Archaeological Parameters for the Beginnings of Beer
By Thomas W. Kavanagh, Ph.D.
Republished from BrewingTechniques' September/October 1994.

Although of great interest to the brewing community, the question of the origin of beer and brewing has received sporadic attention from academic investigators. Home brewer, anthropologist, and museum curator Thomas Kavanagh outlines and evaluates modern archeological research on beer's origins and proposes specific areas for further study.

In the absence of clear archaeological evidence, it is difficult to state with any certainty when or where brewing began, let alone why or how. This article summarizes the two primary articles in the anthropological and related literature on the possible origins of beer, discusses some of the archaeological and zymurlogical parameters within which beer may have originated, and suggests some areas for further investigation.


After World War II, much archaeological attention was focused on the transition from food collecting to food production (4000-6000 B.C.). In a short article in the October 1952 Scientific American, Robert Braidwood of the University of Chicago reported on his excavations in the Zagros mountains of eastern Iraq, noting that by the time the village of Qalat Jarmo was occupied, ca 4750 B.C., several kinds of grains and legumes were being grown and several animals had been domesticated (1). Although at that time Braidwood did not explicitly state a possible grain food product, he later noted that it was "generally assumed that the appearance of domesticated cereals in the Near East was intimately linked with the use of these grains for the preparation of flour for bread making" (2).

In a personal letter to Braidwood, Jonathan D. Sauer, professor of botany at the University of Wisconsin, questioned that assumption, asking Braidwood "whether the earliest utilization of domesticated cereals may have been for beer rather than bread." Braidwood took the opportunity to make use of the new symposium-by-mail section of the American Anthropologist to pose this question to his anthropological colleagues:

Could the discovery that a mash of fermented grain yielded a palatable and nutritious beverage have acted as a greater stimulant toward the experimental selection and breeding of the cereals than the discovery of flour and bread-making? One would assume that the utilization of wild cereals (along with edible roots and berries) as a source of collected food would have been in existence for millennia before their domestication (in a meaningful sense) took place. Was the subsequent impetus to this domestication bread or beer?

In his introduction to the published symposium, Braidwood briefly discussed the technology associated with early Neolithic grain production. Microlithic sickles - small blades of obsidian hafted into a single handle - make their earliest archaeological appearance at the so-called Natufian horizon of what is now Israel, ca. 10,000-8000 B.C. These sickles imply reaping but not necessarily planting. Mortars and pestles came later, but not necessarily solely for grinding grain. Then came "celts" - rubbed stone artifacts - that could have been blades for either axes or hoes. At that time, the earliest evidence for grain - two kinds of wheat, one kind of barley, all "in process" of domestication - was at Jarmo in Iraq and at Jericho in Israel, both ca. 6700 B.C. (see map, Figure 1). Querns and other milling stones, clay ovens, and basketry were associated with these grains. Pottery came last. The timelines in Figure 2 show the emergence of domesticated cultures of four subregions of the ancient Middle East.

Although none of these technological artifacts necessarily implied the existence of beer or bread, Braidwood noted that when recovered archaeologically, cereal grains were almost invariably charred. Thus he asked "whether these charred kernels may have been overfired accidentally in the parching process by which the malt was prepared, although I have seen grain parched for other purposes." Braidwood closed his introduction with the comment, "I am in no way unattracted to the idea, as Sauer put it, that "thirst rather than hunger may have been the stimulus behind the origin of small grain agriculture.'"

Seven scholars replied; unfortunately, although the symposium included several notable anthropologists and paleobotanists of the time, apparently few had any working knowledge of the beer-making process from which to evaluate the Braidwood-Sauer hypothesis.

Three replies, however, are significant. The first comment was from Sauer, who noted that given the preplow technology and the brittleness of the rachis (also called the axis) of wild cereal grains - the attachment between seed and stem which shatters when mature, thus scattering the seed - the collection of wild grain "would seem to me a game scarcely worth the candle except for a more rewarding stake than mere food."

