Joy McCorriston. Cambridge World History of Food. Editor: Kenneth F Kiple & Kriemhild Conee Ornelas. Volume 1. Cambridge, UK: Cambridge University Press, 2000.
Wheat, a grass that today feeds 35 percent of the earth’s population, appeared as a crop among the world’s first farmers 10,000 years ago. It increased in importance from its initial role as a major food for Mediterranean peoples in the Old World to become the world’s largest cereal crop, feeding more than a billion people in the late twentieth century (Feldman 1976: 121). It spread from the Near East, where it first emerged in the nitrogen-poor soils of a semi-arid Mediterranean climate, to flourish in a wide range of environments—from the short summers of far northern latitudes, to cool uplands, to irrigated regions of the tropics. The real story of its origins disappeared from memory many millennia in the past, although some farming peoples still recount tales of how they received other cultivated plants from gods, animate spirits, heroic ancestors, or the earth itself. But today we must use botanical and archaeological evidence to trace the story of wheat’s domestication (implying a change in a plant’s reproduction, making it dependent on humans) and its spread.
Domesticated wheats belong to at least three separate species (Zohary 1971: 238) and hundreds of distinct varieties, a number that continues to increase because the domestication of wheat continues. All domesticated wheat has lost the physical and genetic characteristics that would allow it aggressively to reseed and sprout by itself—losses which clearly distinguish domesticated wheats from wild relatives. Furthermore, both the remarkable geographic distribution of domesticated wheat and the species’ very survival depend on human beings. If no one collected wheat seeds and planted them in cleared, fertile ground, waving fields of grain soon would support hundreds of weeds, to be replaced by wild plants, and perhaps eventually by saplings and forests. Domesticated wheat and humans help each other in a relationship known as “mutualism” (Rindos 1984: 255).
Although humans domesticated wheat, one may argue that dependence on wheat also domesticated humans. The switch from gathering food to producing food, dubbed the “Neolithic Revolution” by V. Gordon Childe (1951: 74, orig. 1936), ultimately and fundamentally altered human development. Both wheat and barley, destined to feed the great civilizations of Mesopotamia, Egypt, Greece, and Rome, originated in the Near East, the earliest cradle of Western civilization. And with food production came great social and technological innovations. For example, because cereals can be stored year-round, early farmers could settle together in larger groups during the seasons when low food availability formerly had forced hunter-gatherers to disperse into small groups. Furthermore, by producing a surplus of cereal food, farmers could support others—people with specialized crafts, administrators, religious castes, and the like. Thousands of years later, cities emerged and empires arose. Clearly, the domestication of a cereal that has fed the Western world (and much of the rest of the world as well) holds a special place in the study of the origins of our foods.
The Origins of Wheat and Barley Agriculture
While archaeologists recognize the momentous developments set in motion by food production in the ancient Near East, they continue to debate the essential factors that first caused people to begin farming wheat and barley.1 How did agriculture begin? Which people first domesticated plants? Why did they do so when they did? And why did farming begin in only a few places? Answers to these questions are significant because with the domestication of wheat, humankind began the shift from hunting and gathering food to producing it. This change in lifestyle set humans on a new evolutionary course, and their society and environment were never the same after farming was established. Because wheat was one of the first crops to be farmed, its role in this fundamental shift has attracted much study, resulting in a variety of models to explain the process of domestication and its causes in the Near East.
The Process of Cereal Domestication
To domesticate wheat, humans must have manipulated wild wheats, either through selective gathering or deliberate cultivation, with the latter implying activities such as preparing ground, sowing, and eliminating competing plants. We owe much of our understanding of the details of this process to the work of several botanists and archaeologists. A Russian botanist, Nikolai Vavilov (1951), for example, discovered that the greatest diversity in the gene pool of wild wheats and barleys is in Southwest Asia. Where diversity is greatest, plants have been growing, fixing mutations, and interbreeding longest. Thus, it can be concluded that Southwest Asia was the ancestral homeland of these plants (Zohary 1970a: 33-5), and subsequent searches for the first farmers have concentrated on this region.
Robert Braidwood, an archaeologist from the University of Chicago, further refined the criteria for the homeland of wild wheats by identifying their modern ecological range, as the semiarid Mediterranean woodland belt known as the “hilly flanks” of the Fertile Crescent (Braidwood 1960: 134). He reasoned that prefarming peoples had adapted culturally and ecologically to specific environments over long periods (a process he dubbed “settling in”) and that the first wheat farmers had already been living among natural stands of wild wheat.
One of Braidwood’s students and a great archaeologist in his own right, Kent V. Flannery, advocated explaining the origins of agriculture in terms of the process of plant domestication. His major contribution to modeling wheat domestication in the Near East stemmed from his recognition that plant domestication may have been the final result in a subtle chain of events that originated with changes in human food procurement occurring much earlier than the actual transition to agriculture (Flannery 1969, 1973: 284). Flannery’s “Broad Spectrum Revolution” portrays a shift in human ecology whereby humans began exploiting many previously minor food sources such as cereals, fish, small game, and water fowl. Ultimately they came to depend on these food sources (1969: 79).
