MANAGED ABUNDANCE

In the earliest Chinese written records, woods appear mostly as obstacles to be removed. As in almost all early societies, fire was the primary means of taming this wild growth.¹¹ But by the sixth or fifth century BCE, nascent states began to see the woods and waters as resources to manage rather than wilds to tame. Soon philosophical texts by Mozi, Mencius, and Lord Shang produced the first coherent conceptions of natural resources in the Chinese tradition. Despite major differences in political philosophies, they agreed on the basic premise of natural abundance and that this abundance could be sapped by human activity. Dozens of new seasonal regulations (yueling) used the same basic principles, limiting the type, frequency, and location of destructive behaviors like logging, hunting, and burning.¹² In the third century, the Qin and Han dynasties codified these rules into the first formal statutes on natural resources in the Chinese tradition.¹³ They reflected a way of thinking about the environment widespread across early Eurasia.¹⁴

In addition to codifying wood-use regulations, the Qin and Han empires established a suite of offices to oversee them, including the imperial forester (yu), which gradually transitioned from an emphasis on hunting to control a broad suite of forest resources. They also established the first state monopolies, including restricted forests (jinshan) to supply fuel to mines and smelters, and a massive complex called Shanglin Park used both for ritual entertainment and to dominate the fuel and timber market of the imperial capital, then the biggest city in the world.¹⁵ The Qin and Han also started the first documented tree-planting programs, principally to shade roads and dikes.¹⁶ Yet while the imperial administration developed key capacities to manage forests, these were largely confined to the immediate hinterland of the capital. In principle, the early empires claimed exclusive oversight of all “mountain groves, ponds, and marshes” (shanlin huze), but in practice, they could only control limited territories. Even here, restrictions had limited application; many rulers issued edicts specifically “relaxing the restrictions on the mountains and marshes” (chi shanze zhi jin).¹⁷ Rather than controlling territory, the Qin and Han usually drafted labor to harvest forest products, including prisoners forced to gather “firewood for the spirits” (guixin).¹⁸ They imported most large timber from the extensive natural woodlands of the south and west.¹⁹ Outside of a few limited times and places, the overwhelming preponderance of woodland was open-access natural growth.

Despite massive political upheaval following the collapse of the Han empire in the early third century CE, the principles of natural resource governance did not change markedly for another five hundred years. Between the third and sixth centuries, China was divided among rival polities, each of which claimed the imperial mandate but had far less capacity to control territory, enforce regulations, or draft labor. The era’s short-lived courts struggled to maintain even limited controls against the claims of lesser nobles, whom they feebly attempted to prevent from “monopolizing the wilds” (zhan shanze).²⁰ During this long period of decentralization, state controls of the wilds were both impractical and largely unnecessary. This era also saw the rise of monastic Buddhism and Daoism, each of which contributed new, often paradoxical, ways of thinking about nature and natural resources. Buddhists cultivated veneration of all life, but they also developed a surprisingly strong profit motive with financial techniques brought from India. This fed rival tendencies toward both conservation and commodification of forests.²¹ Religious Daoists likewise held nonhuman life in particularly high regard, but they also incorporated arcane techniques for taming dangerous and wild natures, derived in part from the non-Sinitic peoples of the south and west.²² Religious figures in both traditions contributed to the expansion of settlements into the wooded periphery, as seen through a literature on monks battling forest demons.²³ Throughout this period, most woodland probably devolved to the control of peasants, nobles, or monks or returned to a wilder state.

