IELTS Academic Reading Practice Test 465

Around 1200 BC the eastern Mediterranean was more closely connected than ever before. The Mycenaean kingdoms of mainland Greece, the Hittite empire of Anatolia, the New Kingdom of Egypt and the island of Cyprus all exchanged goods, letters and skilled craftsmen. Bronze, the metal that gave the age its name, was an alloy of copper and tin. Copper was plentiful in Cyprus, but tin had to be carried from Central Asia or Afghanistan, so that every kingdom depended on trade routes it did not control.

The palaces at the heart of these states did not merely consume; they organised production. Scribes writing in a syllabary now called Linear B recorded flocks, grain, linen, bronze and wine in inventories kept for a single year and then thrown away. Palaces gathered olive oil and wool from surrounding villages, paid craftsmen in rations and shipped the surplus abroad. A merchant at Ugarit on the Syrian coast might sell wheat to the Hittites and Cypriot copper to Egypt, and return with perfume and glass. Because the whole system rested on a handful of royal households, it was efficient but fragile.

Then, within about a century, most of it vanished. The Hittite capital, Hattusa, was abandoned around 1180 BC and never again served as a royal seat; its archive, written on clay tablets, was left where it lay. Ugarit was destroyed and never reoccupied. At Pylos in south-western Greece the palace burned, and the fire, which might have destroyed everything, baked its tablets hard and preserved them, so that archaeologists can still read the final records of a vanished state.

Egypt survived, though it was badly shaken. The pharaoh Merneptah drove off a coalition of foreign peoples in about 1208 BC, and Ramesses III defeated a larger force in the Nile Delta around 1175 BC. The reliefs carved on the walls of Ramesses III's temple at Medinet Habu show the defeated groups, whose names are given as the Peleset, Tjeker, Shekelesh, Denyen and Weshesh, among others. The familiar phrase 'Sea Peoples' is a modern label: the Egyptian texts never use it, and the label was popularised by the French scholar Gaston Maspero in the nineteenth century.

For decades this was the standard account: invaders destroyed the kingdoms of the eastern Mediterranean and then disappeared. Few archaeologists now accept it as the whole story. Some of the groups named by the Egyptians had probably been serving as mercenaries in Mediterranean armies for years; others may have been migrants displaced by drought or warfare. The attackers, in other words, may have included people fleeing the very collapse they are accused of causing.

Attention has since moved to the climate. Pollen and sediment cores from lakes and from the sea floor have been read as evidence of a long dry period across Anatolia, the Aegean and the Levant. Isotope data from caves in Israel point the same way. Such records cannot be dated precisely enough to tie a particular dry year to a particular ruined city, and a single drought would not have destroyed a kingdom by itself. Sustained water shortage, however, would have undermined royal authority, pushed up the price of grain and made the redistribution system impossible to maintain.

Earthquakes are a second candidate. Faults run through the region, and excavators at several sites have uncovered walls that appear to have fallen outwards, as masonry does when the ground moves beneath it. The geophysicist Amos Nur and the archaeologist Eric Cline have suggested that a series of large earthquakes struck over some fifty years, leaving states too weakened to rebuild before the next shock arrived. The proposal is attractive but hard to test. Dating is seldom precise enough to prove that two distant cities fell in a single event, and many ruins show no seismic damage at all.

One influential account, published by Eric Cline in 2014, treats these candidates not as rivals but as a cascade. Cline describes a 'perfect storm' in which drought strained royal treasuries, earthquakes destroyed palaces, displaced populations turned to raiding, trade routes closed, and every failure made the next one more likely. The economic argument is the most compelling. Once the network broke, supplies of tin failed. A state that could not obtain tin could not make bronze, and a ruler who could not arm soldiers or equip farmers had lost the foundation of royal power.

The consequences were lasting. In Greece, Linear B disappeared along with the palaces, and literacy was not regained until an alphabet arrived from the Phoenicians some four centuries later. Populations fell, settlements shrank, and gold and ivory largely vanished from the archaeological record. Egypt survived but exhausted itself, and the New Kingdom came to an end by about 1069 BC. Elsewhere the pattern differed sharply: Assyria suffered comparatively little and later built a far larger empire, while iron, easier to find and to work, gradually replaced bronze for tools and weapons.

