In July 1785, Thomas Jefferson, then American Ambassador to France, paid a visit to the dungeon of the Château de Vincennes. Its three-metre-thick walls had previously imprisoned Diderot and the Marquis de Sade. Now, however, it housed the workshop of a gunsmith, Honoré Blanc, and a dozen assistants. As Jefferson watched, Blanc sorted into bins the pieces of 50 musket flintlocks: ‘tumblers, lock plates, frizzens, pans, cocks, sears, bridals, screws and springs’. From the parts in the bins, Blanc assembled several working gunlocks.
What the astonished Jefferson saw has become commonplace in the 20th century. Guns, cars, ordinary mechanical artefacts are all now made by assembly from interchangeable parts. If one part fails, we can readily replace it with another. In the 18th century, however, interchangeability seemed a distant dream, as utopian as political democracy had seemed before the American Revolution. The final assembly of a mechanical artifact was a laborious process of filing parts until they fitted well enough. In consequence, if a gun failed on the battlefield, it was impossible simply to replace the broken part with another, identical, one.
The modern reader may wonder why Jefferson sought out the gunsmith. In school, we are taught that modernity was brought about by revolutions of two kinds: political (as in America and France) and industrial. If the connections between the two are discussed at all, it is only at the most superficial level. The thesis of Ken Alder’s Engineering the Revolution is that the links go deep: into the subterranean 18th-century processes from which the French Revolution emerged; into the underpinnings of the military successes that almost gave Napoleonic France control of Europe; into the technical heart of guns themselves.
Some of the connections can be seen in the lens of a single day, 10 September 1792. It was a day when the French Revolution could easily have been snuffed out. Advancing towards Paris was a Prussian army, accompanied, as it happens, by Goethe. The Prussians expected a walk-over. They were admired and feared throughout Europe for inspired generalship and brutal discipline, and they thought they faced only a Revolutionary rabble. As the armies met, however, Goethe heard an unnerving sound – ‘humming ... gurgling ... whistling’ – that he was to say signalled ‘the beginning of a new epoch’. It was the sound of the cannonballs of the massive artillery barrage that saved Revolutionary France.
The architect of the cannonade of Valmy – and the patron of Honoré Blanc – was a man who had died three years before the battle: Jean-Baptiste Vaquette de Gribeauval. Born in 1715 to a recently ennobled family, the mathematically talented de Gribeauval rose rapidly via the artillery, a career path open to young men of modest means but one that was vaguely disreputable for an aristocrat, in that it involved the cowardly dealing out of death at a distance. A temporary move to the service of the Habsburg emperor served him well: at Schweidnitz in 1761-62 he held off a large force of besieging Prussians until Frederick the Great came in person to lead the assault. From 1763 onwards, de Gribeauval struggled for control of the French artillery service, becoming its first inspector-general in 1775.
The entrenched artillery faction that de Gribeauval was up against was led by a father and son, both called Jean-Florent de Vallière. The Vallièrists represented ‘the corporate world of absolutist France’. A traditional cannon was a triumph of handiwork. Cast in individual moulds with earthen cores to form the hollow, no two cannon were identical. Typically, French cannon were adorned with elaborate embossing, including the inscription that described their purpose: ultima ratio regis, the last argument of the king. These heavy, slow-firing, weapons were designed for a paradoxically civilised form of warfare: the set-piece siege of a fortress. They were moved slowly into position, often by barge; the farmers on whose land they were deployed were sometimes paid rent to compensate for the inconvenience. There was a consensual rule of thumb to tell defenders how long they should hold out (48 days) before an honourable surrender.
Although there was a developing science of ballistics, the Vallièrists felt that it could not replace the experienced ‘eye’ of the artillerist. Innovation in offensive weaponry was suspect. One supporter of the Vallièrists anticipated the arguments of opponents of the nuclear arms race when he pointed out that such innovation increased the inhumanity of war without contributing to security, since improvements made by one monarch would quickly be copied by others. De Vallière fils even pointed to what we have come to call the social construction of facts. Noting the dependence of experimental tests of new weapons on the skills of the experimenters, he criticised the scientistic notion that ‘one cannot and must not contest the facts (les faits).’
