Academic Reading Articles Practice 12 Test 02

A
Metro systems did not appear because cities admired subterranean engineering; they emerged because the surface stopped working. By the mid-nineteenth century, industrial capitals were swelling with migrants and factories, and the street became a bottleneck of carts, omnibuses, and later trams competing for space. London’s response was radical: in 1863 it opened what is widely regarded as the first modern underground railway, arguing that only by separating mass transit from road traffic could the city maintain economic momentum. Yet building beneath crowded neighbourhoods meant tearing up the very streets that commerce depended upon. The early approach—cut-and-cover—required trenches to be excavated along existing roads, with temporary bridges for pedestrians and a patched-up cityscape after the track was laid. Dust, noise, blocked access, and collapsing shop revenues created political resistance, so the success of the project depended not only on engineers but also on leaders willing to absorb public anger and keep construction moving.

B
The limitations of shallow tunnels soon pushed metro builders toward depth. Steam locomotives could drag carriages through early tunnels, but the smoke and heat made stations unpleasant and ventilation difficult, especially as lines grew longer. The breakthrough was a combination of new tunnelling tools and cleaner traction. In London, the Greathead Shield—an advance in mechanical excavation that stabilised soft ground—made circular, deep-bore tunnels feasible where open trenches were impossible. At roughly the same time, electrification replaced steam on key routes, reducing fumes and allowing trains to operate efficiently in enclosed spaces. Together, these shifts transformed the underground from a shallow trench system into a deeper network threading beneath existing foundations, rivers, and utilities. Deep tunnelling, however, demanded meticulous surveying, stronger linings, and careful management of groundwater, turning metro construction into a high-stakes contest between geology and human ambition.

C
As lines went deeper, operating them safely and comfortably became a second engineering frontier. Stations were no longer simple staircases down from the pavement; they became vertical transport machines. Escalators enabled rapid circulation between street level and deep platforms, while improved passenger flow design reduced congestion at peak times. Inside the tunnels, the objective was to move more trains without expanding the track footprint, and this depended on signalling systems. With better block control and later more advanced train detection, operators could shorten the distance between trains, raising capacity by reducing headways and smoothing traffic through the network. These changes were not cosmetic upgrades: they redefined what an underground railway could do, turning metros into high-frequency arteries capable of absorbing commuting demand that surface transport could not handle.

D
Yet metros have never been purely technical artefacts; they are political and financial bargains set in concrete and steel. Different cities adopted different governance strategies. Some relied on private companies expecting fares to cover costs, while others treated underground rail as a public utility financed through taxation, municipal borrowing, or government bonds. The choice of funding model shaped everything from fare levels to expansion speed, and it influenced which neighbourhoods received stations first. Political leaders justified expensive lines by promising productivity gains, fewer traffic jams, and a more orderly city, but they also faced the recurring problem of overruns and delays. Once excavation begins, unforeseen ground conditions, utility relocations, and contract disputes can inflate budgets, and a metro that was sold as an efficient solution can become a symbol of waste. In this sense, every kilometre of tunnel is also a test of public trust.

E
Historical shocks left distinct layers in metro history. During the Second World War, underground stations and tunnels served as shelters during air raids in several cities, turning transport infrastructure into emergency refuge. At the same time, construction often slowed or stopped as labour and materials were diverted to the war effort, and wartime damage forced planners to rethink urban form. In the post-war decades, reconstruction and modernisation programs revived underground expansion, presenting metros as symbols of a rebuilt future. Later, however, eras of cheap fuel and car-oriented planning weakened the political case for large-scale subterranean projects. Suburban sprawl made fixed-rail investments harder to justify in some regions, while periodic economic crises tightened budgets and postponed ambitious schemes. Metro development, therefore, did not follow a smooth upward curve; it advanced in bursts, shaped by conflict, recovery, and shifting ideologies about how a city should move.

F
In the late twentieth and early twenty-first centuries, the metro’s relationship with the city became even more visibly socioeconomic. New stations can raise surrounding property values, attract investment, and accelerate redevelopment, but this can also trigger gentrification, pushing lower-income residents away from the very accessibility that public transport provides. Meanwhile, network maps often reveal inequality: lines and interchanges may prioritise districts with political influence, while other communities remain underserved or forced into long, indirect journeys. These routing debates expose questions of power—who benefits, who pays, and whose time matters. Alongside these social tensions, automation has become a defining trend. Some systems now use driverless trains to increase frequency, standardise operations, and reduce labour costs, but automation requires sophisticated control centres, rigorous safety design, and substantial capital investment. The technology can enhance reliability, yet it also concentrates risk: a single failure in software or oversight can ripple through the entire network.