Flannery particularly emphasized the importance of moving cultigens (manipulated but not necessarily domesticated plants) on which humans depended “to niches to which [they were] not adapted” (Flannery 1965: 1251; Wright 1971: 460; Rindos 1984: 26-7). Thus, as people relocated to accommodate shifting population densities (Binford 1968: 332), they attempted to produce rich stands of cereals outside their natural range (Flannery 1969: 80). Such an effort would have helped domesticate wild wheats by preventing relatively rare genetic variants from breeding with the large pool of wild wheat types growing in their natural ranges.
David Rindos (1984) has described the general process of plant domestication from an evolutionary perspective, using the principles of natural selection and mutualistic relationships (coevolution) to describe how cereals, for example, would have lost their wild characteristics. In mutualistic relationships, domesticates and humans enhance each other’s fitness, or ability to reproduce. In the case of wheat, women and men began to collect wild seeds in increasing quantities for food, while at the same time inadvertently selecting and replanting seeds from the plants best suited to easy harvesting (Harlan, de Wet, and Price 1973: 311; Kislev 1984: 63). Within a few generations, cultivated wheat plants became dependent on the harvesting process for survival, as wild self-planting mechanisms disappeared from the traits of cultivated wheats (Wilke et al. 1972: 205; Hillman and Davies 1990). The elimination of wild reseeding characteristics from a plant population ultimately accounted for the domestication of wheats.
The Causes of Cereal Domestication
Because the first evidence for agricultural societies occurs at the beginning of the Holocene (our present epoch) after a major climatic change, several archaeologists found a climatic explanation for the origins of agriculture to be plausible. Childe, for example, maintained that agriculture began at the end of the Pleistocene when climate change caused a lush landscape to dry up and become desert. Populations of humans, animals, and plants would have been forced to concentrate at the few remaining sources of water: Their enhanced propinquity would have provided increased opportunity for experimentation and manipulation (Childe 1952: 23, 25). Childe believed that agriculture started in the oases of North Africa and Mesopotamia, and although these locations were probably incorrectly targeted, some of his other hypotheses now seem essentially correct (Byrne 1987; McCorriston and Hole 1991: 60).
Lewis Binford (1968: 332-7) also indicated the importance of climate change when he emphasized the resource stress experienced by permanently settled coastal populations hard-pressed by rising sea levels (also the result of climatic changes at the end of the Pleistocene). He pointed out, however, that population pressure in marginal zones (settled when rising sea levels flooded coastlines and forced populations to concentrate in smaller areas) would “favor the development of more effective means of food production” from lands no longer offering ample resources for scattered hunter-gatherers (1968: 332). Mark Cohen (1977: 23, 40-51) suggested that population growth filled all available land by the end of the Pleistocene; such dense populations eventually would have experienced the population pressure envisioned by Binford.
These ideas, however, only sharpened the question of why agriculture emerged in only a few regions. Accordingly, several archaeologists sought prerequisites—social or technological developments—that may have caused certain “preadapted” groups of hunter-gatherers to adopt farming in the Near East (Hole 1984: 55; Bar-Yosef and Belfer-Cohen 1989: 487; Rosenberg 1990: 409; McCorriston and Hole 1991: 47-9). Robert Braidwood thought that agriculture appeared when “culture was ready to achieve it” (Braidwood in Wright 1971: 457). For example, sedentism, which appeared for the first time just prior to agriculture (Henry 1985: 371-4, 1989: 219), would have profoundly affected the social relations in a group. Sedentary people can store and safeguard larger amounts of food and other goods than can mobile people. Stored goods increase the possibility of prestige being accorded to a relatively few individuals, since more opportunities now exist for redistribution of surplus goods through kinship alliances—the more goods a person distributes to dependents, the greater his prestige (Bender 1978: 213). As we have seen with contemporary sedentary hunter-gatherers, competition between leaders for greater alliance groups arguably stimulates an intensification of productive forces, which in turn provides a “major incentive for the production of surplus” (Godelier 1970: 120; Bender 1978: 213-14, 1981: 154). Perhaps wheat was such a desired surplus.
In parts of Southwest Asia, where sedentism appears to have preceded the development of agriculture (Henry 1981, 1989: 38-9; Bar-Yosef and Belfer-Cohen 1989: 473-4; Moore 1991: 291), it may have been the case that the causes of sedentism also contributed to the shift to food production (Moore 1985: 231; Henry 1989; Watson 1991: 14). On the other hand, Michael Rosenberg (1990: 410-11) argues that increasingly sharper territorial perceptions were the consequences of concentrated resource exploitation in such territories by hunter-gatherer groups already committed to mutualistic exploitation of plant resources.
