IELTS Academic Reading Practice Test 419

Of the seven wonders of the ancient world, the Hanging Gardens of Babylon are the only one whose existence has never been confirmed. The pyramids of Egypt still stand, and the sites of the others have at least been located; the gardens have left no trace on which archaeologists can agree. They are usually imagined as a green mountain rising from the flat plain of Mesopotamia, with trees cascading over a series of terraces. Whether this image records a real building or a story repeated until it acquired the authority of fact has occupied historians for over a century.

The description comes not from Babylon but from the Greek and Roman world, several centuries after the supposed date of construction. Diodorus Siculus, writing in the first century BC, reports that the gardens rose on a series of terraces carried on vaulted arches of brick. Soil was laid on the roofs deep enough for large trees to grow. Strabo, writing a little later, says the terraces were raised on pillars and irrigated with water drawn from the Euphrates, which flows past the city. Philo of Byzantium, an engineer, included the gardens in his list of wonders and described the ingenious machinery used to lift the water to the highest level.

Classical authors also explain why such a structure should have been built. According to their accounts, King Nebuchadnezzar II, who ruled Babylon from 605 to 562 BC, married a princess from Media, a mountainous region to the north-east. The queen, whose name is given as Amytis, found the flat, hot landscape harder to bear than she had expected and longed for the forests and hills of her childhood. The gardens were designed to console her: an artificial mountain, planted with the trees of her homeland, rising above the mud-brick city.

Yet the Babylonians and Assyrians themselves, who recorded their kings' achievements, say nothing of any such garden. Thousands of clay tablets have been recovered from the ruins of Mesopotamian cities, including building inscriptions in which rulers boast of palaces and temples, but none contains a description that matches the classical accounts. Even more telling, perhaps, is the silence of Herodotus, the Greek historian who claimed to have visited Babylon and wrote at length about its walls, its river and its customs, yet never mentions gardens. For a long time scholars explained this absence by assuming that the relevant tablets were still buried or had crumbled away.

The first serious attempt to find the gardens on the ground was made by the German archaeologist Robert Koldewey, who excavated Babylon from 1899 until 1917. Near the royal palace his team uncovered a group of vaulted chambers with a series of deep wells. Koldewey proposed that this was the substructure of the gardens and the means by which they had been watered. The identification was accepted for decades and still appears in guidebooks. Most specialists now doubt it: the vaults look like a storeroom or an administrative building, and the wells could have served any part of the palace complex.

A more radical answer has been proposed by Stephanie Dalley, a scholar of Assyrian and Babylonian texts. In articles and a book published in 2013, she argues that the gardens never existed at Babylon at all, and that the Greeks attached them to the wrong city. In her view they were created by Sennacherib, king of Assyria from 705 to 681 BC, at Nineveh, his capital on the river Tigris, some 400 kilometres to the north. She points out that Sennacherib was a celebrated builder and gardener whose inscriptions describe parks filled with plants brought from distant provinces.

Her case rests on several kinds of evidence. One inscription from Sennacherib's reign records that bronze castings were made for raising water, which Dalley interprets as an early form of the screw later associated with Archimedes. Another describes a canal that carried water from the mountains to the city, and the remains of an aqueduct built to take that channel across a valley still survive at Jerwan.

Royal inscriptions of the period often boast of gardens and of the rare plants collected in them. A carved relief from Nineveh, meanwhile, shows a garden planted on a raised terrace, with an aqueduct visible in the background — precisely the arrangement that the classical writers described.

Other scholars remain unconvinced. They note that the relief may depict a royal park beside the river rather than a terraced garden, and that no excavation at Nineveh has revealed the massive vaulted substructure that the ancient descriptions require. The attribution of the water-raising screws is disputed, since the device is otherwise first described in the third century BC. And the classical tradition, for all its inconsistencies, is unanimous about the city: Strabo, Diodorus and Philo all place the gardens in Babylon. One of the earliest sources for that tradition was a Babylonian priest, Berossus.

