Milk, Wool, and Traction: Secondary Animal Products

views updated


Zooarchaeologists distinguish between primary animal products, such as meat, bone, and marrow, and secondary products, such as milk, wool, and traction (animal labor). Primary products, also known as slaughter products, require the death of the animal and thus can be harvested only once. Secondary products are extracted from the living animal. This is a crucial distinction because secondary products permit a higher yield from the same number of animals. While people can acquire primary products from either wild or domestic animals, secondary products normally are available only from domestic herds. Some researchers have suggested that secondary products may have been the motivation for animal domestication, but the evidence does not support their use to any significant extent until considerably later than the time when animals were domesticated.

It is difficult to study the use of secondary products in prehistory because they typically are not preserved in the archaeological record. Most primary products leave relatively direct evidence in the form of animal bones. Bones are by-products of meat consumption, and bone breakage patterns can indicate their use for marrow. For the most part, secondary product use must be approached indirectly. Sometimes artistic representations portray these products or their use, but it is quite possible for societies to use them without leaving a pictorial record. Indeed, with the exception of the use of animal traction to draw wheeled vehicles, the first artistic depictions of secondary products generally are much later than their earliest use. Thus, the most widely employed method to detect the use of secondary products is the demographic study of the animal bone assemblage.

Such a study focuses on the differing herding strategies that are necessary to achieve significant production of secondary products. If meat is the main concern and secondary product use is absent or insignificant, most males typically are slaughtered at a juvenile or subadult stage, when growth slows and more feed produces little additional weight gain. If herders want milk, they need lactating females, and they must limit competition from the infant animals through slaughter or early weaning. Thus, most males likely will be slaughtered as infants, and the herd will consist mainly of adult females. Both males and females produce wool, so when wool is the desired secondary product, the herd will consist of both sexes, and most animals will live into adulthood. Traction (pulling plows or vehicles) also requires adults, and males or castrates may be better suited to the task. Each strategy creates a distinctive kill-off pattern, or mortality profile. Age and sex information can be derived from the study of the animal bone remains to reconstruct these strategies.


All mammals produce milk, so it is certainly possible that ancient herders used dairy products from the beginning of animal domestication. There are real advantages to dairy products. Animal milk is a good substitute for human milk when a mother dies or cannot produce adequate milk. Dairy products provide a sustainable source of protein and fat that substantially enhances the productivity of the herd. For example, Paul Halstead has calculated that a Greek Neolithic (early farming) village of 40 to 240 inhabitants could meet its caloric needs with 2,400–14,400 sheep if the villagers ate only the meat, but they would need only 1,000–6,000 sheep if they used the milk as well. Dairying thus could be used to reduce herd size and devote more land to agriculture or, alternatively, to keep more animals alive for their wealth value while still deriving protein and calories from the herd. Moreover, processed milk products, such as cheese, can be stored, unlike fresh meat or milk.

There are also drawbacks to dairy production, however. Even with today's electric milking machines, dairy farmers know that maintaining dairy herds is a labor-intensive undertaking. The animals need to be milked regularly (at least once a day) to maintain production. (For maximum yield, modern dairy farmers milk two or even three times daily, at regular intervals.) To accomplish this, either the milking animals must be kept near the settlement—requiring fodder because there probably will not be enough pasture nearby—or a remote dairy-processing camp must be set up in the mountain pastures and some people will have to spend considerable time there. Any milk that is not consumed immediately must be processed and stored.

Lactose intolerance poses a further difficulty. Lactose is the form of sugar found in milk. Mammals normally drink milk only in infancy. The ancestral condition in humans, just as in other mammals, is to lose the ability to digest lactose after infancy as a result of shutting down production of the enzyme lactase. This is still true of most living humans. Human populations with a long history of drinking milk (in East Africa and central and northern Europe) have evolved the capacity to continue producing lactase throughout life. It is safe to assume that the people who first domesticated herd animals would not have been able to digest their milk in adulthood. Even people with lactose intolerance are able to consume dairy products if they are fermented (as are cheese and yogurt), which significantly reduces the lactose content. Little is known about the origins of these fermented products, but the techniques to produce them could not have been developed until after milk already was in use. Most present-day populations that use dairy products, in fact, are not lactose tolerant and rely mainly on fermented dairy products.

