A question is bound to arise how a scientifically ignorant person, like the present writer, can presume to review Charles Gillispie’s remarkable book (itself the continuation and completion of his earlier Science and Polity in France at the End of the Old Regime, published back in 1980). Of course, Gillispie’s is a work not of science but of history, the history of science, and to some extent of political history. It is concerned with the changing shape of what the state seeks from scientists and gives them in return, and the way in which, in response to this, science organizes itself, both as a profession and as a discipline. Gillispie’s example is France just before and after the French Revolution—and for a good reason, for during this period (which spans two generations) French scientists were well ahead of all their rivals.

But the question I opened with had better be faced. For Gillispie is, clearly, thoroughly at home in the language and concepts of the sciences: of pure mathematics, chemistry, mathematical physics and astronomy, biology and natural history, not to mention engineering, mining, and agronomy. He is also a master of lucid explanation, so far as explanation to the ignorant can go, as well being the teller of some highly gripping tales; and he has an admirable, logically taut, often quietly witty, prose style. The uninstructed reader could be said to get everything from him—except science.

Fifty years or so ago the novelist C.P. Snow, in a public lecture, launched a debate about what he called the “two cultures”—meaning literary culture and scientific culture—and the strange and almost total lack of communication between them, which he blamed mainly upon the literary culture. It was a foolish lecture, as may be seen from his complaint that the epoch-making discovery made in 1956 by the American particle physicists Yang and Lee—that parity, or space-reflection symmetry, is not conserved in weak interactions—was most probably not, as it so obviously should have been, “talked about at every High Table in Cambridge.”1

The nonconservation of parity: a pretty topic for general conversation! What on earth could the ignorant be expected to contribute to the talk? The best they could do would be to listen open-mouthed. But then, that is the point that should be made. The gulf between Snow’s two cultures is irremediable because of a quite simple but iron law: that nothing a nonscientific person says about a scientific matter can be of the slightest interest to the scientist. There is no symmetry, and can be no genuine exchange, between scientific and literary culture. Gillispie supplies a rather neat rider to this. To an observer contemplating three centuries of science, he says, it appears—a little surprisingly—that communication has succeeded more nearly among scientists than any other class of person. But that is because their exchanges are “confined to matters to which people in ordinary life are generally quite indifferent.”

The story, as Gillispie tells it, begins with the appointment in 1774 of Anne-Robert Jacques Turgot as controller-general of France. Turgot, a high-minded and dedicated man, knowl-edgeable both in economic theory and certain sciences, was a born administrator rather than a politician, and his ministry was short-lived; nevertheless it represented something quite new in French government, a willingness to employ experts and scientists to find answers to social problems. What Turgot initiated, according to Gillispie, was a long-term progression from bureaucracy to technocracy.2

He illustrates this progression with the career of Antoine Laurent Lavoisier, a founder of modern chemistry, who during Turgot’s ministry became guardian of France’s gunpowder, that is to say of the nation’s munitions. One of the odder features of France’s economy concerned the supply of the mineral saltpeter, an essential element in gunpowder. Saltpeter, the chemical compound potassium nitrate, flourished on the walls of damp passages, cellars, and privies, and its collection had for a century or two been the monopoly of the Salpêtriers du roi (a “guild of gothic harpies,” as Gillispie describes them), who enjoyed the legal right to enter private dwellings to harvest it—to the vast irritation of householders. However, for the purposes of Louis XIV’s great wars, the supply proved inadequate, so Colbert farmed the industry out to private contractors. This led to very complicated trade-offs between the saltpetermen and the private “farmers,” and by Turgot’s day the whole system was near breaking down. He therefore opted for a form of seminationalization, transferring responsibility to a new institution, the Régie des poudres, directed by Lavoisier, under whose meticulous and reforming care the industry prospered.

