Guns, Germs, and Steel: The Fates of Human Societies

—

yams, taro —The table gives major crops, of five crop classes, from early agricultural sites in various parts of the world. Square brackets enclose names of crops first domesticated elsewhere; names not enclosed in brackets refer to local domesticates. Omitted are crops that arrived or became important only later, such as bananas in Africa, corn and beans in the eastern United States, and sweet potato in New Guinea. Cottons are four species of the genus Gossypium, each species being native to a particular part of the world; squashes are five species of the genus Cucurbita. Note that cereals, pulses, and fiber crops launched agriculture in most areas, but that root and tuber crops and melons were of early importance in only some areas.

I Z 8 • GUNS, GERMS,

and steelthrown in handfuls, resulting in a whole field devoted to a single crop. Once cows, horses, and other large mammals were domesticated, they were hitched to plows, and fields were tilled by animal power. In the New World, however, no animal was ever domesticated that could be hitched to a plow. Instead, fields were always tilled by hand-held sticks or hoes, and seeds were planted individually by hand and not scattered as whole handfuls. Most New World fields thus came to be mixed gardens of many crops planted together, rather than monoculture.
Another major difference among agricultural systems involved the main sources of calories and carbohydrates. As we have seen, these were cereals in many areas. In other areas, though, that role of cereals was taken over or shared by roots and tubers, which were of negligible importance in the ancient Fertile Crescent and China. Manioc (alias cassava) and sweet potato became staples in tropical South America, potato and oca in the Andes, African yams in Africa, and Indo-Pacific yams and taro in Southeast Asia and New Guinea. Tree crops, notably bananas and breadfruit, also furnished carbohydrate-rich staples in Southeast Asia and New Guinea.
1 hus, by roman times, almost all of today's leading crops were being cultivated somewhere in the world. Just as we shall see for domestic animals too (Chapter 9), ancient hunter-gatherers were intimately familiar with local wild plants, and ancient farmers evidently discovered and domesticated almost all of those worth domesticating. Of course, medieval monks did begin to cultivate strawberries and raspberries, and modern plant breeders are still improving ancient crops and have added new minor crops, notably some berries (like blueberries, cranberries, and kiwifruit) and nuts (macadamias, pecans, and cashews). But these few modern additions have remained of modest importance compared with ancient staples like wheat, corn, and rice.
Still, our list of triumphs lacks many wild plants that, despite their value as food, we never succeeded in domesticating. Notable among these failures of ours are oak trees, whose acorns were a staple food of Native Americans in California and the eastern United States as well as a fallback food for European peasants in famine times of crop failure. Acorns are nutritionally valuable, being rich in starch and oil. Like many otherwise edible wild foods, most acorns do contain bitter tannins, but acorn lovers

HOW

TO MAKE AN ALMOND • 129learned to deal with tannins in the same way that they dealt with bitter chemicals in almonds and other wild plants: either by grinding and leaching the acorns to remove the tannins, or by harvesting acorns from the occasional mutant individual oak tree low in tannins.
Why have we failed to domesticate such a prized food source as acorns? Why did we take so long to domesticate strawberries and raspberries? What is it about those plants that kept their domestication beyond the reach of ancient farmers capable of mastering such difficult techniques as grafting?
It turns out that oak trees have three strikes against them. First, their slow growth would exhaust the patience of most farmers. Sown wheat yields a crop within a few months; a planted almond grows into a nut-bearing tree in three or four years; but a planted acorn may not become productive for a decade or more. Second, oak trees evolved to make nuts of a size and taste suitable for squirrels, which we've all seen burying, digging up, and eating acorns. Oaks grow from the occasional acorn that a squirrel forgets to dig up. With billions of squirrels each spreading hundreds of acorns every year to virtually any spot suitable for oak trees to grow, we humans didn't stand a chance of selecting oaks for the acorns that we wanted. Those same problems of slow growth and fast squirrels probably also explain why beech and hickory trees, heavily exploited as wild trees for their nuts by Europeans and Native Americans, respectively, were also not domesticated.
Finally, perhaps the most important difference between almonds and acorns is that bitterness is controlled by a single dominant gene in almonds but appears to be controlled by many genes in oaks. If ancient farmers planted almonds or acorns from the occasional nonbitter mutant tree, the laws of genetics dictate that half of the nuts from the resulting tree growing up would also be nonbitter in the case of almonds, but almost all would still be bitter in the case of oaks. That alone would kill the enthusiasm of any would-be acorn farmer who had defeated the squirrels and remained patient.
As for strawberries and raspberries, we had similar trouble competing with thrushes and other berry-loving birds. Yes, the Romans did tend wild strawberries in their gardens. But with billions of European thrushes defecating wild strawberry seeds in every possible place (including Roman gardens), strawberries remained the little berries that thrushes wanted, not the big berries that humans wanted. Only with the recent development of

130 • GUNS, GERMS, AND

steelprotective nets and greenhouses were we finally able to defeat the thrushes, and to redesign strawberries and raspberries according to our own standards.
we've thus seen that the difference between gigantic supermarket strawberries and tiny wild ones is just one example of the various features distinguishing cultivated plants from their wild ancestors. Those differences arose initially from natural variation among the wild plants themselves. Some of it, such as the variation in berry size or in nut bitterness, would have been readily noticed by ancient farmers. Other variation, such as that in seed dispersal mechanisms or seed dormancy, would have gone unrecognized by humans before the rise of modern botany. But whether or not the selection of wild edible plants by ancient hikers relied on conscious or unconscious criteria, the resulting evolution of wild plants into crops was at first an unconscious process. It followed inevitably from our selecting among wild plant individuals, and from competition among plant individuals in gardens favoring individuals different from those favored in the wild.
That's why Darwin, in his great book On the Origin of Species, didn't start with an account of natural selection. His first chapter is instead a lengthy account of how our domesticated plants and animals arose through artificial selection by humans. Rather than discussing the Galapagos Island birds that we usually associate with him, Darwin began by discussing—how farmers develop varieties of gooseberries! He wrote, "I have seen great surprise expressed in horticultural works at the wonderful skill of gardeners, in having produced such splendid results from such poor materials; but the art has been simple, and as far as the final result is concerned, has been followed almost unconsciously. It has consisted in always cultivating the best-known variety, sowing its seeds, and, when a slightly better variety chanced to appear, selecting it, and so onwards." Those principles of crop development by artificial selection still serve as our most understandable model of the origin of species by natural selection.

CHAPTER

8apples or indians

WE HAVE JUST SEEN HOW PEOPLES OF SOME REGIONS began to cultivate wild plant species, a step with momentous unforeseen consequences for their lifestyle and their descendants' place in history. Let us now return to our questions: Why did agriculture never arise independently in some fertile and highly suitable areas, such as California, Europe, temperate Australia, and subequatorial Africa? Why, among the areas where agriculture did arise independently, did it develop much earlier in some than in others?

Two contrasting explanations suggest themselves: problems with the local people, or problems with the locally available wild plants. On the one hand, perhaps almost any well-watered temperate or tropical area of the globe offers enough species of wild plants suitable for domestication. In that case, the explanation for agriculture's failure to develop in some of those areas would lie with cultural characteristics of their peoples. On the other hand, perhaps at least some humans in any large area of the globe would have been receptive to the experimentation that led to domestication. Only the lack of suitable wild plants might then explain why food production did not evolve in some areas.
As we shall see in the next chapter, the corresponding problem for oniestication of big wild mammals proves easier to solve, because there

I 3 2 •

GUNS,GERMS, AND STEELare many fewer species of them than of plants. The world holds only about 148 species of large wild mammalian terrestrial herbivores or omnivores, the large mammals that could be considered candidates for domestication. Only a modest number of factors determines whether a mammal is suitable for domestication. It's thus straightforward to review a region's big mammals and to test whether the lack of mammal domestication in some regions was due to the unavailability of suitable wild species, rather than to local peoples.
That approach would be much more difficult to apply to plants because of the sheer number—200,000—of species of wild flowering plants, the plants that dominate vegetation on the land and that have furnished almost all of our crops. We can't possibly hope to examine all the wild plant species of even a circumscribed area like California, and to assess how many of them would have been domesticable. But we shall now see how to get around that problem.
When one hears that there are so many species of flowering plants, one's first reaction might be as follows: surely, with all those wild plant species on Earth, any area with a sufficiently benign climate must have had more than enough species to provide plenty of candidates for crop development.
But then reflect that the vast majority of wild plants are unsuitable for obvious reasons: they are woody, they produce no edible fruit, and their leaves and roots are also inedible. Of the 200,000 wild plant species, only a few thousand are eaten by humans, and just a few hundred of these have been more or less domesticated. Even of these several hundred crops, most provide minor supplements to our diet and would not by themselves have sufficed to support the rise of civilizations. A mere dozen species account for over 80 percent of the modern world's annual tonnage of all crops. Those dozen blockbusters are the cereals wheat, corn, rice, barley, and sorghum; the pulse soybean; the roots or tubers potato, manioc, and sweet potato; the sugar sources sugarcane and sugar beet; and the fruit banana. Cereal crops alone now account for more than half of the calories consumed by the world's human populations. With so few major crops in the world, all of them domesticated thousands of years ago, it's less surprising that many areas of the world had no wild native plants at all of outstanding potential. Our failure to domesticate even a single major new food

APPLES

OR INDIANS • 133plant in modern times suggests that ancient peoples really may have explored virtually all useful wild plants and domesticated all the ones worth domesticating.
Yet some of the world's failures to domesticate wild plants remain hard to explain. The most flagrant cases concern plants that were domesticated in one area but not in another. We can thus be sure that it was indeed possible to develop the wild plant into a useful crop, and we have to ask why that wild species was not domesticated in certain areas.
A typical puzzling example comes from Africa. The important cereal sorghum was domesticated in Africa's Sahel zone, just south of the Sahara. It also occurs as a wild plant as far south as southern Africa, yet neither it nor any other plant was cultivated in southern Africa until the arrival of the whole crop package that Bantu farmers brought from Africa north of the equator 2,000 years ago. Why did the native peoples of southern Africa not domesticate sorghum for themselves?
Equally puzzling is the failure of people to domesticate flax in its wild range in western Europe and North Africa, or einkorn wheat in its wild range in the southern Balkans. Since these two plants were among the first eight crops of the Fertile Crescent, they were presumably among the most readily domesticated of all wild plants. They were adopted for cultivation in those areas of their wild range outside the Fertile Crescent as soon as they arrived with the whole package of food production from the Fertile Crescent. Why, then, had peoples of those outlying areas not already begun to grow them of their own accord?
Similarly, the four earliest domesticated fruits of the Fertile Crescent all had wild ranges stretching far beyond the eastern Mediterranean, where they appear to have been first domesticated: the olive, grape, and fig occurred west to Italy and Spain and Northwest Africa, while the date palm extended to all of North Africa and Arabia. These four were evidently among the easiest to domesticate of all wild fruits. Why did peoples outside the Fertile Crescent fail to domesticate them, and begin to grow them only when they had already been domesticated in the eastern Mediterranean and arrived thence as crops?
Other striking examples involve wild species that were not domesticated in areas where food production never arose spontaneously, even though those wild species had close relatives domesticated elsewhere. For example, the olive Olea europea was domesticated in the eastern Mediterranean. There are about 40 other species of olives in tropical and southern

134

" GUNS, GERMS, AND STEELAfrica, southern Asia, and eastern Australia, some of them closely related to Olea europea, but none of them was ever domesticated. Similarly, while a wild apple species and a wild grape species were domesticated in Eurasia, there are many related wild apple and grape species in North America, some of which have in modern times been hybridized with the crops derived from their wild Eurasian counterparts in order to improve those crops. Why, then, didn't Native Americans domesticate those apparently useful apples and grapes themselves?
One can go on and on with such examples. But there is a fatal flaw in this reasoning: plant domestication is not a matter of hunter-gatherers' domesticating a single plant and otherwise carrying on unchanged with their nomadic lifestyle. Suppose that North American wild apples really would have evolved into a terrific crop if only Indian hunter-gatherers had settled down and cultivated them. But nomadic hunter-gatherers would not throw over their traditional way of life, settle in villages, and start tending apple orchards unless many other domesticable wild plants and animals were available to make a sedentary food-producing existence competitive with a hunting-gathering existence.
How, in short, do we assess the potential of an entire local flora for domestication? For those Native Americans who failed to domesticate North American apples, did the problem really lie with the Indians or with the apples?
In order to answer this question, we shall now compare three regions that lie at opposite extremes among centers of independent domestication. As we have seen, one of them, the Fertile Crescent, was perhaps the earliest center of food production in the world, and the site of origin of several of the modern world's major crops and almost all of its major domesticated animals. The other two regions, New Guinea and the eastern United States, did domesticate local crops, but these crops were very few in variety, only one of them gained worldwide importance, and the resulting food package failed to support extensive development of human technology and political organization as in the Fertile Crescent. In the light of this comparison, we shall ask: Did the flora and environment of the Fertile Crescent have clear advantages over those of New Guinea and the eastern United States?
one of the central facts of human history is the early importance of the part of Southwest Asia known as the Fertile Crescent (because of the

APPLES

OR INDIANS • 135crescent-like shape of its uplands on a map: see Figure 8.1). That area appears to have been the earliest site for a whole string of developments, including cities, writing, empires, and what we term (for better or worse) civilization. All those developments sprang, in turn, from the dense human populations, stored food surpluses, and feeding of nonfarming specialists made possible by the rise of food production in the form of crop cultivation and animal husbandry. Food production was the first of those major innovations to appear in the Fertile Crescent. Hence any attempt to understand the origins of the modern world must come to grips with the question why the Fertile Crescent's domesticated plants and animals gave it such a potent head start.

tgure 8.1. The Fertile Crescent, encompassing sites of food production

before 7000 b.c.Fortunately, the Fertile Crescent is by far the most intensively studied and best understood part of the globe as regards the rise of agriculture. For most crops domesticated in or near the Fertile Crescent, the wild plant ancestor has been identified; its close relationship to the crop has been proven by genetic and chromosomal studies; its wild geographic range is known; its changes under domestication have been identified and are often understood at the level of single genes; those changes can be observed in

I 3 6 • GUNS, GERMS, AND STEEL

successive layers of the archaeological record; and the approximate place and time of domestication are known. I don't deny that other areas, notably China, also had advantages as early sites of domestication, but those advantages and the resulting development of crops can be specified in much more detail for the Fertile Crescent.
One advantage of the Fertile Crescent is that it lies within a zone of so-called Mediterranean climate, a climate characterized by mild, wet winters and long, hot, dry summers. That climate selects for plant species able to survive the long dry season and to resume growth rapidly upon the return of the rains. Many Fertile Crescent plants, especially species of cereals and pulses, have adapted in a way that renders them useful to humans: they are annuals, meaning that the plant itself dries up and dies in the dry season.
Within their mere one year of life, annual plants inevitably remain small herbs. Many of them instead put much of their energy into producing big seeds, which remain dormant during the dry season and are then ready to sprout when the rains come. Annual plants therefore waste little energy on making inedible wood or fibrous stems, like the body of trees and bushes. But many of the big seeds, notably those of the annual cereals and pulses, are edible by humans. They constitute 6 of the modern world's 12 major crops. In contrast, if you live near a forest and look out your window, the plant species that you see will tend to be trees and shrubs, most of whose body you cannot eat and which put much less of their energy into edible seeds. Of course, some forest trees in areas of wet climate do produce big edible seeds, but these seeds are not adapted to surviving a long dry season and hence to long storage by humans.
A second advantage of the Fertile Crescent flora is that the wild ancestors of many Fertile Crescent crops were already abundant and highly productive, occurring in large stands whose value must have been obvious to hunter-gatherers. Experimental studies in which botanists have collected seeds from such natural stands of wild cereals, much as hunter-gatherers must have been doing over 10,000 years ago, show that annual harvests of up to nearly a ton of seeds per hectare can be obtained, yielding 50 kilocalories of food energy for only one kilocalorie of work expended. By collecting huge quantities of wild cereals in a short time when the seeds were ripe, and storing them for use as food through the rest of the year, some hunting-gathering peoples of the Fertile Crescent had already settled down in permanent villages even before they began to cultivate plants.
Since Fertile Crescent cereals were so productive in the wild, few addi-

APPLES

OR INDIANS • 137tional changes had to be made in them under cultivation. As we discussed in the preceding chapter, the principal changes—the breakdown of the natural systems of seed dispersal and of germination inhibition—evolved automatically and quickly as soon as humans began to cultivate the seeds in fields. The wild ancestors of our wheat and barley crops look so similar to the crops themselves that the identity of the ancestor has never been in doubt. Because of this ease of domestication, big-seeded annuals were the first, or among the first, crops developed not only in the Fertile Crescent but also in China and the Sahel.
Contrast this quick evolution of wheat and barley with the story of corn, the leading cereal crop of the New World. Corn's probable ancestor, a wild plant known as teosinte, looks so different from corn in its seed and flower structures that even its role as ancestor has been hotly debated by botanists for a long time. Teosinte's value as food would not have impressed hunter-gatherers: it was less productive in the wild than wild wheat, it produced much less seed than did the corn eventually developed from it, and it enclosed its seeds in inedible hard coverings. For teosinte to become a useful crop, it had to undergo drastic changes in its reproductive biology, to increase greatly its investment in seeds, and to lose those rock-like coverings of its seeds. Archaeologists are still vigorously debating how many centuries or millennia of crop development in the Americas were required for ancient corn cobs to progress from a tiny size up to the size of a human thumb, but it seems clear that several thousand more years were then required for them to reach modern sizes. That contrast between the immediate virtues of wheat and barley and the difficulties posed by teosinte may have been a significant factor in the differing developments of New World and Eurasian human societies.
A third advantage of the Fertile Crescent flora is that it includes a high percentage of hermaphroditic "setters"—that is, plants that usually pollinate themselves but that are occasionally cross-pollinated. Recall that most wild plants either are regularly cross-pollinated hermaphrodites or consist of separate male and female individuals that inevitably depend on another individual for pollination. Those facts of reproductive biology vexed early farmers, because, as soon as they had located a productive mutant plant, its offspring would cross-breed with other plant individuals and thereby lose their inherited advantage. As a result, most crops belong to the small percentage of wild plants that either are hermaphrodites usually pollinat-mg themselves or else reproduce without sex by propagating vegetatively

138

• GUNS, GERMS, AND STEEL(for example, by a root that genetically duplicates the parent plant). Thus, the high percentage of hermaphroditic selfers in the Fertile Crescent flora aided early farmers, because it meant that a high percentage of the wild flora had a reproductive biology convenient for humans.
Selfers were also convenient for early farmers in that they occasionally did become cross-pollinated, thereby generating new varieties among which to select. That occasional cross-pollination occurred not only between individuals of the same species, but also between related species to produce interspecific hybrids. One such hybrid among Fertile Crescent selfers, bread wheat, became the most valuable crop in the modern world.
Of the first eight significant crops to have been domesticated in the Fertile Crescent, all were selfers. Of the three selfer cereals among them— einkorn wheat, emmer wheat, and barley—the wheats offered the additional advantage of a high protein content, 8-14 percent. In contrast, the most important cereal crops of eastern Asia and of the New World—rice and corn, respectively—had a lower protein content that posed significant nutritional problems.
Those were some of the advantages that the Fertile Crescent's flora afforded the first farmers: it included an unusually high percentage of wild plants suitable for domestication. However, the Mediterranean climate zone of the Fertile Crescent extends westward through much of southern Europe and northwestern Africa. There are also zones of similar Mediterranean climates in four other parts of the world: California, Chile, southwestern Australia, and South Africa (Figure 8.2). Yet those other Mediterranean zones not only failed to rival the Fertile Crescent as early sites of food production; they never gave rise to indigenous agriculture at all. What advantage did that particular Mediterranean zone of western Eurasia enjoy?
It turns out that it, and especially its Fertile Crescent portion, possessed at least five advantages over other Mediterranean zones. First, western Eurasia has by far the world's largest zone of Mediterranean climate. As a result, it has a high diversity of wild plant and animal species, higher than in the comparatively tiny Mediterranean zones of southwestern Australia and Chile. Second, among Mediterranean zones, western Eurasia's experiences the greatest climatic variation from season to season and year to year. That variation favored the evolution, among the flora, of an espe-

APPLES

OR INDIANS • 139

v?-:iJ>*V: Mi \ffK'-' *

:;:.k . v-.K>:?*:> *-;&••::'.•• •.'!.;'" '•:>! .!wSlЈ –.– '>i •; ;.?

Figure 8.2. The world's zones of Mediterranean climate.

cially high percentage of annual plants. The combination of these two factors—a high diversity of species and a high percentage of annuals—means that western Eurasia's Mediterranean zone is the one with by far the highest diversity of annuals.
The significance of that botanical wealth for humans is illustrated by the geographer Mark Blunder's studies of wild grass distributions. Among the world's thousands of wild grass species, Blumler tabulated the 56 with the largest seeds, the cream of nature's crop: the grass species with seeds at least 10 times heavier than the median grass species (see Table 8.1). Virtually all of them are native to Mediterranean zones or other seasonally dry environments. Furthermore, they are overwhelmingly concentrated in the Fertile Crescent or other parts of western Eurasia's Mediterranean zone, which offered a huge selection to incipient farmers: about 32 of the world's 56 prize wild grasses! Specifically, barley and emmer wheat, the two earliest important crops of the Fertile Crescent, rank respectively 3rd and 13th in seed size among those top 56. In contrast, the Mediterranean zone of Chile offered only two of those species, California and southern Africa just one each, and southwestern Australia none at all. That fact alone goes a long way toward explaining the course of human history.
A third advantage of the Fertile Crescent's Mediterranean zone is that

I 4 O • GUNS, GERMS, AND STEEL

table 8.1 World Distribution of Large-Seeded Grass Species

Area

Number of Species

West Asia, Europe, North Africa

33

Mediterranean zone

32

England

1

East Asia

6

Sub-Saharan Africa

4

Americas

11

North America

4

Mesoamerica

5

South America

2

Northern Australia

2

Total:

56

Table 12,1 of Mark Blunder's Ph.D. dissertation, "Seed Weight and Environment in Mediterranean-type Grasslands in California and Israel" (University of California, Berkeley, 1992), listed the world's 56 heaviest-seeded wild grass species (excluding bamboos) for which data were available. Grain weight in those species ranged from 10 milligrams to over 40 milligrams, about 10 times greater than the median value for all of the world's grass species. Those 56 species make up less than 1 percent of the world's grass species. This table shows that these prize grasses are overwhelmingly concentrated in the Mediterranean zone of western Eurasia.

it provides a wide range of altitudes and topographies within a short distance. Its range of elevations, from the lowest spot on Earth {the Dead Sea) to mountains of 18,000 feet {near Teheran), ensures a corresponding variety of environments, hence a high diversity of the wild plants serving as potential ancestors of crops. Those mountains are in proximity to gentle lowlands with rivers, flood plains, and deserts suitable for irrigation agriculture. In contrast, the Mediterranean zones of southwestern Australia and, to a lesser degree, of South Africa and western Europe offer a narrower range of altitudes, habitats, and topographies.
The range of altitudes in the Fertile Crescent meant staggered harvest seasons: plants at higher elevations produced seeds somewhat later than plants at lower elevations. As a result, hunter-gatherers could move up a mountainside harvesting grain seeds as they matured, instead of being overwhelmed by a concentrated harvest season at a single altitude, where all grains matured simultaneously. When cultivation began, it was a simple

APPLES

OR INDIANS • 141matter for the first farmers to take the seeds of wild cereals growing on hillsides and dependent on unpredictable rains, and to plant those seeds in the damp valley bottoms, where they would grow reliably and be less dependent on rain.
The Fertile Crescent's biological diversity over small distances contributed to a fourth advantage—its wealth in ancestors not only of valuable crops but also of domesticated big mammals. As we shall see, there were few or no wild mammal species suitable for domestication in the other Mediterranean zones of California, Chile, southwestern Australia, and South Africa. In contrast, four species of big mammals—the goat, sheep, pig and cow—were domesticated very early in the Fertile Crescent, possibly earlier than any other animal except the dog anywhere else in the world. Those species remain today four of the world's five most important domesticated mammals (Chapter 9). But their wild ancestors were commonest in slightly different parts of the Fertile Crescent, with the result that the four species were domesticated in different places: sheep possibly in the central part, goats either in the eastern part at higher elevations (the Zagros Mountains of Iran) or in the southwestern part (the Levant), pigs in the north-central part, and cows in the western part, including Anatolia. Nevertheless, even though the areas of abundance of these four wild progenitors thus differed, all four lived in sufficiently close proximity that they were readily transferred after domestication from one part of the Fertile Crescent to another, and the whole region ended up with all four species.
Agriculture was launched in the Fertile Crescent by the early domestication of eight crops, termed "founder crops" (because they founded agriculture in the region and possibly in the world). Those eight founders were the cereals emmer wheat, einkorn wheat, and barley; the pulses lentil, pea, chickpea, and bitter vetch; and the fiber crop flax. Of these eight, only two, flax and barley, range in the wild at all widely outside the Fertile Crescent and Anatolia. Two of the founders had very small ranges in the wild, chickpea being confined to southeastern Turkey and emmer wheat to the Fertile Crescent itself. Thus, agriculture could arise in the Fertile Crescent from domestication of locally available wild plants, without having to wait for the arrival of crops derived from wild plants domesticated elsewhere. Conversely, two of the eight founder crops could not have been domesticated anywhere in the world except in the Fertile Crescent, since they did not occur wild elsewhere.
Thanks to this availability of suitable wild mammals and plants, early peoples of the Fertile Crescent could quickly assemble a potent and bal-

142.

" GUNS, GERMS, AND STEELanced biological package for intensive food production. That package comprised three cereals, as the main carbohydrate sources; four pulses, with 20-25 percent protein, and four domestic animals, as the main protein sources, supplemented by the generous protein content of wheat; and flax as a source of fiber and oil (termed linseed oil: flax seeds are about 40 percent oil). Eventually, thousands of years after the beginnings of animal domestication and food production, the animals also began to be used for milk, wool, plowing, and transport. Thus, the crops and animals of the Fertile Crescent's first farmers came to meet humanity's basic economic needs: carbohydrate, protein, fat, clothing, traction, and transport.
A final advantage of early food production in the Fertile Crescent is that it may have faced less competition from the hunter-gatherer lifestyle than that in some other areas, including the western Mediterranean. Southwest Asia has few large rivers and only a short coastline, providing relatively meager aquatic resources (in the form of river and coastal fish and shellfish). One of the important mammal species hunted for meat, the gazelle, originally lived in huge herds but was overexploited by the growing human population and reduced to low numbers. Thus, the food production package quickly became superior to the hunter-gatherer package. Sedentary villages based on cereals were already in existence before the rise of food production and predisposed those hunter-gatherers to agriculture and herding. In the Fertile Crescent the transition from hunting-gathering to food production took place relatively fast: as late as 9000 b.c. people still had no crops and domestic animals and were entirely dependent on wild foods, but by 6000 b.c. some societies were almost completely dependent on crops and domestic animals.
The situation in Mesoamerica contrasts strongly: that area provided only two domesticable animals (the turkey and the dog), whose meat yield was far lower than that of cows, sheep, goats, and pigs; and corn, Meso-america's staple grain, was, as I've already explained, difficult to domesticate and perhaps slow to develop. As a result, domestication may not have begun in Mesoamerica until around 3500 b.c. (the date remains very uncertain); those first developments were undertaken by people who were still nomadic hunter-gatherers; and settled villages did not arise there until around 1500 b.c.
inallthis discussion of the Fertile Crescent's advantages for the early rise of food production, we have not had to invoke any supposed advan-

APPLES

OR INDIANS • 143taees of Fertile Crescent peoples themselves. Indeed, I am unaware of anyone's even seriously suggesting any supposed distinctive biological features of the region's peoples that might have contributed to the potency of its food production package. Instead, we have seen that the many distinctive features of the Fertile Crescent's climate, environment, wild plants, and animals together provide a convincing explanation.
Since the food production packages arising indigenously in New Guinea and in the eastern United States were considerably less potent, might the explanation there lie with the peoples of those areas? Before turning to those regions, however, we must consider two related questions arising in regard to any area of the world where food production never developed independently or else resulted in a less potent package. First, do hunter-gatherers and incipient farmers really know well all locally available wild species and their uses, or might they have overlooked potential ancestors of valuable crops? Second, if they do know their local plants and animals, do they exploit that knowledge to domesticate the most useful available species, or do cultural factors keep them from doing so?
As regards the first question, an entire field of science, termed ethnobiol-ogy, studies peoples' knowledge of the wild plants and animals in their environment. Such studies have concentrated especially on the world's few surviving hunting-gathering peoples, and on farming peoples who still depend heavily on wild foods and natural products. The studies generally show that such peoples are walking encyclopedias of natural history, with individual names (in their local language) for as many as a thousand or more plant and animal species, and with detailed knowledge of those species' biological characteristics, distribution, and potential uses. As people become increasingly dependent on domesticated plants and animals, this traditional knowledge gradually loses its value and becomes lost, until one arrives at modern supermarket shoppers who could not distinguish a wild grass from a wild pulse.
Here's a typical example. For the last 33 years, while conducting biological exploration in New Guinea, I have been spending my field time there constantly in the company of New Guineans who still use wild plants and animals extensively. One day, when my companions of the Fore tribe and were starving in the jungle because another tribe was blocking our return to our supply base, a Fore man returned to camp with a large rucksack full of mushrooms he had found, and started to roast them. Dinner at ast! But then I had an unsettling thought: what if the mushrooms were poisonous?

144 ' GUNS, GERMS, AND STEEL

I patiently explained to my Fore companions that I had read about some mushrooms' being poisonous, that I had heard of even expert American mushroom collectors' dying because of the difficulty of distinguishing safe from dangerous mushrooms, and that although we were all hungry, it just wasn't worth the risk. At that point my companions got angry and told me to shut up and listen while they explained some things to me. After I had been quizzing them for years about names of hundreds of trees and birds, how could I insult them by assuming they didn't have names for different mushrooms? Only Americans could be so stupid as to confuse poisonous mushrooms with safe ones. They went on to lecture me about 29 types of edible mushroom species, each species' name in the Fore language, and where in the forest one should look for it. This one, the tdnti, grew on trees, and it was delicious and perfectly edible.
Whenever I have taken New Guineans with me to other parts of their island, they regularly talk about local plants and animals with other New Guineans whom they meet, and they gather potentially useful plants and bring them back to their home villages to try planting them. My experiences with New Guineans are paralleled by those of ethnobiologists studying traditional peoples elsewhere. However, all such peoples either practice at least some food production or are the partly acculturated last remnants of the world's former hunter-gatherer societies. Knowledge of wild species was presumably even more detailed before the rise of food production, when everyone on Earth still depended entirely on wild species for food. The first farmers were heirs to that knowledge, accumulated through tens of thousands of years of nature observation by biologically modern humans living in intimate dependence on the natural world. It therefore seems extremely unlikely that wild species of potential value would have escaped the notice of the first farmers.
The other, related question is whether ancient hunter-gatherers and farmers similarly put their ethnobiological knowledge to good use in selecting wild plants to gather and eventually to cultivate. One test comes from an archaeological site at the edge of the Euphrates Valley in Syria, called Tell Abu Hureyra. Between 10,000 and 9000 b.c. the people living there may already have been residing year-round in villages, but they were still hunter-gatherers; crop cultivation began only in the succeeding millennium. The archaeologists Gordon Hiliman, Susan Colledge, and David Harris retrieved large quantities of charred plant remains from the site, probably representing discarded garbage of wild plants gathered elsewhere

APPLES OR INDIANS • 145

and brought to the site by its residents. The scientists analyzed over 700 samples* each containing an average of over 500 identifiable seeds belonging to over 70 plant species. It turned out that the villagers were collecting a prodigious variety (157 species!) of plants identified by their charred seeds, not to mention other plants that cannot now be identified.
Were those naive villagers collecting every type of seed plant that they found, bringing it home, poisoning themselves on most of the species, and nourishing themselves from only a few species? No, they were not so silly. While 157 species sounds like indiscriminate collecting, many more species growing wild in the vicinity were absent from the charred remains. The 15" selected species fall into three categories. Many of them have seeds that are nonpoisonous and immediately edible. Others, such as pulses and members of the mustard family, have toxic seeds, but the toxins are easily removed, leaving the seeds edible. A few seeds belong to species traditionally used as sources of dyes or medicine. The many wild species not represented among the 157 selected are ones that would have been useless or harmful to people, including all of the most toxic weed species in the environment.
Thus, the hunter-gatherers of Tell Abu Hureyra were not wasting time and endangering themselves by collecting wild plants indiscriminately. Instead, they evidently knew the local wild plants as intimately as do modern New Guineans, and they used that knowledge to select and bring home only the most useful available seed plants. But those gathered seeds would have constituted the material for the unconscious first steps of plant domestication.
My other example of how ancient peoples apparently used their ethno-biological knowledge to good effect comes from the Jordan Valley in the ninth millennium b.c., the period of the earliest crop cultivation there. The valley's first domesticated cereals were barley and emmer wheat, which are still among the world's most productive crops today. But, as at Tell Abu Hureyra, hundreds of other seed-bearing wild plant species must have grown in the vicinity, and a hundred or more of them would have been edible and gathered before the rise of plant domestication. What was it about barley and emmer wheat that caused them to be the first crops? Were those first Jordan Valley farmers botanical ignoramuses who didn't know what they were doing? Or were barley and emmer wheat actually the best of the local wild cereals that they could have selected?
Two Israeli scientists, Ofer Bar-Yosef and Mordechai Kislev, tackled this

146

• GUNS, GERMS, AND STEELquestion by examining the wild grass species still growing wild in the valley today. Leaving aside species with small or unpalatable seeds, they picked out 23 of the most palatable and largest-seeded wild grasses. Not surprisingly, barley and emmer wheat were on that list.
But it wasn't true that the 21 other candidates would have been equally useful. Among those 23, barley and emmer wheat proved to be the best by many criteria. Emmer wheat has the biggest seeds and barley the second biggest. In the wild, barley is one of the 4 most abundant of the 23 species, while emmer wheat is of medium abundance. Barley has the further advantage that its genetics and morphology permit it to evolve quickly the useful changes in seed dispersal and germination inhibition that we discussed in the preceding chapter. Emmer wheat, however, has compensating virtues: it can be gathered more efficiently than barley, and it is unusual among cereals in that its seeds do not adhere to husks. As for the other 21 species, their drawbacks include smaller seeds, in many cases lower abundance, and in some cases their being perennial rather than annual plants, with the consequence that they would have evolved only slowly under domestication.
Thus, the first farmers in the Jordan Valley selected the 2 very best of the 23 best wild grass species available to them. Of course, the evolutionary changes (following cultivation) in seed dispersal and germination inhibition would have been unforeseen consequences of what those first farmers were doing. But their initial selection of barley and emmer wheat rather than other cereals to collect, bring home, and cultivate would have been conscious and based on the easily detected criteria of seed size, palat-ability, and abundance.
This example from the Jordan Valley, like that from Tell Abu Hureyra, illustrates that the first farmers used their detailed knowledge of local species to their own benefit. Knowing far more about local plants than all but a handful of modern professional botanists, they would hardly have failed to cultivate any useful wild plant species that was comparably suitable for domestication.
we can now examine what local farmers, in two parts of the world (New Guinea and the eastern United States) with indigenous but apparently deficient food production systems compared to that of the Fertile Crescent, actually did when more-productive crops arrived from else-

APPLES

OR INDIANS • 147here If it turned out that such crops did not become adopted for cultural or other reasons, we would be left with a nagging doubt. Despite all our reasoning so far, we would still have to suspect that the local wild flora harbored some ancestor of a potential valuable crop that local farmers failed to exploit because of similar cultural factors. These two examples will also demonstrate in detail a fact critical to history: that indigenous crops from different parts of the globe were not equally productive.
New Guinea, the largest island in the world after Greenland, lies just north of Australia and near the equator. Because of its tropical location and great diversity in topography and habitats, New Guinea is rich in both plant and animal species, though less so than continental tropical areas because it is an island. People have been living in New Guinea for at least 40,000 years—much longer than in the Americas, and slightly longer than anatomically modern peoples have been living in western Europe. Thus, New Guineans have had ample opportunity to get to know their local flora and fauna. Were they motivated to apply this knowledge to developing food production?
I mentioned already that the adoption of food production involved a competition between the food producing and the hunting-gathering lifestyles. Hunting-gathering is not so rewarding in New Guinea as to remove the motivation to develop food production. In particular, modern New Guinea hunters suffer from the crippling disadvantage of a dearth of wild game: there is no native land animal larger than a 100-pound flightless bird (the cassowary) and a 50-pound kangaroo. Lowland New Guineans on the coast do obtain much fish and shellfish, and some lowlanders in the interior still live today as hunter-gatherers, subsisting especially on wild sago palms. But no peoples still live as hunter-gatherers in the New Guinea highlands; all modern highlanders are instead farmers who use wild foods only to supplement their diets. When highlanders go into the forest on hunting trips, they take along garden-grown vegetables to feed themselves. If they have the misfortune to run out of those provisions, even they starve to death despite their detailed knowledge of locally available wild foods. Since the hunting-gathering lifestyle is thus nonviable in much of modern New Guinea, it comes as no surprise that all New Guinea highlanders and most lowlanders today are settled farmers with sophisticated systems of food production. Extensive, formerly forested areas of the highlands were converted by traditional New Guinea farmers to fenced, drained, intensively managed field systems supporting dense human populations.

I 4 8 •

GUNS,GERMS, AND STEELArchaeological evidence shows that the origins of New Guinea agriculture are ancient, dating to around 7000 b.c. At those early dates all the landmasses surrounding New Guinea were still occupied exclusively by hunter-gatherers, so this ancient agriculture must have developed independently in New Guinea. While unequivocal remains of crops have not been recovered from those early fields, they are likely to have included some of the same crops that were being grown in New Guinea at the time of European colonization and that are now known to have been domesticated locally from wild New Guinea ancestors. Foremost among these local domesticates is the modern world's leading crop, sugarcane, of which the annual tonnage produced today nearly equals that of the number two and number three crops combined (wheat and corn). Other crops of undoubted New Guinea origin are a group of bananas known as Australimusa bananas, the nut tree Canarium indicum, and giant swamp taro, as well as various edible grass stems, roots, and green vegetables. The breadfruit tree and the root crops yams and (ordinary) taro may also be New Guinean domesticates, although that conclusion remains uncertain because their wild ancestors are not confined to New Guinea but are distributed from New Guinea to Southeast Asia. At present we lack evidence that could resolve the question whether they were domesticated in Southeast Asia, as traditionally assumed, or independently or even only in New Guinea.However, it turns out that New Guinea's biota suffered from three severe limitations. First, no cereal crops were domesticated in New Guinea, whereas several vitally important ones were domesticated in the Fertile Crescent, Sahel, and China. In its emphasis instead on root and tree crops, New Guinea carries to an extreme a trend seen in agricultural systems in other wet tropical areas (the Amazon, tropical West Africa, and Southeast Asia), whose farmers also emphasized root crops but did manage to come up with at least two cereals (Asian rice and a giant-seeded Asian cereal called Job's tears). A likely reason for the failure of cereal agriculture to arise in New Guinea is a glaring deficiency of the wild starting material: not one of the world's 56 largest-seeded wild grasses is native there.
Second, the New Guinea fauna included no domesticable large mammal 1 species whatsoever. The sole domestic animals of modern New Guinea, the pig and chicken and dog, arrived from Southeast Asia by way of Indonesia within the last several thousand years. As a result, while New Guinea

APPLES

OR INDIANS • I 4 9lowlanders obtain protein from the fish they catch, New Guinea highland farmer populations suffer from severe protein limitation, because the staple crops that provide most of their calories (taro and sweet potato) are low in protein. Taro, for example, consists of barely 1 percent protein, much worse than even white rice, and far below the levels of the Fertile Crescent's wheats and pulses (8-14 percent and 20-25 percent protein, respectively).
Children in the New Guinea highlands have the swollen bellies characteristic of a high-bulk but protein-deficient diet. New Guineans old and young routinely eat mice, spiders, frogs, and other small animals that peoples elsewhere with access to large domestic mammals or large wild game species do not bother to eat. Protein starvation is probably also the ultimate reason why cannibalism was widespread in traditional New Guinea highland societies.
Finally, in former times New Guinea's available root crops were limiting for calories as well as for protein, because they do not grow well at the high elevations where many New Guineans live today. Many centuries ago, however, a new root crop of ultimately South American origin, the sweet potato, reached New Guinea, probably by way of the Philippines, where it had been introduced by Spaniards. Compared with taro and other presumably older New Guinea root crops, the sweet potato can be grown up to higher elevations, grows more quickly, and gives higher yields per acre cultivated and per hour of labor. The result of the sweet potato's arrival was a highland population explosion. That is, even though people had been farming in the New Guinea highlands for many thousands of years before sweet potatoes were introduced, the available local crops had limited them in the population densities they could attain, and in the elevations they could occupy.
In short, New Guinea offers an instructive contrast to the Fertile Crescent. Like hunter-gatherers of the Fertile Crescent, those of New Guinea did evolve food production independently. However, their indigenous food production was restricted by the local absence of domesticable cereals, pulses, and animals, by the resulting protein deficiency in the highlands, and by limitations of the locally available root crops at high elevations. Yet New Guineans themselves know as much about the wild plants and animals available to them as any peoples on Earth today. They can be expected to have discovered and tested any wild plant species worth domesticating. They are perfectly capable of recognizing useful additions

15o

" GUNS, GERMS, AND STEELto their crop larder, as is shown by their exuberant adoption of the sweet potato when it arrived. That same lesson is being driven home again in New Guinea today, as those tribes with preferential access to introduced new crops and livestock (or with the cultural willingness to adopt them) expand at the expense of tribes without that access or willingness. Thus, the limits on indigenous food production in New Guinea had nothing to do with New Guinea peoples, and everything with the New Guinea biota and environment.
OUR other example of indigenous agriculture apparently constrained by the local flora comes from the eastern United States. Like New Guinea, that area supported independent domestication of local wild plants. However, early developments are much better understood for the eastern United States than for New Guinea: the crops grown by the earliest farmers have been identified, and the dates and crop sequences of local domestication are known. Well before other crops began to arrive from elsewhere, Native Americans settled in eastern U.S. river valleys and developed intensified food production based on local crops. Hence they were in a position to take advantage of the most promising wild plants. Which ones did they actually cultivate, and how did the resulting local crop package compare with the Fertile Crescent's founder package?
It turns out that the eastern U.S. founder crops were four plants domesticated in the period 2500-1500 b.c., a full 6,000 years after wheat and barley domestication in the Fertile Crescent. A local species of squash provided small containers, as well as yielding edible seeds. The remaining three founders were grown solely for their edible seeds (sunflower, a daisy relative called sumpweed, and a distant relative of spinach called goose-foot).
But four seed crops and a container fall far short of a complete food production package. For 2,000 years those founder crops served only as minor dietary supplements while eastern U.S. Native Americans continued to depend mainly on wild foods, especially wild mammals and waterbirds, fish, shellfish, and nuts. Farming did not supply a major part of their diet until the period 500-200 b.c., after three more seed crops (knotweed, maygrass, and little barley) had been brought into cultivation.
A modern nutritionist would have applauded those seven eastern U.S.

APPLES

OR INDIANSrops. All of them were high in protein — 17-32 percent, compared with cl4 percent for wheat, 9 percent for corn, and even lower for barley and white rice. Two of them, sunflower and sumpweed, were also high in oil (45-47 percent). Sumpweed, in particular, would have been a nutritionist's ultimate dream, being 32 percent protein and 45 percent oil. Why aren't we still eating those dream foods today?
Alas, despite their nutritional advantage, most of these eastern U.S. crops suffered from serious disadvantages in other respects. Goosefoot, knotweed, little barley, and maygrass had tiny seeds, with volumes only one-tenth that of wheat and barley seeds. Worse yet, sumpweed is a wind-pollinated relative of ragweed, the notorious hayfever-causing plant. Like ragweed's, sumpweed's pollen can cause hayfever where the plant occurs in abundant stands. If that doesn't kill your enthusiasm for becoming a sumpweed farmer, be aware that it has a strong odor objectionable to some people and that handling it can cause skin irritation.
Mexican crops finally began to reach the eastern United States by trade routes after a.d. 1. Corn arrived around a.d. 200, but its role remained very minor for many centuries. Finally, around a.d. 900 a new variety of corn adapted to North America's short summers appeared, and the arrival of beans around a.d. 1 100 completed Mexico's crop trinity of corn, beans, and squash. Eastern U.S. farming became greatly intensified, and densely populated chiefdoms developed along the Mississippi River and its tributaries. In some areas the original local domesticates were retained alongside the far more productive Mexican trinity, but in other areas the trinity replaced them completely. No European ever saw sumpweed growing in Indian gardens, because it had disappeared as a crop by the time that European colonization of the Americas began, in a.d. 1492. Among all those ancient eastern U.S. crop specialties, only two (sunflower and eastern squash) have been able to compete with crops domesticated elsewhere and are still grown today. Our modern acorn squashes and summer squashes are derived from those American squashes domesticated thousands of years ago.
Thus, like the case of New Guinea, that of the eastern United States is instructive. A priori, the region might have seemed a likely one to support productive indigenous agriculture. It has rich soils, reliable moderate rain-rail, and a suitable climate that sustains bountiful agriculture today. The lora is a species-rich one that includes productive wild nut trees (oak and

I 5 2. • •

GUNS,GERMS, AND STEELhickory). Local Native Americans did develop an agriculture based on local domesticates, did thereby support themselves in villages, and even developed a cultural florescence (the Hopewell culture centered on what is today Ohio) around 200 b.c.-a.d. 400. They were thus in a position for several thousand years to exploit as potential crops the most useful available wild plants, whatever those should be.
Nevertheless, the Hopewell florescence sprang up nearly 9,000 years after the rise of village living in the Fertile Crescent. Still, it was not until after a.d. 900 that the assembly of the Mexican crop trinity triggered a larger population boom, the so-called Mississippian florescence, which produced the largest towns and most complex societies achieved by Native Americans north of Mexico. But that boom came much too late to prepare Native Americans of the United States for the impending disaster of European colonization. Food production based on eastern U.S. crops alone had been insufficient to trigger the boom, for reasons that are easy to specify. ; The area's available wild cereals were not nearly as useful as wheat and barley. Native Americans of the eastern United States domesticated no. locally available wild pulse, no fiber crop, no fruit or nut tree. They had* no domesticated animals at all except for dogs, which were probably;: domesticated elsewhere in the Americas.
It's also clear that Native Americans of the eastern United States were not overlooking potential major crops among the wild species around' them. Even 20th-century plant breeders, armed with all the power of modern science, have had little success in exploiting North American wild? plants. Yes, we have now domesticated pecans as a nut tree and blueberries; as a fruit, and we have improved some Eurasian fruit crops (apples, plums, grapes, raspberries, blackberries, strawberries) by hybridizing them with North American wild relatives. However, those few successes havt; changed our food habits far less than Mexican corn changed food habits of Native Americans in the eastern United States after a.d. 900.
The farmers most knowledgeable about eastern U.S. domesticates, the region's Native Americans themselves, passed judgment on them by .d carding or deemphasizing them when the Mexican trinity arrived. That outcome also demonstrates that Native Americans were not constrained | by cultural conservativism and were quite able to appreciate a good plant when they saw it. Thus, as in New Guinea, the limitations on indigenous:; food production in the eastern United States were not due to Native Amer* j

APPLES

ORindians • 153. peoples themselves, but instead depended entirely on the American biota and environment.
have now considered examples of three contrasting areas, in all of which food production did arise indigenously. The Fertile Crescent lies at one extreme; New Guinea and the eastern United States lie at the opposite extreme. Peoples of the Fertile Crescent domesticated local plants much earlier. They domesticated far more species, domesticated far more productive or valuable species, domesticated a much wider range of types of crops, developed intensified food production and dense human populations more rapidly, and as a result entered the modern world with more advanced technology, more complex political organization, and more epidemic diseases with which to infect other peoples.
We found that these differences between the Fertile Crescent, New Guinea, and the eastern United States followed straightforwardly from the differing suites of wild plant and animal species available for domestication, not from limitations of the peoples themselves. When more-productive crops arrived from elsewhere (the sweet potato in New Guinea, the Mexican trinity in the eastern United States), local peoples promptly took advantage of them, intensified food production, and increased greatly in population. By extension, I suggest that areas of the globe where food production never developed indigenously at all—California, Australia, the Argentine pampas, western Europe, and so on—may have offered even less in the way of wild plants and animals suitable for domestication than did New Guinea and the eastern United States, where at least a limited food production did arise. Indeed, Mark Blunder's worldwide survey of locally available large-seeded wild grasses mentioned in this chapter, and the worldwide survey of locally available big mammals to be presented in the next chapter, agree in showing that all those areas of nonexistent or limited indigenous food production were deficient in wild ancestors of domesticable livestock and cereals.
Recall that the rise of food production involved a competition between food production and hunting-gathering. One might therefore wonder whether all these cases of slow or nonexistent rise of food production might instead have been due to an exceptional local richness of resources to be hunted and gathered, rather than to an exceptional availability of

154

* GUNS, GERMS, AND STEELspecies suitable for domestication. In fact, most areas where indigenous food production arose late or not at all offered exceptionally poor rather than rich resources to hunter-gatherers, because most large mammals of Australia and the Americas (but not of Eurasia and Africa) had become extinct toward the end of the Ice Ages. Food production would have faced even less competition from hunting-gathering in these areas than it did in the Fertile Crescent. Hence these local failures or limitations of food production cannot be attributed to competition from bountiful hunting opportunities.
lest these conclusions be misinterpreted, we should end this chapter with caveats against exaggerating two points: peoples' readiness to accept better crops and livestock, and the constraints imposed by locally available wild plants and animals. Neither that readiness nor those constraints are absolute.
We've already discussed many examples of local peoples' adopting more-productive crops domesticated elsewhere. Our broad conclusion is that people can recognize useful plants, would therefore have probably recognized better local ones suitable for domestication if any had existed, and aren't barred from doing so by cultural conservatism or taboos. But a big qualifier must be added to this sentence: "in the long run and over large areas." Anyone knowledgeable about human societies can cite innumerable examples of societies that refused crops, livestock, and other innovations that would have been productive.
Naturally, I don't subscribe to the obvious fallacy that every society promptly adopts every innovation that would be useful for it. The fact is that, over entire continents and other large areas containing hundreds of competing societies, some societies will be more open to innovation, and some will be more resistant. The ones that do adopt new crops, livestock, or technology may thereby be enabled to nourish themselves better and to outbreed, displace, conquer, or kill off societies resisting innovation. That's an important phenomenon whose manifestations extend far beyond the adoption of new crops, and to which we shall return in Chapter 13.
Our other caveat concerns the limits that locally available wild species set on the rise of food production. I'm not saying that food production could never, in any amount of time, have arisen in all those areas where it actually had not arisen indigenously by modern times. Europeans today

APPLES

OR INDIANS • 155who note that Aboriginal Australians entered the modern world as Stone Age hunter-gatherers often assume that the Aborigines would have gone on that way forever.
To appreciate the fallacy, consider a visitor from Outer Space who dropped in on Earth in the year 3000 b.c. The spaceling would have observed no food production in the eastern United States, because food production did not begin there until around 2500 b.c. Had the visitor of 3000 b.c. drawn the conclusion that limitations posed by the wild plants and animals of the eastern United States foreclosed food production there forever, events of the subsequent millennium would have proved the visitor wrong. Even a visitor to the Fertile Crescent in 9500 b.c. rather than in 8500 b.c. could have been misled into supposing the Fertile Crescent permanently unsuitable for food production.
That is, my thesis is not that California, Australia, western Europe, and all the other areas without indigenous food production were devoid of domesticable species and would have continued to be occupied just by hunter-gatherers indefinitely if foreign domesticates or peoples had not arrived. Instead, I note that regions differed greatly in their available pool of domesticable species, that they varied correspondingly in the date when local food production arose, and that food production had not yet arisen independently in some fertile regions as of modern times.
Australia, supposedly the most "backward" continent, illustrates this point well. In southeastern Australia, the well-watered part of the continent most suitable for food production, Aboriginal societies in recent millennia appear to have been evolving on a trajectory that would eventually have led to indigenous food production. They had already built winter villages. They had begun to manage their environment intensively for fish production by building fish traps, nets, and even long canals. Had Europeans not colonized Australia in 1788 and aborted that independent trajectory, Aboriginal Australians might within a few thousand years have become food producers, tending ponds of domesticated fish and growing domesticated Australian yams and small-seeded grasses.
In that light, we can now answer the question implicit in the title of this chapter. I asked whether the reason for the failure of North American Indians to domesticate North American apples lay with the Indians or with the apples.
I'm not thereby implying that apples could never have been domesticated m North America. Recall that apples were historically among the

156

• GUNS, GERMS, AND STEELmost difficult fruit trees to cultivate and among the last major ones to be domesticated in Eurasia, because their propagation requires the difficult technique of grafting. There is no evidence for large-scale cultivation of apples even in the Fertile Crescent and in Europe until classical Greek times, 8,000 years after the rise of Eurasian food production began. If Native Americans had proceeded at the same rate in inventing or acquiring grafting techniques, they too would eventually have domesticated apples—• around the year a.d. 5500, some 8,000 years after the rise of domestication in North America around 2500 b.c.
Thus, the reason for the failure of Native Americans to domesticate North American apples by the time Europeans arrived lay neither with the people nor with the apples. As far as biological prerequisites for apple domestication were concerned, North American Indian farmers were like Eurasian farmers, and North American wild apples were like Eurasian wild apples. Indeed, some of the supermarket apple varieties now being munched by readers of this chapter have been developed recently by crossing Eurasian apples with wild North American apples. Instead, the reason Native Americans did not domesticate apples lay with the entire suite of wild plant and animal species available to Native Americans. That suite's modest potential for domestication was responsible for the late start of food production in North America.