Lactose tolerance seems to have developed where it was important to use fresh milk: in arid East Africa, where the water content may have been useful, and in northern Europe, where the lactose itself was helpful. Lactose enhances calcium absorption and helps prevent rickets in places where vitamin D intake from sunshine or fish is inadequate. In sum, an extended period of cultural or biological adaptation or both would be necessary before animal milk could make a significant contribution to the adult diet.

While milk may have been consumed occasionally from the earliest days of herding (sheep and goats were domesticated c. 8000 b.c.), there is little sign that it was used to any significant extent until much later. The mortality profiles of early herds reflect a meat-oriented herding strategy. At this point there are too few mortality profiles to gain a clear picture of when dairy achieved prominence in various regions. Current evidence suggests not simple diffusion from a center of origin or a single horizon of change, but more piecemeal adoption according to local conditions. Mortality profiles indicate dairy use by about 6000 b.c. in northern Italy (sheep and goats), 5800 b.c. in western Iran (sheep and goats), 5500 b.c. in Greece (sheep), 4000 b.c. in the northern Balkans and the alpine forelands of Switzerland (sheep, goats, and cattle), and 1000 b.c. in Britain but quite possibly as early as 3500 b.c. (cattle).

The mortality evidence has drawbacks, however. The mortality profiles from archaeological sites rarely are a close match to the idealized meat, milk, or wool herding strategies. This probably results from a combination of differential loss of the bones of young animals, which are softer and more fragile, and the fact that prehistoric herders, who were not involved in market economies, practiced less-specialized forms of herding than those seen today. A further difficulty relates to the let-down reflex. Milk is held in the mammary glands until it is "let down" into the milk ducts. Lactating females generally let down their milk through a hormonal reaction in response to the suckling of their infants. Not all animals let down their milk if their own offspring is not present, and some researchers have suggested that this would have been true of early domesticates. There are many stratagems for "tricking" the mother into letting down, however. These schemes include the use of surrogate offspring and the use of a tube to blow air into the mother's vagina (triggering a hormonal reaction). Experts disagree over how great a problem this would have been for early herders. Those who see it as a major impediment suggest that offspring would have had to be kept alive and milk shared with them, which would clearly alter the mortality profile from what is expected for dairy. One instructive study examines medieval Irish cattle mortality profiles. In this case documentary evidence indicates that cattle were kept primarily for dairy, but the mortality profile of the archaeological remains shows later kill-off than expected, between one and two years of age. At best, dairy mortality profiles indicate a herd structure that would support the use of milk but do not provide evidence that it actually was used.

Given these ambiguities, it is useful to seek other lines of evidence. One study of sites in Israel takes a more direct approach. Studies of present-day livestock show that lactating female sheep suffer calcium loss in their bones if they are nutritionally or otherwise stressed. Reasoning that ancient dairy animals would have been stressed at least sometimes, the researchers used X rays to measure the bone mass of sheep and goat specimens from several sites spanning the Neolithic, Chalcolithic, and Bronze Age. While there was some local variation, such calcium loss does not appear before the Chalcolithic (fifth millennium b.c.) and then intensifies in the Middle Bronze Age. So far, this promising but labor-intensive method of analysis has not been applied elsewhere. It is encouraging, however, that it accords well with the mortality profile evidence from the Chalcolithic sites.

