Collapse: How Societies Choose to Fail or Succeed

by Jared Diamond

  The fallacy behind this policy of “filling up” Australia is that there are compelling environmental reasons why, even after more than two centuries of European settlement, Australia has not “filled itself up” to the population density of the U.S. Given Australia’s limited supplies of water and limited potential for food production, it lacks the capacity to support a significantly larger population. An increase in population would also dilute its earnings from mineral exports on a per-capita basis. Australia has recently been receiving immigrants only at the net rate of about 100,000 per year, which yields an annual population growth by immigration of only 0.5%.

  Nevertheless, many influential Australians, including the recent Prime Minister Malcolm Fraser, the leaders of both major political parties, and the Australian Business Council, still argue that Australia should try to increase its population to 50 million people. The reasoning invokes a combination of continued fear of the “Yellow Peril” from overpopulated Asian countries, the aspiration for Australia to become a major world power, and the belief that that goal could not be achieved if Australia had only 20 million people. But those aspirations of a few decades ago have receded to the point where Australians today no longer expect to become a major world power. Even if they did have that expectation, Israel, Sweden, Denmark, Finland, and Singapore provide examples of countries with populations far less than that of Australia (only a few million each) that nevertheless are major economic powers and make big contributions to world technological innovation and culture. Contrary to their government and business leaders, 70% of Australians say that they want less rather than more immigration. In the long run it is doubtful that Australia can even support its present population: the best estimate of a population sustainable at the present standard of living is 8 million people, less than half of the present population.

  Driving inland from the state capital of Adelaide in South Australia, the only Australian state to have originated as a self-supporting colony because of its soils’ decent productivity (high by Australian standards, modest by standards outside Australia), I saw in this prime farmland of Australia one ruin after another of abandoned farms. I was able to visit one of those ruins preserved as a tourist attraction: Kanyaka, a large manor developed as a sheep farm at considerable expense by English nobility in the 1850s, only to fail in 1869, to become abandoned, and never to be reoccupied. Much of that area of inland South Australia was developed for sheep farming during the wet years of the 1850s and early 1860s, when the land was covered with grass and looked lush. With droughts beginning in 1864, the overgrazed landscape became littered with the bodies of dead sheep, and those sheep farms were abandoned. That disaster stimulated the government to send the surveyor-general G. W. Goyder to identify how far inland from the coast the area with rainfall sufficiently reliable to justify farming extended. He defined a line that became known as the Goyder Line, north of which the likelihood of drought made attempts at farming imprudent. Unfortunately, a series of wet years in the 1870s encouraged the government to resell at high prices the abandoned sheep farms of the 1860s, as small overcapitalized wheat farms. Towns sprang up beyond the Goyder Line, railways expanded, and those wheat farms in turn succeeded for a few years of abnormally high rainfall until they too failed and became consolidated into larger holdings that reverted to being large sheep farms in the late 1870s. With the return of drought, many of those sheep farms subsequently failed once again, and those that still survive today cannot support themselves based on sheep: their farmer/owners require second jobs, tourism, or outside investments to make a living.

  There have been more or less similar histories in most other food-producing areas of Australia. What made so many initially profitable food-producing properties become less profitable? The reason is Australia’s number-one environmental problem, land degradation, resulting from a set of nine types of damaging environmental impacts: clearance of native vegetation, overgrazing by sheep, rabbits, soil nutrient exhaustion, soil erosion, man-made droughts, weeds, misguided government policies, and salinization. All of these damaging phenomena operate elsewhere in the world, in some cases with even greater individual impact than in Australia. Briefly, these impacts are as follows:

  I mentioned above that the Australian government formerly required tenants leasing government land to clear native vegetation. While that requirement has now been dropped, Australia still clears more native vegetation per year than any other First World country, and its clearance rates are exceeded in the world only by Brazil, Indonesia, the Congo, and Bolivia. Most of Australia’s current land clearance is going on in the state of Queensland for the purpose of creating pasture land for beef cattle. The Queensland government has announced that it will phase out large-scale clearing—but not until 2006. The resulting damage to Australia includes land degradation through dryland salinization and soil erosion, impairment of water quality by runoff of salt and sediment, loss of agricultural productivity and land values, and damage to the Great Barrier Reef (see below). Rotting and burning of the bulldozed vegetation contribute to Australia’s annual greenhouse gas emissions a gas quantity approximately equal to the country’s total motor vehicle emissions.

  A second major cause of land degradation is overstocking of sheep in numbers that graze down the vegetation faster than it can regrow. In some areas such as in parts of the Murchison District of Western Australia, overgrazing was ruinous and irreversible because it led to loss of the soil. Today, now that overgrazing’s effects are recognized, the Australian government imposes maximum stocking rates for sheep: i.e., farmers are forbidden to stock more than a certain number of sheep per acre on leased land. Formerly, however, the government imposed minimum stocking rates: farmers were obliged to stock a certain minimum number of sheep per acre as a condition of holding the lease. When sheep stocking rates first became well documented in the late 19th century, they were three times higher than the rates considered sustainable today, and before documentation began in the 1890s sheep stocking rates were apparently up to 10 times higher than sustainable rates. That is, the first settlers mined the standing crop of grass, rather than treating it as a potentially renewable resource. Just as was true for land clearance, the government thus required farmers to damage the land and cancelled leases of farmers who failed to damage the land.