A combination of factors probably best explains the domestication of cereals and the shift to agriculture in the Near East. Paleoenvironmental evidence indicates that forests widely replaced a drier steppic cover (van Zeist and Bottema 1982). This has prompted Andrew Moore (1985: 232) to suggest that improved resources (resulting from climatic factors) enabled hunter-gatherers to settle; afterward their populations grew in size, so that ultimately they experienced the sort of resource stress that could have led to intensive manipulation of plants. Donald Henry (1985, 1989) also credits climatic change, several thousand years before agriculture emerged, with causing greater availability of wild cereals, which led increasingly to their exploitation. With this came dependence in the form of a sedentary lifestyle near wild cereal stands, and ultimately domestication during an arid spell when resources grew scarce.
Ecological factors play a major role in another combination model, in which the adaptation of wild cereals to a seasonally stressed environment is viewed as an explanation for the rise of agriculture in the Near East. Herbert Wright, Jr. (1977) was the first to recognize an association between the hot, dry summers and mild, wet winters that characterized Mediterranean climates and the expansion of wild, large-seeded, annual cereals. He suggested that these plants were absent from Southwest Asia in the late Pleistocene. Modern climatic models, in conjunction with archaeological evidence and ecological patterns, distinctly point to the southern Levant—modern Israel and Jordan—as the region where wheat farming first began 10,000 years ago (COHMAP 1988; McCorriston and Hole 1991: 49, 58). There, as summers became hotter and drier, plants already adapted to survive seasonal stress (summer drought), including the wild ancestors of wheat and barley, spread rapidly as the continental flora (adapted to cooler, wetter summers) retreated. Some hunter-gatherer groups living in such regions also experienced seasonal shortages of their erstwhile dependable plant resources. One group, the Natufians, named after the Wadi an Natuf in Israel (where archaeologists first found their remains) probably compensated for seasonal stress by increasingly exploiting the large-seeded annual wild wheats and barleys.
These various approaches, spanning nearly 50 years of research in the Near East, have all contributed to an increasingly sophisticated appreciation of the causes and the process of wheat domestication. Based on data that either fit or fail to fit various models, many specific refutations appeared for each model, but these lie beyond the scope of this chapter. In addition, opinions on the process of domestication in the Near East still differ as follows:
- Was the process fast or slow (Rindos 1984: 138-9; Hillman and Davies 1990: 213)?
- Did domestication take place once or on many independent occasions (Ladizinsky 1989: 387; Zohary 1989: 369; Blumler 1992: 100-2)?
- Was domestication primarily the result of biological processes (Binford 1968: 328-34; Flannery 1969: 75-6; Cohen 1977; Hayden 1981: 528-9; Rindos 1984) or the product of social changes (Bender 1978)?
- Can the domestication process be linked to major changes in the global ecosystem (Childe 1952: 25; Binford 1968: 334; Wright 1977; Byrne 1987)?
Most archaeologists now believe that a complex convergence of multiple factors (climatic changes, plant availability, preadaptive technology, population pressure, and resource stress) accounts for the emergence of agriculture 10,000 years ago in the southern Levant (Hole 1984: 55; Moore 1985; Henry 1989: 40-55, 231-4; McCorriston and Hole 1991: 60; Bar-Yosef and Belfer-Cohen 1992: 39). However, there is still little consensus regarding the rapidity of the shift or the importance to be accorded to any single factor.
Archaeological Evidence for the Domestication of Wheat
The earliest remains of domesticated plants are the charred seeds and plant parts found on archaeological sites that date to the beginning of the Neolithic period. Unfortunately, other evidence for the use of plant foods in the past rarely shows exactly which plants were eaten. For example, grinding stones, sickles, and storage pits all indicate increased plant use and storage during the Early Neolithic period (about 10,000 to 8,000 years ago), but they do not indicate which plants were processed (Wright 1994). In fact, such artifacts could have been used to process many kinds of plants and plant tissues, including many grasses, reeds, nuts, and tubers.
Following the discovery of Neolithic crop plants (Hopf 1969), archaeologists have employed many analytical techniques to determine (1) whether even earlier peoples also cultivated plants, (2) whether such earlier uses of plant resources would have resulted in domestication, and (3) whether the first farmers originated in the region(s) where domesticated wheat first was found. The ultimate aim of such a quest, of course, has been to resolve the question of whether the earliest charred remains of domesticated wheat actually indicate the first wheat farmers.
In aiming at an answer, one must know the plant resources used by preagrarian hunter-gatherers and how the cultivation practices of the first farmers differed from plant use by their predecessors (Hillman 1989; Hillman, Colledge, and Harris 1989: 240-1). This knowledge, however, has proved elusive, largely because most direct evidence for prehistoric human use of plants has decayed or disappeared. Tools and pits may have been used for processing a wide range of plants. Chemical residues that allow archaeologists to specify which plants were eaten during the Early Neolithic period and the preceding Natufian period seldom have been preserved or examined (Hillman et al. 1993). Microscopic studies of the sheen left on flint sickle blades indicate that peoples using these tools reaped cereals (Unger-Hamilton 1989; Anderson 1991: 550), although it is impossible to ascertain which species.
