䉷 2009 by The Wenner-Gren Foundation for Anthropological Research. All rights reserved. 0011-3204/2009/5005-0002$10.00. DOI: 10.1086/605110

Current Anthropology Volume 50, Number 5, October 2009

597

Rethinking the Origins of Agriculture

The Proof Is in the Pudding

Feasting and the Origins of Domestication

by Brian Hayden

Feasting has been proposed as the major context and impetus behind the intensification of production
leading to the domestication of plants and animals. This article examines the way feasting contributes
to fitness in traditional societies through the reduction of risks involving subsistence, reproduction,
and violent confrontations. As other authors have noted, the risk-reduction strategies used by simple
foragers differ significantly from risk-reduction strategies used by transegalitarian hunter-gatherers
and horticulturalists. These differences are examined in more detail and are related to the emergence
of feasting in transegalitarian societies. Surplus-based feasting is proposed as an entirely new element
in community dynamics, probably first developed during the Upper Paleolithic in Europe, but
becoming much more widespread in the world with the development of Mesolithic technology.
Because feasting entails survival and risk-reduction benefits, it creates inherently inflationary food-
production forces. These elements first appear among complex hunter-gatherers and logically lead
to the intensification of food production, ultimately resulting in the domestication of plants and
animals.

In my view, the feasting explanation of domestication falls
under the umbrella of human behavioral ecology (HBE).
However, feasting occupies a rather specialized niche within
the HBE field, not unlike “signaling theory,” as espoused by
Bliege Bird and Smith (2005). I like to refer to this niche as
“paleopolitical ecology” since it focuses on the use of surpluses
to reduce varying kinds of risks (or to promote self-interests)
via political means. Bruce Winterhalder and Douglas Kennett
(2009, in this issue), in this dialogue on domestication, as
well as others, have highlighted the importance of risk-
reduction strategies in adaptive and evolutionary terms. Social
scientists rarely think of feasting in terms of risk reduction,
yet this is an aspect I and my coworkers have tried to em-
phasize as a critical factor in understanding feasting in trans-
egalitarian societies (Adams 2004; Clarke 2001; Hayden 2001).
I would therefore like to describe how feasting operates as a
risk-reduction strategy in transegalitarian societies and how
this leads to powerful pressures to develop political econo-
mies, food production, and ultimately, domesticated plants
and animals. Such risk-reduction strategies are markedly dif-

Brian Hayden is Professor in the Archaeology Department of Simon
Fraser University (Burnaby, British Columbia V5A 1S6, Canada
[[e-mail]]). This paper was submitted 9 IX 08 and accepted
1 V 09.

ferent from risk-reduction strategies used by simple egalitarian
hunter-gatherer societies, so these latter will first be described
to establish a clear reference point for the changes that took
place in transegalitarian societies. It is also worth noting that
risk reduction, as used here, pertains not only to variability
in food resources but also to reduction in risks from attacks
and risks of not being able to reproduce.

The Feasting Model of Domestication

The feasting model of domestication (as developed in Hayden
1990, 2004) posits that a range of technological innovations
(which became widespread in the Mesolithic) made it possible
to produce surpluses on a relatively dependable basis in cer-
tain favorable environments, especially riparian, coastal, and
cereal-rich habitats. These innovations included new mass
fishing technologies (nets, weirs, fishhooks, leisters), mass
seed-gathering and processing technologies involving crush-
ing and boiling, similar mass nut-processing technologies, and
long-term storage technologies. The resulting surpluses, in
turn, underwrote aggrandizer strategies that transformed egal-
itarian bands into transegalitarian complex hunter-gatherers
replete with socioeconomic inequalities, hierarchies, and eco-
nomically based competition in which feasting played a key

598

Current Anthropology

Volume 50, Number 5, October 2009

role. Because feasting was based on surplus production, and
because success in feasting conferred survival and reproduc-
tive benefits (in terms of surviving food shortfalls and ob-
taining military allies and marriage partners), powerful pres-
sures were created to increase food production, especially of
the most desirable foods and luxury foods used to impress
guests. Since individuals were vying with each other to acquire
mates or allies using surplus production, there could never
be enough food produced. Lower-ranked individuals could
always be expected to try to produce more in order to improve
their chances of obtaining better allies or partners, no matter
what the absolute level of production was.

