Advanced reading practice test 5

Reading Passage 1

How Archaeologists Reconstruct Ancient Diets

AFood remains are among the most informative traces left by past societies, but they rarely provide a complete picture on their own. Archaeologists may examine animal bones, charred seeds, shells, pottery residues and the chemical composition of human teeth. Each source answers different questions. A bone with cut marks may indicate how an animal was processed, while a storage vessel can suggest how food was kept. The challenge is to combine these clues without assuming that any single discovery represents the whole diet of a community.

BPlant remains are often preserved unevenly. Seeds and grains may survive when accidentally burned, while softer plant foods usually decay. As a result, a site can appear to contain a diet dominated by cereals even if people also ate leaves, fruits or roots that left little archaeological evidence. Researchers must therefore consider how preservation affects the record. The absence of a particular food does not necessarily mean that people never consumed it.

CChemical analysis can reveal information that visible remains cannot. Stable isotopes in human bones and teeth may provide clues about the kinds of plants or animals people consumed, although interpretation depends on local ecology and the tissues examined. Different tissues form at different times, so teeth can reflect an earlier period of life while some bone measurements provide a more recent average. These methods offer broad patterns rather than a precise menu of every meal.

DPottery can preserve traces of fats and other compounds absorbed during cooking. By analysing these residues, researchers may identify evidence consistent with processing dairy products, aquatic foods or particular plant oils. However, similar chemical signatures can sometimes arise from different sources, and vessels may have been reused. Conclusions are strongest when residue analysis agrees with other evidence, such as animal bones, botanical remains or the shape and location of the vessel.

EDiet also reflects trade and social relationships. An ingredient found far from its natural growing area may suggest exchange, migration or the movement of animals. Yet imported food does not automatically indicate long-distance trade: people may have carried it during travel, or the environment may have changed. Researchers compare several sites and consider the historical context before deciding which explanation is most plausible.

FDifferences in food remains can reveal inequality, although the evidence needs careful interpretation. Some households may have had access to more meat or a wider range of imported goods, while others relied on locally available staples. Burial offerings and elite feasting areas can exaggerate the visibility of wealthy groups, so they should not be treated as representative of everyone. Ordinary domestic rubbish may provide a different, and sometimes more balanced, view of everyday life.

GModern scientific methods have expanded what can be learned from old material, but they also introduce new uncertainties. Samples may be contaminated, laboratory techniques may produce ambiguous results, and researchers must decide which comparisons are appropriate. Independent testing and transparent reporting help others assess whether an interpretation is well supported. New evidence can revise earlier conclusions without making the original researchers careless; it may simply answer questions that could not previously be investigated.

HReconstructing ancient diets is therefore a process of inference rather than a direct reading of the past. The most convincing accounts combine several kinds of evidence, explain the limits of each method and consider alternative interpretations. A collection of food remains can illuminate farming, trade, health and social organisation, but it cannot speak for itself. The researcher’s task is to build the strongest explanation the evidence allows while remaining open to revision.

Reading Passage 2

The Hidden Work of Maintaining Digital Infrastructure

ADigital services often appear effortless to users. A file opens, a payment is processed or a video begins to play, and the complex systems behind the action remain invisible. Yet these services depend on physical equipment, software updates, energy supplies, cooling systems and teams that monitor failures. The apparent simplicity of a digital product can hide a large maintenance operation, much of which becomes noticeable only when something stops working.

BSoftware maintenance is not limited to fixing obvious faults. Developers also update dependencies, adapt systems to new operating environments, respond to security weaknesses and remove obsolete components. A change that seems minor can affect other parts of a system, especially when its original design is poorly documented. For this reason, teams need tests, version control and clear records of decisions. These practices may slow an individual change but reduce the risk of more expensive failures later.

CTechnical debt describes the future cost created when a team chooses a quick solution that makes later work harder. Like financial debt, it is not always harmful: a temporary shortcut may be reasonable when time is limited. Problems arise when shortcuts accumulate without a plan to address them. Developers may then spend increasing amounts of time understanding old behaviour rather than adding useful features. The cost is often gradual, which makes it easy for organisations to underestimate.

DInfrastructure also has a material footprint. Data centres require electricity and cooling, while networks depend on cables, routers and other equipment. The environmental impact varies according to energy sources, hardware efficiency, utilisation and local climate. It is therefore misleading to describe every digital activity as either environmentally harmless or equally damaging. Measuring the footprint requires attention to the whole system, including equipment manufacture and replacement as well as ongoing operation.