Chemical composition of human bone also provides limited clues to plant consumption. For example, the ratio of strontium to calcium (Sr/Ca) in Natufian and Neolithic skeletons indicates that some early farmers eventually relied more heavily on animal foods than did their immediate Natufian predecessors (Sillen 1984; Smith, Bar-Yosef, and Sillen 1984: 126-8; Sillen and Lee-Thorp 1991: 406, 408). None of these isotopic data, however, have come from the very first farming populations (Pre-pottery Neolithic A); furthermore, such analyses cannot identify the specific plants that the first farmers ate.
Neolithic sites with remains of domesticated wheat and other crops are the earliest known farming sites. But practices known to Neolithic farmers surely existed among their Natufian predecessors (Unger-Hamilton 1989) who for the first time in human history used large amounts of cereal processing equipment—grinding stones, sickle blades, storage pits—and lived year-round on one site (Bar-Yosef and Belfer-Cohen 1989: 468-70; Henry 1989: 195, 211-14, 219; Tchernov 1991: 322-9). Yet none of the Natufian sites excavated thus far have revealed domesticated wheat.
Furthermore, the presence of domesticated plants on Neolithic sites, more than any other evidence, has defined our perception of a major economic difference between the first Neolithic farmers and their hunter-gatherer predecessors. Natufians may indeed have cultivated cereals, although they never apparently domesticated them, and traditions of cereal cultivation in conjunction with other gathering and hunting strategies probably persisted long into the Neolithic era when cereal farmers shared the Near East with other groups of people who were not especially committed to cultivation.
A few exceptional excavations have recovered plant remains from pre-Neolithic sites, but most of these have not yet been fully analyzed. The site of Abu Hureyra, along the banks of the Middle Euphrates River in northern Syria, yielded an abundance of charred plant remains reflecting the harvest of many types of wild seeds and fruits: These were gathered primarily in the local environments of the Late Pleistocene—steppe and steppe-forest, wadi banks, and the Euphrates River valley bottom (Hillman et al. 1989: 258-9).
The plant economy of Abu Hureyra’s Epipaleolithic hunter-gatherers, however, does not appear to have led directly to farming. The site was abandoned at the time when farming began elsewhere (Moore 1975: 53, 1979: 68), and the evidence for a wide diversity of plants without evidence of intensive use of any particular one (Hillman et al. 1989: 265) is inconsistent with most models of cereal domestication (e.g., Harlan 1967; Rindos 1984; Henry 1989: 55, 216-17, 228; Hillman and Davies 1990: 212).
Instead, most models assume that cereal domestication followed intensive cereal exploitation by hunter-gatherers. At about the time that people abandoned Abu Hureyra, the sedentary inhabitants of nearby Tell Mureybet began to harvest two-seeded wild einkorn wheat and wild rye with unprecedented intensity (van Zeist and Bakker-Heeres 1984: 176-9; Hillman et al. 1993: 106). Although this type of wild wheat never developed into a domesticated plant (Zohary 1971: 239; van Zeist 1988: 58), the pattern of intensive cereal use at Tell Mureybet mirrors the type of economic pattern suggested for the Natufians from the southern Levant, where no plant remains from Epipaleolithic sites have been fully analyzed (Colledge 1991).
The southern Levant is where the earliest domesticated wheat appears. In the period known as the Pre-potter y Neolithic A (approximately 9,000 to 10,000 years ago2), the early farming site of Jericho (in the Jordan Valley) has two types of domesticated wheat grains, einkorn and emmer (Hopf 1969: 356, 1983: 581). Some of the oldest dates from Jericho can be questioned (Burleigh 1983: 760; Bar-Yosef 1989: 58), and domesticated wheat seeds from Jericho may actually be several hundred years younger than the oldest Neolithic radiocarbon dates (10,500-10,300 years ago) suggest.
Today Jericho lies at the edge of a spring whose outflow creates an oasis in the arid summer landscape of the Jordan Valley. This alluvial fan, created by winter streams flowing from the Judean hills, nourishes palms and summer crops in the midst of a shrubby wasteland, but the area looked different during the Early Neolithic. Most of the sediment accumulated around the site of the early farming village at Jericho has washed downslope since the Neolithic (Bar-Yosef 1986: 161), perhaps because the shady glades of wild trees—pistachio, fig, almond, olive, and pear (Western 1971: 36, 38; 1983)—were stripped from the surrounding hillsides thousands of years ago. The Neolithic inhabitants planted some of the earliest wheat ever farmed, and they depended on the supplemental water provided by the spring and flowing winter streams to ensure their harvests.
The farmers at Jericho necessarily managed water: Floods frequently threatened to damage their habitations and storage areas. They built terrace walls and dug ditches to divert the flow (Bar-Yosef 1986: 161; compare Kenyon 1979: 26-7) from their small round and lozenge-shaped houses with their cobble bases and mud-brick walls (Kenyon 1981: 220-1). Their apparent choice of supplementally watered land to grow wheat was unprecedented, for wild wheats had hitherto thrived on dry slopes at the edge of Mediterranean forest (Limbrey 1990: 46, 48).