It is crucial to observe that such transegalitarian and feast-

ing systems only became widespread and common during the
Mesolithic in most areas of the world, although they appear
to have begun in the Upper Paleolithic at least in the peri-
glacial refugia of Europe. Thus, such systems were a new and
powerful development in human history, and, in fact, such
systems were without parallel elsewhere in the animal king-
dom or among simple foragers (Hayden 2001). How could
feasting systems transform the risk-reduction strategies of ear-
lier times? Let us examine the proposed risk-reduction strat-
egies of simple hunter-gatherer bands as documented by
Wiessner (1982), myself (Hayden 1982, 1987, 1993), and
others.

Food-Risk-Reduction Strategies among
Simple Hunter-Gatherers

Wiessner (1982, 172–173) identified four major types of risk-
reduction strategies concerning food acquisition in a wide
range of traditional societies from bands to chiefdoms. These
can be augmented and grouped into the following cate-
gories:

1. Prevention of resource loss via manipulation (e.g., burn-

ing, transplanting, regulated harvesting) and/or exclusive
territoriality.

2. Risk pooling, variously based on (a) sharing of food

within the band (obligatory reciprocal pooling), (b) reciprocal
feasting networks involving selected segments within and be-
tween communities, and (c) centralized redistribution via cen-
tral administrators (as proposed for chiefdoms).

3. Mobility and the establishment of regional reciprocal

alliances used to access food in times of need. These could
be based on (a) regional kinship networks, including the Aus-
tralian section system; (b) rituals, including common totemic
ancestors; (c) frequent visiting and reciprocal hospitality; (d)
special relationships between people with similar names; (e)
special meat-sharing relationships; and (f ) exchange of hxaro
beadwork or other exotic or prestige items.

4. The creation of enemies who could be displaced if the

need arose (expropriation).

5. Storage (limited under high-mobility conditions).
6. Simple food exchange (transfer of risk).

7. Development of risk-reduction technologies (e.g.,

microlithic armatures).

8. Subsistence diversification.
Simple hunting-gathering bands overwhelmingly relied on

sharing, territoriality, regional alliances, mobility, and the cre-
ation of enemies to cope with high-risk situations. Very slow
rates of change during the Lower and Middle Paleolithic seem
to indicate that this combination of strategies was highly ef-
fective and created a stable adaptation. Due to the overriding
importance of obligatory sharing of resources in most band
societies, Meillassoux (1973) has argued that agriculture was
incompatible with simple forager risk-reduction adaptations.
He maintained that private or semiprivate ownership and
storage had to be established before agriculture could be ex-
pected to develop.

As Wiessner (1982, 173) suggested, it is likely that with the

emergence of complex hunting and gathering societies (where
individual ownership of produce and resource areas was the
norm), a number of new risk-reduction strategies developed
while some previously used strategies underwent major trans-
formations. The new strategies emphasized by complex
hunter-gatherers included storage, manipulation of plants and
animals (encompassing transplanting and cultivation, clam
gardens, stocking of fish, and “fire-stick farming” [Jones 1969;
Peacock 2002; Williams 2006]), exchanges of wealth for food,
improved subsistence procurement and extraction technol-
ogies, feasting, and possibly centralized redistribution. In
place of general reciprocal sharing, strategies such as the long-
term storage of privately owned produce by individual fam-
ilies or kin groups provided the most reliable means to man-
age the variability in resource availability from day to day and
from year to year (Testart 1982). Those who could produce
surpluses on a large enough scale could also acquire prestige
items that could be exchanged for food in times of need
(simple food exchange).