The argument is unlikely to be settled by argument alone. What is needed is a discovery: a tablet from Babylon describing the planting and irrigation of a royal garden, or an excavation at Nineveh that uncovers the terraces their admirers described. Until then, the Hanging Gardens occupy an unusual position among the wonders of the world. Their image is familiar to millions of people, while their reality remains an open question. Work at the site has been hampered by modern building, by the diversion of the Euphrates and by years of conflict in Iraq.

Watch a column of ants crossing a pavement and you are seeing one of the most studied examples of collective problem-solving in the animal kingdom. A colony sends out foragers in many directions; those that find food lay a chemical trail on the way home; other ants follow it, reinforce it, and within minutes a busy highway has formed between the nest and the food. What makes this impressive is that no ant has surveyed the terrain. If two routes of different lengths connect the same two points, a colony will usually end up using the shorter one, even though no individual has compared them. Explaining how this happens occupied biologists for much of the twentieth century, and the answer turns out to involve chemistry, probability and a good deal of random error.

The key ingredient is a pheromone, a chemical secreted by the ant and deposited on the ground from a gland at the tip of its abdomen. Some species use more than one substance: one component attracts nestmates, another repels them, and the balance between the two shapes the traffic. The trail is not a fixed signal but a dynamic one. Each ant that follows a trail adds to it, so that a route in use becomes more attractive and draws still more traffic; this is an example of positive feedback, the same mechanism that produces fashions, bank runs and traffic jams. At the same time the pheromone is volatile: it evaporates, and a trail that nobody walks on for a few minutes simply fades away.

The classic demonstration was carried out in the late 1980s by a group led by Jean-Louis Deneubourg at the Free University of Brussels, working with Argentine ants. The researchers placed a food source on the far side of a bridge made of two branches, one considerably longer than the other, and recorded which route the ants used. At first the foragers chose at random. Within a few minutes, however, the traffic concentrated on one branch, and in most trials it was the shorter one. When the two branches were of equal length, the colony still committed itself to a single branch, but which one varied from trial to trial.

The explanation is statistical rather than clever. An ant travelling a short route completes its journey sooner and therefore makes more trips in the same period, laying pheromone each time. The shorter trail thus receives more deposits per minute, while the pheromone on the longer trail has more time to be lost through evaporation between visits. A small initial advantage, perhaps produced by nothing more than chance, is amplified until the other route is abandoned. This process, in which a random fluctuation is turned into a decisive outcome, is known as symmetry breaking, and it means that the colony's choice depends on its history as well as on the geography of the ground.

Two cautions follow. The first is that ants do not find the shortest path in any mathematical sense; they converge on one of the routes available, which is usually short but need not be the best. Colonies can be trapped by a well-established trail that leads somewhere inferior, and a change to the environment takes time to register. The second is that pheromones are only part of the story. Individual ants also use a sun compass, remember the position of landmarks and keep a running tally of how far they have walked, a form of dead reckoning that desert ants perform with remarkable accuracy. Researchers who fitted the insects with stilts, or trimmed their legs, found that they misjudged distances in proportion to the change, evidence that they count their steps.

Colonies also fail, sometimes spectacularly. Army ants, which hunt by mass raids rather than by laying trails to a fixed food source, can end up following one another in a continuous circle until they die of exhaustion — a phenomenon known as a circular mill. Nor is it fully understood how a colony balances the exploitation of a known route against the exploration that might reveal a better one. Part of the answer lies in the number of ants leaving the nest, part in the way individuals respond to crowding on a trail, and part in the tendency of some foragers to ignore chemical signals altogether. The relative importance of these factors is still disputed.

What began as field biology has become an engineering technique. In the early 1990s the computer scientist Marco Dorigo, then working in Italy, proposed a family of algorithms in which artificial 'ants' — simple programs, not models of real insects — search for solutions to problems of routing and scheduling. Each lays a simulated trail, and successful trails are reinforced much as real pheromone is. The approach, known as ant colony optimisation, is now used to plan delivery rounds, assign aircraft to gates and route data through telecommunications networks. Its designers are careful to say that it is inspired by ants rather than faithful to them: the point is not to imitate the insect but to borrow the principle that many simple agents, communicating through traces left in their environment, can solve problems that no single agent could.