  Three other causes of land degradation have already been mentioned. Rabbits remove vegetation as do sheep, cost farmers by reducing the pasturage available to sheep and cattle, and also cost farmers through the expense of the bulldozers, dynamite, fences, and virus release measures that farms adopt to control rabbit populations. Nutrient exhaustion of soils often develops within the first few years of agriculture, because of the low initial nutrient content of Australian soils. Erosion of topsoil by water and wind increases after its cover of vegetation has been thinned or cleared. The resulting runoff of soil via rivers into the sea, by making coastal waters turbid, is now damaging and killing the Great Barrier Reef, one of Australia’s major tourist attractions (not to mention its biological value in its own right and as a nursery of fish).

  The term “man-made drought” refers to a form of land degradation secondary to land clearance, sheep overgrazing, and rabbits. When the cover of vegetation is removed by any of these means, land that the vegetation had previously shaded now becomes directly exposed to the sun, thereby making the soil hotter and drier. That is, the secondary effects creating hot and dry soil conditions impede plant growth in much the same way as does a natural drought.

  Weeds, discussed in Chapter 1 in connection with Montana, are defined as plants of low value to farmers, either because they are less palatable (or totally unpalatable) to sheep and cattle than preferred pasture plants, or because they compete with useful crops. Some weeds are plant species unintentionally introduced from overseas; about 15% were intentionally but misguidedly introduced for use in agriculture; one-third escaped into the wild from gardens where they had been intentionally introduced as ornamentals; and other we
ed species are Australian native plants. Because grazing animals prefer to eat certain plants, the action of grazing animals tends to increase the abundance of weeds and to convert pasture cover to plant species that are less utilized or unutilizable (in some cases, poisonous to animals). Weeds vary in the ease with which they can be combatted: some weed species are easy to remove and to replace with palatable species or crops, but other weed species are very expensive or prohibitively difficult to eliminate once they have become established.

  About 3,000 plant species are considered weeds in Australia today and cause economic losses of about $2 billion per year. One of the worst is Mimosa, which threatens an especially valuable area, the Kakadu National Park and the World Heritage Area. It is prickly, grows up to 20 feet tall, and produces so many seeds that it can double the area that it covers within a year. Even worse is rubber vine, introduced in the 1870s as an ornamental shrub from Madagascar to make Queensland mining towns prettier. It escaped to become a plant monster of a type depicted in science fiction: besides being poisonous to livestock, smothering other vegetation, and growing into impenetrable thickets, it drops pods that disperse far by floating down rivers, and that eventually pop open to release 300 seeds carried far by the wind. The seeds within one pod suffice to cover two-and-a-half acres with new rubber vines.

  To the misguided government policies of land clearance and sheep overstocking previously mentioned may be added the policies of the government’s Wheat Board. It has tended to make rosy predictions of higher world wheat prices, thereby encouraging farmers to incur debt for capital investments in machinery to plant wheat on land marginal for wheat growing. Many farmers then discovered, to their misfortune after investing much money, that the land could support wheat for only a few years, and that wheat prices dropped.

  The remaining cause of land degradation in Australia, salinization, is the most complex and requires the most explanation. I mentioned previously that large areas of Australia contain much salt in the soil, as legacies of salty sea breezes, former ocean basins, or dried-out lakes. While a few plants can tolerate salty soils, most plants, including almost all of our crops, cannot. If the salt below the root zone just stayed there, it wouldn’t be a problem. But two processes can bring it up towards the surface and start causing problems: irrigation salinization and dryland salinization.

  Irrigation salinization has the potential for arising in dry areas where rainfall is too low or too unreliable for agriculture, and where irrigation is necessary instead, as in parts of southeastern Australia. If a farmer “drip-irrigates,” i.e., installs a small irrigation water fixture at the base of each fruit tree or crop row and allows just enough water to drip out as the tree’s or crop’s roots can absorb, then little water is wasted, and there is no problem. But if the farmer instead follows the commoner practice of “broadcast irrigation,” i.e., flooding the land or else using a sprinkler to distribute the water over a large area, then the ground gets saturated with more water than the roots can absorb. The unabsorbed excess water percolates down to that deeper layer of salty soil, thereby establishing a continuous column of wetted soil through which the deep-lying salt can percolate either up to the shallow root zone and the surface, where it will inhibit or prevent growth of plants other than salt-tolerant species, or else down to the groundwater table and from there into a river. In that sense, the water problems of Australia, which we think of as (and which is) a dry continent, are not problems of too little water but of too much water: water is still sufficiently cheap and available to permit its use in some areas for broadcast irrigation. More exactly, parts of Australia have enough water to permit broadcast irrigation, but not enough water to flush out all the resulting mobilized salt. In principle, problems of irrigation salinization can be partly mitigated by going to the expense of installing drip irrigation instead of broadcast irrigation.