Artifacts also have provided evidence for dairy use. Bowls resembling those now used in dairying have been found at Swiss Neolithic sites, where mortality profiles also suggest the use of milk. Ceramic sieves from the Early Neolithic Linear Pottery culture of central Europe (c. 5500 b.c.) may have been used in cheese making (fig. 1). Such arguments rest on analogies to modern uses of artifacts, however. Chemists now have developed a more direct method. It is possible to detect and identify lipid and protein residues from milk on ancient pottery. Analyses of lipid residues on pottery fragments have provided chemical evidence for the widespread use of milk products during the Neolithic in Britain, about 4100–3500 b.c. The earliest artistic depiction of milking is on a Sumerian cylinder seal from c. 3300 b.c., probably well after the inception of dairy use in the Near East.


Wild sheep (Ovis orientalis) are hairy rather than woolly, and early domestic sheep would have been the same. These sheep have a short woolly undercoat in the winter, which is shed in the spring. Under domestication, this woolly layer became longer and was retained year-round while the outer hair (or kemp) was reduced. The pigment in the coat also was lost. Thus, the development of wool was necessary before sheep could be managed for wool production.

Wool is a perishable material that is rarely preserved in archaeological deposits. There are, however, occasional finds of textiles or textile impressions or other preserved fibers. The earlier finds, from the Upper Palaeolithic through the Neolithic, are all vegetable fibers. In the Neolithic of Europe and the Near East, these fibers usually are flax (linen). Wool appears only c. 3000 b.c. in the Near East and about 500 years later in Europe.

It is difficult, but in some cases possible, to distinguish male and female sheep bones other than the relatively fragile and archaeologically rare horn cores. This problem is compounded by the challenge of distinguishing sheep and goat bones, yet it is primarily sex ratios (the presence of adult males in numbers nearly equal to females) that differentiate wool from dairy mortality profiles. As a result, there are few analyses that can pinpoint wool use on the basis of mortality profiles, and researchers can say only that demographic evidence generally supports the picture derived from fiber remains.

There is more indirect evidence from the bones, however. At the beginning of the Bronze Age, a new population of larger sheep abruptly appeared in Europe, probably spreading rapidly from the steppe zone of eastern Europe. Sheep also became more common in temperate Europe at this time. Given that Bronze Age figurines seem to represent woolly sheep, many researchers believe that these large steppe sheep were the first woolly sheep to reach Europe, largely replacing the earlier hairy sheep. It is worth noting that this pattern of rapid spread differs from the pattern of dairy use, which appeared more patchily and over a much longer period of time. Of course, the situation is somewhat different. In this case a preexisting textile industry based on flax could readily adopt wool, and the switch depended on a new kind of animal rather than simply different herding practices.


Harnessing animals to supplement human muscle power often is hailed as a critical step in the intensification of human energy use. Animal-drawn carts and plows permit higher agricultural yields and facilitate their transport both locally and over long distances. Chariots drawn by horses or donkeys also transformed warfare. There are many ways of using animal labor, but here the focus is on transport and plowing. In prehistoric Europe cattle and horses mainly performed these functions.

Domestic horses made their first appearance in most of Europe at this time. While horses were at times a major meat source on the Eurasian steppe, in most of Europe their adoption appears to have been based primarily on their role in transport as well as their wealth and status value. This is not to say that horses were never eaten, but they did not form a significant part of the diet, and there is no indication that they were raised primarily for meat.

In the case of cattle, an animal already long used in Europe for meat and by this time probably also for milk, traction was a new role, in addition to providing food. The use of cattle for traction meant that more animals were allowed to reach adulthood. Bulls or oxen (castrated bulls) probably were used for this purpose, although female cows pull carts in the modern-day Balkans. Traction is more difficult than dairy or wool use to detect in mortality profiles because only a few animals might have been kept for this purpose (and, like the Balkan cows, they might have been used for meat and milk as well). Animal bone studies can detect the presence of oxen. Castration tends to alter the shape of the horns (reflected in the bony horn cores, which are preserved, whereas the keratinous horn sheath usually is not). It also affects the growth pattern, so that limb bones tend to be longer and narrower than they are in intact males. Recognizing oxen, however, depends on finding a reasonable number of intact horn cores and limb bones. Particularly in the Neolithic, animal bones often were processed heavily for their marrow and fat content, leaving them highly fragmented. Thus mortality profiles do not provide a clear picture of the inception of the use of cattle for traction.