Hans Helbaek, curator of the Danish National Museum and authority on ancient cereal grains, rejected the suggestion that charred grain reflects a brewing/malting operation. He noted that in malting, only enough heat is applied to kill the germ, certainly not enough to carbonize the grain. Furthermore, at Jarmo the carbonized grain was ungerminated. It was not malt. Finally, he noted, "If brewing was the primary utilization of cereals from the beginning, it should be expected that the diffusion of agriculture first and foremost was manifested by a spreading knowledge of brewing." The fact that this was not the case supported the opposite hypothesis.

Paul Mangelsdorf of Harvard gave the longest reply. He said that the idea that beer antedated bread was not particularly new.* His own contribution to the discussion was to note that in the earliest grains - the various strains of wild wheats and barleys - the glume (husks and chaff) adhere to the grain. From this he argued that, without additional processing, the early grains were suited more to beer than to bread. However, he went on to pose several other problems, particularly that beer would not serve as the major source of carbohydrates; thus "man cannot live on beer alone and not too satisfactorily on beer and meat."

Mangelsdorf then noted that too often when Westerners think of bread, they think of yeast-based leavened loaves. He argued that unleavened flat breads would have been technological precursors to both leavened bread and beer and noted that in at least one method of brewing, the partial baking of yeast bread comes before steeping the loaves and fermenting the solution into beer.

To conclude the discussion, Braidwood summarized the various comments. After several in-house archaeological explanations, he focused on two points: first, "if the earliest Near Eastern beer was brewed from germinated grain malt as Mangelsdorf thinks probable," then the ungerminated grain from Jarmo offered "no evidence of the process"; that is, as the Scots say, the hypothesis was "not proven." Second, in reference to the known tendency of ripe wild barley to shatter, Braidwood commented that he knew of at least one village in Syria that made its bulghur - "wheat grains which are first boiled, then sun dried, cracked into grits, and finally served as a hot mush-like dish" - from "not quite ripe" grain.

The "Beer or Bread" symposium was inconclusive; at best, it pointed out that stating the question "whether beer or bread was the prime mover in domestication" was far too simplistic to be answered. Indeed, it is probably better to say that gruel, unleavened bread, and beer were important product points in the process of the domestication of grain, but neither beer nor leavened bread was the impetus. Although archaeologists of the classical periods of ancient civilizations continue to explore the role of beer and brewing in those periods (4-7), the question of beer's relationship to the domestication of grain was all but ignored for the next 35 years, adding to the inconclusiveness of the symposium.

Since that time, however, continued archaeological work has clarified some of the available dates. So far, the earliest traces of Hordeum spontaneum (wild barley) date to 17,000 B.C. at the epi-Paleolithic site of Ohalo II on the south shore of the Sea of Galilee. The earliest domesticated two-row barley, H. distichum, is from the site of Netiv Hagdud, north of Jericho, ca. 7750 B.C. (it was mixed with H. spontaneum). Six-row barley, H. hexastichum, appears at 6000 B.C. at the famous site of Ali Kosh in Iraq (8). The earliest appearance of clay vessels - whether fired or not is unclear - seems to have been ca. 7500 B.C. in the Zagros mountains of Iraq and ca. 6000-5500 B.C. in Syria (9). This early earthenware pottery was relatively porous and was better suited for the storage of dry rather than wet materials. The introduction of burnishing, and of slips and glazes, all of which improved the ability to store liquids, did not come until much later.

The archaeological appearance of the products of civilization - domesticated grains, legumes, animals, and pottery - varies considerably from region to region and from site to site within the greater Near East. Therefore, the "earliest appearance" of a trait does not imply a distribution beyond the point of discovery (note the variances in Figure 1, for example).


In 1986, Solomon Katz and Mary Voigt of the University of Pennsylvania raised the question of the relationship between beer-making and grain domestication when they suggested that "among the factors that led to the domestication of wild cereals [was] . . . the "high' that people obtained from beer" (10). Further, and in contrast to Paul Mangelsdorf's comment in the original symposium, they argued that "individuals who consumed beer were better nourished than those who consumed wheat and barley as gruel or who ignored these wild resources." Finally, as the driving force leading to the domestication of grains, they suggested that "if the discovery of fermentation was made by collectors of wild cereals . . . and if the use of beer was incorporated into the social and/or religious system of these people, then the disruption in the supply of these wild foods would have posed a serious problem. . . ." People with such a problem would try to solve it, thereby leading to the domestication of grain.

