For most of the twentieth century, the Garamantes were dismissed by historians as little more than a band of desert raiders. Greek and Roman writers had described them as a barbarous people who lived deep in the Libyan Sahara and occasionally attacked the settled towns along the Mediterranean coast. Modern scholars repeated these accounts with little curiosity. The Sahara, after all, was widely regarded as too harsh an environment to support a complex civilisation.

That picture began to change in the late 1990s, when archaeologist David Mattingly led a series of surveys in the Fezzan region of south-western Libya. Using satellite images and ground-penetrating radar, his team uncovered the remains of more than a hundred fortified settlements, dense networks of fields and, most surprisingly, an elaborate underground irrigation system. The Garamantes, it turned out, had built a sophisticated kingdom in one of the driest places on Earth, and they had done so without rivers, lakes or regular rainfall.

The key to their success lay several metres below the desert floor. Beneath the Fezzan lies a vast natural reservoir of water known as the Nubian Sandstone Aquifer, which was filled during a much wetter period thousands of years ago. To reach this water, the Garamantes constructed thousands of kilometres of tunnels, known locally as foggaras. Each tunnel sloped gently downwards from a hillside towards an oasis, drawing water through gravity alone. The tunnels were dug entirely by hand, ventilated by vertical shafts spaced every ten metres or so. Some foggaras run for more than five kilometres underground.

Building and maintaining this system was an extraordinary feat of engineering. Mattingly estimates that the foggaras required tens of thousands of labourers working over several centuries. Yet the Garamantian population at its peak is thought to have been only around 100,000 people. This has led some researchers, including the historian Klaus Schmidt, to argue that the foggaras must have been dug largely by enslaved workers captured during raids on neighbouring peoples. Others, such as Anna Lehmann, dispute this interpretation, pointing to the lack of large prisons or barracks at Garamantian sites and suggesting that much of the work may have been carried out by free citizens organised in seasonal labour groups.

Whatever the workforce, the rewards were considerable. The water from the foggaras allowed the Garamantes to grow wheat, barley, dates, grapes and even cotton in carefully laid-out fields. They built towns with multi-storey houses, developed a distinctive script and minted no coins but used Roman ones in trade. Their kingdom sat at the centre of a trans-Saharan trade network that carried gold, ivory and slaves northwards to the Mediterranean coast and salt, glass and pottery southwards into sub-Saharan Africa. Far from being barbarians on the margins of the classical world, the Garamantes were active participants in it.

The civilisation flourished for around a thousand years, from roughly 500 BCE to 500 CE, before going into rapid decline. The cause was not invasion, plague or political collapse — it was the very resource that had made the kingdom possible in the first place. Because the Nubian aquifer is recharged only by rainfall over thousands of years, the Garamantes were essentially mining a non-renewable resource. As the water table fell, the foggaras ran dry one by one. Fields could no longer be irrigated, towns were abandoned, and within two centuries the Garamantian kingdom had vanished beneath the sand.

For modern researchers, the story of the Garamantes is more than an archaeological curiosity. Several countries in North Africa and the Middle East rely heavily on the same aquifer that the Garamantes once tapped, and they too are withdrawing water far faster than it can be replenished. Mattingly has warned that the lesson of the foggaras is a sobering one: a civilisation built entirely on non-renewable groundwater is, in the long run, building on sand.

AFor most of human history, sleep was treated as a kind of small daily death — an unproductive gap between periods of useful activity. It was only in the 1950s, when the American researcher Nathaniel Kleitman and his student Eugene Aserinsky discovered that the sleeping brain passes through distinct, repeating phases, that scientists began to take sleep seriously as a subject of study. Their work, and the decades of research that have followed, has revealed that sleep is not a single state but a sequence of carefully orchestrated processes, each of which appears to serve a different biological purpose.

BA typical adult night of sleep is made up of four to six cycles, each lasting about 90 minutes. Within each cycle, the brain moves through several stages of so-called non-REM sleep, in which brain activity slows and the body relaxes, before entering REM sleep — short for 'rapid eye movement' — when the eyes dart back and forth beneath closed lids and most dreaming takes place. Curiously, the muscles of the limbs become temporarily paralysed during REM, which is thought to prevent the sleeper from physically acting out their dreams.

COne of the most striking findings of recent decades is that deep, non-REM sleep plays a critical role in memory. Studies carried out at Harvard University by Robert Stickgold and colleagues have shown that volunteers who learn a new motor skill perform measurably better on it the morning after a full night's sleep, even without any additional practice. The improvement is largest in those who spent the most time in deep sleep, suggesting that the brain uses this period to consolidate the memories formed during the day, strengthening some neural connections and discarding others.

DREM sleep appears to do something different. Rather than reinforcing memories one by one, it seems to weave them into wider patterns. In one well-known experiment, participants who were allowed REM sleep between two learning sessions were much better at spotting hidden links between apparently unrelated pieces of information than those who were kept awake. This may explain the long-standing intuition, recorded in many cultures, that 'sleeping on a problem' can produce creative solutions that hours of conscious thought cannot.