CHAPTER

9zebras, unhappy
marriages, and the
annakareninaprinciple

DOMESTICABLE ANIMALS ARE ALL ALIKE; EVERY UNDO-mesticable animal is undomesticable in its own way. If you think you've already read something like that before, you're right. Just make a few changes, and you have the famous first sentence of Tolstoy's great novel Anna Karenina:

"Happy families are all alike; every unhappy family is unhappy in its own way." By that sentence, Tolstoy meant that, in order to be happy, a marriage must succeed in many different respects: sexual attraction, agreement about money, child discipline, religion, in-laws, and other vital issues. Failure in any one of those essential respects can doom a marriage even if it has all the other ingredients needed for happiness.This principle can be extended to understanding much else about life besides marriage. We tend to seek easy, single-factor explanations of success. For most important things, though, success actually requires avoiding many separate possible causes of failure. The Anna Karenina principle explains a feature of animal domestication that had heavy consequences tor human history—namely, that so many seemingly suitable big wild mammal species, such as zebras and peccaries, have never been domesticated and that the successful domesticates were almost exclusively Eurasian. Having in the preceding two chapters discussed why so many wild

158 • GUNS, GERMS, AND STEEL

plant species seemingly suitable for domestication were never domesticated, we shall now tackle the corresponding question for domestic mammals. Our former question about apples or Indians becomes a question of zebras or Africans.
inchapter 4 we reminded ourselves of the many ways in which big domestic mammals were crucial to those human societies possessing them. Most notably, they provided meat, milk products, fertilizer, land transport, leather, military assault vehicles, plow traction, and wool, as well as germs that killed previously unexposed peoples.
In addition, of course, small domestic mammals and domestic birds and insects have also been useful to humans. Many birds were domesticated for meat, eggs, and feathers: the chicken in China, various duck and goose species in parts of Eurasia, turkeys in Mesoamerica, guinea fowl in Africa, and the Muscovy duck in South America. Wolves were domesticated in Eurasia and North America to become our dogs used as hunting companions, sentinels, pets, and, in some societies, food. Rodents and other small mammals domesticated for food included the rabbit in Europe, the guinea pig in the Andes, a giant rat in West Africa, and possibly a rodent called the hutia on Caribbean islands. Ferrets were domesticated in Europe to hunt rabbits, and cats were domesticated in North Africa and Southwest Asia to hunt rodent pests. Small mammals domesticated as recently as the 19th and 20th centuries include foxes, mink, and chinchillas grown for fur and hamsters kept as pets. Even some insects have been domesticated, notably Eurasia's honeybee and China's silkworm moth, kept for honey and silk, respectively.
Many of these small animals thus yielded food, clothing, or warmth. But none of them pulled plows or wagons, none bore riders, none except dogs pulled sleds or became war machines, and none of them have been as important for food as have big domestic mammals. Hence the rest of this chapter will confine itself to the big mammals.
1 he importance of domesticated mammals rests on surprisingly few species of big terrestrial herbivores. (Only terrestrial mammals have been domesticated, for the obvious reason that aquatic mammals were difficult to maintain and breed until the development of modern Sea World facili-

ZEBRAS

AND UNHAPPY MARRIAGES • 159• s ) If one defines "big" as "weighing over 100 pounds," then only 14 ch species were domesticated before the twentieth century (see Table 9.1 f r a list). Of those Ancient Fourteen, 9 (the "Minor Nine" of Table 9.1) became important livestock for people in only limited areas of the globe: the Arabian camel, Bactrian camel, llama / alpaca (distinct breeds of the same ancestral species), donkey, reindeer, water buffalo, yak, banteng, and gaur. Only 5 species became widespread and important around the world. Those Major Five of mammal domestication are the cow, sheep, goat, pig,
and horse.
This list may at first seem to have glaring omissions. What about the African elephants with which Hannibal's armies crossed the Alps? What about the Asian elephants still used as work animals in Southeast Asia today? No, I didn't forget them, and that raises an important distinction. Elephants have been tamed, but never domesticated. Hannibal's elephants were, and Asian work elephants are, just wild elephants that were captured and tamed; they were not bred in captivity. In contrast, a domesticated animal is defined as an animal selectively bred in captivity and thereby modified from its wild ancestors, for use by humans who control the animal's breeding and food supply.
That is, domestication involves wild animals' being transformed into something more useful to humans. Truly domesticated animals differ in various ways from their wild ancestors. These differences result from two processes: human selection of those individual animals more useful to humans than other individuals of the same species, and automatic evolutionary responses of animals to the altered forces of natural selection operating in human environments as compared with wild environments. We already saw in Chapter 7 that all of these statements also apply to plant domestication.
The ways in which domesticated animals have diverged from their wild ancestors include the following. Many species changed in size: cows, pigs, and sheep became smaller under domestication, while guinea pigs became larger. Sheep and alpacas were selected for retention of wool and reduction or loss of hair, while cows have been selected for high milk yields. Several species of domestic animals have smaller brains and less developed sense organs than their wild ancestors, because they no longer need the igger brains and more developed sense organs on which their ancestors depended to escape from wild predators.
o appreciate the changes that developed under domestication, just

I 6 O •

GUNS, GERMS,and steeltable 9.1 The Ancient Fourteen Species of Big Herbivorous Domestic Mammals
The Major Five
1. Sheep. Wild ancestor: the Asiatic mouflon sheep of West and Central Asia. Now worldwide.
2. Goat. Wild ancestor: the bezoar goat of West Asia. Now worldwide.
3. Cow, alias ox or cattle. Wild ancestor: the now extinct aurochs, formerly distributed over Eurasia and North Africa. Now worldwide.
4. Pig. Wild ancestor: the wild boar, distributed over Eurasia and North Africa. Now worldwide. Actually an omnivore (regularly eats both animal and plant food), whereas the other 13 of the Ancient Fourteen are more strictly herbivores.
5. Horse. Wild ancestor: now extinct wild horses of southern Russia; a different subspecies of the same species survived in the wild to modern times as Przewalski's horse of Mongolia. Now worldwide.
The Minor Nine
6. Arabian (one-humped) camel. Wild ancestor: now extinct, formerly lived in Arabia and adjacent areas. Still largely restricted to Arabia and northern Africa, though feral in Australia.
7. Bactrian (two-humped) camel: Wild ancestor: now extinct, lived in Central Asia. Still largely confined to Central Asia.
8. Llama and alpaca. These appear to be well-differentiated breeds of the same species, rather than different species. Wild ancestor: the guanaco of the Andes. Still largely confined to the Andes, although some are bred as pack animals in North America.
9. Donkey. Wild ancestor: the African wild ass of North Africa and formerly perhaps the adjacent area of Southwest Asia. Originally confined as a domestic animal to North Africa and western Eurasia, more recently also used elsewhere.
10. Reindeer. Wild ancestor: the reindeer of northern Eurasia. Still largely confined as a domestic animal to that area, though now some are also used in Alaska.
11. Water buffalo. Wild ancestor lives in Southeast Asia. Still used as a domestic animal mainly in that area, though many are also used in Brazil and others have escaped to the wild in Australia and other places.

ZEBRAS

AND UNHAPPY MARRIAGES • I 6 I12 Yak. Wild ancestor: the wild yak of the Himalayas and Tibetan plateau Still confined as a domestic animal to that area.
13 Bali cattle. Wild ancestor: the banteng (a relative of the aurochs) of Southeast Asia. Still confined as a domestic animal to that area.
14 Mithan. Wild ancestor: the gaur (another relative of the aurochs) of Indian and Burma. Still confined as a domestic animal to that area.
compare wolves, the wild ancestors of domestic dogs, with the many breeds of dogs. Some dogs are much bigger than wolves (Great Danes), while others are much smaller (Pekingese). Some are slimmer and built for racing (greyhounds), while others are short-legged and useless for racing (dachshunds). They vary enormously in hair form and color, and some are even hairless. Polynesians and Aztecs developed dog breeds specifically raised for food. Comparing a dachshund with a wolf, you wouldn't even suspect that the former had been derived from the latter if you didn't already know it.
The wild ancestors of the Ancient Fourteen were spread unevenly over the globe. South America had only one such ancestor, which gave rise to the llama and alpaca. North America, Australia, and sub-Saharan Africa had none at all. The lack of domestic mammals indigenous to sub-Saharan Africa is especially astonishing, since a main reason why tourists visit Africa today is to see its abundant and diverse wild mammals. In contrast, the wild ancestors of 13 of the Ancient Fourteen (including all of the Major Five) were confined to Eurasia. (As elsewhere in this book, my use of the term "Eurasia" includes in several cases North Africa, which biogeographically and in many aspects of human culture is more closely related to Eurasia than to sub-Saharan Africa.)
Of course, not all 13 of these wild ancestral species occurred together throughout Eurasia. No area had all 13, and some of the wild ancestors were quite local, such as the yak, confined in the wild to Tibet and adjacent highland areas. However, many parts of Eurasia did have quite a few of these 13 species living together in the same area: for example, seven of the wild ancestors occurred in Southwest Asia.
A very unequal distribution of wild ancestral species among the con-

I 6 Z • GUNS, GERMS, AND STEEL

tinents became an important reason why Eurasians, rather than peoples of other continents, were the ones to end up with guns, germs, and steel. How can we explain the concentration of the Ancient Fourteen in Eurasia? One reason is simple. Eurasia has the largest number of big terrestrial wild mammal species, whether or not ancestral to a domesticated species. Let's define a "candidate for domestication" as any terrestrial herbivorous or omnivorous mammal species (one not predominantly a carnivore) weighing on the average over 100 pounds (45 kilograms). Table 9.2 shows that Eurasia has the most candidates, 72 species, just as it has the most species in many other plant and animal groups. That's because Eurasia is the world's largest landmass, and it's also very diverse ecologically, with habitats ranging from extensive tropical rain forests, through temperate forests, deserts, and marshes, to equally extensive tundras. Sub-Saharan Africa has fewer candidates, 51 species, just as it has fewer species in most other plant and animal groups—because it's smaller and ecologically less diverse than Eurasia. Africa has smaller areas of tropical rain forest than does Southeast Asia, and no temperate habitats at all beyond latitude 37 degrees. As I discussed in Chapter 1, the Americas may formerly have had almost as many candidates as Africa, but most of America's big wild mammals (including its horses, most of its camels, and other species likely to have been domesticated had they survived) became extinct about 13,000 years ago. Australia, the smallest and most isolated continent, has always had far fewer species of big wild mammals than has Eurasia, Africa, or the Americas. Just as in the Americas, in Australia all of those few candidates
table 9.1 Mammalian Candidates for Domestication
ContinentEurasia Sub-Saharan The Australia
Africa Americas
Candidates 72 51 24 1
Domesticated species 13 0 1 0
Percentage of candidates
domesticated 18% 0% 4% 0%
>>.—————
A "candidate" is defined as a species of terrestrial, herbivorous or omnivorous, *•*> mammal weighing on the average over 100 pounds.

ZEBRAS AND UNHAPPY MARRIAGES • 163

t the red kangaroo became extinct around the time of the continent's first colonization by humans.
Thus part of the explanation for Eurasia's having been the main site of hie mammal domestication is that it was the continent with the most candidate species of wild mammals to start out with, and lost the fewest candidates to extinction in the last 40,000 years. But the numbers in Table 9.2 warn us that that's not the whole explanation. It's also true that the percentage of candidates actually domesticated is highest in Eurasia (18 percent) and is especially low in sub-Saharan Africa (no species domesticated out of 51 candidates!). Particularly surprising is the large number of species of African and American mammals that were never domesticated, despite their having Eurasian close relatives or counterparts that were domesticated. Why were Eurasia's horses domesticated, but not Africa's zebras? Why Eurasia's pigs, but not American peccaries or Africa's three species of true wild pigs? Why Eurasia's five species of wild cattle (aurochs, water buffalo, yak, gaur, banteng), but not the African buffalo or American bison? Why the Asian mouflon sheep (ancestor of our domestic sheep), but not North American bighorn sheep?did all those peoples of Africa, the Americas, and Australia, despite their enormous diversity, nonetheless share some cultural obstacles to domestication not shared with Eurasian peoples? For example, did Africa's abundance of big wild mammals, available to kill by hunting, make it superfluous for Africans to go to the trouble of tending domestic stock?
The answer to that question is unequivocal: No! The interpretation is refuted by five types of evidence: rapid acceptance of Eurasian domesticates by non-Eurasian peoples, the universal human penchant for keeping pets, the rapid domestication of the Ancient Fourteen, the repeated independent domestications of some of them, and the limited successes of modern efforts at further domestications.
First, when Eurasia's Major Five domestic mammals reached sub-saharan Africa, they were adopted by the most diverse African peoples wherever conditions permitted. Those African herders thereby achieved a "ge advantage over African hunter-gatherers and quickly displaced them. particular, Bantu farmers who acquired cows and sheep spread out of their homeland in West Africa and within a short time overran the former hunter-gatherers in most of the rest of sub-Saharan Africa. Even without

164

• GUNS, GERMS, AND STEELacquiring crops, Khoisan peoples who acquired cows and sheep around 2,000 years ago displaced Khoisan hunter-gatherers over much of southern Africa. The arrival of the domestic horse in West Africa transformed warfare there and turned the area into a set of kingdoms dependent on cavalry. The only factor that prevented horses from spreading beyond West Africa was trypanosome diseases borne by tsetse flies.
The same pattern repeated itself elsewhere in the world, whenever peoples lacking native wild mammal species suitable for domestication finally had the opportunity to acquire Eurasian domestic animals. European! horses were eagerly adopted by Native Americans in both North and South America, within a generation of the escape of horses from European settlements. For example, by the 19th century North America's Great Plains Indians were famous as expert horse-mounted warriors and bison hunters, but they did not even obtain horses until the late 17th century. Sheep acquired from Spaniards similarly transformed Navajo Indian society and led to, among other things, the weaving of the beautiful woolen blankets for which the Navajo have become renowned. Within a decade of Tasmania's settlement by Europeans with dogs, Aboriginal Tasmanians who had never before seen dogs, began to breed them in large numbers for use in hunting. Thus, among the thousands of culturally diverse native peoples of Australia, the Americas, and Africa, no universal cultural tat stood in the way of animal domestication.
Surely, if some local wild mammal species of those continents had domesticable, some Australian, American, and African peoples won have domesticated them and gained great advantage from them, just they benefited from the Eurasian domestic animals that they immediatel; adopted when those became available. For instance, consider all the pies of sub-Saharan Africa living within the range of wild zebras and 1 falo. Why wasn't there at least one African hunter-gatherer tribe domesticated those zebras and buffalo and that thereby gained sway other Africans, without having to await the arrival of Eurasian horses cattle? All these facts indicate that the explanation for the lack of nath mammal domestication outside Eurasia lay with the locally available mammals themselves, not with the local peoples.
A second type of evidence for the same interpretation comes pets. Keeping wild animals as pets, and taming them, constitute an initiq

ZEBRAS

AND UNHAPPY MARRIAGES • 165in domestication. But pets have been reported from virtually all traditional human societies on all continents. The variety of wild animals thus tamed is far greater than the variety eventually domesticated, and includes some species that we would scarcely have imagined as pets.
For example, in the New Guinea villages where I work, I often see people with pet kangaroos, possums, and birds ranging from flycatchers to ospreys. Most of these captives are eventually eaten, though some are kept just as pets. New Guineans even regularly capture chicks of wild cassowaries (an ostrich-like large, flightless bird) and raise them to eat as a delicacy even though captive adult cassowaries are extremely dangerous and now and then disembowel village people. Some Asian peoples tame eagles for use in hunting, although those powerful pets have also been known on occasion to kill their human handlers. Ancient Egyptians and Assyrians, and modern Indians, tamed cheetahs for use in hunting. Paintings made by ancient Egyptians show that they further tamed (not surprisingly) hoofed mammals such as gazelles and hartebeests, birds such as cranes, more surprisingly giraffes (which can be dangerous), and most astonishingly hyenas. African elephants were tamed in Roman times despite the obvious danger, and Asian elephants are still being tamed today. Perhaps the most unlikely pet is the European brown bear (the same species as the American grizzly bear), which the Ainu people of Japan regularly captured as young animals, tamed, and reared to kill and eat in a ritual ceremony.
Thus, many wild animal species reached the first stage in the sequence of animal-human relations leading to domestication, but only a few emerged at the other end of that sequence as domestic animals. Over a century ago, the British scientist Francis Galton summarized this discrepancy succinctly: "It would appear that every wild animal has had its chance of being domesticated, that note 8 few … were domesticated long ago, but that the large remainder, who failed sometimes in only one small particular, are destined to perpetual wildness."
Datesof domestication provide a third line of evidence confirming galton's view that early herding peoples quickly domesticated all big mammal species suitable for being domesticated. All species for whose dates of domestication we have archaeological evidence were domesticated between about 8000 and 2500 b.c.—that is, within the first few thousand years of the sedentary farming-herding societies that arose after the end

I 6 6 •

GUNS,GERMS, AND STEELof the last Ice Age. As summarized in Table 9.3, the era of big mammal domestication began with the sheep, goat, and pig and ended with camels. Since 2500 b.c. there have been no significant additions.
It's true, of course, that some small mammals were first domesticated long after 2500 b.c. For example, rabbits were not domesticated for food until the Middle Ages, mice and rats for laboratory research not until the 20th century, and hamsters for pets not until the 1930s. The continuing development of domesticated small mammals isn't surprising, because there are literally thousands of wild species as candidates, and because they were of too little value to traditional societies to warrant the effort of raising them. But big mammal domestication virtually ended 4,500 years ago. By then, all of the world's 148 candidate big species must have been tested innumerable times, with the result that only a few passed the test and no other suitable ones remained.
Still a fourth line of evidence that some mammal species are much-more suitable than others is provided by the repeated independent domestications of the same species. Genetic evidence based on the portions of our genetic material known as mitochondrial DNA recently confirmed, as had long been suspected, that humped cattle of India and humpless European cattle were derived from two separate populations of wild ancestral;! cattle that had diverged hundreds of thousands of years ago. That is, Indian peoples domesticated the local Indian subspecies of wild aurochs, Southwest Asians independently domesticated their own Southwest Asiaiy subspecies of aurochs, and North Africans may have independently domesticated the North African aurochs.
Similarly, wolves were independently domesticated to become dogs ii the Americas and probably in several different parts of Eurasia, includii China and Southwest Asia. Modern pigs are derived from independent! sequences of domestication in China, western Eurasia, and possibly other| areas as well. These examples reemphasize that the same few suitable wild ] species attracted the attention of many different human societies.
ihefailures of modern efforts provide a final type of evidence 1 past failures to domesticate the large residue of wild candidate s[ arose from shortcomings of those species, rather than from shortcot

ZEBRAS

ANDunhappy MARRIAGES • 167table9.3 Approximate Dates of First Attested Evidence for Domestication of Large Mammal Species

Species

Date (b.c.)

Place

Dog

10,000

Southwest Asia, China,

North America

Sheep

8,000

Southwest Asia

Goat

8,000

Southwest Asia

Pig

8,000

China, Southwest Asia

Cow

6,000

Southwest Asia, India,

(?)North Africa

Horse

4,000

Ukraine

Donkey

4,000

Egypt

Water buffalo

4,000

China?

Llama / alpaca

3,500

Andes

Bactrian camel

2,500

Central Asia

Arabian camel

2,500

Arabia

For the other four domesticated large mammal species—reindeer, yak, gaur, and ban-teng—there is as yet little evidence concerning the date of domestication. Dates and places shown are merely the earliest ones attested to date; domestication may actually have begun earlier and at a different location.

of ancient humans. Europeans today are heirs to one of the longest traditions of animal domestication on Earth—that which began in Southwest Asia around 10,000 years ago. Since the fifteenth century, Europeans have spread around the globe and encountered wild mammal species not found in Europe. European settlers, such as those that I encounter in New Guinea with pet kangaroos and possums, have tamed or made pets of many local mammals, just as have indigenous peoples. European herders and farmers emigrating to other continents have also made serious efforts to domesticate some local species.
In the 19th and 20th centuries at least six large mammals—the eland, moose, musk ox, zebra, and American bison—have been the subjects especially well-organized projects aimed at domestication, carried out modern scientific animal breeders and geneticists. For example, eland, argest African antelope, have been undergoing selection for meat quality and milk quantity in the Askaniya-Nova Zoological Park in the

I 6 8 •

GUNS,GERMS, AND STEELUkraine, as well as in England, Kenya, Zimbabwe, and South Africa; an experimental farm for elk (red deer, in British terminology) has been operated by the Rowett Research Institute at Aberdeen, Scotland; and an experimental farm for moose has operated in the Pechero-Ilych National Park in Russia. Yet these modern efforts have achieved only very limited successes. While bison meat occasionally appears in some U.S. supermarkets, and while moose have been ridden, milked, and used to pull sleds in Sweden and Russia, none of these efforts has yielded a result of sufficient economic value to attract many ranchers. It is especially striking that recent attempts to domesticate eland within Africa itself, where its disease resistance and climate tolerance would give it a big advantage over intro-1 duced Eurasian wild stock susceptible to African diseases, have not caught;; on.
Thus, neither indigenous herders with access to candidate species ove thousands of years, nor modern geneticists, have succeeded in making \ ful domesticates of large mammals beyond the Ancient Fourteen, were domesticated by at least 4,500 years ago. Yet scientists today couJdB undoubtedly, if they wished, fulfill for many species that part of the defini*! tion of domestication that specifies the control of breeding and food ply. For example, the San Diego and Los Angeles zoos are now subjecting the last surviving California condors to a more draconian control of ing than that imposed upon any domesticated species. All individual cc dors have been genetically identified, and a computer program determir which male shall mate with which female in order to achieve human go (in this case, to maximize genetic diversity and thereby preserve endangered bird). Zoos are conducting similar breeding programs many other threatened species, including gorillas and rhinos. But the ; rigorous selection of California condors shows no prospects of yielding|0 economically useful product. Nor do zoos' efforts with rhinos, althou rhinos offer up to over three tons of meat on the hoof. As we shall see, rhinos (and most other big mammals) present insuperable obstacles I domestication.

in

all, of the world's 148 big wild terrestrial herbivorous mammals the candidates for domestication—only 14 passed the test. Why did other 134 species fail? To which conditions was Francis Gallon re when he spoke of those other species as "destined to perpetual wildnessl

ZEBRAS

AND UNHAPPY MARRIAGES • 169The answer follows from the Anna Karenina principle. To be domesti-ated a candidate wild species must possess many different characteristics. Lack of any single required characteristic dooms efforts at domestication, just as it dooms efforts at building a happy marriage. Playing marriage counselor to the zebra / human couple and other ill-sorted pairs, we can recognize at least six groups of reasons for failed domestication.
Diet. Every time that an animal eats a plant or another animal, the conversion of food biomass into the consumer's biomass involves an efficiency of much less than 100 percent: typically around 10 percent. That is it takes around 10,000 pounds of corn to grow a 1,000-pound cow. If instead you want to grow 1,000 pounds of carnivore, you have to feed it 10,000 pounds of herbivore grown on 100,000 pounds of corn. Even among herbivores and omnivores, many species, like koalas, are too finicky in their plant preferences to recommend themselves as farm animals.
As a result of this fundamental inefficiency, no mammalian carnivore has ever been domesticated for food. (No, it's not because its meat would be tough or tasteless: we eat carnivorous wild fish all the time, and I can personally attest to the delicious flavor of lion burger.) The nearest thing to an exception is the dog, originally domesticated as a sentinel and hunting companion, but breeds of dogs were developed and raised for food in Aztec Mexico, Polynesia, and ancient China. However, regular dog eating has been a last resort of meat-deprived human societies: the Aztecs had no other domestic mammal, and the Polynesians and ancient Chinese had only pigs and dogs. Human societies blessed with domestic herbivorous mammals have not bothered to eat dogs, except as an uncommon delicacy (as in parts of Southeast Asia today). In addition, dogs are not strict carnivores but omnivores: if you are so naive as to think that your beloved pet dog is really a meat eater, just read the list of ingredients on your bag of dog food. The dogs that the Aztecs and Polynesians reared for food were efficiently fattened on vegetables and garbage.
Growth Rate. To be worth keeping, domesticates must also grow quickly. That eliminates gorillas and elephants, even though they are vegetarians with admirably nonfinicky food preferences and represent a lot of meat. What would-be gorilla or elephant rancher would wait 15 years for his herd to reach adult size? Modern Asians who want work elephants find it much cheaper to capture them in the wild and tame them.
Problems of Captive Breeding. We humans don't like to have sex under watchful eyes of others; some potentially valuable animal species don't

17o

" GUNS, GERMS, AND STEELlike to, either. That's what derailed attempts to domesticate cheetahs, the swiftest of all land animals, despite our strong motivation to do so for thousands of years.
As I already mentioned, tame cheetahs were prized by ancient Egyptians and Assyrians and modern Indians as hunting animals infinitely superior to dogs. One Mogul emperor of India kept a stable of a thousand cheetahs. But despite those large investments that many wealthy princes made, all of their cheetahs were tamed ones caught in the wild. The princes' efforts to breed cheetahs in captivity failed, and not until 1960 did even biologists in modern zoos achieve their first successful cheetah birth. In the wild, several cheetah brothers chase a female for several days, and that rough courtship over large distances seems to be required to get the female to ovulate or to become sexually receptive. Cheetahs usually refuse to carry-out that elaborate courtship ritual inside a cage.
A similar problem has frustrated schemes to breed the vicufta, an Andean wild camel whose wool is prized as the finest and lightest of any animal's. The ancient Incas obtained vicuna wool by driving wild vicunas into corrals, shearing them, and then releasing them alive. Modern mer– | chants wanting this luxury wool have had to resort either to this same method or simply to killing wild vicunas. Despite strong incentives of|f money and prestige, all attempts to breed vicunas for wool production in captivity have failed, for reasons that include vicunas' long and elaborate; courtship ritual before mating, a ritual inhibited in captivity; male vicunas' ' fierce intolerance of each other; and their requirement for both a year-1! round feeding territory and a separate year-round sleeping territory.
Nasty Disposition. Naturally, almost any mammal species that is suffi-% ciently large is capable of killing a human. People have been killed by pigs, horses, camels, and cattle. Nevertheless, some large animals have mudr| nastier dispositions and are more incurably dangerous than are others. Tendencies to kill humans have disqualified many otherwise seemingrjS ideal candidates for domestication.
One obvious example is the grizzly bear. Bear meat is an expensive delicacy, grizzlies weigh up to 1,700 pounds, they are mainly vegetarian (though also formidable hunters), their vegetable diet is very broad, thrive on human garbage (thereby creating big problems in Yellowston and Glacier National Parks), and they grow relatively fast. If they would behave themselves in captivity, grizzlies would be a fabulous meat production animal. The Ainu people of Japan made the experiment by routinely

ZEBRAS

AND UNHAPPY MARRIAGES • I 7 Iring grizzly cu^s as Part of a ritual– For understandable reasons, , ugh the Ainu found it prudent to kill and eat the cubs at the age of one Keeping grizzly bears for longer would be suicidal; I am not aware of any adult that has been tamed.
Another otherwise suitable candidate that disqualifies itself for equally obvious reasons is the African buffalo. It grows quickly up to a weight of a ton and lives in herds that have a well-developed dominance hierarchy, a trait whose virtues will be discussed below. But the African buffalo is considered the most dangerous and unpredictable large mammal of Africa. Anyone insane enough to try to domesticate it either died in the effort or was forced to kill the buffalo before it got too big and nasty. Similarly, hippos, as four-ton vegetarians, would be great barnyard animals if they weren't so dangerous. They kill more people each year than do any other African mammals, including even lions.
Few people would be surprised at the disqualification of those notoriously ferocious candidates. But there are other candidates whose dangers are not so well known. For instance, the eight species of wild equids (horses and their relatives) vary greatly in disposition, even though all eight are genetically so close to each other that they will interbreed and produce healthy (though usually sterile) offspring. Two of them, the horse and the North African ass (ancestor of the donkey), were successfully domesticated. Closely related to the North African ass is the Asiatic ass, also known as the onager. Since its homeland includes the Fertile Crescent, the cradle of Western civilization and animal domestication, ancient peoples must have experimented extensively with onagers. We know from Sumerian and later depictions that onagers were regularly hunted, as well as captured and hybridized with donkeys and horses. Some ancient depictions of horselike animals used for riding or for pulling carts may refer to onagers. However, all writers about them, from Romans to modern zookeepers, decry their irascible temper and their nasty habit of biting People. As a result, although similar in other respects to ancestral donkeys, onagers have never been domesticated.
Africa's four species of zebras are even worse. Efforts at domestication went as far as hitching them to carts: they were tried out as draft animals m 19th-century South Africa, and the eccentric Lord Walter Rothschild drove through the streets of London in a carriage pulled by zebras. Alas, ze ras become impossibly dangerous as they grow older. (That's not to eny that many individual horses are also nasty, but zebras and onagers

I 7 Z •

GUNS,GERMS, AND STEELare much more uniformly so.) Zebras have the unpleasant habit of biting a person and not letting go. They thereby injure even more American zoo-keepers each year than do tigers! Zebras are also virtually impossible to lasso with a rope—even for cowboys who win rodeo championships by lassoing horses—because of their unfailing ability to watch the rope noose fly toward them and then to duck their head out of the way.
Hence it has rarely (if ever) been possible to saddle or ride a zebra, and; South Africans' enthusiasm for their domestication waned. UnpredictaWy aggressive behavior on the part of a large and potentially dangerous mammal is also part of the reason why the initially so promising modern experiments in domesticating elk and eland have not been more successful.
Tendency to Panic. Big mammalian herbivore species react to danger J from predators or humans in different ways. Some species are nerve fast, and programmed for instant flight when they perceive a threat.' species are slower, less nervous, seek protection in herds, stand theif| ground when threatened, and don't run until necessary. Most species deer and antelope (with the conspicuous exception of reindeer) are of 1 former type, while sheep and goats are of the latter.
Naturally, the nervous species are difficult to keep in captivity. If into an enclosure, they are likely to panic, and either die of shock or bat themselves to death against the fence in their attempts to escape, true, for example, of gazelles, which for thousands of years were the i frequently hunted game species in some parts of the Fertile Crescent.' is no mammal species that the first settled peoples of that area had opportunity to domesticate than gazelles. But no gazelle species has been domesticated. Just imagine trying to herd an animal that blindly bashes itself against walls, can leap up to nearly 30 feet, and i run at a speed of 50 miles per hour!
Social Structure. Almost all species of domesticated large prove to be ones whose wild ancestors share three social charac they live in herds; they maintain a well-developed dominance among herd members; and the herds occupy overlapping home ra rather than mutually exclusive territories. For example, herds of horses consist of one stallion, up to half a dozen mares, and their Mare A is dominant over mares B, C, D, and E; mare B is submissive tt»| but dominant over C, D, and E; C is submissive to B and A but dc over D and E; and so on. When the herd is on the move, its maintain a stereotyped order: in the rear, the stallion; in the front, the

ZEBRAS AND UNHAPPY MARRIAGES • I 7 3

nking female, followed by her foals in order of age, with the youngest first– and behind her, the other mares in order of rank, each followed by her foals in order of age. In that way, many adults can coexist in the herd without constant fighting and with each knowing its rank.
That social structure is ideal for domestication, because humans in effect take over the dominance hierarchy. Domestic horses of a pack line follow the human leader as they would normally follow the top-ranking female. Herds or packs of sheep, goats, cows, and ancestral dogs (wolves) have a similar hierarchy. As young animals grow up in such a herd, they imprint on the animals that they regularly see nearby. Under wild conditions those are members of their own species, but captive young herd animals also see humans nearby and imprint on humans as well.
Such social animals lend themselves to herding. Since they are tolerant of each other, they can be bunched up. Since they instinctively follow a dominant leader and will imprint on humans as that leader, they can readily be driven by a shepherd or sheepdog. Herd animals do well when penned in crowded conditions, because they are accustomed to living in densely packed groups in the wild.
In contrast, members of most solitary territorial animal species cannot be herded. They do not tolerate each other, they do not imprint on humans, and they are not instinctively submissive. Who ever saw a line of cats (solitary and territorial in the wild) following a human or allowing themselves to be herded by a human? Every cat lover knows that cats are not submissive to humans in the way dogs instinctively are. Cats and ferrets are the sole territorial mammal species that were domesticated, because our motive for doing so was not to herd them in large groups raised for food but to keep them as solitary hunters or pets.
While most solitary territorial species thus haven't been domesticated, it's not conversely the case that most herd species can be domesticated. Most can't, for one of several additional reasons.
First, herds of many species don't have overlapping home ranges but instead maintain exclusive territories against other herds. It's no more possible to pen two such herds together than to pen two males of a solitary species.
Second, many species that live in herds for part of the year are territorial m the breeding season, when they fight and do not tolerate each other's presence. That's true of most deer and antelope species (again with the exception of reindeer), and it's one of the main factors that has disqualified

174 ' GUNS, GERMS, AND STEEL

all the social antelope species for which Africa is famous from being domesticated. While one's first association to African antelope is "vast dense herds spreading across the horizon," in fact the males of those herds space themselves into territories and fight fiercely with each other when breeding. Hence those antelope cannot be maintained in crowded enclosures in captivity, as can sheep or goats or cattle. Territorial behavior similarly combines with a fierce disposition and a slow growth rate to banish rhinos from the farmyard.
Finally, many herd species, including again most deer and antelope, do not have a well-defined dominance hierarchy and are not instinctively prepared to become imprinted on a dominant leader (hence to become misim-printed on humans). As a result, though many deer and antelope species-have been tamed (think of all those true Bambi stories), one never sees such tame deer and antelope driven in herds like sheep. That problem also derailed domestication of North American bighorn sheep, which belong to the same genus as Asiatic mouflon sheep, ancestor of our domestic sheep. Bighorn sheep are suitable to us and similar to mouflons in most respects-] except a crucial one: they lack the mouflon's stereotypical behavior; whereby some individuals behave submissively toward other individuals whose dominance they acknowledge.
let's now return to the problem I posed at the outset of this chapt Initially, one of the most puzzling features of animal domestication is seeming arbitrariness with which some species have been domesticat while their close relatives have not. It turns out that all but a few candii| dates for domestication have been eliminated by the Anna Karenina princi4| pie. Humans and most animal species make an unhappy marriage, for onfrl or more of many possible reasons: the animal's diet, growth rate, matiftp| habits, disposition, tendency to panic, and several distinct features social organization. Only a small percentage of wild mammal sj ended up in happy marriages with humans, by virtue of compatibility' all those separate counts.
Eurasian peoples happened to inherit many more species of don ticable large wild mammalian herbivores than did peoples of the continents. That outcome, with all of its momentous advantages for asian societies, stemmed from three basic facts of mammalian geograf history, and biology. First, Eurasia, befitting its large area and ecolo-

ZEBRAS

AND UNHAPPY MARRIAGES • I 7 5gy started out with the most candidates. Second, Australia and the but not Eurasia or Africa, lost most of their candidates in a Americas) ^i"
sive wave of late-Pleistocene extinctions—possibly because the mam-Is of the former continents had the misfortune to be first exposed to h mans suddenly and late in our evolutionary history, when our hunting skills were already highly developed. Finally, a higher percentage of the rviving candidates proved suitable for domestication on Eurasia than on the other continents. An examination of the candidates that were never domesticated, such as Africa's big herd-forming mammals, reveals particular reasons that disqualified each of them. Thus, Tolstoy would have approved of the insight offered in another context by an earlier author, Saint Matthew: "Many are called, but few are chosen."

CHAPTER

10spacious skies and tilted axes

ON THE MAP OF THE WORLD ON PAGE 177 (FIGURE 10.) compare the shapes and orientations of the continents. You'll 1 struck by an obvious difference. The Americas span a much greater tance north-south (9,000 miles) than east-west: only 3,000 miles at widest, narrowing to a mere 40 miles at the Isthmus of Panama. That the major axis of the Americas is north-south. The same is also though to a less extreme degree, for Africa. In contrast, the major axis* Eurasia is east-west. What effect, if any, did those differences in the ork tation of the continents' axes have on human history?

This chapter will be about what I see as their enormous, son tragic, consequences. Axis orientations affected the rate of spread of and livestock, and possibly also of writing, wheels, and other inver That basic feature of geography thereby contributed heavily to the different experiences of Native Americans, Africans, and Eurasians in.) last 500 years.
food production's spread proves as crucial to understar geographic differences in the rise of guns, germs, and steel as did its orgins, which we considered in the preceding chapters. That's because, as

SPACIOUS

SKIES AND TILTED AXES • 177

Figure 10.1. Major axes of the continents.

saw in Chapter 5, there were no more than nine areas of the globe, perhaps as few as five, where food production arose independently. Yet, already in prehistoric times, food production became established in many other regions besides those few areas of origins. All those other areas became food producing as a result of the spread of crops, livestock, and knowledge of how to grow them and, in some cases, as a result of migrations of farmers and herders themselves.
The main such spreads of food production were from Southwest Asia to Europe, Egypt and North Africa, Ethiopia, Central Asia, and the Indus Valley; from the Sahel and West Africa to East and South Africa; from China to tropical Southeast Asia, the Philippines, Indonesia, Korea, and Japan; and from Mesoamerica to North America. Moreover, food production even in its areas of origin became enriched by the addition of crops, livestock, and techniques from other areas of origin.
Just as some regions proved much more suitable than others for the origins of food production, the ease of its spread also varied greatly around the world. Some areas that are ecologically very suitable for food production never acquired it in prehistoric times at all, even though areas of prehistoric food production existed nearby. The most conspicuous such examples are the failure of both farming and herding to reach Native

I 7 8 •

GUNS,GERMS, AND STEELAmerican California from the U.S. Southwest or to reach Australia from New Guinea and Indonesia, and the failure of farming to spread front South Africa's Natal Province to South Africa's Cape. Even among alt those areas where food production did spread in the prehistoric era, the rates and dates of spread varied considerably. At the one extreme was itt-l rapid spread along east-west axes: from Southwest Asia both west Europe and Egypt and east to the Indus Valley (at an average rate of aboqff 0.7 miles per year); and from the Philippines east to Polynesia (at 3.21 per year). At the opposite extreme was its slow spread along nor axes: at less than 0.5 miles per year, from Mexico northward to the U.1 Southwest; at less than 0.3 miles per year, for corn and beans from ] northward to become productive in the eastern United States around, 900; and at 0.2 miles per year, for the llama from Peru north to j These differences could be even greater if corn was not domesticated i Mexico as late as 3500 b.c., as I assumed conservatively for these tions, and as some archaeologists now assume, but if it was instead * ticated considerably earlier, as most archaeologists used to assume (a many still do).
There were also great differences in the completeness with which! of crops and livestock spread, again implying stronger or weaker to their spreading. For instance, while most of Southwest Asia's fou crops and livestock did spread west to Europe and east to the Indus Va neither of the Andes' domestic mammals (the llama / alpaca and the | pig) ever reached Mesoamerica in pre-Columbian times. That astor failure cries out for explanation. After all, Mesoamerica did develop farming populations and complex societies, so there can be no doubt i Andean domestic animals (if they had been available) would have valuable for food, transport, and wool. Except for dogs, Mesoamerica '\ utterly without indigenous mammals to fill those needs. Some South ican crops nevertheless did succeed in reaching Mesoamerica, such as i ioc, sweet potatoes, and peanuts. What selective barrier let those through but screened out llamas and guinea pigs?A subtler expression of this geographically varying ease of spread i|f phenomenon termed preemptive domestication. Most of the wild species from which our crops were derived vary genetically from : area, because alternative mutations had become established among; wild ancestral populations of different areas. Similarly, the required to transform wild plants into crops can in principle be

SPACIOUS

SKIES AND TILTED AXES • I 7 9t by alternative new mutations or alternative courses of selection to • Id equivalent results. In this light, one can examine a crop widespread historic times and ask whether all of its varieties show the same wild tation or same transforming mutation. The purpose of this examination is to try to figure out whether the crop was developed in just one area or else independently in several areas.
If one carries out such a genetic analysis for major ancient New World rops many of them prove to include two or more of those alternative wild variants, or two or more of those alternative transforming mutations. This suggests that the crop was domesticated independently in at least two different areas, and that some varieties of the crop inherited the particular mutation of one area while other varieties of the same crop inherited the mutation of another area. On this basis, botanists conclude that lima beans (Phaseolus lunatus), common beans (Phaseolus vulgaris), and chili peppers of the Capsicum annuutn I chinense group were all domesticated on at least two separate occasions, once in Mesoamerica and once in South America; and that the squash Cucurbita pepo and the seed plant goosefoot were also domesticated independently at least twice, once in Mesoamerica and once in the eastern United States. In contrast, most ancient Southwest Asian crops exhibit just one of the alternative wild variants or alternative transforming mutations, suggesting that all modern varieties of that particular crop stem from only a single domestication.What does it imply if the same crop has been repeatedly and independently domesticated in several different parts of its wild range, and not just once and in a single area? We have already seen that plant domestication involves the modification of wild plants so that they become more useful to humans by virtue of larger seeds, a less bitter taste, or other qualities. Hence if a productive crop is already available, incipient farmers will surely proceed to grow it rather than start all over again by gathering its not yet so useful wild relative and redomesticating it. Evidence for just a single domestication thus suggests that, once a wild plant had been domesticated, the crop spread quickly to other areas throughout the wild plant's range, preempting the need for other independent domestications o e same plant. However, when we find evidence that the same wild ancestor was domesticated independently in different areas, we infer that e crop spread too slowly to preempt its domestication elsewhere. The evi ence for predominantly single domestications in Southwest Asia, but requent multiple domestications in the Americas, might thus provide

I 8 O •

GUNS,GERMS, AND STEELmore subtle evidence that crops spread more easily out of Southwest Asia than in the Americas.
Rapid spread of a crop may preempt domestication not only of the same wild ancestral species somewhere else but also of related wild species. If you're already growing good peas, it's of course pointless to start from scratch to domesticate the same wild ancestral pea again, but it's also pointless to domesticate closely related wild pea species that for farmers are virtually equivalent to the already domesticated pea species. All of Southwest Asia's founder crops preempted domestication of any of their close relatives throughout the whole expanse of western Eurasia. In contrast, the New World presents many cases of equivalent and closely related, but nevertheless distinct, species having been domesticated in Meso-america and South America. For instance, 95 percent of the cotton grown in the world today belongs to the cotton species Gossypium hirsutum, which was domesticated in prehistoric times in Mesoamerica. However, prehistoric South American farmers instead grew the related cotton Gossypium barbadense. Evidently, Mesoamerican cotton had such difficulty reaching South America that it failed in the prehistoric era to preempt the domestication of a different cotton species there (and vice versa). Chili peppers, squashes, amaranths, and chenopods are other crops of which different but related species were domesticated in Mesoamerica and South America, since no species was able to spread fast enough to preempt the others.We thus have many different phenomena converging on the same conclusion: that food production spread more readily out of Southwest Asia than in the Americas, and possibly also than in sub-Saharan Africa. Those phenomena include food production's complete failure to reach some ecologically suitable areas; the differences in its rate and selectivity of spread; and the differences in whether the earliest domesticated crops preempted redomestications of the same species or domestications of close relatives. What was it about the Americas and Africa that made the spread of food production more difficult there than in Eurasia?
To answer this question, let's begin by examining the rapid spread of food production out of Southwest Asia (the Fertile Crescent). Soon after food production arose there, somewhat before 8000 b.c., a centrifugal wave of it appeared in other parts of western Eurasia and North Africa

SPACIOUS

SKIES AND TILTED AXES • I 8 Ifarther and farther removed from the Fertile Crescent, to the west and east. On this page I have redrawn the striking map (Figure 10.2) assembled by the geneticist Daniel Zohary and botanist Maria Hopf, in which they illustrate how the wave had reached Greece and Cyprus and the Indian subcontinent by 6500 b.c., Egypt soon after 6000 b.c., central Europe by 5400 b.c., southern Spain by 5200 b.c., and Britain around 3500 b.c. In each of those areas, food production was initiated by some of the same suite of domestic plants and animals that launched it in the Fertile Crescent. In addition, the Fertile Crescent package penetrated Africa southward to Ethiopia at some still-uncertain date. However, Ethiopia also developed many indigenous crops, and we do not yet know whether it was these crops or the arriving Fertile Crescent crops that launched Ethiopian food production.
The spread of Fertile Crescent crops across western Eurasia

o before 7000 bc

• 7000-6000 bc
O 6000-5000 BC
5000-3800 bc
3800-2500 BC

Figure 10.2. The symbols show early radiocarbon-dated sites where

remains of Fertile Crescent crops have been found. D = the Fertile Crescent itself (sites before 7000 b.c.). Note that dates become progressively later as one gets farther from the Fertile Crescent. This map is based onMap 20 of Zohary and Hopf's Domestication of Plants in the Old World but substitutes calibrated radiocarbon dates for their uncalibrated dates.

