A For most of recorded history, money meant metal. The first true coins are generally credited to the kingdom of Lydia, in what is now western Turkey, some time in the seventh century BC, and for the next two thousand years the worth of a coin was understood to lie largely in the substance from which it was struck. Gold, silver and copper were scarce, durable and difficult to imitate convincingly, which made them ideal stores of wealth. Their great drawback was physical. Metal is heavy, and a merchant settling a substantial debt in a distant city faced the prospect of hauling a small fortune along roads where bandits were a constant hazard.

B It was in China that this practical problem first produced a lasting solution. During the Tang dynasty, in the ninth century, merchants who sold tea and other goods in the capital began depositing their takings with the offices that provincial governments maintained there, receiving in return a certificate that could be redeemed for coins back home. Because wealth seemed to travel without anyone carrying it, these documents became known as 'flying money'. Strictly speaking, they were not currency at all, since they were not passed from hand to hand in ordinary trade; they were closer to what we would now call a money order. Nevertheless, they accustomed people to the notion that a piece of paper could stand for metal stored somewhere else.

C The decisive step came roughly two centuries later in Sichuan, a province in the south-west whose authorities, short of copper, had for some time been minting coins of iron. Iron coins were worth very little relative to their weight, so that even a fairly modest purchase could require a load too heavy for one person to carry comfortably. Around the turn of the eleventh century, a group of wealthy merchant houses in the city of Chengdu started to issue notes known as jiaozi, which were accepted in exchange for goods and could be cashed in for iron coins. Inevitably, some issuers proved unable to honour their notes, and in 1023 the Song government took the business out of private hands, establishing a state office to print notes of its own. This is usually regarded as the birth of the world's first government-issued paper money.

D Later rulers embraced the innovation with even greater enthusiasm. After the Mongols under Kublai Khan conquered China in the thirteenth century, they made paper the principal currency of their empire, and the Venetian merchant Marco Polo, who claimed to have spent many years at the Khan's court, devoted an entire chapter of his account to it. He explained that the notes were made from the inner bark of the mulberry tree and stamped with the ruler's seal, and that anyone who refused them risked execution. Many of his readers dismissed the description as fantasy. The system, however, contained a hidden weakness. Since printing more notes cost almost nothing, governments facing military or financial pressure found the temptation to do so irresistible, and the result was repeated bouts of inflation. By the middle of the fifteenth century, under the Ming dynasty, the notes had lost so much of their value that they had largely fallen out of use.

E Europe arrived at paper money by a different route. In seventeenth-century London, goldsmiths who kept customers' valuables in their strongrooms issued receipts for the deposits, and these receipts gradually began to circulate as a means of payment. In 1661, Stockholms Banco in Sweden went further and issued what are generally considered the first true banknotes in Europe. Sweden's own currency at the time included enormous copper plates, the largest of which weighed nearly twenty kilograms, so the attraction of paper was obvious. The experiment was short-lived: the bank issued more notes than it could redeem and collapsed within a few years, and its founder, Johan Palmstruch, was sentenced to death, although the sentence was never carried out. The Bank of England, founded in 1694, proved more durable. Its early notes were partly handwritten, and fully printed notes did not appear until 1855.

F What made European paper acceptable was a guarantee. A note was, in effect, a promise by the bank to pay the bearer its face value in gold or silver on demand, and in the nineteenth century this promise was formalised in the gold standard, which Britain adopted officially in 1821 and much of the industrialised world later followed. The arrangement inspired confidence, but it also constrained governments, because the quantity of money in circulation was tied to the quantity of gold in the vaults. In wartime, when states needed to spend far beyond their means, convertibility was repeatedly suspended. Britain abandoned the gold standard for good in 1931, and in 1971 the United States ended the right of foreign governments to exchange dollars for gold, severing the last formal link between the world's major currencies and the metal.

G Today's money is what economists call 'fiat' currency: it is backed by nothing physical, and its value rests on the authority of the state and the willingness of the public to accept it. Paradoxically, just as paper has freed itself entirely from metal, it is itself being displaced. In several countries, Sweden among them, cash now accounts for only a small fraction of everyday payments, and some commentators predict that banknotes will eventually become collectors' items. If so, the long story that began with flying money will have reached its logical conclusion: money that is nothing more than a trusted record.

