Millstone: IELTS Academic Reading practice test

Grain is not food until it has been ground, and until about two thousand years ago the grinding was done by hand. A quern is two circular stones, the upper turned against the lower by a handle, and the flour that results has to be produced daily because it will not keep. Estimates based on surviving querns and on modern reconstructions suggest that feeding a single household took several hours of turning every day, and in almost every society the turning fell to women. Against that background, a machine that would grind while nobody stood over it was not a convenience. It was the return of a substantial part of the waking life of half the population.

The Romans knew how to build one. Vitruvius, writing shortly before the birth of Christ, describes a vertical waterwheel connected by a pair of gears to a horizontal millstone, which is the arrangement that would still be in use seventeen centuries later. The most spectacular surviving example stands at Barbegal in southern France, where sixteen wheels were arranged in two rows down a hillside, each pair driving its own stones, in what was probably a single industrial flour works. Yet mills of this kind remained the exception across the empire. Where labour is cheap and plentiful there is less pressure to mechanise, and the incentive to replace hand grinding with machinery was weaker than it would later become.

In medieval Europe that incentive grew, and the mill spread with remarkable speed. The English survey of 1086 known as Domesday Book lists more than five thousand mills in a country of perhaps a million and a half people, which works out at one for every few hundred inhabitants. The commonest design turned an undershot wheel, whose paddles dip into the current and are pushed by it. An undershot wheel needs no engineering works beyond a race, but it takes its power from the speed of the stream alone and is markedly less efficient than the alternative. The overshot wheel, into whose buckets water is delivered from above, takes its power from the weight of the falling water as well, and can extract two or three times as much from the same flow. The price is that the water has to be raised to the top of the wheel, which means building a dam, a holding pond and a channel to bring the supply along the contour.

Wind arrived later. The first windmills in north-western Europe appear in the records towards the end of the twelfth century, the earliest English reference being to a mill at Weedley in Yorkshire in 1185. The problem that had to be solved is that wind, unlike a river, changes direction. The first answer was the post mill, in which the entire body of the mill, machinery and stones and all, is balanced on a massive upright post and swung round by a tail beam until the sails face the wind. It works, but a building that must be pushed round by hand cannot be made very large. The tower mill solved that by fixing the body in masonry and rotating only the cap that carries the sails, at the cost of having to lift every stone and shaft up the tower during construction.

A machine that is expensive to build and cheap to run invites a particular kind of ownership, and in much of medieval Europe the mill belonged to the lord of the manor. Tenants were not merely encouraged to bring their grain to it; they were obliged to, and obliged to leave behind a fixed proportion of the flour, known as multure, as payment. Hand querns were therefore not simply an inconvenient alternative but an illegal one. The most celebrated quarrel of this kind took place at St Albans, where in 1331 the abbot confiscated the querns his tenants had been using and had the stones set into the floor of his own parlour, where they remained as a paved reminder for the next fifty years.

By then the waterwheel had long since stopped being a machine only for grain. The same rotating shaft, fitted with cams that lift and drop heavy hammers, will beat woollen cloth in water to thicken it, a process called fulling that reached England in the twelfth century and moved the cloth industry out of the towns and into the valleys where the streams were. Other mills drove saws, crushed ore, worked the bellows of iron furnaces, pulped rags for paper and ground pigments. Each of these began as a modification of the same drive train, which is why a river valley in medieval Europe was less a source of water than a line of power sockets.

The end came from two directions. Steam freed a factory from having to sit beside falling water, and the roller mill, which passes grain between sets of steel rollers rather than crushing it between stones, produced a whiter flour faster and separated the bran and the oily germ so cleanly that the result keeps for months rather than weeks. Wind and water mills survived longest where neither coal nor capital was easy to come by. But the knowledge did not disappear with the buildings, because the gearing, the shafts, the governors and the whole vocabulary of transmitting rotary power from one place to another had been worked out at the mill, and the factory simply inherited it.

It is tempting to see all this as a story about machines, and it is more interesting as a story about where effort goes when it is no longer needed. The hours released by the mill did not vanish; they were spent on other work, and the households that had ground their own grain became households that bought flour and sold something else. Every labour-saving device carries that second effect with it, and the second effect is usually larger than the first, though it is always harder to see.