By the time of the Revolutionary wars, saltpeter had become celebrated in France. Students were dispatched to Paris for instruction in saltpeter production and the fabrication of cannon, and at the height of the Terror, a grand saltpeter pageant was laid on. Borne aloft before marching students was a “draped and garlanded framework on which rested a basket of saltpeter.” Another group of students placed on the altar of the Fatherland a model of a Mountain made of saltpeter, which was topped by a saltpeter bust of Marat.


Nothing could better illustrate Gillispie’s chosen double subject, science and the “polity.” For it was Lavoisier’s work on gunpowder in his laboratory at the Arsenal in Paris that, more than anything else, led to the radical reform he was to make in pure science. In Gillispie’s words,

he had found that sulfur, and also phosphorus, far from losing weight when burned, as commonly supposed, on the contrary gain considerably, and that the increment comes from a prodigious quantity of air “fixed” during the combustion.”

This led him to the discovery of the nature of the atmosphere and the process of combustion, to the dethroning of the “phlogiston” theory, and ultimately to a general reform of the science of chemistry.

Scientists were conspicuous public figures in France at the time of the Revolution, as they were not, or not so much, in Britain. There were a number of them in the Legislative Assembly, and the experiment was tried of employing them in positions of politi-cal power. Thus Gaspard Monge, a Jacobin mathematician—a great teacher, though in the eyes of elegant hostesses a barbarian, and with a wide connection among men of science—was in 1792 appointed minister of the navy. He was not a success, being too much inclined to make well-meaning muddles. However, as advisers, scientists continued to exert great influence, both then and in the Napoleonic period.

There is nothing in previous history, or even in the British rule in India, quite to compare with Napoleon’s Egyptian expedition between 1798 and 1801. The intention was to make of Egypt not just a colony but a province of France, and the invaders were, supposedly, bringing not just liberty but the glories of French science. A Commission of Science and Arts was set up, led by the physicist Joseph Fourier, the chemist Pierre-Eugène Berthollet, and Monge, who was a close friend of Napoleon’s. Various other eminent scientists also joined the party, as well as a group of students from the École Polytechnique, who were expected to undertake road and bridge building and cartography.

In fact, the young polytechnicians found the ruins of ancient Egypt so enthralling that they neglected their practical duties in order to compile material for a scientific Description of Egypt, which appeared, subsidized by the state, over the next thirty years, in many folio volumes with magnificent engravings (in the tradition of the Encyclopédie). By 1801 the French had been driven out of Egypt, and Napoleon’s vision of empire, on the model of Alexander, quickly faded; thus this work would be the most tangible legacy of the venture.

A figure who looms large in Gillispie’s book is the Marquis de Condorcet, the mathematician and Revolutionary martyr, author of A Sketch for a Historical Picture of the Progress of the Human Mind. He was in a sense the inspirer of one of the most lasting achievements of the Revolutionary period, the founding (in 1795–1796) of the Institut de France, a corporation gathering all the cultural disciplines under one roof. It replaced the Académie des Sciences and other academies of the Old Regime (“academy” like “university” having become a taboo word), and, significantly, it dropped out of the scheme altogether the most famous academy of all, the Académie Française (which was not revived until 1815). The conception of the Institut de France followed the prescriptions of a famous Plan of Public Instruction drawn up by Condorcet, in which he argued that the culture of France ought from now on to be essentially scientific. Science would focus French minds on utility, and the beauty of scientific logic would be a panacea against medieval superstition.

Here we need, perhaps, to consider semantics. The French language does not employ the word “Science” in quite the way that English does, spelling it with a capital letter and giving it the status of a mysterious abstract entity. The equivalent in French is the more matter-of-fact sciences, in the plural. Nor does French possess the word “scientist”; but it has the word scientifique, and this is a term that in fact greatly fosters superstition (much as its English counterpart “scientific” does) and was a fatal snare for Condorcet. For he assumed that the exactness of the “exact” sciences could be extended to politics and social matters. He was a specialist in the mathematics of probability, and he held that probability analysis could be applied to elections and voting procedures. This throws much light on his conception of politics (and for that matter on Turgot’s too). For Condorcet did not think of legislative assemblies as designed to represent conflicting interests or factions; nor was this the purpose of voting. Instead, in a pro-foundly Jacobin notion, he thought of voting as a collective device for determining truth: the truth about where justice and humanity lay.


