IELTS Academic Reading Practice Test 43

By the early nineteenth century, London had become the largest and busiest port in the world, and the Thames was crowded with vessels loading and unloading cargo. Crossing the river, however, was a slow business. Upstream, a handful of bridges linked the two banks, but in the docklands to the east any new bridge would have had to be extremely high to let the masts of sailing ships pass beneath it, and its approaches would have swallowed valuable land on either side. Thousands of dock workers relied instead on small ferry boats. A tunnel beneath the river was an obvious alternative, but an attempt in the first decade of the century, in which the Cornish engineer Richard Trevithick was involved, had been abandoned after repeated flooding. The soft, waterlogged ground beneath the Thames seemed impossible to dig through safely.

The man who eventually succeeded was Marc Isambard Brunel, a French-born engineer who had left France during the Revolution and settled in Britain in 1799. According to a well-known account, his key idea came from studying the shipworm, a mollusc that bores through the timbers of ships while protecting its soft body with a pair of hard shells at its head. In 1818, together with Thomas Cochrane, he patented a tunnelling shield: a structure that would support the earth at the face of a tunnel while men excavated it in small, protected sections.

The shield that was eventually used, constructed by the engineering firm of Henry Maudslay, was a massive iron framework pressed against the face of the excavation. It was divided into 36 separate cells, arranged in three tiers, each of which held a single miner. The face itself was covered by heavy wooden boards, which the miners removed one at a time; each man dug away a few inches of earth behind a board and then replaced it before moving on to the next.

When the entire face had been excavated in this way, the shield was pushed forward by powerful jacks, and bricklayers working immediately behind it lined the newly exposed walls with brick. This arrangement meant that only a small area of unsupported ground was exposed at any moment, greatly reducing the danger of a sudden collapse, although it could not prevent water from forcing its way through the loose gravel and silt above.

Work began in 1825 at Rotherhithe, on the south bank. To create the vertical shaft from which the tunnel would start, Brunel had a brick cylinder about 15 metres across constructed on the surface. Soil was then removed from beneath it, allowing the structure to sink gradually into the ground under its own weight. His son, Isambard Kingdom Brunel, who was barely twenty years old, soon became resident engineer, and he spent long periods underground directing the work, often sleeping only a few hours at a time.

Progress was painfully slow, and conditions in the tunnel were appalling. The river above was heavily polluted with sewage, and the foul water that seeped into the workings made many of the men seriously ill. The tunnel flooded several times. In May 1827 the river broke through, and the younger Brunel descended in a diving bell to inspect the hole in the riverbed, which was then blocked by dropping bags of clay onto it. A far worse flood on 12 January 1828 killed six workers, and Isambard himself was badly injured and needed months to recover. By that time the company had exhausted its funds, and the workings were sealed off behind a brick wall.

For seven years the half-finished tunnel stood idle, and critics mocked it as 'the Great Bore'. Work resumed only after the government agreed in 1834 to lend the project £247,000 through the Treasury. Brunel's original shield, badly corroded after years of exposure to river water, had to be dismantled, and a new and heavier one was installed, and digging restarted in 1836. Even then the work took several more years. The tunnel finally reached the north bank at Wapping in 1841, and it opened to the public on 25 March 1843, almost two decades after the first shaft had been begun.

The opening was a sensation. Tens of thousands of visitors are reported to have paid a penny each to walk through the tunnel during its first day, and it quickly became one of London's most popular attractions, with stalls selling souvenirs set up in the arches between its two passages. Commercially, though, the tunnel was a disappointment. It had been intended for horse-drawn traffic, but the large spiral ramps needed to bring carts down to it were never built because the money had run out, and so it was used only by pedestrians. Within a few years its novelty had faded, and it acquired a poor reputation.

In 1865 the tunnel was sold to the East London Railway, and trains have run through it since 1869; today it forms part of the London Overground network. It is recognised as the first tunnel known to have been built successfully beneath a navigable river. More importantly, the principle of the shield that Brunel developed, much refined over the following two centuries, remains the basis of the tunnel-boring machines used to construct modern underground railways around the world.