G
Environmental pressures now dominate the strategic argument for underground rail, but they also introduce new vulnerabilities. Electrified metros can cut per-passenger emissions compared with private cars, especially as electricity grids decarbonise, making rail a cornerstone of climate policy. However, the environmental ledger is complicated by the carbon footprint of excavation, concrete, and steel, and by the growing threat of extreme weather. Flooding, in particular, can overwhelm entrances, shafts, and low-lying tunnel sections, damaging equipment and forcing closures. As a result, contemporary projects increasingly include resilience planning—pumps, barriers, drainage redesign, and protective upgrades to critical systems—so that underground infrastructure can survive climate volatility. Looking forward, the future of metros may depend less on grand standalone lines and more on integration: coordinated timetables, unified ticketing, and seamless links with buses, suburban rail, cycling, and walking. Metros persist because they solve a fundamental urban problem, but they endure only by adapting to new technologies, political realities, and environmental constraints.

Autonomous vehicles (AVs) are commonly presented as a technological remedy for an old problem: human error on the road. Fewer collisions, less congestion, and wider access for people who cannot drive are the headline benefits. Yet the ethical landscape does not begin and end with spectacular “trolley problem” thought experiments. Such crash dilemmas are rhetorically powerful but statistically rare, and they can distract from the routine decisions that distribute risk every second an AV operates. Ethical evaluation therefore extends to mundane engineering choices, the evidence required for deployment, and the social and legal arrangements that determine who benefits and who bears harm when systems fail.

At the core of AV ethics lies a safety philosophy built less on utilitarian calculus in emergencies than on preventing emergencies in the first place. Engineers emphasise that most safety is produced by cautious behaviour: maintaining generous following distances, managing speed to preserve stopping margins, and refusing manoeuvres that exceed a vehicle’s confidence in its sensors or control. Even here, however, values are embedded. A system tuned to minimise crashes may tolerate more false positives—braking or yielding unnecessarily—which can frustrate surrounding drivers, degrade traffic flow, or provoke risky human responses. A more assertive system may appear smoother but accept a higher background probability of conflict. The “ethical setting” is therefore not a single switch but a web of thresholds that quietly determine whose time is prioritised, how uncertainty is treated, and how risk is apportioned across passengers, pedestrians, and other road users.

Proving safety becomes hardest at the margins of experience. AVs can be trained and tested extensively in typical environments, yet the most consequential failures often occur in rare, messy situations: an erratic pedestrian stepping off a curb mid-block, a cyclist swerving around debris, a temporary sign placed at an odd angle, or a burst of glare that confuses vision systems. These long-tail events are often called edge cases, and they resist straightforward enumeration. Road testing accumulates real mileage but cannot ethically “hunt” for dangerous scenarios, while simulation can generate vast numbers of events but may miss subtle behaviours that real humans produce. As a result, developers face an ethical question of sufficiency: how much evidence counts as “safe enough,” what uncertainty is acceptable, and how to monitor performance once software updates and new data distributions change what the system encounters.

Data sits at the centre of this debate because AV safety is inseparable from measurement. Vehicles generate and store enormous quantities of sensor logs from cameras, radar, and lidar, capturing not only the vehicle’s actions but also fine-grained traces of surrounding people and environments. Companies argue that retaining these logs enables debugging, post-incident analysis, and continuous improvement, especially when rare failures must be reconstructed precisely. Privacy advocates respond that persistent recording can normalise surveillance, particularly if retention periods are long, access rules are unclear, or data is shared with third parties. In practice, policies on retention and access vary widely between firms and jurisdictions, shaping public trust and willingness to accept AVs in everyday life. The ethical challenge is to balance the legitimate need for safety evidence with strict limits on collection, storage, de-identification, and secondary use.

Liability attribution is equally unsettled, especially in partially automated systems that instruct a human to supervise. If the machine performs most steering and braking, the human may become complacent, and their capacity to intervene quickly may be more imagined than real. Critics describe a “moral crumbling” effect: when something goes wrong, responsibility collapses onto the human operator because they are the easiest target, even if meaningful control had already been ceded to automation. Manufacturers, regulators, and insurers must therefore decide whether AV harms resemble product defects, driver negligence, or a hybrid category where responsibility is distributed across designers, operators, and those who authorised deployment. Transparent documentation—training data sources, test coverage, update histories, and incident response procedures—can make liability decisions less arbitrary, but it also raises commercial tensions when firms are reluctant to reveal proprietary details.

