by Mr. E. Montoriol
Inspector of Posts and Telegraphs
Annales des Postes, Telegraphes, et Telephones,
Vol. 5, No. 4, December, 1916
English translation by Eric Fischer, enf@pobox.com
In recent tragic circumstances, France has revealed to the world the energy resources in which it is rich, and which have been unsuspected even to many French. Among those who recognize this richness, many have accused it -- sometimes justly -- of scattering ideas, and leaving to others, less inventive but more practical, the care of pursuing their realization, thus permitting them to collect the moral and material advantages. There is at the same time a branch of science where this reproach is not merited: telegraphy. Delayed, at the start, with the Chappe system, which preceded electric telegraphy, it regained and has soon ascended to the place that suits it, and from the point of view of scientific research, to this the tool neatly says: multiple telegraphy, of French origin, thanks to Rouview, Meyer, and Baudot, is again, at this time, our uncontested domain.
Exactly forty years ago, the French Administration decided to construct the first quintuple Baudot apparatus; during this long period, the system, continually perfected, has stayed in the first rank among those of the entire world, and one does not forsee anything that will ever be able to dethrone it. At the occasion of this sort of anniversary, it is interesting to recall, as briefly as possible, some traits of the life of the inventor, and, at the same time, the stages his work followed from the initial conception to the present epoch.
Baudot (Jean-Maurice-Emile) was born in Magneux (Haute-Marne) the 11th of September, 1845. After receiving only a primary education, he spent his adolescence in agricultural work on his father's farm. Until his admission, as an alternate, to the Administration of Telegraph Lines (July 16, 1870), he lived a rural life, in an environment that in no way seemed to predestine him to become the genial inventor that he later was. It was only after his entry into the Administration that, seduced by the scientific side of his new profession, he enterprised to perfect his general education: he had everything to learn, in electricity and in mechanics! He put himself resolutely to this task, and in the short instants of leisure that his professional occupations left him, at the Central Post in Paris, he put his forehead, with a marvelous power of work, to the research that came to him from his enterprise on telegraphic apparatus and the study of science, without which the first would have been sterile. Promoted to Controller in 1880, after the first sucess of his system, he had the ambition to become an Engineer: without neglecting for this the work again in infancy, and without being frightened of the difficulties that it gave him to surmount, he prepared for the test, took it with success, and was named Inspector-Engineer in 1882.
At the moment when Baudot made his start, telegraphy had just been revolutionized by several inventions, notably those of the English Hughes and the French Meyer. The first had created his marvellous printing apparatus; the second, taking ideas expressed by Rouvier in 1858, had just imagined his "multiple" system, permitting the simultaneous exchange of four telegrams on a single conductor. For all who were interested in the progress of the telegraphic art, a seductive question posed itself, which consisted of researching a combination of these two inventions, which is to say to obtain multiple transmissions over a single wire, with the automatic translation of the signals into typographic characters. In 1872, Baudot enterprised to pursue the solution and delivered, to this end, a profound study of the systems successively invented before him.
The Hughes apparatus could be compared, in its principle, to the one described by Ampere in 1820: where the former system needed the use of 25 wires to distinguish the 25 letters of the alphabet, Hughes used 25 moments (or more exactly 28) to obtain this differentiation. These successive simplifications had permitted him to reduce to one the number of primitive wires necessary for a single transmission. Couldn't one, in an analogous condition, reduce the number of moments?
In 1832 Morse, after the invention of his system of dots and dashes, had attempted to reduce his telegraphic alphabet to a simpler form, by linking his manipulator by a variable number of wires, each of which commanded a pen; the pens, operated together or separately, wrote on different lines across the width of a band of paper, and the combination of traces thus obtained represented the letters of the alphabet. The same year, Schilling imagined a system of six magnetic needles, each commanded by a wire; this number was reduced to five by the Highton brothers and Davy in 1838, again to two in 1842, by Cooke and Wheatstone, and finally to one, by Bain, in 1843.
Among the most interesting alphabets, Baudot stopped first of all on that used by Davy in 1838: each of three wires was linked, at the start, to two keys, permitting sending a positive or negative current. At the corresponding station, each of these wires terminated at two galvanic relays, one functioning under the influence of the positive, the other of the negative; the functioning of each of these relays determined the path of a local current, which left a trace, by electrochemical decomposition, on a band of paper prepared for this effect; these traces occupied, along the width of the band, six different positions, following the relays by which they were produced and representing the combinations effected by the six keys of the keyboard at the beginning. One thus obtained 26 distinct combinations. The Highton brothers imagined, in 1848, a printing apparatus, using this same alphabet of three elements, with two directions of current. Witehouse, in 1855, attained 63 combinations, with six wires and a single direction of current; the reception effected, as in Davy's system, by an electrochemical decomposition.