Nonetheless, the proliferation of institutions with claims to woodland contributed to the spread and elaboration of forest cultivation. The period’s most important manual on estate management, Jia Sixie’s sixth-century Essential Arts to Nourish the People, describes a number of techniques for cultivating trees for commercial purposes. It includes extended sections on orchards and hedges (yuan li), with instructions on growing jujubes (zao), peaches (tao), crab apples (nai), plums and apricots (li, mei, xing), and various other fruit-bearing trees.²⁴ Jia gives instructions on cultivating mulberry (sang) and Chinese mulberry (zhe) to feed silkworms.²⁵ He also details methods for planting elms and poplars (yu, baiyang) on three- to ten-year coppicing cycles, providing the most important sources of fuel and small poles.²⁶ Shorter sections detail willows (liu), pagoda trees (huai), catalpas (zi, qiu), and bamboo (zhu).²⁷ Anecdotes from the era describe both temples and noble estates cultivating orchards, tea plantations, and fuel coppices along the lines that Jia described.²⁸ Yet while the scattered estates of the period bristled with orchards, woodlots, and hedges, control of human labor remained the principal mechanism for managing woodland resources. Like the imperial government, noble and monastic estates conducted their own labor drafts, dominating labor to the point of contention with the imperial government.²⁹ In stark contrast to its extensive coverage of other commercially valuable tree products, Essential Arts does not include advice on planting timber trees, offering only a section on logging (famu). This suggests that estates derived most of their timber from logging the extensive naturally seeded woodlands at the periphery.³⁰ As long as woods were plentiful, woodcutting levies were both parsimonious and effective. They placed the locus of woodland governance on labor that was scarce relative to the wood it cut.

In the early seventh century, the Tang dynasty (618–907) formalized this principle of managing woodlands through human labor. With the compilation of the Tang Code of 624, the young dynasty organized the scattered regulations of earlier centuries into a universal penal law. Two statutes in the code established specific guidelines around the use of woodlands. The first law prohibited any private entity from monopolizing the bounties of the wilds (zhan shanye pohu li).³¹ This reflected the principle of exclusive wildland oversight seen in edicts since the Qin and Han. The second article established a fundamentally new law governing “products of the wilds with labor already invested [in them]” (shanye wu yi jia gongli). It defined as theft (dao) the taking of natural products that had already been gathered, including wood that had already been cut.³² This meant that it was human labor that turned natural products into property. The Tang even deployed this principle to manage its own wood supply. An eighth-century contract confirms that woodcutting was a common form of labor service through the midpoint of the dynasty.³³

Far from a unique development, the conceptualization of wildland seen in the Tang Code was roughly comparable to late Roman law. In translating the Chinese laws into their equivalent Latin terms, standing timber was treated as fructus naturales—“fruits of nature” that could not be owned. Cut timber became fructus separati—“fruits separated” from their conditions of production that become the property of the person who cut them. Together, these laws reflected the principle of separatio fructuum, or “cutting the fruit,” which held that it was removing fruits of nature from their conditions of growth that rendered them property.³⁴ While stated most clearly in the Tang Code, these legal principles formalized inchoate concepts dating from before the seventh century. Later the Tang Code was adopted almost verbatim into the penal law of the Song dynasty with these articles unchanged.³⁵ Just as the Roman law would form the basis of regulations on the European continent, Tang law laid the foundations for later Chinese law. Most importantly, the articulation of a clear principle of “cutting the fruit” provided an avenue to turn rights to use wildland into rights to own wild products. This idea would eventually enable the ownership of woodland itself.

The Tang also saw further developments in silvicultural technique. The poet and essayist Liu Zongyuan (773–819) even wrote a biography of “tree planter Guo the hunchback” (zhongshu Guo tuotuo), a professional gardener who is also attributed a Book of Tree Planting.³⁶ Whether or not “hunchback Guo” actually wrote it, this book is notable for its extensive, intimate knowledge of tree planting. It shows that knowledgeable cultivators of the ninth century had access to a broad suite of silvicultural techniques, including planting from seeds and cuttings, transplanting, grafting, pruning, and logging. The Book of Tree Planting also includes the first clear instructions for planting timber trees, such as pine (song) and fir (shan).³⁷ Nonetheless, as late as the tenth century, anecdotal evidence suggests that most timber was still cut from the wild growth. Edwin H. Schafer argues, based on his survey of Tang literature, that “medieval forests must still have seemed inexhaustible” and that officials saw tree planting as a waste of time.³⁸

Despite substantial changes during the first millennium CE, the fundamental premise of natural bounties prevailed. States restricted small patches of forest, peasant communities managed woodlots for fuel, monks and aristocrats planted fruit and shade trees, but the wilds were always waiting at the fringes of this cultivated tapestry. Simple regulations were still enough to promote the abundance of nature. Nonetheless, in the second half of the millennium the precepts governing woodlands began to change. While reinforcing earlier standards of open access, the Tang Code also laid the foundations for later concepts of property rights. While practiced only on small estates, the premodern suite of silvicultural techniques and tree species was essentially complete by the ninth century. These developments suggest that the assumptions of abundance were beginning to fail, first gradually and then with growing urgency. In the period that followed, this system entered its own crisis, a crisis only resolved through a fundamentally new conception of the natural world.