Cybersecurity adds another layer of ethical urgency because AVs are connected systems, not isolated machines. Wireless updates, cloud connectivity, and complex sensor interfaces create attack surfaces that traditional vehicles did not have. A successful intrusion could disrupt perception, alter control commands, immobilise fleets, or exfiltrate sensitive information—turning a safety technology into a public hazard. Ethical deployment therefore requires security engineering from the outset: hardened software pipelines, strong authentication, careful key management, and rapid patching when vulnerabilities are discovered. It also requires clear reporting norms so that security flaws are not hidden until they are exploited. In this context, “safety” is not only about preventing crashes but also about preventing malicious manipulation of the system’s behaviour.

The social consequences of automation broaden the ethical lens beyond engineering and law. Large-scale AV deployment may disrupt employment in trucking, delivery, taxis, and logistics support, even if new roles emerge in fleet operations and maintenance. The transition costs are likely to be unevenly distributed across regions and worker groups, raising questions about who should fund retraining, wage support, or phased adoption. At the same time, AVs can reproduce or amplify inequality through algorithmic bias. If training data under-represents certain neighbourhoods, lighting conditions, body types, or mobility aids, detection and prediction may degrade precisely where vulnerability is higher. Fairness thus involves more than one demographic metric; it includes which areas are prioritised for service, who is excluded by design choices, and whether safety improvements are shared broadly rather than concentrated in affluent districts.

Regulation increasingly responds by defining operational limits rather than accepting blanket claims of autonomy. Many systems are certified only within an operational design domain—a set of road types, speeds, weather conditions, and mapping quality where performance is validated. One common tool is geofencing, which restricts vehicles to mapped zones or conditions where risks are better understood. When the system detects it is leaving that domain—due to heavy rain, sensor obstruction, unfamiliar road layouts, or ambiguous markings—it should not improvise beyond evidence. Instead, it is expected to execute a minimum-risk manoeuvre, such as slowing down, signalling, and pulling over to a safe stop. This regulatory stance embodies a practical ethic: reducing service is preferable to operating beyond proven capability, even if it occasionally disappoints users or limits commercial scale.

A Megacities are routinely described as “large cities,” yet such a label conceals their more consequential character: they function as complex adaptive systems whose behaviour emerges from dense interdependencies rather than from any single sectoral plan. When populations exceed ten million, the urban fabric becomes a tightly coupled assemblage of housing markets, labour flows, transport networks, energy grids, water cycles, waste streams, and public health regimes. Interventions in one domain are therefore rarely isolated; they propagate across the system through feedback loops, time lags, and distributional effects. In such contexts, governance is less a matter of issuing policies than of managing the unintended consequences produced when interconnected subsystems are pushed beyond capacity. The resulting dilemma is structural: the very agglomeration that generates productivity and specialised services also amplifies congestion, inequality, and exposure to climate hazards, meaning that “success” and “risk” are co-produced by the same urban mechanisms.

B The capacity to govern these interdependencies is frequently undermined by fragmentation, a condition that can be conceptualised as jurisdictional mismatch between functional urban systems and administrative boundaries. Megacities commonly incorporate multiple municipalities, semi-autonomous agencies, private utilities, and special-purpose authorities whose mandates overlap but do not align. Transport planning may be conducted independently of land-use zoning; drainage may be designed without reference to housing densification; health systems may be expected to respond to crises whose drivers sit in environmental or labour policy. This silo effect produces not only duplication and policy gaps but also delayed response when emergencies demand coordinated action across borders that are bureaucratic rather than ecological. When a flood, heatwave, or epidemic escalates, the question is not merely what should be done, but which institution has the authority, data access, and budget to act—while the hazard ignores municipal lines.

C What is feasible is further conditioned by finance, and financing is never neutral. Megacities require long-horizon investment in transit corridors, drainage retrofits, affordable housing, and distributed energy, yet revenue capacity is often uneven across districts and politically constrained at the metropolitan scale. Where property values are concentrated in central zones, tax bases can be spatially skewed, and where informal settlements dominate, large populations may fall outside formal revenue systems despite acute service needs. Debt instruments and land-value capture can expand fiscal space, but their benefits and burdens are distributed unevenly and can intensify exclusion if repayment relies on regressive fees. Public–private partnerships (PPPs) are frequently advanced as a pragmatic solution for mobilising capital and expertise; however, the governance literature has repeatedly noted that PPPs can generate accountability deficits when contracts are opaque, risk is socialised, and essential services become more expensive for households least able to pay. Thus, funding choices create both opportunity and risk: the same financial architecture that accelerates infrastructure delivery can embed long-term affordability problems and weaken democratic oversight.

