London is smother'd with sulph'rous fires.
Still she wears a black hood and cloak
Of sea-coal smoak
As if she mourn'd for brewers and dyers.10
Since the sea-coal used by brewers, dyers and other manufacturers contained twice as much sulphur as that used today, its smoke darkened the air, dirtied clothes and curtains, stunted trees and flowers, blackened buildings and statues, and choked and killed the inhabitants. In an early condemnation of air pollution, published in 1661, John Evelyn compared the ‘columns and clouds of smoke which are belched forth from the sooty throats’ of London's chimneys with ‘the picture of Troy sacked by the Greeks’. The capital's inhabitants, he claimed, ‘breathe nothing but an impure and thick mist accompanied by a fulginous and filthy vapour’. Ladies used ground almonds to clean their complexion, while preachers in churches had to compete with the constant coughing and spitting of their congregations.11 The situation was even worse in those Dutch towns where industrial plants burned peat for brewing, dyeing, soap factories and brick kilns, because (although far cheaper than coal) peat created toxic fumes.
The presence of industrial enterprises in the heart of cities greatly increased the risk of fire; so did the use of wood and other flammable material to build cheap and shoddy high-density housing for the influx of immigrants. In East Asia, the use of wood not only for residences and shops but also for temples, government offices and covered markets further increased the risk; so did the use of palm and bamboo for roofs and floors and the practice of cooking on an open brazier, using oil lamps for light, and setting off fireworks during celebrations. Even the occupants of stone and brick temples, tombs, fortresses and merchant warehouses remained nervous. ‘Oh, this word fire!’ wrote an English merchant in Java. ‘Had it been spoken near me, either in English, Malay, Javanese or Chinese, although I had been sound asleep, yet I should have leaped out of my bed.’ He recalled that while the merchants slept, ‘our men many times have sounded a drum at our chamber doors and we never heard them; yet presently after, they have but whispered to themselves of fire and we all have run out of our chambers.‘12
As Christopher Friedrichs has noted: ‘Of all the elements, it was not earth, water or air that most persistently threatened the well-being of the early modern city. The most dangerous element was fire’ – and, in the mid-seventeenth century, major fires became both more frequent and more destructive. A ‘gazetteer’ of accidental urban fires in England listed over one hundred between 1640 and 1689, at least ten of which consumed over a hundred buildings. London experienced so many fires in 1655 that many thought they presaged the Last Judgement; while six years later, when faced after the Sunday sermon for the fifteenth consecutive week with pleas for charity from those whose homes had burned down, Samuel Pepys became irritated and ‘resolve[d] to give no more to them’.13 He changed his mind in 1666, when the ‘Great Fire’ of London destroyed St Paul's Cathedral, the Guildhall, the Royal Exchange, 84 churches and 13,000 houses, leaving 80,000 people homeless and causing £8 million of damage. Although Londoners blamed the Lord Mayor, who had initially jested that ‘a woman might piss it out’ and failed to create fire breaks, the true culprit was the climate: after an unusually hot and dry spring, temperatures in the summer of 1666 rose 1.5°C above normal, and a precipitation shortfall of 6 inches turned London into a tinderbox. The same conditions prevailed in much of north-western Europe, giving rise to fires in a score of German cities. Only the spectacular destruction of so much of London has overshadowed the frequency of urban fires elsewhere in 1666.14
7. The fires of Istanbul 1600–1700.
Three maps show the areas of the Ottoman capital destroyed by fire in 1600–50, 1650–75 and 1675–1700. Note that the third quarter of the century experienced the worst damage, with major conflagrations in 1652, 1660, 1665 and 1673.
