Attacked in public by his friend and mentor over the compromise solution that he had crafted, Poincaré too now had to intervene in print, not least because the legitimacy of his commission had been thrown into question. Like Cornu, he acknowledged the competing interests. Unlike Cornu, however, Poincaré saw such discord as the mandate for a rough and ready settlement. As always he aimed for an engineer’s progressive middle course, avoiding both revolution and reaction. The essential point, in his eyes, was to drum out of existence such monstrosities as 8h25m40s or 25°17' 14". As far as he was concerned, the important point was the adoption of any decimal system.
As Poincaré knew full well, the physicists (that is, the electricians) had fled the room. Poincaré now tried, gently, to usher them back. Surely they had exaggerated the inconvenience of the decimal system. If they’d only read the report, they would come over to his side (or so Poincaré thought). So he began to cite from it. After so many years of struggle to establish an acceptable system of electrical units, it was understandable that the electricians could not easily abandon it now. But, he urged them, be reasonable. For all practical purposes, by any measurement one simply is comparing one thing to another: one time to another time, one resistor to another resistor. In any such industrial operation there is no need to go back to the underlying definition of a unit. The storekeeper measures cloth with a meter and has absolutely no need to remember that the meter is the forty-millionth part of the terrestrial meridian. Only the so-called absolute units would be damaged by the reform. (An absolute unit of electrical current, for example, was defined not in relation to any particular materials, but by identifying an ampere as just that amount of current which when passing through two parallel, infinitely thin bars a meter apart would repel each other with a certain force.) In effect, Poincaré said: Let the English worry about the natural theology of absolutes; it was convenience, not divine sanction that mattered.
Indeed, Poincaré insisted that the object of the physicists’ complaint was inconsequential. True, the old sexagesimal clocks with their sixty seconds and sixty minutes could only awkwardly be compared to the new observatory clocks that would display centihours (hundredths of hours, equal to thirty-six seconds). So what? Chronometers need only indicate intervals of time, indeed very small intervals. There is simply no need to reset them to a specific time of day. But push the thought experiment to the limit. Imagine, Poincaré continued, that the astronomers adopted the decimal system, followed by such a spectacular diffusion among the general public that there was not a clock based on seconds still to be found anywhere. How much bother might those rare physicists be subject to who wanted to determine the absolute value of electrical resistance (the ohm)? They would have to multiply by 36. “And for them to avoid this operation we are daily to impose tedious calculations on thousands of mariners along with millions of schoolchildren and former schoolchildren?” Which do we do more often—determine the absolute value of the electric resistance, fix our position at sea, or add two angles, or two times? In sum, Poincaré accused the physicists of blocking progress for the astronomers and the public just because they, the physicists, would not profit. As far as Poincaré was concerned, some progress was better than none. If units differed in astronomical treatises and electrical books, that was a small price to pay to rid the world of the absurdity of having three units in a single number like 8h,14m,25s.26
Despite the prodigious amount of work they put into their campaign, Poincaré’s time commission stalled. Finding open foreign hostility to the idea of such time reform, the Ministry of Foreign Affairs let the Bureau of Longitude know in July 1900 that State was not ready to back the effort. After a hundred years, the revolutionary struggle to rationalize time expired.27
Although they lost the attempt to decimalize time, many participants in Poincaré’s commission argued passionately over time zones and time distribution. Sarrauton, for example (never one to avoid polemics), turned his guns toward the zone system. He took one of his scathing reviews (on the proposed time zones) and sent it personally, on 25 April 1899, to Loewy at the Bureau of Longitude. It began, as so many time-pleas did, with an homage to the rail and cable: “The surface of the globe is criss-crossed by trains riding on rails, by fast boats full of voyagers and merchandise, and on the aerial and submarine telegraphic lines, news circulates with the speed of light. . . . The surface of the planet has been, in a certain sense contracted.” Time zones of synchronized clocks were the answer, but certainly not these English time zones.
Sarrauton demanded that the wedge-shaped time coordinated divisions of the earth be liberated from the grasping claws of the British Empire. His ire had been roused by a proposed “Boudenoot” law that would have delayed Paris time by 9 minutes and 21 seconds: “This is immediately the time of Greenwich; the English meridian, soon, ‘France towed by England’ and the collapse of the metric system!” Happily, as far as Sarrauton was concerned, there was an alternative—the law of Gouzy and Delaune—that would do the right thing by zones, decimalization, and the long-lost prime meridian: it would divide the hour into 100 minutes, the minute into 100 seconds, set civil time into twenty-four time zones, and count longitude from the Bering straits beginning on 1 January 1900. “This is the achievement of the system of decimal units, France realizing one of the most important reforms of modern times, and in scientific matters, the preponderant French influence in the world. We have arrived at a crossroads and these two projects of law mark the two roads that open before us. We must choose.”28 Whatever its advantages for rational France, Gouzy and Delaune’s proposal never passed. France adopted the Greenwich meridian on 9 March 1911.