I 8 2. • GUNS, GERMS, AND

steelOf course, not all pieces of the package spread* to all those outlying areas: for example, Egypt was too warm for einkorn wheat to become established. In some outlying areas, elements of the package arrived at different times: for instance, sheep preceded cereals in southwestern Europe. Some outlying areas went on to domesticate a few local crops of their own, such as poppies in western Europe and watermelons possibly in Egypt. But most food production in outlying areas depended initially on Fertile Crescent domesticates. Their spread was soon followed by that of other innovations originating in or near the Fertile Crescent, including the wheel, writing, metalworking techniques, milking, fruit trees, and beet and wine production.
Why did the same plant package launch food production throughout western Eurasia? Was it because the same set of plants occurred in the wild in many areas, were found useful there just as in the Fertile Crescent, and were independently domesticated? No, that's not the reason. First, many of the Fertile Crescent's founder crops don't even occur in the wild outside Southwest Asia. For instance, none of the eight main founder crops except-barley grows wild in Egypt. Egypt's Nile Valley provides an environment similar to the Fertile Crescent's Tigris and Euphrates Valleys. Hence the$ package that worked well in the latter valleys also worked well enough in the Nile Valley to trigger the spectacular rise of indigenous Egyptian civilization. But the foods to fuel that spectacular rise were originally absent in Egypt. The sphinx and pyramids were built by people fed on crops originally native to the Fertile Crescent, not to Egypt.
Second, even for those crops whose wild ancestor does occur outside of Southwest Asia, we can be confident that the crops of Europe and India were mostly obtained from Southwest Asia and were not local domesticates. For example, wild flax occurs west to Britain and Algeria and east to the Caspian Sea, while wild barley occurs east even to Tibet. However, for most of the Fertile Crescent's founding crops, all cultivated varieties in the world today share only one arrangement of chromosomes out of the multiple arrangements found in the wild ancestor; or else they share only a single mutation (out of many possible mutations) by which the cultivated varieties differ from the wild ancestor in characteristics desirable TO humans. For instance, all cultivated peas share the same recessive gene that prevents ripe pods of cultivated peas from spontaneously popping ope* and spilling their peas, as wild pea pods do.
Evidently, most of the Fertile Crescent's founder crops were never

SPACIOUS

SKIES AND TILTED AXES • 183domesticated again elsewhere after their initial domestication in the Fertile Crescent. Had they been repeatedly domesticated independently, they would exhibit legacies of those multiple origins in the form of varied chromosomal arrangements or varied mutations. Hence these are typical examples of the phenomenon of preemptive domestication that we discussed above. The quick spread of the Fertile Crescent package preempted any possible other attempts, within the Fertile Crescent or elsewhere, to domesticate the same wild ancestors. Once the crop had become available, there was no further need to gather it from the wild and thereby set it on the path to domestication again.
The ancestors of most of the founder crops have wild relatives, in the Fertile Crescent and elsewhere, that would also have been suitable for domestication. For example, peas belong to the genus Pisum, which consists of two wild species: Pisum sativum, the one that became domesticated to yield our garden peas, and Pisum fulvum, which was never domesticated. Yet wild peas of Pisum fulvum taste good, either fresh or dried, and are common in the wild. Similarly, wheats, barley, lentil, chickpea, beans, and flax all have numerous wild relatives besides the ones that became domesticated. Some of those related beans and barleys were indeed domesticated independently in the Americas or China, far from the early site of domestication in the Fertile Crescent. But in western Eurasia only one of several potentially useful wild species was domesticated—probably because that one spread so quickly that people soon stopped gathering the other wild relatives and ate only the crop. Again as we discussed above, the crop's rapid spread preempted any possible further attempts to domesticate its relatives, as well as to redomesticate its ancestor.why was the spread of crops from the Fertile Crescent so rapid? The answer depends partly on that east-west axis of Eurasia with which I opened this chapter. Localities distributed east and west of each other at the same latitude share exactly the same day length and its seasonal variations. To a lesser degree, they also tend to share similar diseases, regimes of temperature and rainfall, and habitats or biomes (types of vegetation). For example, southern Italy, northern Iran, and Japan, all located at about the same latitude but lying successively 4,000 miles east or west of each other, are more similar to each other in climate than each is to a location lying even a mere 1,000 miles due south. On all the continents the habitat

184

" GUNS, GERMS, AND STEELtype known as tropical rain forest is confined to within about 10 degrees latitude of the equator, while Mediterranean scrub habitats (such as California's chaparral and Europe's maquis) lie between about 30 and 40 degrees of latitude.
But the germination, growth, and disease resistance of plants are adapted to precisely those features of climate. Seasonal changes of day length, temperature, and rainfall constitute signals that stimulate seeds to germinate, seedlings to grow, and mature plants to develop flowers, seeds, and fruit. Each plant population becomes genetically programmed, through natural selection, to respond appropriately to signals of the seasonal regime under which it has evolved. Those regimes vary greatly with latitude. For example, day length is constant throughout the year at the equator, but at temperate latitudes it increases as the months advance from the winter solstice to the summer solstice, and it then declines again through the next half of the year. The growing season—that is, the months with temperatures and day lengths suitable for plant growth—is shortest at high latitudes and longest toward the equator. Plants are also adapted to the diseases prevalent at their latitude.
Woe betide the plant whose genetic program is mismatched to the latitude of the field in which it is planted! Imagine a Canadian farmer foolish enough to plant a race of corn adapted to growing farther south, in Mexico. The unfortunate corn plant, following its Mexico-adapted genetic program, would prepare to thrust up its shoots in March, only to find itself still buried under 10 feet of snow. Should the plant become genetically reprogrammed so as to germinate at a time more appropriate to Canada— say, late June—the plant would still be in trouble for other reasons. Its genes would be telling it to grow at a leisurely rate, sufficient only to bring it to maturity in five months. That's a perfectly safe strategy in Mexico's mild climate, but in Canada a disastrous one that would guarantee the plant's being killed by autumn frosts before it had produced any mature corn cobs. The plant would also lack genes for resistance to diseases of northern climates, while uselessly carrying genes for resistance to diseases of southern climates. All those features make low-latitude plants poorly adapted to high-latitude conditions, and vice versa. As a consequence, most Fertile Crescent crops grow well in France and Japan but poorly at the equator.
Animals too are adapted to latitude-related features of climate. In that respect we are typical animals, as we know by introspection. Some of us

SPACIOUS

SKIES AND TILTED AXES • 185can't stand cold northern winters with their short days and characteristic germs, while others of us can't stand hot tropical climates with their own characteristic diseases. In recent centuries overseas colonists from cool northern Europe have preferred to emigrate to the similarly cool climates of North America, Australia, and South Africa, and to settle in the cool highlands within equatorial Kenya and New Guinea. Northern Europeans who were sent out to hot tropical lowland areas used to die in droves of diseases such as malaria, to which tropical peoples had evolved some genetic resistance.
That'i part of the reason why Fertile Crescent domesticates spread west and east so rapidly: they were already well adapted to the climates of the regions to which they were spreading. For instance, once farming crossed from the plains of Hungary into central Europe around 5400 b.c., it spread so quickly that the sites of the first farmers in the vast area from Poland west to Holland (marked by their characteristic pottery with linear decorations) were nearly contemporaneous. By the time of Christ, cereals of Fertile Crescent origin were growing over the 10,000-mile expanse from the Atlantic coast of Ireland to the Pacific coast of Japan. That west-east expanse of Eurasia is the largest land distance on Earth.
Thus, Eurasia's west-east axis allowed Fertile Crescent crops quickly to launch agriculture over the band of temperate latitudes from Ireland to the Indus Valley, and to enrich the agriculture that arose independently in eastern Asia. Conversely, Eurasian crops that were first domesticated far from the Fertile Crescent but at the same latitudes were able to diffuse back to the Fertile Crescent. Today, when seeds are transported over the whole globe by ship and plane, we take it for granted that our meals are a geographic mishmash. A typical American fast-food restaurant meal would include chicken (first domesticated in China) and potatoes (from the Andes) or corn (from Mexico), seasoned with black pepper (from India) and washed down with a cup of coffee (of Ethiopian origin). Already, though, by 2,000 years ago, Romans were also nourishing themselves with their own hodgepodge of foods that mostly originated elsewhere. Of Roman crops, only oats and poppies were native to Italy. Roman staples were the Fertile Crescent founder package, supplemented by quince (originating in the Caucasus); millet and cumin (domesticated in Central Asia); cucumber, sesame, and citrus fruit (from India); and chicken, rice, apricots, peaches, and foxtail millet (originally from China). Even though Rome's apples were at least native to western Eurasia, they were grown

I 8 6 • GUNS, GERMS, AND

steelby means of grafting techniques that had developed in China and spread westward from there.
While Eurasia provides the world's widest band of land at the same latitude, and hence the most dramatic example of rapid spread of domesticates, there are other examples as well. Rivaling in speed the spread of the Fertile Crescent package was the eastward spread of a subtropical package that was initially assembled in South China and that received additions on reaching tropical Southeast Asia, the Philippines, Indonesia, and New Guinea. Within 1,600 years that resulting package of crops (including bananas, taro, and yams) and domestic animals (chickens, pigs, and dogs) had spread more than 5,000 miles eastward into the tropical Pacific to reach the islands of Polynesia. A further likely example is the east-west spread of crops within Africa's wide Sahel zone, but paleobotanists have yet to work out the details.
contrast the ease of east-west diffusion in Eurasia with the difficulties of diffusion along Africa's north-south axis. Most of the Fertile Crescent founder crops reached Egypt very quickly and then spread as far south as the cool highlands of Ethiopia, beyond which they didn't spread. South Africa's Mediterranean climate would have been ideal for them, but the 2,000 miles of tropical conditions between Ethiopia and South Africa posed an insuperable barrier. Instead, African agriculture south of the Sahara was launched by the domestication of wild plants (such as sorghum and African yams) indigenous to the Sahel zone and to tropical West Africa, and adapted to the warm temperatures, summer rains, and relatively constant day lengths of those low latitudes.
Similarly, the spread southward of Fertile Crescent domestic animals through Africa was stopped or slowed by climate and disease, especially by trypanosome diseases carried by tsetse flies. The horse never became established farther south than West Africa's kingdoms north of the equator. The advance of cattle, sheep, and goats halted for 2,000 years at the northern edge of the Serengeti Plains, while new types of human economies and livestock breeds were being developed. Not until the period a.d. 1-200, some 8,000 years after livestock were domesticated in the Fertile Crescent, did cattle, sheep, and goats finally reach South Africa. Tropical African crops had their own difficulties spreading south in Africa, arriving in South Africa with black African farmers (the Bantu) just after those

SPACIOUS

SKIES AND TILTED AXES • 187Fertile Crescent livestock did. However, those tropical African crops could never be transmitted across South Africa's Fish River, beyond which they were stopped by Mediterranean conditions to which they were not adapted.
The result was the all-too-familiar course of the last two millennia of South African history. Some of South Africa's indigenous Khoisan peoples (otherwise known as Hottentots and Bushmen) acquired livestock but remained without agriculture. They became outnumbered and were replaced northeast of the Fish River by black African farmers, whose southward spread halted at that river. Only when European settlers arrived by sea in 1652, bringing with them their Fertile Crescent crop package, could agriculture thrive in South Africa's Mediterranean zone. The collisions of all those peoples produced the tragedies of modern South Africa: the quick decimation of the Khoisan by European germs and guns; a century of wars between Europeans and blacks; another century of racial oppression; and now, efforts by Europeans and blacks to seek a new mode of coexistence in the former Khoisan lands.
contrast also the ease of diffusion in Eurasia with its difficulties along the Americas' north-south axis. The distance between Mesoamerica and South America—say, between Mexico's highlands and Ecuador's—is only 1,200 miles, approximately the same as the distance in Eurasia separating the Balkans from Mesopotamia. The Balkans provided ideal growing conditions for most Mesopotamian crops and livestock, and received those domesticates as a package within 2,000 years of its assembly in the Fertile Crescent. That rapid spread preempted opportunities for domesticating those and related species in the Balkans. Highland Mexico and the Andes would similarly have been suitable for many of each other's crops and domestic animals. A few crops, notably Mexican corn, did indeed spread to the other region in the pre-Columbian era.
But other crops and domestic animals failed to spread between Mesoamerica and South America. The cool highlands of Mexico would have provided ideal conditions for raising llamas, guinea pigs, and potatoes, all domesticated in the cool highlands of the South American Andes. Yet the northward spread of those Andean specialties was stopped completely by the hot intervening lowlands of Central America. Five thousand years after lamas had been domesticated in the Andes, the Olmecs, Maya, Aztecs,

I 8 8 •

GUNS,GERMS, AND STEELand all other native societies of Mexico remained without pack animals and without any edible domestic mammals except for dogs.
Conversely, domestic turkeys of Mexico and domestic sunflowers of the eastern United States might have thrived in the Andes, but their southward spread was stopped by the intervening tropical climates. The mere 700 miles of north-south distance prevented Mexican corn, squash, and beans from reaching the U.S. Southwest for several thousand years after their domestication in Mexico, and Mexican chili peppers and chenopods never did reach it in prehistoric times. For thousands of years after corn was domesticated in Mexico, it failed to spread northward into eastern North America, because of the cooler climates and shorter growing season prevailing there. At some time between a.d. 1 and a.d. 200, corn finally appeared in the eastern United States but only as a very minor crop. Not until around a.d. 900, after hardy varieties of corn adapted to northern climates had been developed, could corn-based agriculture contribute to the flowering of the most complex Native American society of North America, the Mississippian culture—a brief flowering ended by European-introduced germs arriving with and after Columbus.
Recall that most Fertile Crescent crops prove, upon genetic study, to derive from only a single domestication process, whose resulting crop spread so quickly that it preempted any other incipient domestications of the same or related species. In contrast, many apparently widespread Native American crops prove to consist of related species or even of genetically distinct varieties of the same species, independently domesticated in Mesoamerica, South America, and the eastern United States. Closely related species replace each other geographically among the amaranths, beans, chenopods, chili peppers, cottons, squashes, and tobaccos. Different varieties of the same species replace each other among the kidney beans, lima beans, the chili pepper Capsicum annuum I chinense, and the squash Cucurbita pepo. Those legacies of multiple independent domestications may provide further testimony to the slow diffusion of crops along the Americas' north-south axis.Africa and the Americas are thus the two largest landmasses with a predominantly north-south axis and resulting slow diffusion. In certain other parts of the world, slow north-south diffusion was important on a smaller scale. These other examples include the snail's pace of crop exchange between Pakistan's Indus Valley and South India, the slow spread of South Chinese food production into Peninsular Malaysia, and

SPACIOUS SKIES AND TILTED AXES • 189

the failure of tropical Indonesian and New Guinean food production to arrive in prehistoric times in the modern farmlands of southwestern and southeastern Australia, respectively. Those two corners of Australia are now the continent's breadbaskets, but they lie more than 2,000 miles south of the equator. Farming there had to await the arrival from faraway Europe, on European ships, of crops adapted to Europe's cool climate and short growing season.
I have been dwelling on latitude, readily assessed by a glance at a map, because it is a major determinant of climate, growing conditions, and ease of spread of food production. However, latitude is of course not the only such determinant, and it is not always true that adjacent places at the same latitude have the same climate (though they do necessarily have the same day length). Topographic and ecological barriers, much more pronounced on some continents than on others, were locally important obstacles to diffusion.
For instance, crop diffusion between the U.S. Southeast and Southwest was very slow and selective although these two regions are at the same latitude. That's because much of the intervening area of Texas and the southern Great Plains was dry and unsuitable for agriculture. A corresponding example within Eurasia involved the eastern limit of Fertile Crescent crops, which spread rapidly westward to the Atlantic Ocean and eastward to the Indus Valley without encountering a major barrier. However, farther eastward in India the shift from predominantly winter rainfall to predominantly summer rainfall contributed to a much more delayed extension of agriculture, involving different crops and farming techniques, into the Ganges plain of northeastern India. Still farther east, temperate areas of China were isolated from western Eurasian areas with similar climates by the combination of the Central Asian desert, Tibetan plateau, and Himalayas. The initial development of food production in China was therefore independent of that at the same latitude in the Fertile Crescent, and gave rise to entirely different crops. However, even those barriers between China and western Eurasia were at least partly overcome during the second millennium b.c., when West Asian wheat, barley, and horses reached China.
By the same token, the potency of a 2,000-mile north-south shift as a barrier also varies with local conditions. Fertile Crescent food production

19o

' GUNS, GERMS, AND STEELspread southward over that distance to Ethiopia, and Bantu food production spread quickly from Africa's Great Lakes region south to Natal, because in both cases the intervening areas had similar rainfall regimes and were suitable for agriculture. In contrast, crop diffusion from Indonesia south to southwestern Australia was completely impossible, and diffusion over the much shorter distance from Mexico to the U.S. Southwest and Southeast was slow, because the intervening areas were deserts hostile to agriculture. The lack of a high-elevation plateau in Mesoamerica south of Guatemala, and Mesoamerica's extreme narrowness south of Mexico and especially in Panama, were at least as important as the latitudinal gradient in throttling crop and livestock exchanges between the highlands of Mexico and the Andes.
Continental differences in axis orientation affected the diffusion not only of food production but also of other technologies and inventions. For example, around 3,000 b.c. the invention of the wheel in or near Southwest Asia spread rapidly west and east across much of Eurasia within a few centuries, whereas the wheels invented independently in prehistoric Mexico never spread south to the Andes. Similarly, the principle of alphabetic writing, developed in the western part of the Fertile Crescent by 1500 b.c., spread west to Carthage and east to the Indian subcontinent within about a thousand years, but the Mesoamerican writing systems that flourished in prehistoric times for at least 2,000 years never reached the Andes.
Naturally, wheels and writing aren't directly linked to latitude and day length in the way crops are. Instead, the links are indirect, especially via food production systems and their consequences. The earliest wheels were parts of ox-drawn carts used to transport agricultural produce. Early writing was restricted to elites supported by food-producing peasants, and it served purposes of economically and socially complex food-producing societies (such as royal propaganda, goods inventories, and bureaucratic record keeping). In general, societies that engaged in intense exchanges of crops, livestock, and technologies related to food production were more likely to become involved in other exchanges as well.
America's patriotic song "America the Beautiful" invokes our spacious skies, our amber waves of grain, from sea to shining sea. Actually, that song reverses geographic realities. As in Africa, in the Americas the spread of native crops and domestic animals was slowed by constricted skies and environmental barriers. No waves of native grain ever stretched from the Atlantic to the Pacific coast of North America, from Canada to Patagonia,

SPACIOUS

SKIES AND TILTED AXES • 191or from Egypt to South Africa, while amber waves of wheat and barley came to stretch from the Atlantic to the Pacific across the spacious skies of Eurasia. That faster spread of Eurasian agriculture, compared with that of Native American and sub-Saharan African agriculture, played a role (as the next part of this book will show) in the more rapid diffusion of Eurasian writing, metallurgy, technology, and empires.
To bring up all those differences isn't to claim that widely distributed crops are admirable, or that they testify to the superior ingenuity of early Eurasian farmers. They reflect, instead, the orientation of Eurasia's axis compared with that of the Americas or Africa. Around those axes turned the fortunes of history.

CHAPTER

11lethal gift of livestock

WE HAVE NOW TRACED HOW FOOD PRODUCTION AROSE in a few centers, and how it spread at unequal rates from there to other areas. Those geographic differences constitute important ultimate answers to Yali's question about why different peoples ended up with disparate degrees of power and affluence. However, food production itself is not a proximate cause. In a one-on-one fight, a naked farmer would have no advantage over a naked hunter-gatherer.

Instead, one part of the explanation for farmer power lies in the much denser populations that food production could support: ten naked farmers certainly would have an advantage over one naked hunter-gatherer in a fight. The other part is that neither farmers nor hunter-gatherers are naked, at least not figuratively. Farmers tend to breathe out nastier germs, to own better weapons and armor, to own more-powerful technology in general, and to live under centralized governments with literate elites better able to wage wars of conquest. Hence the next four chapters will explore how the ultimate cause of food production led to the proximate causes of germs, literacy, technology, and centralized government.
Ine links connecting livestock and crops to germs were unforgettably 1 ustrated for me by a hospital case about which I learned through a physi-
n tnend. When my friend was an inexperienced young doctor, he was

I 9 6 •

GUNS,GERMS, AND STEELcalled into a hospital room to deal with a married couple stressed-out by a mysterious illness. It did not help that the couple was also having difficulty communicating with each other, and with my friend. The husband was a small, timid man, sick with pneumonia caused by an unidentified microbe, and with only limited command of the English language. Acting as translator was his beautiful wife, worried about her husband's condition and frightened by the unfamiliar hospital environment. My friend was also stressed-out from a long week of hospital work, and from trying to figure out what unusual risk factors might have brought on the strange illness. The stress caused my friend to forget everything he had been taught about patient confidentiality: he committed the awful blunder of requesting the woman to ask her husband whether he'd had any sexual experiences that could have caused the infection.
As the doctor watched, the husband turned red, pulled himself together so that he seemed even smaller, tried to disappear under his bedsheets, and stammered out words in a barely audible voice. His wife suddenly screamed in rage and drew herself up to tower over him. Before the doctor could stop her, she grabbed a heavy metal bottle, slammed it with full force onto her husband's head, and stormed out of the room. It took a while for the doctor to revive her husband and even longer to elicit, through the man's broken English, what he'd said that so enraged his wife. The answer slowly emerged: he had confessed to repeated intercourse with : sheep on a recent visit to the family farm; perhaps that was how he had'lf contracted the mysterious microbe.
This incident sounds bizarrely one-of-a-kind and of no possible broader significance. In fact, it illustrates an enormous subject of great importance: human diseases of animal origins. Very few of us love sheep in the carnal sense that this patient did. But most of us platonically love our pet animals, such as our dogs and cats. As a society, we certainly appear to have an inordinate fondness for sheep and other livestock, to judge from the vast numbers of them that we keep. For example, at the time of a recent census, Australia's 17,085,400 people thought so highly of sheep that they kept 161,600,000 of them.
Some of us adults, and even more of our children, pick up infectious diseases from our pets. Usually they remain no more than a nuisance, but a few have evolved into something far more serious. The major killers of humanity throughout our recent history—smallpox, flu, tuberculosis* malaria, plague, measles, and cholera—are infectious diseases that evolved

LETHAL

GIFT OF LIVESTOCK • 197from diseases of animals, even though most of the microbes responsible for our own epidemic illnesses are paradoxically now almost confined to humans. Because diseases have been the biggest killers of people, they have also been decisive shapers of history. Until World War II, more victims of war died of war-borne microbes than of battle wounds. All those military histories glorifying great generals oversimplify the ego-deflating truth: the winners of past wars were not always the armies with the best generals and weapons, but were often merely those bearing the nastiest germs to transmit to their enemies.
The grimmest examples of germs' role in history come from the European conquest of the Americas that began with Columbus's voyage of 1492. Numerous as were the Native American victims of the murderous Spanish conquistadores, they were far outnumbered by the victims of murderous Spanish microbes. Why was the exchange of nasty germs between the Americas and Europe so unequal? Why didn't Native American diseases instead decimate the Spanish invaders, spread back to Europe, and wipe out 95 percent of Europe's population? Similar questions arise for the decimation of many other native peoples by Eurasian germs, as well as for the decimation of would-be European conquistadores in the tropics of Africa and Asia.
Thus, questions of the animal origins of human disease lie behind the broadest pattern of human history, and behind some of the most important issues in human health today. (Think of AIDS, an explosively spreading human disease that appears to have evolved from a virus resident in wild African monkeys.) This chapter will begin by considering what a "disease" is, and why some microbes have evolved so as to "make us sick," whereas most other species of living things don't make us sick. We'll examine why many of our most familiar infectious diseases run in epidemics, such as our current AIDS epidemic and the Black Death (bubonic plague) epidemics of the Middle Ages. We'll then consider how the ancestors of microbes now confined to us transferred themselves from their original animal hosts. Finally, we'll see how insight into the animal origins of our infectious diseases helps explain the momentous, almost one-way exchange of germs between Europeans and Native Americans.
Naturally, we're disposed to think about diseases just from our own point of view: what can we do to save ourselves and to kill the

I 9 8 •

GUNS,GERMS, AND STEELmicrobes? Let's stamp out the scoundrels, and never mind what their motives are! In life in general, though, one has to understand the enemy in order to beat him, and that's especially true in medicine.
Hence let's begin by temporarily setting aside our human bias and considering disease from the microbes' point of view. After all, microbes are as much a product of natural selection as we are. What evolutionary benefit does a microbe derive from making us sick in bizarre ways, like giving us genital sores or diarrhea? And why should microbes evolve so as to kill us? That seems especially puzzling and self-defeating, since a microbe that kills its host kills itself.
Basically, microbes evolve like other species. Evolution selects for those individuals most effective at producing babies and at helping them spread to suitable places to live. For a microbe, spread may be defined mathematically as the number of new victims infected per each original patient. That number depends on how long each victim remains capable of infecting new victims, and how efficiently the microbe is transferred from one victim to the next.
Microbes have evolved diverse ways of spreading from one person to another, and from animals to people. The germ that spreads better leaves more babies and ends up favored by natural selection. Many of our "symptoms" of disease actually represent ways in which some damned clever microbe modifies our bodies or our behavior such that we become enlisted to spread microbes.
The most effortless way a germ could spread is by just waiting to be transmitted passively to the next victim. That's the strategy practiced by microbes that wait for one host to be eaten by the next host: for instance, salmonella bacteria, which we contract by eating already infected eggs or meat; the worm responsible for trichinosis, which gets from pigs to us by waiting for us to kill the pig and eat it without proper cooking; and the worm causing anisakiasis, with which sushi-loving Japanese and Americans occasionally infect themselves by consuming raw fish. Those parasites pass to a person from an eaten animal, but the virus causing laughing sickness (kuru) in the .New Guinea highlands used to pass to a person from another person who was eaten. It was transmitted by cannibalism, when highland babies made the fatal mistake of licking their fingers after playing with raw brains that their mothers had just cut out of dead kuru victims awaiting cooking.
Some microbes don't wait for the old host to die and get eaten, but

LETHAL

GIFT OF LIVESTOCK • 199instead hitchhike in the saliva of an insect that bites the old host and flies off to find a new host. The free ride may be provided by mosquitoes, fleas, lice, or tsetse flies that spread malaria, plague, typhus, or sleeping sickness, respectively. The dirtiest of all tricks for passive carriage is perpetrated by microbes that pass from a woman to her fetus and thereby infect babies already at birth. By playing that trick, the microbes responsible for syphilis rubella, and now AIDS pose ethical dilemmas with which believers in a fundamentally just universe have had to struggle desperately.
Other germs take matters into their own hands, figuratively speaking. They modify the anatomy or habits of their host in such a way as to accelerate their transmission. From our perspective, the open genital sores caused by venereal diseases like syphilis are a vile indignity. From the microbes' point of view, however, they're just a useful device to enlist a host's help in inoculating microbes into a body cavity of a new host. The skin lesions caused by smallpox similarly spread microbes by direct or indirect body contact (occasionally very indirect, as when U.S. whites bent on wiping out "belligerent" Native Americans sent them gifts of blankets previously used by smallpox patients).
More vigorous yet is the strategy practiced by the influenza, common cold, and pertussis (whooping cough) microbes, which induce the victim to cough or sneeze, thereby launching a cloud of microbes toward prospective new hosts. Similarly, the cholera bacterium induces in its victim a massive diarrhea that delivers bacteria into the water supplies of potential new victims, while the virus responsible for Korean hemorrhagic fever broadcasts itself in the urine of mice. For modification of a host's behavior, nothing matches rabies virus, which not only gets into the saliva of an infected dog but drives the dog into a frenzy of biting and thus infecting many new victims. But for physical effort on the bug's own part, the prize still goes to worms such as hookworms and schistosomes, which actively burrow through a host's skin from the water or soil into which their larvae had been excreted in a previous victim's feces.
Thus, from our point of view, genital sores, diarrhea, and coughing are symptoms of disease." From a germ's point of view, they're clever evolutionary strategies to broadcast the germ. That's why it's in the germ's interests to "make us sick." But why should a germ evolve the apparently self-defeating strategy of killing its host?
From the germ's perspective, that's just an unintended by-product (fat consolation to us!) of host symptoms promoting efficient transmission of

Z O O •

GUNS,GERMS, AND STEELmicrobes. Yes, an untreated cholera patient may eventually die from producing diarrheal fluid at a rate of several gallons per day. At least for a while, though, as long as the patient is still alive, the cholera bacterium profits from being massively broadcast into the water supplies of its next victims. Provided that each victim thereby infects on the average more than one new victim, the bacterium will spread, even though the first host happens to die.
'*
So much FOR our dispassionate examination of the germ's interests. Now let's get back to considering our own selfish interests: to stay alive and healthy, best done by killing the damned germs. One common response of ours to infection is to develop a fever. Again, we're used to considering fever as a "symptom of disease," as if it developed inevitably without serving any function. But regulation of body temperature is under our genetic control and doesn't just happen by accident. A few microbes are more sensitive to heat than our own bodies are. By raising our body temperature, we in effect try to bake the germs to death before we get baked ourselves.
Another common response of ours is to mobilize our immune system. White blood cells and other cells of ours actively seek out and kill foreign microbes. The specific antibodies that we gradually build up against a particular microbe infecting us make us less likely to get reinfected once we become cured. As we all know from experience, there are some illnesses, such as flu and the common cold, to which our resistance is only temporary; we can eventually contract the illness again. Against other illnesses, though—including measles, mumps, rubella, pertussis, and the now defeated smallpox—our antibodies stimulated by one infection confer lifelong immunity. That's the principle of vaccination: to stimulate our antibody production without our having to go through the actual experience of the disease, by inoculating us with a dead or weakened strain of microbe.
Alas, some clever microbes don't just cave in to our immune defenses. Some have learned to trick us by changing those molecular pieces of the microbe (its so-called antigens) that our antibodies recognize. The constant evolution or recycling of new strains of flu, with differing antigens, explains why your having gotten flu two years ago didn't protect you

LETHAL

GIFT OF LIVESTOCK • 2.OIeainst the different strain that arrived this year. Malaria and sleeping sickness are even more slippery customers in their ability rapidly to change their antigens– Among the slipperiest of all is AIDS, which evolves new antigens even as it sits within an individual patient, thereby eventually overwhelming his or her immune system.
Our slowest defensive response is through natural selection, which changes our gene frequencies from generation to generation. For almost any disease, some people prove to be genetically more resistant than are others. In an epidemic those people with genes for resistance to that particular microbe are more likely to survive than are people lacking such genes. As a result, over the course of history, human populations repeatedly exposed to a particular pathogen have come to consist of a higher proportion of individuals with those genes for resistance—just because unfortunate Individuals without the genes were less likely to survive to pass their genes on to babies.
Fat consolation, you may be thinking again. This evolutionary response is not one that does the genetically susceptible dying individual any good. It does mean, though, that a human population as a whole becomes better protected against the pathogen. Examples of those genetic defenses include the protections (at a price) that the sickle-cell gene, Tay-Sachs gene, and cystic fibrosis gene may confer on African blacks, Ashkenazi Jews, and northern Europeans against malaria, tuberculosis, and bacterial diarrheas, respectively.
In short, our interaction with most species, as exemplified by hummingbirds, doesn't make us or the hummingbird "sick." Neither we nor hummingbirds have had to evolve defenses against each other. That peaceful relationship was able to persist because hummingbirds don't count on us to spread their babies or to offer our bodies for food. Hummingbirds evolved instead to feed on nectar and insects, which they find by using their own wings.
But microbes evolved to feed on the nutrients within our own bodies, and they don't have wings to let them reach a new victim's body once the original victim is dead or resistant. Hence many germs have had to evolve tricks to let them spread between potential victims, and many of those tricks are what we experience as "symptoms of disease." We've evolved countertricks of our own, to which the germs have responded by evolving counter-couritertricks. We and our pathogens are now locked in an escalat-

1 O 2 •

GUNS,GERMS, AND STEELing evolutionary contest, with the death of one contestant the price of defeat, and with natural selection playing the role of umpire. Now let's consider the form of the contest: blitzkrieg or guerrilla war?
Suppose that one counts the number of cases of some particular infectious disease in some geographic area, and watches how the numbers change with time. The resulting patterns differ greatly among diseases. For certain diseases, like malaria or hookworm, new cases appear any month of any year in an affected area. So-called epidemic diseases, though, produce no cases for a long time, then a whole wave of cases, then no more cases again for a while.
Among such epidemic diseases, influenza is one personally familiar to most Americans, certain years being particularly bad years for us (but great years for the influenza virus). Cholera epidemics come at longer intervals, the 1991 Peruvian epidemic being the first one to reach the New World during the 20th century. Although today's influenza and cholera epidemics make front-page stories, epidemics used to be far more terrifying before the rise of modern medicine. The greatest single epidemic in human history was the one of influenza that killed 21 million people at the end of the First World War. The Black Death (bubonic plague) killed one-quarter of Europe's population between 1346 and 1352, with death tolls ranging up to 70 percent in some cities. When the Canadian Pacific Railroad was being built through Saskatchewan in the early 1880s, that province's Native Americans, who had previously had little exposure to whites and their germs, died of tuberculosis at the incredible rate of 9 percent per year.
The infectious diseases that visit us as epidemics, rather than as a steady trickle of cases, share several characteristics. First, they spread quickly and efficiently from an infected person to nearby healthy people, with the result that the whole population gets exposed within a short time. Second, they're "acute" illnesses: within a short time, you either die or recover completely. Third, the fortunate ones of us who do recover develop antibodies that leave us immune against a recurrence of the disease for a long time, possibly for the rest of our life. Finally, these diseases tend to be restricted to humans; the microbes causing them tend not to live in the soil or in other animals. All four of these traits apply to what Americans think

LETHAL

GIFT OF LIVESTOCK • Z O 3of as the familiar acute epidemic diseases of childhood, including measles, rubella, mumps, pertussis, and smallpox.
The reason why the combination of those four traits tends to make a disease run in epidemics is easy to understand. In simplified form, here's what happens. The rapid spread of microbes, and the rapid course of symptoms, mean that everybody in a local human population is quickly infected and soon thereafter is either dead or else recovered and immune. No one is left alive who could still be infected. But since the microbe can't survive except in the bodies of living people, the disease dies out, until a new crop of babies reaches the susceptible age—and until an infectious person arrives from the outside to start a new epidemic.
A classic illustration of how such diseases occur as epidemics is the history of measles on the isolated Atlantic islands called the Faeroes. A severe epidemic of measles reached the Faeroes in 1781 and then died out, leaving the islands measles free until an infected carpenter arrived on a ship from Denmark in 1846. Within three months, almost the whole Faeroes population (7,782 people) had gotten measles and then either died or recovered, leaving the measles virus to disappear once again until the next epidemic. Studies show that measles is likely to die out in any human population numbering fewer than half a million people. Only in larger populations can the disease shift from one local area to another, thereby persisting until enough babies have been born in the originally infected area that measles can return there.
What's true for measles in the Faeroes is true of our other familiar acute infectious diseases throughout the world. To sustain themselves, they need a human population that is sufficiently numerous, and sufficiently densely packed, that a numerous new crop of susceptible children is available for infection by the time the disease would otherwise be waning. Hence measles and similar diseases are also known as" crowd diseases.
vJBviousLY, crowd diseases could not sustain themselves in small bands of hunter-gatherers and slash-and-burn farmers. As tragic modern experience with Amazonian Indians and Pacific Islanders confirms, almost an entire tribelet may be wiped out by an epidemic brought by an outside visitor—because no one in the tribelet had any antibodies against the microbe. For example, in the winter of 1902 a dysentery epidemic brought

Z O 4 *

GUNS,GERMS, AND STEELby a sailor on the whaling ship Active killed 51 out of the 56 Sadlermiut Eskimos, a very isolated band of people living on Southampton Island in the Canadian Arctic. In addition, measles and some of our other "childhood" diseases are more likely to kill infected adults than children, and all adults in the tribelet are susceptible. (In contrast, modern Americans rarely contract measles as adults, because most of them get either measles or the vaccine against it as children.) Having killed most of the tribelet, the epidemic then disappears. The small population size of tribelets explains not only why they can't sustain epidemics introduced from the outside, but also why they never could evolve epidemic diseases of their own to give back to visitors.
That's not to say, though, that small human populations are free from all infectious diseases. They do have infections, but only of certain types. Some are caused by microbes capable of maintaining themselves in animals or in the soil, with the result that the disease doesn't die out but remains constantly available to infect people. For example, the yellow fever virus is carried by African wild monkeys, whence it can always infect rural human populations of Africa, whence it was carried by the transatlantic slave trade to infect New World monkeys and people.
Still other infections of small human populations are chronic diseases such as leprosy and yaws. Since the disease may take a very long time to kill its victim, the victim remains alive as a reservoir of microbes to infect other members of the tribelet. For instance, the Karimui Basim of the New Guinea highlands, where I worked in the 1960s, was occupied by an isolated population of a few thousand people, suffering from the world's highest incidence of leprosy—about 40 percent! Finally, small human populations are also susceptible to nonfatal infections against which we don't develop immunity, with the result that the same person can become rein-fected after recovering. That happens with hookworm and many other parasites.
All these types of diseases, characteristic of small isolated populations, must be the oldest diseases of humanity. They were the ones we could evolve and sustain through the early millions of years of our evolutionary history, when the total human population was tiny and fragmented. These diseases are also shared with, or similar to the diseases of, our closest wild relatives, the African great apes. In contrast, the crowd diseases, which we discussed earlier, could have arisen only with the buildup of large, dem*,; human populations. That buildup began with the rise of agriculture start*

LETHAL

GIFT OF LIVESTOCK • 2 O 5ing about 10,000 years ago and then accelerated with the rise of cities starting several thousand years ago. In fact, the first attested dates for many familiar infectious diseases are surprisingly recent: around 1600 b.c. for smallpox (as deduced from pockmarks on an Egyptian mummy), 400 b c. for mumps, 200 b.c. for leprosy, a.d. 1840 for epidemic polio, and 1959 for AIDS.
DID THEse o^ agriculture launch the evolution of our crowd infectious diseases? One reason just mentioned is that agriculture sustains much higher human population densities than does the hunting-gathering lifestyle—on the average, 10 to 100 times higher. In addition, hunter-gatherers frequently shift camp and leave behind their own piles of feces with accumulated microbes and worm larvae. But farmers are sedentary and live amid their own sewage, thus providing microbes with a short path from one person's body into another's drinking water.
Some farming populations make it even easier for their own fecal bacteria and worms to infect new victims, by gathering their feces and urine and spreading them as fertilizer on the fields where people work. Irrigation agriculture and fish farming provide ideal living conditions for the snails carrying schistosomiasis and for flukes that burrow through our skin as we wade through the feces-laden water. Sedentary farmers become surrounded not only by their feces but also by disease transmitting rodents, attracted by the farmers' stored food. The forest clearings made by African farmers also provide ideal breeding habitats for malaria-transmitting mosquitoes.
If the rise of farming was thus a bonanza for our microbes, the rise of cities was a greater one, as still more densely packed human populations festered under even worse sanitation conditions. Not until the beginning of the 20th century did Europe's urban populations finally become self-sustaining: before then, constant immigration of healthy peasants from the countryside was necessary to make up for the constant deaths of city dwellers from crowd diseases. Another bonanza was the development of world trade routes, which by Roman times effectively joined the populations of Europe, Asia, and North Africa into one giant breeding ground for microbes. That's when smallpox finally reached Rome, as the Plague of Antoninus, which killed millions of Roman citizens between a.d. 165 and 180.

1

O 6 •GUNS,GERMS, AND STEELSimilarly, bubonic plague first appeared in Europe as the Plague of Justinian (a.d. 542-43). But plague didn't begin to hit Europe with full force as the Black Death epidemics until a.d. 1346, when a new route for overland trade with China provided rapid transit, along Etfrasia's east-west axis, for flea-infested furs from plague-ridden areas of Central Asia to Europe. Today, our jet planes have made even the longest intercontinental flights briefer than the duration of any human infectious disease. That's how an Aerolineas Argentinas airplane, stopping in Lima (Peru) in 1991, managed to deliver dozens of cholera-infected people that same day to my city of Los Angeles, over 3,000 miles from Lima. The explosive increase in world travel by Americans, and in immigration to the United States, is turning us into another melting pot—this time, of microbes that we previously dismissed as just causing exotic diseases in far-off countries.
Thus,whenthe human population became sufficiently large and concentrated, we reached the stage in our history at which we could at last evolve and sustain crowd diseases confined to our own species. But that conclusion presents a paradox: such diseases could never have existed before then! Instead, they had to evolve as new diseases. Where did those new diseases come from?
Evidence has recently been emerging from molecular studies of the disease-causing microbes themselves. For many of the microbes responsible for our unique diseases, molecular biologists can now identify the microbe's closest relatives. These also prove to be agents of crowd infectious diseases—but ones confined to various species of our domestic animals and pets! Among animals, too, epidemic diseases require large, dense populations and don't afflict just any animal: they're confined mainly to social animals providing the necessary large populations. Hence when we domesticated social animals, such as cows and pigs, they were already afflicted by epidemic diseases just waiting to be transferred to us.
For example, measles virus is most closely related to the virus causing rinderpest. That nasty epidemic disease affects cattle and many wild cud-chewing mammals, but not humans. Measles in turn doesn't afflict cattle. The close similarity of the measles virus to the rinderpest virus suggests that the latter transferred from cattle to humans and then evolved into the measles virus by changing its properties to adapt to" us. That transfer is not at all surprising, considering that many peasant farmers live and sleep

LETHAL

GIFT OF LIVESTOCK • Z O 7close to cows and their feces, urine, breath, sores, and blood. Our intimacy with cattle has been going on for the 9,000 years since we domesticated them—ample time for the rinderpest virus to discover us nearby. As Table 11.1 illustrates, others of our familiar infectious diseases can similarly be traced back to diseases of our animal friends.
cjiven our proximity to the animals we love, we must be getting constantly bombarded by their microbes. Those invaders get winnowed by natural selection, and only a few of them succeed in establishing themselves as human diseases. A quick survey of current diseases lets us trace out four stages in the evolution of a specialized human disease from an animal precursor.
The first stage is illustrated by dozens of diseases that we now and then pick up directly from our pets and domestic animals. They include cat-scratch fever from our cats, leptospirosis from our dogs, psittacosis from our chickens and parrots, and brucellosis from our cattle. We're similarly liable to pick up diseases from wild animals, such as the tularemia that hunters can get from skinning wild rabbits. All those microbes are still at an early stage in their evolution into specialized human pathogens. They still don't get transmitted directly from one person to another, and even their transfer to us from animals remains uncommon.
In the second stage a former animal pathogen evolves to the point where it does get transmitted directly between people and causes epidemics.
tablen.i Deadly Gifts from Our Animal FrendsAnimal with Most Closely
Human Disease Related PathogenMeasles cattle (rinderpest)
Tuberculosis cattle
Smallpox cattle (cowpox) or other livestock with
related pox viruses
Flu pigs and ducks
Pertussis pigs, dogs
Falciparum malaria birds (chickens and ducks?)

X O 8 •

GUNS,GERMS, AND STEELHowever, the epidemic dies out for any of several reasons, such as being cured by modern medicine, or being stopped when everybody around has already been infected and either becomes immune or dies. For example, a previously unknown fever termed O'nyong-nyong fever appeared in East Africa in 1959 and proceeded to infect several million Africans. It probably arose from a virus of monkeys and was transmitted to humans by mosquitoes. The fact that patients recovered quickly and became immune to further attack helped the new disease die out quickly. Closer to home for Americans, Fort Bragg fever was the name applied to a new leptospiral disease that broke out in the United States in the summer of 1942 and soon disappeared.
A fatal disease vanishing for another reason was New Guinea's laughing sickness, transmitted by cannibalism and caused by a slow-acting virus from which no one has ever recovered. Kuru was on its way to exterminating New Guinea's Fore tribe of 20,000 people, until the establishment of Australian government control around 1959 ended cannibalism and thereby the transmission of kuru. The annals of medicine are full of accounts of diseases that sound like no disease known today, but that once caused terrifying epidemics and then disappeared as mysteriously as they had come. The "English sweating sickness," which swept and terrified Europe between 1485 and 1552, and the "Picardy sweats" of 18th– and 19th-century France, are just two of the many epidemic illnesses that vanished long before modern medicine had devised methods for identifying the responsible microbes.
A third stage in the evolution of our major diseases is represented by former animal pathogens that did establish themselves in humans, that have not (not yet?) died out, and that may or may not still become major killers of humanity. The future remains very uncertain for Lassa fever, caused by a virus derived probably from rodents. Lassa fever was first observed in 1969 in Nigeria, where it causes a fatal illness so contagious that Nigerian hospitals have been closed down if even a single case appears. Better established is Lyme disease, caused by a spirochete that we get from the bite of ticks carried by mice and deer. Although the first known human cases in the United States appeared only as recently as 1962, Lyme disease is already reaching epidemic proportions in many parts of our country. The future of AIDS, derived from monkey viruses and first documented in humans around 1959, is even more secure (from the virus's perspective).

LETHAL

GIFT OF LIVESTOCK • Z O 9The final stage of this evolution is represented by the major, long-established epidemic diseases confined to humans. These diseases must have been the evolutionary survivors of far more pathogens that tried to make the jump to us from animals—and mostly failed.
What is actually going on in those stages, as an exclusive disease of animals transforms itself into an exclusive disease of humans? One transformation involves a change of intermediate vector: when a microbe relying on some arthropod vector for transmission switches to a new host, the microbe may be forced to find a new arthropod as well. For example, typhus was initially transmitted between rats by rat fleas, which sufficed for a while to transfer typhus from rats to humans. Eventually, typhus microbes discovered that human body lice offered a much more efficient method of traveling directly between humans. Now that Americans have mostlv deloused themselves, typhus has discovered a new route into us: by infecting eastern North American flying squirrels and then transferring to people whose attics harbor flying squirrels.
In short, diseases represent evolution in progress, and microbes adapt by natural selection to new hosts and vectors. But compared with cows' bodies, ours offer different immune defenses, lice, feces, and chemistries. In that new environment, a microbe must evolve new ways to live and to propagate itself. In several instructive cases doctors or veterinarians have actually been able to observe microbes evolving those new ways.
The best-studied case involves what happened when myxomatosis hit Australian rabbits. The myxo virus, native to a wild species of Brazilian rabbit, had been observed to cause a lethal epidemic in European domestic rabbits, which are a different species. Hence the virus was intentionally introduced to Australia in 1950 in the hopes of ridding the continent of its plague of European rabbits, foolishly introduced in the nineteenth century. In the first year, myxo produced a gratifying (to Australian farmers) 99.8 percent mortality rate in infected rabbits. Unfortunately for the farmers, the death rate then dropped in the second year to 90 percent and eventually to 25 percent, frustrating hopes of eradicating rabbits completely from Australia. The problem was that the myxo virus evolved to serve its own interests, which differed from ours as well as from those of the rabbits. The virus changed so as to kill fewer rabbits and to permit lethally infected ones to live longer before dying. As a result, a less lethal myxo virus spreads baby viruses to more rabbits thaa did the original, highly virulent

2. I O •

GUNS,GERMS, AND STEELFor a similar example in humans, we have only to consider the surprising evolution of syphilis. Today, our two immediate associations to syphilis are genital sores and a very slowly developing disease, leading to the death of many untreated victims only after many years. However, when syphilis was first definitely recorded in Europe in 1495, its pustules often covered the body from the head to the knees, caused flesh to fall off people's faces, and led to death within a few months. By 1546, syphilis had evolved into the disease with the symptoms so well known to us today. Apparently, just as with myxomatosis, those syphilis spirochetes that evolved so as to keep their victims alive for longer were thereby able to transmit their spirochete offspring into more victims.
The importance of lethal microbes in human history is well illustrated by Europeans' conquest and depopulation of the New World. Far more Native Americans died in bed from Eurasian germs than on the battlefield from European guns and swords. Those germs undermined Indian resistance by killing most Indians and their leaders and by sapping the survivors' morale. For instance, in 1519 Cortes landed on the coast of Mexico with 600 Spaniards, to conquer the fiercely militaristic Aztec Empire with a population of many millions. That Cortes reached the Aztec capital of Tenochtitlan, escaped with the loss of "only" two-thirds of his force, and managed to fight his way back to the coast demonstrates both Spanish military advantages and the initial naivete of the Aztecs. But when Cortes's next onslaught came, the Aztecs were no longer naive and fought street by street with the utmost tenacity. What gave the Spaniards a decisive advantage was smallpox, which reached Mexico in 1520 with one infected slave arriving from Spanish Cuba. The resulting epidemic proceeded to kill nearly half of the Aztecs, including Emperor Cuitlahuac. Aztec survivors were demoralized by the mysterious illness that killed Indians and spared Spaniards, as if advertising the Spaniards' invincibility. By 1618, Mexico's initial population of about 20 million had plummeted to about 1.6 million.
Pizarro had similarly grim luck when he landed on the coast of Peru in 1531 with 168 men to conquer the Inca Empire of millions. Fortunately for Pizarro and unfortunately for the Incas, smallpox had arrived overland around 1526, killing much of the Inca population, including both the emperor Huayna Capac and his designated successor. As we saw in Chap-

LETHAL

GIFT OF LIVESTOCK • IIIter 3 the result of the throne's being left vacant was that two other sons of Huayna Capac, Atahuallpa and Huascar, became embroiled in a civil war that Pizarro exploited to conquer the divided Incas.
When we in the United States think of the most populous New World societies existing in 1492, only those of the Aztecs and the Incas tend to come to our minds. We forget that North America also supported populous Indian societies in the most logical place, the Mississippi Valley, which contains some of our best farmland today. In that case, however, conquis-tadores contributed nothing directly to the societies' destruction; Eurasian germs, spreading in advance, did everything. When Hernando de Soto became the first European conquistador to march through the southeastern United States, in 1540, he came across Indian town sites abandoned two years earlier because the inhabitants had died in epidemics. These epidemics had been transmitted from coastal Indians infected by Spaniards visiting the coast. The Spaniards' microbes spread to the interior in advance of the Spaniards themselves.
De Soto was still able to see some of the densely populated Indian towns lining the lower Mississippi. After the end of his expedition, it was a long time before Europeans again reached the Mississippi Valley, but Eurasian microbes were now established in North America and kept spreading. By the time of the next appearance of Europeans on the lower Mississippi, that of French settlers in the late 1600s, almost all of those big Indian towns had vanished. Their relics are the great mound sites of the Mississippi Valley. Only recently have we come to realize that many of the mound-building societies were still largely intact when Columbus reached the New World, and that they collapsed (probably as a result of disease) between 1492 and the systematic European exploration of the Mississippi.
When I was young, American schoolchildren were taught that North America had originally been occupied by only about one million Indians. That low number was useful in justifying the white conquest of what could be viewed as an almost empty continent. However, archaeological excavations, and scrutiny of descriptions left by the very first European explorers on our coasts, now suggest an initial number of around 20 million Indians. For the New World as a whole, the Indian population decline in the century or two following Columbus's arrival is estimated to have been as large as 95 percent.
The main killers were Old World germs to which Indians had never een exposed, and against which they therefore had neither immune nor

ill

• GUNS, GERMS, AND STEELgenetic resistance. Smallpox, measles, influenza, and typhus competed for top rank among the killers. As if these had not been enough, diphtheria, malaria, mumps, pertussis, plague, tuberculosis, and yellow fever came up close behind. In countless cases, whites were actually there to witness the destruction occurring when the germs arrived. For example, in 1837 the Mandan Indian tribe, with one of the most elaborate cultures in our Great Plains, contracted smallpox from a steamboat traveling up the Missouri River from St. Louis. The population of one Mandan village plummeted from 2,000 to fewer than 40 within a few weeks.
while over a dozen major infectious diseases of Old World origins became established in the New World, perhaps not a single major killer reached Europe from the Americas. The sole possible exception is syphilis, whose area of origin remains controversial. The one-sidedness of that exchange of germs becomes even more striking when we recall that large, dense human populations are a prerequisite for the evolution of our crowd infectious diseases. If recent reappraisals of the pre-Columbian New World population are correct, it was not far below the contemporary population of Eurasia. Some New World cities like Tenochtitlan were among the world's most populous cities at the time. Why didn't Tenochtitlan have awful germs waiting for the Spaniards?
One possible contributing factor is that the rise of dense human populations began somewhat later in the New World than in the Old World. Another is that the three most densely populated American centers—the Andes, Mesoamerica, and the Mississippi Valley—never became connected by regular fast trade into one huge breeding ground for microbes, in the way that Europe, North Africa, India, and China became linked in Roman times. Those factors still don't explain, though, why the New World apparently ended up with no lethal crowd epidemics at all. (Tuberculosis DNA has been reported from the mummy of a Peruvian Indian who died 1,000 years ago, but the identification procedure used did not distinguish human tuberculosis from a closely related pathogen (Mycobacteriutn •; bovis) that is widespread in wild animals.)Instead, what must be the main reason for the failure of lethal crowd epidemics to arise in the Americas becomes clear when we pause to ask a simple question. From what microbes could they conceivably have evolved? We've seen that Eurasian crowd diseases evolved out of disease*^

LETHAL

GIFT OF LIVESTOCK • 2. I 3of Eurasian herd animals that became domesticated. Whereas many such animals existed in Eurasia, only five animals of any sort became domesticated in the Americas: the turkey in Mexico and the U.S. Southwest, the llama / alpaca and the guinea pig in the Andes, the Muscovy duck in tropical South America, and the dog throughout the Americas.
In turn, we also saw that this extreme paucity of domestic animals in the New World reflects the paucity of wild starting material. About 80 percent of the big wild mammals of the Americas became extinct at the end of the last Ice Age, around 13,000 years ago. The few domesticates that remained to Native Americans were not likely sources of crowd diseases, compared with cows and pigs. Muscovy ducks and turkeys don't live in enormous flocks, and they're not cuddly species (like young lambs) with which we have much physical contact. Guinea pigs may have contributed a trypanosome infection like Chagas' disease or leishmaniasis to our catalog of woes, but that's uncertain. Initially, most surprising is the absence of any human disease derived from llamas (or alpacas), which it's tempting to consider the Andean equivalent of Eurasian livestock. However, llamas had four strikes against them as a source of human pathogens: they were kept in smaller herds than were sheep and goats and pigs; their total numbers were never remotely as large as those of the Eurasian populations of domestic livestock, since llamas never spread beyond the Andes; people don't drink (and get infected by) llama milk; and llamas aren't kept indoors, in close association with people. In contrast, human mothers in the New Guinea highlands often nurse piglets, and pigs as well as cows are frequently kept inside the huts of peasant farmers.
The historical importance of animal-derived diseases extends far beyond the collision of the Old and the New Worlds. Eurasian germs played a key role in decimating native peoples in many other parts of the world, including Pacific islanders, Aboriginal Australians, and the Khoisan peoples (Hottentots and Bushmen) of southern Africa. Cumulative mortalities of these previously unexposed peoples from Eurasian germs ranged from 50 percent to 100 percent. For instance, the Indian population of Hispaniola declined from around 8 million, when Columbus arrived in a.d. 1492, to zero by 1535. Measles reached Fiji with a Fijian chief returning from a visit to Australia in 1875, and proceeded to kill about one-quarter of all Fijians then alive (after most Fijians had already been

114

' GUNS, GERMS, AND STEELkilled by epidemics beginning with the first European visit, in 1791). Syphilis, gonorrhea, tuberculosis, and influenza arriving with Captain Cook in 1779, followed by a big typhoid epidemic in 1804 and numerous "minor" epidemics, reduced Hawaii's population from around half a million in 1779 to 84,000 in 1853, the year when smallpox finally reached Hawaii and killed around 10,000 of the survivors. These examples could be multiplied almost indefinitely.
However, germs did not act solely to Europeans' advantage. While the New World and Australia did not harbor native epidemic diseases awaiting Europeans, tropical Asia, Africa, Indonesia, and New Guinea certainly did. Malaria throughout the tropical Old World, cholera in tropical Southeast Asia, and yellow fever in tropical Africa were (and still are) the most notorious of the tropical killers. They posed the most serious obstacle to European colonization of the tropics, and they explain why the European colonial partitioning of New Guinea and most of Africa was not accomplished until nearly 400 years after European partitioning of the New World began. Furthermore, once malaria and yellow fever did become transmitted to the Amencas by European ship traffic, they emerged as the major impediment to colonization of the New World tropics as well. A familiar example is the role of those two diseases in aborting the French effort, and nearly aborting the ultimately successful American effort, to construct the Panama Canal.
Bearing all these facts in mind, let's try to regain our sense of perspective about the role of germs in answering Yali's question. There is no doubt that Europeans developed a big advantage in weaponry, technology, and political organization over most of the non-European peoples that they conquered. But that advantage alone doesn't fully explain how initially so few European immigrants came to supplant so much of the native population of the Americas and some other parts of the world. That might not have happened without Europe's sinister gift to other continents—the germs evolving from Eurasians' long intimacy with domestic animals.

CHAPTER

12blueprints and borrowed letters

NINETEENTH-CENTURY AUTHORS TENDED TO INTERPRET history as a progression from savagery to civilization. Key hallmarks of this transition included the development of agriculture, metallurgy, complex technology, centralized government, and writing. Of these, writing was traditionally the one most restricted geographically: until the expansions of Islam and of colonial Europeans, it was absent from Australia, Pacific islands, subequatorial Africa, and the whole New World except for a small part of Mesoamerica. As a result of that confined distribution, peoples who pride themselves on being civilized have always viewed writing as the sharpest distinction raising them above "barbarians" or "savages."