The only other site that has yielded domesticated wheat from the same general era as Jericho is the site of Tell Aswad about 25 kilometers southeast of Damascus, Syria (Contenson et al. 1979; Contenson 1985). In the earliest midden layers of this prehistoric settlement, along the margins of a now-dried lake, archaeologists recovered domesticated emmer wheat along with barley and domesticated legumes such as lentils and peas. Any former dwellings had long since been destroyed, perhaps because structures consisted largely of wattle (from reeds) and daub (Contenson et al.1979: 153-5).
Today Tell Aswad lies outside the green Damascus oasis on a dusty, treeless plain occupied by the modern international airport, but its former setting was quite different. We know from charred seeds of marshy plants, historical accounts of the environment (van Zeist in Contenson et al. 1979: 167-8), and pollen studies (Leroi-Gourhan in Contenson et al. 1979: 170) that the lake once adjacent to the site was much larger; in addition, there were many wild trees adapted to a semiarid Mediterranean forest-steppe (pistachios, figs, and almonds). Pollen of species such as myrtle and buckthorn (Rhamnus spp.) may indicate rainfall greater than the annual 200 millimeters today (Leroi-Gourhan in Contenson et al. 1979: 170). Under wetter conditions, farmers were probably able to grow wheat and other crops. When it was drier, they probably used the extra moisture afforded by the lake and autumn flooding to grow wheats beside the lake shores.
Tell Aswad and Jericho are critical sites in the history of wheat agriculture. To be sure, we cannot be certain that the farmers who settled at the edge of Lake Ateibe (Tell Aswad) and near the spring feeding into the Jordan Valley at Jericho were the first people ever to grow domesticated wheat because archaeologists will never know for certain if earlier evidence awaits discovery elsewhere.
It is interesting to note, however, that contemporary evidence in adjacent regions suggests that people had not domesticated plants by 8000 B.C. In the Nile Valley of Egypt, for example, farming appears much later, around 5000 B.C. (Wenke 1989: 136), and in northern Syria on such early settlements as Tell Mureybet, people exploited wild, not domesticated, wheats and rye (van Zeist 1970: 167-72; van Zeist and Bakker-Heeres 1984: 183-6, 198; Hillman et al. 1993: 106). Recent research in the Taurus Mountains of southeastern Turkey has focused on early settled communities that apparently were not intensively exploiting wild or domesticated cereals (Rosenberg et al. 1995). Southern Mesopotamia, where the first cities emerged, saw agricultural settlements only in later times (Adams 1981: 54), and the surrounding mountains continued to support pastoralists and hunter-gatherers long after farming appeared in the southern Levant.
Botanical and ecological evidence for the domestication of wheat and its differentiation into many species also partially contributes to an understanding of where and when the first domestication occurred. Many different morphological forms of wheat appear in the archaeological record, even as early as the Neolithic deposits at Jericho, Tell Aswad, and Tell Mureybet along the northern Euphrates River (van Zeist 1970: 167-72; van Zeist and Bakker-Heeres 1984: 183-6, 198). The different forms of wild and domesticated wheats are of incalculable value to archaeologist and botanist alike, for these different plants that can be distinguished from archaeological contexts allow botanists and ecologists to identify wild and domesticated species and the conditions under which they must have grown.
The forms recognized archaeologically, moreover, may not always conform to wheat classification schemes used by modern breeders and geneticists. Wheat classification is complex and confusing, for hundreds of varieties have appeared as wheat farming spread around the world. Although many different kinds of wheat can be readily distinguished by their morphological characteristics (such as red or black awns, hairy glume keels, spikelet density), other varieties can cross-fertilize to combine characters in a perplexing array of new plants. The great variability in the visible characteristics of wheats has led to confusion over how to classify different species—a term employed in its strictest sense to describe reproductively isolated organisms (Mayr 1942; Baker 1970: 50-1, 65-6). In the case of wheat, many botanists commonly accept as distinct species morphologically distinct types that can readily cross to form fertile hybrids with other so-called species (Zohary 1971: 238).
Because botanists rely on both morphological and genetic characteristics to identify different wheats, classificatory schemes (of which many exist, for example, Percival 1921; Schiemann 1948; Morris and Sears 1967; Löve 1982) must take both aspects into account (Zohary 1971: 236-7; but compare Baker 1970). Using morphological traits, taxonomists originally split wild and cultivated wheats into at least a dozen different taxa, many of which are highly inter-fertile. Geneticists, however, maintain that all domesticated wheats belong to four major groups that produce only sterile crosses; furthermore, they include the wild grass genus, Aegilops, in the Triticum genus, since several taxa of wild Aegilops contributed chromosome sets (genomes) to domesticated wheats by crossing with wild wheat plants (Zohary 1971: 236).