In these ferocious debates over zones, decimalization, and the prime meridian, time conventions crossed arenas that we tend to think of as far from each other. Laws, maps, science, industry, daily life, and the legacy of the French Revolution all collided, drawing in leading figures from the French technical, intellectual, and scientific establishments. At the Bureau of Longitude, Poincaré’s philosophical hope for “conventions” and “convenience” ran smack into the everyday realities of navigators, electricians, astronomers, and railroaders. Just before Poincaré published “The Measure of Time” in 1898, physical, conventional, and coordinated time had converged in whirling reformist debate that was thoroughly abstract and yet altogether concrete.
Of Time and Maps
If one of the Bureau of Longitude’s projects in 1897 was the establishment of decimalized time, even more pressing was the orchestration of one of the most difficult time-synchronized mapping projects in the Bureau’s illustrious history. Already in 1885, the Ministry of the Navy had assigned the Bureau the task of determining the exact positions of Dakar and Saint-Louis in “our colony” of Senegal.29 Years in the making, the Senegal report only appeared among the Bureau’s publications in 1897—coming into Poincaré’s hands just before he wrote “The Measure of Time” and took on the presidency of the Bureau.
Mapping “our colony” was no easy task. Governor Louis Faidherbe had violently annexed the Wolof and Cayor regions of Senegal by 1865, enabling the colonizers, at least intermittently, to clear the road from Saint-Louis to the Cape Verde peninsula. By 1885, the colonial government had a railway link under construction joining Saint-Louis and Dakar.
But laying iron tracks did not stop fierce anticolonial resistance. French forces, still battling in the east and south, never completely suppressed the rebellion against them. Insurgencies were still erupting on the eve of World War I. In the heat of these colonial battles, the Bureau’s men were there to fix the position of Senegal’s two main cities, in part to extend mapping into the interior of the colony alongside conquering troops. But the French colonial project had even grander ambitions. By providing exact coordinates at Dakar, the French authorities intended to extend their cable system from that port down the length of the west coast of Africa to the Cape of Good Hope. Longitude, train tracks, telegraphy, and time-synchronization reinforced each other
. Each showed a different facet of a new global grid.
By the time the Dakar–Saint-Louis expedition set off from Bordeaux, the Bureau had already extended its telegraphic reach to Cadiz’s San Fernando Observatory (about fifty miles northwest of Gibraltar). From there, the French relied on British Cable’s link joining the Cadiz and Tenerife observatories (Tenerife Observatory had just recently been cabled to Senegal). One hour each evening, the astronomers controlled the cable. For France, as for all of Continental Europe, renting cable time from the British had entered into the order of things. Controlling the vast majority of the world’s submarine cables, the British passed messages between France and its colonies everywhere but North Africa. Alone the Tenerife-Senegal, West African, Saigon-Haiphong, and Obock-Perim cables cost France almost 2.5 million francs per year, paid, gallingly enough, to their imperial competition, the British. Such dependence burnt slow and deep in the French establishment; the military, commercial, and journalistic sectors all chafed during the 1880s and 1890s under British communications dominance. But the Chamber of Deputies blocked French cable proposals one after another: One casualty was an 1886 link from Reunion to Madagascar, Djibouti, and Tunis; another, in 1887, from the French West Indies to New York, and a third from Brest to Haiti, in 1892–93.30
Figure 4.1 Cabling Simultaneity: Paris, Cadiz, Tenerife, and Dakar. Undersea cables proliferated in the last part of the nineteenth century, and longitude finders instantly seized the opportunity of using each new link to wire time across the oceans. The French Bureau of Longitude made use of British and Spanish submarine cables to get their time signal from the observatory at Cadiz, Spain, over Tenerife and thence to Dakar. SOURCE: VIVIEN DE SAINT-MARTIN, ATLAS UNIVERSEL (1877).
Just as they relied on the British cable-laying firm, the French authorities needed the Spanish. Cecilion Pujazon, astronomer at San Fernando, offered his observatory as a relay point, linking the Paris-launched signal to Senegal through Tenerife. Taking the steamship Orenoque of the Compagnie Transatlantique, the astronomers arrived, worse for the wear, in Dakar on 15 March 1895. Governor Seignac-Lesseps immediately put his troops at their disposal. The artillery captain commanding the Dakar garrison offered laborers and indigenous masons, and housed the scientists in the military’s mess halls—there being (according to the querulous chief astronomer) no decent hotel to be found anywhere. With horror the chief astronomer noted that passengers were sleeping in straw bedding like natives.
Military aid to the map troops did not end with food and lodging. The Bureau of Longitude astronomers took up their posts in the fort’s blockhouse, looming on the point of Cape Verde to protect the coal park and anchorage. Thick concrete walls designed to protect artillerists against attack allowed the astronomers to install their timekeeping pendulum in the bunker’s sheltered room. This kept the temperature of their all-important clock under control, an absolute necessity in Dakar’s ferocious heat. Establishing their meridian telescope outside the blockhouse, astronomers mounted the collimator on the parapet of the gun battery. Five days later, on 20 March, the team steamed for Saint-Louis. Welcomed by the governor’s aide-de-camp at “the political center of our colony,” they hammered into place a Saint-Louis station.31
A gubernatorial edict prevented people from entering or leaving their operations building during the observations (footsteps disturbed the pendulum). Those few natives allowed to pass by did so without carts, in bare feet, on sandy soil. Observations ran from 26 March until 11 April 1895; the team launched their first signals to Dakar on 29 April and to San Fernando on 2 May.