London was not the only capital city where unusual drought in the mid-seventeenth century produced a ‘Great Fire’. In Moscow in 1648, after several months without rain, ‘within a few hours more than half the city inside the White Wall, and about half the city outside the wall, went up in flames'; while a large part of the new Mughal capital Shahjahanabad (now Delhi) burnt down in 1662. Istanbul suffered more (and more devastating fires) in the seventeenth century than in any other period of its history: one in 1660, once again after a prolonged drought, burned down 280,000 houses and several public buildings15 (Fig. 7). Major blazes also regularly devastated Edo, the largest city in Japan, notably the Meireki fire of 1657 – which, like those in Moscow in 1648, Istanbul in 1660 and London in 1666, broke out after an abnormal drought. Three separate conflagrations combined to destroy three-quarters of Edo, including 50,000 homes of merchants and artisans, almost 1,000 noble mansions and over 350 temples and shrines. Even the shogun's magnificent new castle, the tallest building in Japan, ‘which can be compared to one of the largest walled cities in Europe, has been completely destroyed by this horrendous fire’. Perhaps 160,000 people died. In the words of a Japanese contemporary account: ‘When the fire drew upon them, burning up completely all that was near at hand, some people could no longer bear the heat and they formed themselves into a human shield to try to ward off the fire, but they were choked by the billowing smoke. Others were consumed by the fire, their limbs burned to ashes.’ A Dutch eyewitness described how ‘with horror and dread I saw this immense city ablaze, like Troy'; and how, the next day, ‘passing through the streets’, he saw ‘innumerable burnt people, either completely or partly consumed, of which at least a third were small children, spread all over and lying dead on each other’. He also captured the desolation in a striking painting that showed empty city blocks, charred trees and scores of dead bodies in the streets16 (Plate 4). No sooner had rebuilding commenced than another major fire ‘destroyed an area about 1.5 miles in circumference’, followed by a third in 1661 and a fourth in 1668 which ‘devoured so many houses of nobles and civilians that it is estimated that two-thirds of the city of Edo has been destroyed’. According to one Dutch visitor, ‘it seems that it has become customary for that all-consuming element (fire) to rage there around Japanese New Year’.17
It is possible to demonstrate the extraordinary intensity of the four Edo fires of the mid-seventeenth century. A ‘core’ of earth excavated in 1975 at a building site in Hitotsubashi, not far from the shogun's castle, revealed three prominent layers of ash. The most recent, representing the firestorm caused by the bombing of Tokyo in 1945, measured 4 inches; the second, caused by the fire that followed the Kanto earthquake of 1923, measured 6 inches; the third, representing the fires of the mid-seventeenth century, measured 8 inches. The fact that the burned debris from 1657–68 was twice as thick as that created by the most advanced pyrotechnics of the twentieth century is both striking and sobering.18
All these urban fires were apparently accidental; many more occurred because of wars. Thus during the 1640s, although 13 English towns experienced ‘accidental’ fires, soldiers deliberately caused at least 80 more, some of them large (over 80 houses destroyed in Birmingham and almost 250 in Gloucester, both in 1643). War also destroyed towns in other ways: constructing or extending fortifications, preparing for a siege, and fire from siege artillery brought down many buildings. At Exeter, the third largest city in England, between 1642 and 1646 the defenders deliberately razed all the suburbs, where one-third of the pre-war population had lived, while bombardment during two sieges left ‘whole streets converted to ashes’. Although Exeter successfully resisted capture, it did not attain its pre-war extent for 60 years.19
The prevalence of war meant that every urban space needed walls – indeed, the Chinese character most commonly used for ‘city’ (cheng: ) literally signifies ‘city walls’ because it is made up of the characters that signify ‘earth’ () plus ‘complete’ () – but they did not always save the community within. Mainz, in western Germany, which surrendered without a struggle in 1631, over the next five years lost 25 per cent of its houses, 40
