Phasing the Material Revolutions
Discussion
Phasing the Material Revolutions
Jeff Holden:
The first transformation
'For most of the Stone Age, humans shaped matter but never changed it. Flint was knapped, wood carved, hide scraped: the material at the end was the material at the beginning. The first great exception to become a technology of civilization was fired clay. Put clay in fire and something irreversible happens: it becomes a new substance, hard and durable, that did not exist in nature’s inventory.2 Ceramic vessels made storage cheap; cheap storage made surplus administrable; administered surplus made cities. And when civilization needed a medium durable enough to carry its first writing, the medium was fired clay. We can read the administrative records of Uruk five thousand years later because the same command of matter that made the granary jar made the tablet.3
Bronze was the first engineered alloy: copper deliberately alloyed with tin, two ores that rarely occur together, which is why the Bronze Age was also civilization’s first continent-spanning supply chain.4 Iron then won not because it was better (early iron was often inferior to good bronze) but because its ore is everywhere. What gated the Iron Age was furnace practice and the black art of carburization: a pure command-of-matter problem.5 When smiths solved it, metal stopped being an aristocratic monopoly. The cheap iron axe and plowshare opened the forests and heavy soils of Europe, and the population followed the plow.6
Glass, or the Scientific Revolution’s material precondition
My favorite version of the thought experiment: imagine civilization before and after our ability to make and shape glass.
Before, the human sensorium ended at the unaided eye, and a scholar’s working life ended when his vision did. Spectacles (ground glass, northern Italy, the 1280s) added decades to the productive lives of scribes, jewelers, and weavers past forty.78 Then, as furnace craft refined over centuries from Murano to the Dutch and English workshops, glass delivered the instruments: the telescope in 1608, the microscope soon after, the prism, the barometer tube, the thermometer, the sealed retort.9 Glass gave science two things at once—a sensorium and a laboratory. Very little in nature is detectable by unaided human senses, and no chemistry can be watched inside a clay pot or an iron kettle; it took an inert, transparent, heat-tolerant vessel to make reactions visible. Galileo’s moons, Hooke’s cells, Leeuwenhoek’s animalcules, Newton’s spectrum, Torricelli’s vacuum: it is difficult to name a pivotal observation of seventeenth-century science that did not pass through a piece of shaped silica.10 Some historians have gone further, arguing that Europe’s uniquely deep glass tradition is part of why the Scientific Revolution happened there and not elsewhere.11 However far one takes that, the sequence is not in dispute: centuries of furnace craft first, instruments second, revolution third.
Hold that thought about glass. It returns.
Steam waited seventeen centuries for iron
In the first century AD, Hero of Alexandria built the aeolipile: a hollow metal sphere, pivoted on the tubes that fed it steam from a cauldron below, that spun on its axis from the recoil of steam escaping through two bent nozzles—a working reaction turbine, seventeen centuries early.12 The idea of steam power is two thousand years old. What did not exist was the material command to do anything with it: no metallurgy that could contain working pressure, no seals, no way to bore a cylinder true, no fitted piston to convert pressure into force. For seventeen centuries the idea sat idle, a temple novelty, not for want of imagination but for want of iron worked to tolerance.
Then the matter arrived, and the transition came in a single burst. Newcomen’s atmospheric engine of 1712 was leaky and desperately inefficient, but it was buildable at blacksmith tolerances, so it was built.13 James Watt had the decisive insight, the separate condenser, in 1765, and then carried it in his head for a decade, unable to realize it, because his engine demanded a precision Newcomen’s never did: a piston that sealed against steam.14 What stood between the idea and the engine was John Wilkinson’s boring mill, patented in 1774 and turning out cylinders by 1775, which could bore a large cylinder round and straight. Watt reported that a fifty-inch Wilkinson cylinder erred by no more than the thickness of an old shilling.15 By 1776 the first commercial Watt engines were running.16 The Industrial Revolution began, on the ground, not when someone imagined a better engine but when someone could machine a true cylinder.
And precision, once achieved, revealed itself to be the rare material capability that manufactures more of itself. Maudslay’s screw-cutting lathe made accurate screws that made more accurate lathes; Whitworth’s standardized threads made parts interchangeable; interchangeable parts made mass production.17 It is worth pausing on how this is even possible, because “a machine making something more accurate than itself” sounds like it should violate a conservation law. The resolution is that accuracy is not copied from a master; it is generated. Lap any two plates against each other and they will mate, but mating only proves they are complementary curves, one convex and one concave. Lap three plates against one another in rotation, and the only geometry consistent with all three pairings is a true plane: flatness created from nothing, with mathematics itself as the master.18 Likewise, a nut riding on dozens of threads of an imperfect screw moves more uniformly than any single thread is cut, because its many contacts average their random errors (the law of large numbers, cast in metal), and Maudslay closed the loop by measuring each screw’s residual error and correcting it into the next.19 Self-checking geometry, averaging, and measurement feeding correction: that is how each generation of machine tools generates accuracy beyond its own, and why precision compounds. Machine tools are the machines that make machines, and from Wilkinson’s bore forward, every generation of them has been built by the generation before.
The press was a metallurgy project
We file printing under the history of information. It was executed as a project in matter. Gutenberg was a goldsmith, and his revolution was made of metal: the hardened steel punch, the copper matrix, the adjustable hand mould, and a lead-tin-antimony alloy formulated to cast sharp and resist shrinking, an engineered type-metal that remained the standard for five hundred years.20 And the press waited on a second material capability: paper, which took roughly a thousand years to travel from Han China through the Islamic world to European mills.21 The reason is arithmetic. A Bible printed on vellum consumed the skins of some 170 calves; on paper, the marginal book collapsed in cost.22 The information regime changed when, and only when, the matter was ready.