THE SONG WOOD CRISIS

Like the development of early imperial regulations, the transition to large-scale silviculture began with a crisis within the old patterns of wood use. Before this crisis, communities had maintained a reliable supply of wood products through three main mechanisms: seasonal restrictions on open-access woodlands, trade between wood-rich and wood-poor regions, and limited cultivation of forests in strategic areas. By the Song dynasty (960–1279), this system became unstable, as rising demand led to both intensified use of communal woodlots and extensive logging at the periphery. Song officials initially tried to resolve threats of wood shortage by escalating the first two mechanisms—imposing stronger logging restrictions and importing more timber—but these interventions were no longer enough. By the early twelfth century, the demand for timber was so high that people increasingly resorted to the last tool in their repertoire: tree planting. Formerly practiced in highly localized conditions, silviculture became widespread. The spread of tree planting, the most intensive mechanism for managing woodlands, was a clear indication that the other, more parsimonious forest systems had failed, and with them the assumptions that limited management was enough to secure natural abundance.

The crisis of the old system of wood oversight was the product of several overlapping shifts in the economic and geopolitical environment in the early Song. When the Song dynasty was declared in 960, it was merely the latest of six courts to control North China since the fall of the Tang in 907. But unlike their short-lived predecessors, Song emperors ruled for over three centuries, first from Kaifeng (Bianjing), where the Grand Canal joins the Yellow River (the Northern Song). After losing most of North China to a rival state in 1127, the dynasty continued until 1279, ruling from Hangzhou (Lin’an), where the southern terminus of the Grand Canal meets a large bay on the East China Sea (the Southern Song).

Smaller than the great empires that preceded it, the Song derived power more from commerce and centralization than from extensive territory. With an increasingly specialized bureaucracy, growing printing and popular literacy, a state-backed paper currency, and extensive use of bituminous coal, the Song appears in retrospect much like the states of western Europe more than five hundred years later. Some historians have considered it the beginnings of the “early modern” period in China.³⁹ Like these later states, the Song spent three centuries striving against the circumstances from which it emerged: military competition with powerful regional rivals, internal upheavals brought on by urbanization and an expanding commercial economy, and a wave of nearly unprecedented environmental threats.

When the Song emerged in the late tenth century, it controlled only the North China Plain. While large compared to most European states, it lacked huge territories that had formed part of the great empires of the past. While the Song was able to take possession of the south by 980, it contended against major non-Han rivals on its northern borders, including the Khitan-ruled Liao (907–1125), the Tangut-ruled Xi Xia (1038–1227), and the Jurchen-ruled Jin (1115–1234). Not only did enemy states cut off Song access to some of the richest woods on its periphery; they also posed major military threats that led all sides to escalate their wood use. In the northeast, the Song grew forests to defend its border with the Liao.⁴⁰ In the northwest, both the Song and the Xi Xia logged extensively to build forts during their mid-eleventh-century wars.⁴¹ Both interventions removed large swaths of woodland from other use. In the meantime, iron production expanded by an order of magnitude in the eleventh century, in large part for military purposes.⁴² Yet despite large investments in fort construction and weapons production, the Song could not effectively defend its borders. The Jin ultimately grew so strong that it forced the Song to retreat from its northern capital in 1127. Ironically, this invasion was made possible, in part, by the removal of the border forest that had been planted to slow nomadic cavalry.⁴³ Throughout these wars, the Song military struggled to balance escalating demand for timber and fuel with limited reserves, especially in the strategic border regions of the northeast and northwest.

The Song was far more urban than its predecessors, demanding large volumes of timber for shops, houses, and government buildings and for the ships that carried their supplies. In earlier empires, urban development had focused overwhelmingly on the imperial capitals, positioned near large, state-monopolized forests in the northwest. The Northern Song capital was also large: during its peak in the late eleventh century, Kaifeng probably reached 750,000 urban residents.⁴⁴ But unlike earlier capitals that abutted on wooded mountains, Kaifeng was in the middle of the wood-poor North China Plain and had to import essentially all of its wood. In fact, the homes and workshops of the Song capital consumed so much fuel that the region could not supply enough firewood, and Kaifeng switched almost entirely to mineral coal by the end of the eleventh century.⁴⁵ Dozens of other urban centers emerged in the eleventh century as well, far more than in earlier periods.⁴⁶ All of these cities brought their own demands for timber and fuel.