Knowledge brings power. Hence writing brings power to modern societies, by making it possible to transmit knowledge with far greater accuracy and in far greater quantity and detail, from more distant lands and more remote times. Of course, some peoples (notably the Incas) managed to administer empires without writing, and "civilized" peoples don't always defeat "barbarians," as Roman armies facing the Huns learned. But the European conquests of the Americas, Siberia, and Australia illustrate the typical recent outcome.
writing marched together with weapons, microbes, and centralized

2. I 6 •

GUNS,GERMS, AND STEELpolitical organization as a modern agent of conquest. The commands of the monarchs and merchants who organized colonizing fleets were conveyed in writing. The fleets set their courses by maps and written sailing directions prepared by previous expeditions. Written accounts of earlier expeditions motivated later ones, by describing the wealth and fertile lands awaiting the conquerors. The accounts taught subsequent explorers what conditions to expect, and helped them prepare themselves. The resulting empires were administered with the aid of writing. While all those types of information were also transmitted by other means in preliterate societies, writing made the transmission easier, more detailed, more accurate, and more persuasive.
Why, then, did only some peoples and not others develop writing, given its overwhelming value? For example, why did no traditional hunters-gatherers evolve or adopt writing? Among island empires, why did writing arise in Minoan Crete but not in Polynesian Tonga? How many separate times did writing evolve in human history, under what circumstances, and for what uses? Of those peoples who did develop it, why did some do so much earlier than others? For instance, today almost all Japanese and Scandinavians are literate but most Iraqis are not: why did writing nevertheless arise nearly four thousand years earlier in Iraq?
The diffusion of writing from its sites of origin also raises important questions. Why, for instance, did it spread to Ethiopia and Arabia from the Fertile Crescent, but not to the Andes from Mexico? Did writing systems spread by being copied, or did existing systems merely inspire neighboring peoples to invent their own systems? Given a writing system that works well for one language, how do you devise a system for a different language? Similar questions arise whenever one tries to understand the origins and spread of many other aspects of human culture—such as technology, religion, and food production. The historian interested in such questions about writing has the advantage that they can often be answered in unique detail by means of the written record itself. We shall therefore trace writing's development not only because of its inherent importance, but also for the general insights into cultural history that it provides.
Thethree basic strategies underlying writing systems differ in the size of the speech unit denoted by one written sign: either a single bask sound, a whole syllable, or a whole word. Of these, the one employed

BLUEPRINTS

AND BORROWED LETTERS • 2. I 7today by most peoples is the alphabet, which ideally would provide a unique sign (termed a letter) for each basic sound of the language (a phoneme). Actually, most alphabets consist of only about 20 or 30 letters, and most languages have more phonemes than their alphabets have letters. For example, English transcribes about 40 phonemes with a mere 26 letters. Hence most alphabetically written languages, including English, are forced to assign several different phonemes to the same letter and to represent some phonemes by combinations of letters, such as the English two-letter combinations sh and th (each represented by a single letter in the Russian and Greek alphabets, respectively).The second strategy uses so-called logograms, meaning that one written sign stands for a whole word. That's the function of many signs of Chinese writing and of the predominant Japanese writing system (termed kanji). Before the spread of alphabetic writing, systems making much use of logograms were more common and included Egyptian hieroglyphs, Maya glyphs, and Sumerian cuneiform.
The third strategy, least familiar to most readers of this book, uses a sign for each syllable. In practice, most such writing systems (termed syllabaries) provide distinct signs just for syllables of one consonant followed by one vowel (like the syllables of the word "fa-mi-ly"), and resort to various tricks in order to write other types of syllables by means of those signs. Syllabaries were common in ancient times, as exemplified by the Linear B writing of Mycenaean Greece. Some syllabaries persist today, the most important being the kana syllabary that the Japanese use for telegrams, bank statements, and texts for blind readers.
I've intentionally termed these three approaches strategies rather than writing systems. No actual writing system employs one strategy exclusively. Chinese writing is not purely logographic, nor is English writing purely alphabetic. Like all alphabetic writing systems, English uses many logograms, such as numerals, $, %, and + : that is, arbitrary signs, not made up of phonetic elements, representing whole words. "Syllabic" Linear B had many logograms, and "logographic" Egyptian hieroglyphs included many syllabic signs as well as a virtual alphabet of individual letters for each consonant.
inventing a writing system from scratch must have been incomparably more difficult than borrowing and adapting one. The first scribes

Z I 8 •

GUNS,GERMS, AND STEELhad to settle on basic principles that we now take for granted. For example, they had to figure out how to decompose a continuous utterance into speech units, regardless of whether those units were taken as words, syllables, or phonemes. They had to learn to recognize the same sound or speech unit through all our normal variations in speech volume, pitch, speed, emphasis, phrase grouping, and individual idiosyncrasies of pronunciation. They had to decide that a writing system should ignore all of that variation. They then had to devise ways to represent sounds by symbols.
Somehow, the first scribes solved all those problems, without having in front of them any example of the final result to guide their efforts. That task was evidently so difficult that there have been only a few occasions in history when people invented writing entirely on their own. The two indisputably independent inventions of writing were achieved by the Sumerians of Mesopotamia somewhat before 3000 b.c. and by Mexican Indians before 600 b.c. (Figure 12.1); Egyptian writing of 3000 b.c. and Chinese writing (by 1300 b.c.) may also have arisen independently. Probably all other peoples who have developed writing since then have borrowed, adapted, or at least been inspired by existing systems.
The independent invention that we can trace in greatest detail is history's oldest writing system, Sumerian cuneiform (Figure 12.1). For thousands of years before it jelled, people in some farming villages of the Fertile Crescent had been using clay tokens of various simple shapes for accounting purposes, such as recording numbers of sheep and amounts of grain. In the last centuries before 3000 b.c., developments in accounting technology, format, and signs rapidly led to the first system of writing. One such technological innovation was the use of flat clay tablets as a convenient writing surface. Initially, the clay was scratched with pointed tools, which gradually yielded to reed styluses for neatly pressing a mark into the tablet. Developments in format included the gradual adoption of conventions whose necessity is now universally accepted: that writing should be organized into ruled rows or columns (horizontal rows for the Sumerians, as for modern Europeans); that the lines should be read in a constant direction (left to right for Sumerians, as for modern Europeans); and that the lines should be read from top to bottom of the tablet rather than vice versa.
But the crucial change involved the solution of the problem basic to

BLUEPRINTS

AND BORROWED LETTERS • 2, I 9Locations of some scripts mentioned in the text

1. Sumer 9.

West Semitic, Phoenician 5. Proto-Elamite2. Mesoamerica 10. Ethiopian 7. Hittite
?3. China 11. Korea (han'g&l) 8. Indus Valley
??4. Egypt 13. Italy (Roman, Etruscan) 17. Easter Island
14. Greece
15. Ireland (ogham) Syllabaries
6. Crete (Linear A and B) 12. Japan (kana) 16. Cherokee
Figure 12.1. The question marks next to China and Egypt denote somedoubt whether early writing in those areas arose completely independently or was stimulated by writing systems that arose elsewhere earlier."Other" refers to scripts that were neither alphabets nor syllabaries andthat probably arose under the influence of earlier scripts.virtually all writing systems: how to devise agreed-on visible marks that represent actual spoken sounds, rather than only ideas or else words independent of their pronunciation. Early stages in the development of the solution have been detected especially in thousands of clay tablets excavated from the ruins of the former Sumerian city of Uruk, on the Euphrates

220 •

GUNS,GERMS, AND STEELRiver about 200 miles southeast of modern Baghdad. The first Sumerian writing signs were recognizable pictures of the object referred to (for instance, a picture of a fish or a bird). Naturally, those pictorial signs consisted mainly of numerals plus nouns for visible objects; the resulting texts were merely accounting reports in a telegraphic shorthand devoid of grammatical elements. Gradually, the forms of the signs became more abstract, especially when the pointed writing tools were replaced by reed styluses. New signs were created by combining old signs to produce new meanings: for example, the sign for head was combined with the sign for bread in order to produce a sign signifying eat.The earliest Sumerian writing consisted of nonphonetic logograms. That's to say, it was not based on the specific sounds of the Sumerian language, and it could have been pronounced with entirely different sounds to yield the same meaning in any other language—just as the numeral sign 4 is variously pronounced four, chetwtre, neljd, and empat by speakers of English, Russian, Finnish, and Indonesian, respectively. Perhaps the most important single step in the whole history of writing was the Sumerians' introduction of phonetic representation, initially by writing an abstract noun (which could not be readily drawn as a picture) by means of the sign for a depictable noun that had the same phonetic pronunciation. For instance, it's easy to draw a recognizable picture of arrow, hard to draw a recognizable picture of life, but both are pronounced ti in Sumerian, so a picture of an arrow came to mean either arrow or life. The resulting ambiguity was resolved by the addition of a silent sign called a determinative, to indicate the category of nouns to which the intended object belonged. Linguists term this decisive innovation, which also underlies puns today, the rebus principle.Once Sumerians had hit upon this phonetic principle, they began to use it for much more than just writing abstract nouns. They employed it to write syllables or letters constituting grammatical endings. For instance, in English it's not obvious how to draw a picture of the common syllable -tion, but we could instead draw a picture illustrating the verb shun, which has the same pronunciation. Phonetically interpreted signs were also used to "spell out" longer words, as a series of pictures each depicting the sound of one syllable. That's as if an English speaker were to write the word believe as a picture of a bee followed by a picture of a leaf. Phonetic signs also permitted scribes to use the same pictorial sign for a set of related words (such as tooth, speech, and speaker), but to resolve the ambiguity

BLUEPRINTS

AND BORROWED LETTERS • Z Z I

An example of Babylonian cuneiform writing, derived ultimately from

Sumerian cuneiform.with an additional phonetically interpreted sign (such as selecting the sign for two, each, or peak).Thus, Sumerian writing came to consist of a complex mixture of three types of signs: logograms, referring to a whole word or name; phonetic signs, used in effect for spelling syllables, letters, grammatical elements, or

Z Z Z •

GUNS,GERMS, ANDsteelparts of words; and determinatives, which were not pronounced but were used to resolve ambiguities. Nevertheless, the phonetic signs in Sumerian writing fell far short of a complete syllabary or alphabet. Some Sumerian syllables lacked any written signs; the same sign could be pronounced in different ways; and the same sign could variously be read as a word, a syllable, or a letter.
Besides Sumerian cuneiform, the other certain instance of independent origins of writing in human history comes from Native American societies of Mesoamerica, probably southern Mexico. Mesoamerican writing is believed to have arisen independently of Old World writing, because there is no convincing evidence for pre-Norse contact of New World societies with Old World societies possessing writing. In addition, the forms of Mesoamerican writing signs were entirely different from those of any Old World script. About a dozen Mesoamerican scripts are known, all or most of them apparently related to each other (for example, in their numerical and calendrical systems), and most of them still only partially deciphered. At the moment, the earliest preserved Mesoamerican script is from the Zapotec area of southern Mexico around 600 b.c., but by far the best-understood one is of the Lowland Maya region, where the oldest known written date corresponds to a.d. 292.
Despite its independent origins and distinctive sign forms, Maya writing is organized on principles basically similar to those of Sumerian writing and other western Eurasian writing systems that Sumerian inspired. Like Sumerian, Maya writing used both logograms and phonetic signs. Logograms for abstract words were often derived by the rebus principle. That is, an abstract word was written with the sign for another word pronounced similarly but with a different meaning that could be readily depicted. Like the signs of Japan's kana and Mycenaean Greece's Linear B syllabaries, Maya phonetic signs were mostly signs for syllables of one consonant plus one vowel (such as ta, te, ti, to, tu). Like letters of the early Semitic alphabet, Maya syllabic signs were derived from pictures of the object whose pronunciation began with that syllable (for example, the Maya syllabic sign "tie" resembles a tail, for which the Maya word is neh).All of these parallels between Mesoamerican and ancient western Eurasian writing testify to the underlying universality of human creativity. While Sumerian and Mesoamerican languages bear no special relation to each other among the world's languages, both raised similar basic issues in reducing them to writing. The solutions that Sumerians invented before

Z 2. 4 "

GUNS,GERMS, AND STEEL3000 b.c. were reinvented, halfway around the world, by early Mesoamer-ican Indians before 600 b.c.
with the possible exceptions of the Egyptian, Chinese, and Easter Island writing to be considered later, all other writing systems devised anywhere in the world, at any time, appear to have been descendants of systems modified from or at least inspired by Sumerian or early Mesoamerican writing. One reason why there were so few independent origins of writing is the great difficulty of inventing it, as we have already discussed. The other reason is that other opportunities for the independent invention of writing were preempted by Sumerian or early Mesoamerican writing and their derivatives.
We know that the development of Sumerian writing took at least hundreds, possibly thousands, of years. As we shall see, the prerequisites for those developments consisted of several features of human society that determined whether a society would find writing useful, and whether the society could support the necessary specialist scribes. Many other human societies besides those of the Sumerians and early Mexicans—such as those of ancient India, Crete, and Ethiopia—evolved these prerequisites. However, the Sumerians and early Mexicans happened to have been the first to evolve them in the Old World and the New World, respectively. Once the Sumerians and early Mexicans had invented writing, the details or principles of their writing spread rapidly to other societies, before they could go through the necessary centuries or millennia of independent experimentation with writing themselves. Thus, that potential for other, independent experiments was preempted or aborted.
The spread of writing has occurred by either of two contrasting methods, which find parallels throughout the history of technology and ideas. Someone invents something and puts it to use. How do you, another would-be user, then design something similar for your own use, knowing that other people have already got their own model built and working?
Such transmission of inventions assumes a whole spectrum of forms. At the one end lies "blueprint copying," when you copy or modify an available detailed blueprint. At the opposite end lies "idea diffusion," when you receive little more than the basic idea and have to reinvent the details. Knowing that it can be done stimulates you to try to do it yourself, but

1. A woman and child from New Guinea's north coastal lowlands

(Siar Island).

BLUEPRINTS

AND BORROWED LETTERS • 2 2. 5your eventual specific solution may or may not resemble that of the first inventor.
To take a recent example, historians are still debating whether blueprint copying or idea diffusion contributed more to Russia's building of an atomic bomb. Did Russia's bomb-building efforts depend critically on blueprints of the already constructed American bomb, stolen and transmitted to Russia by spies? Or was it merely that the revelation of America's A-bomb at Hiroshima at last convinced Stalin of the feasibility of building such a bomb, and that Russian scientists then reinvented the principles in an independent crash program, with little detailed guidance from the earlier American effort? Similar questions arise for the history of the development of wheels, pyramids, and gunpowder. Let's now examine how blueprint copying and idea diffusion contributed to the spread of writing systems.

Today, professional linguists design writing systems for unwritten languages by the method of blueprint copying. Most such tailor-made systems modify existing alphabets, though some instead design syllabaries. For example, missionary linguists are working on modified Roman alphabets for hundreds of New Guinea and Native American languages. Government linguists devised the modified Roman alphabet adopted in 1928 by Turkey for writing Turkish, as well as the modified Cyrillic alphabets designed for many tribal languages of Russia.

In a few cases, we also know something about the individuals who designed writing systems by blueprint copying in the remote past. For instance, the Cyrillic alphabet itself (the one still used today in Russia) is descended from an adaptation of Greek and Hebrew letters devised by Saint Cyril, a Greek missionary to the Slavs in the ninth century a.d. The first preserved texts for any Germanic language (the language family that includes English) are in the Gothic alphabet created by Bishop Ulfilas, a missionary living with the Visigoths in what is now Bulgaria in the fourth century a.d. Like Saint Cyril's invention, Ulfilas's alphabet was a mishmash of letters borrowed from different sources: about 20 Greek letters, about five Roman letters, and two letters either taken from the runic alphabet or invented by Ulfilas himself. Much more often, we know noth-mg about the individuals responsible for devising famous alphabets of the

2. Z 6 •

GUNS, GERMS,and steelpast. But it's still possible to compare newly emerged alphabets of the past with previously existing ones, and to deduce from letter forms which existing ones served as models. For the same reason, we can be sure that the Linear B syllabary of Mycenaean Greece had been adapted by around 1400 b.c. from the Linear A syllabary of Minoan Crete.
At all of the hundreds of times when an existing writing system of one language has been used as a blueprint to adapt to a different language, some problems have arisen, because no two languages have exactly the same sets of sounds. Some inherited letters or signs may simply be dropped, when the sounds that those letters represent in the lending language do not exist in the borrowing language. For example, Finnish lacks the sounds that many other European languages express by the letters b,c, f, g, w, x, and z, so the Finns dropped these letters from their version of the Roman alphabet. There has also been a frequent reverse problem, of devising letters to represent "new" sounds present in the borrowing language but absent in the lending language. That problem has been solved in several different ways: such as using an arbitrary combination of two or more letters (like the English th to represent a sound for which the Greek and runic alphabets used a single letter); adding a small distinguishing mark to an existing letter (like the Spanish tilde n, the German umlaut 6, and the proliferation of marks dancing around Polish and Turkish letters); co-opting existing letters for which the borrowing language had no use (such as modern Czechs recycling the letter c of the Roman alphabet to express the Czech sound ts); or just inventing a new letter (as our medieval ancestors did when they created the new letters /', u, and w).The Roman alphabet itself was the end product of a long sequence of blueprint copying. Alphabets apparently arose only once in human history: among speakers of Semitic languages, in the area from modern Syria to the Sinai, during the second millennium b.c. All of the hundreds of historical and now existing alphabets were ultimately derived from that ancestral Semitic alphabet, in a few cases (such as the Irish ogham alphabet) by idea diffusion, but in most by actual copying and modification of letter forms.
That evolution of the alphabet can be traced back to Egyptian hieroglyphs, which included a complete set of 24 signs for the 24 Egyptian consonants. The Egyptians never took the logical (to us) next step of discarding all their logograms, determinatives, and signs for pairs and triosof consonants, and using just their consonantal alphabet. Starting around

BLUEPRINTS

AND BORROWED LETTERS • 2. 2 71700 b.c., though, Semites familiar with Egyptian hieroglyphs did begin to experiment with that logical step.
Restricting signs to those for single consonants was only the first of three crucial innovations that distinguished alphabets from other writing systems. The second was to help users memorize the alphabet by placing the letters in a fixed sequence and giving them easy-to-remember names. Our English names are mostly meaningless monosyllables ("a," "bee," "cee " "dee," and so on). But the Semitic names did possess meaning in Semitic languages: they were the words for familiar objects ('aleph = ox, beth = hot>se, gimel = camel, daleth = door, and so on). These Semitic words were related "acrophonically" to the Semitic consonants to which they refer: that is, the first letter of the word for the object was also the letter named for the object ('a, b, g, d, and so on). In addition, the earliest forms of the Semitic letters appear in many cases to have been pictures of those same objects. All these features made the forms, names, and sequence of Semitic alphabet letters easy to remember. Many modern alphabets, including ours, retain with minor modifications that original sequence (and, in the case of Greek, even the letters' original names: alpha, beta, gamma, delta, and so on) over 3,000 years later. One minor modification that readers will already have noticed is that the Semitic and Greek g became the Roman and English c, while the Romans invented a new g in its present position.The third and last innovation leading to modern alphabets was to provide for vowels. Already in the early days of the Semitic alphabet, experiments began with methods for writing vowels by adding small extra letters to indicate selected vowels, or else by dots, lines, or hooks sprinkled over the consonantal letters. In the eighth century b.c. the Greeks became the first people to indicate all vowels systematically by the same types of letters used for consonants. Greeks derived the forms of their vowel letters or – e –17 – i – o by "co-opting" five letters used in the Phoenician alphabet for consonantal sounds lacking in Greek.
From those earliest Semitic alphabets, one line of blueprint copying and evolutionary modification led via early Arabian alphabets to the modern Ethiopian alphabet. A far more important line evolved by way of the Aramaic alphabet, used for official documents of the Persian Empire, into the modern Arabic, Hebrew, Indian, and Southeast Asian alphabets. But the me most familiar to European and American readers is the one that led Vla tne Phoenicians to the Greeks by the early eighth century b.c., thence

Z 2 8 • GUNS, GERMS, AND

steelto the Etruscans in the same century, and in the next century to the Romans, whose alphabet with slight modifications is the one used to print this book. Thanks to their potential advantage of combining precision with simplicity, alphabets have now been adopted in most areas of the modern world.
while blueprint copying and modification are the most straightforward option for transmitting technology, that option is sometimes unavailable. Blueprints may be kept secret, or they may be unreadable to someone not already steeped in the technology. Word may trickle through about an invention made somewhere far away, but the details may not get transmitted. Perhaps only the basic idea is known: someone has succeeded, somehow, in achieving a certain final result. That knowledge may nevertheless inspire others, by idea diffusion, to devise their own routes to such a result.
A striking example from the history of writing is the origin of the syllabary devised in Arkansas around 1820 by a Cherokee Indian named Sequoyah, for writing the Cherokee language. Sequoyah observed that white people made marks on paper, and that they derived great advantage by using those marks to record and repeat lengthy speeches. However, the detailed operations of those marks remained a mystery to him, since (like most Cherokees before 1820) Sequoyah was illiterate and could neither speak nor read English. Because he was a blacksmith, Sequoyah began by ;| devising an accounting system to help him keep track of his customers' debts. He drew a picture of each customer; then he drew circles and lines of various sizes to represent the amount of money owed.
Around 1810, Sequoyah decided to go on to design a system for writing the Cherokee language. He again began by drawing pictures, but gave them up as too complicated and too artistically demanding. He next started to invent separate signs for each word, and again became dissatisfied when he had coined thousands of signs and still needed more.
Finally, Sequoyah realized that words were made up of modest numbers | of different sound bites that recurred in many different words—what wewould call syllables. he initially devised 200 syllabic signs and gradually 1 reduced them to 85, most of them for combinations of one consonant and | one vowel.
As one source of the signs themselves, Sequoyah practiced copying the

BLUEPRINTS

AND BORROWED LETTERS • 2. 2. 9

The set of signs that Sequoyah devised to represent syllables of the Chero-

kee language.letters from an English spelling book given to him by a schoolteacher. About two dozen of his Cherokee syllabic signs were taken directly from those letters, though of course with completely changed meanings, since Sequoyah did not know the English meanings. For example, he chose the shapes D, R, b, h to represent the Cherokee syllables a, e, si, and ni, respectively, while the shape of the numeral 4 was borrowed for the syllable se. He coined other signs by modifying English letters, such as designing thesigns (j , W, and U to represent the syllables yu, sa, and na, respectively.Still other signs were entirely of his creation, such as H, fo, and ^ for ho, li, and nu, respectively. Sequoyah's syllabary is widely admired by professional linguists for its good fit to Cherokee sounds, and for the ease with which it can be learned. Within a short time, the Cherokees achieved almost 100 percent literacy in the syllabary, bought a printing press, had Sequoyah's signs cast as type, and began printing books and newspapers. Cherokee writing remains one of the best-attested examples of a script that arose through idea diffusion. We know that Sequoyah received paper

23o ' GUNS, GERMS,

and steeland other writing materials, the idea of a writing system, the idea of using separate marks, and the forms of several dozen marks. Since, however, he could neither read nor write English, he acquired no details or even principles from the existing scripts around him. Surrounded by alphabets he could not understand, he instead independently reinvented a syllabary, unaware that the Minoans of Crete had already invented another syllabary 3,500 years previously.
Sequoya's example can serve as a model for how idea diffusion probably led to many writing systems of ancient times as well. The han'gul alphabet devised by Korea's King Sejong in a.d. 1446 for the Korean language was evidently inspired by the block format of Chinese characters and by the alphabetic principle of Mongol or Tibetan Buddhist writing. However, King Sejong invented the forms of han'gul letters and several unique features of his alphabet, including the grouping of letters by sylla-bles into square blocks, the use of related letter shapes to represent related* i vowel or consonant sounds, and shapes of consonant letters that depict the position in which the lips or tongue are held to pronounce that consonant. The ogham alphabet used in Ireland and parts of Celtic Britain from i around the fourth century a.d. similarly adopted the alphabetic principle (in this case, from existing European alphabets) but again devised unique letter forms, apparently based on a five-finger system of hand signals.
We can confidently attribute the han'gul and ogham alphabets to idea diffusion rather than to independent invention in isolation, because we know that both societies were in close contact with societies possessing writing and because it is clear which foreign scripts furnished the inspiration. In contrast, we can confidently attribute Sumerian cuneiform and the earliest Mesoamerican writing to independent invention, because at the fj times of their first appearances there existed no other script in their respective hemispheres that could have inspired them. Still debatable are the origins of writing on Easter Island, in China, and in Egypt.
The Polynesians living on Easter Island, in the Pacific Ocean, had unique script of which the earliest preserved examples date back only to,! about a.d. 1851, long after Europeans reached Easter in 1722. Perhaps writing arose independently on Easter before the arrival of Europeant although no examples have survived. But the most straightforward interpretation is to take the facts at face value, and to assume that Easfeip

BLUEPRINTS AND BORROWED

letters • z 3 I

A Korean sign illustrating the remarkable han'gul writing system. Each square block represents a syllable, but each component sign within the

block represents a letter.Islanders were stimulated to devise a script after seeing the written proclamation of annexation that a Spanish expedition handed to them in the year 1770.
As for Chinese writing, first attested around 1300 b.c. but with possible earlier precursors, it too has unique local signs and some unique principles, and most scholars assume that it evolved independently. Writing had developed before 3000 b.c. in Sumer, 4,000 miles west of early Chinese urban centers, and appeared by 2200 b.c. in the Indus Valley, 2,600 miles west, but no early writing systems are known from the whole area between the Indus Valky and China. Thus, there is no evidence that the earliest Chinese scribes could have had knowledge of any other writing system to inspire them.
Egyptian hieroglyphics, the most famous of all ancient writing systems, are also usually assumed to be the product of independent invention, but the alternative interpretation of idea diffusion is more feasible than in the

Z 3 2. • GUNS, GERMS,

and steelcase of Chinese writing. Hieroglyphic writing appeared rather suddenly, in nearly full-blown form, around 3000 b.c. Egypt lay only 800 miles west of Sumer, with which Egypt had trade contacts. I find it suspicious that no evidence of a gradual development of hieroglyphs has come down to us, even though Egypt's dry climate would have been favorable for preserving earlier experiments in writing, and though the similarly dry climate of Sumer has yielded abundant evidence of the development of Sumerian cuneiform for at least several centuries before 3000 b.c. Equally suspicious is the appearance of several other, apparently independently designed, writing systems in Iran, Crete, and Turkey (so-called proto-Elamite writing, Cretan pictographs, and Hieroglyphic Hittite, respectively), after the rise of Sumerian and Egyptian writing. Although each of those systems used distinctive sets of signs not borrowed from Egypt or Sumer, the peoples involved could hardly have been unaware of the writing of their neighboring trade partners.

An example of Chinese writing: a handscroll by

W« Li, from a.d. 1679.It would be a remarkable coincidence if, after millions of years of human existence without writing, all those Mediterranean and Near Eastern societies had just happened to hit independently on the idea of writing within a few centuries of each other. Hence a possible interpretation seems to me idea diffusion, as in the case of Sequoyah's syllabary. That is, Egyptians and other peoples may have learned from Sumerians about the idea

BLUEPRINTS

AND BORROWED LETTERS • 2. 3 3

An example of Egyptian hieroglyphs: the funerary papyrus of Princess

Entiu-ny.of writing and possibly about some of the principles, and then devised other principles and all the specific forms of the letters for themselves.
let us now return to the main question with which we began this chapter: why did writing arise in and spread to some societies, but not to many others? Convenient starting points for our discussion are the limited capabilities, uses, and users of early writing systems.
Early scripts were incomplete, ambiguous, or complex, or all three. For

Z 3 4 •

GUNS,GERMS, AND STEELexample, the oldest Sumerian cuneiform writing could not render normal prose but was a mere telegraphic shorthand, whose vocabulary was restricted to names, numerals, units of measure, words for objects counted, and a few adjectives. That's as if a modern American court clerk were forced to write "John 27 fat sheep," because English writing lacked the necessary words and grammar to write "We order John to deliver the 27 fat sheep that he owes to the government." Later Sumerian cuneiform did become capable of rendering prose, but it did so by the messy system that I've already described, with mixtures of logograms, phonetic signs, and unpronounced determinatives totaling hundreds of separate signs. Linear B, the writing of Mycenaean Greece, was at least simpler, being based on a syllabary of about 90 signs plus logograms. Offsetting that virtue, Linear B was quite ambiguous. It omitted any consonant at the end of a word, and it used the same sign for several related consonants (for instance, one sign for both / and r, another for p and b and pb, and still another for g and k and kh). We know how confusing we find it when native-born Japanese people speak English without distinguishing / and r: imagine the confusion if our alphabet did the same while similarly homogenizing the other consonants that I mentioned! It's as if we were to spell the words "rap," "lap," "lab," and "laugh" identically.A related limitation is that few people ever learned to write these early scripts. Knowledge of writing was confined to professional scribes in the employ of the king or temple. For instance, there is no hint that Linear B was used or understood by any Mycenaean Greek beyond small cadres of palace bureaucrats. Since individual Linear B scribes can be distinguished by their handwriting on preserved documents, we can say that all preserved Linear B documents from the palaces of Knossos and Pylos are the work of a mere 75 and 40 scribes, respectively.
The uses of these telegraphic, clumsy, ambiguous early scripts were as restricted as the number of their users. Anyone hoping to discover how Sumerians of 3000 b.c. thought and felt is in for a disappointment. Instead, the first Sumerian texts are emotionless accounts of palace and temple bureaucrats. About 90 percent of the tablets in the earliest known Sumerian archives, from the city of Uruk, are clerical records of goods paid in, workers given rations, and agricultural products distributed. Only later, as Sumerians progressed beyond logograms to phonetic writing, did they begin to write prose narratives, such as propaganda and myths.
Mycenaean Greeks never even reached that propaganda-and-myths

BLUEPRINTS AND BORROWED LETTERS • Z 3 5

stage. One-third of all Linear B tablets from the palace of Knossos are accountants' records of sheep and wool, while an inordinate proportion of writing at the palace of Pylos consists of records of flax. Linear B was inherently so ambiguous that it remained restricted to palace accounts, whose context and limited word choices made the interpretation clear. Not a trace of its use for literature has survived. The Iliad and Odyssey were composed and transmitted by nonliterate bards for nonliterate listeners, and not committed to writing until the development of the Greek alphabet hundreds of years later.Similarly restricted uses characterize early Egyptian, Mesoamerican, and Chinese writing. Early Egyptian hieroglyphs recorded religious and state propaganda and bureaucratic accounts. Preserved Maya writing was similarly devoted to propaganda, births and accessions and victories of kings, and astronomical observations of priests. The oldest preserved Chinese writing of the late Shang Dynasty consists of religious divination about dynastic affairs, incised into so-called oracle bones. A sample Shang text: "The king, reading the meaning of the crack note 9, said: 'If the child is born on a keng day, it will be extremely auspicious.' "
To us today, it is tempting to ask why societies with early writing systems accepted the ambiguities that restricted writing to a few functions and a few scribes. But even to pose that question is to illustrate the gap between ancient perspectives and our own expectations of mass literacy. The intended restricted uses of early writing provided a positive disincentive for devising less ambiguous writing systems. The kings and priests of ancient Sumer wanted writing to be used by professional scribes to record numbers of sheep owed in taxes, not by the masses to write poetry and hatch plots. As the anthropologist Claude Levi-Strauss put it, ancient writing's main function was "to facilitate the enslavement of other human beings." Personal uses of writing by nonprofessionals came only much later, as writing systems grew simpler and more expressive.
For instance, with the fall of Mycenaean Greek civilization, around 1200 b.c., Linear B disappeared, and Greece returned to an age of preliter-acy. When writing finally returned to Greece, in the eighth century B.C., the new Greek writing, its users, and its uses were very different. The writ-mg was no longer an ambiguous syllabary mixed with logograms but an alphabet borrowed from the Phoenician consonantal alphabet and improved by the Greek invention of vowels. In place of lists of sheep, legi-

2 3 6 '

GUNS,GERMS, AND STEELble only to scribes and read only in palaces, Greek alphabetic writing from the moment of its appearance was a vehicle of poetry and humor, to be read in private homes. For instance, the first preserved example of Greek alphabetic writing, scratched onto an Athenian wine jug of about 740 b.c., is a line of poetry announcing a dancing contest: "Whoever of all dancers performs most nimbly will win this vase as a prize." The next example is three lines of dactylic hexameter scratched onto a drinking cup: "I am Nestor's delicious drinking cup. Whoever drinks from this cup swiftly will the desire of fair-crowned Aphrodite seize him." The earliest preserved examples of the Etruscan and Roman alphabets are also inscriptions on drinking cups and wine containers. Only later did the alphabet's easily learned vehicle of private communication become co-opted for public or bureaucratic purposes. Thus, the developmental sequence of uses for alphabetic writing was the reverse of that for the earlier systems of logograms and syllabaries.
The limited uses and users of early writing suggest why writing appeared so late in human evolution. All of the likely or possible independent inventions of writing (in Sumer, Mexico, China, and Egypt), and all of the early adaptations of those invented systems (for example, those in Crete, Iran, Turkey, the Indus Valley, and the Maya area), involved socially stratified societies with complex and centralized political institutions, whose necessary relation to food production we shall explore in a later chapter. Early writing served the needs of those political institutions (such as record keeping and royal propaganda), and the users were full-time bureaucrats nourished by stored food surpluses grown by food-producing peasants. Writing was never developed or even adopted by hunter-gatherer societies, because they lacked both the institutional uses of early writing and the social and agricultural mechanisms for generating the food surpluses required to feed scribes.
Thus, food production and thousands of years of societal evolution following its adoption were as essential for the evolution of writing as for the evolution of microbes causing human epidemic diseases. Writing arose independently only in the Fertile Crescent, Mexico, and probably China precisely because those were the first areas where food production emerged in their respective hemispheres. Once writing had been invented by those

BLUEPRINTS

AND BORROWED LETTERS • 2. 3 7few societies, it then spread, by trade and conquest and religion, to other societies with similar economies and political organizations.
While food production was thus a necessary condition for the evolution or early adoption of writing, it was not a sufficient condition. At the beginning of this chapter, I mentioned the failure of some food-producing societies with complex political organization to develop or adopt writing before modern times. Those cases, initially so puzzling to us moderns accustomed to viewing writing as indispensable to a complex society, included one of the world's largest empires as of a.d. 1520, the Inca Empire of South America. They also included Tonga's maritime proto-empire, the Hawaiian state emerging in the late 18th century, all of the states and chiefdoms of subequatorial Africa and sub-Saharan West Africa before the arrival of Islam, and the largest native North American societies, those of the Mississippi Valley and its tributaries. Why did all those societies fail to acquire writing, despite their sharing prerequisites with societies that did do so?
Here we have to remind ourselves that the vast majority of societies with writing acquired it by borrowing it from neighbors or by being inspired by them to develop it, rather than by independently inventing it themselves. The societies without writing that I just mentioned are ones that got a later start on food production than did Sumer, Mexico, and China. (The only uncertainty in this statement concerns the relative dates for the onset of food production in Mexico and in the Andes, the eventual Inca realm.) Given enough time, the societies lacking writing might also have eventually developed it on their own. Had they been located nearer to Sumer, Mexico, and China, they might instead have acquired writing or the idea of writing from those centers, just as did India, the Maya, and most other societies with writing. But they were too far from the first centers of writing to have acquired it before modern times.
The importance of isolation is most obvious for Hawaii and Tonga, both of which were separated by at least 4,000 miles of ocean from the nearest societies with writing. The other societies illustrate the important point that distance as the crow flies is not an appropriate measure of isolation for humans. The Andes, West Africa's kingdoms, and the mouth of the Mississippi River lay only about 1,200, 1,500, and 700 miles, respectively, from societies with writing in Mexico, North Africa, and Mexico, respectively. These distances are considerably less than the distances the

2. 3 8 • GUNS, GERMS, AND STEEL

alphabet had to travel from its homeland on the eastern shores of the Mediterranean to reach Ireland, Ethiopia, and Southeast Asia within 2,000 years of its invention. But humans are slowed by ecological and water barriers that crows can fly over. The states of North Africa (with writing) and West Africa (without writing) were separated from each other by Saharan desert unsuitable for agriculture and cities. The deserts of northern Mexico similarly separated the urban centers of southern Mexico from the chiefdoms of the Mississippi Valley. Communication between southern Mexico and the Andes required either a sea voyage or else a long chain of overland contacts via the narrow, forested, never urbanized Isthmus of Darien. Hence the Andes, West Africa, and the Mississippi Valley were effectively rather isolated from societies with writing.
That's not to say that those societies without writing were totally isolated. West Africa eventually did receive Fertile Crescent domestic animals across the Sahara, and later accepted Islamic influence, including Arabic writing. Corn diffused from Mexico to the Andes and, more slowly, from Mexico to the Mississippi Valley. But we already saw in Chapter 10 that the north-south axes and ecological barriers within Africa and the Ameri-cas retarded the diffusion of crops and domestic animals. The history of writing illustrates strikingly the similar ways in which geography and ecology influenced the spread of human inventions.

CHAPTER 13

necessity's mother
ON JULY 3, 1908, ARCHAEOLOGISTS EXCAVATING THE ancient Minoan palace at Phaistos, on the island of Crete, chanced upon one of the most remarkable objects in the history of technology. At first glance it seemed unprepossessing: just a small, flat, unpainted, circular disk of hard-baked clay, 6l/2 inches in diameter. Closer examination showed each side to be covered with writing, resting on a curved line that spiraled clockwise in five coils from the disk's rim to its center. A total of 241 signs or letters was neatly divided by etched vertical lines into groups of several signs, possibly constituting words. The writer must have planned and executed the disk with care, so as to start writing at the rim and fill up all the available space along the spiraling line, yet not run out of space on reaching the center (page 240).
Ever since it was unearthed, the disk has posed a mystery for historians of writing. The number of distinct signs (45) suggests a syllabary rather than an alphabet, but it is still undeciphered, and the forms of the signs are unlike those of any other known writing system. Not another scrap of the strange script has turned up in the 89 years since its discovery. Thus, it remains unknown whether it represents an indigenous Cretan script or a foreign import to Crete.
For historians of technology, the Phaistos disk is even more baffling; its

Z 4 O •

GUNS,GERMS, AND STEEL

One side of the two-sided Phaistos Disk.

estimated date of 1700 b.c. makes it by far the earliest printed document in the world. Instead of being etched by hand, as were all texts of Crete's later Linear A and Linear B scripts, the disk's signs were punched into soft clay (subsequently baked hard) by stamps that bore a sign as raised type. The printer evidently had a set of at least 4.5 stamps, one for each sign appearing on the disk. Making these stamps must have entailed a great deal of work, and they surely weren't manufactured just to print this single document. Whoever used them was presumably doing a lot of writing. With those stamps, their owner could make copies much more quickly and neatly than if he or she had written out each of the script's complicated signs at each appearance.
The Phaistos disk anticipates humanity's next efforts at printing, which similarly used cut type or blocks but applied them to paper with ink, not

NECESSITY'S

MOTHER • 24 Ito clay without ink. However, those next efforts did not appear until 2,500 vears later in China and 3,100 years later in medieval Europe. Why was the disk's precocious technology not widely adopted in Crete or elsewhere in the ancient Mediterranean? Why was its printing method invented around 1700 b.c. in Crete and not at some other time in Mesopotamia, Mexico, or any other ancient center of writing? Why did it then take thousands of years to add the ideas of ink and a press and arrive at a printing press? The disk thus constitutes a threatening challenge to historians. If inventions are as idiosyncratic and unpredictable as the disk seems to suggest, rHen efforts to generalize about the history of technology may be doomed from the outset.
Technology, in the form of weapons and transport, provides the direct means by which certain peoples have expanded their realms and conquered other peoples. That makes it the leading cause of history's broadest pattern. But why were Eurasians, rather than Native Americans or sub-Saharan Africans, the ones to invent firearms, oceangoing ships, and steel equipment? The differences extend to most other significant technological advances, from printing presses to glass and steam engines. Why were all those inventions Eurasian? Why were all New Guineans and Native Australians in a.d. 1800 still using stone tools like ones discarded thousands of years ago in Eurasia and most of Africa, even though some of the world's richest copper and iron deposits are in New Guinea and Australia, respectively? All those facts explain why so many laypeople assume that Eurasians are superior to other peoples in inventiveness and intelligence.
If, on the other hand, no such difference in human neurobiology exists to account for continental differences in technological development, what does account for them? An alternative view rests on the heroic theory of invention. Technological advances seem to come disproportionately from a few very rare geniuses, such as Johannes Gutenberg, James Watt, Thomas Edison, and the Wright brothers. They were Europeans, or descendants of European emigrants to America. So were Archimedes and other rare geniuses of ancient times. Could such geniuses have equally well been born in Tasmania or Namibia? Does the history of technology depend on nothing more than accidents of the birthplaces of a few inventors?
Still another alternative view holds that it is a matter not of individual inventiveness but of the receptivity of whole societies to innovation. Some societies seem hopelessly conservative, inward looking, and hostile to

2 4 2 '

GUNS,GERMS,and steelchange. That's the impression of many Westerners who have attempted to help Third World peoples and ended up discouraged. The people seem perfectly intelligent as individuals; the problem seems instead to lie with their societies. How else can one explain why the Aborigines of northeastern Australia failed to adopt bows and arrows, which they saw being used by Torres Straits islanders with whom they traded? Might all the societies of an entire continent be unreceptive, thereby explaining technology's slow pace of development there? In this chapter we shall finally come to grips with a central problem of this book: the question of why technology did evolve at such different rates on different continents.
The starting point for our discussion is the common view expressed in the saying "Necessity is the mother of invention." That is, inventions supposedly arise when a society has an unfulfilled need: some technology is widely recognized to be unsatisfactory or limiting. Would-be inventors, motivated by the prospect of money or fame, perceive the need and try to meet it. Some inventor finally comes up with a solution superior to the existing, unsatisfactory technology. Society adopts the solution if it is compatible with the society's values and other technologies.
Quite a few inventions do conform to this commonsense view of necessity as invention's mother. In 1942, in the middle of World War II, the U.S. government set up the Manhattan Project with the explicit goal of inventing the technology required to build an atomic bomb before Nazi Germany could do so. That project succeeded in three years, at a cost of $2 billion (equivalent to over $20 billion today). Other instances are Eli Whitney's 1794 invention of his cotton gin to replace laborious hand cleaning of cotton grown in the U.S. South, and James Watt's 1769 invention of his steam engine to solve the problem of pumping water out of British coal mines.
These familiar examples deceive us into assuming that other major inventions were also responses to perceived needs. In fact, many or most inventions were developed by people driven by curiosity or by a love of tinkering, in the absence of any initial demand for the product they had in mind. Once a device had been invented, the inventor then had to find an application for it. Only after it had been in use for a considerable time did consumers come to feel that they "needed" it. Still other devices, invented to serve one purpose, eventually found most of their use for other, unantic-

NECESSITY'S MOTHER • 2-43

ipated purposes. It may come as a surprise to learn that these inventions in search of a use include most of the major technological breakthroughs of modern times, ranging from the airplane and automobile, through the internal combustion engine and electric light bulb, to the phonograph and transistor. Thus, invention is often the mother of necessity, rather than vice versa.
A good example is the history of Thomas Edison's phonograph, the most original invention of the greatest inventor of modern times. When Edison built his first phonograph in 1877, he published an article proposing ten uses to which his invention might be put. They included preserving the last words of dying people, recording books for blind people to hear, announcing clock time, and teaching spelling. Reproduction of music was not high on Edison's list of priorities. A few years later Edison told his assistant that his invention had no commercial value. Within another few years he changed his mind and did enter business to sell phonographs— but for use as office dictating machines. When other entrepreneurs created jukeboxes by arranging for a phonograph to play popular music at the drop of a coin, Edison objected to this debasement, which apparently detracted from serious office use of his invention. Only after about 20 years did Edison reluctantly concede that the main use of his phonograph was to record and play music.
The motor vehicle is another invention whose uses seem obvious today. However, it was not invented in response to any demand. When Nikolaus Otto built his first gas engine, in 1866, horses had been supplying people's land transportation needs for nearly 6,000 years, supplemented increasingly by steam-powered railroads for several decades. There was no crisis in the availability of horses, no dissatisfaction with railroads.
Because Otto's engine was weak, heavy, and seven feet tall, it did not recommend itself over horses. Not until 1885 did engines improve to the point that Gottfried Daimler got around to installing one on a bicycle to create the first motorcycle; he waited until 1896 to build the first truck.
In 1905, motor vehicles were still expensive, unreliable toys for the rich. Public contentment with horses and railroads remained high until World War I, when the military concluded that it really did need trucks. Intensive postwar lobbying by truck manufacturers and armies finally convinced the public of its own needs and enabled trucks to begin to supplant horse-drawn wagons in industrialized countries. Even in the largest American cities, the changeover took 50 years.

2. 4 4 "

GUNS,GERMS, AND STEELInventors often have to persist at their tinkering for a long time in the absence of public demand, because early models perform too poorly to be useful. The first cameras, typewriters, and television sets were as awful as Otto's seven-foot-tall gas engine. That makes it difficult for an inventor to foresee whether his or her awful prototype might eventually find a use and thus warrant more time and expense to develop it. Each year, the United States issues about 70,000 patents, only a few of which ultimately reach the stage of commercial production. For each great invention that ultimately found a use, there are countless others that did not. Even inventions that meet the need for which they were initially designed may later prove more valuable at meeting unforeseen needs. While James Watt designed his steam engine to pump water from mines, it soon was supplying power to cotton mills, then (with much greater profit) propelling locomotives and boats.
thus, thecommonsense view of invention that served as our starting point reverses the usual roles of invention and need. It also overstates the importance of rare geniuses, such as Watt and Edison. That "heroic theory of invention," as it is termed, is encouraged by patent law, because an applicant for a patent must prove the novelty of the invention submitted. Inventors thereby have a financial incentive to denigrate or ignore previous work. From a patent lawyer's perspective, the ideal invention is one that arises without any precursors, like Athene springing fully formed from the forehead of Zeus.
In reality, even for the most famous and apparently decisive modern inventions, neglected precursors lurked behind the bald claim "X invented Y." For instance, we are regularly told, "James Watt invented the steam engine in 1769," supposedly inspired by watching steam rise from a teakettle's spout. Unfortunately for this splendid fiction, Watt actually got the idea for his particular steam engine while repairing a model of Thomas Newcomen's steam engine, which Newcomen had invented 57 years earlier and of which over a hundred had been manufactured in England by the time of Watt's repair work. Newcomen's engine, in turn, followed the steam engine that the Englishman Thomas Savery patented in 1698, which followed the steam engine that the Frenchman Denis Papin designed (but did not build) around 1680, which in turn had precursors in the ideas of

NECESSITY'S MOTHER • 2-45

the Dutch scientist Christiaan Huygens and others. All this is not to deny that Watt greatly improved Newcomen's engine (by incorporating a separate steam condenser and a double-acting cylinder), just as Newcomen had greatly improved Savery's.
Similar histories can be related for all modern inventions that are adequately documented. The hero customarily credited with the invention followed previous inventors who had had similar aims and had already produced designs, working models, or (as in the case of the Newcomen steam engine) commercially successful models. Prison's famous "invention" of the incandescent light bulb on the night of October 21, 1879, improved on many other incandescent light bulbs patented by other inventors between 1841 and 1878. Similarly, the Wright brothers' manned powered airplane was preceded by the manned unpowered gliders of Otto Lilienthal and the unmanned powered airplane of Samuel Langley; Samuel Morse's telegraph was preceded by those of Joseph Henry, William Cooke, and Charles Wheatstone; and Eli Whitney's gin for cleaning short-staple (inland) cotton extended gins that had been cleaning long-staple (Sea Island) cotton for thousands of years.
All this is not to deny that Watt, Edison, the Wright brothers, Morse, and Whitney made big improvements and thereby increased or inaugurated commercial success. The form of the invention eventually adopted might have been somewhat different without the recognized inventor's contribution. But the question for our purposes is whether the broad pattern of world history would have been altered significantly if some genius inventor had not been born at a particular place and time. The answer is clear: there has never been any such person. All recognized famous inventors had capable predecessors and successors and made their improvements at a time when society was capable of using their product. As we shall see, the tragedy of the hero who perfected the stamps used for the Phaistos disk was that he or she devised something that the society of the time could not exploit on a large scale.
MY examples so far have been drawn from modern technologies, because their histories are well known. My two main conclusions are that technology develops cumulatively, rather than in isolated heroic acts, and that it finds most of its uses after it has been invented, rather than being

2. 4 6 •

GUNS,GERMS, AND STEELinvented to meet a foreseen need. These conclusions surely apply with much greater force to the undocumented history of ancient technology. When Ice Age hunter-gatherers noticed burned sand and limestone residues in their hearths, it was impossible for them to foresee the long, serendipitous accumulation of discoveries that would lead to the first Roman glass windows (around a.d. 1), by way of the first objects with surface glazes (around 4000 b.c.), the first free-standing glass objects of Egypt and Mesopotamia (around 2500 b.c.), and the first glass vessels (around 1500
B.C.).
We know nothing about how those earliest known surface glazes themselves were developed. Nevertheless, we can infer the methods of prehistoric invention by watching technologically "primitive" people today, such as the New Guineans with whom I work. I already mentioned their knowledge of hundreds of local plant and animal species and each species: edibility, medical value, and other uses. New Guineans told me similarly about dozens of rock types in their environment and each type's hardness, color, behavior when struck or flaked, and uses. All of that knowledge is acquired by observation and by trial and error. I see that process of "invention" going on whenever I take New Guineans to work with me in an area away from their homes. They constantly pick up unfamiliar things in the forest, tinker with them, and occasionally find them useful enough to bring home. I see the same process when I am abandoning a campsite, and local people come to scavenge what is left. They play with my discarded objects and try to figure out whether they might be useful in New Guinea society. Discarded tin cans are easy: they end up reused as containers. Other objects are tested for purposes very different from the one for which they were manufactured. How would that yellow number 2 pencil look as an ornament, inserted through a pierced ear-lobe or nasal septum? Is that piece of broken glass sufficiently sharp and strong to be useful as a knife? Eureka!
The raw substances available to ancient peoples were natural materials such as stone, wood, bone, skins, fiber, clay, sand, limestone, and minerals, all existing in great variety. From those materials, people gradually learned to work particular types of stone, wood, and bone into tools; to convert particular clays into pottery and bricks; to convert certain mixtures of sand, limestone, and other "dirt" into glass; and to work available pure soft metals such as copper and gold, then to extract metals from ores, and finally to work hard metals such as bronze and iron.