Another approach is to examine changes in the bones themselves. Extensive use for traction stresses the bones and joints of the animal, causing remodeling of the bones and such pathological conditions as osteoarthritis. Studies applied to faunal remains of the northern Balkans have suggested the use of cattle in traction (probably plowing) in the Late Neolithic, c. 4500 b.c. Other studies of later and modern cattle have developed criteria for the alterations caused by use in traction, but they have not yet been applied widely to early animal bone assemblages.

Other evidence has been brought to bear on the appearance of plowing in the archaeological record. The plows themselves probably were made of perishable materials, such as wood (although some Late Bronze Age plows are preserved in bog deposits). Plow marks, however, often are preserved below burial mounds in northern Europe in the Late Neolithic and Bronze Age (from c. 4000 b.c.). This probably was not simply a result of happening to build a mound on a plowed field but rather a ritual turning or penetration of the earth as part of the funerary ceremony itself. It is also most likely the ritual significance of plowing that has led to its frequent representation in Bronze Age rock art in much of western Europe (fig. 2). In any case, these are clear indications that plowing was practiced by 2500 b.c. The location of sites on heavy soils where they had not been found previously also has been seen as indirect evidence for plowing in the Late Neolithic of the northern Balkans, c. 4500 b.c.

In sum, much detail remains to be filled in, but drawing on the various lines of evidence it is reasonable to suggest that plowing began in southeast Europe in about 4500 b.c. and was practiced in northwest Europe by 4000 b.c. Clearly, plowing was entrenched across Europe by 2500 b.c., probably earlier. Thus, the plow, too, may have spread fairly rapidly, although somewhat earlier than wool.

Wheeled vehicles, which are inherently mobile, seem to have spread even faster. On current evidence, they appeared at about the same time in Europe and the Near East, c. 3500 b.c. (fig. 3). At this point wheeled vehicles are depicted in pictographs and models in Mesopotamia and surrounding regions. Likewise, in northern Germany a burial mound covers a set of wagon ruts, and in Poland a depiction of a wagon appears on a pot of the Middle Neolithic Funnel Beaker culture. To date, the earliest evidence of wheeled vehicles on the eastern European steppe is slightly later, in about 3100 b.c., but perhaps earlier finds will come to light, as this seems one possible route connecting Mesopotamia and the northern European plain. All of these early vehicles are slow, four-wheeled carts, apparently drawn by cattle. At this same time burials of pairs of cattle appear in eastern and northern Europe, probably yoked pairs sacrificed as part of ceremonies.

There is considerable debate concerning whether horses were ridden before or after they were used to draw carts and chariots. It is clear, at least, that the domestic horses that reached Europe from the steppe zone to the east were not used primarily for food and presumably were for some kind of transport. Animal bone remains at archaeological sites suggest that they were rare and probably were kept only by the elite classes that emerged in the fourth millennium b.c. A few horses appeared in graves of the Tiszapolgár culture of the Carpathian Basin shortly before 4000 b.c. They then disappeared, so perhaps they were isolated imports that did not establish a local breeding population. They turned up again in this area c. 3500 b.c. and slightly later in northern Europe. They did not become common throughout Europe until the Early Bronze Age, in about 2500 b.c.


Gordon Childe, in a metaphor of lasting power, characterized the major economic, social, and ideological changes accompanying the origins of agriculture as the Neolithic Revolution. In an analogy to this concept, Andrew Sherratt has proposed a similarly crucial Secondary Products Revolution (SPR). The premise of the SPR is that dairy, wool, and traction appeared at roughly the same time in the Near East and Europe and that the use of products derived from living animals rendered animal husbandry dramatically more productive, with profound consequences.