A Each autumn, several billion birds leave their breeding grounds in Europe and northern Asia and head south, and a remarkable proportion of the survivors return the following spring not merely to the same region but to the same wood, marsh or even nest box. Some feats are almost beyond belief. In 2007, a female bar-tailed godwit fitted with a satellite transmitter flew non-stop from Alaska to New Zealand, a distance of more than 11,000 kilometres, across open ocean with no landmarks whatsoever. For much of the twentieth century, how such journeys are possible was one of the great unsolved problems of biology. It is now clear that the answer is not a single sense but a set of overlapping systems, which researchers have identified one at a time, frequently by means of ingenious experiments.

B The first of these mechanisms was demonstrated in the early 1950s by the German ornithologist Gustav Kramer. Kramer noticed that caged starlings became restless during the migration season and tended to flutter in the direction they would have flown in the wild. When he used mirrors to shift the apparent position of the sun by 90 degrees, the birds' preferred direction shifted by a similar angle. Since the sun crosses the sky at roughly 15 degrees an hour, however, a sun compass is useless unless it is linked to an accurate sense of time. This was confirmed by so-called clock-shift experiments: when homing pigeons were kept for several days under lighting that put their internal rhythm six hours out of step, they made errors of about 90 degrees when released, exactly as the theory predicted.

C Many songbirds migrate at night, and for these species the American biologist Stephen Emlen identified a second guide in the late 1960s. Emlen placed indigo buntings in funnel-shaped cages lined with paper, with an ink pad at the base, so that each hop left a footprint revealing the direction the bird was attempting to take. Tested beneath the artificial sky of a planetarium, the buntings oriented themselves correctly, but became confused when constellations near the pole star were hidden. Most strikingly, in a study published in 1970, young birds raised under a planetarium sky that rotated around a different star, Betelgeuse, later behaved as if that star marked north. The implication was that the star compass is not inherited in fixed form; instead, the young learn it by observing which part of the sky remains stationary while the rest revolves.

D Neither the sun nor the stars are of any help under thick cloud, yet birds continue to orient successfully in overcast conditions. Work in Frankfurt during the 1960s showed that European robins kept in a windowless room responded to changes in an artificial magnetic field, and in 1972 Wolfgang and Roswitha Wiltschko demonstrated that this compass is unlike the one a hiker carries. Instead of detecting the polarity of the field, it registers the angle at which field lines dip towards the Earth's surface, which allows a bird to distinguish 'towards the pole' from 'towards the equator' but not north from south in the conventional sense. How the sense actually works remains the subject of intense research. The leading hypothesis, which grew out of work by the physicist Klaus Schulten in the late 1970s, holds that light-sensitive proteins in the eye called cryptochromes undergo chemical reactions affected by magnetism, so that the bird may literally see the field as a pattern superimposed on its view.

E A compass, though, only tells an animal which way it is heading; it does not reveal where it is. A celebrated experiment by the Dutch researcher A. C. Perdeck in the 1950s illustrated the difference. Perdeck captured around 11,000 starlings passing through the Netherlands on their way to wintering grounds in Britain and northern France, transported them by aircraft to Switzerland and released them. Adult birds corrected their course and headed for their usual destinations. First-year birds, by contrast, continued in their original compass direction and ended up in Spain and southern France. Experience, it seemed, provides a kind of 'map' that inexperienced birds lack.

F What that map consists of is still disputed, but smell appears to play a part. In the 1970s the Italian scientist Floriano Papi found that pigeons deprived of their sense of smell often failed to find their way home, and a 2013 study led by Anna Gagliardo reported that Cory's shearwaters whose sense of smell had been temporarily blocked could no longer navigate across the open sea, while those carrying magnets navigated normally. Sceptics doubt that odours alone could carry enough information over such vast distances, and many researchers suspect that birds combine smells with magnetic cues and familiar landmarks.

G These discoveries have acquired an urgency that Kramer could hardly have anticipated. In 2014, Henrik Mouritsen's group at the University of Oldenburg reported that robins on the university campus could not use their magnetic compass at all until their huts were screened with aluminium against the weak electromagnetic noise produced by electrical equipment. Artificial light presents a more visible threat: a 2017 study of a memorial installation in New York, whose vertical beams are switched on once a year, found that it drew migrating birds down in large numbers, and that they dispersed when the lights were briefly turned off. Understanding how birds navigate is no longer simply a scientific curiosity; it may prove essential to keeping their journeys possible.