To make matters worse, the Song presided over the greatest environmental crisis in a thousand years, itself both a cause and a result of woodland degradation. The Yellow River, the flood-prone lifeblood of North China, had been relatively stable since the second century CE. But for hundreds of years, sediment accumulation had gradually raised the river’s banks above the surrounding countryside, threatening catastrophe when the river ran high. Extensive wood clearance only worsened the problem by exposing more soil to flow into the river and speed sedimentation. In the late tenth century, the river started to flood regularly, culminating in a massive deluge in 1048 that inundated large swaths of countryside and shifted the river’s course far to the north. To manage the unruly river, Song hydrocrats ordered extensive logging for fascines to rebuild the dikes. This only worsened the regional wood shortage while further depleting nearby mountains of their soil-retaining woodlands, leading to further rounds of sedimentation and flooding. In 1128, in an attempt to slow the Jin invasion, Song troops breached the Yellow River dikes, causing another massive flood, which shifted the course of the river to the south, far past its original course. Throughout this “environmental drama,” the river conservancy consumed timber on an unprecedented scale.⁴⁷

Finally, the Song oversaw a major growth in the commercial economy, fueled in part by a major expansion in the money supply. After overseeing the greatest minting of coin in over a thousand years, the Song backed the first official printing of paper money, expanding the currency beyond the supply of copper for the first time in Chinese history.⁴⁸ With so much money in circulation, timber and fuel became market commodities, priced in cash. This transformed wood from a distinctive, locally situated product into a standardized commodity. By rendering timber, fuel, and other products fungible, commodification provided the third key mechanism for bringing wood into circulation. In the cash-rich economy, merchants multiplied the influence of the state and cities, traveling the empire looking for more wood to bring into the commercial markets.

OFFICIAL FOREST MANAGEMENT

The overlapping climaxes in military, urban, hydrological, and commercial pressures challenged the Song state’s ability to find enough timber to build forts to defend its border, fleets to ply its waterways, and dikes to protect its farmland from deluge. These crises fed increasingly strident debates about the proper role of the state in taxing and regulating society and in managing the natural world. As the state realized the extent of its crisis, it doubled down on old forms of management, including restrictions on woodcutting, attempts to economize, expansion of the logging frontier, and an extension of direct forest oversight. Yet the key features of Northern Song forest policy were confusion and contention. Bureaucrats were on fundamentally new ground and disagreed about the proper course of action.

Song bureaucrats first attempted to regulate excessive wood extraction by imposing more and stricter logging bans, especially in the densely populated North China Plain. In 1049, a merchant requested a moratorium on logging in the northern portion of Dingzhou to allow its woods to recover.⁴⁹ The Dingzhou forest appears to have recovered somewhat by 1074, when its timber was cut again, resulting in another logging ban.⁵⁰ In 1080, a community in Huizhou 惠州 reported that its woods had dwindled to 12 percent of their original size and requested a total restriction on use until they recovered.⁵¹ The growing frequency of these restrictions, especially in the densely populated North China Plain, suggests that their effectiveness declined. While scattered, these reports also evidenced a growing conflict between official demand for timber and the government’s role in restricting overuse, as it was often official requisitions that led to excessive logging in the first place.

With logging restrictions proving increasingly ineffective, the most obvious means to alleviate the growing timber shortage was to expand the logging frontier. Government efforts to do so focused largely on the northwest, because it offered direct water routes to Kaifeng, and because the substantial military presence in the region offered the possibility of using soldiers as loggers. By the mid-eleventh century, the “woods that blocked out the sky” (linmu can tian) in the Qinglin mountains—long a preferred source of timber—had grown more scarce.⁵² To replace this supply, Song officials were increasingly tempted by the further reaches of the Loess Plateau, which was relatively well forested.⁵³ In fact, Song officials had logged the western fringes in the early decades of the dynasty, until Tibetan (fan) reprisals had led Emperor Zhenzong to cancel these operations in 1017.⁵⁴ In the 1030s, the buildup of Tangut power on the Loess Plateau led to extensive logging for military construction. By 1044, the Song and the Xi Xia collectively built more than three hundred stockades.⁵⁵ Logging in the region therefore risked exposure to enemy forces on two fronts. Nonetheless, the presence of rich woodlands upstream of Kaifeng offered a tempting source of timber.