NECESSITY'S

MOTHER • Z 4 JA good illustration of the histories of trial and error involved is furnished by the development of gunpowder and gasoline from raw materials. Combustible natural products inevitably make themselves noticed, as when a resinous log explodes in a campfire. By 2000 b.c., Mesopotamians were extracting tons of petroleum by heating rock asphalt. Ancient Greeks discovered the uses of various mixtures of petroleum, pitch, resins, sulfur, and quicklime as incendiary weapons, delivered by catapults, arrows, firebombs, and ships. The expertise at distillation that medieval Islamic alchemists developed to produce alcohols and perfumes also let them dis-till petroleum into fractions, some of which proved to be even more powerful incendiaries. Delivered in grenades, rockets, and torpedoes, those incendiaries played a key role in Islam's eventual defeat of the Crusaders. By then, the Chinese had observed that a particular mixture of sulfur, charcoal, and saltpeter, which became known as gunpowder, was especially explosive. An Islamic chemical treatise of about a.d. 1100 describes seven gunpowder recipes, while a treatise from a.d. 1280 gives more than 70 recipes that had proved suitable for diverse purposes (one for rockets, another for cannons).
As for postmedieval petroleum distillation, 19th-century chemists found the middle distillate fraction useful as fuel for oil lamps. The chemists discarded the most volatile fraction (gasoline) as an unfortunate waste product—until it was found to be an ideal fuel for internal-combustion engines. Who today remembers that gasoline, the fuel of modern civilization, originated as yet another invention in search of a use?
Once an inventor has discovered a use for a new technology, the next step is to persuade society to adopt it. Merely having a bigger, faster, more powerful device for doing something is no guarantee of ready acceptance. Innumerable such technologies were either not adopted at all or adopted only after prolonged resistance. Notorious examples include the U.S. Congress's rejection of funds to develop a supersonic transport in 1971, the world's continued rejection of an efficiently designed typewriter keyboard, and Britain's long reluctance to adopt electric lighting. What is it that promotes aa invention's acceptance by a society?
Let's begin by comparing the acceptability of different inventions within the same society. It turns out that at least four factors influence acceptance.
The first and most obvious factor is relative economic advantage com-

Z 4 8 • GUNS, GERMS, AND STEEL

pared with existing technology. While wheels are very useful in modern industrial societies, that has not been so in some other societies. Ancient Native Mexicans invented wheeled vehicles with axles for use as toys, but not for transport. That seems incredible to us, until we reflect that ancient Mexicans lacked domestic animals to hitch to their wheeled vehicles, which therefore offered no advantage over human porters.
A second consideration is social value and prestige, which can override economic benefit (or lack thereof). Millions of people today buy designer jeans for double the price of equally durable generic jeans—because the social cachet of the designer label counts for more than the extra cost. Similarly, Japan continues to use its horrendously cumbersome kanji writing system in preference to efficient alphabets or Japan's own efficient kana syllabary—because the prestige attached to kanji is so great.
Still another factor is compatibility with vested interests. This book, like probably every other typed document you have ever read, was typed with a QWERTY keyboard, named for the left-most six letters in its upper row. Unbelievable as it may now sound, that keyboard layout was designed in 1873 as a feat of anti-engineering. It employs a whole series of perverse tricks designed to force typists to type as slowly as possible, such as scattering the commonest letters over all keyboard rows and concentrating them on the left side (where right-handed people have to use their weaker hand). The reason behind all of those seemingly counterproductive features is that the typewriters of 1873 jammed if adjacent keys were struck in quick succession, so that manufacturers had to slow down typists. When improvements in typewriters eliminated the problem of jamming, trials in 1932 with an efficiently laid-out keyboard showed that it would let us double our typing speed and reduce our typing effort by 95 percent. But QWERTY keyboards were solidly entrenched by then. The vested interests of hundreds of millions of QWERTY typists, typing teachers, typewriter and computer salespeople, and manufacturers have crushed all moves toward keyboard efficiency for over 60-years.
While the story of the QWERTY keyboard may sound funny, many similar cases have involved much heavier economic consequences. Why does Japan now dominate the world market for transistorized electronic consumer products, to a degree that damages the United States's balance of payments with Japan, even though transistors were invented and patented in the United States? Because Sony bought transistor licensing rights from Western Electric at a time when the American electronics consumer

NECESSITY'SMOTHER • 249

industry was churning out vacuum tube models and reluctant to compete with its own products. Why were British cities still using gas street lighting into the 1920s, long after U.S. and German cities had converted to electric street lighting? Because British municipal governments had invested heavily in gas lighting and placed regulatory obstacles in the way of the competing electric light companies.
The remaining consideration affecting acceptance of new technologies is the ease with which their advantages can be observed. In a.d. 1340, when firearms had not yet reached most of Europe, England's earl of Derby and earl of Salisbury happened to be present in Spain at the battle of Tarifa, where Arabs used cannons against the Spaniards. Impressed by what they saw, the earls introduced cannons to the English army, which adopted them enthusiastically and already used them against French soldiers at the battle of Crecy six years later.
thus, wheels, designer jeans, and QWERTY keyboards illustrate the varied reasons why the same society is not equally receptive to all inventions. Conversely, the same invention's reception also varies greatly among contemporary societies. We are all familiar with the supposed generalization that rural Third World societies are less receptive to innovation than are Westernized industrial societies. Even within the industrialized world, some areas are much more receptive than others. Such differences, if they existed on a continental scale, might explain why technology developed faster on some continents than on others. For instance, if all Aboriginal Australian societies were for some reason uniformly resistant to change, that might account for their continued use of stone tools after metal tools had appeared on every other continent. How do differences in receptivity among societies arise?
A laundry list of at least 14 explanatory factors has been proposed by historians of technology. One is long life expectancy, which in principle should give prospective inventors the years necessary to accumulate technical knowledge, as well as the patience and security to embark on long development programs yielding delayed rewards. Hence the greatly increased life expectancy brought by modern medicine may have contributed to the recently accelerating pace of invention.
The next five factors involve economics or the organization of society: (1) The availability of cheap slave labor in classical times supposedly dis-

15o ' GUNS, GERMS, AND STEEL

couraged innovation then, whereas high wages or labor scarcity now stimulate the search for technological solutions. For example, the prospect of changed immigration policies that would cut off the supply of cheap Mexican seasonal labor to Californian farms was the immediate incentive for the development of a machine-harvestable variety of tomatoes in California. (2) Patents and other property laws, protecting ownership rights of inventors, reward innovation in the modern West, while the lack of such protection discourages it in modern China. (3) Modern industrial societies provide extensive opportunities for technical training, as medieval Islam did and modern Zaire does not. (4) Modern capitalism is, and the ancient Roman economy was not, organized in a way that made it potentially rewarding to invest capital in technological development. (5) The strong individualism of U.S. society allows successful inventors to keep earnings for themselves, whereas strong family ties in New Guinea ensure that someone who begins to earn money will be joined by a dozen relatives expecting to move in and be fed and supported.
Another four suggested explanations are ideological, rather than economic or organizational: (1) Risk-taking behavior, essential for efforts at innovation, is more widespread in some societies than in others. (2) The scientific outlook is a unique feature of post-Renaissance European society that has contributed heavily to its modern technological preeminence. (3) Tolerance of diverse views and of heretics fosters innovation, whereas a strongly traditional outlook (as in China's emphasis on ancient Chinese classics) stifles it. (4) Religions vary greatly in their relation to technological innovation: some branches of Judaism and Christianity are claimed to be especially compatible with it, while some branches of Islam, Hinduism, and Brahmanism may be especially incompatible with it.
All ten of these hypotheses are plausible. But none of them has any necessary association with geography. If patent rights, capitalism, and certain religions do promote technology, what selected for those factors in postmedieval Europe but not in contemporary China or India?
At least the direction in which those ten factors influence technology seems clear. The remaining four proposed factors—war, centralized government, climate, and resource abundance—appear to act inconsistently: sometimes they stimulate technology, sometimes they inhibit it. (1) Throughout history, war has often been a leading stimulant of technological innovation. For instance, the enormous investments made in nuclear weapons during World War II and in airplanes and trucks during World

NECESSITY'S MOTHER • Z 5 I

War I launched whole new fields of technology. But wars can also deal devastating setbacks to technological development. (2) Strong centralized government boosted technology in late-19th-century Germany and Japan, and crushed it in China after a.d. 1500. (3) Many northern Europeans assume that technology thrives in a rigorous climate where survival is impossible without technology, and withers in a benign climate where clothing is unnecessary and bananas supposedly fall off the trees. An opposite view is that benign environments leave people free from the constant struggle for existence, free to devote themselves to innovation. (4) There has also been debate over whether technology is stimulated by abundance or by scarcity of environmental resources. Abundant resources might stimulate the development of inventions utilizing those resources, such as water mill technology in rainy northern Europe, with its many rivers—but why didn't water mill technology progress more rapidly in even rainier New Guinea? The destruction of Britain's forests has been suggested as the reason behind its early lead in developing coal technology, but why didn't deforestation have the same effect in China?
This discussion does not exhaust the list of reasons proposed to explain why societies differ in their receptivity to new technology. Worse yet, all of these proximate explanations bypass the question of the ultimate factors behind them. This may seem like a discouraging setback in our attempt to understand the course of history, since technology has undoubtedly been one of history's strongest forces. However, I shall now argue that the diversity of independent factors behind technological innovation actually makes it easier, not harder, to understand history's broad pattern.
for the purposes of this book, the key question about the laundry list is whether such factors differed systematically from continent to continent and thereby led to continental differences in technological development. Most laypeople and many historians assume, expressly or tacitly, that the answer is yes. For example, it is widely believed that Australian Aborigines as a group shared ideological characteristics contributing to their technological backwardness: they were (or are) supposedly conservative, living in an imagined past Dreamtime of the world's creation, and not focused on practical ways to improve the present. A leading historian of Africa characterized Africans as inward looking and lacking Europeans' drive for expansion.

2. 5 2. •

GUNS,GERMS, AND STEELBut all such claims are based on pure speculation. There has never been a study of many societies under similar socioeconomic conditions on each of two continents, demonstrating systematic ideological differences between the two continents' peoples. The usual reasoning is instead circular: because technological differences exist, the existence of corresponding ideological differences is inferred.
In reality, I regularly observe in New Guinea that native societies there differ greatly from each other in their prevalent outlooks. Just like industrialized Europe and America, traditional New Guinea has conservative societies that resist new ways, living side by side with innovative societies that selectively adopt new ways. The result, with the arrival of Western technology, is that the more entrepreneurial societies are now exploiting Western technology to overwhelm their conservative neighbors.
For example, when Europeans first reached the highlands of eastern New Guinea, in the 1930s, they "discovered" dozens of previously uncon-tacted Stone Age tribes, of which the Chimbu tribe proved especially aggressive in adopting Western technology. When Chimbus saw white settlers planting coffee, they began growing coffee themselves as a cash crop. In 1964 I met a 50-year-old Chimbu man, unable to read, wearing a traditional grass skirt, and born into a society still using stone tools, who had become rich by growing coffee, used his profits to buy a sawmill for $100,000 cash, and bought a fleet of trucks to transport his coffee and timber to market. In contrast, a neighboring highland people with whom I worked for eight years, the Daribi, are especially conservative and uninterested in new technology. When the first helicopter landed in the Daribi area, they briefly looked at it and just went back to what they had been doing; the Chimbus would have been bargaining to charter it. As a result, Chimbus are now moving into the Daribi area, taking it over for plantations, and reducing the Daribi to working for them.
On every other continent as well, certain native societies have proved very receptive, adopted foreign ways and technology selectively, and integrated them successfully into their own society. In Nigeria the Ibo people became the local entrepreneurial equivalent of New Guinea's Chimbus. Today the most numerous Native American tribe in the United States is the Navajo, who on European arrival were just one of several hundred tribes. But the Navajo proved especially resilient and able to deal selectively with innovation. They incorporated Western dyes into their weav-

NECESSITY'S

MOTHER • 153ing, became silversmiths and ranchers, and now drive trucks while continuing to live in traditional dwellings.
Among the supposedly conservative Aboriginal Australians as well, there are receptive societies along with conservative ones. At the one extreme, the Tasmanians continued to use stone tools superseded tens of thousands of years earlier in Europe and replaced in most of mainland Australia too. At the opposite extreme, some aboriginal fishing groups of southeastern Australia devised elaborate technologies for managing fish populations, including the construction of canals, weirs, and standing traps.
Thus, the development and reception of inventions vary enormously from society to society on the same continent. They also vary over time within the same society. Nowadays, Islamic societies in the Middle East are relatively conservative and not at the forefront of technology. But medieval Islam in the same region was technologically advanced and open to innovation. It achieved far higher literacy rates than contemporary Europe; it assimilated the legacy of classical Greek civilization to such a degree that many classical Greek books are now known to us only through Arabic copies; it invented or elaborated windmills, tidal mills, trigonometry, and lateen sails; it made major advances in metallurgy, mechanical and chemical engineering, and irrigation methods; and it adopted paper and gunpowder from China and transmitted them to Europe. In the Middle Ages the flow of technology was overwhelmingly from Islam to Europe, rather than from Europe to Islam as it is today. Only after around a.d. 1500 did the net direction of flow begin to reverse.
Innovation in China too fluctuated markedly with time. Until around a.d. 1450, China was technologically much more innovative and advanced than Europe, even more so than medieval Islam. The long list of Chinese inventions includes canal lock gates, cast iron, deep drilling, efficient animal harnesses, gunpowder, kites, magnetic compasses, movable type, paper, porcelain, printing (except for the Phaistos disk), sternpost rudders, and wheelbarrows. China then ceased to be innovative for reasons about which we shall speculate in the Epilogue. Conversely, we think of western Europe and its derived North American societies as leading the modern world in technological innovation, but technology was less advanced in western Europe than in any other "civilized" area of the Old World until the late Middle Ages.

Z 5 4 "

GUNS,GERMS. AND STEELThus, it is untrue that there are continents whose societies have tended to be innovative and continents whose societies have tended to be conservative. On any continent, at any given time, there are innovative societies and also conservative ones. In addition, receptivity to innovation fluctuates in time within the same region.
On reflection, these conclusions are precisely what one would expect if a society's innovativeness is determined by many independent factors. Without a detailed knowledge of all of those factors, innovativeness becomes unpredictable. Hence social scientists continue to debate the specific reasons why receptivity changed in Islam, China, and Europe, and why the Chimbus, Ibos, and Navajo were more receptive to new technology than were their neighbors. To the student of broad historical patterns, though, it makes no difference what the specific reasons were in each of those cases. The myriad factors affecting innovativeness make the historian's task paradoxically easier, by converting societal variation in innovativeness into essentially a random variable. That means that, over a large enough area (such as a whole continent) at any particular time, some proportion of societies is likely to be innovative.
where do innovations actually come from? For all societies except the few past ones that were completely isolated, much or most new technology is not invented locally but is instead borrowed from other societies. The relative importance of local invention and of borrowing depends mainly on two factors: the ease of invention of the particular technology, and the proximity of the particular society to other societies.
Some inventions arose straightforwardly from a handling of natural raw materials. Such inventions developed on many independent occasions in world history, at different places and times. One example, which we have already considered at length, is plant domestication, with at least nine independent origins. Another is pottery, which may have arisen from observations of the behavior of clay, a very widespread natural material, when dried or heated. Pottery appeared in Japan around 14,000 years ago, in the Fertile Crescent and China by around 10,000 years ago. and in Amazonia, Africa's Sahel zone, the U.S. Southeast, and Mexico thereafter.
An example of a much more difficult invention is writing, which does not suggest itself by observation of any natural material. As we saw in Chapter 12, it had only a few independent origins, and the alphabet arose

NECESSITY'S

MOTHER • Z 5 5apparently only once in world history. Other difficult inventions include the water wheel, rotary quern, tooth gearing, magnetic compass, windmill, and camera obscura, all of which were invented only once or twice in the Old World and never in the New World.
Such complex inventions were usually acquired by borrowing, because they spread more rapidly than they could be independently invented locally. A clear example is the wheel, which is first attested around 3400 b.c. near the Black Sea, and then turns up within the next few centuries over much of Europe and Asia. All those early Old World wheels are of a peculiar design: a solid wooden circle constructed of three planks fastened together, rather than a rim with spokes. In contrast, the sole wheels of Native American societies (depicted on Mexican ceramic vessels) consisted of a single piece, suggesting a second independent invention of the wheel— as one would expect from other evidence for the isolation of New World from Old World civilizations.
No one thinks that that same peculiar Old World wheel design appeared repeatedly by chance at many separate sites of the Old World within a few centuries of each other, after 7 million years of wheelless human history. Instead, the utility of the wheel surely caused it to diffuse rapidly east and west over the Old World from its sole site of invention. Other examples of complex technologies that diffused east and west in the ancient Old World, from a single West Asian source, include door locks, pulleys, rotary querns, windmills—and the alphabet. A New World example of technological diffusion is metallurgy, which spread from the Andes via Panama to Mesoamerica.
When a widely useful invention does crop up in one society, it then tends to spread in either of two ways. One way is that other societies see or learn of the invention, are receptive to it, and adopt it. The second is that societies lacking the invention find themselves at a disadvantage visa-vis the inventing society, and they become overwhelmed and replaced if the disadvantage is sufficiently great. A simple example is the spread of muskets among New Zealand's Maori tribes. One tribe, the Ngapuhi, adopted muskets from European traders around 1818. Over the course of the next 15 years, New Zealand was convulsed by the so-called Musket Wars, as musketless tribes either acquired muskets or were subjugated by tribes already armed with them. The outcome was that musket technology had spread throughout the whole of New Zealand by 1833: all surviving Maori tribes now had muskets.

256

• GUNS, GtRMS, AND STEELWhen societies do adopt a new technology from the society that invented it, the diffusion may occur in many different contexts. They include peaceful trade (as in the spread of transistors from the United States to Japan in 1954), espionage (the smuggling of silkworms from Southeast Asia to the Mideast in a.d. 552), emigration (the spread of French glass and clothing manufacturing techniques over Europe by the 200,000 Huguenots expelled from France in 1685), and war. A crucial case of the last was the transfer of Chinese papermaking techniques to Islam, made possible when an Arab army defeated a Chinese army at the battle of Talas River in Central Asia in a.d. 751, found some papermakers among the prisoners of war, and brought them to Samarkand to set up paper manufacture.
In Chapter 12 we saw that cultural diffusion can involve either detailed "blueprints" or just vague ideas stimulating a reinvention of details. While Chapter 12 illustrated those alternatives for the spread of writing, they also apply to the diffusion of technology. The preceding paragraph gave examples of blueprint copying, whereas the transfer of Chinese porcelain technology to Europe provides an instance of long-drawn-out idea diffusion. Porcelain, a fine-grained translucent pottery, was invented in China around the 7th century a.d. When it began to reach Europe by the Silk Road in the 14th century (with no information about how it was manufactured), it was much admired, and many unsuccessful attempts were made to imitate it. Not until 1707 did the German alchemist Johann Bottger, after lengthy experiments with processes and with mixing various minerals and clays together, hit upon the solution and establish the now famous Meissen porcelain works. More or less independent later experiments in France and England led to Sevres, Wedgwood, and Spode porcelains. Thus, European potters had to reinvent Chinese manufacturing methods for themselves, but they were stimulated to do so by having models of the desired product before them.
depending on their geographic location, societies differ in how readily they can receive technology by diffusion from other societies. The most isolated people on Earth in recent history were the Aboriginal Tasma-nians, living without oceangoing watercraft on an island 100 miles from Australia, itself the most isolated continent. The Tasmanians had no contact with other societies for 10,000 years and acquired no new technology

NECESSITY'S

MOTHER • Z 5 7other than what they invented themselves. Australians and New Guineans, separated from the Asian mainland by the Indonesian island chain, received only a trickle of inventions from Asia. The societies most accessible to receiving inventions by diffusion were those embedded in the major continents. In these societies technology developed most rapidly, because they accumulated not only their own inventions but also those of other societies. For example, medieval Islam, centrally located in Eurasia, acquired inventions from India and China and inherited ancient Greek
learning.
The importance of diffusion, and of geographic location in making it possible, is strikingly illustrated by some otherwise incomprehensible cases of societies that abandoned powerful technologies. We tend to assume that useful technologies, once acquired, inevitably persist until superseded by better ones. In reality, technologies must be not only acquired but also maintained, and that too depends on many unpredictable factors. Any society goes through social movements or fads, in which economically useless things become valued or useful things devalued temporarily. Nowadays, when almost all societies on Earth are connected to each other, we cannot imagine a fad's going so far that an important technology would actually be discarded. A society that temporarily turned against a powerful technology would continue to see it being used by neighboring societies and would have the opportunity to reacquire it by diffusion (or would be conquered by neighbors if it failed to do so). But such fads can persist in isolated societies.
A famous example involves Japan's abandonment of guns. Firearms reached Japan in a.d. 1543, when two Portuguese adventurers armed with harquebuses (primitive guns) arrived on a Chinese cargo ship. The Japanese were so impressed by the new weapon that they commenced indigenous gun production, greatly improved gun technology, and by a.d. 1600 owned more and better guns than any other country in the world.
But there were also factors working against the acceptance of firearms in Japan. The country had a numerous warrior class, the samurai, for whom swords rated as class symbols and works of an (and as means for subjugating the lower classes). Japanese warfare had previously involved single combats between samurai swordsmen, who stood in the open, made ritual speeches, and then took pride in fighting gracefully. Such behavior became lethal in the presence of peasant soldiers ungracefully blasting away with guns. In addition, guns were a foreign invention and grew to

Z 5 8 •

GUNS,GERMS, AND STEELbe despised, as did other things foreign in Japan after 1600. The samurai-controlled government began by restricting gun production to a few cities, then introduced a requirement of a government license for producing a gun, then issued licenses only for guns produced for the government, and finally reduced government orders for guns, until Japan was almost without functional guns again.
Contemporary European rulers also included some who despised guns and tried to restrict their availability. But such measures never got far in Europe, where any country that temporarily swore off firearms would be promptly overrun by gun-toting neighboring countries. Only because Japan was a populous, isolated island could it get away with its rejection of the powerful new military technology. Its safety in isolation came to an end in 1853, when the visit of Commander Perry's U.S. fleet bristling with cannons convinced Japan of its need to resume gun manufacture.
That rejection and China's abandonment of oceangoing ships (as well as of mechanical clocks and water-driven spinning machines) are well-known historical instances of technological reversals in isolated or semi-isolated societies. Other such reversals occurred in prehistoric times. The extreme case is that of Aboriginal Tasmanians, who abandoned even bone tools and fishing to become the society with the simplest technology in the modern world (Chapter 15). Aboriginal Australians may have adopted and then abandoned bows and arrows. Torres Islanders abandoned canoes, while Gaua Islanders abandoned and then readopted them. Pottery was abandoned throughout Polynesia. Most Polynesians and many Melane-sians abandoned the use of bows and arrows in war. Polar Eskimos lost the bow and arrow and the kayak, while Dorset Eskimos lost the bow and arrow, bow drill, and dogs.
These examples, at first so bizarre to us, illustrate well the roles of geography and of diffusion in the history of technology. Without diffusion, fewer technologies are acquired, and more existing technologies are lost.
because technology begets more technology, the importance of an invention's diffusion potentially exceeds the importance of the original invention. Technology's history exemplifies what is termed an autocata-lytic process: that is, one that speeds up at a rate that increases with time, because the process catalyzes itself. The explosion of technology since the Industrial Revolution impresses us today, but the medieval explosion was

NECESSITY'S MOTHER • 159

equally impressive compared with that of the Bronze Age, which in turn dwarfed that of the Upper Paleolithic.
One reason why technology tends to catalyze itself is that advances depend upon previous mastery of simpler problems. For example, Stone Age farmers did not proceed directly to extracting and working iron, which requires high-temperature furnaces. Instead, iron ore metallurgy grew out of thousands of years of human experience with natural outcrops of pure metals soft enough to be hammered into shape without heat (copper and gold). It also grew out of thousands of years of development of simple furnaces to make pottery, and then to extract copper ores and work copper alloys (bronzes) that do not require as high temperatures as does iron. In both the Fertile Crescent and China, iron objects became common only after about 2,000 years of experience of bronze metallurgy. New World societies had just begun making bronze artifacts and had not yet started making iron ones at the time when the arrival of Europeans truncated the New World's independent trajectory.
The other main reason for autocatalysis is that new technologies and materials make it possible to generate still other new technologies by recombination. For instance, why did printing spread explosively in medieval Europe after Gutenberg printed his Bible in a.d. 1455, but not after that unknown printer printed the Phaistos disk in 1700 b.c.? The explanation is partly that medieval European printers were able to combine six technological advances, most of which were unavailable to the maker of the Phaistos disk. Of those advances—in paper, movable type, metallurgy, presses, inks, and scripts—paper and the idea of movable type reached Europe from China. Gutenberg's development of typecasting from metal dies, to overcome the potentially fatal problem of nonuniform type size, depended on many metallurgical developments: steel for letter punches, brass or bronze alloys (later replaced by steel) for dies, lead for molds, and a tin-zinc-lead alloy for type. Gutenberg's press was derived from screw presses in use for making wine and olive oil, while his ink was an oil-based improvement on existing inks. The alphabetic scripts that medieval Europe inherited from three millennia of alphabet development lent themselves to printing with movable type, because only a few dozen letter forms had to be cast, as opposed to the thousands of signs required for Chinese writing.
In all six respects, the maker of the Phaistos disk had access to much less powerful technologies to combine into a printing system than did Gutenberg. The disk's writing medium was clay, which is much bulkier

2. 6 O •

GUNS,GERMS, AND STEELand heavier than paper. The metallurgical skills, inks, and presses of 1700 b.c. Crete were more primitive than those of a.d. 1455 Germany, so the disk had to be punched by hand rather than by cast movable type locked into a metal frame, inked, and pressed. The disk's script was a syllabary with more signs, of more complex form, than the Roman alphabet used by Gutenberg. As a result, the Phaistos disk's printing technology was much clumsier, and offered fewer advantages over writing by hand, than Gutenberg's printing press. In addition to all those technological drawbacks, the Phaistos disk was printed at a time when knowledge of writing was confined to a few palace or temple scribes. Hence there was little demand for the disk maker's beautiful product, and little incentive to invest in making the dozens of hand punches required. In contrast, the potential mass market for printing in medieval Europe induced numerous investors to lend money to Gutenberg.
humantechnology developed from the first stone tools, in use by two and a half million years ago, to the 1996 laser printer that replaced my already outdated 1992 laser printer and that was used to print this book's manuscript. The rate of development was undetectably slow at the beginning, when hundreds of thousands of years passed with no discernible change in our stone tools and with no surviving evidence for artifacts made of other materials. Today, technology advances so rapidly that it is reported in the daily newspaper.
In this long history of accelerating development, one can single out two especially significant jumps. The first, occurring between 100,000 and 50,000 years ago, probably was made possible by genetic changes in our bodies: namely, by evolution of the modern anatomy permitting modern speech or modern brain function, or both. That jump led to bone tools, single-purpose stone tools, and compound tools. The second jump resulted from our adoption of a sedentary lifestyle, which happened at different times in different parts of the world, as early as 13,000 years ago in some areas and not even today in others. For the most part, that adoption was linked to our adoption of food production, which required us to remain close to our crops, orchards, and stored food surpluses.
Sedentary living was decisive for the history of technology, because it enabled people to accumulate nonportable possessions. Nomadic hunter-

NECESSITY'S

MOTHER • Z 6 Igatherers are limited to technology that can be carried. If you move often and lack vehicles or draft animals, you confine your possessions to babies, weapons, and a bare minimum of other absolute necessities small enough to carry. You can't be burdened with pottery and printing presses as you shift camp. That practical difficulty probably explains the tantalizingly early appearance of some technologies, followed by a long delay in their further development. For example, the earliest attested precursors of ceramics are fired clay figurines made in the area of modern Czechoslovakia 27,000 years ago, long before the oldest known fired clay vessels (from Japan 14,000 years ago). The same area of Czechoslovakia at the same time has yielded the earliest evidence for weaving, otherwise not attested until the oldest known basket appears around 13,000 years ago and the oldest known woven cloth around 9,000 years ago. Despite these very early first steps, neither pottery nor weaving took off until people became sedentary and thereby escaped the problem of transporting pots and looms.
Besides permitting sedentary living and hence the accumulation of possessions, food production was decisive in the history of technology for another reason. It became possible, for the first time in human evolution, to develop economically specialized societies consisting of non-food-producing specialists fed by food-producing peasants. But we already saw, in Part 2 of this book, that food production arose at different times in different continents. In addition, as we've seen in this chapter, local technology depends, for both its origin and its maintenance, not only on local invention but also on the diffusion of technology from elsewhere. That consideration tended to cause technology to develop most rapidly on continents with few geographic and ecological barriers to diffusion, either within that continent or on other continents. Finally, each society on a continent represents one more opportunity to invent and adopt a technology, because societies vary greatly in their innovativeness for many separate reasons. Hence, all other things being equal, technology develops fastest in large productive regions with large human populations, many potential inventors, and many competing societies.
Let us now summarize how variations in these three factors—time of onset of food production, barriers to diffusion, and human population size—led straightforwardly to the observed intercontinental differences in the development of technology. Eurasia (effectively including North

2. 6 Z •

GUNS,GERMS, AND STEELAfrica) is the world's largest landmass, encompassing the largest number of competing societies. It was also the landmass with the two centers where food production began the earliest: the Fertile Crescent and China. Its east-west major axis permitted many inventions adopted in one part of Eurasia to spread relatively rapidly to societies at similar latitudes and climates elsewhere in Eurasia. Its breadth along its minor axis (north-south) contrasts with the Americas' narrowness at the Isthmus of Panama. It lacks the severe ecological barriers transecting the major axes of the Americas and Africa. Thus, geographic and ecological barriers to diffusion of technology were less severe in Eurasia than in other continents. Thanks to all these factors, Eurasia was the continent on which technology started its post-Pleistocene acceleration earliest and resulted in the greatest local accumulation of technologies.
North and South America are conventionally regarded as separate continents, but they have been connected for several million years, pose similar historical problems, and may be considered together for comparison with Eurasia. The Americas form the world's second-largest landmass, significantly smaller than Eurasia. However, they are fragmented by geography and by ecology: the Isthmus of Panama, only 40 miles wide, virtually transects the Americas geographically, as do the isthmus's Darien rain forests and the northern Mexican desert ecologically. The latter desert separated advanced human societies of Mesoamerica from those of North America, while the isthmus separated advanced societies of Mesoamerica from those of the Andes and Amazonia. In addition, the main axis of the Americas is north-south, forcing most diffusion to go against a gradient of latitude (and climate) rather than to operate within the same latitude. For example, wheels were invented in Mesoamerica, and llamas were domesticated in the central Andes by 3000 b.c., but 5,000 years later the Americas' sole beast of burden and sole wheels had still not encountered each other, even though the distance separating Mesoamerica's Maya societies from the northern border of the Inca Empire (1,200 miles) was far less than the 8,000 miles separating wheel– and horse-sharing France and China. Those factors seem to me to account for the Americas' technological lag behind Eurasia.
Sub-Saharan Africa is the world's third largest landmass, considerably smaller than the Americas. Throughout most of human history it was far more accessible to Eurasia than were the Americas, but the Saharan desert is still a major ecological barrier separating sub-Saharan Africa from

NECESSITY'S

MOTHER • 2. 6 3Eurasia plus North Africa. Africa's north-south axis posed a further obstacle to the diffusion of technology, both between Eurasia and sub-Saharan Africa and within the sub-Saharan region itself. As an illustration of the latter obstacle, pottery and iron metallurgy arose in or reached sub-Saharan Africa's Sahel zone (north of the equator) at least as early as they reached western Europe. However, pottery did not reach the southern tip of Africa until around a.d. 1, and metallurgy had not yet diffused overland to the southern tip by the time that it arrived there from Europe on ships.
Finally, Australia is the smallest continent. The very low rainfall and productivity of most of Australia makes it effectively even smaller as regards its capacity to support human populations. It is also the most isolated continent. In addition, food production never arose indigenously in Australia. Those factors combined to leave Australia the sole continent still without metal artifacts in modern times.
Table 13.1 translates these factors into numbers, by comparing the continents with respect to their areas and their modern human populations. The continents' populations 10,000 years ago, just before the rise of food production, are not known but surely stood in the same sequence, since many of the areas producing the most food today would also have been productive areas for hunter-gatherers 10,000 years ago. The differences in population are glaring: Eurasia's (including North Africa's) is nearly 6 times that of the Americas, nearly 8 times that of Africa's, and 230 times that of Australia's. Larger populations mean more inventors and more competing societies. Table 13.1 by itself goes a long way toward explaining the origins of guns and steel in Eurasia.
All these effects that continental differences in area, population, ease of
table13.1Human Populations of the ContinentsContinent 1990 AreaPopulation (square miles)Eurasia and North Africa 4,120,000,000 24,200,000
(Eurasia) (4,000,000,000) (21,500,000)
(North Africa) (120,000,000) (2,700,000)
North America and South America 736,000,000 16,400,000
Sub-Saharan Africa 535,000,000 9,100,000
Australia 18,000,000 3,000,000

Z 6 4 "

GUNS,GERMS, AND STEELdiffusion, and onset of food production exerted on the rise of technology became exaggerated, because technology catalyzes itself. Eurasia's considerable initial advantage thereby was translated into a huge lead as of a.d. 1492—for reasons of Eurasia's distinctive geography rather than of distinctive human intellect. The New Guineans whom I know include potential Edisons. But they directed their ingenuity toward technological problems appropriate to their situations: the problems of surviving without any imported items in the New Guinea jungle, rather than the problem of inventing phonographs.

CHAPTER

14from egalitarianism to kleptocracy

IN 1979, WHILE I

WAS FLYING WITH MISSIONARY FRIENDS over a remote swamp-filled basin of New Guinea, I noticed a few huts many miles apart. The pilot explained to me that, somewhere in that muddy expanse below us, a group of Indonesian crocodile hunters had recently come across a group of New Guinea nomads. Both groups had panicked, and the encounter had ended with the Indonesians shooting several of the nomads.My missionary friends guessed that the nomads belonged to an uncon-tacted group called the Fayu, known to the outside world only through accounts by their terrified neighbors, a missionized group of erstwhile nomads called the Kirikiri. First contacts between outsiders and New Guinea groups are always potentially dangerous, but this beginning was especially inauspicious. Nevertheless, my friend Doug flew in by helicopter to try to establish friendly relations with the Fayu. He returned, alive but shaken, to tell a remarkable story.
It turned out that the Fayu normally lived as single families, scattered through the swamp and coming together once or twice each year to negotiate exchanges of brides. Doug's visit coincided with such a gathering, of a few dozen Fayu. To us, a few dozen people constitute a small, ordinary gathering, but to the Fayu it was a rare, frightening event. Murderers sud-

1 6 6 •

GUNS,GERMS, AND STEELdenly found themselves face-to-face with their victim's relatives. For example, one Fayu man spotted the man who had killed his father. The son raised his ax and rushed at the murderer but was wrestled to the ground by friends; then the murderer came at the prostrate son with an ax and was also wrestled down. Both men were held, screaming in rage, until they seemed sufficiently exhausted to be released. Other men periodically shouted insults at each other, shook with anger and frustration, and pounded the ground with their axes. That tension continued for the several days of the gathering, while Doug prayed that the visit would not end in violence.
The Fayu consist of about 400 hunter-gatherers, divided into four clans and wandering over a few hundred square miles. According to their own account, they had formerly numbered about 2,000, but their population had been greatly reduced as a result of Fayu killing Fayu. They lacked political and social mechanisms, which we take for granted, to achieve peaceful resolution of serious disputes. Eventually, as a result of Doug's visit, one group of Fayu invited a courageous husband-and-wife missionary couple to live with them. The couple has now resided there for a dozen years and gradually persuaded the Fayu to renounce violence. The Fayu are thereby being brought into the modern world, where they face an uncertain future.
Many other previously uncontacted groups of New Guineans and Amazonian Indians have similarly owed to missionaries their incorporation into modern society. After the-missionaries come teachers and doctors, bureaucrats and soldiers. The spreads of government and of religion have thus been linked to each other throughout recorded history, whether the spread has been peaceful (as eventually with the Fayu) or by force. In the latter case it is often government that organizes the conquest, and religion that justifies it. While nomads and tribespeople occasionally defeat organized governments and religions, the trend over the past 13,000 years has been for the nomads and tribespeople to lose.
At the end of the last Ice Age, much of the world's population lived in societies similar to that of the Fayu today, and no people then lived in a much more complex society. As recently as a.d. 1500, less than 20 percent of the world's land area was marked off by boundaries into states run by bureaucrats and governed by laws. Today, all land except Antarctica's is so divided. Descendants of those societies that achieved centralized government and organized religion earliest ended up dominating the modern

FROM

EGALITARIANISM TO KLEPTOCRACY • 2. 6 Jworld. The combination of government and religion has thus functioned, together with germs, writing, and technology, as one of the four main sets of proximate agents leading to history's broadest pattern. How did government and religion arise?
fayu bands and modern states represent opposite extremes along the spectrum of human societies. Modern American society and the Fayu differ in the presence or absence of a professional police force, cities, money, distinctions between rich and poor, and many other political, economic, and social institutions. Did all of those institutions arise together, or did some arise before others? We can infer the answer to this question by comparing modern societies at different levels of organization, by examining written accounts or archaeological evidence about past societies, and by observing how a society's institutions change over time.
Cultural anthropologists attempting to describe the diversity of human societies often divide them into as many as half a dozen categories. Any such attempt to define stages of any evolutionary or developmental continuum—whether of musical styles, human life stages, or human societies— is doubly doomed to imperfection. First, because each stage grows out of some previous stage, the lines of demarcation are inevitably arbitrary. (For example, is a 19-year-old person an adolescent or a young adult?) Second, developmental sequences are not invariant, so examples pigeonholed under the same stage are inevitably heterogeneous. (Brahms and Liszt would turn in their graves to know that they are now grouped together as composers of the romantic period.) Nevertheless, arbitrarily delineated stages provide a useful shorthand for discussing the diversity of music and of human societies, provided one bears in mind the above caveats. In that spirit, we shall use a simple classification based on just four categories— band, tribe, chiefdom, and state (see Table 14.1)—to understand societies.
Bands are the tiniest societies, consisting typically of 5 to 80 people, most or all of them close relatives by birth or by marriage. In effect, a band is an extended family or several related extended families. Today, bands still living autonomously are almost confined to the most remote parts of New Guinea and Amazonia, but within modern times there were many others that have only recently fallen under state control or been assimi-ated or exterminated. They include many or most African Pygmies, southern African San hunter-gatherers (so-called Bushmen), Aboriginal

Z 6 8 • GUNS, GERMS,

and steeltable 14.1 Types of Societies
Band Tribe Chiefdom State
MembershipNumber of dozens hundreds thousands over 50,000
people
Settlement nomadic fixed: 1 fixed: 1 or more fixed: many
pattern village villages villages
and cities
Basis of relation– kin kin-based class and resi– class and
ships clans dence residence
Ethnicities and 1 1 1 1 or more
languages
Government
Decision making, "egalitarian" "egalitarian" centralized, centralized
leadership or hereditary
big-man
Bureaucracy none none none, or 1 or many levels
2 levels
Monopoly of no no yes yes
force and
information
Conflict resolu– informal informal centralized laws, judges
tion
Hierarchy of no no no-> para– capital
settlement mount village
Australians, Eskimos (Inuit), and Indians of some resource-poor areas of the Americas such as Tierra del Fuego and the northern boreal forests. All those modern bands are or were nomadic hunter-gatherers rather than settled food producers. Probably all humans lived in bands until at least 40,000 years ago, and most still did as recently as 11,000 years ago.
Bands lack many institutions that we take for granted in our own society. They have no permanent single base of residence. The band's land is used jointly by the whole group, instead of being partitioned among subgroups or individuals. There is no regular economic specialization, except by age and sex: all able-bodied individuals forage for food. There are no formal institutions, such as laws, police, and treaties, to resolve conflicts within and between bands. Band organization is often described as

FROM

EGAHTARIANISM TO KLEPTOCRACY • 169Band Tribe Chtefdom StateReligionJustifies kiepto– no no yes yes —no
cracy?
Economy
Food production no no-*yes yes-*intensive intensive
Division of labor no no no— yes yes
Exchange reciprocal reciprocal redistributive redistribu-
(" tribute") tive
("taxes")
Control of land band clan chief various
Society
Stratified no no yes, by kin yes. not
by kin
Slavery no no small-scale large-scale
Luxury goods no no yes yes
for elite
Public architec– no no no —yes yes
ture
Indigenous lit– no no no often
eracv
A horizontal arrow indicates rhat the attribute vanes between less and more complex societies of that type.
"egalitarian": there is no formalized social stratification into upper and lower classes, no formalized or hereditary leadership, and no formalized monopolies of information and decision making. However, the term egalitarian" should not be taken to mean that all band members are equal n prestige and contribute equally to decisions. Rather, the term merely means that any band "leadership" is informal and acquired through qualities such as personality, strength, intelligence, and fighting skills.
My own experience with bands comes from the swampy lowland area ew kumea where the Fayu live, a region known as the Lakes Plains. ere, I still encounter extended families of a few adults with their dependent children and elderly, living in crude temporary shelters along streams traveling by canoe and or. foot. Why do peoples of the Lakes Plains

Z 7 O •

GUNS,GERMS, AND STEELcontinue to live as nomadic bands, when most other New Guinea peoples and almost all other peoples elsewhere in the world, now live in settled larger groups? The explanation is that the region lacks dense local concentrations of resources that would permit many people to live together, and that (until the arrival of missionaries bringing crop plants) it also lacked native plants that could have permitted productive farming. The bands' food staple is the sago palm tree, whose core yields a starchy pith when the palm reaches maturity. The bands are nomadic, because they must move when they have cut the mature sago trees in an area. Band numbers are kept low by diseases (especially malaria), by the lack of raw materials in the swamp (even stone for tools must be obtained by trade), and by the limited amount of food that the swamp yields for humans. Similar limitations on the resources accessible to existing human technology prevail in the regions of the world recently occupied by other bands.
Our closest animal relatives, the gorillas and chimpanzees and bonobos of Africa, also live in bands. All humans presumably did so too, until improved technology for extracting food allowed some hunter-gatherers to settle in permanent dwellings in some resource-rich areas. The band is the political, economic, and social organization that we inherited from our millions of years of evolutionary history. Our developments beyond it all took place within the last few tens of thousands of years.
The first of those stages beyond the band is termed the tribe, which differs in being larger (typically comprising hundreds rather than dozens of people) and usually having fixed settlements. However, some tribes and even chiefdoms consist of herders who move seasonally.
Tribal organization is exemplified by New Guinea highlanders, whose political unit before the arrival of colonial government was a village or else a close-knit cluster of villages. This political definition of "tribe" is thus often much smaller than what linguists and cultural anthropologists would define as a tribe—namely, a group that shares language and culture. For example, in 1964 I began to work among a group of highlanders known as the Fore. By linguistic and cultural standards, there were then 12,000 Fore, speaking two mutually intelligible dialects and living in 65 villages of several hundred people each. But there was no political unity whatsoever among villages of the Fore language group. Each hamlet was involved in a kaleidoscopically changing pattern of war and shifting alu*

FROM

EGALITARIANISM TO KLEPTOCRACY • i 7 Iances with all neighboring hamlets, regardless of whether the neighbors were Fore or speakers of a different language.
Tribes recently independent and now variously subordinated to national states, still occupy much of New Guinea, Melanesia, and Amazonia. Similar tribal organization in the past is inferred from archaeological evidence of settlements that were substantial but lacked the archaeological hallmarks of chiefdoms that I shall explain below. That evidence suggests that tribal organization began to emerge around 13,000 years ago in the Fertile Crescent and later in some other areas. A prerequisite for lining in settlements is either food production or else a productive environment with especially concentrated resources that can be hunted and gathered within a small area. That's why settlements, and by inference tribes, began to proliferate in the Fertile Crescent at that time, when climate changes and improved technology combined to permit abundant harvests of wild cereals.
Besides differing from a band by virtue of its settled residence and its larger numbers, a tribe also differs in that it consists of more than one formally recognized kinship group, termed clans, which exchange marriage partners. Land belongs to a particular clan, not to the whole tribe. However, the number of people in a tribe is still low enough that everyone knows everyone else by name and relationships.
For other types of human groups as well, "a few hundred" seems to be an upper limit for group size compatible with everyone's knowing everybody. In our state society, for instance, school principals are likely to know all their students by name if the school contains a few hundred children, but not if it contains a few thousand children. One reason why the organization of human government tends to change from that of a tribe to that of a chiefdom in societies with more than a few hundred members is that the difficult issue of conflict resolution between strangers becomes increasingly acute in larger groups. A fact further diffusing potential problems of conflict resolution in tribes is that almost everyone is related to everyone else, by blood or marriage or both. Those ties of relationships binding all tribal members make police, laws, and other conflict-resolving institutions 0 larger societies unnecessary, since any two villagers getting into an argument will share many kin, who apply pressure on them to keep it from becoming violent. In traditional New Guinea society, if a New Guinean appened to encounter an unfamiliar New Guinean while both were away F0m their respective villages, the two engaged in a long discussion of their

2 7 2, • GUNS, GERMS,

and steelrelatives, in an attempt to establish some relationship and hence some reason why the two should not attempt to kill each other.
Despite all of these differences between bands and tribes, many similarities remain. Tribes still have an informal, "egalitarian" system of government. Information and decision making are both communal. In the New Guinea highlands, I have watched village meetings where all adults in the village were present, sitting on the ground, and individuals made speeches, without any appearance of one person's "chairing" the discussion. Many highland villages do have someone known as the "big-man," the most influential man of the village. But that position is not a formal office to be filled and carries only limited power. The big-man has no independent decision-making authority, knows no diplomatic secrets, and can do no more than attempt to sway communal decisions. Big-men achieve that status by their own attributes; the position is not inherited.
Tribes also share with bands an "egalitarian" social system, without ranked lineages or classes. Not only is status not inherited; no member of a traditional tribe or band can become disproportionately wealthy by his or her own efforts, because each individual has debts and obligations to many others. It is therefore impossible for an outsider to guess, from appearances, which of all the adult men in a village is the big-man: he lives in the same type of hut, wears the same clothes or ornaments, or is as naked, as everyone else.
Like bands, tribes lack a bureaucracy, police force, and taxes. Their economy is based on reciprocal exchanges between individuals or families, rather than on a redistribution of tribute paid to some central authority. Economic specialization is slight: full-time crafts specialists are lacking, and every able-bodied adult (including the big-man) participates in growing, gathering, or hunting food. I recall one occasion when I was walking past a garden in the Solomons Islands, saw a man digging and waving at me in the distance, and realized to my astonishment that it was a friend or mine named Faletau. He was the most famous wood carver of the Solomons, an artist of great originality—but that did not free him of the necessity to grow his own sweet potatoes. Since tribes thus lack economic specialists, they also lack slaves, because there are no specialized menial jobs for a slave to perform.
Just as musical composers of the classical period range from C. P. Bach to Schubert and thereby cover the whole spectrum from baroque composers to romantic composers, tribes also shade into bands at otie >|

FROM

EGALITARIANISM TO KLEPTOCRACY • 2 7 3extreme and into chiefdoms at the opposite extreme. In particular, a tribal big-man's role in dividing the meat of pigs slaughtered for feasts points to the role of chiefs in collecting and redistributing food and goods—now reconstrued as tribute—in chiefdoms. Similarly, presence or absence of oublic architecture is supposedly one of the distinctions between tribes and chiefdoms, but large New Guinea villages often have cult houses (known as haus tamburan, on the Sepik River) that presage the temples of chiefdoms.
a.lthough a few bands and tribes survive today on remote and ecologically marginal lands outside state control, fully independent chiefdoms had disappeared by the early twentieth century, because they tended to occupy prime land coveted by states. However, as of a.d. 1492, chiefdoms were still widespread over much of the eastern United States, in productive areas of South and Central America and sub-Saharan Africa that had not yet been subsumed under native states, and in all of Polynesia. The archaeological evidence discussed below suggests that chiefdoms arose by around 5500 b.c. in the Fertile Crescent and by around 1000 b.c. in Mesoamerica and the Andes. Let us consider the distinctive features of chiefdoms, very different from modern European and American states and, at the same time, from bands and simple tribal societies.
As regards population size, chiefdoms were considerably larger than tribes, ranging from several thousand to several tens of thousands of people. That size created serious potential for internal conflict because, for any person living in a chiefdom, the vast majority of other people in the chiefdom were neither closely related by blood or marriage nor known by name. With the rise of chiefdoms around 7,500 years ago, people had to learn, for the first time in history, how to encounter strangers regularly without attempting to kill them.
Part of the solution to that problem was for one person, the chief, to exercise a monopoly on the right to use force. In contrast to a tribe's big-man, a chief held a recognized office, filled by hereditary right. Instead of the decentralized anarchy of a village meeting, the chief was a permanent centralized authority, made all significant decisions, and had a monopoly on critical information (such as what a neighboring chief was privately threatening, or what harvest the gods had supposedly promised). Unlike 'g-men, chiefs could be recognized from afar by visible distinguishing

2. 7 4 * GUNS, GERMS, AND

steelfeatures, such as a large fan worn over the back on Rennell Island in the Southwest Pacific. A commoner encountering a chief was obliged to perform ritual marks of respect, such as (on Hawaii) prostrating oneself. The chief's orders might be transmitted through one or two levels of bureaucrats, many of whom were themselves low-ranked chiefs. However, in contrast to state bureaucrats, chiefdom bureaucrats had generalized rather than specialized roles. In Polynesian Hawaii the same bureaucrats (termed konohiki) extracted tribute and oversaw irrigation and organized labor corvees for the chief, whereas state societies have separate tax collectors, water district managers, and draft boards.A chiefdom's large population in a small area required plenty of food, obtained by food production in most cases, by hunting-gathering in a few especially rich areas. For example, American Indians of the Pacific Northwest coast, such as the Kwakiutl, Nootka, and Tlingit Indians, lived under chiefs in villages without any agriculture or domestic animals, because the rivers and sea were so rich in salmon and halibut. The food surpluses generated by some people, relegated to the rank of commoners, went to feed the chiefs, their families, bureaucrats, and crafts specialists, who variously made canoes, adzes, or spittoons or worked as bird catchers or tattooers.
Luxury goods, consisting of those specialized crafts products or else rare objects obtained by long-distance trade, were reserved for chiefs. For example, Hawaiian chiefs had feather cloaks, some of them consisting of tens of thousands of feathers and requiring many human generations for their manufacture (by commoner cloak makers, of course). That concentration of luxury goods often makes it possible to recognize chiefdoms archaeologically, by the fact that some graves (those of chiefs) contain much richer goods than other graves (those of commoners), in contrast to the egalitarian burials of earlier human history. Some ancient complex chiefdoms can also be distinguished from tribal villages by the remains of elaborate public architecture (such as temples) and by a regional hierarchy of settlements, with one site (the site of the paramount chief) being obviously larger and having more administrative buildings and artifacts than other sites.
Like tribes, chiefdoms consisted of multiple hereditary lineages living at one site. However, whereas the lineages of tribal villages are equal-ranked clans, in a chiefdom all members of the chief's lineage had hereditary perquisites. In effect, the society was divided into hereditary chief and commoner classes, with Hawaiian chiefs themselves subdivided into eight

FROM

EGALITARIAN1SM TO KLEPTOCRACY • 275hierarchically ranked lineages, each concentrating its marriages within its own lineage. Furthermore, since chiefs required menial servants as well as specialized craftspeople, chiefdoms differed from tribes in having many jobs that could be filled by slaves, typically obtained by capture in raids.
The most distinctive economic feature of chiefdoms was their shift from reliance solely on the reciprocal exchanges characteristic of bands and tribes by which A gives B a gift while expecting that B at some unspecified future time will give a gift of comparable value to A. We modern state dwellers indulge in such behavior on birthdays and holidays, but most of our flow of goods is achieved instead by buying and selling for money according to the law of supply and demand. While continuing reciprocal exchanges and without marketing or money, chiefdoms developed an additional new system termed a redistributive economy. A simple example would involve a chief receiving wheat at harvest time from every farmer in the chiefdom, then throwing a feast for everybody and serving bread or else storing the wheat and gradually giving it out again in the months between harvests. When a large portion of the goods received from commoners was not redistributed to them but was retained and consumed by the chiefly lineages and craftspeople, the redistribution became tribute, a precursor of taxes that made its first appearance in chiefdoms. From the commoners the chiefs claimed not only goods but also labor for construction of public works, which again might return to benefit the commoners (for example, irrigation systems to help feed everybody) or instead benefited mainly the chiefs (for instance, lavish tombs).
We have been talking about chiefdoms generically, as if they were all the same. In fact, chiefdoms varied considerably. Larger ones tended to have more powerful chiefs, more ranks of chiefly lineages, greater distinctions between chiefs and commoners, more retention of tribute by the chiefs, more layers of bureaucrats, and grander public architecture. For instance, societies on small Polynesian islands were effectively rather similar to tribal societies with a big-man, except that the position of chief was hereditary. The chief's hut looked like any other hut, there were no bureaucrats or public works, the chief redistributed most goods he received back to the commoners, and land was controlled by the community. But on the largest Polynesian islands, such as Hawaii, Tahiti, and Tonga, chiefs were cognizable at a glance by their ornaments, public works were erected by arge labor forces, most tribute was retained by the chiefs, and all land as controlled by them. A further gradation among societies with ranked

Z 7 *> "

GUNS,GERMS, AND STEELlineages was from those where the political unit was a single autonomous village, to those consisting of a regional assemblage of villages in which the largest village with a paramount chief controlled the smaller villages with lesser chiefs.
Bynow, it should be obvious that chiefdoms introduced the dilemma fundamental to all centrally governed, nonegalitarian societies. At best, they do good by providing expensive services impossible to contract for on an individual basis. At worst, they function unabashedly as kleptocracies, transferring net wealth from commoners to upper classes. These noble and selfish functions are inextricably linked, although some governments emphasize much more of one function than of the other. The difference between a kleptocrat and a wise statesman, between a robber baron and a public benefactor, is merely one of degree: a matter of just how large a percentage of the tribute extracted from producers is retained by the elite, and how much the commoners like the public uses to which the redistributed tribute is put. We consider President Mobutu of Zaire a kleptocrat because he keeps too much tribute (the equivalent of billions of dollars) and redistributes too little tribute (no functioning telephone system in Zaire). We consider George Washington a statesman because he spent tax money on widely admired programs and did not enrich himself as president. Nevertheless, George Washington was born into wealth, which is much more unequally distributed in the United States than in New Guinea villages.
For any ranked society, whether a chiefdom or a state, one thus has to ask: why do the commoners tolerate the transfer of the fruits of their hard labor to kleptocrats? This question, raised by political theorists from Plato to Marx, is raised anew by voters in every modern election. Kleptocracies with little public support run the risk of being overthrown, either by downtrodden commoners or by upstart would-be replacement kleptocrats seeking public support by promising a higher ratio of services rendered to fruits stolen. For example, Hawaiian history was repeatedly punctuated by revolts against repressive chiefs, usually led by younger brothers promising less oppression. This may sound funny to us in the context of old Hawaii, until we reflect on all the misery still being caused by such struggles in the modern world.
What should an elite do to gain popular support while still maintaining