In the SPR model, dairy, wool, and traction are linked into an integrated system quite different from the mixed farming model (with animals raised only for meat) that preceded it. Plowing increased agricultural productivity by permitting cultivation of larger areas and the use of heavier (and often more fertile) soils. Animal transport facilitated agriculture and made it possible for cities to draw their sustenance from a larger surrounding area, promoting economic integration at a regional level. In this view, it is no accident that animals used primarily for transport (horses, donkeys, and camels) were domesticated at about this time.

Wool provides a valuable, nonperishable, and easily transportable product that can be raised in areas that are marginal for agriculture. This probably contributed to the development of specialized pastoralism. The protein, fat, and calories of dairy products offered a source of animal nutrients that was an alternative to meat. Thus, herders could afford to keep animals alive for wool production and traction. Use of living animal products made the animals more valuable, adding to their overall worth. This value encouraged raiding and so may have contributed to increased warfare.

Comparative studies of contemporary societies have shown that plow agriculture leads to a greater investment in a particular plot of land, and, in this context, land tenure and inheritance rules take on new significance. In general, women perform most of the labor in hoe agriculture, whereas men do most of the work in plow agriculture. If this was true in prehistoric Europe, it means that this period marked a major change in gender roles. Rock art that shows men plowing supports this argument, although the association of ards (early plows) with female-associated artifacts in bog hoards in Late Bronze Age Denmark might mean that women plowed in some cases. If the generalizations from contemporary societies are followed, hoe agriculture typically is associated with matriliny (tracing descent through the female line) and plow agriculture with patriliny (tracing descent through the male line). Together with the new importance of wealth in both land and livestock, providing higher stakes for inheritance, the kinship system may have experienced considerable change.

These changing roles may have led to an imbalance in power between men and women. Sherratt suggests that female labor may have been devoted increasingly to the weaving of woolen textiles; women also may have been involved in dairy production. Sherratt also believes that men came to dominate the economy, whereas women were relegated to the domestic sphere. Growing textile production for exchange, however, may have given women considerable economic power.

The enhancement of land and livestock wealth brings with it greater opportunity to create inequalities of wealth and power. Along with improved transport, Sherratt believes that this underlies another of Childe's concepts: the Urban Revolution, or the rise of the first cities in the Near East. He also suggests that the use of animal traction had important long-term effects. Regarding traction as the first step in the mechanization of agriculture, leading to further mechanization, Sherratt claims that it ultimately explains why the Industrial Revolution happened in the Old World rather than the New World.

Many researchers have critiqued the concept of the SPR, particularly the claim that dairy, wool, and traction appeared more or less simultaneously. As already noted, dairy, in particular, may have a rather longer history, and the various elements of the SPR seem to have spread at different rates and perhaps by different routes. Nevertheless, most researchers agree that there seemed to be a significant intensification of secondary product use starting in about 3500 b.c. There also are signs of changes in settlement patterns, inequality, and gender roles at roughly this time and indications that herding may have taken on greater importance in relation to plant agriculture. It is important to remember, however, that this was not a unified phenomenon across Europe but instead was locally variable. Wheeled vehicles seemed to be more important in eastern and northern Europe, whereas plows were more significant in western Europe, for example.

Moreover, as is often the case, the direction of causality is not clear. Sherratt thinks that the use of secondary products drove the other changes and that secondary product use, in turn, was necessitated by population growth that required intensified food production. Because, however, secondary products permitted herders to slaughter fewer animals, it may be that it was not the secondary products that converted livestock into wealth but the wealth value of living animals that motivated the use of secondary products. In any case, extensive use of these living animal products had wide-ranging consequences for the societies keeping the animals.

See alsoDomestication of the Horse (vol. 1, part 4).


Amoroso, Emmanuel C., and Peter A. Jewell. "The Exploitation of the Milk-Ejection Reflex by Primitive Peoples." In Man and Cattle. Edited by Arthur E. Mourant and Frederick E. Zeuner, pp. 126–137. London: Royal Anthropological Institute of Great Britain and Ireland, 1963.

Bakker, Jan Albert, Janusz Kruk, Albert E. Lanting, and Sarunas Milisauskas. "The Earliest Evidence of Wheeled Vehicles in Europe and the Near East." Antiquity 73 (1999): 778–790.