Many of the wheats distinguished by taxonomists, however, lose their identifying genetic signatures when charred, abraded, and preserved for thousands of years in archaeological sites. Because fragile genetic material only recently has been demonstrated to have survived this process (Brown et al. 1993), morphological features that can be used to distinguish different wheat types have made traditional taxonomic schemes (based on morphology) of great value to archaeologists. Furthermore, some of the major behavioral characteristics of cultivated and wild wheats do have morphological correlates that endure in the archaeological record. These features also reflect significant events in the domestication of wheat.
The most significant of these morphological features is rachis (segmented stem) durability. Wild wheats and wild Aegilops, a morphologically distinct grass genus with species capable of crossing with many wheats, have a rachis capable of shattering, once the grains have matured, into pieces bearing one or two grains. These pieces, or spikelets, taper at their bases and carry stiff hairs that act as barbs to facilitate the spikelets’ entry into cracks in the soil. In wild wheats, grains are tightly enclosed in tough glumes that protect them from predation.
In domesticated wheats, these features vanish. The rachis fails to shatter when ripe, a feature particularly important to humans who harvest using sickles—the tools introduced by Natufian and early Neolithic groups (Hillman and Davies 1990: 172-7). In the relatively pure stands of wild wheats, at the margins of Mediterranean oak forests where agriculture began, harvesting methods would fundamentally affect the domestication process (Harlan 1967, 1989; Bohrer 1972; Wilke et al. 1972: 205; Hill-man and Davies 1990: 172-7). Harvesters use fairly violent motions when equipped with sickles or when uprooting plants to harvest straw and seed. These methods tend to shatter ripe ears, leaving for collection either immature seed (unfit for germination the following year if replanted) or relatively rare genetic mutants with tough rachises. Although these rare plants reproduce poorly in the wild, they are ideal for cultivation, as ripe seed can regenerate if replanted (Helbaek in Braidwood and Howe 1960: 112-13). By unconscious selection (Rindos 1984: 86-9) for a tough rachis gene, harvesters may replace a wild population with a domesticated one in as few as 20 to 30 years (Hillman and Davies 1990: 189).
Wild and domesticated cereals often can be distinguished when examining rachis fragments in archaeo-logical plant remains (for example, Bar-Yosef and Kislev 1986; Kislev, Bar Yosef, and Gopher 1986: 198-9; compare Bar-Yosef and Belfer-Cohen 1992: 37-8). The earliest known domesticated wheats from Tell Aswad exhibit tough rachises (van Zeist and Bakker-Heeres 1982: 192-6). At the same period, the wheats intensively harvested along the Middle Euphrates River at Tell Mureybet (van Zeist 1970: 167-72; van Zeist and Bakker-Heeres 1984: 183-6, 198) and in northern Mesopotamia at the site of Qeremez Dere (Watkins, Baird, and Betts 1989: 21) remained wild, perhaps partly because of a harvesting technique that favored the proliferation of brittle-rachis types in the population. For example, beating wild grass heads over baskets to collect seed was a technique widely employed in many parts of the world where no domestication occurred (Bohrer 1972: 145-7; Wilke et al. 1972: 205-6; Harlan 1989; Nabhan 1989: 112-18). Although baskets and wooden beaters have a low probability of surviving in archaeological sites in the Near East, the remarkable paucity of sickle blades at Tell Mureybet (Cauvin 1974: 59) would support a suggestion that people may have harvested wild cereals by a different method from that used at Jericho and Tell Aswad, where sickle blades are more common.
The results of modern genetic studies have also contributed incomparably to disentangling the history of domesticated wheat. In an effort to improve modern strains of bread wheat and to discover new genetic combinations, biologists have compared the genetic signatures of different varieties, types, and species of wheats. Genetic differences and similarities have allowed specialists to trace relationships among various forms of wild and domesticated wheats and to determine which wild wheats were ancestral to domesticates.
All of the relationships described in Figure II.A.10.1 and Table II.A.10.2 have been confirmed by genetic tests (Zohary 1989: 359). Of particular importance to domestication, the work of H. Kihara has largely defined the cytogenetic relationships between emmer, durum, and hexaploid wheats (Lilienfeld 1951). Domesticated emmer wheat shares close genetic affinities with its wild progenitor (Triticum dicoccoides = Triticum turgidum subsp. dicoccoides) and is largely a product of unconscious human selection for a tough rachis. Durum wheats and rivet wheats likewise received their 2 chromosome sets from wild emmer (Zohary 1971: 239) and probably are secondarily derived from domesticated emmer through selection for free-threshing characteristics, larger seeds, and various ecological tolerances (for example, Percival 1921: 207, 230-1, 241-3).