For as long as farmers have depended on rain, people have dreamed of controlling it. Until the middle of the twentieth century such hopes belonged largely to folklore and showmanship, and travelling 'rainmakers' toured drought-stricken districts promising downpours in return for a fee. Today, however, the World Meteorological Organization reports that operational weather modification programmes are under way in more than fifty countries. Cloud seeding, the deliberate introduction of particles into clouds to alter the way in which they produce precipitation, has become a routine activity in regions as different as the western United States, China and the Arabian Peninsula. Yet after almost eighty years of practice, scientists are still debating how well it actually works.

The principle rests on a peculiarity of cloud physics. Many clouds contain droplets of water that remain liquid at temperatures well below freezing, a state known as supercooling. These droplets need a tiny solid particle, or nucleus, on which to freeze, and in the atmosphere suitable particles can be surprisingly scarce. Once ice crystals do form, they grow rapidly at the expense of the surrounding droplets, eventually becoming heavy enough to fall as snow, which may melt into rain on its way to the ground. This mechanism is often called the Bergeron process, after the Swedish meteorologist Tor Bergeron, who helped to describe it in the 1930s. If such nuclei are missing, however, a cloud may drift across a parched landscape without releasing much of its moisture at all.

The modern science began almost by accident. In July 1946, Vincent Schaefer, a researcher at the General Electric laboratory in Schenectady, New York, was studying ice formation in a chilled chamber when he added dry ice, or solid carbon dioxide, to lower the temperature. A cloud of ice crystals immediately appeared. On 13 November that year, Schaefer scattered roughly three kilograms of crushed dry ice from an aircraft into a supercooled cloud, and snow was seen falling from it near Mount Greylock in western Massachusetts. Shortly afterwards, his colleague Bernard Vonnegut found that crystals of silver iodide, whose structure closely resembles that of ice, could also act as nuclei. The discovery attracted intense interest from government agencies and the armed forces, and General Electric was soon collaborating with them in a research effort known as Project Cirrus. Because it can be burned in generators on the ground or released from flares attached to aircraft, silver iodide remains the most widely used seeding agent.

A second technique targets clouds that are too warm for ice to form. In hygroscopic seeding, fine particles of salt are released into the rising air beneath a cloud, where they attract water vapour and swell into unusually large droplets. These collide with smaller droplets and merge with them, accelerating the growth of raindrops. The method is favoured in hot, arid countries such as the United Arab Emirates, which has run a national seeding programme since the 1990s and sends aircraft to fire salt flares into the towering convective clouds that build up over its mountains in summer.

The fundamental difficulty lies in proving that seeding has had any effect. Precipitation varies enormously from one storm to the next, and nobody can know how much snow a seeded cloud would have produced if it had been left untouched. Early programmes often announced impressive results that later analysis could not confirm. Randomised experiments, in which comparable clouds are either seeded or left alone by chance, have produced mixed findings. A nine-year pilot programme in the mountains of Wyoming, which ended in 2014, estimated that seeding could raise precipitation from suitable winter storms by around 5 to 15 per cent. Many storms were unsuitable, however, so the overall gain to the snowpack was considerably smaller. Critics have also warned that seeding one cloud might simply deprive areas further downwind of rain they would otherwise have received, although the Wyoming study found little evidence of such an effect.

The clearest evidence so far came from an experiment in the mountains of Idaho in early 2017. Researchers released silver iodide from an aircraft flying a zigzag course and then used radar to follow the resulting bands of snow as they formed and fell. Because natural snowfall was negligible on several of the days studied, the team could attribute the snow directly to seeding and calculate how much had fallen. The results, published in 2020, confirmed that the technique works under the right conditions. They also highlighted its limits, however: the effect is easiest to detect precisely when there is little natural snow, and the volumes of water produced were modest.

Cloud seeding also raises questions that are not purely scientific. During the Vietnam War, the United States secretly seeded clouds over the Ho Chi Minh Trail in an attempt to prolong the monsoon season and make supply routes impassable. The disclosure of this programme contributed to an international convention, opened for signature in 1977, that prohibits the hostile use of environmental modification techniques. More recently, after record rainfall flooded Dubai in April 2024, seeding was widely blamed on social media, although meteorologists pointed out that the storm had been forecast days in advance and was far too large to have been created artificially. Concerns have also been raised about the environmental effects of silver iodide, although the quantities released are very small.