EReliability depends on preparation for failure, not merely confidence that failure will not occur. Teams may maintain backups, test recovery procedures and design services so that one component can fail without bringing down everything else. A backup that has never been tested, however, may not work when it is needed. Similarly, a recovery plan may depend on staff or access credentials that are unavailable during an emergency. Rehearsal exposes weaknesses before a real incident does.

FMaintenance work is frequently undervalued because its success is difficult to see. A new feature has a clear launch date and can be demonstrated, whereas a vulnerability that never becomes an incident leaves no visible event to celebrate. Organisations that reward only rapid delivery may encourage teams to postpone necessary upkeep. Over time, the resulting fragility can make innovation slower because each change becomes riskier and more expensive.

GAutomation can help monitor systems and identify unusual patterns, but it does not remove the need for human judgement. Alerts may be too frequent, too vague or based on thresholds that no longer fit current conditions. Staff need to distinguish a genuine threat from harmless variation and decide which problems deserve immediate attention. Good monitoring therefore combines reliable tools with clear responsibilities and a process for learning from incidents.

HDigital infrastructure should be understood as a continuing service rather than a finished object. Its value depends on the less visible work that keeps it secure, reliable and adaptable. Budgets and schedules that recognise maintenance as essential can reduce future disruption, while transparent documentation helps new staff understand existing systems. The goal is not to prevent every failure, which is impossible, but to detect problems early, recover effectively and improve the system over time.

Reading Passage 3

Why Some Languages Gain Influence

AThe influence of a language is often described in terms of the number of people who speak it, but speaker totals alone tell only part of the story. A language may be widely used in family life yet have limited reach in international business or scientific publishing. Another may be used by fewer people as a first language but serve as a common medium among people with different native languages. Influence depends on where a language is useful, who controls important institutions and what opportunities it provides.

BTrade and political power have historically helped languages spread. Merchants may adopt a language to negotiate across regions, while administrations may use one language for taxation, law or education. These uses can continue after the political circumstances that created them have changed. Once a language becomes established in institutions, learning it can provide access to employment and further education, giving individuals a practical reason to continue using it.

CEducation can reinforce this process, particularly when examinations and textbooks favour one language. Students may learn it to gain qualifications even if they use another language at home. Over time, the language associated with formal education may become linked with professional status. However, access to high-quality instruction is uneven, and a policy that promotes one language without supporting others can restrict participation or weaken the transmission of local languages between generations.

DTechnology can widen a language's reach, but digital presence is not evenly distributed. Search tools, speech-recognition systems and online learning materials work best for languages with sufficient written data, funding and technical development. Languages with fewer digital resources may be less visible online, which can reduce incentives to create new content for them. This can produce a feedback loop: more resources attract more users, while limited resources make a language harder to use in some digital settings.

ELanguage influence is not necessarily a zero-sum contest. People often use different languages for different purposes, and multilingual communities may move between them depending on context. A global language can provide access to international networks while a local language carries family knowledge, humour and cultural memory. Difficulties arise when institutions treat one language as inherently more valuable and provide too little support for people to maintain other languages they need or wish to use.

FLanguage policy can influence whether smaller languages remain part of public life. Broadcasting, public services, teacher training and the availability of written materials can make it easier for people to use a language beyond the home. Such support does not guarantee that every language will survive, since language change is also shaped by migration, personal choice and economic pressures. Nevertheless, public decisions can either widen or narrow the settings in which a language is practical.

GPredictions about language decline are complicated by the difference between knowing a language and using it regularly. A language may still be understood by many people while fewer children learn to speak it fluently. In other cases, community initiatives lead to renewed teaching and use. Researchers therefore examine transmission between generations, the range of settings in which a language is used and speakers' own attitudes, rather than relying only on census totals.

HThe influence of a language is best understood as the outcome of social, political, educational and technological conditions. Languages gain reach when they offer access to valued opportunities, but they remain meaningful for reasons that cannot be reduced to economic advantage. Supporting multilingualism requires recognising both the practical benefits of shared languages and the cultural value of linguistic diversity. The future of a language depends not only on how many people speak it today, but also on whether future generations can use it in the settings that matter to them.