Hexaploid wheats, which belong in a single cyto-genetic taxon (Zohary 1971: 238; Zohary and Hopf 1993: 24), have no wild hexaploid ancestors: They emerged as a result of a cross between domesticated tetraploid wheats (which may or may not have been free-threshing) and a wild grass native to continental and temperate climates of central Asia (Zohary and Hopf 1988: 46). This implies that hexaploid wheats emerged only when tetraploid wheats spread from the Mediterranean environment to which they were adapted. From archaeological evidence of the spread of farming, one assumes that hexaploid wheats appeared after 7500 B.C.33 True bread wheats with loose glumes probably came from spelt ancestors, but only two slight genetic changes produce loose glumes (Zohary and Hopf 1988: 46), implying that the mutations may occur easily and become rapidly fixed in a domesticated population.
Cytogenetic studies also have suggested that domestication occurred in only one population of wild wheats, from which all modern conspecific cultigens (of the same species) are derived. All the varieties and species of tetraploid wheats have the same basic genetic constitution as wild emmer wheat (AABB genomes)4 rather than timopheevii wheat (AAGG). This indicates that if multiple domestications had occurred, timopheevii wheat, which is morphologically indistinguishable from wild emmer, would have been systematically ignored. A more parsimonious explanation is that of Daniel Zohary (1989: 369), who suggests that emmer wheat was domesticated once and passed from farming community to community (see also Runnels and van Andel 1988). Archaeological evidence on Crete and in Greece (Barker 1985: 63-5) indicates that fully domesticated wheats were introduced to Europe from the Near East (Kislev 1984: 63-5). An alternative hypothesis—that hunter-gatherers in Europe independently domesticated emmer and einkorn from native wild grasses (Dennell 1983: 163)—has little supporting evidence. Botanists using cytogenetic evidence, however, may more easily recognize evidence for single domestication than for multiple events, genetic traces of which can be obscured by other biological and historical processes (Blumler 1992: 99, 105).
Ecology of Wheats
Perhaps it will never be possible to determine unequivocally whether wheat species were domesticated in one place or in several locations. Nevertheless, the ecological constraints limiting the growth of different species, varieties, and forms of wild and domesticated wheats narrow greatly the possibilities of where and under what ecological circumstances wheat may have been domesticated. Ecological constraints have been examined both on a macro and micro scale, and both scales contribute significantly to our understanding of wheat domestication.
On a macro scale, the geographic distributions of discrete species or varieties of wild wheats provide ecological ranges within which, or indeed adjacent to which, researchers locate wheat domestication and the origins of agriculture (Harlan and Zohary 1966). Using modern wild wheat distributions, botanists and archaeologists have singled out the southern Levant and Taurus range as the most likely source of domesticated emmer and einkorn (Harlan and Zohary 1966; Zohary 1971: 239-42), although bread wheats may have quickly evolved from spelt wheat somewhere in the Caspian region (Zohary 1971: 244; Zeven 1980: 32). Timopheevii wheats represent merely a later independent domestication of tetraploids in eastern Anatolia and Georgia. The conclusions of Vavilov, Braidwood, Flannery, Harlan, and D. Zohary depend greatly on modern geographical distributions of wild wheats.
Nevertheless, a serious problem in using the modern ranges of wild wheats is the assumption that these ranges reflect the former natural extent of wild species. In the past 10,000 years in the Near East, climates as well as human land use patterns have fluctuated. Grazing, deforestation, and suppression of natural forest fires have had a profound effect on vegetation (Naveh and Dan 1973; Naveh 1974; Le Houérou 1981; Zohary 1983), altering not only plant distributions but the character of entire vegetation zones (McCorriston 1992).
Water, light, and soil properties determine growth on a local scale within the geographic ranges of wheats. Wheat plants, like many other crops, require land free of competition from established plants that tap water and block light and where the seed may embed itself in the earth before ants or other predators discover it (Hillman and Davies 1990: 164; Limbrey 1990: 46).
Truly wild einkorn and emmer typically thrive on the open slopes at the margins of scrub-oak forests: They are poor competitors in nitrogen-rich soils typical of weedy habitats, such as field margins, habitation sites, and animal pens (Hillman and Davies 1990: 159, 160; Hillman 1991; McCorriston 1992: 217; compare Blumler and Byrne 1991). These latter sites readily support domesticated glume wheats (emmer and einkorn). Their wild relatives prefer clay soils forming on “basalt or other base-rich fine-grained rocks and sediments under warm climates with a marked dry season” (Limbrey 1990: 46). Indeed, some evidence suggests that wild wheats were first harvested from such soils (Unger-Hamilton 1989: 100; Limbrey 1990: 46).
Early farming sites like Jericho, Tell Aswad, and Mureybet, however, were located beside alluvial soils in regions of low rainfall where supplemental watering from seasonal flooding and high water tables would have greatly enhanced the probability that a wheat crop would survive in any given year. If the wild wheats originally were confined to the upland basaltic soils, they must have been deliberately moved to alluvial fields in the first stages of domestication (Sherratt 1980: 314-15; McCorriston 1992: 213-24; compare Hillman and Davies 1990). Removal of a plant from its primary habitat often causes a genetic bottleneck (Lewis 1962; Grant 1981) whereby the newly established population is fairly homogeneous because its genetic ancestry comes from only a few plants. This tends greatly to facilitate domestication (Ladizinsky 1985: 196-7; Hillman and Davies 1990: 177-81).