FROM

EGALITARIANISM TO KLEPTOCRACY • 2. 7 7more comfortable lifestyle than commoners? Kleptocrats throughout the ages have resorted to a mixture of four solutions:
1 Disarm the populace, and arm the elite. That's much easier in these davs of high-tech weaponry, produced only in industrial plants and easily monopolized by an elite, than in ancient times of spears and clubs easily made at home.
2 Make the masses happy by redistributing much of the tribute received, in popular ways. This principle was as valid for Hawaiian chiefs as it is for American politicians today.
3. Use the monopoly of force to promote happiness, by maintaining public order and curbing violence. This is potentially a big and underap-preciated advantage of centralized societies over noncentralized ones. Anthropologists formerly idealized band and tribal societies as gentle and non"iolent, because visiting anthropologists observed no murder in a band of 25 people in the course of a three-year study. Of course they didn't: it's easy to calculate that a band of a dozen adults and a dozen children, subject to the inevitable deaths occurring anyway for the usual reasons other than murder, could not perpetuate itself if in addition one of its dozen adults murdered another adult every three years. Much more extensive long-term information about band and tribal societies reveals that murder is a leading cause of death. For example, I happened to be visiting New Guinea's lyau people at a time when a woman anthropologist was interviewing lyau women about their life histories. Woman after woman, when asked to name her husband, named several sequential husbands who had died violent deaths. A typical answer went like this: "My first husband was killed by Elopi raiders. My second husband was killed by a man who wanted me, and who became my third husband. That husband was killed by the brother of my second husband, seeking to avenge his murder." Such biographies prove common for so-called gentle tribespeople and contributed to the acceptance of centralized authority as tribal societies grew larger.
4. The remaining way for kleptocrats to gain public support is to construct an ideology or religion justifying kleptocracy. Bands and tribes already had supernatural beliefs, just as do modern established religions. But the supernatural beliefs of bands and tribes did not serve to justify central authority, justify transfer of wealth, or maintain peace between unrelated individuals. When supernatural beliefs gained those functions and became institutionalized, they were thereby transformed into what we

2. 7

8 •GUNS,GERMS, AND STEELterm a religion. Hawaiian chiefs were typical of chiefs elsewhere, in asserting divinity, divine descent, or at least a hotline to the gods. The chief claimed to serve the people by interceding for them with the gods and reciting the ritual formulas required to obtain rain, good harvests, and success in fishing.
Chiefdoms characteristically have an ideology, precursor to an institutionalized religion, that buttresses the chief's authority. The chief may either combine the offices of political leader and priest in a single person, or may support a separate group of kleptocrats (that is, priests) whose function is to provide ideological justification for the chiefs. That is why chiefdoms devote so much collected tribute to constructing temples and other public works, which serve as centers of the official religion and visible signs of the chief's power.
Besides justifying the transfer of wealth to kleptocrats, institutionalized religion brings two other important benefits to centralized societies. First, shared ideology or religion helps solve the problem of how unrelated individuals are to live together without killing each other—by providing them with a bond not based on kinship. Second, it gives people a motive, other than genetic self-interest, for sacrificing their lives on behalf of others. At the cost of a few society members who die in battle as soldiers, the whole society becomes much more effective at conquering other societies or resisting attacks.
The political, economic, and social institutions most familiar to us today are those of states, which now rule all of the world's land area except for Antarctica. Many early states and all modern ones have had literate elites, and many modern states have literate masses as well. Vanished states tended to leave visible archaeological hallmarks, such as ruins of temples with standardized designs, at least four levels of settlement sizes, and pottery styles covering tens of thousands of square miles. We thereby know that states arose around 3700 b.c. in Mesopotamia and around 300 B.C. in Mesoamerica, over 2,000 years ago in the Andes, China, and Southeast Asia, and over 1,000 years ago in West Africa. In modern times the formation of states out of chiefdoms has been observed repeatedly. Thus, we possess much more information about past states and their formation than about past chiefdoms, tribes, and bands.
Protostates extend many features of large paramount (multivillage)

FROM

EG ALITARIANISM TO KLEPTOCRACY • Z 7 9hiefdoms. They continue the increase in size from bands to tribes to chief-doms. Whereas chiefdoms' populations range from a few thousand to a few tens of thousands, the populations of most modern states exceed one million and China's exceeds one billion. The paramount chief's location may become the state's capital city. Other population centers of states outside the capital may also qualify as true cities, which are lacking in chiefdoms. Cities differ from villages in their monumental public works, palaces of rulers, accumulation of capital from tribute or taxes, and concentration of people other than food producers.
Eariv states had a hereditary leader with a title equivalent to king, like a super paramount chief and exercising an even greater monopoly of information, decision making, and power. Even in democracies today, crucial knowledge is available to only a few individuals, who control the flow of information to the rest of the government and consequently control decisions. For instance, in the Cuban Missile Crisis of 1963, information and discussions that determined whether nuclear war would engulf half a billion people were initially confined by President Kennedy to a ten-member executive committee of the National Security Council that he himself appointed; then he limited final decisions to a four-member group consisting of himself and three of his cabinet ministers.
Central control is more far-reaching, and economic redistribution in the form of tribute (renamed taxes) more extensive, in states than in chiefdoms. Economic specialization is more extreme, to the point where today not even farmers remain self-sufficient. Hence the effect on society is catastrophic when state government collapses, as happened in Britain upon the removal of Roman troops, administrators, and coinage between a.d. 407 and 411. Even the earliest Mesopotamian states exercised centralized control of their economies. Their food was produced by four specialist groups (cereal farmers, herders, fishermen, and orchard and garden growers), from each of which the state took the produce and to each of which it gave out the necessary supplies, tools, and foods other than the type of food that this group produced. The state supplied seeds and plow animals to the cereal farmers, took wool from the herders, exchanged the wool by long-distance trade for metal and other essential raw materials, and paid out food rations to the laborers who maintained the irrigation systems on which the farmers depended.
Many, perhaps most, early states adopted slavery on a much larger scale an did chiefdoms. That was not because chiefdoms were more kindly

2. 8 O •

GUNS,GERMS, AND STEELdisposed toward defeated enemies but because the greater economic specialization of states, with more mass production and more public works provided more uses for slave labor. In addition, the larger scale of state warfare made more captives available.
A chiefdom's one or two levels of administration are greatly multiplied in states, as anyone who has seen an organizational chart of any government knows. Along with the proliferation of vertical levels of bureaucrats, there is also horizontal specialization. Instead of konohiki carrying out every aspect of administration for a Hawaiian district, state governments have several separate departments, each with its own hierarchy, to handle water management, taxes, military draft, and so on. Even small states have more complex bureaucracies than large chiefdoms. For instance, the West African state of Maradi had a central administration with over 130 titled offices.
Internal conflict resolution within states has become increasingly formalized by laws, a judiciary, and police. The laws are often written, because many states (with conspicuous exceptions, such as that of the Incas) have Iiad literate elites, writing having been developed around the same time as the formation of the earliest states in both Mesopotamia and Mesoamerica. In contrast, no early chiefdom not on the verge of statehood developed writing.
Early states had state religions and standardized temples. Many early kings were considered divine and were accorded special treatment in innumerable respects. For example, the Aztec and Inca emperors were both carried about in litters; servants went ahead of the Inca emperor's litter and swept the ground clear; and the Japanese language includes special forms of the pronoun "you" for use only in addressing the emperor. Early kings were themselves the head of the state religion or else had separate high priests. The Mesopotamian temple was the center not only of religion but also of economic redistribution, writing, and crafts technology.
All these features of states carry to an extreme the developments that led from tribes to chiefdoms. In addition, though, states have diverged from chiefdoms m several new directions. The most fundamental such distinction is that states are organized on political and territorial lines, not on the kinship lines that denned bands, tribes, and simple chiefdoms. Furthermore, bands and tribes always, and chiefdoms usually, consist of a single ethnic and linguistic group. States, though—especially so-called empires

FROM

EG ALITARIAN1SM TO KLEPTOCRACY • 2. 8 Iformed by amalgamation or conquest of states—are regularly multiethnic and multilingual. State bureaucrats are not selected mainly on the basis of kinship, as in chiefdoms, but are professionals selected at least partly on the basis of training and ability. In later states, including most today, the leadership often became nonhereditary, and many states abandoned the entire system of formal hereditary classes carried over from chiefdoms.
over the past 13,000 years the predominant trend in human society has been .lie replacement of smaller, less complex units by larger, more corr.olex ones. Obviously, that is no more than an average long-term trend, with innumerable shifts in either direction: 1,000 amalgamations for 999 reversals. We know from our daily newspaper that large units (for instance, the former USSR, Yugoslavia, and Czechoslovakia) can disintegrate inco smaller units, as did Alexander of Macedon's empire over 2,000 years ago. More complex units don't always conquer less complex ones but may succumb to them, as when the Roman and Chinese Empires were overrun by "barbarian" and Mongol chiefdoms, respectively. But the long-term trend has stil! been toward large, complex societies, culminating in states.
Obviously, too, part of the reason for states' triumphs over simpler entities when the two collide is that states usually enjoy an advantage of weaponry and other technology, and a large numerical advantage in population. But there are also two other potential advantages inherent in chiefdoms and states. First, a centralized decision maker has the advantage at concentrating troops and resources. Second, the official religions and patriotic fervor of many states make their troops willing to fight suicidally.
The latter willingness is one so strongly programmed into us citizens of modern states, by our schools and churches and governments, that we forget what a radical break it marks with previous human history. Every state has its slogan urging its citizens to be prepared to die if necessary for the state: Britain's '"For King and Country," Spain's "For Dios y Espana," and so on. Similar sentiments motivated 16th-century Aztec warriors: Ihere is nothing like death in war, nothing like the flowery death so precious to Him note 10 who gives life: for off I see it. my heart yearns for it!"
Such sentiments are unthinkable in bands and tribes. In all the accounts

2. 8 Z •

GUNS,GERMS, AND STEELthat my New Guinea friends have given me of their former tribal wars there has been not a single hint of tribal patriotism, of a suicidal charge or of any other military conduct carrying an accepted risk of being killed. Instead, raids are initiated by ambush or by superior force, so as to minimize at all costs the risk that one might die for one's village. But that attitude severely limits the military options of tribes, compared with state societies. Naturally, what makes patriotic and religious fanatics such dangerous opponents is not the deaths of the fanatics themselves, but their willingness to accept the deaths of a fraction of their number in order to annihilate or crush their infidel enemy. Fanaticism in war, of the type that drove recorded Christian and Islamic conquests, was probably unknown on Earth until chiefdoms and especially states emerged within the last 6,000 years.
rlow did small, noncentralized, kin-based societies evolve into large centralized ones in which most members are not closely related to each other? Having reviewed the stages in this transformation from bands to states, we now ask what impelled societies thus to transform themselves.
At many moments in history, states have arisen independently—or, as cultural anthropologists say, "pristinely," that is, in the absence of any preexisting surrounding states. Pristine state origins took place at least once, possibly many times, on each of the continents except Australia and North America. Prehistoric states included those of Mesopotamia, North China, the Nile and Indus Valleys, Mesoamerica, the Andes, and West Africa. Native states in contact with European states have arisen from chiefdoms repeatedly in the last three centuries in Madagascar, Hawaii, Tahiti, and many parts of Africa. Chiefdoms have arisen pristinely even more often, in all of the same regions and in North America's Southeast and Pacific Northwest, the Amazon, Polynesia, and sub-Saharan Africa. All these origins of complex societies give us a rich database for understanding their development.
Of the many theories addressing the problem of state origins, the simplest denies that there is any problem to solve. Aristotle considered states the natural condition of human society, requiring no explanation. his error was understandable, because all the societies with which he would have been acquainted—Greek societies of the fourth century B.C.—were

FROM

EG ALITARIANISM TO KLEPTOCRACY • 183states. However, we now know that, as of a.d. 1492, much of the world was instead organized into chiefdoms, tribes, or bands. State formation does demand an explanation.
The next theory is the most familiar one. The French philosopher Jean-Tacques Rousseau speculated that states are formed by a social contract, a rational decision reached when people calculated their self-interest, came to the agreement that they would be better off in a state than in simpler societies, and voluntarily did away with their simpler societies. But observation and historical records have failed to uncover a single case of a state's being formed in that ethereal atmosphere of dispassionate farsightedness. Smaller units do not voluntarily abandon their sovereignty and merge into larger units. They do so only by conquest, or under external duress.
A third theory, still popular with some historians and economists, sets out from the undoubted fact that, in both Mesopotamia and North China and Mexico, large-scale irrigation systems began to be constructed around the time that states started to emerge. The theory also notes that any big, complex system for irrigation or hydraulic management requires a centralized bureaucracy to construct and maintain it. The theory then turns an observed rough correlation in time into a postulated chain of cause and effect. Supposedly, Mesopotamians and North Chinese and Mexicans foresaw the advantages that a large-scale irrigation system would bring them, even though there was at the time no such system within thousands of miles (or anywhere on Earth) to illustrate for them those advantages. Those farsighted people chose to merge their inefficient little chiefdoms into a larger state capable of blessing them with large-scale irrigation.
However, this "hydraulic theory" of state formation is subject to the same objections leveled against social contract theories in general. More specifically, it addresses only the final stage in the evolution of complex societies. It says nothing about what drove the progression from bands to tribes to chiefdoms during all the millennia before the prospect of large-scale irrigation loomed up on the horizon. When historical or archaeological dates are examined in detail, they fail to support the view of irrigation as the driving force for state formation. In Mesopotamia, North China, Mexico, and Madagascar, small-scale irrigation systems already existed e ore tne "se of states. Construction of large-scale irrigation systems did not accompany the emergence of states but came only significantly later in ac of those areas. In most of the states formed over the Maya area of

2. 8 4 '

GUNS,GERMS, AND STEELMesoamerica and the Andes, irrigation systems always remained small-scale ones that local communities could build and maintain themselves. Thus, even in those areas where complex systems of hydraulic management did emerge, they were a secondary consequence of states that must have formed for other reasons.
What seems to me to point to a fundamentally correct view of state formation is an undoubted fact of much wider validity than the correlation between irrigation and the formation of some states—namely, that the size of the regional population is the strongest single predictor of societal complexity. As we have seen, bands number a few dozen individuals, tribes a few hundred, chiefdoms a few thousand to a few tens of thousands, and states generally over about 50,000. In addition to that coarse correlation between regional population size and type of society (band, tribe, and so on), there is a finer trend, within each of those categories, between population and societal complexity: for instance, that chiefdoms with large populations prove to be the most centralized, stratified, and complex ones.
These correlations suggest strongly that regional population size or population density or population pressure has something to do with the formation of complex societies. But the correlations do not tell us precisely how population variables function in a chain of cause and effect whose outcome is a complex society. To trace out that chain, let us now remind ourselves how large dense populations themselves arise. Then we can examine why a large but simple society could not maintain itself. With that as background, we shall finally return to the question of how a simpler society actually becomes more complex as the regional population increases.
we have seen that large or dense populations arise only under conditions of food production, or at least under exceptionally productive conditions for hunting-gathering. Some productive hunter-gatherer societies reached the organizational level of chiefdoms, but none reached the level of states: all states nourish their citizens by food production. These considerations, along with the just mentioned correlation between regional population size and societal complexity, have led to a protracted chicken-or-egg debate about the causal relations between food production, population variables, and societal complexity. Is it intensive food production that is the cause, triggering population growth and somehow leading to a com-

FROM

egal1tarianism TO KLEPTOCRACY • Z 8 5lex society? Or are large populations and complex societies instead the cause, somehow leading to intensification of food production?
Posing the question in that either-or form misses the point. Intensified food production and societal complexity stimulate each other, by autoca-talvsis. That is, population growth leads to societal complexity, by mechanisms that we shall discuss, while societal complexity in turn leads to intensified food production and thereby to population growth. Complex centralized societies are uniquely capable of organizing public works (including irrigation systems), long-distance trade (including the importation of m: rals to make better agricultural tools), and activities of different groups of economic specialists (such as feeding herders with farmers' cereal, and transferring the herders' livestock to farmers for use as plow animals}. Al! of these capabilities of centralized societies have fostered intens'fied food production and hence population growth throughout history.
In addition, food production contributes in at least three ways to specific features of complex societies. First, it involves seasonally pulsed inputs of labor. When the harvest has been stored, the farmers' labor becomes available for a centralized political authority to harness—in order to build public works advertising state power (such as the Egyptian pyramids), or to build public works that could feed more mouths (such as Polynesian Hawaii's irrigation systems or fishponds), or to undertake wars of conquest to form larger political entities.
Second, food production may be organized so as to generate stored food surpluses, which permit economic specialization and social stratification. The surpluses can be used to feed all tiers of a complex society: the chiefs, bureaucrats, and other members of the elite; the scribes, craftspeople, and other non-food-producing specialists; and the farmers themselves, during times that they are drafted to construct public works.
Finally, food production permits or requires people to adopt sedentary living, which is a prerequisite for accumulating substantial possessions, developing elaborate technology and crafts, and constructing public works. The importance of fixed residence to a complex society explains wny missionaries and governments, whenever they make first contact with Previously uncontacted nomadic tribes or bands in New Guinea or the mazon, universally have two immediate goals. One goal, of course, is the , . VIOus one o* "pacifying" the nomads: that is, dissuading them from
lng missionaries, bureaucrats, or each other. The other goal is to induce

1 8 6 •

GUNS,GERMS, AND STEELthe nomads to settle in villages, so that the missionaries and bureaucrats can find the nomads, bring them services such as medical care and schools, and proselytize and control them.
Thus, food production, which increases population size, also acts in many ways to make features of complex societies possible. But that doesn't prove that food production and large populations make complex societies inevitable. How can we account for the empirical observation that band or tribal organization just does not work for societies of hundreds of thousands of people, and that all existing large societies have complex centralized organization? We can cite at least four obvious reasons.One reason is the problem of conflict between unrelated strangers. That problem grows astronomically as the number of people making up the society increases. Relationships within a band of 20 people involve only 190 two-person interactions (20 people times 19 divided by 2), but a band of 2,000 would have 1,999,000 dyads. Each of those dyads represents a potential time bomb that could explode in a murderous argument. Each murder in band and tribal societies usually leads to an attempted revenge killing, starting one more unending cycle of murder and countermurder that destabilizes the society.
In a band, where everyone is closely related to everyone else, people related simultaneously to both quarreling parties step in to mediate quarrels. In a tribe, where many people are still close relatives and everyone at least knows everybody else by name, mutual relatives and mutual friends mediate the quarrel. But once the threshold of "several hundred," below which everyone can know everyone else, has been crossed, increasing numbers of dyads become pairs of unrelated strangers. When strangers fight, few people present will be friends or relatives of both combatants, with self-interest in stopping the fight. Instead, many onlookers will be friends or relatives of only one combatant and will side with that person, escalating the two-person fight into a general brawl. Hence a large society that continues to leave conflict resolution to all of its members is guaranteed to blow up. That factor alone would explain why societies of thousands can exist only if they develop centralized authority to monopolize force and resolve conflicts.
A second reason is the growing impossibility of communal decision

FROM

EGALITARIANISM TO KLEPTOCRACY • 2. 8 7making with increasing population size. Decision making by the entire adult population is still possible in New Guinea villages small enough that news and information quickly spread to everyone, that everyone can hear everyone else in a meeting of the whole village, and that everyone who wants to speak at the meeting has the opportunity to do so. But all those prerequisites for communal decision making become unattainable in much larger communities. Even now, in these days of microphones and loudspeakers, we all know that a group meeting is no way to resolve issues for a group of thousands of people. Hence a large society must be structured and centralized if it is to reach decisions effectively.
A third reason involves economic considerations. Any society requires means to transfer goods between its members. One individual may happen to acquire more of some essential commodity on one day and less on another. Because individuals have different talents, one individual consistently tends to wind up with an excess of some essentials and a deficit of others. In small societies with few pairs of members, the resulting necessary transfers of goods can be arranged directly between pairs of individuals or families, by reciprocal exchanges. But the same mathematics that makes direct pairwise conflict resolution inefficient in large societies makes direct pairwise economic transfers also inefficient. Large societies can function economically only if they have a redistributive economy in addition to a reciprocal economy. Goods in excess of an individual's needs must be transferred from the individual to a centralized authority, which then redistributes the goods to individuals with deficits.
A final consideration mandating complex organization for large societies has to do with population densities. Large societies of food producers have not only more members but also higher population densities than do small bands of hunter-gatherers. Each band of a few dozen hunters occupies a large territory, within which they can acquire most of the resources essential to them. They can obtain their remaining necessities by trading with neighboring bands during intervals between band warfare. As population density increases, the territory of that band-sized population of a rew dozen would shrink to a small area, with more and more of life's necessities having to be obtained outside the area. For instance, one couldn't just divide Holland's 16,000 square miles and 16,000,000 people »nto 800,000 individual territories, each encompassing 13 acres and serv-In8 as home to an autonomous band of 20 people who remained self-sufficient confined within their 13 acres, occasionally taking advantage of

Z 8 8 • GUNS, GERMS, AND STEEL

a temporary truce to come to the borders of their tiny territory in order to exchange some trade items and brides with the next band. Such spatial realities require that densely populated regions support large and complexly organized societies.
Considerations of conflict resolution, decision making, economics, and space thus converge in requiring large societies to be centralized. But centralization of power inevitably opens the door—for those who hold the power, are privy to information, make the decisions, and redistribute the goods—to exploit the resulting opportunities to reward themselves and their relatives. To anyone familiar with any modern grouping of people, that's obvious. As early societies developed, those acquiring centralized power gradually established themselves as an elite, perhaps originating as one of several formerly equal-ranked village clans that became "more equal" than the others.
Thosearethe reasons why large societies cannot function with band organization and instead are complex kleptocracies. But we are still left with the question of how small, simple societies actually evolve or amalgamate into large, complex ones. Amalgamation, centralized conflict resolution, decision making, economic redistribution, and kleptocratic religion don't just develop automatically through a Rousseauesque social contract. What drives the amalgamation?
In part, the answer depends upon evolutionary reasoning. I said at the outset of this chapter that societies classified in the same category are not all identical to each other, because humans and human groups are infinitely diverse. For example, among bands and tribes, the big-men of some are inevitably more charismatic, powerful, and skilled in reaching decisions than the big-men of others. Among large tribes, those with stronger big-men and hence .greater centralization tend to have an advantage over those with less centralization. Tribes that resolve conflicts as poorly as did the Fayu tend to blow apart again into bands, while ill-governed chief-doms blow apart into smaller chiefdoms or tribes. Societies with effective conflict resolution, sound decision making, and harmonious economic redistribution can develop better technology, concentrate their military power, seize larger and more productive territories, and crush autonomous smaller societies one by one.
Thus, competition between societies at one level of complexity tends to

FROM

EGALITARIANISM TO KLEPTOCRACY • 189lead to societies on the next level of complexity if conditions permit. Tribes conquer or combine with tribes to reach the size of chiefdoms, which conquer or combine with other chiefdoms to reach the size of states, which conquer or combine with other states to become empires. More generally, large units potentially enjoy an advantage over individual small units if— and that's a big "if"—the large units can solve the problems that come with their larger size, such as perennial threats from upstart claimants to leadership, commoner resentment of kleptocracy, and increased problems associated with economic integration.The amalgamation of smaller units into larger ones has often been documented historically or archaeologically. Contrary to Rousseau, such amalgamations never occur by a process of unthreatened little societies freely deciding to merge, in order to promote the happiness of their citizens. Leaders of little societies, as of big ones, are jealous of their independence and prerogatives. Amalgamation occurs instead in either of two ways: by merger under the threat of external force, or by actual conquest. Innumerable examples are available to illustrate each mode of amalgamation.
Merger under the threat of external force is well illustrated by the formation of the Cherokee Indian confederation in the U.S. Southeast. The Cherokees were originally divided into 30 or 40 independent chiefdoms, each consisting of a village of about 400 people. Increasing white settlement led to conflicts between Cherokees and whites. When individual Cherokees robbed or assaulted white settlers and traders, the whites were unable to discriminate among the different Cherokee chiefdoms and retaliated indiscriminately against any Cherokees, either by military action or by cutting off trade. In response, the Cherokee chiefdoms gradually found themselves compelled to join into a single confederacy in the course of the 18th century. Initially, the larger chiefdoms in 1730 chose an overall leader, a chief named Moytoy, who was succeeded in 1741 by his son. The first task of these leaders was to punish individual Cherokees who attacked whites, and to deal with the white government. Around 1758 the Chero-kees regularized their decision making with an annual council modeled on previous village councils and meeting at one village (Echota), which thereby became a de facto "capital." Eventually, the Cherokees became "terate (as we saw in Chapter 12) and adopted a written constitution.
The Cherokee confederacy was thus formed not by conquest but by the malgamation of previously jealous smaller entities, which merged only

290 •

GUNS,GERMS, AND STEELwhen threatened with destruction by powerful external forces. In much the same way, in an example of state formation described in every American history textbook, the white American colonies themselves, one of which (Georgia) had precipitated the formation of the Cherokee state, were impelled to form a nation of their own when threatened with the powerful external force of the British monarchy. The American colonies were initially as jealous of their autonomy as the Cherokee chiefdoms, and their first attempt at amalgamation under the Articles of Confederation (1781) proved unworkable because it reserved too much autonomy to the ex-colonies. Only further threats, notably Shays's Rebellion of 1786 and the unsolved burden of war debt, overcame the ex-colonies' extreme reluctance to sacrifice autonomy and pushed them into adopting our current strong federal constitution in 1787. The 19th-century unification of Germany's jealous principalities proved equally difficult. Three early attempts (the Frankfurt Parliament of 1848, the restored German Confederation of 1850, and the North German Confederation of 1866) failed before the external threat of France's declaration of war in 1870 finally led to the princelets' surrendering much of their power to a central imperial German government in 1871.
The other mode of formation of complex societies, besides merger under threat of external force, is merger by conquest. A well-documented example is the origin of the Zulu state, in southeastern Africa. When first observed by white settlers, the Zulus were divided into dozens of little chiefdoms. During the late 1700s, as population pressure rose, fighting between the chiefdoms became increasingly intense. Among all those chiefdoms, the ubiquitous problem of devising centralized power structures was solved most successfully by a chief called Dingiswayo, who gained ascendancy of the Mtetwa chiefdom by killing a rival around 1807. Dingiswayo developed a superior centralized military organization by drafting young men from all villages and grouping them into regiments by age rather than by their village. He also developed superior centralized political organization by abstaining from slaughter as he conquered other chiefdoms, leaving the conquered chief's family intact, and limiting himself to replacing the conquered chief himself with a relative willing to cooperate with Dingiswayo. He developed superior centralized conflict resolution by expanding the adjudication of quarrels. In that way Dingiswayo was able to conquer and begin the integration of 30 other Zulu chiefdoms. His sue-

FROM

EGALITARIANISM TO KLEPTOCRACY • 2. 9 Icessors strengthened the resulting embryonic Zulu state by expanding its judicial system, policing, and ceremonies.
This Zulu example of a state formed by conquest can be multiplied almost indefinitely. Native states whose formation from chiefdoms happened to be witnessed by Europeans in the 18th and 19th centuries include the Polynesian Hawaiian state, the Polynesian Tahitian state, the Merina state of Madagascar, Lesotho and Swazi and other southern African states besides that of the Zulus, the Ashanti state of West Africa, and the Ankole and Buganda states of Uganda. The Aztec and Inca Empires were formed by 15th-century conquests, before Europeans arrived, but we know much about their formation from Indian oral histories transcribed by early Spanish settlers. The formation of the Roman state and the expansion of the Macedonian Empire under Alexander were described in detail by contemporary classical authors.
All these examples illustrate that wars, or threats of war, have played a key role in most, if not all, amalgamations of societies. But wars, even between mere bands, have been a constant fact of human history. Why is it, then, that they evidently began causing amalgamations of societies only within the past 13,000 years? We had already concluded that the formation of complex societies is somehow linked to population pressure, so we should now seek a link between population pressure and the outcome of war. Why should wars tend to cause amalgamations of societies when populations are dense but not when they are sparse? The answer is that the fate of defeated peoples depends on population density, with three possible outcomes:
Where population densities are very low, as is usual in regions occupied by hunter-gatherer bands, survivors of a defeated group need only move farther away from their enemies. That tends to be the result of wars between nomadic bands in New Guinea and the Amazon.
Where population densities are moderate, as in regions occupied by rood-producing tribes, no large vacant areas remain to which survivors of a defeated band can flee. But tribal societies without intensive food production have no employment for slaves and do not produce large enough food surpluses to be able to yield much tribute. Hence the victors have no use tor survivors of a defeated tribe, unless to take the women in marriage. The defeated men are killed, and their territory may be occupied by the victors.

2 9 i *

GUNS, GERMS, AND STEELWhere population densities are high, as in regions occupied by states or chiefdoms, the defeated still have nowhere to flee, but the victors now have two options for exploiting them while leaving them alive. Because chiefdoms and state societies have economic specialization, the defeated can be used as slaves, as commonly happened in biblical times. Alternatively, because many such societies have intensive food production systems capable of yielding large surpluses, the victors can leave the defeated in place but deprive them of political autonomy, make them pay regular tribute in food or goods, and amalgamate their society into the victorious state or chiefdom. This has been the usual outcome of battles associated with the founding of states or empires throughout recorded history. For example, the Spanish conquistadores wished to exact tribute from Mexico's defeated native populations, so they were very interested in the Aztec Empire's tribute lists. It turned out that the tribute received by the Aztecs each year from subject peoples had included 7,000 tons of corn, 4,000 tons of beans, 4,000 tons of grain amaranth, 2,000,000 cotton cloaks, and huge quantities of cacao beans, war costumes, shields, feather headdresses, and amber.
Thus, food production, and competition and diffusion between societies, led as ultimate causes, via chains of causation that differed in detail but that all involved large dense populations and sedentary living, to the proximate agents of conquest: germs, writing, technology, and centralized political organization. Because those ultimate causes developed differently on different continents, so did those agents of conquest. Hence those agents tended to arise in association with each other, but the association was not strict: for example, an empire arose without writing among the Incas, and writing with few epidemic diseases among the Aztecs. Dingis-wayo's Zulus illustrate that each of those agents contributed somewhat independently to history's pattern. Among the dozens of Zulu chiefdoms, the Mtetwa chiefdom enjoyed no advantage whatsoever of technology, writing, or germs over the other chiefdoms, which it nevertheless succeeded in defeating. Its advantage lay solely in the spheres of government and ideology. The resulting Zulu state was thereby enabled to conquer a fraction of a continent for nearly a century.

PART FOUR

AROUND THE
WORLD IN FIVE
CHAPTERS

CHAPTER 15

yali’s people

WHEN MY WIFE, MARIE, AND I WERE VACATIONING IN Australia one summer, we decided to visit a site with well-preserved Aboriginal rock paintings in the desert near the town of Men-indee. While I knew of the Australian desert's reputation for dryness and summer heat, I had already spent long periods working under hot, dry conditions in the Californian desert and New Guinea savanna, so I considered myself experienced enough to deal with the minor challenges we would face as tourists in Australia. Carrying plenty of drinking water, Marie and I set off at noon on a hike of a few miles to the paintings.

The trail from the ranger station led uphill, under a cloudless sky, through open terrain offering no shade whatsoever. The hot, dry air that we were breathing reminded me of how it had felt to breathe while sitting in a Finnish sauna. By the time we reached the cliff site with the paintings, we had finished our water. We had also lost our interest in art, so. we pushed on uphill, breathing slowly and regularly. Presently I noticed a bird that was unmistakably a species of babbler, but it seemed enormous compared with any known babbler species. At that point, I realized that I was experiencing heat hallucinations for the first time in my life. Marie and I decided that we had better head straight back.

2. 9 6 • GUNS, GERMS, AND STEEL

Both of us stopped talking. As we walked, we concentrated on listening to our breathing, calculating the distance to the next landmark, and estimating the remaining time. My mouth and tongue were now dry, and Marie's face was red. When we at last reached the air-conditioned ranger station, we sagged into chairs next to the water cooler, drank down the cooler's last half-gallon of water, and asked the ranger for another bottle. Sitting there exhausted, both physically and emotionally, I reflected that the Aborigines who had made those paintings had somehow spent their entire lives in that desert without air-conditioned retreats, managing to find food as well as water.
To white Australians, Menindee is famous as the base camp for two whites who had suffered worse from the desert's dry heat over a century earlier: the Irish policeman Robert Burke and the English astronomer William Wills, ill-fated leaders of the first European expedition to cross Australia from south to north. Setting out with six camels packing food enough for three months, Burke and Wills ran out of provisions while in the desert north of Menindee. Three successive times, they encountered and were rescued by well-fed Aborigines whose home was that desert, and who plied the explorers with fish, fern cakes, and roasted fat rats. But then Burke foolishly shot his pistol at one of the Aborigines, whereupon the whole group fled. Despite their big advantage over the Aborigines in possessing guns with which to hunt, Burke and Wills starved, collapsed, and died within a month after the Aborigines' departure.
My wife's and my experience at Menindee, and the fate of Burke and Wills, made vivid for me the difficulties of building a human society in Australia. Australia stands out from all the other continents: the differences between Eurasia, Africa, North America, and South America fade into insignificance compared with the differences between Australia and any of those other landmasses. Australia is by far the driest, smallest, flattest, most infertile, climatically most unpredictable, and biologically most impoverished continent. It was the last continent to be occupied by Europeans. Until then, it had supported the most distinctive human societies, and the least numerous human population, of any continent.
Australia thus provides a crucial test of theories about intercontinental differences in societies. It had the most distinctive environment, and also the most distinctive societies. Did the former cause the latter? If so, how? Australia is the logical continent with which to begin our around-the-

YALI'S

PEOPLE • i 9 7world tour, applying the lessons of Parts 2 and 3 to understanding the differing histories of all the continents.
most lay people would describe as the most salient feature of Native Australian societies their seeming "backwardness." Australia is the sole continent where, in modern times, all native peoples still lived without any of the hallmarks of so-called civilization—without farming, herding, metal, bows and arrows, substantial buildings, settled villages, writing, chiefdorm, or states. Instead, Australian Aborigines were nomadic or seminomadic hunter-gatherers, organized into bands, living in temporary shelters or huts, and still dependent on stone tools. During the last 13,000 years less cultural change has accumulated in Australia than in any other continent. The prevalent European view of Native Australians was already typified by the words of an early French explorer, who wrote, "They are the most miserable people of the world, and the human beings who approach closest to brute beasts."
Yet, as of 40,000 years ago, Native Australian societies enjoyed a big head start over societies of Europe and the other continents. Native Australians developed some of the earliest known stone tools with ground edges, the earliest hafted stone tools (that is, stone ax heads mounted on handles), and by far the earliest watercraft, in the world. Some of the oldest known painting on rock surfaces comes from Australia. Anatomically modern humans may have settled Australia before they settled western Europe. Why, despite that head start, did Europeans end up conquering Australia, rather than vice versa?
Within that question lies another. During the Pleistocene Ice Ages, when much ocean water was sequestered in continental ice sheets and sea level dropped far below its present stand, the shallow Arafura Sea now separating Australia from New Guinea was low, dry land. With the melting of ice sheets between around 12,000 and 8,000 years ago, sea level rose, that low land became flooded, and the former continent of Greater Australia became sundered into the two hemi-continents of Australia and New Guinea (Figure 15.1 on page 299).
The human societies of those two formerly joined landmasses were in modern times very different from each other. In contrast to everything that just said about Native Australians, most New Guineans, such as Yali's

198 ' GUNS, GERMS,

and steelpeople, were farmers and swineherds. They lived in settled villages and were organized politically into tribes rather than bands. All New Guineans had bows and arrows, and many used pottery. New Guineans tended to have much more substantial dwellings, more seaworthy boats, and more numerous and more varied utensils than did Australians. As a consequence of being food producers instead of hunter-gatherers, New Guineans lived at much higher average population densities than Australians: New Guinea has only one-tenth of Australia's area but supported a native population several times that of Australia's.
Why did the human societies of the larger landmass derived from Pleistocene Greater Australia remain so "backward" in their development, while the societies of the smaller landmass "advanced" much more rapidly? Why didn't all those New Guinea innovations spread to Australia, which is separated from New Guinea by only 90 miles of sea at Torres Strait? From the perspective of cultural anthropology, the geographic distance between Australia and New Guinea is even less than 90 miles, because Torres Strait is sprinkled with islands inhabited by farmers using bows and arrows and culturally resembling New Guineans. The largest Torres Strait island lies only 10 miles from Australia. Islanders carried on a lively trade with Native Australians as well as with New Guineans. How could two such different cultural universes maintain themselves across a calm strait only 10 miles wide and routinely traversed by canoes?
Compared with Native Australians, New Guineans rate as culturally "advanced." But most other modern people consider even New Guineans "backward." Until Europeans began to colonize New Guinea in the late 19th century, all New Guineans were nonliterate, dependent on stone tools, and politically not yet organized into states or (with few exceptions) chiefdoms. Granted that New Guineans had "progressed" beyond Native Australians, why had they not yet "progressed" as far as many Eurasians, Africans, and Native Americans? Thus, Yali's people and their Australian cousins pose a puzzle inside a puzzle.
When asked to account for the cultural "backwardness" of Aboriginal Australian society, many white Australians have a simple answer: supposed deficiencies of the Aborigines themselves. In facial structure and skin color, Aborigines certainly look different from Europeans, leading some late-19th century authors to consider them a missing link between apes and humans. How else can one account for the fact that white English colonists created a literate, food-producing, industrial democracy, within

Figure 15.1. Map of the region from Southeast Asia to Australia and

ew Guinea. Solid lines denote the present coastline; the dashed lines are e coastline during Pleistocene times when sea level dropped to below ^ Present stand-that is, the edge of the Asian and Greater Australianext>VeSj *,* ^^ time' NeW Guinea and Australia we Joined in anPonded Greater Australia, while Borneo, Java, Sumatra, and Taiwanere part of the Asian mainland.

3 O O • GUNS, GERMS, AND STEEL

a few decades of colonizing a continent whose inhabitants after more than 40,000 years were still nonliterate hunter-gatherers? It is especially striking that Australia has some of the world's richest iron and aluminum deposits, as well as rich reserves of copper, tin, lead, and zinc. Why, then, were Native Australians still ignorant of metal tools and living in the Stone Age?
It seems like a perfectly controlled experiment in the evolution of human societies. The continent was the same; only the people were different. Ergo, the explanation for the differences between Native Australian and European-Australian societies must lie in the different people composing them. The logic behind this racist conclusion appears compelling. We shall see, however, that it contains a simple error.
As the first step in examining this logic, let us examine the origins of the peoples themselves. Australia and New Guinea were both occupied by at least 40,000 years ago, at a time when they were both still joined as Greater Australia. A glance at a map (Figure 15.1) suggests that the colonists must have originated ultimately from the nearest continent, Southeast Asia, by island hopping through the Indonesian Archipelago. This conclusion is supported by genetic relationships between modern Australians, New Guineans, and Asians, and by the survival today of a few populations of somewhat similar physical appearance in the Philippines, Malay Peninsula, and Andaman Islands off Myanmar.
Once the colonists had reached the shores of Greater Australia, they spread quickly over the whole continent to occupy even its farthest reaches and most inhospitable habitats. By 40,000 years ago, fossils and stone tools attest to their presence in Australia's southwestern corner; by 35,000 years ago, in Australia's southeastern corner and Tasmania, the corner of Australia most remote from the colonists' likely beachhead in western Australia or New Guinea (the parts nearest Indonesia and Asia); and by 30,000 years ago, in the cold New Guinea highlands. All of those areas could have been reached overland from a western beachhead. However, the colonization of both the Bismarck and the Solomon Archipelagoes northeast of New Guinea, by 35,000 years ago, required further overwater crossings of dozens of miles. The occupation could have been even more rapid than that apparent spread of dates from 40,000 to 30,000 years ago,

YALI'S

PEOPLE "301since the various dates hardly differ within the experimental error of the radiocarbon method.
At the Pleistocene times when Australia and New Guinea were initially occupied, the Asian continent extended eastward to incorporate the modern islands of Borneo, Java, and Bali, nearly 1,000 miles nearer to Australia and New Guinea than Southeast Asia's present margin. However, at least eight channels up to 50 miles wide still remained to be crossed in getting from Borneo or Bali to Pleistocene Greater Australia. Forty thousand years ago, those crossings may have been achieved by bamboo rafts, low-tech but seaworthy watercraft still in use in coastal South China today. The crossings must nevertheless have been difficult, because after that initial landfall by 40,000 years ago the archaeological record provides no compelling evidence of further human arrivals in Greater Australia from Asia for tens of thousands of years. Not until within the last few thousand years do we encounter the next firm evidence, in the form of the appearance of Asian-derived pigs in New Guinea and Asian-derived dogs in Australia.
Thus, the human societies of Australia and New Guinea developed in substantial isolation from the Asian societies that founded them. That isolation is reflected in languages spoken today. After all those millennia of isolation, neither modern Aboriginal Australian languages nor the major group of modern New Guinea languages (the so-called Papuan languages) exhibit any clear relationships with any modern Asian languages.
The isolation is also reflected in genes and physical anthropology. Genetic studies suggest that Aboriginal Australians and New Guinea high-landers are somewhat more similar to modern Asians than to peoples of other continents, but the relationship is not a close one. In skeletons and physical appearance, Aboriginal Australians and New Guineans are also distinct from most Southeast Asian populations, as becomes obvious if one compares photos of Australians or New Guineans with those of Indonesians or Chinese. Part of the reason for all these differences is that the initial Asian colonists of Greater Australia have had a long time in which to diverge from their stay-at-home Asian cousins, with only limited genetic exchanges during most of that time. But probably a more important reason is that the original Southeast Asian stock from which the colonists of
reater Australia were derived has by now been largely replaced by other Asians expanding out of China.

302.

• GUNS, GERMS, AND STEELAboriginal Australians and New Gumeans have also diverged genetically, physically, and linguistically from each other. For instance, among the major (genetically determined) human blood groups, groups B of the so-called ABO system and S of the MNS system occur in New Guinea as well as in most of the rest of the world, but both are virtually absent in Australia. The tightly coiled hair of most New Guineans contrasts with the straight or wavy hair of most Australians. Australian languages and New Guinea's Papuan languages are unrelated not only to Asian languages but also to each other, except for some spread of vocabulary in both directions across Torres Strait.
All that divergence of Australians and New Guineans from each other reflects lengthy isolation in very different environments. Since the rise of the Arafura Sea finally separated Australia and New Guinea from each other around 10,000 years ago, gene exchange has been limited to tenuous contact via the chain of Torres Strait islands. That has allowed the populations of the two hemi-continents to adapt to their own environments. While the savannas and mangroves of coastal southern New Guinea are fairly similar to those of northern Australia, other habitats of the hemi-continents differ in almost all major respects.
Here are some of the differences. New Guinea lies nearly on the equator, while Australia extends far into the temperate zones, reaching almost 40 degrees south of the equator. New Guinea is mountainous and extremely rugged, rising to 16,500 feet and with glaciers capping the highest peaks, while Australia is mostly low and flat—94 percent of its area lies below 2,000 feet of elevation. New Guinea is one of the wettest areas on Earth, Australia one of the driest. Most of New Guinea receives over 100 inches of rain annually, and much of the highlands receives over 200 inches, while most of Australia receives less than 20 inches. New Guinea's equatorial climate varies only modestly from season to season and year to year, but Australia's climate is highly seasonal and varies from year to year far more than that of any other continent. As a result, New Guinea is laced with permanent large rivers, while Australia's permanently flowing rivers are confined in most years to eastern Australia, and even Australia's largest river system (the Murray-Darling) has ceased flowing for months during droughts. Must of New Guinea's land area is clothed in dense rain forest, while most of Australia's supports only desert and open dry woodland.
New Guinea is covered with young fertile soil, as a consequence of volcanic activity, glaciers repeatedly advancing and retreating and scouring

YALI'S

PEOPLE • 303the highlands, and mountain streams carrying huge quantities of silt to the lowlands. In contrast, Australia has by far the oldest, most infertile, most nutrient-leached soils of any continent, because of Australia's little volcanic activity and its lack of high mountains and glaciers. Despite having only one-tenth of Australia's area, New Guinea is home to approximately as many mammal and bird species as is Australia—a result of New Guinea's equatorial location, much higher rainfall, much greater range of elevations, and greater fertility. All of those environmental differences influenced the two hemi-continents' very disparate cultural histories, which we shall now consider.
theearliest and most intensive food production, and the densest populations, of Greater Australia arose in the highland valleys of New Guinea at altitudes between 4,000 and 9,000 feet above sea level. Archaeological excavations uncovered complex systems of drainage ditches dating back to 9,000 years ago and becoming extensive by 6,000 years ago, as well as terraces serving to retain soil moisture in drier areas. The ditch systems were similar to those still used today in the highlands to drain swampy areas for use as gardens. By around 5,000 years ago, pollen analyses testify to widespread deforestation of highland valleys, suggesting forest clearance for agriculture.
Today, the staple crops of highland agriculture are the recently introduced sweet potato, along with taro, bananas, yams, sugarcane, edible grass stems, and several leafy vegetables. Because taro, bananas, and yams are native to Southeast Asia, an undoubted site of plant domestication, it used to be assumed that New Guinea highland crops other than sweet potatoes arrived from Asia. However, it was eventually realized that the wild ancestors of sugarcane, the leafy vegetables, and the edible grass stems are New Guinea species, that the particular types of bananas grown in New Guinea have New Guinea rather than Asian wild ancestors, and that taro and some yams are native to New Guinea as well as to Asia. If New Guinea agriculture had really had Asian origins, one might have expected to find highland crops derived unequivocally from Asia, but there are none. For those reasons it is now generally acknowledged that agriculture arose indigenously in the New Guinea highlands by domestication of New Guinea wild plant species.
New Guinea thus joins the Fertile Crescent, China, and a few other

304 " GUNS, GERMS, AND STEEL

regions as one of the world's centers of independent origins of plant domestication. No remains of the crops actually being grown in the highlands 6,000 years ago have been preserved in archaeological sites. However, that is not surprising, because modern highland staple crops are plant species that do not leave archaeologically visible residues except under exceptional conditions. Hence it seems likely that some of them were also the founding crops of highland agriculture, especially as the ancient drainage systems preserved are so similar to the modern drainage systems used for growing taro.
The three unequivocally foreign elements in New Guinea highland food production as seen by the first European explorers were chickens, pigs, and sweet potatoes. Chickens and pigs were domesticated in Southeast Asia and introduced around 3,600 years ago to New Guinea and most other Pacific islands by Austronesians, a people of ultimately South Chinese origin whom we shall discuss in Chapter 17. (Pigs may have arrived earlier.) As for the sweet potato, native to South America, it apparently reached New Guinea only within the last few centuries, following its introduction to the Philippines by Spaniards. Once established in New Guinea, the sweet potato overtook taro as the highland's leading crop, because of its shorter time required to reach maturity, higher yields per acre, and greater tolerance of poor soil conditions.
The development of New Guinea highland agriculture must have triggered a big population explosion thousands of years ago, because the highlands could have supported only very low population densities of hunter-gatherers after New Guinea's original megafauna of giant marsupials had been exterminated. The arrival of the sweet potato triggered a further explosion in recent centuries. When Europeans first flew over the highlands in the 1930s, they were astonished to see below them a landscape similar to Holland's. Broad valleys were completely deforested and dotted with villages, and drained and fenced fields for intensive food production covered entire valley floors. That landscape testifies to the population densities achieved in the highlands by farmers with stone tools.
Steep terrain, persistent cloud cover, malaria, and risk of drought at lower elevations confine New Guinea highland agriculture to elevations above about 4,000 feet. In effect, the New Guinea highlands are an island of dense farming populations thrust up into the sky and surrounded below by a sea of clouds. Lowland New Guineans on the seacoast and rivers are villagers depending heavily on fish, while those on dry ground away from

YALI'S

PEOPLE • 305the coast and rivers subsist at low densities by slash-and-burn agriculture based on bananas and yams, supplemented by hunting and gathering. In contrast, lowland New Guinea swamp dwellers live as nomadic hunter-gatherers dependent on the starchy pith of wild sago palms, which are very productive and yield three times more calories per hour of work than does gardening. New Guinea swamps thus provide a clear instance of an environment where people remained hunter-gatherers because farming could not compete with the hunting-gathering lifestyle.
The sago eaters persisting in lowland swamps exemplify the nomadic hunter-gatherer band organization that must formerly have characterized all New Guineans. For all the reasons that we discussed in Chapters 13 and 14, the farmers and the fishing peoples were the ones to develop more-complex technology, societies, and political organization. They live in permanent villages and tribal societies, often led by a big-man. Some of them construct large, elaborately decorated, ceremonial houses. Their great art, in the form of wooden statues and masks, is prized in museums around the world.
New guinea thus became the part of Greater Australia with the most-advanced technology, social and political organization, and art. However, from an urban American or European perspective, New Guinea still rates as "primitive" rather than "advanced." Why did New Guineans continue to use stone tools instead of developing metal tools, remain non-literate, and fail to organize themselves into chiefdoms and states? It turns out that New Guinea had several biological and geographic strikes against it.
First, although indigenous food production did arise in the New Guinea highlands, we saw in Chapter 8 that it yielded little protein. The dietary staples were low-protein root crops, and production of the sole domesticated animal species (pigs and chickens) was too low to contribute much to people's protein budgets. Since neither pigs nor chickens can be harnessed to pull carts, highlanders remained without sources of power other than human muscle power, and also failed to evolve epidemic diseases to repel the eventual European invaders.
A second restriction on the size of highland populations was the limited available area: the New Guinea highlands have only a few broad valleys, notably the Wahgi and Baliem Valleys, capable of supporting dense popu-

3 O 6 •

GUNS,GERMS, AND STEELlations. Still a third limitation was the reality that the mid-montane zone between 4,000 and 9,000 feet was the sole altitudinal zone in New Guinea suitable for intensive food production. There was no food production at all in New Guinea alpine habitats above 9,000 feet, little on the hillslopes between 4,000 and 1,000 feet, and only low-density slash-and-burn agriculture in the lowlands. Thus, large-scale economic exchanges of food, between communities at different altitudes specializing in different types of food production, never developed in New Guinea. Such exchanges in the Andes, Alps, and Himalayas not only increased population densities in those areas, by providing people at all altitudes with a more balanced diet, but also promoted regional economic and political integration.
For all these reasons, the population of traditional New Guinea never exceeded 1,000,000 until European colonial governments brought Western medicine and the end of intertribal warfare. Of the approximately nine world centers of agricultural origins that we discussed in Chapter 5, New Guinea remained the one with by far the smallest population. With a mere 1,000,000 people, New Guinea could not develop the technology, writing, and political systems that arose among populations of tens of millions in China, the Fertile Crescent, the Andes, and Mesoamerica.
New Guinea's population is not only small in aggregate, but also fragmented into thousands of micropopulations by the rugged terrain: swamps in much of the lowlands, steep-sided ridges and narrow canyons alternating with each other in the highlands, and dense jungle swathing both the lowlands and the highlands. When I am engaged in biological exploration in New Guinea, with teams of New Guineans as field assistants, I consider excellent progress to be three miles per day even if we are traveling over existing trails. Most highlanders in traditional New Guinea never went more than 10 miles from home in the course of their lives.
Those difficulties of terrain, combined with the state of intermittent warfare that characterized relations between New Guinea bands or villages, account for traditional New Guinea's linguistic, cultural, and political fragmentation. New Guinea has by far the highest concentration of languages in the world: 1,000 out of the world's 6,000 languages, crammed into an area only slightly larger than that of Texas, and divided into dozens of language families and isolated languages as different from each other as English is from Chinese. Nearly half of all New Guinea languages have fewer than 500 speakers, and even the largest language groups (still with a mere 100,000 speakers) were politically fragmented into hun-

YALI'S

PEOPLE • 307dreds of villages, fighting as fiercely with each other as with speakers of other languages. Each of those microsocieties alone was far too small to support chiefs and craft specialists, or to develop metallurgy and writing.
Besides a small and fragmented population, the other limitation on development in New Guinea was geographic isolation, restricting the inflow of technology and ideas from elsewhere. New Guinea's three neighbors were all separated from New Guinea by water gaps, and until a few thousand years ago they were all even less advanced than New Guinea (especially the New Guinea highlands) in technology and food production. Of those three neighbors, Aboriginal Australians remained hunter-gatherers with almost nothing to offer New Guineans that New Guineans did not already possess. New Guinea's second neighbor was the much smaller islands of the Bismarck and the Solomon Archipelagoes to the east. That left, as New Guinea's third neighbor, the islands of eastern Indonesia. But that area, too, remained a cultural backwater occupied by hunter-gatherers for most of its history. There is no item that can be identified as having reached New Guinea via Indonesia, after the initial colonization of New Guinea over 40,000 years ago, until the time of the Austronesian expansion around 1600 B.C.
With that expansion, Indonesia became occupied by food producers of Asian origins, with domestic animals, with agriculture and technology at least as complex as New Guinea's, and with navigational skills that served as a much more efficient conduit from Asia to New Guinea. Austronesians settled on islands west and north and east of New Guinea, and in the far west and on the north and southeast coasts of New Guinea itself. Austronesians introduced pottery, chickens, and probably dogs and pigs to New Guinea. (Early archaeological surveys claimed pig bones in the New Guinea highlands by 4000 b.c., but those claims have not been confirmed.) For at least the last thousand years, trade connected New Guinea to the technologically much more advanced societies of Java and China. In return for exporting bird of paradise plumes and spices, New Guineans received Southeast Asian goods, including even such luxury items as Dong Son bronze drums and Chinese porcelain.
With time, the Austronesian expansion would surely have had more impact on New Guinea. Western New Guinea would eventually have been incorporated politically into the sultanates of eastern Indonesia, and metal tools might have spread through eastern Indonesia to New Guinea. But— that hadn't happened by A.D. 1511, the year the Portuguese arrived in the

3 O 8 • GUNS, GERMS,

and steelMoluccas and truncated Indonesia's separate train of developments. When Europeans reached New Guinea soon thereafter, its inhabitants were still living in bands or in fiercely independent little villages, and still using stone tools.
while the new Guinea hemi-continent of Greater Australia thus developed both animal husbandry and agriculture, the Australian hemi-continent developed neither. During the Ice Ages Australia had supported even more big marsupials than New Guinea, including diprotodonts (the marsupial equivalent of cows and rhinoceroses), giant kangaroos, and giant wombats. But all those marsupial candidates for animal husbandry disappeared in the wave of extinctions (or exterminations) that accompanied human colonization of Australia. That left Australia, like New Guinea, with no domesticable native mammals. The sole foreign domesticated mammal adopted in Australia was the dog, which arrived from Asia (presumably in Austronesian canoes) around 1500 b.c. and established itself in the wild in Australia to become the dingo. Native Australians kept captive dingos as companions, watchdogs, and even as living blankets, giving rise to the expression "five-dog night" to mean a very cold night. But they did not use dingos / dogs for food, as did Polynesians, or for cooperative hunting of wild animals, as did New Guineans.
Agriculture was another nonstarter in Australia, which is not only the driest continent but also the one with the most infertile soils. In addition, Australia is unique in that the overwhelming influence on climate over most of the continent is an irregular nonannual cycle, the ENSO (acronym for El Nifio Southern Oscillation), rather than the regular annual cycle of the seasons so familiar in most other parts of the world. Unpredictable severe droughts last for years, punctuated by equally unpredictable torrential rains and floods. Even today, with Eurasian crops and with trucks and railroads to transport produce, food production in Australia remains a risky business. Herds build up in good years, only to be killed off by drought. Any incipient farmers in Aboriginal Australia would have faced similar cycles in their own populations. If in good years they had settled in villages, grown crops, and produced babies, those large populations would have starved and died off in drought years, when the land could support far fewer people.
The other major obstacle to the development of food production in

YALI'S

PEOPLE • 309Australia was the paucity of domesticable wild plants. Even modern European plant geneticists have failed to develop any crop except macadamia nuts from Australia's native wild flora. The list of the world's potential prize cereals—the 56 wild grass species with the heaviest grains—includes only two Australian species, both of which rank near the bottom of the list (grain weight only 13 milligrams, compared with a whopping 40 milligrams for the heaviest grains elsewhere in the world). That's not to say that Australia had no potential crops at all, or that Aboriginal Australians would never have developed indigenous food production. Some plants, such as certain species of yams, taro, and arrowroot, are cultivated in southern New Guinea but also grow wild in northern Australia and were gathered by Aborigines there. As we shall see, Aborigines in the climatically most favorable areas of Australia were evolving in a direction that might have eventuated in food production. But any food production that did arise indigenously in Australia would have been limited by the lack of domesticable animals, the poverty of domesticable plants, and the difficult soils and climate.
Nomadism, the hunter-gatherer lifestyle, and minimal investment in shelter and possessions were sensible adaptations to Australia's ENSO-driven resource unpredictability. When local conditions deteriorated, Aborigines simply moved to an area where conditions were temporarily better. Rather than depending on just a few crops that could fail, they minimized risk by developing an economy based on a great variety of wild foods, not all of which were likely to fail simultaneously. Instead of having fluctuating populations that periodically outran their resources and starved, they maintained smaller populations that enjoyed an abundance of food in good years and a sufficiency in bad years.
The Aboriginal Australian substitute for food production has been termed "firestick farming." The Aborigines modified and managed the surrounding landscape in ways that increased its production of edible plants and animals, without resorting to cultivation. In particular, they intentionally burned much of the landscape periodically. That served several purposes: the fires drove out animals that could be killed and eaten immediately; fires converted dense thickets into open parkland in which people could travel more easily; the parkland was also an ideal habitat for kangaroos, Australia's prime game animal; and the fires stimulated the growth both of new grass on which kangaroos fed and of fern roots on which Aborigines themselves fed.