Some strategies used by simple hunter-gatherers continued

to be employed, albeit in a much-attenuated fashion. For
instance, obligatory sharing of food or wealth with all com-
munity members might no longer have been the norm, but
reciprocal sharing generally continued to be practiced among
close kin as well as during times of famine. Transegalitarian
ethnographic stories are replete with incidents of impover-
ished families begging for, and expecting to get, food from
wealthy households during famines. Families also generally
continued to maintain kinship ties in other communities so
that their place of residence could be moved to new locations
if family economic fortunes declined or if resources became
scarce in their foraging ranges. This strategy is characteristic
of most societies (e.g., Clarke 2001), including modern in-
dustrial ones, although the absolute amount of time, effort,
and resources spent on maintaining such kinship networks
in distant locations varies considerably. Elite members of
transegalitarian societies created unusually long-distance net-
works based on kinship, ritual ties, and exchange relation-
ships. Maintaining such networks entailed great expenditures

Hayden

Feasting and the Origins of Domestication

599

and considerable time. In contrast, normal community mem-
bers maintained, on average, much smaller and more local
networks. Thus, elites would have had greater flexibility to
cope with resource fluctuations.

In addition to kinship networks, ritual continued to be

used as a means of establishing or reinforcing close reciprocal
cooperative bonds between members of different commu-
nities. However, it seems likely that ritual ties with members
of the most distant communities only involved elite and
wealthy individuals. I suspect that such elite ritual organi-
zations took the form of regional (and local) secret societies
with costly membership fees and specialized facilities and par-
aphernalia (hence the regional archaeological ritual centers
such as Gobekli Tepe in Turkey and Kfar HaHoresh in the
Levant). Ritual links to distant communities were probably
not possible to establish for most nonelites due to the high
costs involved.

Many of the elite strategies (involving rituals, exchange of

prestige items, and maintenance of distant kinship ties) could
also have been used to forge alliances for defense and un-
doubtedly served as channels for acquiring spouses, thus re-
ducing risks in reproductive and defense terms. In addition
to persisting or transformed risk-reduction strategies, I sug-
gest that a strong division emerged in transegalitarian societies
for the first time between risk-reduction strategies used by
those with more surpluses and wealth versus strategies used
by those with minimal or no surpluses.

Feasting as a Risk-Reduction Strategy

With Mesolithic technology, the exploitation of relatively
inexhaustible types of rapidly reproducing, resilient resources
(r-selected species such as fish, grass seeds, and nuts) appears
to have greatly increased average production in certain eco-
logically favorable areas, but it did not completely eliminate
occasional large-scale fluctuations leading to famines (Hayden
1992, 531). Yet, most years, surpluses could be produced.
Given the possibility of producing surpluses in many, if not
most, years, how could extra food be used to reduce risks?
Storage is an obvious short-term strategy. Storage not only
can involve dry or cold storage of harvested and hunted foods
but it can also involve the transformation of extra harvest
into meat on the hoof. Falvey (1977) points out that Hill
Tribes in Thailand explicitly view the use of crop surpluses
in the raising of domestic animals in terms of “putting money
in the bank.” However, storage of whatever form has its own
risks, costs, and limitations. If families store far more than
will be needed in a certain cycle, the amount that remains
after consumption will eventually spoil and be lost. However,
it is always prudent to store more than is actually anticipated
to be needed because of uncertainties in the fluctuations in
resources, as well as risks of spoilage, insect or animal dep-
redations, or theft. Thus, it should be adaptive for most fam-
ilies to store somewhat more than needed on an annual basis

but not so much that the extra surplus can never be used.
This leaves many families with a regular overproduction of
surplus in normal years. Can this unneeded surplus be used
to further reduce risks? In order to do so, some means must
be found that is capable of transforming food surpluses into
other useful goods, services, debts, or relationships. I argue
that the transegalitarian feasting complex constitutes the pri-
mary mechanism for doing this. It is easy to understand how
surplus food can be used to hold feasts; however, what is not
so obvious is how feasts can reduce subsistence or other risks.