Katz and Voigt suggested a scenario that followed a series of accidental steps from the use of cereals in a gruel or porridge through the "accidental sprouting" [malting] of the grain and the "discovery of the sweet taste of the sprout" [mashing]. Then, came the discovery that "if this special gruel was heated and then allowed to stand overnight or longer, wild yeasts would have started the process of fermentation."


To my mind, both the Braidwood-Sauer initial hypothesis and Katz and Voigt's more recent discussion overlooked the primary problem. In both cases, the hypothesis was that the discovery of fermentation was the impetus to domestication. This hypothesis assumes that fermentation was possible with pre- or proto-Neolithic technology. A major step toward the confirmation of the hypothesis must therefore come through the demonstration that a controlled and reproducible fermentation was possible given the technology available at the dawn of the Neolithic period.

Although brewing is a complex process, its technology need not be. I believe, however, that the absence of particular elements of connective technology at two points in the process might have prevented the whole trajectory from developing during pre-Neolithic times. The first point is simply, was it possible to produce an efficient malt given the conditions at the dawn of the Neolithic period? Second, did a container exist in which fermentation could take place?

The grain: In the Braidwood symposium, Sauer argued that the collection of wild grain "would seem to me a game scarcely worth the candle except for a more rewarding stake than mere food." Since then, based on analogy with modern hunter-gatherers and on ethnohistorical reconstructions, it has been shown that reliance on wild grain crops with only pre-Neolithic technology was not particularly arduous, and it was rewarding. Paleoethnobotanist Jack Harlan cited examples, including his own experiment that suggests that in harvests of wild grasses, "yields of 500-800 kg/hectare were not uncommon and 1 ton/hectare could easily be obtained" (11). Moreover, the time required for such harvests was minimal; one person could harvest ~1-2 kg/h, and "one adult could easily gather 10 kg in a morning's effort" (12).

Scheduling, however, would have been one of the major problems with such an economy; as Sauer noted, the maturation process is progressive on each wild plant, with the grains on the head maturing at different times. In addition, the basic problem with harvesting wild grasses is that as they near maturity, the point of attachment between seed and stalk (the rachis or axis) becomes brittle; this feature enables the plant to distribute its seed at the slightest jiggle.

Given these facts, the most efficient way to harvest a wild grain crop would be to make daily passes through the field, knocking the grain heads into a basket or other container with a stick or flail-like implement (11). In such a harvest scheme, the fully mature grains would be dispersed during the time between passes, the almost mature grains would be collected, and the clearly immature grains would remain attached. Each patch would be visited several times during the maturation period. In this context, Braidwood's comment about his unnamed Syrian village acquires significance: they made their bulghur from "not quite ripe" grain.

But this bio-ecological system is also a primary branching point in the bread/beer hypothesis. For the larger history of cereal-grain cultivation, the question is whether "not quite ripe" grain was viable. Would it sprout? Could it be used for seed? For the history of brewing, the point is whether one could sprout "not quite ripe" grain and convert its starches to sugars. That is, could one malt and mash it? Although Helbeck noted in 1953 that none of the grains recovered from Jarmo was malt, that does not mean that malt could not have been produced. If, however, because of the way the grain was harvested malt could not have been produced, it is fair to conclude that one could make flat bread and gruel but not beer.

Therefore, one point of investigation should focus on the degree of ripeness of wild seed in proportion to the brittleness of the rachis: Does wild seed become fertile before the rachis embrittles, or do the two progress together? This question would also have a bearing on the question of the efficiency with which a mash could convert starches to sugars. Furthermore, given that scheduled and repetitive harvests of a progressively maturing grain patch would result in the harvesting of grain at slightly different stages of maturity, what minimum percentage of mature/immature (maltable to unmaltable) grain in a gruel would enable starch-to-sugar conversion to occur?