Against the Vallièrist worldview de Gribeauval mobilised professionalism, new technology, a new view of war, a thoroughgoing rationalisation of production, implacable standardisation and technical drawing. Though generally aristocratic in their origins, artillerists increasingly were educated at schools dedicated to the teaching of science (arguably the first such schools in Europe) and tested by written examination, especially in mathematics. Their education was designed to develop their pride in ‘the cold and fastidious details of their craft’ and distanced them from the decadence of Versailles. Gradually, they became professionals, ‘a self-organising, self-disciplining, and self-promoting social body whose loyalty was to the state, rather than to the king’.
At the technical core of de Gribeauval’s reforms were new, smaller, lighter cannon, shorn of all adornment, cast solid in a standard, reusable mould and hollowed out by a horizontal boring machine that de Gribeauval adopted in the 1760s. To the Vallièrists, these ‘mass-produced’ cannon were inferior, inaccurate weapons, but their great virtue was the relative ease with which they could be moved. Cannon could accompany manoeuvring troops, in a pale prefiguring of Blitzkrieg, rather than be restricted to use in set-piece sieges.
Quality control over production was no longer to rest on the subjective judgment of officials, often influenced, the Gribeauvalists suspected, by excessively close relationships, personal and monetary, to those whose work they were examining. For example, over-large cannonballs had long been detected by checking to see whether they would pass through a gauge of the same diameter as the barrel of the cannon that would fire them, but whether a ball was too small was left as a matter for the experienced ‘eye’. The Gribeauvalists introduced a further gauge with a slightly smaller diameter through which the ball must not pass, plus special test equipment to cope with departures from sphericity, and insisted that all gauges be made centrally at the arsenal at Strasbourg.
The greatly intensified use of gauges and the associated, new, notion of ‘tolerance’ (maximum allowable deviation from standard size) were tools of depersonalisation and uniformity. Another tool was drawing. Sketches had long been used to convey the ‘idea’ of a new device (as in Leonardo’s notebooks), and perspectival drawing had developed to help the human viewer ‘see’ the object being represented. The reforming engineers of 18th-century France developed a different approach to representation. Their preferred form, owing much to the geometer and political radical Gaspar Monge, was what has become modern technical drawing, with its separate plan view and two elevations, and its abandonment of easy subjective interpretability in favour of faithful representation of length. The appeal of this visual objectivity was that it allowed the separation of conception and execution. Engineers could draw precisely what they wanted built, and, in principle, leave workers no discretion in how to turn technical drawings into uniform artefacts.
The commitment of French military engineers to a ‘mechanical ideal’ was an aspect of their conflict with both groups with whom they interacted in the process of production: the entrepreneurs from whom the state purchased weapons and who subcontracted the work to artisan workforces; and those workforces themselves. The engineers tended to feel that both entrepreneurs and artisans were wily and venal, preferring private gain and an easy life to predictable, high-quality production. Though the engineers had legal authority on their side, and could in principle have recalcitrant workers imprisoned, the struggle was far from one-sided. Entrepreneurs, some of whom were very wealthy, could appeal over the heads of the engineers direct to the Court, or could simply refuse to supply weapons to the state (commercial alternatives were open to them, notably the market in the Americas, where technical standards were less demanding). Artisans could riot, and their ready access to weapons meant that revolts by gunsmiths were a threat that could not be taken lightly.
The production of interchangeable parts offered the promise of reduced costs and the practical advantages of apparent ease of repair on the battlefield. Crucially, however, it also offered the prospect of liberating weapons production from constant, ad hoc, local negotiations. That made it attractive to engineers in their pursuit of rationalisation, but equally unattractive to those whose livelihoods depended on these negotiations. Blanc was in the Vincennes dungeon in part for his own protection: his patrons in the artillery knew well that his technical work was politically sensitive.