Our ethnoarchaeological work in Southeast Asia and Poly-

nesia has revealed that feasting in transegalitarian hill tribe
societies is used to create “social security nets” usually in-
volving lineage members, allied kin groups, and sometimes
other families concerned about risks and willing to produce/
invest surpluses toward those ends. Because of endemic war-
fare in the region prior to state-level rule, mobility and travel
were much more restricted than with simple egalitarian bands,
thus limiting the possibility of creating, maintaining, and us-
ing extensive intercommunity alliances to reduce risks. As
previously noted, private ownership of resources and produce
were incompatible with community-wide obligatory sharing
as a risk-reduction strategy. However, pooling risk through
restricted reciprocal-sharing networks involving selected pro-
ductive and reliable individuals would have been a very ef-
fective risk-reduction strategy. The result, I suggest, was the
emergence of adaptive feasting-based “social safety networks”
that pooled risks within certain segments of communities but
demanded continuous surplus contributions from members.
These networks became important in coping with individual
family fluctuations in subsistence fortunes but equally im-
portant in coping with the frequently cutthroat village politics
that characterized these communities.

One of the defining characteristics of transegalitarian

societies is the unstable and fluctuating subsistence base that
can favor some family fortunes at one time and other family
fortunes at other times. Without outside assistance, the oc-
currence of sickness, injury, or litigation involving principal
food providers could lead to starvation. Damage claims
against a household due to injuries or theft (or even destruc-
tion of crops or property by household animals) constituted
socioeconomic types of calamities. In many transegalitarian
societies, claims for recompense might even be trumped up
by ruthless individuals who seek to gain from families that
do not have adequate security networks (Condominas 1977).

Without a strong social safety net to back up families in

times of need, individuals may be unable to borrow food,
they may be unable to raise sufficient bride prices or dowries
to marry, they may have to pay exorbitant compensations,
their household land may be seized, and individuals may even
be sold into slavery or at least driven from the community.
As Condominas (1977) noted, litigation is endemic within
these communities. At the community level, lack of powerful
allies can invite attacks, force relocations, and result in mas-
sacres. These are the very considerable risks involved in not

600

Current Anthropology

Volume 50, Number 5, October 2009

creating mutual-support networks. Feasting is certainly the
main means of establishing such networks. Because feasting
is repeated at regular intervals and because it demands surplus
economic production and contributions, it is performance
grounded and capable of easily and effectively eliminating
freeloaders and cheaters. They are simply not invited to par-
ticipate in subsequent events and are thus excluded from
support networks. Undesirable noncontributors can be, and
are, eliminated from the network even if they are kin. Thus,
successful feasting dramatically reduces the risks surrounding
food production, procreation, and survival in unstable and
conflict-riddled environments but effectively ties this to being
productive in subsistence and committed to contributing
goods and labor to networking endeavors. Feasting establishes
the personal ties upon which transegalitarian social safety net-
works are predicated. Feasting in transegalitarian societies
takes a number of different forms: work feasts sponsored by
individuals for their own gain, aggressively competitive feasts
like the potlatch, solidarity feasts in which everyone contrib-
utes equally, and reciprocal alliance feasts that are essentially
contractual in terms of obligatory return feasts and return
support (debts).

It is, above all else, this last form, the alliance feast, that is

instrumental in creating social safety networks and is most
directly implicated in risk-reduction strategies; other feasting
forms may also serve this function but in more indirect fash-
ions. Alliance feasts establish reciprocal obligations and debts;
they create mutual networks of support. Anyone within the
network who hosts a feast or who confronts an adverse sit-
uation can generally rely on labor and material support from
other members. In order to enter into such a network and
in order to remain an active member, a family must contribute
to the major feasts sponsored by the network heads or key
members. Leaders of these networks (usually lineage or clan
heads) often control considerable labor and goods, enabling
them to develop more privileged “elite” risk-reduction strat-
egies for themselves, and I suspect that it is ultimately their
progenitors who were responsible for the fundamental
changes in risk-reduction strategies in transegalitarian
societies.