The Spread of Domesticated Wheats from the Near East
Deliberate planting of wheat in new habitats (supplementally watered alluvial soils) is one of only a few known events in the earliest spread of wheat farming. Archaeologists understand very poorly the processes and constraints that led to the spread of wheat into different environments and regions of the Near East. They understand far better the spread of agriculture in Europe because of research priorities set by Childe and other Western archaeologists who identified the arrival of domesticated wheat in Europe around 6000 B.C. (Barker 1985: 64; Zohary and Hopf 1988: 191) as an event of critical significance in the history of humankind (viewed from a European perspective).
Thus, in contrast with the Near East, in Europe the process of transition from different styles of hunting and gathering to a predominantly agricultural economy and the adaptation of crops such as wheat to new environments has received intense archaeological and theoretical consideration. The progress of Neolithic settlement across the plains of Greece and Italy, up the Balkans and into the river basins of central and eastern Europe, across the forested lands, and into the low countries encompasses many regional archaeologies and localized theoretical explanations. Wheat accompanied the Neolithic nearly everywhere in Europe, although by the time farming took hold in Britain around 3500 B.C., the cultivated varieties of einkorn, emmer, and spelt must have tolerated colder and longer winters, longer daylight during the ripening season, and greatly different seasonal rainfall than that of the Mediterranean lands from which the crops originated.
Wheat also spread out of the Near East to Africa, where it could be found in northern Egypt after 5000 B.C. (Wenke 1989: 136; Wetterstrom 1993: 203-13). With other introduced crops, wheat fueled the emergence of cultural complexity and replaced any indigenous attempts at agriculture initiated in sub-tropical arid lands to the south (Close and Wendorf 1992: 69). Because most wheats require cool winters with plentiful rain, the plant never spread widely in tropical climates where excessive moisture during growing and ripening seasons inhibits growth and spurs disease (Lamb 1967: 199; Purseglove 1985: 293). But the grain also spread to South Asia where as early as 4000 B.C. hexaploid wheats were cultivated at the Neolithic site of Mehrgarh (Pakistan) (Costantini 1981). By the third millennium B.C., the Indus Valley civilization, city-states in Mesopotamia, and dynastic Egypt all depended on domesticated wheat and other cereals.
In the sixteenth century, colonists from the Old World brought wheat to the New World: The Spanish introduced it to Argentina, Chile, and California where the cereal flourished in climates and soils that closely resembled the lands where it already had been grown for thousands of years (Crosby 1986; Aschmann 1991: 33-5). The political and social dominance of European imperialists in these new lands and the long history of wheat farming in the Old World—where crop and weeds had adapted to a wide range of temperature, light, and rainfall conditions (Crosby 1986)—largely accounts for the fact that wheat is one of the world’s most significant crops today. Its domestication is a continuing process, with yearly genetic improvements of different strains through breeding and new gene-splicing techniques (Heyne and Smith 1967).
Botanical and archaeological evidence for wheat domestication constitutes one of the most comprehensive case studies in the origins of agriculture. Some of the issues discussed in this chapter remain unresolved. For many of the domesticated wheats, including the primitive first wheats (einkorn and emmer), the integration of botanical and archaeological evidence indicates where and approximately when people entered into the mutualistic relationship that domesticated both humans and their wheats. Less is understood about the origins of hexaploid wheats, largely because the results of archaeological investigations in central Asia have long been inaccessible to Western prehistorians.5 From botanical studies, however, we can (1) trace the wild ancestors of modern wheats, (2) suggest that isolated events led to the domestication of each species, and (3) reconstruct the environmental constraints within which the first farmers planted and reaped.
Yet, some debate still continues over the question of how wheat was domesticated. Did people in many communities independently select easily harvested plants from weedy wild emmer growing on their dump heaps (Blumler and Byrne 1991)? Or did they, as most believe, deliberately harvest and sow fully wild wheats and in the process domesticate them (Unger-Hamilton 1989; Hillman and Davies 1990; Anderson-Gerfaud, Deraprahamian, and Willcox 1991: 217)? Related questions, of course, follow: Did this latter process happen once or often, and did the first true farmers move these wild plants to seasonally flooded or supplementally watered soils as part of the domestication process?
Scholars also continue to discuss why people began to domesticate wheat in the first place some 10,000 years ago as part of the larger question of the origins of agriculture. Although most agree that many factors were involved in the Near East and that they converged at the end of the Pleistocene, there is no agreement about which factor or combination of factors was most important (for example, Graber 1992; Hole and McCorriston 1992). But because there are no written Neolithic versions of the original crop plants (compare Hodder 1990: 20-1), it is up to us—at the interdisciplinary junction of archaeology and botany—to continue to reconstruct this evolutionary and cultural process in human history.