Elsewhere, Gauss and Weber in 1834 and Witehouse in 1855 had obtained 31 distinct signs by the combination of 5 elementary signals, transmitted successively on *a single line of wire* and translated like the preceding.
Among all these alphabets, Baudot did not just have to choose which of these seemed best adapted to the new usage to which he wanted to put it; he tried, first of all, Witehouse's, with six signalling elements, and gained his first patent, the 17th of June, 1874; but with the principle of inversion [shifting], previously used by Hughes, exaggerating the the number of 63 combinations furnished by this system, he examined whether he should adopt that of Gauss and Weber, with five elements, or that of Davy, with three. This last seemed the most advantageous, from the point of view of efficiency of the line, but it required using two directions of current and, between two successive elements, left the line in a neutral state, which is to say exposed to exterior influences; these, acting on a sensitive receiver, were likely to create a great disturbance in the registration of the signals; also, Baudot preferred the alphabet of five units, which, if it was inferior so far as the efficiency of the line was concerned, permitted the use of a single direction of current, leaving the other for the idle condition, which is to say to keep the receiver from parasitic currents which no signal was crossing the line. More will be said below about these first two systems.
During this time, other inventors were searching to solve analogous problems. It was thus that Mr. Mimault, telegraph employee at Poitiers, patented, on January 17, 1874, a device called "Printing telegraph system" (no. 101,939); he made use of the 31 combination code, but over *five lines of wire.* The qualifier "printing" in truth was not applicable to this system, because the impression was made in conventional signs, needing a manual translation; but at the same time in a variation on this patent, the translation was given automatically and had as a result pointilistic figures, imitating the contours of ordinary letters, bringing them together, as a general aspect, like a work of tapestry; this pointilism was obtained by electrochemical decomposition. The administration rejected this project as overall a step backwards, since it required five wires. The inventor modified it and received, on July 4, 1874, a second patent for a new system, which he called "Printing telegraph with multiple transmission on a single wire" (no. 104,153). The word "printing" had, here, the same meaning as previously: the signals, received with the aid of the alphabet of 31 combinations, were constituted as rectilinear traces, as in Meyer's apparatus, from which it borrowed even the distributor and the receiver with inked nerves[?] in a helix. The Administration did not estimate that this second system had any marked superiority to Meyer's, and did not take it into consideration.
Now, as it was said above, it was June 17, 1874 when Baudot patented, under the number 103,898 and the title "System of rapid telegraphy" his first apparatus, which was both multiple and really printing, since the conventional signals were translated automatically *into typographic characters.* The code of signals was the one from Witehouse's (1855) six-wire apparatus. The manipulator was formed with six keys, linked to six contacts of a distributor, of which the brooms were synchronized with those of of another distributor, placed at the point of arrival; six contacts of this were each linked to an electromagnet. The brooms, in turning, successively put each key in temporary contact with the same-numbered electromagnet at the receiving post, and the armatures reproduced each of the 63 combinations that could be executed with the manipulator at the source. The functioning of the electromagnets determined the rotation of a typewheel: the first made it turn one division; the second two; the third four; the fourth eight; the fifth sixteen; and the sixth thirty-two; the various combinations of these movements permitted the chosen character to be brought above the strip of paper: the impression thus effected, the armatures were released, and everything returned to the waiting position, to receive and register the next combination. This geometric progression had previously been used in 1859 by Wheatstone, except that the English inventor had used four emissions instead of six, and used two directions of current. Mimault had used a similar progression in his systems where the letters were pointilized electrochemically.
The preparation of combinations at the start, and their automatic translation at their arrival, being purely local operations, it was indicated to use the necessary interval between two consecutive signals to distribute a line of other apparatus, as in the Meyer system.
A committee, designated by the Administration, and composed of:
| Messrs. | Blavier, Divisional inspector |
| Hequet, Chief of the Central Station in Paris | |
| Raynaud, Deputy of technical studies | |
| Clerac, Chief of the auditing service | |
| Charles, Chief of the workshops |
who had previously ruled on the first of Mimault's projects, were charged with examining Baudot's; after study, they proposed to the administration to open a credit of 2000 francs for the construction of a test apparatus: the experiments commenced in December, 1875; they were encouraging enough that it was decided to construct a quintuple installation, destined to give regular service. This was executed at the Dumoulin-Froment workshop and patented March 2, 1876, under the name of "Printing multiple telegraph system apparatus" (no. 111,719). For the reasons enumerated above, the code definitively adopted was that of five signalling elements; in addition, the translation at receipt was done with the help of a new part, to which Baudot had given the name of combinator, and which no longer used the geometric progression.
