate relay stations. One of these was restored in 1989 and is now a small museum devoted to the history of this form of telegraphy. Figure 8.6 (colour plate) shows the hexagonal five-storey brick tower at Chatley Heath in Surrey, just over one kilometre southeast of the M25/A3 interchange, and a fifteen-minute walk through the woods from the nearest car park. The pub named 'The Telegraph' in Telegraph Road on Putney Heath in southwest London provides another clue to the route.

Although long-distance mechanical semaphore communication had been discarded well before the end of the 19th century, its little brother lasted well into the twentieth for naval and military communications. This form of semaphore used a pair of hand-held flags, usually with two bright and contrasting colours. The British army used it at the start of the First World War, but so many signallers were killed by enemy snipers that semaphore from the trenches had been abandoned by 1916. However, Lord Baden-Powell, the Chief Scout, continued to regard hand semaphore as a useful skill for members of the scout movement. Competence in semaphore was required to obtain the

C rib

Numbers Letters V W X Z

Figure 8.7. Flag positions for semaphore. The 28 flag combinations used by the Royal Navy in the First World War are presented here systematically. They did not exactly match alphabetical order, J and Y being anomalous. Flags were usually coloured red and yellow.

Second Class Badge until the syllabus was radically revised in the 1960s.

The semaphore code used angles at 45° intervals. This provided 28 possible combinations (illustrated in figure 8.7), more than enough for the alphabet. The letters of the alphabet followed a simple sequence, clockwise from the point of view of the observer. For some arcane reason the positions of the letters J and Y were anomalous. The ten numerals had to share symbols with ten letters, though confusion was avoided by using two symbols to indicate a change from letters to numerals or the reverse. Although the basic code had worldwide agreement, the special symbols for error cancellation, start of message and suchlike had local variations.

Was this article helpful?

0 0

Post a comment