Bartosiewicz, László, Wim Van Neer, and An Lentacker. Draught Cattle: Their Osteological Identification and History. Tervueren, Belgium: Musée Royale de l'Afrique Centrale, 1997.

Bogucki, Peter I. "Animal Traction and Household Economies in Neolithic Europe." Antiquity 67 (1993): 492–503.

——. "The Antiquity of Dairying in Temperate Europe." Expedition 28, no. 2 (1986): 51–58.

Bökönyi, Sándor. History of Domestic Mammals in Central and Eastern Europe. Translated by Lili Halápy. Budapest, Hungary: Akadémiai Kiadó, 1974.

Childe, V. Gordon. Man Makes Himself. New York: New American Library, 1951.

Copley, M. S., et al. "The Earliest Direct Chemical Evidence for Widespread Dairying in Prehistoric Britain." Proceedings of the National Academy of Science 100, no. 4 (2003): 1524–1529.

Craig, Oliver E., Jacqui Mulville, Michael Parker Pearson, Robert J. Sokol, Keith Gelsthorpe, Rebecca Stacey, and Matthew J. Collins. "Detecting Milk Proteins in Ancient Pots." Nature 408, no. 6810 (2000): 312.

Gibbs, Liv. "Identifying Gender Representation in the Archaeological Record: A Contextual Study." In Reader in Gender Archaeology. Edited by Kelley Hays-Gilpin and David S. Whitley, pp. 231–254. London: Routledge, 1998.

Halstead, Paul L. J. "Counting Sheep in Neolithic and Bronze Age Greece." In Pattern of the Past: Studies in Honour of David Clarke. Edited by Ian Hodder, Glynn L. Isaac, and Norman Hammond, pp. 307–339. Cambridge, U.K.: Cambridge University Press, 1981.

Harris, Marvin. "Lactophiles and Lactophobes: Milk Lovers and Milk Haters." In his Good to Eat: Riddles of Food and Culture. New York: Simon & Schuster, 1985.

McCormick, Finbar. "Early Faunal Evidence for Dairying." Oxford Journal of Archaeology 11, no. 2 (1992): 201–209.

Payne, Sebastian. "Kill-Off Patterns in Sheep and Goats: The Mandibles from Asvan Kale." Anatolian Studies 23 (1973): 281–303.

Ryder, Michael L. Sheep and Man. London: Duckworth, 1983.

Sherratt, Andrew G. "Plough and Pastorialism: Aspects of the Secondary Products Revolution." In Pattern of the Past: Studies in Honour of David Clarke. Edited by H. Hammond, I. Hodder, and G. Isaac, pp. 261–305. Cambridge, U.K.: Cambridge University Press, 1981. (Reprinted with new information in Economy and Society in Prehistoric Europe: Changing Perspectives. Edited by A. Sherratt, pp. 158–198. Princeton, N.J.: Princeton University Press, 1997.)

——. "The Secondary Exploitation of Animals in the Old World." World Archaeology 15 (1983): 90–104. (Reprinted with new information in Economy and Society in Prehistoric Europe: Changing Perspectives. Edited by A. Sherratt, pp. 199–228. Princeton, N.J.: Princeton University Press, 1997.)

——. "Mobile Resources: Settlement and Exchange in Early Agricultural Europe." In his Economy and Society in Prehistoric Europe: Changing Perspectives. Princeton, N.J.: Princeton University Press, 1997. First published in Ranking, Resource, and Exchange: Aspects of the Archaeology of Early European Society. Edited by A. Colin Renfrew and Stephen J. Shennan, pp. 13–26. Cambridge, U.K.: Cambridge University Press, 1982.

Smith, Patricia, and Liora R. Kolska Horwitz. "Radiographic Evidence for Changing Patterns of Animal Exploitation in the Southern Levant." Journal of Archaeological Science 11, no. 6 (1984): 467–475.

Nerissa Russell