310

• GUNS, GERMS, AND STEELWe think of Australian Aborigines as desert people, but most of them were not. Instead, their population densities varied with rainfall (because it controls the production of terrestrial wild plant and animal foods) and with abundance of aquatic foods in the sea, rivers, and lakes. The highest population densities of Aborigines were in Australia's wettest and most productive regions: the Murray-Darling river system of the Southeast, the eastern and northern coasts, and the southwestern corner. Those areas also came to support the densest populations of European settlers in modern Australia. The reason we think of Aborigines as desert people is simply that Europeans killed or drove them out of the most desirable areas, leaving the last intact Aboriginal populations only in areas that Europeans didn't want.
Within the last 5,000 years, some of those productive regions witnessed an intensification of Aboriginal food-gathering methods, and a buildup of Aboriginal population density. Techniques were developed in eastern Australia for rendering abundant and starchy, but extremely poisonous, cycad seeds edible, by leaching out or fermenting the poison. The previously unexploited highlands of southeastern Australia began to be visited regularly during the summer, by Aborigines feasting not only on cycad nuts and yams but also on huge hibernating aggregations of a migratory moth called the bogong moth, which tastes like a roasted chestnut when grilled. Another type of intensified food-gathering activity that developed was the freshwater eel fisheries of the Murray-Darling river system, where water levels in marshes fluctuate with seasonal rains. Native Australians constructed elaborate systems of canals up to a mile and a half long, in order to enable eels to extend their range from one marsh to another. Eels were caught by equally elaborate weirs, traps set in dead-end side canals, and stone walls across canals with a net placed in an opening of the wall. Traps at different levels in the marsh came into operation as the water. level rose and fell. While the initial construction of those "fish farms" must have involved a lot of work, they then fed many people. Nineteenth-century European observers found villages of a dozen Aboriginal houses at the eel farms, and there are archaeological remains of villages of up to 146 stone houses, implying at least seasonally resident populations of hundreds of people.
Still another development in eastern and northern Australia was the harvesting of seeds of a wild millet, belonging to the same genus as the broomcorn millet that was a staple of early Chinese agriculture. The millet

YALI'S

PEOPLE • 311was reaped with stone knives, piled into haystacks, and threshed to obtain the seeds, which were then stored in skin bags or wooden dishes and finally ground with millstones. Several of the tools used in this process, such as the stone reaping knives and grindstones, were similar to the tools independently invented in the Fertile Crescent for processing seeds of other wild grasses. Of all the food-acquiring methods of Aboriginal Australians, millet harvesting is perhaps the one most likely to have evolved eventually into crop production.
Along with intensified food gathering in the last 5,000 years came new types of tools. Small stone blades and points provided more length of sharp edge per pound of tool than the large stone tools they replaced. Hatchets with ground stone edges, once present only locally in Australia, became widespread. Shell fishhooks appeared within the last thousand years.
why did australia not develop metal tools, writing, and politically complex societies? A major reason is that Aborigines remained hunter-gatherers, whereas, as we saw in Chapters 12-14, those developments arose elsewhere only in populous and economically specialized societies of food producers. In addition, Australia's aridity, infertility, and climatic unpredictability limited its hunter-gatherer population to only a few hundred thousand people. Compared with the tens of millions of people in ancient China or Mesoamerica, that meant that Australia had far fewer potential inventors, and far fewer societies to experiment with adopting innovations. Nor were its several hundred thousand people organized into closely interacting societies. Aboriginal Australia instead consisted of a sea of very sparsely populated desert separating several more productive ecological "islands," each of them holding only a fraction of the continent's population and with interactions attenuated by the intervening distance. Even within the relatively moist and productive eastern side of the continent, exchanges between societies were limited by the 1,900 miles from Queensland's tropical rain forests in the northeast to Victoria's temperate rain forests in the southeast, a geographic and ecological distance as great as that from Los Angeles to Alaska.
Some apparent regional or continentwide regressions of technology in Australia may stem from the isolation and relatively few inhabitants of its population centers. The boomerang, that quintessential Australian

312 • GUNS, GERMS, AND STEEL

weapon, was abandoned in the Cape York Peninsula of northeastern Australia. When encountered by Europeans, the Aborigines of southwestern Australia did not eat shellfish. The function of the small stone points that appear in Australian archaeological sites around 5,000 years ago remains uncertain: while an easy explanation is that they may have been used as spearpoints and barbs, they are suspiciously similar to the stone points and barbs used on arrows elsewhere in the world. If they really were so used, the mystery of bows and arrows being present in modern New Guinea but absent in Australia might be compounded: perhaps bows and arrows actually were adopted for a while, then abandoned, across the Australian continent. All these examples remind us of the abandonment of guns in Japan, of bows and arrows and pottery in most of Polynesia, and of other technologies in other isolated societies (Chapter 13).
The most extreme losses of technology in the Australian region took place on the island of Tasmania, 130 miles off the coast of southeastern Australia. At Pleistocene times of low sea level, the shallow Bass Strait now separating Tasmania from Australia was dry land, and the people occupying Tasmania were part of the human population distributed continuously over an expanded Australian continent. When the strait was at last flooded around 10,000 years ago, Tasmanians and mainland Australians became cut off from each other because neither group possessed watercraft capable of negotiating Bass Strait. Thereafter, Tasmania's population of 4,000 hunter-gatherers remained out of contact with all other humans on Earth, living in an isolation otherwise known only from science fiction novels.
When finally encountered by Europeans in a.d. 1642, the Tasmanians had the simplest material culture of any people in the modern world. Like mainland Aborigines, they were hunter-gatherers without metal tools. But they also lacked many technologies and artifacts widespread on the mainland, including barbed spears, bone tools of any type, boomerangs, ground or polished stone tools, hafted stone tools, hooks, nets, pronged spears, traps, and the practices of catching and eating fish, sewing, and starting a fire. Some of these technologies may have arrived or been invented in mainland Australia only after Tasmania became isolated, in which case we can conclude that the tiny Tasmanian population did not independently invent these technologies for itself. Others of these technologies were brought to Tasmania when it was still part of the Australian mainland, and were subsequently lost in Tasmania's cultural isolation. For example,

YALI'S

PEOPLE • 313the Tasmanian archaeological record documents the disappearance of fishing, and of awls, needles, and other bone tools, around 1500 b.c. On at least three smaller islands (Flinders, Kangaroo, and King) that were isolated from Australia or Tasmania by rising sea levels around 10,000 years ago, human populations that would initially have numbered around 200 to 400 died out completely.
Tasmania and those three smaller islands thus illustrate in extreme form a conclusion of broad potential significance for world history. Human populations of only a few hundred people were unable to survive indefinitely in complete isolation. A population of 4,000 was able to survive for 10,000 years, but with significant cultural losses and significant failures to invent, leaving it with a uniquely simplified material culture. Mainland Australia's 300,000 hunter-gatherers were more numerous and less isolated than the Tasmanians but still constituted the smallest and most isolated human population of any of the continents. The documented instances of technological regression on the Australian mainland, and the example of Tasmania, suggest that the limited repertoire of Native Australians compared with that of peoples of other continents may stem in part from the effects of isolation and population size on the development and maintenance of technology—like those effects on Tasmania, but less extreme. By implication, the same effects may have contributed to differences in technology between the largest continent (Eurasia) and the next smaller ones (Africa, North America, and South America).
why didn't mo re-advanced technology reach Australia from its neighbors, Indonesia and New Guinea? As regards Indonesia, it was separated from northwestern Australia by water and was very different from it ecologically. In addition, Indonesia itself was a cultural and technological backwater until a few thousand years ago. There is no evidence of any new technology or introduction reaching Australia from Indonesia, after Australia's initial colonization 40,000 years ago, until the dingo appeared around 1500 b.c.
The dingo reached Australia at the peak of the Austronesian expansion from South China through Indonesia. Austronesians succeeded in settling all the islands of Indonesia, including the two closest to Australia—Timor and Tanimbar (only 275 and 205 miles from modern Australia, respectively). Since Austronesians covered far greater sea distances in the course

314 " GUNS, GERMS, AND STEEL

of their expansion across the Pacific, we would have to assume that they repeatedly reached Australia, even if we did not have the evidence of the dingo to prove it. In historical times northwestern Australia was visited each year by sailing canoes from the Macassar district on the Indonesian island of Sulawesi (Celebes), until the Australian government stopped the visits in 1907. Archaeological evidence traces the visits back until around a.d. 1000, and they may well have been going on earlier. The main purpose of the visits was to obtain sea cucumbers (also known as beche-de-mer or trepang), starfish relatives exported from Macassar to China as a reputed aphrodisiac and prized ingredient of soups.
Naturally, the trade that developed during the Macassans' annual visits left many legacies in northwestern Australia. The Macassans planted tamarind trees at their coastal campsites and sired children by Aboriginal women. Cloth, metal tools, pottery, and glass were brought as trade goods, though Aborigines never learned to manufacture those items themselves. Aborigines did acquire from the Macassans some loan words, some ceremonies, and the practices of using dugout sailing canoes and smoking tobacco in pipes.
But none of these influences altered the basic character of Australian society. More important than what happened as a result of the Macassan visits is what did not happen. The Macassans did not settle in Australia— undoubtedly because the area of northwestern Australia facing Indonesia is much too dry for Macassan agriculture. Had Indonesia faced the tropical rain forests and savannas of northeastern Australia, the Macassans could have settled, but there is no evidence that they ever traveled that far. Since the Macassans thus came only in small numbers and for temporary visits and never penetrated inland, just a few groups of Australians on a small stretch of coast were exposed to them. Even those few Australians got to see only a fraction of Macassan culture and technology, rather than a full Macassan society with rice fields, pigs, villages, and workshops. Because the Australians remained nomadic hunter-gatherers, they acquired only those few Macassan products and practices compatible with their lifestyle. Dugout sailing canoes and pipes, yes; forges and pigs, no.
Apparently much more astonishing than Australians' resistance to Indonesian influence is their resistance to New Guinea influence. Across the narrow ribbon of water known as Torres Strait, New Guinea farmers who spoke New Guinea languages and had pigs, pottery, and bows and arrows faced Australian hunter-gatherers who spoke Australian languages and

YALI'S

PEOPLE • 315lacked pigs, pottery, and bows and arrows. Furthermore, the strait is not an open-water barrier but is dotted with a chain of islands, of which the largest (Muralug Island) lies only 10 miles from the Australian coast. There were regular trading visits between Australia and the islands, and between the islands and New Guinea. Many Aboriginal women came as wives to Muralug Island, where they saw gardens and bows and arrows. How was it that those New Guinea traits did not get transmitted to Australia?
This cultural barrier at Torres Strait is astonishing only because we may mislead ourselves into picturing a full-fledged New Guinea society with intensive agriculture and pigs 10 miles off the Australian coast. In reality, Cape York Aborigines never saw a mainland New Guinean. Instead, there was trade between New Guinea and the islands nearest New Guinea, then between those islands and Mabuiag Island halfway down the strait, then between Mabuiag Island and Badu Island farther down the strait, then between Badu Island and Muralug Island, and finally between Muralug and Cape York.
New Guinea society became attenuated along that island chain. Pigs were rare or absent on the islands. Lowland South New Guineans along Torres Strait practiced not the intensive agriculture of the New Guinea highlands but a slash-and-burn agriculture with heavy reliance on seafoods, hunting, and gathering. The importance of even those slash-and-burn practices decreased from southern New Guinea toward Australia along the island chain. Muralug Island itself, the island nearest Australia, was dry, marginal for agriculture, and supported only a small human population, which subsisted mainly on seafood, wild yams, and mangrove fruits.
The interface between New Guinea and Australia across Torres Strait was thus reminiscent of the children's game of telephone, in which children sit in a circle, one child whispers a word into the ear of the second child, who whispers what she thinks she has just heard to the third child, and the word finally whispered by the last child back to the first child bears no resemblance to the initial word. In the same way, trade along the Torres Strait islands was a telephone game that finally presented Cape York Aborigines with something very different from New Guinea society. In addition, we should not imagine that relations between Muralug Islanders and Cape York Aborigines were an uninterrupted love feast at which Aborigines eagerly sopped up culture from island teachers. Trade instead alter-

3 I 6 •

GUNS,GERMS, AND STEELnated with war for the purposes of head-hunting and capturing women to become wives.
Despite the dilution of New Guinea culture by distance and war, some New Guinea influence, did manage to reach Australia. Intermarriage carried New Guinea physical features, such as coiled rather than straight hair, down the Cape York Peninsula. Four Cape York languages had phonemes unusual for Australia, possibly because of the influence of New Guinea languages. The most important transmissions were of New Guinea shell fishhooks, which spread far into Australia, and of New Guinea outrigger canoes, which spread down the Cape York Peninsula. New Guinea drums, ceremonial masks, funeral posts, and pipes were also adopted on Cape York. But Cape York Aborigines did not adopt agriculture, in part because what they saw of it on Muralug Island was so watered-down. They did not adopt pigs, of which there were few or none on the islands, and which they would in any case have been unable to feed without agriculture. Nor did they adopt bows and arrows, remaining instead with their spears and spear-throwers.
Australia is big, and so is New Guinea. But contact between those two big landmasses was restricted to those few small groups of Torres Strait islanders with a highly attenuated New Guinea culture, interacting with those few small groups of Cape York Aborigines. The latter groups' decisions, for whatever reason, to use spears rather than bows and arrows, and not to adopt certain other features of the diluted New Guinea culture they saw, blocked transmission of those New Guinea cultural traits to all the rest of Australia. As a result, no New Guinea trait except shell fishhooks spread far into Australia. If the hundreds of thousands of farmers in the cool New Guinea highlands had been in close contact with the Aborigines in the cool highlands of southeastern Australia, a massive transfer of intensive food production and New Guinea culture to Australia might have followed. But the New Guinea highlands are separated from the Australian highlands by 2,000 miles of ecologically very different landscape. The New Guinea highlands might as well have been the mountains of the moon, as far as Australians' chances of observing and adopting New Guinea highland practices were concerned.
In short, the persistence of Stone Age nomadic hunter-gatherers in Australia, trading with Stone Age New Guinea farmers and Iron Age Indonesian farmers, at first seems to suggest singular obstinacy on the part of Native Australians. On closer examination, it merely proves to reflect the

YALl'S

PEOPLE • 317ubiquitous role of geography in the transmission of human culture and technology.
itremains for usto consider the encounters of new guinea's and australia's stone age societies with iron age Europeans. A Portuguese navigator "discovered" New Guinea in 1526, Holland claimed the western half in 1828, and Britain and Germany divided the eastern half in 1884. The first Europeans settled on the coast, and it took them a long time to penetrate into the interior, but by 1960 European governments had established political control over most New Guineans.
The reasons that Europeans colonized New Guinea, rather than vice versa, are obvious. Europeans were the ones who had the oceangoing ships and compasses to travel to New Guinea; the writing systems and printing presses to produce maps, descriptive accounts, and administrative paperwork useful in establishing control over New Guinea; the political institutions to organize the ships, soldiers, and administration; and the guns to shoot New Guineans who resisted with bow and arrow and clubs. Yet the number of European settlers was always very small, and today New Guinea is still populated largely by New Guineans. That contrasts sharply with the situation in Australia, the Americas, and South Africa, where European settlement was numerous and lasting and replaced the original native population over large areas. Why was New Guinea different?
A major factor was the one that defeated all European attempts to settle the New Guinea lowlands until the 1880s: malaria and other tropical diseases, none of them an acute epidemic crowd infection as discussed in Chapter 11. The most ambitious of those failed lowland settlement plans, organized by the French marquis de Rays around 1880 on the nearby island of New Ireland, ended with 930 out of the 1,000 colonists dead within three years. Even with modern medical treatments available today, many of my American and European friends in New Guinea have been forced to leave because of malaria, hepatitis, or other diseases, while my own health legacy of New Guinea has been a year of malaria and a year of dysentery.
As Europeans were being felled by New Guinea lowland germs, why were Eurasian germs not simultaneously felling New Guineans? Some New Guineans did become infected, but not on the massive scale that

3 I 8 •

GUNS,GERMS, ANDsteelkilled off most of the native peoples of Australia and the Americas. One lucky break for New Guineans was that there were no permanent European settlements in New Guinea until the 1880s, by which time public health discoveries had made progress in bringing smallpox and other infectious diseases of European populations under control. In addition, the Austronesian expansion had already been bringing a stream of Indonesian settlers and traders to New Guinea for 3,500 years. Since Asian mainland infectious diseases were well established in Indonesia, New Guineans thereby gained long exposure and built up much more resistance to Eurasian germs than did Aboriginal Australians.
The sole part of New Guinea where Europeans do not suffer from severe health problems is the highlands, above the altitudinal ceiling for malaria. But the highlands, already occupied by dense populations of New Guineans, were not reached by Europeans until the 1930s. By then, the Australian and Dutch colonial governments were no longer willing to open up lands for white settlement by killing native people in large numbers or driving them off their lands, as had happened during earlier centuries of European colonialism.
The remaining obstacle to European would-be settlers was that European crops, livestock, and subsistence methods do poorly everywhere in the New Guinea environment and climate. While introduced tropical American crops such as squash, corn, and tomatoes are now grown in small quantities, and tea and coffee plantations have been established in the highlands of Papua New Guinea, staple European crops, like wheat, barley, and peas, have never taken hold. Introduced cattle and goats, kept in small numbers, suffer from tropical diseases, just as do European people themselves. Food production in New Guinea is still dominated by the crops and agricultural methods that New Guineans perfected over the course of thousands of years.
All those problems of disease, rugged terrain, and subsistence contributed to Europeans' leaving eastern New Guinea (now the independent nation of Papua New Guinea) occupied and governed by New Guineans, who nevertheless use English as their official language, write with the alphabet, live under democratic governmental institutions modeled on those of England, and use guns manufactured overseas. The outcome was different in western New Guinea, which Indonesia took over from Holland in 1963 and renamed Irian Jaya province. The province is now governed by Indonesians, for Indonesians. Its rural population is still

YALI'S

PEOPLE • 319overwhelmingly New Guinean, but its urban population is Indonesian, as a result of government policy aimed at encouraging Indonesian immigration. Indonesians, with their long history of exposure to malaria and other tropical diseases shared with New Guineans, have not faced as potent a germ barrier as have Europeans. They are also better prepared than Europeans for subsisting in New Guinea, because Indonesian agriculture already included bananas, sweet potatoes, and some other staple crops of New Guinea agriculture. The ongoing changes in Irian Jaya represent the continuation, backed by a centralized government's full resources, of the Austronesian expansion that began to reach New Guinea 3,500 years ago. Indonesians are modern Austronesians.
europeans colonized australia, rather than Native Australians colonizing Europe, for the same reasons that we have just seen in the case of New Guinea. However, the fates of New Guineans and of Aboriginal Australians were very different. Today, Australia is populated and governed by 20 million non-Aborigines, most of them of European descent, plus increasing numbers of Asians arriving since Australia abandoned its previous White Australia immigration policy in 1973. The Aboriginal population declined by 80 percent, from around 300,000 at the time of European settlement to a minimum of 60,000 in 1921. Aborigines today form an underclass of Australian society. Many of them live on mission stations or government reserves, or else work for whites as herdsmen on cattle stations. Why did Aborigines fare so much worse than New Guineans?
The basic reason is Australia's suitability (in some areas) for European food production and settlement, combined with the role of European guns, germs, and steel in clearing Aborigines out of the way. While I already stressed the difficulties posed by Australia's climate and soils, its most productive or fertile areas can nevertheless support European farming. Agriculture in the Australian temperate zone is now dominated by the Eurasian temperate-zone staple crops of wheat (Australia's leading crop), barley, oats, apples, and grapes, along with sorghum and cotton of African Sahel origins and potatoes of Andean origins. In tropical areas of northeastern Australia (Queensland) beyond the optimal range of Fertile Crescent crops, European farmers introduced sugarcane of New Guinea origins, bananas and citrus fruit of tropical Southeast Asian origins, and peanuts of tropical South American origins. As for livestock, Eurasian sheep made

3 2. O •

GUNS,GERMS, AND STEELit possible to extend food production to arid areas of Australia unsuitable for agriculture, and Eurasian cattle joined crops in moister areas.
Thus, the development of food production in Australia had to await the arrival of non-native crops and animals domesticated in climatically similar parts of the world too remote for their domesticates to reach Australia until brought by transoceanic shipping. Unlike New Guinea, most of Australia lacked diseases serious enough to keep out Europeans. Only in tropical northern Australia did malaria and other tropical diseases force Europeans to abandon their 19th-century attempts at settlement, which succeeded only with the development of 20th-century medicine.
Australian Aborigines, of course, stood in the way of European food production, especially because what was potentially the most productive farmland and dairy country initially supported Australia's densest populations of Aboriginal hunter-gatherers. European settlement reduced the number of Aborigines by two means. One involved shooting them, an option that Europeans considered more acceptable in the 19th and late 18th centuries than when they entered the New Guinea highlands in the 1930s. The last large-scale massacre, of 31 Aborigines, occurred at Alice Springs in 1928. The other means involved European-introduced germs to which Aborigines had had no opportunity to acquire immunity or to evolve genetic resistance. Within a year of the first European settlers' arrival at Sydney, in 1788, corpses of Aborigines who had died in epidemics became a common sight. The principal recorded killers were smallpox, influenza, measles, typhoid, typhus, chicken pox, whooping cough, tuberculosis, and syphilis.
In these two ways, independent Aboriginal societies were eliminated in all areas suitable for European food production. The only societies that survived more or less intact were those in areas of northern and western Australia useless to Europeans. Within one century of European colonization, 40,000 years of Aboriginal traditions had been mostly swept away.
we can now return to the problem that I posed near the beginning of this chapter. How, except by postulating deficiencies in the Aborigines themselves, can one account for the fact that white English colonists apparently created a literate, food-producing, industrial democracy, within a few decades of colonizing a continent whose inhabitants after more than 40,000 years were still nonliterate nomadic hunter-gatherers? Doesn't that

YALI'S PEOPLE • 311

constitute a perfectly controlled experiment in the evolution of human societies, forcing us to a simple racist conclusion?
The resolution of this problem is simple. White English colonists did not create a literate, food-producing, industrial democracy in Australia. Instead, they imported all of the elements from outside Australia: the livestock, all of the crops (except macadamia nuts), the metallurgical knowledge, the steam engines, the guns, the alphabet, the political institutions, even the germs. All these were the end products of 10,000 years of development in Eurasian environments. By an accident of geography, the colonists who landed at Sydney in 1788 inherited those elements. Europeans have never learned to survive in Australia or New Guinea without their inherited Eurasian technology. Robert Burke and William Wills were smart enough to write, but not smart enough to survive in Australian desert regions where Aborigines were living.
The people who did create a society in Australia were Aboriginal Australians. Of course, the society that they created was not a literate, food-producing, industrial democracy. The reasons follow straightforwardly from features of the Australian environment.

CHAPTER

16How china became chinese

IMMIGRATION, AFFIRMATIVE ACTION, MULTILINGUALISM, ethnic diversity—my state of California was among the pioneers of these controversial policies and is now pioneering a backlash against them. A glance into the classrooms of the Los Angeles public school system, where my sons are being educated, fleshes out the abstract debates with the faces of children. Those children represent over 80 languages spoken in the home, with English-speaking whites in the minority. Every single one of my sons' playmates has at least one parent or grandparent who was born outside the United States; that's true of three of my own sons' four grandparents. But immigration is merely restoring the diversity that America held for thousands of years. Before European settlement, the mainland United States was home to hundreds of Native American tribes and languages and came under control of a single government only within the last hundred years.

In these respects the United States is a thoroughly "normal" country. All but one of the world's six most populous nations are melting pots that achieved political unification recently, and that still support hundreds of languages and ethnic groups. For example, Russia, once a small Slavic state centered on Moscow, did not even begin its expansion beyond the Ural Mountains until a.d. 1582. From then until the 19th century, Russia

HOW

CHINA BECAME CHINESE • 3 2. 3proceeded to swallow up dozens of non-Slavic peoples, many of which retain their original language and cultural identity. Just as American history is the story of how our continent's expanse became American, Russia's history is the story of how Russia became Russian. India, Indonesia, and Brazil are also recent political creations (or re-creations, in the case of India), home to about 850, 670, and 210 languages, respectively.
The great exception to this rule of the recent melting pot is the world's most populous nation, China. Today, China appears politically, culturally, and linguistically monolithic, at least to laypeople. It was already unified politically in 221 B.C. and has remained so for most of the centuries since then. From the beginnings of literacy in China, it has had only a single writing system, whereas modern Europe uses dozens of modified alphabets. Of China's 1.2 billion people, over 800 million speak Mandarin, the language with by far the largest number of native speakers in the world. Some 300 million others speak seven other languages as similar to Mandarin, and to each other, as Spanish is to Italian. Thus, not only is China not a melting pot, but it seems absurd to ask how China became Chinese. China has been Chinese, almost from the beginnings of its recorded history.
We take this seeming unity of China so much for granted that we forget how astonishing it is. One reason why we should not have expected such unity is genetic. While a coarse racial classification of world peoples lumps all Chinese people as so-called Mongoloids, that category conceals much more variation than the differences between Swedes, Italians, and Irish within Europe. In particular, North and South Chinese are genetically and physically rather different: North Chinese are most similar to Tibetans and Nepalese, while South Chinese are similar to Vietnamese and Filipinos. My North and South Chinese friends can often distinguish each other at a glance by physical appearance: the North Chinese tend to be taller, heavier, paler, with more pointed noses, and with smaller eyes that appear more "slanted" (because of what is termed their epicanthic fold).
North and South China differ in environment and climate as well: the north is drier and colder; the south, wetter and hotter. Genetic differences arising in those differing environments imply a long history of moderate isolation between peoples of North and South China. How did those peoples nevertheless end up with the same or very similar languages and cultures?
China's apparent linguistic near-unity is also puzzling in view of the

3 Z 4 '

GUNS,GERMS, AND STEELlinguistic disunity of other long-settled parts of the world. For instance, we saw in the last chapter that New Guinea, with less than one-tenth of China's area and with only about 40,000 years of human history, has a thousand languages, including dozens of language groups whose differences are far greater than those among the eight main Chinese languages. Western Europe has evolved or acquired about 40 languages just in the 6,000-8,000 years since the arrival of Indo-European languages, including languages as different as English, Finnish, and Russian. Yet fossils attest to human presence in China for over half a million years. What happened to the tens of thousands of distinct languages that must have arisen in China over that long time span?
These paradoxes hint that China too was once diverse, as all other populous nations still are. China differs only by having been unified much earlier. Its "Sinification" involved the drastic homogenization of a huge region in an ancient melting pot, the repopulation of tropical Southeast Asia, and the exertion of a massive influence on Japan, Korea, and possibly even India. Hence the history of China offers the key to the history of all of East Asia. This chapter will tell the story of how China did become Chinese.
a convenient starting point is a detailed linguistic map of China (see Figure 16.1). A glance at it is an eye-opener to all of us accustomed to thinking of China as monolithic. It turns out that, in addition to China's eight "big" languages—Mandarin and its seven close relatives (often referred to collectively simply as "Chinese"), with between 11 million and 800 million speakers each—China also has over 130 "little" languages, many of them with just a few thousand speakers. All these languages, "big" and "little," fall into four language families, which differ greatly in the compactness of their distributions.
At the one extreme, Mandarin and its relatives, which constitute the Chinese subfamily of the Sino-Tibetan language family, are distributed continuously from North to South China. One could walk through China, from Manchuria in the north to the Gulf of Tonkin in the south, while remaining entirely within land occupied by native speakers of Mandarin and its relatives. The other three families have fragmented distributions, being spoken by "islands" of people surrounded by a "sea" of speakers of Chinese and other language families.

HOW

CHINA BECAME CHINESE • 3 2. 5Especially fragmented is the distribution of the Miao-Yao (alias Hmong-Mien) family, which consists of 6 million speakers divided among about five languages, bearing the colorful names of Red Miao, White Miao (alias Striped Miao), Black Miao, Green Miao (alias Blue Miao), and Yao. Miao-Yao speakers live in dozens of small enclaves, all surrounded by speakers of other language families and scattered over an area of half a million square miles, extending from South China to Thailand. More than 100,000 Miao-speaking refugees from Vietnam have carried this language family to the United States, where they are better known under the alternative name of Hmong.
Another fragmented language group is the Austroasiatic family, whose most widely spoken languages are Vietnamese and Cambodian. The 60 million Austroasiatic speakers are scattered from Vietnam in the east to the Malay Peninsula in the south and to northern India in the west. The fourth and last of China's language families is the Tai-Kadai family (including Thai and Lao), whose 50 million speakers are distributed from South China southward into Peninsular Thailand and west to Myanmar (Figure 16.1).
Naturally, Miao-Yao speakers did not acquire their current fragmented distribution as a result of ancient helicopter flights that dropped them here and there over the Asian landscape. Instead, one might guess that they once had a more nearly continuous distribution, which became fragmented as speakers of other language families expanded or induced Miao-Yao speakers to abandon their tongues. In fact, much of that process of linguistic fragmentation occurred within the past 2,500 years and is well documented historically. The ancestors of modern speakers of Thai, Lao, and Burmese all moved south from South China and adjacent areas to their present locations within historical times, successively inundating the settled descendants of previous migrations. Speakers of Chinese languages were especially vigorous in replacing and linguistically converting other ethnic groups, whom Chinese speakers looked down upon as primitive and inferior. The recorded history of China's Zhou Dynasty, from 1100 to 221 b.c., describes the conquest and absorption of most of China's non-Chinese-speaking population by Chinese-speaking states.
We can use several types of reasoning to try to reconstruct the linguistic map of East Asia as of several thousand years ago. First, we can reverse the historically known linguistic expansions of recent millennia. Second, we can reason that modern areas with just a single language or related

3X8

• GUNS, GERMS. AND STEELlanguage group occupying a large, continuous area testify to a recent geographic expansion of that group, such that not enough historical time has elapsed for it to differentiate into many languages. Finally, we can reason conversely that modern areas with a high diversity of languages within a given language family lie closer to the early center of distribution of that language family.
Using those three types of reasoning to turn back the linguistic clock, we conclude that North China was originally occupied by speakers of Chinese and other Sino-Tibetan languages; that different parts of South China were variously occupied by speakers of Miao-Yao, Austroasiatic, and Tai-Kadai languages; and that Sino-Tibetan speakers have replaced most speakers of those other families over South China. An even more drastic linguistic upheaval must have swept over tropical Southeast Asia to the south of China—in Thailand, Myanmar, Laos, Cambodia, Vietnam, and Peninsular Malaysia. Whatever languages were originally spoken there must now be entirely extinct, because all of the modern languages of those countries appear to be recent invaders, mainly from South China or, in a few cases, from Indonesia. Since Miao-Yao languages barely survived into the present, we might also guess that South China once harbored still other language families besides Miao-Yao, Austroasiatic, and Tai-Kadai, but that those other families left no modern surviving languages. As we shall see, the Austronesian language family (to which all Philippine and Polynesian languages belong) may have been one of those other families that vanished from the Chinese mainland, and that we know only because it spread to Pacific islands and survived there.
These language replacements in East Asia remind us of the spread of European languages, especially English and Spanish, into the New World, formerly home to a thousand or more Native American languages. We know from our recent history that English did not come to replace U.S. Indian languages merely because English sounded musical to Indians' ears. Instead, the replacement entailed English-speaking immigrants' killing most Indians by war, murder, and introduced diseases, and the surviving Indians' being pressured into adopting English, the new majority language. The immediate causes of that language replacement were the advantages in technology and political organization, stemming ultimately from the advantage of an early rise of food production, that invading Europeans held over Native Americans. Essentially the same processes accounted for the replacement of Aboriginal Australian languages by English, and of

HOW

CHINA BECAME CHINESE • 3^9subequatorial Africa's original Pygmy and Khoisan languages by Bantu languages.
Hence East Asia's linguistic upheavals raise a corresponding question: what enabled Sino-Tibetan speakers to spread from North China to South China, and speakers of Austroasiatic and the other original South China language families to spread south into tropical Southeast Asia? Here, we must turn to archaeology for evidence of the technological, political, and agricultural advantages that some Asians evidently gained over other Asians.
As everywhere else in the world, the archaeological record in East Asia for most of human history reveals only the debris of hunter-gatherers using unpolished stone tools and lacking pottery. The first East Asian evidence for something different comes from China, where crop remains, bones of domestic animals, pottery, and polished (Neolithic) stone tools appear by around 7500 b.c. That date is within a thousand years of the beginning of the Neolithic Age and food production in the Fertile Crescent. But because the previous millennium in China is poorly known archaeologically, one cannot decide at present whether the origins of Chinese food production were contemporaneous with those in the Fertile Crescent, slightly earlier, or slightly later. At the least, we can say that China was one of the world's first centers of plant and animal domestication.
China may actually have encompassed two or more independent centers of origins of food production. I already mentioned the ecological differences between China's cool, dry north and warm, wet south. At a given latitude, there are also ecological distinctions between the coastal lowlands and the interior uplands. Different wild plants are native to these disparate environments and would thus have been variously available to incipient farmers in various parts of China. In fact, the earliest identified crops were two drought-resistant species of millet in North China, but rice in South China, suggesting the possibility of separate northern and southern centers of plant domestication.
Chinese sites with the earliest evidence of crops also contained bones of domestic pigs, dogs, and chickens. These domestic animals and crops were gradually joined by China's many other domesticates. Among the animals, water buffalo were most important (for pulling plows), while silkworms,

33o " GUNS, GERMS, AND

steelducks, and geese were others. Familiar later Chinese crops include soybeans, hemp, citrus fruit, tea, apricots, peaches, and pears. In addition, just as Eurasia's east-west axis permitted many of these Chinese animals and crops to spread westward in ancient times, West Asian domesticates also spread eastward to China and became important there. Especially significant western contributions to ancient China's economy have been wheat and barley, cows and horses, and (to a lesser extent) sheep and goats.
As elsewhere in the world, in China food production gradually led to the other hallmarks of "civilization" discussed in Chapters 11-14. A superb Chinese tradition of bronze metallurgy had its origins in the third millennium b.c. and eventually resulted in China's developing by far the earliest cast-iron production in the world, around 500 b.c. The following 1,500 years saw the outpouring of Chinese technological inventions, mentioned in Chapter 13, that included paper, the compass, the wheelbarrow, and gunpowder. Fortified towns emerged in the third millennium b.c., with cemeteries whose great variation between unadorned and luxuriously furnished graves bespeaks emerging class differences. Stratified societies whose rulers could mobilize large labor forces of commoners are also attested by huge urban defensive walls, big palaces, and eventually the Grand Canal (the world's longest canal, over 1,000 miles long), linking North and South China, Writing is preserved from the second millennium b.c. but probably arose earlier. Our archaeological knowledge of China's emerging cities and states then becomes supplemented by written accounts of China's first dynasties, going back to the Xia Dynasty, which arose around 2000 b.c.
As for food production's more sinister by-product of infectious diseases, we cannot specify where within the Old World most major diseases of Old World origin arose. However, European writings from Roman and medieval times clearly describe the arrival of bubonic plague and possibly smallpox from the east, so these germs could be of Chinese or East Asian origin. Influenza (derived from pigs) is even more likely to have arisen in China, since pigs were domesticated so early and became so important there.
China's size and ecological diversity spawned many separate local cultures, distinguishable archaeologically by their differing styles of pottery and artifacts. In the fourth millennium b.c. those local cultures expanded geographically and began to interact, compete with each other, and

MOW

CHINA BECAME CHINESE • 331coalesce. Just as exchanges of domesticates between ecologically diverse regions enriched Chinese food production, exchanges between culturally diverse regions enriched Chinese culture and technology, and fierce competition between warring chiefdoms drove the formation of ever larger and more centralized states (Chapter 14).
While China's north-south gradient retarded crop diffusion, the gradient was less of a barrier there than in the Americas or Africa, because China's north-south distances were smaller; and because China's is transected neither by desert, as is Africa and northern Mexico, nor by a narrow isthmus, as is Central America. Instead, China's long east-west rivers (the Yellow River in the north, the Yangtze River in the south) facilitated diffusion of crops and technology between the coast and inland, while its broad east-west expanse and relatively gentle terrain, which eventually permitted those two river systems to be joined by canals, facilitated north-south exchanges. All these geographic factors contributed to the early cultural and political unification of China, whereas western Europe, with a similar area but a more rugged terrain and no such unifying rivers, has resisted cultural and political unification to this day.
Some developments spread from south to north in China, especially iron smelting and rice cultivation. But the predominant direction of spread was from north to south. That trend is clearest for writing: in contrast to western Eurasia, which produced a plethora of early writing systems, such as Sumerian cuneiform, Egyptian hieroglyphics, Hittite, Minoan, and the Semitic alphabet, China developed just a single well-attested writing system. It was perfected in North China, spread and preempted or replaced any other nascent system, and evolved into the writing still used in China today. Other major features of North Chinese societies that spread southward were bronze technology, Sino-Tibetan languages, and state formation. All three of China's first three dynasties, the Xia and Shang and Zhou Dynasties, arose in North China in the second millennium b.c.
Preserved writings of the first millennium b.c. show that ethnic Chinese already tended then (as many still do today) to feel culturally superior to non-Chinese "barbarians," while North Chinese tended to regard even South Chinese as barbarians. For example, a late Zhou Dynasty writer of the first millennium b.c. described China's other peoples as follows: "The people of those five regions—the Middle states and the Rong, Yi, and other wild tribes around them—had all their several natures, which they could not be made to alter. The tribes on the east were called Yi. They had their

3 3 2 "

GUNS,GERMS, AND STEELhair unbound, and tattooed their bodies. Some of them ate their food without its being cooked by fire." The Zhou author went on to describe wild tribes to the south, west, and north as indulging in equally barbaric practices, such as turning their feet inward, tattooing their foreheads, wearing skins, living in caves, not eating cereals, and, of course, eating their food raw.
States organized by or modeled on that Zhou Dynasty of North China spread to South China during the first millennium b.c., culminating in China's political unification under the Qin Dynasty in 221 b.c. Its cultural unification accelerated during that same period, as literate "civilized" Chinese states absorbed, or were copied by, the illiterate "barbarians." Some of that cultural unification was ferocious: for instance, the first Qin emperor condemned all previously written historical books as worthless and ordered them burned, much to the detriment of our understanding of early Chinese history and writing. Those and other draconian measures must have contributed to the spread of North China's Sino-Tibetan languages over most of China, and to reducing the Miao-Yao and other language families to their present fragmented distributions.
Within East Asia, China's head start in food production, technology, writing, and state formation had the consequence that Chinese innovations also contributed heavily to developments in neighboring regions. For instance, until the fourth millennium b.c. most of tropical Southeast Asia was still occupied by hunter-gatherers making pebble and flake stone tools belonging to what is termed the Hoabinhian tradition, named after the site of Hoa Binh, in Vietnam. Thereafter, Chinese-derived crops, Neolithic technology, village living, and pottery similar to that of South China spread into tropical Southeast Asia, probably accompanied by South China's language families. The historical southward expansions of Burmese, Laotians, and Thais from South China completed the Sinification of tropical Southeast Asia. All those modern peoples are recent offshoots of their South Chinese cousins.
So overwhelming was this Chinese steamroller that the former peoples of tropical Southeast Asia have left behind few traces in the region's modern populations. Just three relict groups of hunter-gatherers—the Semang Negritos of the Malay Peninsula, the Andaman Islanders, and the Veddoid Negritos of Sri Lanka—remain to suggest that tropical Southeast Asia's former inhabitants may have been dark-skinned and curly-haired, like modern New Guineans and unlike the light-skinned, straight-haired South

HOW

CHINA BECAME CHINESE • 333Chinese and the modern tropical Southeast Asians who are their offshoots. Those relict Negritos of Southeast Asia may be the last survivors of the source population from which New Guinea was colonized. The Semang Negritos persisted as hunter-gatherers trading with neighboring farmers but adopted an Austroasiatic language from those farmers—much as, we shall see, Philippine Negrito and African Pygmy hunter-gatherers adopted languages from their farmer trading partners. Only on the remote Andaman Islands do languages unrelated to the South Chinese language families persist—the last linguistic survivors of what must have been hundreds of now extinct aboriginal Southeast Asian languages.
Even Korea and Japan were heavily influenced by China, although their geographic isolation from it ensured that they did not lose their languages or physical and genetic distinctness, as did tropical Southeast Asia. Korea and Japan adopted rice from China in the second millennium b.c., bronze metallurgy by the first millennium b.c., and writing in the first millennium a.d. China also transmitted West Asian wheat and barley to Korea and Japan.
In thus describing China's seminal role in East Asian civilization, we should not exaggerate. It is not the case that all cultural advances in East Asia stemmed from China and that Koreans, Japanese, and tropical Southeast Asians were noninventive barbarians who contributed nothing. The ancient Japanese developed some of the oldest pottery in the world and settled as hunter-gatherers in villages subsisting on Japan's rich seafood resources, long before the arrival of food production. Some crops were probably domesticated first or independently in Japan, Korea, and tropical Southeast Asia.
But China's role was nonetheless disproportionate. For example, the prestige value of Chinese culture is still so great in Japan and Korea that Japan has no thought of discarding its Chinese-derived writing system despite its drawbacks for representing Japanese speech, while Korea is only now replacing its clumsy Chinese-derived writing with its wonderful indigenous han'gul alphabet. That persistence of Chinese writing in Japan and Korea is a vivid 20th-century legacy of plant and animal domestication in China nearly 10,000 years ago. Thanks to the achievements of East Asia's first farmers, China became Chinese, and peoples from Thailand to (as we shall see in the next chapter) Easter Island became their cousins.

CHAPTER

17speedboat to polynesia

PACIFIC ISLAND HISTORY IS ENCAPSULATED FOR ME IN AN incident that happened when three Indonesian friends and I walked into a store in Jayapura, the capital of Indonesian New Guinea. My friends' names were Achmad, Wiwor, and Sauakari, and the store was run by a merchant named Ping Wah. Achmad, an Indonesian government officer, was acting as the boss, because he and I were organizing an ecological survey for the government and had hired Wiwor and Sauakari as local assistants. But Achmad had never before been in a New Guinea mountain forest and had no idea what supplies to buy. The results were comical.