In 1068, a minor supernumerary official named Wang Shao suggested a new tack in the decades-long conflict with the Xi Xia. He argued that the Song should recruit Tibetans as clients by offering them trade goods and titles. This would solve the Song’s strategic weakness while presenting the Xi Xia with enemies on multiple fronts.⁵⁶ By 1072, Song armies conquered the Xi Xia prefecture of Hezhou and incorporated it into the Song empire as Xihe Circuit, with Wang Shao appointed supreme circuit commander.⁵⁷ Xihe Circuit soon became a site of substantial institutional experimentation. Following its final pacification in 1074, more than a dozen new county- and prefecture-level towns were built to administer the region.⁵⁸ The state created official markets at the frontier to trade Sichuan tea for Tibetan horses.⁵⁹ It established markets to trade for Tibetan timber as well.⁶⁰ In 1080, the emperor appointed Li Xian, a eunuch supporter of Wang Shao, as the head administrator of these nascent timber markets.⁶¹ Noting that Xihe was the only location in the empire with timbers large enough for imperial construction, he gave Li authority to control the timber trade from the frontier markets all the way to the capital.⁶²

Over the next several years, Li Xian built the Xihe Logging and Timber Purchase Bureau (Xihe Cai Mai Muzhi Si) into a small but notable moneymaking operation in the northwestern borderlands. In 1081, the logging bureau was disbursed two hundred thousand strings of cash as the principal to buy timber.⁶³ It used profits from selling the logs downstream to finance the transport costs of grain and fodder to supply the frontier, with the principal reinvested in further timber purchase.⁶⁴ The returns on this investment were apparently significant: in 1084, Li Xian was able to borrow fifty thousand strings of cash from the logging bureau to buy stores for the military.⁶⁵ Like the larger and more famous tea and horse markets, the timber markets became an independent source of revenue and authority for a class of military and eunuch bureaucrats in the borderlands of the northwest.⁶⁶ They also increased state capacity to obtain timber in an increasingly tight Yellow River market. Yet this expansion of frontier logging was only one element in the Song’s changing forest oversight.

As reports of wood shortages grew in scale and frequency, Song officials began to reconsider some of the theoretical foundations of the centuries-old wood policies inherited from the Tang. Under the patronage of Emperor Shenzong in the late 1060s and early 1070s, the reformer Wang Anshi rose to the apex of Song political power, where he began to rethink the basis of state control of the environment. Citing the Offices of Zhou, a classical text that had been used to justify strong forest bureaucracies in the Qin and Han, Wang argued that forest regulation was well within the ambit of the classical state.⁶⁷ He argued that “in antiquity there was not just a single tax of ten percent [on farm production] … there were foresters and wardens in the wilds [shanze], and many varieties of [other officials].”⁶⁸ Despite the wood crisis, Wang explicitly refused to tax areas used for communal fuel collection (xide qiaocai) or any wildlands of public benefit (zhonghu zhili), including mountain forests. He also forbade landlords from enclosing these lands or renting them out on false pretenses.⁶⁹ But while Wang’s specific policies did not overhaul the rules governing wood use, his radical reading of the Chinese classics began to shift the underlying principles of natural resource governance. Yet this push for greater state authority was soon halted. Wang was forced to resign from office in 1076, and in 1085 his patron died. The new emperor, Zhezong, appointed Wang’s archrival Sima Guang as grand councillor, whereupon Sima abolished most of Wang’s policies.⁷⁰

