IELTS Academic Reading Practice Test 24

By the first decade of the twentieth century, the South Pole was one of the last great geographical goals left on Earth. Antarctica had first been sighted only in 1820, and its interior, a vast plateau of ice rising to around 3,000 metres, remained almost entirely unknown. Earlier British expeditions had pushed far inland: in 1909 Ernest Shackleton's party turned back just 180 kilometres short of the Pole, forced to retreat because their food supplies were running out. Two expeditions that arrived on the continent in early 1911, one led by the British naval officer Robert Falcon Scott and the other by the Norwegian Roald Amundsen, would finally settle who would be first.

Amundsen had not originally intended to go south at all. He had planned to drift across the Arctic Ocean in the Fram, a ship designed to withstand the pressure of sea ice, and perhaps to reach the North Pole. In 1909, however, two Americans, Frederick Cook and Robert Peary, each claimed to have reached the North Pole first, and Amundsen quietly changed his plans. He kept the new destination secret, even from most of his crew, until the Fram stopped at Madeira in September 1910. Scott, who was already on his way to Antarctica, learned of the Norwegian challenge from a brief telegram.

The two leaders prepared in very different ways. Amundsen had spent two winters among the Netsilik Inuit of northern Canada during an earlier voyage through the Northwest Passage, and he had learned from them how to dress in loose fur clothing and how to handle teams of dogs. He relied entirely on dogs and skis, and several of his men were chosen because they were expert skiers. His plan was ruthlessly practical: as the loads lightened, many of the dogs would be killed to feed the others and the men. Scott, by contrast, placed his trust in a mixture of motor sledges, ponies, dogs and men pulling sledges themselves.

Amundsen built his base, which he named Framheim, on the Ross Ice Shelf at the Bay of Whales, about 100 kilometres closer to the Pole than Scott's hut at Cape Evans. During the Antarctic autumn of 1911 both parties laid depots of food and fuel along their intended routes. Amundsen made a false start in September, when extreme cold forced his party back, but on 20 October he set out again with four companions, four sledges and around fifty dogs. Scott's main party followed on 1 November, after his motor sledges had gone ahead; within days both machines had broken down and been abandoned.

Amundsen's route took him up a previously unknown glacier, which he named after Axel Heiberg, one of the expedition's financial backers, and onto the high plateau. Travelling fast on skis while the dogs hauled the loads, his party navigated across the featureless plateau using a sextant and distances recorded by wheels mounted on the sledges. They reached the Pole on 14 December 1911. They spent three days there taking measurements to make sure of their position, left a small tent containing letters, one of them addressed to Scott, and turned for home. They arrived back at Framheim on 25 January 1912, all five men in good health.

Scott's experience was very different. Once the ponies had been shot, as planned, and the dog teams sent back to base, his men hauled their sledges themselves up the Beardmore Glacier and across the plateau. On 17 January 1912 Scott and four companions reached the Pole, only to find the Norwegian tent and flag. The return journey became a disaster.

Edgar Evans died in February after a series of falls, and in March Lawrence Oates, suffering from severe frostbite, walked out of the tent into a blizzard rather than slow down his companions. Scott, Edward Wilson and Henry Bowers continued until a storm trapped them in their tent, where they died around the end of March.

Their bodies were found by a search party on 12 November 1912, only about 18 kilometres from a large supply depot. Beside them were Scott's diaries and around 16 kilograms of rock samples collected on the Beardmore Glacier, including fossils of an extinct plant called Glossopteris. The searchers collected these records, collapsed the tent over the bodies and built a cairn of snow above them. These fossils later helped to show that Antarctica had once been joined to other southern continents and had enjoyed a much warmer climate. Scott's journal, published in 1913, turned him into a national hero in Britain, and for decades his death was presented as a model of courage and self-sacrifice.

Later historians have been less generous. Some have argued that Scott's reliance on ponies and motor sledges, and his doubts about dogs, reflected poor judgement, while others have stressed the role of the weather. The atmospheric scientist Susan Solomon, analysing temperature records in a study published in 2001, concluded that conditions on the Ross Ice Shelf in March 1912 were far colder than normal and that this was a major cause of the tragedy. Most historians now accept that both preparation and bad luck played a part. Amundsen's success, meanwhile, owed much to his willingness to learn from people who had lived in polar environments for centuries.