Few findings in psychology have travelled as far from their origin as the ten thousand hour rule. In its popular form it holds that the difference between the expert and the rest is accumulated practice, that the quantity required is roughly ten thousand hours, and that anybody prepared to put in the time may expect to arrive. The idea has an obvious appeal, since it replaces an accident of birth with an act of will, and it has been repeated in enough management books, sports columns and school assemblies to have acquired the status of common knowledge. It rests, however, on a single study of a single conservatoire, and on a reading of that study which its own author spent the rest of his life disputing.

The study, published in 1993 by Anders Ericsson and two colleagues, compared violin students at a Berlin music academy who had been sorted by their teachers into three groups by ability. The researchers reconstructed how much solitary practice each student had done since first picking up the instrument. The best group had accumulated, on average, about ten thousand hours by the age of eighteen; the middle group rather less; the least accomplished group considerably less again. That is the whole of the celebrated result. It is a comparison of group averages, not a threshold; no individual student was found to have crossed a line at ten thousand hours; and the paper says nothing about whether practice alone would have carried a student from the bottom group to the top.

What Ericsson did insist on was a distinction that the popular version drops entirely. Practice, in his sense, is not playing through pieces one can already play. It is deliberate practice: effortful, usually solitary work aimed at the specific component of performance one is currently worst at, guided by feedback precise enough to show whether each attempt was better or worse than the last, and continued past the point where it stops being enjoyable. Most people, in most activities, never do this at all. They reach a level that serves their purposes and then repeat it for years, which is why thirty years of driving does not make an expert driver.

The serious challenge to the rule came in 2014, when Brooke Macnamara and her colleagues pooled the results of every study they could find that had measured both practice and performance. Practice, they reported, accounted for about a quarter of the variation between individuals in games such as chess, about a fifth in music, rather less in sport, and almost none at all in professions such as law and medicine. The authors were careful to say that this makes practice important. It also makes it very far from decisive, and it leaves three quarters of the difference between a strong chess player and a weak one to be explained by something other than the hours each has put in.

Ericsson's reply was that the meta-analysis had measured the wrong thing. Most of the studies it pooled recorded experience of an activity rather than deliberate practice of it, and the two are not remotely the same; a player who has spent four thousand hours in casual games has not been doing what a student with a coach and a list of weaknesses has been doing. He also pointed out that the domains in which practice explained least were precisely those with no established training method, no agreed measure of performance and no immediate feedback, which is to say the domains in which deliberate practice is barely possible. The disagreement was never fully resolved, and both positions have reasonable versions.

What the two camps do agree on is that domains differ, and the dividing line is a useful one. Chess, music and competitive swimming share three features: the task is stable, the standard of a good outcome is agreed, and the feedback is immediate and unambiguous. Under those conditions, hours accumulate into skill. Where the task keeps changing, where success is judged years later, and where the same decision may be right or wrong depending on events nobody controls, experience buys far less; studies of forecasters, clinicians making rare diagnoses and senior managers repeatedly find practitioners of thirty years' standing performing no better than those of five. The uncomfortable implication is that in some fields the expert is a person who has had more opportunities to form confident habits, rather than better ones.

It is also worth noticing that the opposition between practice and talent is less clean than either side's slogans suggest. Whatever is left over when practice has been accounted for is not a single thing called ability. It includes the age at which a person began, the resources of the household that paid for lessons, physical characteristics that matter enormously in some activities and not at all in others, and, most awkwardly, the disposition to tolerate several thousand hours of difficult and unrewarding work. If the capacity to practise in the way Ericsson describes is itself unevenly distributed, then practice and talent are not two competing explanations but two descriptions of the same tangle.

My own view is that the useful part of the idea was never the number. What Ericsson's work established, and what survives every subsequent criticism, is that a great deal of what looks like a gift is in fact a construction, assembled over years by methods that can be described and taught. That is a genuine contribution, and it has changed how music, sport and surgery are taught for the better. The damage was done by the promise that came attached to it. A rule that tells everybody they may arrive is also a rule that tells those who worked hard and did not arrive that the fault was theirs, and it is difficult to think of a less accurate or less kind thing to tell them.