Despite the incomplete and short-lived nature of Wang Anshi’s reforms, they laid the groundwork for more radical policies yet to come. In 1102, the young emperor Huizong, appointed another reformer, Cai Jing, as grand councillor. Cai soon revived interventionist policies modeled on Wang Anshi’s administration. Using the 1070s reforms as precedent, Cai initiated a wide-reaching program to extend state oversight and generate revenue from non-agrarian land, including the restoration of monopolies on goods like tea and salt.⁷¹ Under Wang’s reforms, assistant magistrates had been key to extending additional oversight to counties with large populations.⁷² Cai had a far more specific vision: he made assistant magistrates the first point of contact between the state bureaucracy and the productive landscape, enumerating new interventions that had not been part of Wang’s suite of reforms. In a major policy proposal, Cai wrote: “Copper, lead, gold, silver, iron, tin, and mercury mines and smelters and timber forests should be established; woodcutting should be restricted; barren mountains should be planted, etcetera. In each county, establish an assistant magistrate to manage these affairs.”⁷³

Building on Wang’s ideas, Cai specified a group of officials to manage state mines and forests throughout the empire. Acts from the next several decades clarified their functions. In 1105, an official in Jiangxi suggested to limit the post to counties that actually had mines and forests to manage. Following the elimination of unnecessary positions, about two-thirds of Jiangxi’s counties merited the additional staff.⁷⁴ This suggests that assistant magistrates took control of preexisting woodlands, probably areas that had been common-access prior to the policy. In the absence of other directives, they were responsible for enforcing existing logging restrictions rather than any radically new policies. Nonetheless, they became the first officials with specific responsibilities for managing forests at the local level.

Like Wang’s reforms, many of Cai’s policies were curtailed following his retirement in 1120.⁷⁵ But state forestry projects actually grew increasingly specific and closely managed over the next several years. An act of 1123 made each assistant magistrate responsible for maintaining twenty thousand timber trees (linmu) in his county, with provisions to punish those who kept fewer and reward those who kept more.⁷⁶ Two years later, local officials were made responsible for including these tree counts in their regular reports on the local economy.⁷⁷ Several other undated forest regulations were probably products of this period as well.⁷⁸ One specified clear punishments for anyone cutting wood from a state forest without license.⁷⁹ Another slowed the promotion schedule of assistant magistrates who permitted the destruction of forests under their supervision and rewarded those who expanded forests with faster professional advancement.⁸⁰ Collectively, these policies shifted forest oversight from preventative to positive policies. In addition to imposing increasingly strict restrictions on logging, the state tasked county officials with surveying standing timber and rewarded them for growing the size of their forests. While these rules do not specifically mention tree planting, this would have been one way for officials to meet production targets. Regardless of whether or not officials planted trees, the new regulations shifted the emphasis on wood oversight from logging to progressively earlier stages in the growth cycle. The ultimate fate of these county forests is not clear, in part because most of Cai Jing’s writings were destroyed by his critics. Indeed, his opponents were so successful in shaping the narrative that a pseudo-historical version of Cai Jing appears as a villain in the classic novel Outlaws of the Marsh.⁸¹ Nonetheless, it appears that the forest policies were soon rendered defunct by the retreat of the Song court in 1127.⁸²

PRIVATE FOREST MANAGEMENT

Like the Song state, private entrepreneurs developed two overarching responses to the wood crisis: logging new frontiers and intensifying forest management in the most densely populated regions. As the state focused on expanding logging into the northwest, which offered river routes to the capital and a large military presence, private merchants focused on the interconnected riverine and coastal markets of the south. While some of these regions had been logged to excess, others still had dense natural woodlands. An eleventh-century materia medica noted that “the deep mountains of the interior south” had plenty of fir in the natural growth.⁸³ In the early twelfth century, another text reported that the immediate hinterland of Hangzhou was “lush with lacquer, paper-mulberry, pine, and fir and frequented by merchants.”⁸⁴ Even in the late twelfth century, a Song minister described portions of the Hangzhou area as “dense with old firs.”⁸⁵ But as scarcities emerged in the oldest and best-known timber markets, merchants went further afield, buying timber from itinerant loggers throughout the Yangzi River basin.⁸⁶ Along the coast, cities like Ningbo (Mingzhou) and Quanzhou became particularly important centers of maritime trade in the twelfth century. As their industries developed, these ports were interlinked into an extensive trading realm that bought timber from as far afield as Guangzhou and southwestern Japan.⁸⁷ Collectively, the timber markets of the southeast coast and southern interior were probably an order of magnitude larger than the state-dominated logging regions of the north and northwest.