No single cause is likely to be found. The collapse of the Bronze Age world is better understood as the failure of an interdependent system, in which no kingdom could survive the breakdown of the others for long. That is what makes the period so unsettling to study: the states that fell were not primitive, and their rulers were not foolish.

When a product succeeds with very little advertising, the explanation usually offered is word of mouth. The phrase is used loosely, but it describes something that can be studied: people telling other people about a purchase, and those listeners changing what they do as a result. Marketing researchers have spent more than half a century trying to model the process, partly to forecast sales and partly to decide where money should be spent. Their findings are more interesting, and much less certain, than the advertising industry's enthusiasm for 'buzz' suggests.

The foundational work came from rural sociology rather than from business. In the 1950s a young researcher named Everett Rogers built on a classic study of why farmers in Iowa had adopted hybrid seed corn at such different speeds. Some sowed it within a year or two of its release; others held out for a decade. From such studies he built a general theory of the diffusion of innovations, published in 1962, which divided adopters into innovators, early adopters, an early majority, a late majority and laggards. Plotted over time, the number of new adopters usually traces a bell-shaped curve, and the cumulative total an S-shaped one.

Rogers also argued that the fate of an innovation depends less on its merits than on how people perceive five of its attributes. These are its relative advantage over whatever it replaces, its compatibility with existing habits, its complexity, whether it can be tried on a small scale, and how visible its benefits are. A device that neighbours can be seen using has an advantage that a hidden improvement, however clever, does not. Diffusion, in his account, is above all a social process, carried on through conversations between people who trust one another.

Quantitative modelling followed. In 1969 Frank Bass proposed an equation describing the first purchase of a new consumer durable as the product of two forces. One, conventionally labelled p, stands for outside influence such as advertising and press coverage; the other, q, stands for influence from people who already own the product, in other words word of mouth. Fitted to the past sales of televisions, refrigerators and air conditioners, the model reproduced the familiar slow start, rapid growth and eventual plateau with surprising accuracy. Versions of it are still used to forecast demand.

A different line of research began with a study of the 1940 presidential election in the United States. Paul Lazarsfeld and his colleagues found that voters were influenced far more by other people than by radio or newspapers. The idea that became known as the two-step flow held that media messages reach most people only indirectly, filtered through friends and acquaintances who pay close attention. Those intermediaries, later called opinion leaders, were not necessarily wealthy or powerful; they were simply interested. The finding gave word of mouth a central place in explanations of how ideas and products travel.

Popular accounts often conclude that adoption depends on a small number of highly connected people, variously called influentials, opinion leaders or influencers. Duncan Watts and Peter Dodds tested this assumption with computer simulations in which artificial people passed messages to one another. They found that whether a message spread widely depended less on the presence of exceptional individuals than on the willingness of ordinary people to pass it on. In their model, large cascades began among a critical mass of easily influenced people, and campaigns aimed at the influential were not reliably more effective than random targeting. The finding remains contested, not least because a simulation is not the world.

Structure matters as much as the people in it. In a much-cited paper published in 1973, Mark Granovetter interviewed men who had recently found work through personal contacts and discovered that most had heard of their jobs from people they saw only occasionally rather than from close friends. His explanation was that weak ties bridge otherwise separate social groups, and therefore carry information that a tight circle of friends already possesses. This has since been applied to everything from job hunting to the spread of rumour. A closely knit group can be an efficient echo chamber and a poor channel for novelty.

Online platforms have made some of this visible and much of it measurable: reviews, ratings, shares and reposts leave a trail that earlier researchers could only dream of. Yet the evidence is harder to interpret than it looks. People who write reviews are not a random sample of buyers. A product with many positive reviews may be popular because it is good, because it is cheap, or because a campaign asked its buyers to post. Separating the effect of what people say from the effect of the product itself requires experiments that firms rarely run and seldom publish.