  The other process responsible for salinization, besides irrigation salinization, is dryland salinization, potentially operating in areas where rainfall suffices for agriculture. That’s true especially in the areas of Western Australia and parts of South Australia with reliable (or formerly reliable) winter rains. As long as ground in such areas is still covered with its natural vegetation, which is present all year, the plants’ roots take up most of the rain falling, and little rainwater remains to percolate down through the soil to establish contact with the deeper salt layers. But suppose a farmer clears the natural vegetation and replaces it with crops, which are planted seasonally and then harvested, leaving the ground bare for part of the year. Rain soaking the ground when it is bare does percolate down to the deep-lying salt, permitting it to diffuse up to the surface. Unlike irrigation salinization, dryland salinization is difficult, expensive, or essentially impossible to reverse once the natural vegetation has been cleared.

  One can think of salt mobilized by either irrigation or dryland salinization into soil water as like a salty underground river, which in some parts of Australia has salt concentrations three times those of the ocean. That underground river flows downhill just as does a normal above-ground river, but much more slowly. Eventually, it may seep out into a downhill depression, creating hypersaline ponds that I saw in South Australia. If a farmer on a hilltop adopts bad land management practices that cause his land to become salinized, the salt may slowly flow through the ground to the land of farms lying downhill, even if those farms are well managed. In Australia there is no mechanism whereby the owner of a downhill farm that has been thus ruined can collect compensation from the owner of an uphill farm responsible for his ruin. Some of the underground river doesn’t emerge in downhill depressions but instead flows down into above-ground rivers, including Australia’s largest river system, the Murray/Darling.

  Salinization inflicts heavy financial losses on the Australian economy, in three ways. First, it is rendering much farmland, including some of the most valuable land in Australia, less productive or useless to grow crops and raise livestock. Second, some of the salt is carried into city drinking water supplies. For instance, the Murray/Darling River provides between 40% and 90% of the drinking water of Adelaide, South Australia’s capital, but the river’s rising salt levels could eventually make it unsuitable for human consumption or crop irrigation without the added expense of desalination. Even more expensive than either of those two problems are the damages caused by salt corroding infrastructure, including roads, railroads, airfields, bridges, buildings, water pipes, hot water systems, rainwater systems, sewers, household and industrial appliances, power and telecommunication lines, and water treatment plants. Overall, it is estimated that only about a third of Australia’s economic losses arising from salinization are the direct costs to Australian agriculture; the losses “beyond the farm gate” and downstream, to Australia’s water supplies and infrastructure, cost twice as much.

  As for the extent of salinization, it already affects about 9% of all cleared land in Australia, and that percentage is projected under present trends to rise to about 25%. Salinization is currently especially serious in the states of Western Australia and South Australia; the former state’s wheat belt is considered one of the worst examples of dryland salinization in the world. Of its original native vegetation, 90% has now been cleared, mostly between 1920 and 1980, culminating in the “Million Acres a Year” program pushed by the Western Australia state government in the 1960s. No other equally large area of land in the world was cleared of its natural vegetation so quickly. The proportion of the wheat belt sterilized by salinization is expected to reach one-third within the next two decades.

  The total area in Australia to which salinization has the potential for spreading is more than 6 times the current extent and includes a 4-fold increase in Western Australia, 7-fold increase in Queensland, 10-fold increase in Victoria, and 60-fold increase in New South Wales. In addition to the wheat belt, another major problem area is the basin of the Murray/Darling River, which accounts for nearly half of Australia’s agricultural production but which
now gets progressively saltier downstream towards Adelaide because of more salty underground water entering and more water being extracted for irrigation by humans along its length. (In some years so much water is extracted that no water is left in the river to enter the ocean.) That salt input into the Murray/Darling arises not just from irrigation practices along the river’s lower reaches but also from the impact of increasingly extensive industrial-scale cotton farming along its headwaters in Queensland and New South Wales. Those cotton operations are considered Australia’s biggest single dilemma of land and water management, because on the one hand cotton by itself is Australia’s most valuable crop after wheat, but on the other hand the mobilized salt and applied pesticides associated with cotton-growing damage other types of agriculture downstream in the Murray/Darling Basin.

  Once salinization has been initiated, it is often either poorly reversible (especially in the case of dryland salinization), or prohibitively expensive to solve, or solutions take a prohibitively long time. Underground rivers flow very slowly, such that once one has mobilized salt through bad land management, it may take 500 years to flush that mobilized salt out of the ground even if one switches overnight to drip irrigation and stops mobilizing further salt.