per cent of its population and 60 per cent of its wealth while it served as the headquarters of the Swedish expeditionary force. Although its massive walls allowed Pavia (Lombardy) to withstand an eight-week siege in 1655, success still ruined the city: lack of demand destroyed its industries; buying flour before and subsidizing bread prices during the siege bankrupted its treasury; and the besiegers’ destruction of all municipal assets beyond the walls seriously impeded economic recovery. Nonetheless, Mainz and Pavia were lucky: cities taken by force during the mid-seventeenth century suffered far greater losses and might take over a generation to recover. The siege and sack of Mantua in 1629 reduced its population from 29,000 to 9,000; by 1647 it still had only 15,000 inhabitants and in 1676 only 20,000. The population of Warsaw, the capital of the Polish-Lithuanian Commonwealth, numbered perhaps 30,000 in the 1630s (and up to 100,000 whenever the Diet met there), but this fell below 6,000 after occupation by Transylvanian and Swedish forces in 1655–7. They also left over half its buildings ruined.20 Perhaps the worst man-made urban catastrophe of the period occurred in 1642 when, after a siege that lasted a year, the Chinese rebel leader Li Zicheng decided to force the surrender of Kaifeng (capital of Henan province) by breaching the dikes on the nearby Yellow river. By a fateful coincidence, at exactly the same time the defenders broke another set of dikes hoping to flood Li's encampment and thus drive him off. According to a contemporary, water from both breaches poured through one of the city's gates, creating floods that ‘suddenly rose twenty feet’. The following day Li sent men into the city on boats in search of ransoms and plunder, but they found not a living soul.21
Floods also contributed to the ‘urban graveyard effect’. Because many cities grew up beside rivers and lakes, even without military intervention higher precipitation could cause immense flood damage. The worst inundation in the history of Mexico City occurred in 1629, when a combination of torrential rains and inadequate drainage caused the surrounding lakes to rise suddenly, submerging considerable parts of the city for five years. The catastrophe led some to consider relocating the capital, and although on this occasion the Spanish central government rejected the option, 30 years later repeated flooding led it to approve abandoning the regional capital of Santa Fé la Vieja, Argentina, and relocating it to higher ground. In Europe, the Seine burst its banks and flooded Paris 18 times in the seventeenth century, with particularly serious inundations in 1649, 1651 and 1658; towns in the low-lying province of Holland suffered even more frequently because storms in the North Sea periodically drove water over or through the dikes (as in 1651, flooding Amsterdam).22
A final cause of the ‘urban graveyard effect’ was the dependence of early modern cities on food produced far away. A Chinese magistrate near Shanghai foresaw with absolute clarity the danger inherent in this situation:
Our county does not produce rice, but relies for its food upon other areas. When the summer wheat is reaching ripeness and the autumn crops are already rising, the boats of the merchants that come loaded with rice form an unbroken line … [But] if by chance there were to be an outbreak of hostilities … such that the city gates did not open for ten days, and the hungry people raised their voices in clamour, how could there fail to be riot and disorder?
His fears turned into reality in 1641–2 when, even without ‘an outbreak of hostilities’, global cooling destroyed the rice harvest throughout South China. Perhaps 500,000 people starved to death and public order collapsed.23
The ‘Palace cities’
‘Palace cities’, those with a large population of otherwise unproductive government officials to feed, were most vulnerable because they normally had to import a high proportion of their food, and therefore sought supplies farther afield – and the longer the supply chain, the more susceptible it was to disruption. Thus, every year, huge convoys of barges carried 450,000 tonnes of rice (as well as vast quantities of wheat, millet, beans and other foodstuffs) to Beijing along the Grand Canal, which stretched almost 1,200 miles down to the fertile rice paddies of the Yangzi valley. In 1641 drought in Shandong caused the Grand Canal to dry up (for the only time in its history); while, after 1642, fear of bandit attacks interfered with routine maintenance (dredging, diking and repairing the locks) and disrupted the sailing of the convoys. Since most of the imported rice nourished the 300,000 inhabitants of the Inner City, the failure of the last Ming emperor to feed his own people no doubt contributed to their decision to surrender his capital, scarcely firing a shot, in 1644.