But even if some noticeable and desirable result could be produced by malting grains using pre-Neolithic technology as hypothesized by Katz and Voigt, a major technological invention of the Neolithic might have brought these pre-Neolithic gains to naught and, indeed, might have produced exactly the evolutionary stress they suggest as leading toward the desire to domesticate. With a Mesolithic technology, semiripe grains might have been harvested by a flail-like implement, but Braidwood noted in 1953 that one of the earliest features of Natufian technology (proto-Neolithic) was the introduction of the microlithic sickle. With this implement, instead of harvesting just the almost-mature grains, the whole seed head was harvested all at once; some grains would have been close to maturity, most would have been immature, and some very immature.

Again, the problem for both the larger history of grain cultivation and for brewing is clear: Without some way of determining which grains were viable and which were not, a random sample taken from such a harvest would have a greater than random proportion of immature to mature grains. The questions become: Could such a sample be malted? Could it be mashed? Were enough enzymes present to convert the starches to sugars?

It would have been in this context that the primary genetic mutation involved in cereal-grain domestication may have begun through natural selection - the first farmers would have selected seed for their tough rachises, those that would keep the seed on the head well into maturity.* The importance of such mutant grain would have been its predictable maltability, of particular value to any Mesolithic/early Neolithic maltster.

At the same time, it must be noted that, according to archaeologist Charles Redman, the growth in the use of such a product required a long time: "Although the earliest evidence for domesticated barley is at about 7000 B.C., . . . of the axes recovered from . . . [near] the end of the seventh millennium . . . only from 10-15% were tough" (13). It took the 1000 years of the seventh millennium B.C. to result in a 15% increase in the use of domesticated - that is, readily maltable - grains. Stated another way, even if we hypothesize that at the close of the seventh millennium all of the domesticated grain had been reserved for malting, we are still only speaking of 15% of the total grain product.

The container: I believe a more serious problem with the Braidwood-Sauer and Katz-Voigt hypotheses surrounds the early Neolithic fermentor. Although a pre-Neolithic gatherer could make a grain gruel, and even a mash - both of which are heated suspensions of crushed grains in liquids - could she make it ferment without the use of pottery?
     For Katz and Voigt:

The container the Natufians would have used would have been a skin bag or perhaps a basket or a wooden vessel. . . . Fermentation . . . will produce alcohol (rather than some other metabolic product) if . . . the fermentation medium [is] anaerobic. . . . Anaerobic conditions are relatively easy to obtain. . . . If the beer is in a container shielded from the air, carbon dioxide will itself shut off enough oxygen to ensure alcohol production. . . . In the Middle East, where summer temperatures can reach 120 °F or more, there would have been little need to heat the brew. Even in winter, putting the mixture in a sunny spot would provide adequate heat for the yeast to work during the day and at night it might have been placed near a fire.
Katz and Voigt seem to have, first, confused the temperature requirements of the mashing and fermentation processes and then overlooked the archaeological evidence for containers, their differential characteristics, and thus their suitability for beer-making. At the first point in the mashing process, the problem is not "adequate heat for the yeast," but adequate heat for the enzymes. Further, although it may be possible for a liquid to reach equilibrium with an ambient temperature of 120 °F, such temperatures would be inadequate for saccharification; rather, a temperature of ~150 °F would be needed. Unfortunately, direct heating of a basket, skin, or wood container is not possible.

One possible method for adding heat could have been "stone boiling" - dropping hot rocks into a liquid-filled container. This method would work for animal paunches suspended from tripods and for tightly woven baskets whose elements swell up with the moisture; in addition, many prehistoric Near Eastern baskets were covered with pitch to further enhance their watertight characteristics. Although it is difficult to simmer for long periods using stone boiling - sooner or later the volume of the added hot rocks will exceed the volume of the container - it should be possible to achieve and maintain saccharification temperatures for 20 min or so, and thus achieve a minimal degree of starch-sugar conversion of malted grain.