Bread from steel
In 1798 Thomas Malthus stated the trap: population, when unchecked, grows geometrically, while the food supply grows at best arithmetically, so scarcity is not a policy failure but a mathematical destiny.23 For the following century he looked right about the past and wrong about the present only on borrowed time. The nineteenth century escaped him by spending down one-time material windfalls: fresh plowland in the Americas, guano shipped from Peruvian islands, nitrate mined from the Chilean desert.24 Deposits, not income. In 1898 William Crookes stood before the British Association and made the Malthusian forecast specific and dated: the wheat-eating world was exhausting its fixed nitrogen, the deposits were finite, and civilized nations stood “in deadly peril of not having enough to eat.” His prescribed escape was explicit: fix nitrogen from the air. He assigned the task to chemistry.25
The chemistry community understood the problem completely; understanding fed no one. In 1909 Fritz Haber synthesized ammonia from air on a tabletop; the chemistry was solved.26 Scaling it was a matter problem, and a brutal one: the reaction demands hundreds of atmospheres at hundreds of degrees, and under those conditions hydrogen invades carbon steel and embrittles it from within. Reactors burst. Carl Bosch’s answer was a materials invention, a double-walled reactor with a soft iron liner that surrendered to the hydrogen while an outer steel shell carried the pressure, and by 1913 the first plant was running at industrial scale.27 Four years from tabletop to world-changing plant: that is what lockstep looks like when the matter problem yields. Today roughly half the nitrogen atoms in a human body have passed through a Haber–Bosch reactor, and roughly half of humanity is fed by crops it fertilizes.28 Malthus was not refuted by an argument; he was refuted by a pressure vessel. The Malthusian ceiling turned out to be neither divine nor mathematical. It was a statement about a civilization’s command of matter, and when the command moved, the ceiling moved with it.
The metal that fell to earth
Aluminum is the most abundant metal in the Earth’s crust, and in the 1850s it was more precious than gold; the story is told that Napoleon III reserved aluminum cutlery for his most honored guests and left mere gold for the rest.29 As late as 1884 it still traded at the price of silver. Then the Hall–Héroult process of 1886 turned it from a precious metal into a commodity within a generation.30 Seventeen years later the Wright Flyer flew, its homebuilt engine kept light enough to fly by a crankcase cast in the new metal.31 Every era of flight since has been named, internally, for its material: wood and fabric, aluminum monocoque, titanium, the single-crystal superalloy turbine blades that gate the jet engine by surviving in gas streams hotter than their own melting point, carbon composite.32 Aviation is materials science with wings, and each new regime of flight waited on a new regime of matter.
December 1947, and the province we still live in
The transistor is remembered as the founding event of the Information Age, which causes us to misfile it. It was a triumph of material command: germanium purified to unprecedented cleanliness, impurity atoms introduced in controlled numbers, two gold contacts set a hair’s breadth apart.33 Everything since has been nearly eighty years of ever-finer arrangement of atoms: silicon refined to nine nines of purity, single crystals pulled flawless from the melt, features now specified in counts of atoms.34
And glass returns, as promised. In 1970, Corning drew the first low-loss optical fiber, and the internet became, at its physical layer, light traveling through glass.35 Why glass won is worth spelling out, because the incumbent was copper, and copper was losing on both dimensions that matter. The bandwidth a channel can carry rises with the frequency of its carrier, and light oscillates tens of thousands of times faster than the highest-frequency signals copper can carry; and electrical signals attenuate in coaxial cable so quickly that Bell’s highest-capacity coax lines needed an amplifier roughly every mile, while sufficiently pure glass carries light for tens of kilometers between repeaters.36 The operative word is pure. In 1966, Charles Kao showed that the crippling losses in existing optical glass came not from the glass itself but from its parts-per-million impurities, which converted the future of telecommunications into a materials-purification problem.37 Corning solved it four years later. The Scientific Revolution’s material, refined three centuries further, gated a second information regime.
Even now, the largest technology buildout in history strains not against algorithms but against matter: advanced packaging, stacked memory, power delivery.38 The gate has not moved.
The pattern, stated plainly
In every case the sequence is the same. The idea arrives early, often absurdly early. Hero had the steam. Watt had the physics. Crookes had the warning and Haber had the chemistry. The knowledge was in hand; the matter was not. And each time the material capability finally arrived (Wilkinson’s bore, Bosch’s reactor, Hall’s cell, Bell Labs’ germanium), the transition followed within years, not centuries. That is the precise meaning of lockstep: the command of matter is not one input among many. It is the pacing item. Civilization does not advance at the speed of its ideas; it advances at the speed at which its ideas can be embodied.
There is a fair objection: haven’t some revolutions been pure thought? Calculus, Maxwell’s equations, the fast Fourier transform, the modern algorithm—none required a new ability to shape atoms. True, within a regime ideas run free, and gloriously. But the regime changes, the moments when civilization jumps to a new physical medium, have been, without exception, command-of-matter events. The historical record does not contain a counterexample."
< " So if you want to forecast a civilization, do not ask what it believes or even what it knows. Watch what it is learning to do with matter." >
Source
* Article: The Lockstep: Civilization and the Command of Matter. Jeff Holden.