ESleep also plays an important role in physical health. During deep sleep, the body releases growth hormone, repairs tissues and regulates the immune system. A landmark study published in 2009 followed more than 150 healthy adults who were deliberately exposed to a common cold virus. Those who had slept less than seven hours a night in the preceding fortnight were almost three times more likely to develop symptoms than those who had slept eight hours or more. Other studies have linked chronic short sleep to a higher risk of obesity, diabetes and heart disease, though the exact mechanisms are still debated.

FDespite this growing body of evidence, the average adult in many industrialised countries now sleeps significantly less than their grandparents did. Surveys in the United States, the United Kingdom and Japan suggest that average nightly sleep duration has fallen by roughly an hour since the 1960s. Researchers point to several culprits: longer working hours, the spread of artificial light, and, most recently, the widespread use of bright electronic screens in the hours before bed. Light, especially the blue-rich light emitted by phones and tablets, suppresses the production of melatonin, the hormone that signals to the body that it is time to sleep.

GSome scientists argue that we are now living through a quiet public-health crisis driven by sleep deprivation. Others are more cautious, pointing out that sleep needs vary considerably from person to person and that the link between short sleep and disease may sometimes run the other way — illness may itself disrupt sleep. What almost all researchers do agree on, however, is that sleep is no longer the empty interval that previous generations believed it to be. It is one of the most biologically active and important periods of the day.

Urban planners have long thought of trees as decoration — pleasant additions to a street, but ultimately a matter of taste. Over the past two decades, however, a growing body of research has begun to reframe the urban tree as something closer to essential infrastructure: a piece of equipment that delivers measurable economic, environmental and health benefits to the cities that invest in it.

The most easily quantified of these benefits is temperature. As global temperatures rise, cities suffer disproportionately from a phenomenon known as the urban heat island, in which dense concentrations of concrete, asphalt and metal can raise air temperatures by as much as ten degrees Celsius above those of the surrounding countryside. Trees mitigate this effect in two ways. Their canopies provide shade, blocking up to ninety per cent of incoming solar radiation; and their leaves release water vapour through a process called transpiration, which cools the surrounding air much as sweat cools the skin. A 2019 study of forty European cities found that streets with continuous tree cover were on average two to three degrees cooler than nearby treeless streets during summer afternoons.

The implications of this cooling effect extend well beyond comfort. Heatwaves are among the deadliest of all weather events, and their toll falls most heavily on the elderly, the poor and those with chronic illnesses. A team of researchers led by epidemiologist Tamara Iungman estimated, in a study published in The Lancet in 2023, that increasing tree cover in European cities to thirty per cent of urban area could prevent roughly a third of all heat-related deaths during summer. In some southern cities, the figure could be even higher.

Trees also clean the air, although the picture here is more complicated than it first appears. Leaves intercept fine particles produced by traffic and industry, and some species absorb gaseous pollutants such as ozone and nitrogen dioxide through tiny openings in their leaves. However, certain species, particularly some oaks and poplars, also release compounds known as volatile organic compounds, or VOCs, which can react with vehicle emissions to form additional ozone at ground level. Choosing the wrong species, or placing trees in the wrong locations — for example, in a narrow street where they trap polluted air rather than allowing it to disperse — can therefore worsen the very problem they are intended to solve.

Less obvious are the mental-health benefits. A 2019 study published in Scientific Reports analysed the medical records of nearly 100,000 residents of Leipzig, Germany, and found that those who lived within a hundred metres of street trees were significantly less likely to be prescribed antidepressants. The effect held even after controlling for factors such as income, education and access to parks. The authors were careful not to claim that trees alone cause better mental health, but argued that the association was strong enough to deserve attention from urban planners.

Despite their benefits, urban trees face an increasingly difficult environment. Compacted soils, limited root space, road salt, vehicle damage and underground utilities all shorten their lives; the average street tree in a major Western city now survives for only around twenty years, compared with several centuries for the same species growing in a forest. Climate change adds further stress, exposing trees to droughts and pests for which native species may not be well adapted. Some cities, including Melbourne and Barcelona, have responded by deliberately diversifying their tree populations and importing species from warmer regions in anticipation of future climates.

There is also a question of equity. Studies in both the United States and Europe have repeatedly found that wealthier neighbourhoods tend to have far more tree cover than poorer ones, even within the same city. Since the neighbourhoods with the fewest trees are often the same ones with the worst air quality and the highest exposure to heat, the unequal distribution of urban trees can amplify existing health inequalities. A handful of cities have begun to tackle this directly — Los Angeles, for example, has explicitly committed to planting more trees in its lowest-income districts first — but for the most part, the gap remains wide.

What seems clear is that the question is no longer whether cities should plant trees, but how they should plant them — which species, where, for whose benefit, and how to keep them alive long enough for those benefits to be felt. Treating trees as infrastructure means budgeting for their maintenance, planning their locations as carefully as bus routes, and recognising that, like any infrastructure, they will fail without sustained investment.