D Technological solutions are increasingly presented as a substitute for institutional capacity, a move consistent with technocratic governance traditions. Smart sensors, predictive analytics, digital payments, and real-time dashboards can indeed increase operational efficiency by identifying leaks, optimising bus frequencies, or detecting equipment failure before breakdown occurs. Yet the deployment of “smart city” tools is also socially selective: benefits may accrue first to connected districts with reliable coverage and political influence, while marginal areas experience little improvement or face new burdens. Moreover, datafication raises privacy concerns when surveillance expands without oversight, especially in contexts where legal protections are weak or where data is shared across agencies and vendors. The crucial issue is not whether data can be collected—at megacity scale it inevitably will be—but whether transparency exists regarding what is gathered, how it is used, and who can contest errors or misuse. Without such safeguards, efficiency gains may be purchased at the cost of legitimacy and equity.

E Climate risk functions as the defining stress test because it targets multiple subsystems simultaneously and exposes the limits of incremental adaptation. Many megacities face compound hazards: heatwaves intensified by the urban heat island effect; flooding driven by extreme rainfall, sea-level rise, and inadequate drainage; drought that constrains water supply; and coastal storms that disrupt energy and transport. These hazards are often worsened by land subsidence, the sealing of soils, and the loss of wetlands that once absorbed water. Resilience planning typically includes drainage upgrades, protective barriers, cooling strategies, and early-warning systems, yet the most contested decisions involve land and exposure. A “defend” strategy seeks to fortify existing development, while a “retreat” strategy restricts construction in floodplains, relocates vulnerable communities, or redesigns zoning to reduce future risk. These choices are politically explosive because they translate climate science into distributional outcomes: who must move, who receives protection, and whose property rights are constrained.

F Social equity is not an ethical add-on but an operational precondition for sustainability, because policies that ignore distributional realities often fail through backlash. Environmental measures such as congestion charging, fuel price reform, or strict emissions controls may reduce pollution and carbon output, yet they can be perceived as regressive if low-income workers lack viable alternatives in transport or housing. When daily survival depends on informal employment, long commutes, or precarious shelter, a technically rational reform can be experienced as punitive. Consequently, inclusive governance requires compensatory design: targeted rebates, improved public transport, phased implementation, and participatory mechanisms that give affected communities meaningful influence. Without such provisions, enforcement becomes politically costly and compliance deteriorates, undermining the policy’s stated environmental goals.

G The megacity’s metabolism—its flows of food, water, materials, and waste—reveals dependence on hinterlands that are often governed by separate jurisdictions with different incentives. Cities import vast quantities of food and construction inputs and export waste to landfills, incinerators, or informal recycling networks that may sit far beyond municipal borders. Attempts to create circular economy systems—composting, reuse markets, industrial symbiosis, and repair infrastructures—can reduce pressure on landfills and supply chains, but only if coordination extends across the region and incentives reshape production and consumption upstream. If waste is treated as an afterthought, circularity becomes a slogan; if it is treated as a supply-chain redesign problem, it demands new regulatory alignments, cross-border agreements, and stable markets for secondary materials. Thus, sustainability is partly determined by governance beyond the city itself.

H Because the benefits and burdens of policy are uneven, accountability becomes central: trust is the scarce resource that determines whether plans survive contact with politics. Large-scale coordination can be achieved through metropolitan authorities that set standards and integrate transport and housing, but such bodies must be coupled with local flexibility to adapt policies to neighbourhood realities. Transparency mechanisms—open budgets, performance audits, and accessible citizen feedback channels—help make trade-offs visible and enable corrective action when projects drift or corruption emerges. In the absence of such mechanisms, even technically sound plans can fail, not because the engineering is wrong, but because legitimacy collapses and enforcement becomes contested. Governance, in this view, is a continual process of justification and revision rather than a one-time institutional design.

I Ultimately, sustainable governance in megacities is a long-term negotiation between growth and limits, conducted under uncertainty and contested values. The most effective approaches resist sectoral separation: finance, infrastructure, equity, and climate resilience are treated as linked issues rather than parallel agendas. A transit expansion, for instance, is simultaneously a mobility project, a housing strategy, a labour-market intervention, and a climate policy, and its success depends on whether institutional arrangements can manage these interactions. While certain principles—coordination, transparency, and inclusive design—appear repeatedly in comparative research, the passage does not claim that a single governance model fits every megacity. Instead, it implies that governance capacity must be assembled pragmatically, balancing metropolitan integration with local adaptation, and distributing benefits in ways that sustain legitimacy over decades of urban expansion.