At the moment that my friends entered the store, Ping Wah was reading a Chinese newspaper. When he saw Wiwor and Sauakari, he kept reading it but then shoved it out of sight under the counter as soon as he noticed Achmad. Achmad picked up an ax head, causing Wiwor and Sauakari to laugh, because he was holding it upside down. Wiwor and Sauakari showed him how to hold it correctly and to test it. Achmad and Sauakari then looked at Wiwor's bare feet, with toes splayed wide from a lifetime of not wearing shoes. Sauakari picked out the widest available shoes and held them against Wiwor's feet, but the shoes were still too narrow, sending Achmad and Sauakari and Ping Wah into peals of laughter. Achmad picked up a plastic comb with which to comb out his straight, coarse black

SPEEDBOAT

TO POLYNESIA • 335hair. Glancing at Wiwor's tough, tightly coiled hair, he handed the comb to Wiwor. It immediately stuck in Wiwor's hair, then broke as soon as Wiwor pulled on the comb. Everyone laughed, including Wiwor. Wiwor responded by reminding Achmad that he should buy lots of rice, because there would be no food to buy in New Guinea mountain villages except sweet potatoes, which would upset Achmad's stomach—more hilarity.
Despite all the laughter, I could sense the underlying tensions. Achmad was Javan, Ping Wah Chinese, Wiwor a New Guinea highlander, and Sauakari a New Guinea lowlander from the north coast. Javans dominate the Indonesian government, which annexed western New Guinea in the 1960s and used bombs and machine guns to crush New Guinean opposition. Achmad later decided to stay in town and to let me do the forest survey alone with Wiwor and Sauakari. He explained his decision to me by pointing to his straight, coarse hair, so unlike that of New Guineans, and saying that New Guineans would kill anyone with hair like his if they found him far from army backup.
Ping Wah had put away his newspaper because importation of Chinese writing is nominally illegal in Indonesian New Guinea. In much of Indonesia the merchants are Chinese immigrants. Latent mutual fear between the economically dominant Chinese and politically dominant Javans erupted in 1966 in a bloody revolution, when Javans slaughtered hundreds of thousands of Chinese. As New Guineans, Wiwor and Sauakari shared most New Guineans' resentment of Javan dictatorship, but they also scorned each other's groups. Highlanders dismiss lowlanders as effete sago eaters, while lowlanders dismiss highlanders as primitive big-heads, referring both to their massive coiled hair and to their reputation for arrogance. Within a few days of my setting up an isolated forest camp with Wiwor and Sauakari, they came close to fighting each other with axes.
Tensions among the groups that Achmad, Wiwor, Sauakari, and Ping Wah represent dominate the politics of Indonesia, the world's fourth-most-populous nation. These modern tensions have roots going back thousands of years. When we think of major overseas population movements, we tend to focus on those since Columbus's discovery of the Americas, and on the resulting replacements of non-Europeans by Europeans within historic times. But there were also big overseas movements long before Columbus, and prehistoric replacements of non-European peoples by other non-European peoples. Wiwor, Achmad, and Sauakari represent three prehistorical waves of people that moved overseas from the Asian mainland into the

3 3 6 •

GUNS,GERMS, AND STEELPacific. Wiwor's Highlanders are probably descended from an early wave that had colonized New Guinea from Asia by 40,000 years ago. Achmad's ancestors arrived in Java ultimately from the South China coast, around 4,000 years ago, completing the replacement there of people related to Wiwor's ancestors. Sauakari's ancestors reached New Guinea around 3,600 years ago, as part of that same wave from the South China coast, while Ping Wah's ancestors still occupy China.
The population movement that brought Achmad's and Sauakari's ancestors to Java and New Guinea, respectively, termed the Austronesian expansion, was among the biggest population movements of the last 6,000 years. One prong of it became the Polynesians, who populated the most remote islands of the Pacific and were the greatest seafarers among Neolithic peoples. Austronesian languages are spoken today as native languages over more than half of the globe's span, from Madagascar to Easter Island. In this book on human population movements since the end of the Ice Ages, the Austronesian expansion occupies a central place, as one of the most important phenomena to be explained. Why did Austronesian people, stemming ultimately from mainland China, colonize Java and the rest of Indonesia and replace the original inhabitants there, instead of Indonesians colonizing China and replacing the Chinese? Having occupied all of Indonesia, why were the Austronesians then unable to occupy more than a narrow coastal strip of the New Guinea lowlands, and why were they completely unable to displace Wiwor's people from the New Guinea highlands? How did the descendants of Chinese emigrants become transformed into Polynesians?
Today, the population of Java, most other Indonesian islands (except the easternmost ones), and the Philippines is rather homogeneous. In appearance and genes those islands' inhabitants are similar to South Chinese, and even more similar to tropical Southeast Asians, especially those of the Malay Peninsula. Their languages are equally homogeneous: while 374 languages are spoken in the Philippines and western and central Indonesia, all of them are closely related and fall within the same sub-subfamily (Western Malayo-Polynesian) of the Austronesian language family. Austronesian languages reached the Asian mainland on the Malay Peninsula and in small pockets in Vietnam and Cambodia, near the westernmost Indonesian islands of Sumatra and Borneo, but they occur nowhere else on the mainland (Figure 17.1). Some Austronesian words

SPEEDBOAT

TO POLYNESIA • 337Distribution of Austronesian languages
Figure 17.1. The Austronesian language family consists of foursubfamilies, three of them confined to Taiwan and one (Malayo-Polynesian) widespread. The latter subfamily in turn consists of two sub-subfamilies, Western Malayo-Polynesian (= W M-P) and Central-Eastern Malayo-Polynesian (= C-E M-P). The latter sub-subfamily inturn consists of four sub-sub-subfamilies, the very widespread Oceanic one to the east and three others to the west in a much smallerarea comprising Halmahera, nearby islands of eastern Indonesia, andthe west end of New Guinea.borrowed into English include "taboo" and "tattoo" (from a Polynesian language), "boondocks" (from the Tagalog language of the Philippines), and "amok," "batik," and "orangutan" (from Malay).
That genetic and linguistic uniformity of Indonesia and the Philippines is initially as surprising as is the predominant linguistic uniformity of China. The famous Java Homo erectus fossils prove that humans have occupied at least western Indonesia for a million years. That should have given ample time for humans to evolve genetic and linguistic diversity and tropical adaptations, such as dark skins like those of many other tropical peoples—but instead Indonesians and Filipinos have light skins.
It is also surprising that Indonesians and Filipinos are so similar to trop-

3 3 6 •

GUNS,GERMS, AND STEELPacific. Wiwor's Highlanders are probably descended from an early wave that had colonized New Guinea from Asia by 40,000 years ago. Achmad's ancestors arrived in Java ultimately from the South China coast, around 4,000 years ago, completing the replacement there of people related to Wiwor's ancestors. Sauakari's ancestors reached New Guinea around 3,600 years ago, as part of that same wave from the South China coast, while Ping Wah's ancestors still occupy China.
The population movement that brought Achmad's and Sauakari's ancestors to Java and New Guinea, respectively, termed the Austronesian expansion, was among the biggest population movements of the last 6,000 years. One prong of it became the Polynesians, who populated the most remote islands of the Pacific and were the greatest seafarers among Neolithic peoples. Austronesian languages are spoken today as native languages over more than half of the globe's span, from Madagascar to Easter Island. In this book on human population movements since the end of the Ice Ages, the Austronesian expansion occupies a central place, as one of the most important phenomena to be explained. Why did Austronesian people, stemming ultimately from mainland China, colonize Java and the rest of Indonesia and replace the original inhabitants there, instead of Indonesians colonizing China and replacing the Chinese? Having occupied all of Indonesia, why were the Austronesians then unable to occupy more than a narrow coastal strip of the New Guinea lowlands, and why were they completely unable to displace Wiwor's people from the New Guinea highlands? How did the descendants of Chinese emigrants become transformed into Polynesians?
ioday, the population of Java, most other Indonesian islands (except the easternmost ones), and the Philippines is rather homogeneous. In appearance and genes those islands' inhabitants are similar to South Chinese, and even more similar to tropical Southeast Asians, especially those of the Malay Peninsula. Their languages are equally homogeneous: while 374 languages are spoken in the Philippines and western and central Indonesia, all of them are closely related and fall within the same sub-subfamily (Western Malayo-Polynesian) of the Austronesian language family. Austronesian languages reached the Asian mainland on the Malay Peninsula and in small pockets in Vietnam and Cambodia, near the westernmost Indonesian islands of Sumatra and Borneo, but they occur nowhere else on the mainland (Figure 17.1). Some Austronesian words

SPEEDBOAT

TO POLYNESIA • 337Distribution of Austronesian languages
Figure 17.1. The Austronesian language family consists of foursubfamilies, three of them confined to Taiwan and one (Malayo-Polynesian) widespread. The latter subfamily in turn consists of two sub-subfamilies, Western Malayo-Polynesian (= W M-P) and Central-Eastern Malayo-Polynesian (= C-E M-P). The latter sub-subfamily inturn consists of four sub-sub-subfamilies, the very widespread Oceanic one to the east and three others to the west in a much smallerarea comprising Halmahera, nearby islands of eastern Indonesia, andthe west end of New Guinea.borrowed into English include "taboo" and "tattoo" (from a Polynesian language), "boondocks" (from the Tagalog language of the Philippines), and "amok," "batik," and "orangutan" (from Malay).
That genetic and linguistic uniformity of Indonesia and the Philippines is initially as surprising as is the predominant linguistic uniformity of China. The famous Java Homo erectus fossils prove that humans have occupied at least western Indonesia for a million years. That should have given ample time for humans to evolve genetic and linguistic diversity and tropical adaptations, such as dark skins like those of many other tropical peoples—but instead Indonesians and Filipinos have light skins.
It is also surprising that Indonesians and Filipinos are so similar to trop-

338

• GUNS, GERMS, AND STEELical Southeast Asians and South Chinese in other physical features besides light skins and in their genes. A glance at a map makes it obvious that Indonesia offered the only possible route by which humans could have reached New Guinea and Australia 40,000 years ago, so one might naively have expected modern Indonesians to be like modern New Guineans and Australians. In reality, there are only a few New Guinean-like populations in the Philippine / western Indonesia area, notably the Negritos living in mountainous areas of the Philippines. As is also true of the three New Guinean-like relict populations that I mentioned in speaking of tropical Southeast Asia (Chapter 16), the Philippine Negritos could be relicts of populations ancestral to Wiwor's people before they reached New Guinea. Even those Negritos speak Austronesian languages similar to those of their Filipino neighbors, implying that they too (like Malaysia's Semang Negritos and Africa's Pygmies) have lost their original language.
All these facts suggest strongly that either tropical Southeast Asians or South Chinese speaking Austronesian languages recently spread through the Philippines and Indonesia, replacing all the former inhabitants of those islands except the Philippine Negritos, and replacing all the original island languages. That event evidently took place too recently for the colonists to evolve dark skins, distinct language families, or genetic distinctiveness or diversity. Their languages are of course much more numerous than the eight dominant Chinese languages of mainland China, but are no more diverse. The proliferation of many similar languages in the Philippines and Indonesia merely reflects the fact that the islands never underwent a political and cultural unification, as did China.Details of language distributions provide valuable clues to the route of this hypothesized Austronesian expansion. The whole Austronesian language family consists of 959 languages, divided among four subfamilies. But one of those subfamilies, termed Malayo-Polynesian, comprises 945 of those 959 languages and covers almost the entire geographic range of the Austronesian family. Before the recent overseas expansion of Europeans speaking Indo-European languages, Austronesian was the most widespread language family in the world. That suggests that the Malayo-Polynesian subfamily differentiated recently out of the Austronesian family and spread far from the Austronesian homeland, giving rise to many local languages, all of which are still closely related because there has been too little time to develop large linguistic differences. For the location of that Austronesian homeland, we should therefore look not to Malayo-

SPEEDBOAT

TO POLYNESIA • 339Polynesian but to the other three Austronesian subfamilies, which differ considerably more from each other and from Malayo-Polynesian than the sub-subfamilies of Malayo-Polynesian differ among each other.
It turns out that those three other subfamilies have coincident distributions all of them tiny compared with the distribution of Malayo-Polynesian. They are confined to aborigines of the island of Taiwan, lying only 90 miles from the South China mainland. Taiwan's aborigines had the island largely to themselves until mainland Chinese began settling in large numbers within the last thousand years. Still more mainlanders arrived after 1945, especially after the Chinese Communists defeated the Chinese Nationalists in 1949, so that aborigines now constitute only 2 percent of Taiwan's population. The concentration of three out of the four Austronesian subfamilies on Taiwan suggests that, within the present Austronesian realm, Taiwan is the homeland where Austronesian languages have been spoken for the most millennia and have consequently had the longest time in which to diverge. All other Austronesian languages, from those on Madagascar to those on Easter Island, would then stem from a population expansion out of Taiwan.
we can now turn to archaeological evidence. While the debris of ancient village sites does not include fossilized words along with bones and pottery, it does reveal movements of people and cultural artifacts that could be associated with languages. Like the rest of the world, most of the present Austronesian realm—Taiwan, the Philippines, Indonesia, and many Pacific islands—was originally occupied by hunter-gatherers lacking pottery, polished stone tools, domestic animals, and crops. (The sole exceptions to this generalization are the remote islands of Madagascar, eastern Melanesia, Polynesia, and Micronesia, which were never reached by hunter-gatherers and remained empty of humans until the Austronesian expansion.) The first archaeological signs of something different within the Austronesian realm come from—Taiwan. Beginning around the fourth millennium b.c., polished stone tools and a distinctive decorated pottery style (so-called Ta-p'en-k'eng pottery) derived from earlier South China mainland pottery appeared on Taiwan and on the opposite coast of the South China mainland. Remains of rice and millet at later Taiwanese sites provide evidence of agriculture.
Ta-p'en-k'eng sites of Taiwan and the South China coast are full of fish

340

" GUNS, GERMS, AND STEELbones and mollusk shells, as well as of stone net sinkers and adzes suitable for hollowing out a wooden canoe. Evidently, those first Neolithic occupants of Taiwan had watercraft adequate for deep-sea fishing and for regular sea traffic across Taiwan Strait, separating that island from the China coast. Thus, Taiwan Strait may have served as the training ground where mainland Chinese developed the open-water maritime skills that would permit them to expand over the Pacific.
One specific type of artifact linking Taiwan's Ta-p'en-k'eng culture to later Pacific island cultures is a bark beater, a stone implement used for pounding the fibrous bark of certain tree species into rope, nets, and clothing. Once Pacific peoples spread beyond the range of wool-yielding domestic animals and fiber plant crops and hence of woven clothing, they became dependent on pounded bark "cloth" for their clothing. Inhabitants of Rennell Island, a traditional Polynesian island that did not become Westernized until the 1930s, told me that Westernization yielded the wonderful side benefit that the island became quiet. No more sounds of bark beaters everywhere, pounding out bark cloth from dawn until after dusk every day!
Within a millennium or so after the Ta-p'en-k'eng culture reached Taiwan, archaeological evidence shows that cultures obviously derived from it spread farther and farther from Taiwan to fill up the modern Austrone-sian realm (Figure 17.2). The evidence includes ground stone tools, pottery, bones of domestic pigs, and crop remains. For example, the decorated Ta-p'en-k'eng pottery on Taiwan gave way to undecorated plain or red pottery, which has also been found at sites in the Philippines and on the Indonesian islands of Celebes and Timor. This cultural "package" of pottery, stone tools, and domesticates appeared around 3000 b.c. in the Philippines, around 2500 b.c. on the Indonesian islands of Celebes and North Borneo and Timor, around 2000 b.c. on Java and Sumatra, and around 1600 b.c. in the New Guinea region. There, as we shall see, the expansion assumed a speedboat pace, as bearers of the cultural package raced eastward into the previously uninhabited Pacific Ocean beyond the Solomon Archipelago. The last phases of the expansion, during the millennium after a.d. 1, resulted in the colonization of every Polynesian and Micronesian island capable of supporting humans. Astonishingly, it also swept westward across the Indian Ocean to the east coast of Africa, resulting in the colonization of the island of Madagascar.

SPEEDBOAT

TO POLYNESIA • 341

Figure 17.2. The paths of the Austronesian expansion, with approxi

mate dates when each region was reached. 4a — Borneo, 4b = Celebes,4c = Timor (around 2500 b.c.). 5a – Halmahera (around 1600 b.c.).5b = Java, 5c — Sumatra (around 2000 b.c.). 6a = Bismarck Archipelago (around 1600 b.c.). 6b = Malay Peninsula, 6c = Vietnam (around 1000 b.c.). 7 = Solomon Archipelago (around 1600 b.c.). 8 – Santa Cruz, 9c = Tonga, 9d = New Caledonia (around 1200 b.c.).lOb= Society Islands, lOc = Cook Islands, 11 a = Tuamotu Archipelago(around A.D. 1).At least until the expansion reached coastal New Guinea, travel between islands was probably by double-outrigger sailing canoes, which are still widespread throughout Indonesia today. That boat design represented a major advance over the simple dugout canoes prevalent among traditional peoples living on inland waterways throughout the world. A dugout canoe is just what its name implies: a solid tree trunk "dug out" (that is, hollowed out), and its ends shaped, by an adze. Since the canoe is as round-bottomed as the trunk from which it was carved, the least imbalance m weight distribution tips the canoe toward the overweighted side.

3 4 2 '

GUNS,GERMS, AND STEELWhenever I've been paddled in dugouts up New Guinea rivers by New Guineans, I have spent much of the trip in terror: it seemed that every slight movement of mine risked capsizing the canoe and spilling out me and my binoculars to commune with crocodiles. New Guineans manage to look secure while paddling dugouts on calm lakes and rivers, but not even New Guineans can use a dugout in seas with modest waves. Hence some stabilizing device must have been essential not only for the Austrone-sian expansion through Indonesia but even for the initial colonization of Taiwan.
The solution was to lash two smaller logs ("outriggers") parallel to the hull and several feet from it, one on each side, connected to the hull by poles lashed perpendicular to the hull and outriggers. Whenever the hull starts to tip toward one side, the buoyancy of the outrigger on that side prevents the outrigger from being pushed under the water and hence makes it virtually impossible to capsize the vessel. The invention of the double-outrigger sailing canoe may have been the technological breakthrough that triggered the Austronesian expansion from the Chinese mainland.
Iwo striking coincidences between archaeological and linguistic evidence support the inference that the people bringing a Neolithic culture to Taiwan, the Philippines, and Indonesia thousands of years ago spoke Austronesian languages and were ancestral to the Austronesian speakers still inhabiting those islands today. First, both types of evidence point unequivocally to the colonization of Taiwan as the first stage of the expansion from the South China coast, and to the colonization of the Philippines and Indonesia from Taiwan as the next stage. If the expansion had proceeded from tropical Southeast Asia's Malay Peninsula to the nearest Indonesian island of Sumatra, then to other Indonesian islands, and finally to the Philippines and Taiwan, we would find the deepest divisions (reflecting the greatest time depth) of the Austronesian language family among the modern languages of the Malay Peninsula and Sumatra, and the languages of Taiwan and the Philippines would have differentiated only recently within a single subfamily. Instead, the deepest divisions are in Taiwan, and the languages of the Malay Peninsula and Sumatra fall together in the same sub-sub-subfamily: a recent branch of the Western Malayo-Polyne-

SPEEDBOAT

TO POLYNESIA • 343sian sub-subfamily, which is in turn a fairly recent branch of the Malayo-Polynesian subfamily. Those details of linguistic relationships agree perfectly with the archaeological evidence that the colonization of the Malay Peninsula was recent, and followed rather than preceded the colonization of Taiwan, the Philippines, and Indonesia.
The other coincidence between archaeological and linguistic evidence concerns the cultural baggage that ancient Austronesians used. Archaeology provides us with direct evidence of culture in the form of pottery, pig and fish bones, and so on. One might initially wonder how a linguist, studying only modern languages whose unwritten ancestral forms remain unknown, could ever figure out whether Austronesians living on Taiwan 6,000 years ago had pigs. The solution is to reconstruct the vocabularies of vanished ancient languages (so-called protolanguages) by comparing vocabularies of modern languages derived from them.
For instance, the words meaning "sheep" in many languages of the Indo-European language family, distributed from Ireland to India, are quite similar: "avis," "avis," "ovis," "oveja," "ovtsa," "owis," and "oi" in Lithuanian, Sanskrit, Latin, Spanish, Russian, Greek, and Irish, respectively. (The English "sheep" is obviously from a different root, but English retains the original root in the word "ewe.") Comparison of the sound shirts that the various modern Indo-European languages have undergone during their histories suggests that the original form was "owis" in the ancestral Indo-European language spoken around 6,000 years ago. That unwritten ancestral language is termed Proto-Indo-European.
Evidently, Proto-Indo-Europeans 6,000 years ago had sheep, in agreement with archaeological evidence. Nearly 2,000 other words of their vocabulary can similarly be reconstructed, including words for "goat," "horse," "wheel," "brother," and "eye." But no Proto-Indo-European word can be reconstructed for "gun," which uses different roots in different modern Indo-European languages: "gun" in English, "fusil" in French, "ruzhyo" in Russian, and so on. That shouldn't surprise us: people 6,000 years ago couldn't possibly have had a word for guns, which were invented only within the past 1,000 years. Since there was thus no inherited shared root meaning "gun," each Indo-European language had to invent or borrow its own word when guns were finally invented.
Proceeding in the same way, we can compare modern Taiwanese, Philippine, Indonesian, and Polynesian languages to reconstruct a Proto-Aus-

344

" GUNS, GERMS, AND STEELtronesian language spoken in the distant past. To no one's surprise, that reconstructed Proto-Austronesian language had words with meanings such as "two," "bird," "ear," and "head louse": of course, Proto-Aus-tronesians could count to 2, knew of birds, and had ears and lice. More interestingly, the reconstructed language had words for "pig," "dog," and "rice," which must therefore have been part of Proto-Austronesian culture. The reconstructed language is full of words indicating a maritime economy, such as "outrigger canoe," "sail," "giant clam," "octopus," "fish trap," and "sea turtle." This linguistic evidence regarding the culture of Proto-Austronesians, wherever and whenever they lived, agrees well with the archaeological evidence regarding the pottery-making, sea-oriented, food-producing people living on Taiwan around 6,000 years ago.
The same procedure can be applied to reconstruct Proto-Malayo-Poly-nesian, the ancestral language spoken by Austronesians after emigrating from Taiwan. Proto-Malayo-Polynesian contains words for many tropical crops like taro, breadfruit, bananas, yams, and coconuts, for which no word can be reconstructed in Proto-Austronesian. Thus, the linguistic evidence suggests that many tropical crops were added to the Austronesian repertoire after the emigration from Taiwan. This conclusion agrees with archaeological evidence: as colonizing farmers spread southward from Taiwan (lying about 23 degrees north of the equator) toward the equatorial tropics, they came to depend increasingly on tropical root and tree crops, which they proceeded to carry with them out into the tropical Pacific.
How could those Austronesian-speaking farmers from South China via Taiwan replace the original hunter-gatherer population of the Philippines and western Indonesia so completely that little genetic and no linguistic evidence of that original population survived? The reasons resemble the reasons why Europeans replaced or exterminated Native Australians within the last two centuries, and why South Chinese replaced the original tropical Southeast Asians earlier: the farmers' much denser populations, superior tools and weapons, more developed watercraft and maritime skills, and epidemic diseases to which the farmers but not the hunter-gatherers had some resistance. On the Asian mainland Austronesian-speaking farmers were able similarly to replace some of the former hunter-gatherers of the Malay Peninsula, because Austronesians colonized the peninsula from the south and east (from the Indonesian islands of Sumatra and Borneo) around the same time that Austroasiatic-speaking farmers were colonizing the peninsula from the north (from Thailand). Other Austrone-

SPEEDBOAT

to polynesia • 345sians managed to establish themselves in parts of southern Vietnam and Can*b.0dia to become the ancestors of the modern Chamic minority of those countries.
However, Austronesian farmers could spread no farther into the Southeast A.sian mainland, because Austroasiatic and Tai-Kadai farmers had already replaced the former hunter-gatherers there, and because Austronesian farmers had no advantage over Austroasiatic and Tai-Kadai farmers. Although we infer that Austronesian speakers originated from coastal South China, Austronesian languages today are not spoken anywhere in mainland China, possibly because they were among the hundreds of former Chinese languages eliminated by the southward expansion of Sino-Tibetan speakers. But the language families closest to Austronesian are thought to be Tai-Kadai, Austroasiatic, and Miao-Yao. Thus, while Austronesian languages in China may not have survived the onslaught of Chinese dynasties, some of their sister and cousin languages did.
We have now followed the initial stages of the Austronesian expansion for 2,500 miles from the South China coast, through Taiwan and the Philippines, to western and central Indonesia. In the course of that expansion, Austronesians came to occupy all habitable areas of those islands, from the seacoast to the interior, and from the lowlands to the mountains. By 1500 b.c. their familiar archaeological hallmarks, including pig bones and plain red-slipped pottery, show that they had reached the eastern Indonesian island of Halmahera, less that 200 miles from the western end of the big mountainous island of New Guinea. Did they proceed to overrun that island, just as they had already overrun the big mountainous islands of Celebes, Borneo, Java, and Sumatra?
They did not, as a glance at the faces of most modern New Guineans makes obvious, and as detailed studies of New Guinean genes confirm. My friend Wiwor and all other New Guinea highlanders differ obviously from Indonesians, Filipinos, and South Chinese in their dark skins, tightly coiled hair, and face shapes. Most lowlanders from New Guinea's interior and south coast resemble the highlanders except that they tend to be taller. Geneticists have failed to find characteristic Austronesian gene markers in blood samples from New Guinea highlanders.
But peoples of New Guinea's north and east coasts, and of the Bismarck and Solomon Archipelagoes north and east of New Guinea, present a more

3 4 6 '

GUNS,GERMS, AND STEELcomplex picture. In appearance, they are variably intermediate between highlanders like Wiwor and Indonesians like Achmad, though on the average considerably closer to Wiwor. For instance, my friend Sauakari from the north coast has wavy hair intermediate between Achmad's straight hair and Wiwor's coiled hair, and skin somewhat paler than Wiwor's, though considerably darker than Achmad's. Genetically, the Bismarck and Solomon islanders and north coastal New Guineans are about 15 percent Austronesian and 85 percent like New Guinea highlanders. Hence Aus-tronesians evidently reached the New Guinea region but failed completely to penetrate the island's interior and were genetically diluted by New Guinea's previous residents on the north coast and islands.
Modern languages tell essentially the same story but add detail. In Chapter 151 explained that most New Guinea languages, termed Papuan languages, are unrelated to any language families elsewhere in the world. Without exception, every language spoken in the New Guinea mountains, the whole of southwestern and south-central lowland New Guinea, including the coast, and the interior of northern New Guinea is a Papuan language. But Austronesian languages are spoken in a narrow strip immediately on the north and southeast coasts. Most languages of the Bismarck and Solomon islands are Austronesian: Papuan languages are spoken only in isolated pockets on a few islands.
Austronesian languages spoken in the Bismarcks and Solomons and north coastal New Guinea are related, as a separate sub-sub-subfamily termed Oceanic, to the sub-sub-subfamily of languages spoken on Hal-mahera and the west end of New Guinea. That linguistic relationship confirms, as one would expect from a map, that Austronesian speakers of the New Guinea region arrived by way of Halmahera. Details of Austronesian and Papuan languages and their distributions in North New Guinea testify to long contact between the Austronesian invaders and the Papuan-speaking residents. Both the Austronesian and the Papuan languages of the region show massive influences of each other's vocabularies and grammars, making it difficult to decide whether certain languages are basically Austronesian languages influenced by Papuan ones or the reverse. As one travels from village to village along the north coast or its fringing islands, one passes from a village with an Austronesian language to a village with a Papuan language and then to another Austronesian-speaking village, without any genetic discontinuity at the linguistic boundaries.

SPEEDBOAT

TO POLYNESIA • 347All this suggests that descendants of Austronesian invaders and of original New Guineans have been trading, intermarrying, and acquiring each other's genes and languages for several thousand years on the North New Guinea coast and its islands. That long contact transferred Austronesian languages more effectively than Austronesian genes, with the result that most Bismarck and Solomon islanders now speak Austronesian languages, even though their appearance and most of their genes are still Papuan. But neither the genes nor the languages of the Austronesians penetrated New Guinea's interior. The outcome of their invasion of New Guinea was thus very different from the outcome of their invasion of Borneo, Celebes, and other big Indonesian islands, where their steamroller eliminated almost all traces of the previous inhabitants' genes and languages. To understand what happened in New Guinea, let us now turn to the evidence from archaeology.
around 1600 b.c., almost simultaneously with their appearance on Halmahera, the familiar archaeological hallmarks of the Austronesian expansion—pigs, chickens, dogs, red-slipped pottery, and adzes of ground stone and of giant clamshells—appear in the New Guinea region. But two features distinguish the Austronesians' arrival there from their earlier arrival in the Philippines and Indonesia.
The first feature consists of pottery designs, which are aesthetic features of no economic significance but which do let archaeologists immediately recognize an early Austronesian site. Whereas most early Austronesian pottery in the Philippines and Indonesia was undecorated, pottery in the New Guinea region was finely decorated with geometric designs arranged in horizontal bands. In other respects the pottery preserved the red slip and the vessel forms characteristic of earlier Austronesian pottery in Indonesia. Evidently, Austronesian settlers in the New Guinea region got the idea of tattooing" their pots, perhaps inspired by geometric designs that they had already been using on their bark cloth and body tattoos. This style is termed Lapita pottery, after an archaeological site named Lapita, where it was described.
The much more significant distinguishing feature of early Austronesian sites m the New Guinea region is their distribution. In contrast to those in the Philippines and Indonesia, where even the earliest known Austronesian

348 • GUNS, GERMS, AND STEEL

sites are on big islands like Luzon and Borneo and Celebes, sites with Lapita pottery in the New Guinea region are virtually confined to small islets fringing remote larger islands. To date, Lapita pottery has been found at only one site (Aitape) on the north coast of New Guinea itself, and at a couple of sites in the Solomons. Most Lapita sites of the New Guinea region are in the Bismarcks, on islets off the coast of the larger Bismarck islands, occasionally on the coasts of the larger islands themselves. Since (as we shall see) the makers of Lapita pottery were capable of sailing thousands of miles, their failure to transfer their villages a few miles to the large Bismarck islands, or a few dozen miles to New Guinea, was certainly not due to inability to get there.
The basis of Lapita subsistence can be reconstructed from the garbage excavated by archaeologists at Lapita sites. Lapita people depended heavily on seafood, including fish, porpoises, sea turtles, sharks, and shellfish. They had pigs, chickens, and dogs and ate the nuts of many trees (including coconuts). While they probably also ate the usual Austronesian root crops, such as taro and yams, evidence of those crops is hard to obtain, because hard nut shells are much more likely than soft roots to persist for thousands of years in garbage heaps.
Naturally, it is impossible to prove directly that the people who made Lapita pots spoke an Austronesian language. However, two facts make this inference virtually certain. First, except for the decorations on the pots, the pots themselves and their associated cultural paraphernalia are similar to the cultural remains found at Indonesian and Philippine sites ancestral to modern Austronesian-speaking societies. Second, Lapita pottery also appears on remote Pacific islands with no previous human inhabitants, with no evidence of a major second wave of settlement subsequent to that bringing Lapita pots, and where the modern inhabitants speak an Austronesian language (more of this below). Hence Lapita pottery may be safely assumed to mark Austronesians' arrival in the New Guinea region.
What were those Austronesian pot makers doing on islets adjacent to bigger islands? They were probably living in the same way as modern pot makers lived until recently on islets in the New Guinea region. In 1972 I visited such a village on Malai Islet, in the Siassi island group, off the medium-sized island of Umboi, off the larger Bismarck island of New Britain. When I stepped ashore on Malai in search of birds, knowing nothing about the people there, I was astonished by the sight that greeted me.

SPEEDBOAT

TOpolynesia • 349Instead of the usual small village of low huts, surrounded by large gardens sufficient to feed the village, and with a few canoes drawn up on the beach, most of the area of Malai was occupied by two-story wooden houses side by side, leaving no ground available for gardens—the New Guinea equivalent of downtown Manhattan. On the beach were rows of big canoes. It turned out that Malai islanders, besides being fishermen, were also specialized potters, carvers, and traders, who lived by making beautifully decorated pots and wooden bowls, transporting them in their canoes to larger islands and exchanging their wares for pigs, dogs, vegetables, and other necessities. Even the timber for Malai canoes was obtained by trade from villagers on nearby Umboi Island, since Malai does not have trees big enough to be fashioned into canoes.
In the days before European shipping, trade between islands in the New Guinea region was monopolized by such specialized groups of canoe-building potters, skilled in sailing without navigational instruments, and living on offshore islets or occasionally in mainland coastal villages. By the time I reached Malai in 1972, those indigenous trade networks had collapsed or contracted, partly because of competition from European motor vessels and aluminum pots, partly because the Australian colonial government forbade long-distance canoe voyaging after some accidents in which traders were drowned. I would guess that the Lapita potters were the inter-island traders of the New Guinea region in the centuries after 1600 b.c.
The spread of Austronesian languages to the north coast of New Guinea itself, and over even the largest Bismarck and Solomon islands, must have occurred mostly after Lapita times, since Lapita sites themselves were concentrated on Bismarck islets. Not until around a.d. 1 did pottery derived from the Lapita style appear on the south side of New Guinea's southeast peninsula. When Europeans began exploring New Guinea in the late 19th century, all the remainder of New Guinea's south coast still supported populations only of Papuan-language speakers, even though Austronesian-speaking populations were established not only on the southeastern peninsula but also on the Aru and Kei Islands (lying 70-80 miles off western New Guinea's south coast). Austronesians thus had thousands of years in which to colonize New Guinea's interior and its southern coast from nearby bases, but they never did so. Even their colonization of North New Guinea's coastal fringe was more linguistic than genetic: all northern coastal peoples remained predominantly New Guineans in their genes. At

35o

" GUNS, GERMS, AND STEELmost, some of them merely adopted Austronesian languages, possibly in order to communicate with the long-distance traders who linked societies.
THUS, the outcome of the Austronesian expansion in the New Guinea region was opposite to that in Indonesia and the Philippines. In the latter region the indigenous population disappeared—presumably driven off, killed, infected, or assimilated by the invaders. In the former region the indigenous population mostly kept the invaders out. The invaders (the Austronesians) were the same in both cases, and the indigenous populations may also have been genetically similar to each other, if the original Indonesian population supplanted by Austronesians really was related to New Guineans, as I suggested earlier. Why the opposite outcomes?
The answer becomes obvious when one considers the differing cultural circumstances of Indonesia's and New Guinea's indigenous populations. Before Austronesians arrived, most of Indonesia was thinly occupied by hunter-gatherers lacking even polished stone tools. In contrast, food production had already been established for thousands of years in the New Guinea highlands, and probably in the New Guinea lowlands and in the Bismarcks and Solomons as well. The New Guinea highlands supported some of the densest populations of Stone Age people anywhere in the modern world.
Austronesians enjoyed few advantages in competing with those established New Guinean populations. Some of the crops on which Austronesians subsisted, such as taro, yams, and bananas, had probably already been independently domesticated in New Guinea before Austronesians arrived. The New Guineans readily integrated Austronesian chickens, dogs, and especially pigs into their food-producing economies. New Guineans already had polished stone tools. They were at least as resistant to tropical diseases as were Austronesians, because they carried the same five types of genetic protections against malaria as did Austronesians, and some or all of those genes evolved independently in New Guinea. New Guineans were already accomplished seafarers, although not as accomplished as the makers of Lapita pottery. Tens of thousands of years before the arrival of Austronesians, New Guineans had colonized the Bismarck and Solomon Archipelagoes, and a trade in obsidian (a volcanic stone suitable for making sharp tools) was thriving in the Bismarcks at least 18,000

SPEEDBOAT

TOpolynesia • 351ears before the Austronesians arrived. New Guineans even seem to have xpanded recently westward against the Austronesian tide, into eastern Indonesia, where languages spoken on the islands of North Halmahera and of Timor are typical Papuan languages related to some languages of western New Guinea.
In short, the variable outcomes of the Austronesian expansion strikingly llustrate the role of food production in human population movements. Austronesian food-producers migrated into two regions (New Guinea and Indonesia) occupied by resident peoples who were probably related to each other. The residents of Indonesia were still hunter-gatherers, while the residents of New Guinea were already food producers and had developed many of the concomitants of food production (dense populations, disease resistance, more advanced technology, and so on). As a result, while the Austronesian expansion swept away the original Indonesians, it failed to make much headway in the New Guinea region, just as it also failed to make headway against Austroasiatic and Tai-Kadai food producers in tropical Southeast Asia.
We have now traced the Austronesian expansion through Indonesia and up to the shores of New Guinea and tropical Southeast Asia. In Chapter 19 we shall trace it across the Indian Ocean to Madagascar, while in Chapter 15 we saw that ecological difficulties kept Austronesians from establishing themselves in northern and western Australia. The expansion's remaining thrust began when the Lapita potters sailed far eastward into the Pacific beyond the Solomons, into an island realm that no other humans had reached previously. Around 1200 b.c. Lapita potsherds, the familiar triumvirate of pigs and chickens and dogs, and the usual other archaeological hallmarks of Austronesians appeared on the Pacific archipelagoes of Fiji, Samoa, and Tonga, over a thousand miles east of the Solomons. Early in the Christian era, most of those same hallmarks (with the notable exception of pottery) appeared on the islands of eastern Polynesia, including the Societies and Marquesas. Further long overwater canoe voyages brought settlers north to Hawaii, east to Pitcairn and Easter Islands, and southwest to New Zealand. The native inhabitants of most of those islands today are the Polynesians, who thus are the direct descendants of the Lapita potters. They speak Austronesian languages closely related to those of the New Guinea region, and their main crops are the Austronesian Package that included taro, yams, bananas, coconuts, and breadfruit.
With the occupation of the Chatham Islands off New Zealand around

3 5 2 ' GUNS, GERMS, AND

steela.d. 1400, barely a century before European "explorers" entered the Pacific, the task of exploring the Pacific was finally completed by Asians. Their tradition of exploration, lasting tens of thousands of years, had begun when Wiwor's ancestors spread through Indonesia to New Guinea and Australia. It ended only when it had run out of targets and almost every habitable Pacific island had been occupied.
Toanyone interested in world history, human societies of East Asia and the Pacific are instructive, because they provide so many examples of how environment molds history. Depending on their geographic homeland, East Asian and Pacific peoples differed in their access to domesticable wild plant and animal species and in their connectedness to other peoples. Again and again, people with access to the prerequisites for food production, and with a location favoring diffusion of technology from elsewhere, replaced peoples lacking these advantages. Again and again, when a single wave of colonists spread out over diverse environments, their descendants developed in separate ways, depending on those environmental differences.
For instance, we have seen that South Chinese developed indigenous food production and technology, received writing and still more technology and political structures from North China, and went on to colonize tropical Southeast Asia and Taiwan, largely replacing the former inhabitants of those areas. Within Southeast Asia, among the descendants or relatives of those food-producing South Chinese colonists, the Yumbri in the mountain rain forests of northeastern Thailand and Laos reverted to living as hunter-gatherers, while the Yumbri's close relatives the Vietnamese (speaking a language in the same sub-subfamily of Austroasiatic as the Yumbri language) remained food producers in the rich Red Delta and established a vast metal-based empire. Similarly, among Austronesian emigrant farmers from Taiwan and Indonesia, the Punan in the rain forests of Borneo were forced to turn back to the hunter-gatherer lifestyle, while their relatives living on Java's rich volcanic soils remained food producers, founded a kingdom under the influence of India, adopted writing, and built the great Buddhist monument at Borobudur. The Austronesians who went on to colonize Polynesia became isolated from East Asian metallurgy and writing and hence remained without writing or metal. As we saw m Chapter 2, though, Polynesian political and social organization and econo-

SPEEDBOAT

TO POLYNESIA '353mies underwent great diversification in different environments. Within a millennium, East Polynesian colonists had reverted to hunting-gathering on the Chathams while building a protostate with intensive food production on Hawaii.
When Europeans at last arrived, their technological and other advantages enabled them to establish temporary colonial domination over most of tropical Southeast Asia and the Pacific islands. However, indigenous germs and food producers prevented Europeans from settling most of this region in significant numbers. Within this area, only New Zealand, New Caledonia, and Hawaii—the largest and most remote islands, lying farthest from the equator and hence in the most nearly temperate (Europe-like) climates—now support large European populations. Thus, unlike Australia and the Americas, East Asia and most Pacific islands remain occupied by East Asian and Pacific peoples.

CHAPTER

18hemispheres colliding

THE LARGEST POPULATION REPLACEMENT OF THE LAST 13,000 years has been the one resulting from the recent collision between Old World and New World societies. Its most dramatic and decisive moment, as we saw in Chapter 3, occurred when Pizarro's tiny army of Spaniards captured the Inca emperor Atahuallpa, absolute ruler of the largest, richest, most populous, and administratively and technologically most advanced Native American state. Atahuallpa's capture symbolizes the European conquest of the Americas, because the same mix of proximate factors that caused it was also responsible for European conquests of other Native American societies. Let us now return to that collision of hemispheres, applying what we have learned since Chapter 3. The basic question to be answered is: why did Europeans reach and conquer the lands of Native Americans, instead of vice versa? Our starting point will be a comparison of Eurasian and Native American societies as of a.d. 1492, the year of Columbus's "discovery" of the Americas.

ourcomparison begins with food production, a major determinant of local population size and societal complexity—hence an ultimate factor behind the conquest. The most glaring difference between American

HEMISPHERES

COLLIDING • 355and Eurasian food production involved big domestic mammal species. In Chapter 9 we encountered Eurasia's 13 species, which became its chief source of animal protein (meat and milk), wool, and hides, its main mode of land transport of people and goods, its indispensable vehicles of warfare and (by drawing plows and providing manure) a big enhancer of crop production. Until waterwheels and windmills began to replace Eurasia's mammals in medieval times, they were also the major source of its "industrial" power beyond human muscle power—for example, for turning grindstones and operating water lifts. In contrast, the Americas had only one species of big domestic mammal, the llama/alpaca, confined to a snail area of the Andes and the adjacent Peruvian coast. While it was used for meat, wool, hides, and goods transport, it never yielded milk for human consumption, never bore a rider, never pulled a cart or a plow, and never served as a power source or vehicle of warfare.
That's an enormous set of differences between Eurasian and Native American societies—due largely to the Late Pleistocene extinction (extermination?) of most of North and South America's former big wild mammal species. If it had not been for those extinctions, modern history might have taken a different course. When Cortes and his bedraggled adventurers landed on the Mexican coast in 1519, they might have been driven into the sea by thousands of Aztec cavalry mounted on domesticated native American horses. Instead of the Aztecs' dying of smallpox, the Spaniards might have been wiped out by American germs transmitted by disease-resistant Aztecs. American civilizations resting on animal power might have been sending their own conquistadores to ravage Europe. But those hypothetical outcomes were foreclosed by mammal extinctions thousands of years earlier.
Those extinctions left Eurasia with many more wild candidates for domestication than the Americas offered. Most candidates disqualify themselves as potential domesticates for any of half a dozen reasons. Hence Eurasia ended up with its 13 species of big domestic mammals and the Americas with just its one very local species. Both hemispheres also had domesticated species of birds and small mammals—the turkey, guinea P'g, and Muscovy duck very locally and the dog more widely in the Americas; chickens, geese, ducks, cats, dogs, rabbits, honeybees, silkworms, and some others m Eurasia. But the significance of all those species of small domestic animals was trivia! compared with that of the big ones.
Eurasia and the Americas also differed with respect to plant food pro-

3 5 6 '

GUNS,GERMS, AND STEELduction, though the disparity here was less marked than for animal food production. In 1492 agriculture was widespread in Eurasia. Among the few Eurasian hunter-gatherers lacking both crops and domestic animals were the Ainu of northern Japan, Siberian societies without reindeer, and small hunter-gatherer groups scattered through the forests of India and tropical Southeast Asia and trading with neighboring farmers. Some other Eurasian societies, notably the Central Asian pastoralists and the reindeer-herding Lapps and Samoyeds of the Arctic, had domestic animals but little or no agriculture. Virtually all other Eurasian societies engaged in agriculture as well as in herding animals.
Agriculture was also widespread in the Americas, but hunter-gatherers occupied a larger fraction of the Americas' area than of Eurasia's. Those regions of the Americas without food production included all of northern North America and southern South America, the Canadian Great Plains, and all of western North America except for small areas of the U.S. Southwest that supported irrigation agriculture. It is striking that the areas of Native America without food production included what today, after Europeans' arrival, are some of the most productive farmlands and pastures of both North and South America: the Pacific states of the United States, Canada's wheat belt, the pampas of Argentina, and the Mediterranean zone of Chile. The former absence of food production in these lands was due entirely to their local paucity of domesticable wild animals and plants, and to geographic and ecological barriers that prevented the crops and the few domestic animal species of other parts of the Americas from arriving. Those lands became productive not only for European settlers but also, in some cases, for Native Americans, as soon as Europeans introduced suitable domestic animals and crops. For instance, Native American societies became renowned for their mastery of horses, and in some cases of cattle and sheepherding, in parts of the Great Plains, the western United States, and the Argentine pampas. Those mounted plains warriors and Navajo sheepherders and weavers now figure prominently in white Americans image of American Indians, but the basis for that image was created only after 1492. These examples demonstrate that the sole missing ingredients required to sustain food production in large areas of the Americas were domestic animals and crops themselves.
In those parts of the Americas that did support Native American agriculture, it was constrained by five major disadvantages vis-a-vis Eurasian agriculture: widespread dependence on protein-poor corn, instead or

HEMISPHERES

COLLIDING • 357Eurasia's diverse and protein-rich cereals; hand planting of individual seeds, instead of broadcast sowing; tilling by hand instead of plowing by animals, which enables one person to cultivate a much larger area, and hich also permits cultivation of some fertile but tough soils and sods that difficult to till by hand (such as those of the North American Great Plains); lack of animal manuring to increase soil fertility; and just human muscle power, instead of animal power, for agricultural tasks such as threshing, grinding, and irrigation. These differences suggest that Eurasian agriculture as of 1492 may have yielded on the average more calories and protein per person-hour of labor than Native American agriculture did.
such differences in food production constituted a major ultimate cause of the disparities between Eurasian and Native American societies. Among the resulting proximate factors behind the conquest, the most important included differences in germs, technology, political organization, and writing. Of these, the one linked most directly to the differences in food production was germs. The infectious diseases that regularly visited crowded Eurasian societies, and to which many Eurasians consequently developed immune or genetic resistance, included all of history's most lethal killers: smallpox, measles, influenza, plague, tuberculosis, typhus, cholera, malaria, and others. Against that grim list, the sole crowd infectious diseases that can be attributed with certainty to pre-Columbian Native American societies were nonsyphilitic treponemas. (As I explained in Chapter 11, it remains uncertain whether syphilis arose in Eurasia or in the Americas, and the claim that human tuberculosis was present in the Americas before Columbus is in my opinion unproven.)
This continental difference in harmful germs resulted paradoxically from the difference in useful livestock. Most of the microbes responsible for the infectious diseases of crowded human societies evolved from very similar ancestral microbes causing infectious diseases of the domestic animals with which food producers began coming into daily close contact around 10,000 years ago. Eurasia harbored many domestic animal species and hence developed many such microbes, while the Americas had very ew of each. Other reasons why Native American societies evolved so few ethal microbes were that villages, which provide ideal breeding grounds for epidemic diseases, arose thousands of years later in the Americas than urasia; and that the three regions of the New World supporting urban

3 5 8 '

GUNS,GERMS, AND STEELsocieties (the Andes, Mesoamerica, and the U.S. Southeast) were never connected by fast, high-volume trade on the scale that brought plague influenza, and possibly smallpox to Europe from Asia. As a result, even malaria and yellow fever, the infectious diseases that eventually became major obstacles to European colonization of the American tropics, and that posed the biggest barrier to the construction of the Panama Canal are not American diseases at all but are caused by microbes of Old World tropical origin, introduced to the Americas by Europeans.
Rivaling germs as proximate factors behind Europe's conquest of the Americas were the differences in all aspects of technology. These differences stemmed ultimately from Eurasia's much longer history of densely populated, economically specialized, politically centralized, interacting and competing societies dependent on food production. Five areas of technology may be singled out:
First, metals—initially copper, then bronze, and finally iron—were used for tools in all complex Eurasian societies as of 1492. In contrast, although copper, silver, gold, and alloys were used for ornaments in the Andes and some other parts of the Americas, stone and wood and bone were still the principal materials for tools in all Native American societies, which made only limited local use of copper tools.
Second, military technology was far more potent in Eurasia than in the Americas. European weapons were steel swords, lances, and daggers, supplemented by small firearms and artillery, while body armor and helmets were also made of solid steel or else of chain mail. In place of steel, Native Americans used clubs and axes of stone or wood (occasionally copper in the Andes), slings, bows and arrows, and quilted armor, constituting much less effective protection and weaponry. In addition, Native American armies had no animals to oppose to horses, whose value for assaults and fast transport gave Europeans an overwhelming advantage until some Native American societies themselves adopted them.
Third, Eurasian societies enjoyed a huge advantage in their sources or power to operate machines. The earliest advance over human muscle power was the use of animals—cattle, horses, and donkeys—to pull plows and to turn wheels for grinding grain, raising water, and irrigating or draining fields. Waterwheels appeared in Roman times and then proliferated, along with tidal mills and windmills, in the Middle Ages. Coupled to systems of geared wheels, those engines harnessing water and wind powe were used not only to grind grain and move water but also to serve myna

HEMISPHERES

COLLIDING • 359manufacturing purposes, including crushing sugar, driving blast furnace Tellows, grinding ores, making paper, polishing stone, pressing oil, producing salt, producing textiles, and sawing wood. It is conventional to define the industrial Revolution arbitrarily as beginning with the harnessing of steam power in 18th-century England, but in fact an industrial revolution based on water and wind power had begun already in medieval times in many parts of Europe. As of 1492, all of those operations to which animal, water, and wind power were being applied in Eurasia were still being carried out by human muscle power in the Americas.
Lone before the wheel began to be used in power conversion in Eurasia, it had become the basis of most Eurasian land transport—not only for animal-drawn vehicles but also for human-powered wheelbarrows, which enabled one or more people, still using just human muscle power, to transport much greater weights than they could have otherwise. Wheels were also adopted in Eurasian pottery making and in clocks. None of those uses of the wheel was adopted in the Americas, where wheels are attested only in Mexican ceramic toys.
The remaining area of technology to be mentioned is sea transport. Many Eurasian societies developed large sailing ships, some of them capable of sailing against the wind and crossing the ocean, equipped with sextants, magnetic compasses, sternpost rudders, and cannons. In capacity, speed, maneuverability, and seaworthiness, those Eurasian ships were far superior to the rafts that carried out trade between the New World's most advanced societies, those of the Andes and Mesoamerica. Those rafts sailed with the wind along the Pacific coast. Pizarro's ship easily ran down and captured such a raft on his first voyage toward Peru.
inaddition to their germs and technology, Eurasian and Native American societies differed in their political organization. By late medieval or Renaissance times, most of Eurasia had come under the rule of organized states. Among these, the Habsburg, Ottoman, and Chinese states, the Mogul state of India, and the Mongol state at its peak in the 13th century started out as large polyglot amalgamations formed by the conquest of other states. For that reason they are generally referred to as empires. Many Eurasian states and empires had official religions that con-»• uted to state cohesion, being invoked to legitimize the political leader-s »P and to sanction wars against other peoples. Tribal and band societies

360

• GUNS, GERMS, AND STEELin Eurasia were largely confined to the Arctic reindeer herders, the Siberian hunter-gatherers, and the hunter-gatherer enclaves in the Indian subcontinent and tropical Southeast Asia.
The Americas had two empires, those of the Aztecs and Incas, which resembled their Eurasian counterparts in size, population, polyglot makeup, official religions, and origins in the conquest of smaller states. In the Americas those were the sole two political units capable of mobilizing resources for public works or war on the scale of many Eurasian states, whereas seven European states (Spain, Portugal, England, France, Holland, Sweden, and Denmark) had the resources to acquire American colonies between 1492 and 1666. The Americas also held many chiefdoms (some of them virtually small states) in tropical South America, Meso-america beyond Aztec rule, and the U.S. Southeast. The rest of the Americas was organized only at the tribal or band level.
The last proximate factor to be discussed is writing. Most Eurasian states had literate bureaucracies, and in some a significant fraction of the populace other than bureaucrats was also literate. Writing empowered European societies by facilitating political administration and economic exchanges, motivating and guiding exploration and conquest, and making available a range of information and human experience extending into remote places and times. In contrast, use of writing in the Americas was confined to the elite in a small area of Mesoamerica. The Inca Empire employed an accounting system and mnemonic device based on knots (termed quipu), but it could not have approached writing as a vehicle for transmitting detailed information.
THUS, eurasian societies in the time of Columbus enjoyed big advantages over Native American societies in food production, germs, technology (including weapons), political organization, and writing. These were the main factors tipping the outcome of the post-Columbian collisions. But those differences as of a.d. 1492 represent just one snapshot of historical trajectories that had extended over at least 13,000 years in the Americas, and over a much longer time in Eurasia. For the Americas, in particular, the 1492 snapshot captures the end of the independent trajectory of Native Americans. Let us now trace out the earlier stages of those trajectories.

HEMISPHERES

COLLIDING • 361Table 18-1 summarizes approximate dates of the appearance of key developments in the main "homelands" of each hemisphere (the Fertile Crescent and China in Eurasia, the Andes and Amazonia and Mesoamerica the Americas). It also includes the trajectory for the minor New World homeland of the eastern United States, and that for England, which Ot a homeland at all but is listed to illustrate how rapidly developments
spread from the Fertile Crescent.
This table is sure to horrify any knowledgeable scholar, because it reduces exceedingly complex histories to a few seemingly precise dates. In reality, all of those dates are merely attempts to label arbitrary points along a continuum. For example, more significant than the date of the first metal tool found by some archaeologist is the time when a significant fraction of all tools was made of metal, but how common must metal tools be to rate as "widespread"? Dates for the appearance of the same development may differ among different parts of the same homeland. For instance, within the Andean region pottery appears about 1,300 years earlier in coastal Ecuador (3100 b.c.) than in Peru (1800 b.c.). Some dates, such as those for the rise of chiefdoms, are more difficult to infer from the archaeological record than are dates of artifacts like pottery or metal tools. Some of the dates in Table 18.1 are very uncertain, especially those for the onset of American food production. Nevertheless, as long as one understands that the table is a simplification, it is useful for comparing continental histories.
The table suggests that food production began to provide a large fraction of human diets around 5,000 years earlier in the Eurasian homelands than in those of the Americas. A caveat must be mentioned immediately: while there is no doubt about the antiquity of food production in Eurasia, there is controversy about its onset in the Americas. In particular, archaeologists often cite considerably older claimed dates for domesticated plants at Coxcatlan Cave in Mexico, at Guitarrero Cave in Peru, and at some other American sites than the dates given in the table. Those claims are now being reevaluated for several reasons: recent direct radiocarbon dat-mg of crop remains themselves has in some cases been yielding younger ates; the older dates previously reported were based instead on charcoal ought to be contemporaneous with the plant remains, but possibly not so, and the status of some of the older plant remains as crops or just as co lected wild plants is uncertain. Still, even if plant domestication did egm earlier in the Americas than the dates shown in Table 18.1, agricul-

3 6 1 •

GUNS,GERMS, AND STEELtable18.1 Historical Trajectories of Eurasia and the Americas

Approximate Date of Adoption

Eurasia

Fertile Crescent

China

England

Plant domestication

8500 b.c.

by 7500 b.c.

3500 b.c.

Animal domestication

8000 b.c.

by 7500 b.c.

3500 b.c.

Pottery

7000 b.c.

by 7500 b.c.

3500 b.c.

Villages

9000 b.c.

by 7500 b.c.

3000 b.c.

Chiefdoms

5500 b.c.

4000 b.c.

2500 b.c.

Widespread metal tools

4000 b.c.

2000 b.c.

2000 b.c.

or artifacts (copper

and/or bronze)

States

3700 b.c.

2000 b.c.

500 A.D.

Writing

3200 b.c.

by 1300 b.c.

A.D. 43

Widespread iron tools

900 b.c.

500 b.c.

650 b.c.

This table gives approximate dates of widespread adoption of significant developments in three Eurasian and four Native American areas. Dates for animal domestication neglect dogs, which were domesticated earlier than food-producing animals in both Eurasia and

ture surely did not provide the basis for most human calorie intake and sedentary existence in American homelands until much later than in Eurasian homelands.
As we saw in Chapters 5 and 10, only a few relatively small areas of each hemisphere acted as a "homeland" where food production first arose and from which it then spread. Those homelands were the Fertile Crescent and China in Eurasia, and the Andes and Amazonia, Mesoamerica, and the eastern United States in the Americas. The rate of spread of key developments is especially well understood for Europe, thanks to the many archaeologists at work there. As Table 13.1 summarizes for England, once food production and village living had arrived from the Fertile Crescent after a long lag (5,000 years), the subsequent lag for England's adoption of chiefdoms, states, writing, and especially metal tools was much shorter: 2,000 years for the first widespread metal tools of copper and bronze, and only 250 years for widespread iron tools. Evidently, it was much easier for one society of already sedentary farmers to "borrow" metallurgy from

HEMISPHERES

COLLIDING • 363

Native America

Andes

Amazonia

Mesoamerica Eastern U.S.

by 3000 B.C.

3000 b.c.

by 3000 b.c. 2500 b.c.

3500 b.c.

p

500 b.c. —

3100-1800 b.c.

6000 b.c.

1500 b.c. 2500 b.c.

3100-1800 b.c.

6000 b.c.

1500 b.c. 500 b.c.

by 1500 b.c.

A.D. 1

1500 b.c. 200 b.c.

A.D. 1000