Not everything went well. The river flowed north-south, making the Saint-Louis meridian sighting extremely difficult. It was good to be on the riverbank to avoid obstacles, but they couldn’t very well plant themselves in the middle of the river to sight due north toward the polestar. “Marauders” began stealing anything metal left in the open, including, almost daily, the spike used for collimating the shooting of the stars.
Figure 4.2 Measuring Dakar. The overlap of military structures and geographical work was extensive—for the French, the British, and the Americans. At Dakar, the French longitude expedition used every aspect of the fort overlooking the vital coal yards to establish the fundamental longitude measurement of the colony. SOURCE: ANNALES DU BUREAU DES LONGITUDES, VOL. 5 (1897).
The 260 kilometers of wire that link Dakar to Saint-Louis were established in the bush-covered plains of Cayor, with primitive means and in the midst of a population that was hostile if not in open war with us. The wire was so often cut, the poles so often knocked down and then re-established more or less well, that one does not find the normal [physical] conditions of an ordinary line. Let me add that at sunset the dew was so strong that the poles dripped, and where 5 insulators ordinarily would suffice [back in France], even 70 failed to give a trace of electrical wave.32
Even once the signal made it along the train tracks through Senegal, the signals from Santa Cruz (Tenerife) to San Fernando failed because, during the crucial nights of observation, there were active and official transmissions on the Spanish land lines following the election of deputies. Then came corrections. There were the usual corrections for temperature variations; there were corrections through “personal equations,” the characteristic psychophysiological lags of each observer as he signaled a star crossing the meridian. There were corrections for the behavior of the delicate mirror that registered the signal’s arrival. In the end, after doubling the weight of one set of observations and compensating for various other instrumental and human errors, the team settled on a San Fernando–Saint-Louis longitudinal difference of 41 minutes, 12.207 seconds.33 Not wanting to miss the opportunity of a rare steamboat leaving Dakar for Tenerife, the astronomers corralled some homesick sailors to quickly pack their instruments for the voyage back to Paris. Their report to the Bureau appeared in print in 1897.
By the time the Senegal report saw light of day, French-British relations over their colonies were deteriorating, and telegraphic cables ran through the center of many disputes. British cable companies refused to employ foreigners and were clearly reading French messages between Paris and Dakar or Saigon even before the French authorities had them in hand. At the 1885 Cable Convention, Britain insisted on the rights of warring nations to tear up their opponents’ lines—clearly an advantage to the country that owned twenty-four of the thirty extant cable ships. In 1898, tensions increased to the brink of war. Racing to control a part of the Nile in Sudan, Captain Marchand’s French squadron was ready to stake a claim; but while the British troops remained in continuous contact with London, the French cable inexplicably went silent. Only when the French Governor loaded the cannon in Dakar (where the French Bureau of Longitude men had just been) did the cable, magically, come alive.34
Back in Paris, the longitude battle between the French and British capitols would not end and could not be contained in the domain of the politicians. Even as the Bureau of Longitude continued to vex itself about whether French clocks should be set by Greenwich time, the astronomers went clock-to-clock over the most basic of questions: where was Paris in relation to London, or rather, how far east did the Paris Observatory stand in relation to the Royal Observatory? Back in July 1825, the two national rivals had tried to settle the problem by launching rockets from the English Channel, using their explosions to synchronize clocks. Le Verrier and Sir George Airy then tried again by telegraph in 1854, settling on a difference of 9 minutes and 20.51 seconds. Unfortunately that consensus collapsed in 1872 with the American Coast Survey telegraph campaign that accompanied the laying of the transatlantic cable. It determined that Paris was nearly a half-second farther away from London—9 minutes and 20.97 seconds. Half a second was far too large an error for anyone to tolerate in the era of hundredths, if not thousandths, of a second, so the Astronomer Royal, along with General Perrier (director of the Geographical Service of the French Army) and Admiral Mouchez (director of the Paris Observatory) joined forces in 1888 to settl
e the question (so they hoped) for good.
Appointing two observers from each country, the astronomers planned to conduct measurements shoulder to shoulder. A Frenchman and an Englishman would literally stand next to each other, with one pair in England and the other in France. They shared a telegraph line, they traveled together back and forth across La Manche, and they coordinated their procedures down to fine corrections of personal errors. They even shared a single battery-run electric light to avoid needless heat that might distort their instruments. At Montsouris, the French and English instruments stood on piers twenty feet apart. Yet the results caused dismay:
Einstein's Clocks and Poincare's Maps Page 17