The next step is fermentation itself. Although Katz and Voigt state that a "container shielded from the air" was necessary for yeast-based fermentation, presumably the only containers that were available during the proto-Neolithic period were the skin, basket, or wood containers they mention. And although open fermentation is indeed possible - one might note the excellent product still produced by Samuel Smith in Yorkshire - open fermentation in a nonsterile, organic, and permeable container would at first glance seem as often to lead to an interesting science project as to beer.

On the other hand, it should be noted that the very permeability of such a container might have been of value. Although Katz and Voigt suggest that in 120 °F temperatures "there would have been little need to heat the brew," the problem at this stage of the process is not heating the brew - indeed a temperature of 120 °F would kill the yeast - but cooling it. Liquid-filled baskets and paunches sweat, perspire, and otherwise lose heat through evaporation - as do the earliest earthenwares. Whether such evaporative cooling would allow the liquid in a basket to maintain adequate fermentation temperatures is proposed as an experiment to brewers in Tucson and Baghdad.


We should remember that the Braidwood-Sauer and Katz-Voigt hypothesis was not "which came first," beer or bread, but was whether cereal-grain domestication was propelled by the ability and desire to make a fermentable mash at the dawn of the Neolithic period. In light of the currently available archaeological evidence, interpreted with a knowledge of the biochemical processes of brewing, I suggest that although it may have long been possible to malt grain and indeed to make a fermentable mash, it was probably difficult, if not impossible, to make a reproducible beer, or any other fermented beverage, before the invention of pottery. Therefore, it may not have been until well after the invention of pottery that the production of beer would have significantly influenced the use of domesticated cereal grains.


(1) R.J. Braidwood, "From Cave to Village," Scientific American, October 1952, 62-66.

(2) "Symposium: Did Man Once Live by Bread Alone," American Anthropologist 55 (4), 15-526 (1953).

(3) J.M. Renfrew, Paleobotany: The Prehistoric Food Plants of the Near East and Europe (Columbia University Press, New York, 1973).

(4) A.L. Oppenheim, On Beer and Brewing Techniques in Ancient Mesopotamia, supplement to the Journal of the American Oriental Society, 1950.

(5) H.F. Lutz, Viticulture and Brewing in the Ancient Orient (J. C. Hinrich, G.E. Stechert, Leipzig, Germany, 1922).

(6) A. Lucas, Ancient Egyptian Materials and Industries (E. Arnold, London, England, 1948).

(7) E. Huber, Bier und Bierbereitung bei den Volkern der Urzeir I, Babylonien und Aegypten (Gesellschaft fur die Geschichte und Bibliographie des Brauwesens, Berlin, Germany, 1926).

(8) D. Zohary and M. Hopf, The Domestication of Plants in the Old World (Clarendon Press, Oxford, 1993).

(9) P.M. Rice, Pottery Analysis: A Source Book (University of Chicago Press, 1987).

(10) S.H. Katz and M.M. Voigt, "Bread and Beer: The Early Use of Cereals in the Human Diet," Expeditions 28 (2), 23-34 (1986).

(11) J.R. Harlan, "A Wild Wheat Harvest in Turkey," Archaeology 10, 197-201 (1967).

(12) J.R. Harlan, "Wild-Grass Seed Harvesting in the Sahara and Sub-Sahara of Africa," in D.R. Harris and G.C. Hillman, Foraging and Farming: The Evolution of Plant Exploitation (Unwin Hyman, London, 1993).

(13) Charles L. Redman, The Rise of Civilization: From Early Farmers to Urban Society in the Ancient Near East (W.H. Freeman, San Francisco, 1978).

(14) W.J. Darby, P. Ghalioungui, and L. Grivetti, Food, the Gift of Osiris (Academic Press, Cambridge, Massachusetts, 1977).

(15) L.F. Hartman and A.L. Oppenheim, On Beer and Brewing Techniques in Ancient Mesopotamia, supplement to the Journal of the American Oriental Society, 1950.

Issue 2.5 Table Of Contents
[Home]  [BrewingTechniques Library]  [Contact Us]  [Order]