Meanwhile, the same pressures that led the state to intensify forest management prompted landowners to do the same. For the first time, demand was great enough that landowners began to invest in planting trees, not just for fruit and fuel, but for timber as well. Timber-planting techniques were well known before the twelfth century. As noted above, the ninth-century Book of Tree Planting records techniques for planting both pine and fir, the two principal timber trees of southern China. The famous poet and statesman Su Shi (1037–1101) also recorded a method of planting pines.⁸⁸ Fir planting was likewise attested for ritual and ornamental purposes. A temple near Hangzhou boasted two enormous fir trees that local tradition held had been transplanted in 893, while the firs at the “ten thousand fir” temple (wanshan si) in northwestern Jiangxi were planted no later than the early eleventh century.⁸⁹ In 1173, the philosopher Zhu Xi (1130–1200) planted firs on his grandmother’s grave in Huizhou 徽州; sixteen of his twenty-four plantings were still alive as of 1999.⁹⁰ But while these texts demonstrate the expertise to plant pine and fir, they do not suggest that commercial planting was widespread.

In this respect, 1100 represents a key turning point, from limited plantings on private estates and temples to large-scale investments in timber plantations. Ye Mengde (1077–1148) spent his later years in Huzhou, in western Zhejiang, where he wrote of his plans to grow large stands of pine, fir, and tung trees on thirty-year cycles.⁹¹ This is the first clear mention of staggering timber plantings across different plots so that the trees mature at different times. I have already quoted Yuan Cai, a native of nearby Quzhou, who wrote repeatedly of the profits from planting trees in the late twelfth century.⁹² A contemporary gazetteer recorded that in Huizhou, “the hills are well suited for fir, the locals do little work in the fields, and many plant fir as their vocation.”⁹³ Deeds from Huizhou further attest that plantations of fir seedlings (shanmiao) were widespread by the early thirteenth century.⁹⁴ Several agricultural manuals from the thirteenth century gave advice on how to grow fir, further documenting this expertise.⁹⁵ By then, the demand for timber was so great that it supported both a huge expansion in imports from far abroad and a fundamentally new market in purpose-grown timber in the Jiangnan interior. This marked a point of departure in biome modification—not just selective pressures on naturally seeded woodlands, but direct human intervention to plant and propagate timber.

PATHS NOT TAKEN

By the time Jiangnan landowners began planting large stands of timber trees in the twelfth century, people had been modifying China’s woodland biomes for thousands of years, first through fire and then through simple rules predicated on the notion of regulated abundance. Despite substantial continuities in this basic framework, it persisted across a millennium of change. But eventually the growing scale of cities, ships, armies, dikes, and markets generated demands for timber and fuel that could not be fulfilled through existing mechanisms. There was no single response to this crisis. Some officials turned to the proven solutions of the past, placing logging moratoriums on depleted woodlots and establishing timber markets in newly conquered regions. Others promoted more expansive views of state oversight, extending official forest management to counties throughout the empire. In the meantime, the merchants and landlords of the south developed their own responses, extending the timber trade far upriver and overseas and supplementing natural growth with the first extensive plantings of timber trees.

Collectively, these developments offered at least three different roads out of the Song wood crisis. First, China could continue to expand its resource frontiers, with a timber monopoly in the upper Yellow River basin and private timber markets along the Yangzi River and the fringes of the East China Sea. Following these trajectories, it could have developed much like Holland, with a riverine timber frontier on one side and a maritime timber frontier on the other.⁹⁶ Second, it could develop an extensive and powerful forest bureaucracy. This would have taken China along a similar path to those later followed by Venice, Korea, France, or Prussia.⁹⁷ We might remember Cai Jing as China’s Colbert, or even think of Colbert as France’s Cai Jing. Third, China could follow southern landlords like Ye Mengde and Yuan Cai who pioneered commercial silviculture. Until 1127, any of these roads might have led out of the wood crisis. All this changed when a foreign army occupied the north, literally forcing the Song along a southern route. When the dust settled, the court was left ruling territory centered on the fir-planting regions of Jiangnan, having lost both the wood-poor North China Plain and the timber monopoly of the northwest. As a result, it was private merchants, not official monopolies, that would drive the timber trade, and it was private planting, not state management, that would transform China’s woodland canopy.