The provisioning of seventeenth-century Istanbul, another ‘palace city’, strikingly resembled that of Beijing. The Ottoman capital imported thousands of sheep and lambs, over 500 cattle and 500 tonnes of bread daily, because the sultan (like the Chinese emperor) needed to feed not only the imperial family, bureaucrats, eunuchs, artisans, guards, merchants, and their households, but also students in the colleges and medreses attached to the imperial mosques. Like Beijing, Istanbul also boasted a proven supply network – Egypt, the Balkans and the lands around the Aegean and Black Seas all regularly sent food to the city, some of it as tribute, just as they had done since Roman times – but this network too was subject to disruption by natural and human agency. Thus in 1620–1, the Bosporus froze over, while in 1641–3 unusually weak Nile floods caused an epic drought in Egypt: both climatically induced events dramatically reduced the supply of food shipped to Istanbul. War likewise disrupted the supply of food: between 1645 and 1658, when enemy fleets repeatedly prevented ships from passing through the Dardanelles, food prices in the Ottoman capital soared. On each occasion, whatever the cause, as in Beijing the households of those normally fed by the state suffered first and most – which helps to explain why palace personnel led the revolts that culminated in regicide in both 1622 and 1648.24
In Madrid, the capital of the Spanish Monarchy, magistrates imposed a daily schedule according to which each nearby village had to deliver a specified amount of wheat to the special granary maintained to feed the court. When supplies dwindled during the disastrous harvest of 1630, the magistrates extended the tribute system to include over 500 communities within a radius of 60 miles. Every house in each of these communities became responsible for providing a fixed share of the court's total requirement of 30 tonnes of wheat a day.25 In 1647, when ‘the torrential and persistent rain made traffic impossible on the roads to Madrid’, a senior minister warned the king that ‘The stocks of flour have almost run out and people cannot get into the countryside to find dry firewood to heat the ovens. Very few mills still have a wheel that works, because of the floods.’ The city's granaries soon emptied and the minister worried that ‘if the bread supply fails for a single day, instead of the 100 people who are protesting today, Your Majesty will find the entire population in front of the palace’.26 To avert disaster, the government once again unilaterally extended the grain tribute system, rescinded all exemptions, dismissed all appeals from oppressed villages for relief, and sent agents 120 miles and more from the capital to requisition bread. Thanks to these rapid and radical responses, Philip IV never did ‘find the entire population in front of the palace’.27
‘Palace cities’ were not alone in creating a sophisticated but vulnerable supply network. When other urban centres outgrew the capacity of their immediate hinterland to feed them, they too became dependent on distant market forces. Thus the burgeoning populations of the largest port cities around Europe's Atlantic and North Sea coasts (including London, Amsterdam, Antwerp, Lisbon and Seville) depended for their daily bread on importing large quantities of grain from the Polish-Lithuanian Commonwealth, which boasted fertile soil, cheap labour and easy access to water transport. By the early seventeenth century, between 150,000 and 200,000 tonnes of grain came down the Vistula annually for sale at Danzig, where an average of 1,500 ships loaded and shipped it to western Europe. Any interference in this commerce – for example when war broke out between Poland and Sweden in the 1620s, or when the Danish Sound froze over early in 1658 – caused the price of bread im
mediately to soar in the leading cities of Atlantic Europe, where the poorer members of their populations starved.
The Macro-Regions
No early modern settlement was entirely self-sufficient: all of them needed to import at least some items. Even the inhabitants of upland villages isolated for part of each year by winter snows or monsoon rains periodically trudged to the nearest market town to sell handicrafts or surplus agricultural produce and acquire such essentials as salt for food preservation and iron for tools. As population density increased in the sixteenth century, the number of market towns multiplied spectacularly. In China, the number of markets in Zhangzhou prefecture (Fujian) increased from 11 in 1491 to 38 in 1573, and to 65 in 1628. In Japan, market towns in most parts of the coastal plains stood by the 1630s between two and four miles apart; while by then, in England, men and women had to travel an average of only eight miles to the nearest market.
Markets achieved their greatest density around major cities – in the counties around London, markets were on average less than a mile apart – because they formed part of a zone of economic activity that contained the best arable land, the densest population, the hubs of communication and transport, and the largest capital accumulations. Economists call them ‘macro-regions’. Ming China contained eight macro-regions, each one centred on a river system and separated from the others by natural barriers. The Indian subcontinent also included several macro-regions, including Gujarat and the Ganges valley; the Ottoman empire boasted Egypt, the lands around the Aegean, and the Black Sea region; in the Americas, Mexico City lay at the centre of a macro-region; so did the Kinai and Kantô plains in Japan. The macro-regions of Europe included the Genoa–Turin–Venice–Florence quadrilateral in Italy; the Home Counties in southeast England; the adjacent provinces of Holland, Zealand and Utrecht; and the Île-de-France. Many of the settlements within these macro-regions adopted a high-risk, high-reward economic strategy: they concentrated on producing cash crops which they sold to merchants and manufacturers, and imported the food they needed from farther afield.28
Global Crisis Page 14