The strength of a network is generally gauged by the size

and sumptuousness of the major feasts sponsored by the net-
work, usually in the form of funeral feasts, marriages, or
house-building feasts. In this sense, such feasts serve as epi-
deictic displays to signal, for all to see, the relative power and
rank of the network group. This is close to the heart of sig-
naling theory as developed by Bliege Bird and Smith (2005).
However, alliance feasts operate at much higher levels, and
in far more powerful ways, than mere communication or
signaling devices. Due to their obligatory reciprocal material
and labor exchanges, feasts actually create debts, powerful
social units or factions, and political power.

Because the quality of life, reproduction, and survival often

depend on the strength of these networks in transegalitarian
societies, there are enormous pressures to periodically hold

feasts made as impressive as possible, to provide as much
food (especially luxury foods) as possible, and to serve such
foods in impressive containers to influential guests—heads of
feasting factions. Ethnographer after ethnographer has ob-
served that the real pressures to intensify and augment food
production in transegalitarian societies is not from subsistence
needs but from the sociopolitical demands for more pigs,
more cattle, more food to successfully compete in feasting
and marriage payments, whether in California, the Northwest
Coast, New Guinea, Southeast Asia, Polynesia, East Africa, or
elsewhere, whether among complex hunter-gatherers, simple
horticulturalists, or incipient chiefdoms (see Hayden 2001,
2007). The surplus-based feasting strategies of risk reduction
are largely responsible for creating Earle’s (1997) “political”
economy. This is a new dimension of food demand and pro-
duction that simply is lacking among egalitarian foragers. It
first appears among the complex hunter-gatherers that
emerged at the end of the Pleistocene, prior to the domesti-
cation of plants and animals. Contrary to the misrepresen-
tations of Smith (2001) and Winterhalder and Kennett (2006),
no “marked status inequalities” or “high” degrees of seden-
tism, or any “two-tiered settlement hierarchies” need exist for
these transegalitarian feasting systems to operate among com-
plex hunter-gatherers. Modest archaeological indicators of
these developments (without settlement hierarchies, exces-
sively rich burials, or lavish durable prestige goods) charac-
terize many ethnographic transegalitarian groups (e.g., those
in New Guinea) that engage in feasting and other surplus-
based strategies geared to reduce risks and enhance the sur-
vival interests of certain factions within communities.

The important point is that transegalitarian feasting based

on the production of surpluses is a new, powerful force that
relentlessly pushes food production to its maximum limit
under favorable technological and environmental conditions.
Feasting and the other characteristics of complex hunter-
gatherers (increased sedentism, storage, socioeconomic in-
equalities, prestige goods) are well documented in the Na-
tufian culture of the Near East (Hayden 2004) and in the
Jomon culture of Japan, while good cases can be made for
similar developments in China (Hayden 2007), and north-
eastern North America. I argue that domestication is not
simply the ultimate end product of incremental population
growth over the span of 2 million years, or of gradual im-
provements in human intelligence, or other similar cumu-
lative changes. Rather, domestication was one element of a
constellation of entirely new traits that emerged rather
abruptly with the appearance of complex hunter-gatherers
involving economically based competition and the conversion
of economic surpluses into other desirable goods, debts, and
services. Underlying all these developments were the new
strategies adopted by transegalitarian societies to reduce risks
involving subsistence, reproduction, and conflicts. The strat-
egies with the most far-reaching consequences were storage
and feasting, and under favorable conditions, these strategies
repeatedly resulted in domestication in many places in the

Hayden

Feasting and the Origins of Domestication

601

world in order to meet the inherently escalating demands for
more staples and more prestige foods (and ordinary foods)
used in feasts.