Gillispie was withering about Condorcet in his earlier volume, saying that he

made his career in the two worlds of science and politics, claiming importance in each on the basis of his promise or importance in the other, while exhibiting real effectiveness in neither.

He is a fraction kinder to him in the present volume, and in both volumes he seems happy to accept the general concept, dear to Condorcet, of “political science.” But here, I feel, we de-tect an obfuscating effect of the noun “Science.” It would really be better, though university departments of “political science” are long established, to admit that the concept is absurd. It is noticeable that in the middle of the last century, when “political science” was much in the air, its spokesmen, such as the American scholars Charles Merriam and David Easton, did not claim that it existed but only that it ought to exist, that the day for it had come, that the world was in desperate need of it.

One of the most celebrated achievements of the Revolutionary period was the metric system, and this forms the subject of a long, immensely detailed, and absorbing set piece in Gillispie’s new book. Under the Old Regime, the system of measurements of length and volume was chaotic, the same terms meaning different things in different parts of France. Legally the fundamental linear unit was the pied du roi (the King’s foot), which was contained six times in the cubit, or Toise de Paris, and the actual standard was a graduated iron bar mortised into the wall of the Châtelet in Paris, against which citizens might verify their measuring rods and rulers. (This Toise du Châtelet, since it was a unit of measurement, could not be said to have a measurement, for, evidently, it could not be measured by itself; though replicas could be, and were, made of it.)3

The Revolutionary government, however, wanting to give the world an example of the clearing up of ancient muddles, directed the Académie des Sciences to devise a new unit of measurement. Its introduction was to coincide with adoption of the decimal system, and the new unit was to be named the “meter,” being a length divisible and multipliable by tens. To lend authority to this new unit (which it was hoped the rest of the world would also adopt) it was decided to relate it to some feature of the natural world. This was not easy, but the Académie discerned two possibilities: first, the length of a pendulum which would, at a given latitude and at a given temperature, mark out seconds of time; secondly, a fraction of the length of the meridian, the distance on the earth’s surface between the equator and either pole (the “quadrant of the meridian”). It was decided to go for the second of these, and in 1792 two scien-tists, Jean-Baptiste Delambre and André Méchain, were dispatched to survey, by triangulation, the meridian from Dunkirk to Barcelona. They were equipped with a new and highly ingenious surveying instrument, the “repeating circles” of Jean Charles de Borda, and they pursued their task, which took six years, with heroic determination.

But as it must be remembered, this was an exceedingly fraught era. On August 10, 1792, King Louis XVI was suspended and took refuge in the National Assembly, and a mob invaded the Tuileries, massacring the Swiss Guards; and in the following January the King went to the guillotine. The country was full of wild rumors, and when Delambre or Méchain set up their “repeating circles” in a provincial town, a hostile crowd was likely to form around them. They found that the high points—church steeples and the like—from which previous surveyors had conducted their triangulations had often collapsed, and when they replaced them with wooden scaffolding, this might well be destroyed by vandals. Méchain, who was in charge of the southern part of the route, suffered a freak accident which put him out of action for six months, and then, France having declared war against Spain, he was interned in Barcelona. Also, to his horror, he discovered an inconsistency in the calculations he had already reported. He ought to have confessed to it, since it might have revealed some hitherto unknown scientific phenomenon, whereas he concealed it and devoted the rest of his life to correcting it.