Meanwhile, Mimault had raised complaints, claiming that his patent of January 17, 1874 assured him the exclusive right to the geometric progression and to the alphabet of five signalling elements, of which Baudot had made use, although the principles had been in the public domain since 1859 and 1835, respectively. The geometric progression having been abandoned in Baudot's second patent, the litigation was limited to the use of the alphabet previously employed by Gauss and Weber and by Witehouse. Invited by the committee to make known his new projects, Mimault promised to submit his plans to them, but he did not present them; Count Th. du Moncel, from whom Mimault had solicited support, tried in a friendly letter dated January 15, 1877 to dissuade him from judiciary action, developing to him the technical considerations that left no doubt of the checkmate that awaited him; but despite this sage advice, Mimault brought a charge of infringement against Baudot on June 17, 1877. A first judgement, rendered May 15, 1879, by the Civil Tribunal of the Seine, formally discarding the charge of infringement, awarded Mimault the proprietorship of the principle of using a code of 31 combinations with the aid of five elementary signals; Baudot's patents were considered simple "patents of perfection."
This judgement did not satisfy either party in the case, who appealed, including Mimault himself. Mr. Cochery, who had become in the meantime the Minister of Posts and Telegraphs, wanted to clear himself on the well-foundedness of Mimault's claims; for this he brought together an extra-administrative committee composed of:
| Messrs. | Mascard, Member of the Institute, Professor at the College of France |
| Cornu, Member of the Institute, Professor at the Polytechnic School | |
| Marie, Engineer in Chief of the Paris-Lyon-Mediterranean Company |
The consultation written up by this committee had the conclusion "that the claims of Mr. Mimault are unfounded; Mr. Baudot has borrowed nothing from the principles of the Mimault system." Finally, the Court of Appeals of Paris, and then that of Amiens, on May 7, 1882 and May 27, 1884, respectively, annulled the judgement of the Civil Tribunal of the Seine, declaring Mimault "ill founded in his demands, ends, and conclusions," uprighting[?] him and condemning him to all the expenses. This end was confirmed May 4, 1885 by the Supreme Court of Appeals.
One cannot pass silently by this incident in the life of Baudot, without the trouble of seeming to attribute to it a well-foundedness that is demonstrated not to exist by the impartial examination of facts, the high moral authority of the personalities successively called to formulate their opinion, and the final authority of the the definitive judgement, after the exhaustion of all jurisdictions. However, the lines that precede are not at all meant to weigh down Mimault, who protected his double title of worker and vanquished; turned sour by the receipt of the one who he considered his rival, in a place into which he had worked himself; knowing that this success made useless the perfections that he had forseen, and irredeemably barred the route toward which he had made all his hopes; without doubt did he believe in good faith that the principles on which he had attempted his application, and of which no use had been made for twenty years, really belonged to him; without doubt again would he not want to see the profound differences that, in realization, cleanly separated Baudot's invention and his own. "One never persuades an inventor," Du Moncel amicably said to him, in his previously cited letter. He moreover cruelly attoned for the moment of frenzy that, on January 4, 1888, cut short the life of the unfortunate Raynaud... peace, then, to his memory.
From this time until his death, Baudot never ceased to devote himself exclusively to the perfection of his work, employing in this job all his time and all the credit that the Administration put at his disposal, completing this even with the help of his own modest personal resources. These sometimes were insufficient, and it was thus that in 1880, it was learned that he had been forced to pawn in Mont-de-Piete the great gold medal that had been awarded to him by the jury of the International Exhibition of Paris in 1878. The Administration, as soon as they heard, hastened to come to the aid of the inventor. (The author of these lines has this detail from Baudot himself, who, in informal chats, liked to recount thus stories from the rural life of his youth or of the difficulties of his administrative beginnings, never blushing from his modest origins and never taking vanity from the situation that he had acquired from his work.)
The Baudot system was understood progressively in France, then imposed itself soon in foreign countries; Italy was the first to introduce it, in 1887, in its interior service; Holland, in 1895; Switzerland in 1896, Austria and Brasil in 1897, England in 1898, Germany in 1900, Russia in 1904, the British Indies in 1905, Spain in 1906, Belgium in 1909, the Republic of Argentina in 1912, and Romania in 1913. The inventor obtained the knight's cross from the Legion of Honor in 1879, and was made an officer in 1898.
Baudot died in 1903, worn out prematurely by a life of labor, from which he never knew any rest; but, privileged among so many inventors who succumbed to trouble, he had the rare happiness of assisting in the generalization of his system and of seeing his genius radiate to the ends of the earth.
[This is followed in the original by a long technical discussion, which I have not translated, of the evolution of the Baudot telegraph system and some of its descendants.]
Thanks to Ric Werme for correcting a typo.