When the first grey wolves were carried into Yellowstone National Park in January 1995, the species had been absent from the park for around seventy years. Government hunters and park staff had deliberately eliminated wolves in the early twentieth century, when predators were regarded as a threat to livestock and to the elk and deer that visitors came to see. By the 1920s no breeding packs remained. Their return followed decades of scientific argument and political controversy, and it has become one of the most closely studied wildlife reintroductions in history. The legal basis for the project was the Endangered Species Act of 1973, which required the government to plan for the recovery of threatened species, but the plan was strongly opposed by many ranchers in the surrounding states, who feared losses of cattle and sheep. Between 1995 and 1997, 41 wolves captured in Canada and Montana were released into the park.

The wolves multiplied quickly. With abundant elk to hunt and no competition from other wolves, the population inside the park reached around 170 animals in the early 2000s before settling at about 100, divided among roughly ten packs. Wolves in the park usually hunt in packs, and elk make up most of their prey in winter, although they also take deer, bison and smaller animals. Elk numbers declined sharply over the same period. The herd that spends the winter on the park's northern range numbered around 17,000 in the mid-1990s, but fewer than 6,000 by 2019. Biologists led for many years by Douglas Smith, who directed the park's wolf project, have followed individual wolves and packs since the release, producing an unusually detailed record of how a returning predator affects its prey.

What made the Yellowstone wolves famous, however, was a further claim: that by reducing and frightening the elk, they had set off a chain of effects through the whole ecosystem. Ecologists call this a trophic cascade. In a series of studies, William Ripple and Robert Beschta reported that willow, aspen and cottonwood trees along some streams had begun to grow again after decades in which elk browsing had kept young shoots short. They proposed that elk were avoiding places where they could easily be ambushed, such as narrow valleys and riverbanks, a pattern sometimes called the landscape of fear. Recovering trees, in turn, would stabilise riverbanks, shade the water, and provide food for beavers and nesting sites for songbirds.

This account proved enormously popular. A short online video arguing that wolves had even changed the course of rivers was watched by millions of people, and the Yellowstone story became a standard example in textbooks and in public debates about rewilding. For some scientists, however, its very popularity was a warning sign. Writing in 2014, the ecologist Arthur Middleton argued that the version of the story told to the public had become far simpler than the science behind it, and that much of the evidence pointed to a messier reality. Critics of this kind did not deny that wolves mattered. Their concern was that a single dramatic narrative might be applied to other ecosystems where conditions, and the evidence, were quite different.

One problem concerns the cause of the elk decline. Wolves were not the only predators in the park: grizzly bears and cougars, whose numbers also grew, kill large numbers of elk calves each spring. Severe droughts reduced the food available to elk, and hunters killed many animals that migrated out of the park in winter. Separating these causes is difficult, because they all acted at the same time. Another problem concerns the landscape of fear. When Matthew Kauffman and his colleagues measured how much young aspen was eaten in areas where the risk of meeting wolves was high or low, they found no difference, suggesting that fear alone was not protecting the trees. Where aspen and willow have recovered, a reduction in elk numbers seems the more likely explanation.

The streams themselves add a further complication. Beavers were common on the northern range in the early twentieth century, and their dams raised the water table and created wetlands in which willows thrived. When the beavers disappeared, partly because elk had eaten the willows they depended on, streams cut deeper channels and the land beside them dried out. In an experiment in which fences kept elk away from willow plants, Kristin Marshall, Tom Hobbs and David Cooper found that the protected willows grew tall only where water levels had also been raised artificially, imitating the effect of a beaver dam. Removing elk was not enough, by itself, to restore the original ecosystem.

None of this means that the wolves have had no effect. Their presence has changed where and how elk feed, provided carcasses for scavengers such as ravens, eagles and bears, and contributed to the fall in elk numbers that has allowed some vegetation to recover. Bison, which wolves rarely hunt, have increased in number. The lesson that many ecologists draw is that ecosystems rarely respond to a single change in a simple, predictable way. Removing a top predator for seventy years produced changes that the return of that predator cannot simply reverse. For the managers of other parks and reserves, this is an important caution: reintroducing a predator can be valuable, but it should not be expected to restore a landscape on its own.