There is much more of this in Gillispie’s book; it is almost a picaresque novel. Also, how beautifully clearly and simply he explains the principle of triangulation! Still, one cannot help wondering whether the entire enterprise, with its appeal to a “natural” standard, stands up logically. For it cannot be said that the meter was “based” on the length of the meridian. It had already been decided that it should represent one ten-millionth of the quadrant of the meridian, whatever that should turn out to be. That the meter should be defined by this length, when discovered, seems not much more than cosmological window-dressing. The great scientist the Marquis de Laplace told Napoleon that the chief benefit of the metrical system was not the destruction of feudal metrology but the division by tens; and in fact the meter was later defined as the distance between two marks on a metal bar kept at Sevres, near Paris.4

Gillispie, indeed, is perhaps a little overkind toward the period’s concepts of “nature” and the “natural.” He defends the notion of a “naturalistic metrology” and speaks of the “natural” classification of flowering plants launched by Antoine-Laurent de Jussieu, implicitly contrasting it, to its advantage, with the scheme originated by the Swedish botanist Carolus Linnaeus, which first described plants and animals in terms of genera or species. But one cannot help thinking that a “natural” classification is a will-o’-the-wisp. All one can fairly ask of a classification is that it should have a useful purpose and be carried out consistently.

The role of classification, or taxonomy, figures a good deal in Gillispie’s book. One of his central themes is the progression, over his chosen period, between two distinct modes of doing science: the “encyclopaedic” mode, which works by separating a complex sector of experience into its elements and arranging these “according to the natural connections inherent in the phenomena,” and the “positivist” mode, which is a study of how things work, an investigation of “actions in nature, and indeed on nature, rather than arrangements.” Once a science enters the positivist stage, he writes,

its goal is no longer a metaphysical quest for truth nor a rational theory purporting to represent physical reality. There is no longer any question of classifying information about the world in a manner consonant with the nature of things.

However, another way of describing this progression would be that of Michel Foucault, who regards only the earlier of these approaches as taxonomy and describes the zoologist Georges Cuvier’s new and revolutionary focus on function (by which he means something akin to Gillispie’s “positivist” science) as a movement away from taxonomy, “a passage from the taxonomic to the synthetic notion of life.”5 It moreover marks, for Foucault, the disappearance of the concept of “nature” altogether, along with the possibility of a “general taxonomy,” in which all the elements in the world would be arranged in a single order. Foucault is wonderfully persuasive on the subject of classification, and I feel drawn to his idea.

Another of Gillispie’s large generalizations is that the famous quarrels between scientists, like the acrimonious dispute about the nature of species between Cuvier and Étienne Geoffroy Saint-Hilaire (for a long time close friends and colleagues at the Museum of Natural History), may distract us from what is more important, their great and acknowledged dependence on one another and their cooperation in founding a discipline. In his Le Règne animal (1817), the first scientists cited by Cuvier are Lamarck (of whom later he would write a malicious éloge) and Geoffroy; and Geoffroy himself, in his Philosophie anatomique (1818–1822), in which he argues for the theory that all animals share a single plan of structure, proceeds, after a formal obeisance to Newton, to give generous thanks to Cuvier.

Moreover, when Cuvier’s and Lamarck’s successors took the trouble to republish their principal works in the 1830 and 1840s it was not from any sympathy for their best-known theories. It was simply, says Gillispie, because “nothing as complete, as clear, as descriptive, and as comprehensible as were the two great syntheses had appeared in the interval since their initial publication.” Cooperation, of the kind Gillispie is talking about, is, as he makes clear, much more basic to science than it is to literary scholarship—which is no doubt the reason why, conversely, scientists tend to have such a passion for priority.

It might seem a weakness in what Gillispie says about cooperation that it does not allow for scientific revolutions, for these will more probably originate with a single person. But on this question, though he was a close friend and colleague of the late Thomas Kuhn, famous as the author of The Structure of Scientific Revolutions (1962), Gillispie is reticent.

This Issue

May 26, 2005