What can be said with confidence is that talk between buyers matters and that it can be encouraged. Free samples, trial offers and referral schemes all try to start conversations by putting a product into someone's hands. The evidence suggests that people talk most about things that are noticeable, that are easy to describe and that make them look good to the person they are telling. What remains genuinely uncertain is how much of the success of any particular product can be attributed to conversation, and whether attempts to engineer it work better than simply making something worth talking about.

Few pieces of management advice are repeated as often as the instruction to hire for attitude and train for skill. It sounds generous and hard-headed at the same time, and it is popular in industries where staff turnover is high and the tasks themselves can be learned in a few weeks. The claim deserves close examination, because it bundles together two quite different questions: whether a person's dispositions predict how well they will do a job, and whether those dispositions can be identified reliably at interview. A decision based on a twenty-minute conversation needs the second to be true.

The slogan is most often heard in retail, hospitality and call centres. There the specific skills involved can genuinely be taught in days, and vacancies are expensive to leave open. Employers in these sectors reasonably point out that a cheerful and reliable worker who knows nothing about the product can be trained, whereas a knowledgeable worker with no interest in customers cannot easily be changed. Formal qualifications, they add, often say very little about how someone will behave when a queue is long and a customer is angry.

There is real evidence behind the first half of the claim. Reviews of selection research have consistently found that general mental ability is the strongest single predictor of performance across a wide range of jobs. Measures of conscientiousness add a modest amount of predictive power on top of it. The same reviews tend to report that structured interviews predict performance considerably better than the informal conversations that many managers still prefer. In a structured interview every candidate is asked the same job-related questions and scored against a fixed scale.

Perhaps the most quoted illustration comes from outside academia. An internal study at Google, reported in 2011, examined the qualities that separated the company's most effective managers from its least effective ones. Technical expertise came last among the eight behaviours identified. What mattered more were coaching, clear communication, empathy and an interest in the team's success. The study is often cited as proof that character beats competence, but it shows something narrower: among people already selected for high technical ability, technical ability no longer distinguishes the best managers.

Critics of attitude-based hiring raise two objections. The first is that the term is too vague to bear the weight employers place on it: 'attitude' is not a single quality but a bundle of habits, and the instruments used to measure it are weak. Candidates asked to rate their own conscientiousness can guess what the employer wants, and can present themselves in a favourable light without much difficulty. Such questionnaires may also say little about how well someone will actually do the job. A strong impression at interview is a poor substitute, since interviewers tend to reward people who resemble themselves and often form judgements within the first minutes of a meeting.

The second objection is economic. Firms that pay for training create skills that a competitor can use without having paid for them. For this reason, economists have long predicted that employers will under-invest in general training and concentrate on skills specific to their own operations. Hiring people who already possess the required skills is, from this point of view, a way of passing the cost of training to somebody else — a previous employer, a college or the individual. Read in this light, 'hire for attitude' may describe not a theory of performance but a reluctance to spend.

Labour-market evidence points in a similar direction. Work by the economist David Deming has found that, in the United States since the 1980s, employment and wages have grown fastest in occupations that demand social skills as well as cognitive ones, while routine work has declined. If that is right, the qualities employers label attitude are becoming more valuable rather than less. It does not follow that they can be identified at the door; it follows that they have to be developed, like any other competence.

Management researchers who study how competence develops at work make a related point: the distinction between attitude and skill is largely artificial. Much of what looks like a personal quality — patience with a difficult customer, care in checking a delivery, willingness to ask for help — is produced by doing the job alongside people who already do it well. In this view, employing someone for their manner and then providing no structured training is not a strategy. It is an omission that places the burden of learning on the new recruit and on colleagues who were never asked to teach.

None of this means that dispositions are irrelevant. It means that they are difficult to measure before someone starts work, and that they are partly produced by the job itself. The sensible conclusion is to assess what can be assessed well: the ability to learn, evidence of how a candidate has behaved in relevant situations, and performance on a sample of the actual work. Trial periods, scored interviews and references checked against specific claims are all imperfect, but they are more informative than a conversation about someone's attitude.

The slogan, then, is half right and half evasive. Personal qualities do matter enormously, and no classroom teaches them. But they are not reliably detected in a short meeting either, and hiring for attitude while refusing to train is a way of not paying for something the employer needs. The aim should be to select people who can learn, and then to teach them.