For most of the twentieth century, pay in the private sector was treated as confidential. Contracts discouraged staff from discussing their salaries, and many employers regarded any comparison between colleagues as a breach of loyalty. That norm is now under pressure. Several governments require firms to publish figures on the gap between what men and women earn, and a few oblige employers to disclose the pay range for a job before the first interview. The logical endpoint of this trend would be to publish what every individual earns. Whether that would be a step towards fairness or a costly mistake is worth examining.

The case for openness begins with asymmetries of information. An employer negotiating with a hundred employees knows far more about the going rate than any one of them does, and that advantage translates into lower wages. Where staff cannot compare their pay with that of colleagues, discrimination is also harder to detect: two people doing the same job may be paid differently without either of them realising it. In the United States, a law passed in 1935 protects most private-sector employees who discuss their pay with one another, and the board that enforces it has repeatedly ruled that blanket bans on such discussion are unlawful. Supporters of full transparency argue that secrecy is, in effect, a subsidy to the employer.

The legislation that exists so far is narrower than that. In Britain, employers with 250 or more staff have had to publish annual statistics on the gender pay gap since 2017, and Denmark has required larger firms to produce gender-segregated pay figures for considerably longer. The European Union adopted a directive in 2023 that goes further, obliging employers to tell applicants the starting pay or pay range for a post and to report their gaps; member states have until 2026 to bring their laws into line. Economists who study the gender pay gap generally agree that such disclosures have narrowed it, though modestly, and largely by restraining the growth of pay at the top rather than by raising it at the bottom.

Evidence about the effect on individuals comes from an unusual natural experiment. When a newspaper in California obtained and published the salaries of employees of the state university system, researchers were able to observe what happened next. They found that staff whose pay fell below the median for their department became markedly less satisfied with their jobs and more likely to look for another one, while those paid above the median did not become measurably happier. The lesson is not that people dislike information; it is that they compare themselves with those closest to them — the colleagues in the next office — rather than with the population at large or with figures in a national report.

Employers' organisations raise more practical objections. Publication of individual salaries, they argue, invites resentment between colleagues, makes recruitment harder in a competitive market and removes the flexibility that small firms rely on when cash is short. Privacy matters too: a salary reveals a great deal about a person's circumstances, and an employee may have good reasons for not wanting a neighbour, a former partner or a rival company to know what they earn. Some of these concerns could be met by publishing ranges rather than exact figures, but the underlying tension remains.

A subtler objection concerns the direction in which published pay may move. When differences between colleagues become visible, employers may respond not by raising the lowest pay but by holding down or even cutting the pay of those near the top, so that the whole structure is compressed. Employees, for their part, may be reluctant to accept extra responsibility if the reward for doing so is open to public inspection and therefore harder to justify. Some labour economists have argued that transparency can depress average earnings even while it makes the distribution fairer. The evidence on this point is limited and comes mostly from the public sector, where pay is already constrained by budgets and by politics.

A middle course exists, and it is the one most often adopted. Rather than publishing individual salaries, an employer can publish the graded scales on which those salaries rest, together with summary figures for the gaps between groups. Public-sector organisations have done this for decades: a teacher or a nurse can look up the grade and step that determines their pay, and knows that colleagues on the same grade receive the same amount. The advantage is that the rules become visible without any individual being exposed. Employees can see whether the system treats them consistently, and can challenge a decision that appears to break the rules.

My own view is that the case for publishing every individual salary is weaker than its advocates suggest. The evidence that it raises pay is thin, and there is a real risk that it changes behaviour — hiring, promotion, even ordinary conversation between colleagues — in ways that are hard to foresee. The case for publishing the structure of pay is much stronger. A company that publishes its grades and its gap figures is telling employees, and applicants, how it values different kinds of work, and that is a matter of basic fairness. Beyond that, whether an individual's salary is visible to colleagues seems to me a decision best left to those concerned.