Every wave of new technology has brought fears that machines will make human workers redundant. In the early nineteenth century, English textile workers known as Luddites smashed the machinery that threatened their livelihoods. In 1930 the economist John Maynard Keynes warned of what he called technological unemployment, although he also predicted that his grandchildren's generation might need to work only fifteen hours a week. The arrival of artificial intelligence systems capable of writing text, analysing images and answering questions has revived these fears in a particularly acute form. My argument is that the evidence so far justifies concern, but not the predictions of mass joblessness that dominate public discussion.

Much of the alarm can be traced to a single study. In 2013 Carl Benedikt Frey and Michael Osborne of the University of Oxford estimated that 47 per cent of jobs in the United States were at high risk of being automated, perhaps within a decade or two. Their method was to assess whole occupations, asking whether the work involved could be performed by computers. Later researchers pointed out a weakness in this approach: most occupations consist of many different tasks, only some of which can be automated. When economists at the OECD analysed the tasks that individual workers actually perform, they found that only around 14 per cent of jobs in its member countries faced a high risk of automation, although a further third were likely to change significantly.

History also offers some reassurance. When automated teller machines spread through American banks from the 1970s, many expected bank tellers to disappear. Instead, because the machines made it cheaper to run each branch, banks opened more branches, and the number of tellers continued to grow for many years, although their work shifted from handling cash towards dealing with customers. More broadly, the economist David Autor and his colleagues have estimated that around 60 per cent of American workers in 2018 were employed in occupations that did not exist in 1940. Technology destroys particular tasks, but it also creates demand for new kinds of work that are difficult to imagine in advance.

There are, however, reasons to think that artificial intelligence may differ from earlier technologies. Most previous waves of automation replaced physical or routine work, such as weaving, assembling or calculating. Generative AI, by contrast, performs tasks that were once considered the preserve of educated professionals: drafting documents, writing computer code and summarising research. An analysis published by the International Monetary Fund in 2024 estimated that around 40 per cent of jobs worldwide, and some 60 per cent in advanced economies, were exposed to AI in some way. Exposure, it must be stressed, is not the same as replacement; in many cases the technology may complement workers rather than substitute for them.

Early studies of AI in the workplace support this distinction. In a study of more than 5,000 customer-support agents at a large software company, Erik Brynjolfsson, Danielle Li and Lindsey Raymond found that access to an AI assistant raised the number of problems resolved per hour by 14 per cent on average. Strikingly, the largest gains, around 34 per cent, went to the least experienced and least skilled workers, while the most skilled improved very little. Similar results have emerged from experiments with professional writing tasks, in which AI tools reduced the time taken and narrowed the gap between stronger and weaker writers. Such findings suggest that AI may sometimes reduce, rather than widen, inequality within occupations.

Yet complementing workers is not the only possibility. The economists Daron Acemoglu and Pascual Restrepo studied the spread of industrial robots in the United States between 1990 and 2007 and found that local labour markets more exposed to robots experienced lower employment and wages. Each additional robot per thousand workers, they estimated, reduced the proportion of the population in work by about 0.2 percentage points. In their view, the effect of a technology depends heavily on how firms choose to use it: to replace workers in order to cut costs, or to make workers more productive and create new tasks. There is, they argue, nothing inevitable about the direction of technological change.

Even if total employment holds up, the transition may be painful. Workers whose skills are made less valuable rarely move smoothly into the new jobs that technology creates, since those jobs may be in different places, require different qualifications or pay less. Studies of factory closures have repeatedly found that displaced workers suffer lasting losses of income, and older workers have tended to fare worst. The central question is therefore not whether there will be work in the future, but who will bear the costs of change and how quickly people can adapt.

This suggests where policy should focus. Rather than preparing for a world without work, governments would do better to make retraining genuinely accessible to adults, to strengthen support for people who lose their jobs, and to encourage the use of AI in ways that extend human skills rather than replace them. Schools and universities, too, will need to emphasise abilities that complement machines, such as judgement, communication and the capacity to keep learning. Predictions about artificial intelligence have often been wrong, in both directions. Humility about the future is not a reason for complacency, but it is a reason to prepare for several possible outcomes rather than one.