Initially, the Revolution promised to free the project of interchangeable parts production from Ancien Régime impediments. Valmy brought only a temporary respite: by the summer of 1793, the new Jacobin Government was under extreme pressure. Robespierre and his colleagues turned to two military engineers – Lazare Carnot and Claude-Antione Prieur – and together the technocrats and the Jacobins sought to construct a massive new centre of weapons production, the Manufacture of Paris, with, at its heart, an experimental workshop, the Atelier de Perfectionnement, the goal of which was the perfecting of arms production, specifically the development of machine tools to make interchangable parts, tools that would be usable by ‘all sorts of citizens’, not just skilled artisans.
The technological vision of deskilled uniformity was, Alder points out, as utopian as Robespierre’s Republic of Virtue, and both collapsed after Thermidor. Blanc, however, found a relatively tranquil provincial haven at Roanne, and began large-scale production. His costs were high; he was still dependent on the skill of his workers; and uncomfortably much final hand-filing was still needed. But the dream of interchangeability was becoming real: at the start of the 19th century, the Roanne factory was producing ten thousand gunlocks and two thousand muskets a year, at a time when Blanc’s American counterpart, Eli Whitney, was struggling to produce much smaller numbers.
With the most famous artillerist of them all, Napoleon Bonaparte, in power, success might have seemed assured. But pragmatism and compromise with existing vested interests were now the order of the day, not state-imposed uniformity. Napoleon, for example, abandoned the metric system, symbol of Revolutionary, rational standardisation, complaining that it had ‘thrown local markets into disarray’. He promoted his crony and old commanding officer, Jean-Jacques-Basilien Gassendi, to the head of the artillery service. Though he was the descendant of a famous natural philosopher, Gassendi had been lukewarm about de Gribeauval’s reforms, and was opposed to nationalised production.
Against Blanc’s atomistic vision of identical, interchangeable parts, Gassendi offered an alternative view. ‘Harmony’, both technical and social, was key. Imperfections in the parts did not matter if the weapon worked as a whole. State inspectors should, therefore, not be too rigorous, and they should realise that both entrepreneurs and skilled artisans had a legitimate place in a harmonious social system of production. Entrepreneurs deserved their profits, and artisans’ tacit knowledge should be respected. Interchangeability was socially dangerous and a technological chimera. Against Blanc’s experimental demonstrations of close-to-perfect interchangeability, Gassendi staged his own tests, which seemed to show that as few as 3 per cent of Blanc’s gunlocks were interchangeable. Nor could the test of experience – interchangeability on the battlefield – come to the aid of Blanc’s supporters. Blanc was not allowed to engrave his factory’s name on his locks, so they could not easily be distinguished from ordinary gun-locks with no claim to interchangeability.
The facts might be ambiguous, but Gassendi had Napoleon’s ear. Roanne was closed in 1807, and, with Gassendi’s supporters writing the historical record, interchangeability was deemed a failure. ‘Technological amnesia’, as Alder puts it, set in.
Dead and forgotten in France, interchangeability became ‘the American system of manufactures’. Jefferson’s letter home describing what he had seen in the dungeon of Vincennes was one spark that led to the American interchangeable-parts rifles that dazzled the British public in the 1851 Crystal Palace Exhibition; to the Colt revolver and Singer sewing-machine; to Henry Ford, the assembly line and 20th-century mass production; and to the swamping of Nazi Germany by the mobilisation of American productive capacity in the Second World War. The shift, Alder notes, was more than a matter of geography. Though the Army was intimately involved in the history of interchangeability in the United States, the defining impetus was to come from profit-seeking, cost-minimising capitalists, not perfection-seeking state engineers.
Even at nearly five hundred pages, Alder’s book sometimes seems too compressed, with several over-arching themes – his ‘technological’ interpretation of the French Revolution; the relationship of technology to society; the nature of modernity – which jostle for attention and occasionally break the narrative flow. Nevertheless, Engineering the Revolution is a triumph. It deserves to be read widely, and not just as an inquiry into the origins of modern France, with its distinctive combination of capitalism, relatively small-scale production, and state-led technocracy. As Alder says, ‘until we understand the development of science and technology from the point of view of its human creators – and take seriously the claims of their opponents – we will never come to grips with our own technological life. Nor will we understand how things might be different.’
Send Letters To:
London Review of Books,
28 Little Russell Street
London, WC1A 2HN
Please include name, address, and a telephone number.