In the late 1960s and early 1970s, the psychologist Walter Mischel and his colleagues at Stanford University carried out a series of experiments with young children at the university's nursery school. The procedure was simple. A child of about four or five was seated at a table and shown a small treat, such as a marshmallow or a pretzel. The researcher explained that he needed to leave the room, and that if the child waited until he returned, they would receive two treats; if they could not wait, they could ring a bell and eat the single treat at once. Some children gave in within seconds. Others sang to themselves, covered their eyes or turned their chairs away from the table, and managed to hold out for a quarter of an hour or more.

The experiments were originally designed to explore the strategies that children use to resist temptation, and on that subject they proved genuinely illuminating. Children who distracted themselves, or who were encouraged to imagine the marshmallow as a fluffy cloud rather than something sweet to eat, waited far longer than those who stared at the reward. But the studies became famous for a different reason. When Mischel and his collaborators contacted some of the original participants years later, they found that those who had waited longer as preschoolers tended, as adolescents, to be described by their parents as more competent. They also achieved higher scores on the SAT, the test used for admission to American universities. The 1990 paper reporting this link, by Yuichi Shoda, Mischel and Philip Peake, turned a laboratory curiosity into one of the best-known findings in psychology.

It is easy to understand why. The study appeared to identify a simple, measurable quality, self-control, that could be observed in a four-year-old and that predicted success more than a decade later. Popular books, business seminars and parenting guides seized on it, and some educators built entire programmes around teaching young children to delay gratification. Yet the evidence on which this edifice rested was surprisingly thin. The follow-up samples were small, and the children had been drawn from a nursery school attended mainly by the sons and daughters of Stanford academics and staff, a group that was hardly representative of society as a whole.

The first serious challenge concerned what the test actually measures. In a study published in 2013, Celeste Kidd, Holly Palmeri and Richard Aslin at the University of Rochester showed that children's decisions were strongly influenced by whether they trusted the adult involved. Before the marshmallow task, half of the children in their study dealt with an experimenter who promised them better art materials and then failed to provide them; the other half dealt with one who kept her word. Those in the reliable condition waited, on average, about four times as long as the others: roughly twelve minutes compared with three. For a child whose experience has taught them that promises are often broken, the researchers argued, eating the marshmallow straight away is not a failure of willpower but a rational choice.

A second challenge came in 2018, when Tyler Watts, Greg Duncan and Haonan Quan re-examined the long-term predictions using a far larger and more varied group: around 900 children who had taken part in a national study of child development in the United States. The ability to wait at the age of four and a half was indeed associated with later achievement in reading and mathematics, but the association was considerably weaker than earlier reports had suggested. Once the researchers took account of factors such as family background, the home environment and early cognitive ability, it shrank to almost nothing. In other words, the test seemed largely to be detecting advantages that children already possessed, rather than an independent trait that caused later success.

None of this means that self-control is unimportant, and it would be a mistake to swing from uncritical enthusiasm to outright dismissal. Interestingly, a study led by Stephanie Carlson, also published in 2018 and co-authored by Mischel himself, compared children tested in the 1960s, the 1980s and the 2000s. Contrary to the expectations of most of the adults whom the researchers surveyed, the more recent generation waited longer, on average about two minutes longer than children in the 1960s. Whatever the explanation, the finding suggests that the capacity to wait is shaped by a changing environment rather than fixed at birth. The researchers proposed several possible explanations, including the expansion of preschool education and changes in children's capacity for abstract thought, but they stressed that further research in more diverse populations was needed.

What the history of the marshmallow test illustrates, I would argue, is less a failure of the original research than a failure of interpretation. Mischel's early work was careful and modest in its claims about how children manage temptation; it was the later leap to a single predictive number, and the eagerness of popular culture to embrace it, that went too far. Children are not simply born patient or impatient. How they behave in a room with a sweet reflects their age, their experience of adults, the security of their home and a great deal else besides. If the test teaches us anything, it is that a child's choices make sense only in the context in which they are made.