References Cited

Adams, Ron. 2004. An ethnoarchaeological study of feasting

in Sulawesi, Indonesia. Journal of Anthropological Ar-
chaeology 23:56–78.

Bliege Bird, Rebecca, and Eric Alden Smith. 2005. Signaling

theory, strategic interaction, and symbolic capital. Current
Anthropology 46:221–248.

Clarke, Michael. 2001. Akha feasting. In Feasts: archaeological

and ethnographic perspectives on food, politics, and power.
M. Dietler and B. Hayden, eds. Pp. 144–167. Washington,
DC: Smithsonian Institution.

Condominas, Georges. 1977. We have eaten the forest. New

York: Hill & Wang.

Earle, Timothy. 1997. How chiefs come to power: the political

economy in prehistory. Stanford, CA: Stanford University
Press.

Falvey, L. 1977. Ruminants in the highlands of northern Thai-

land. Thai-Australian Highland Agronomy Project. Chiang
Mai: Tribal Research Institute, Chiang Mai University.

Hayden, Brian. 1982. Interaction parameters and the demise

of Paleo-Indian craftsmanship. Plains Anthropologist 27:
109–123.

———. 1987. Alliances and ritual ecstasy: human responses

to resource stress. Journal for the Scientific Study of Re-
ligion 26:81–91.

———. 1990. Nimrods, piscators, pluckers, and planters: the

emergence of food production. Journal of Anthropological
Archaeology 9:31–69.

———. 1992. Ecology and complex hunter/gatherers. In A

complex culture of the British Columbia Plateau: tradi-
tional Stl’atl’imx resource use. Brian Hayden, ed. Pp.
525–564. Vancouver: University of British Columbia Press.

———. 1993. Archaeology: the science of once and future

things. New York: W. H. Freeman.

———. 2001. Fabulous feasts: a prolegomenon to the im-

portance of feasting. In Feasts: archaeological and ethno-
graphic perspectives on food, politics, and power. M. Diet-
ler and B. Hayden, eds. Pp. 23–64. Washington, DC:
Smithsonian Institution.

———. 2004. Sociopolitical organization in the Natufian: a

view from the Northwest. In The last hunter-gatherer so-
cieties in the Near East. C. Delage, ed. Pp. 263–308. British
Archaeological Reports International Series. Oxford: John
and Erica Hedges.

———. 2007. The origins of domesticated rice. Paper pre-

sented at the 2007 EuroSEAS meetings. Naples, Italy.

Jones, Rhys. 1969. Fire-stick farming. Australian Natural His-

tory 16:224–228.

Meillassoux, Claude. 1973. On the mode of production of the

hunting band. Oxford: Oxford University Press.

Peacock, Sandra. 2002. Perusing the pits: evidence for pre-

historic geophyte processing on the Canadian Plateau. In
Hunter-gatherer archaeobotany. S. Mason and J. Hather,
eds. Pp. 45–63. London: Institute of Archaeology.

Smith, Bruce. 2001. The transition to food production. In

Archaeology at the Millennium. G. Feinman and T. Price,
eds. Pp. 199–230. New York: Kluwer Academic/Plenum.

Testart, Alain. 1982. The significance of food storage among

hunter-gatherers. Current Anthropology 23:523–537.

Wiessner, Polly. 1982. Beyond willow smoke and dogs’ tails.

American Antiquity 47:171–178.

Williams, Judith. 2006. Clam gardens. Vancouver, British Co-

lumbia: New Star.

Winterhalder, Bruce, and Douglas Kennett. 2006. Behavioral

ecology and the transition from hunting and gathering to
agriculture. In Behavioral ecology and the transition to ag-
riculture. D. Kennett and B. Winterhalder, eds. Pp. 1–21.
Berkeley: University of California Press.

———. 2009. Four neglected concepts with a role to play in

explaining the origins of agriculture. Current Anthropology
50:645–648.