A Within a generation, the car-free city centre has moved from the margins of urban planning to its mainstream. Pontevedra, a Spanish city of some 80,000 people, banned most private traffic from its historic core in 1999; Oslo removed around 700 parking spaces from its centre in the late 2010s; and Barcelona has been converting groups of city blocks into so-called 'superblocks' from which through-traffic is excluded. To their advocates, such schemes represent a long-overdue correction to a century in which the private car was permitted to colonise streets that had once belonged to everyone. To their critics, they are an expensive form of social engineering, imposed by a comfortable minority on people whose lives depend on driving. Both descriptions, I would argue, contain more truth than their proponents care to admit, which is precisely why the debate deserves more careful treatment than it usually receives.

B The case for closure begins with health, and here the evidence is hard to dispute. Road traffic is among the principal sources of nitrogen dioxide and fine particulate matter in most large cities, and both pollutants are associated with respiratory and cardiovascular disease. Traffic is also the dominant source of urban noise, whose effects on sleep and stress are increasingly recognised by public-health bodies. Proponents add a less measurable but no less important argument: that streets cleared of vehicles can be reclaimed for walking, play, markets and conversation, restoring a form of civic life that heavy traffic had gradually driven out. This last claim is sometimes dismissed as nostalgia, but anyone who has watched a once-congested square fill with people within months of its closure will find it difficult to regard as sentimental.

C The most persistent objection comes from shopkeepers, who fear that customers unable to drive to the door will simply go elsewhere. Their anxiety is understandable but, on the whole, poorly founded. Retailers consistently overestimate the share of their trade that arrives by car: a study of one shopping street in Bristol in the mid-2000s found that traders believed roughly twice as many of their customers drove as actually did. People who arrive on foot or by bicycle tend to spend less on each visit but return more often, so that over a month their spending compares favourably. None of this means that pedestrianisation is costless for business. A poorly designed scheme, particularly one that leaves a district hard to reach by any means, can and does drive trade away.

D A related fear is that traffic banished from one area will simply reappear in neighbouring streets, producing gridlock elsewhere. This does not necessarily happen. A review of road-closure schemes led by the transport researcher Sally Cairns in the late 1990s concluded that when road space is taken away, a proportion of the traffic does not divert but disappears, as people alter their routes, the timing of their trips, their mode of transport or even whether they travel at all. The effect, sometimes called 'traffic evaporation', is the mirror image of the well-documented tendency for new roads to fill with new traffic.

E Far more serious, in my view, is the objection from fairness. Not everyone can walk long distances or ride a bicycle, and public transport does not serve every journey equally well. For elderly and disabled residents, for parents managing small children and shopping, and for plumbers and electricians carrying equipment, a car may be a necessity rather than an indulgence. Schemes designed around the habits of young and able-bodied professionals can, without anyone intending it, shift the costs of change onto those least able to bear them. It is no answer to say that such people are a minority; a policy that improves the average while making life markedly harder for the most vulnerable is not one that a city should be proud of.

F The experience of cities that have made car-free centres work suggests that this objection can be met, but only deliberately. Pontevedra did not simply prohibit cars; it guaranteed access for emergency vehicles, for deliveries at set hours and for residents with garages, and it provided free car parks on the edge of the centre. Ghent, in Belgium, divided its centre into sectors in 2017, so that drivers could still reach any address but could no longer cut straight across the city. Where such provision has been neglected, the political consequences can be severe: in Madrid, a low-emission zone introduced in 2018 became the object of prolonged political and legal battles after a change of city government.

G What emerges, then, is that the question of whether city centres should be closed to cars is the wrong one. The more useful question is how, and for whom. A ban announced suddenly, without alternatives in place, is likely to provoke a backlash and may well be reversed; a gradual programme that invests first in public transport, safe cycling routes and well-managed access for those who genuinely need to drive has a far better record. Cities that wish to reclaim their streets would do well to treat the car not as an enemy to be defeated but as a guest whose privileges are being renegotiated.