ROBERT Mee. NETTING
Smallholders, Householders
Farm Families and the Ecology of
Intensive, Sustainable Agriculture
STANFORD UNIVERSITY PRESS
STANFORD, CALIFORNIA
1993
Stanford University Press
Stanford, California
© 1993 by the Board of Trustees of the
Leland Stanford Junior University
Printed in the United States of America
CJP data appear at the end of the book
Library of Congress Catalogmg-in-Pubhcation Data
Nerting, Robert McC.
Smallholders, householders: farm fam1hes and the ecology of intensive, sustainable
agriculture I Robert McC. Netting.
p. cm.
Includes b1bhograph1cal references and index.
ISBN 0-8047-2001-4 (alk. paper):
ISBN 0-8047-2102-5 (pbk.: alk. paper):
1. Traditional farmmg. 2. Agricultural ecology. 3. Agriculture-Economic aspects.
4. Sustainable agriculture. 5. Family farms.
I. Title.
CN407.4.N48 1993
630-dc20
92-20376
CIP
@This book 1s prmted on acid-free paper.
Prologue: An Ethnological Essay
in Practical Reason
ANTHROPOLOGICAL THEORIES of"practical reason" that are natural or
ecological, utilitarian or economic like Steward's have in recent years
been opposed by scholars who emphasize "culture" as symbolic or meaningful. In the words of Marshall Sahlins:
The distinctive quality of man [is) not that he must live in a material world, [a)
circumstance he shares with all organisms, but that he does so according to a
meaningful scheme of his own devising, in which capacity mankind is unique.
[This approach J therefore takes as the decisive quality of culture ... not that this
culture must conform to material constraints but that 1t does so according to a
definite symbolic scheme which is never the only one possible. Hence it is culture
which constitutes utility. (Sahlins 1976: viii)
This book on the practice of smallholder intensive agriculture by farmfamily households relates elements of environment, technology, and human social organization in the tradition of cultural ecology pioneered by
Steward (1938, 1955). It is not a study of "culture" in the widely accepted
anthropological sense of a distinctive system of shared meanings and a
symbolic organization of experience characterizing a particular society or
social group. The focus on differences in ways of thinking expressed in
language, beliefs, rituals, and myths, and interpreted from a wide range
of cultural texts, has been and necessarily remains a central concern of the
discipline. But anthropology has also been an empirical social science of
practical reason, grounded in an Enlightenment faith that there are regularities in human behavior and institutions that can be understood as filling human biological and psychological needs under particular circumstances of geography, demography, technology, and history. These com monalities can be discerned cross-culturally in groups separated by space
and time and displaying a splendid variety of cultural values, religions,
kinship systems, and political structures.
The systematic comparison of practices and institutions that reoccur in
different societies, and the analysis of how they function and change in a
2
Prologue
functional and comprehensible manner is ethnology, a somewhat quaint
and old-fashioned designation chat many contemporary anthropologists
would never use to characterize their own work. What follows is not an
attempt to interpret "culture," a project of eliciting and perhaps creating
meaning so grand that only the artist or the literary critic would confidently attempt it. Rather it examines a limited sec of social and economic
factors that are regularly associated with a definable type of productive
activity, despite considerable variation in a number of other "cultural"
features that may themselves cohere internally in meaningful, patterned
ways.
Smallholders: Characterizing a Type
Smallholders are rural cultivators practicing intensive, permanent, diversified agriculture on relatively small farms in areas of dense population. The family household is the major corporate social unit for mobilizing agricultural labor, managing productive resources, and organizing
consumption. The household produces a significant part of its own subsistence, and it generally participates in the market, where it sells some
agricultural goods as well as carrying on cottage industry or other offfarm employment. Choices of allocating time and effort, tools, land, and
capital to specific uses, in a context of changing climate, resource availability, and markets must be made daily, and these economic decisions are
intelligible in rational, utilitarian terms. Smallholders have ownership or
ocher well-defined tenure rights in land that are long-term and often heritable. They are also members of communities with common property
and accompanying institutions for sharing, monitoring, and protecting
such resources. The existence of separate household enterprises, with a
measure of autonomy and self-determination, in a larger economy with
institutionalized property rights and market exchange, presents the likelihood of economic inequality, both among households in the community at any point in time and in the changing status of a single household
at different times in its developmental cycle. But inequality is not equivalent to enduring class stratification within the farming community, and
neither does it exclude socioeconomic mobility. The argument of chis
book is chat these characteristics regularly co-occur, and that their systematic articulation and changing relationships can be reliably observed,
described, and explained .
Not all food producers are smallholders. The characteristics put forward here do not apply to shifting cultivators practicing long-fallow or
slash-and-burn farming where land is still plentiful and population density low, as in some pares of the humid tropics today. Nor does the des-
Prologue
3
ignacion smallholder fit herders, whether they be the nomadic pastoralists
of East Africa or the ranchers of Texas. It does not match geographically,
economically, or socially with the farming systems of dry monocropping
of wheat, sugarcane estates, cotton plantations with slaves, or California
agribusinesses. Smallholders practice intensive agriculture, producing relatively high annual or multicrop yields from permanent fields that are
seldom or never rested, with fertility restored and sustamed by practices
such as thorough tillage, crop diversification and rotation, animal husbandry, fertilization, irrigation, drainage, and terracing. I am not talking
here about amber waves of grain but about gardens and orchards, about
rice paddies, dairy farms, and cl1inampas.
Even the casual observer has little difficulty in recognizing a landscape
domesticated by intensive agriculture. The stepped stone walls and mirrored, ponded fields of Balinese wet-rice cultivation and the neat, fenced,
manicured pastures of Dutch farmsteads bespeak high, dependable yields
and diligent stewardship. But that these are, in fact, representatives of a
distinctive type of land use regularly associated with specific demographic, social, and institutional factors may require something more
than a leap of faith in practical reason. It is the virtue ofJulian Steward's
approach that consistent cross-cultural relationships can be demonstrated
empirically despite striking variability in local environment, technology,
culture, and politics. The common features form a definable cultural ecosystem with its own evolutionary patterns and probabilities of change.
The smallholder as depicted here may be what Max Weber (1949: 90)
called an ideal type-chat is, a "conceptual pattern [that] brings together
certain relationships and events of historical life into a complex, which is
conceived as an internally consistent system." As in Geertz's (1963) characterization of agricultural involution or Popkin's (1979) of the rational
peasant, "the researcher posits a structured representation of a social category that singles out certain feat ures and abstracts from others" (Little
1989: 194). The smallholding householders that I examine in this book
are alike in that for all of them land is objectively a scarce good, agrarian
production per unit area is relatively high and sustainable, fields are permanent, work takes skill and relatively long periods of time, decisions
muse be made frequently, and the farm family has some continuing rights
to the land and its fruits. In these type traits, the smallholder differs in
kind or in degree both from other food producers and from those who
pursue other occupations. Drawing principally on ethnography, agricultural economics, and geography, I first describe what smallholders do
and then attempt to account for the systematic commonalities of behavior
and institutio ns that make a kind of sense according co the plebeian, but
still powerful, canons of practical reason. There is no shared culture of
4
Prologue
meanings among the many disparate groups of smallholders, but the
quest for functionally meaningful and coherent systems that transcend
the distinctions of societies and regions is also part of the anthropological
calling.
Why Study Smallholders? Some Subjective Reflections
on Objective Research
It would be misleading and disingenuous to argue that scholarly work
that styles itself "social science" arises from a single-minded search for
timeless truths existing out there in the real world, or that those of us
who essay this approach to knowledge believe that we shall discover natural laws of society, test hypotheses in some irrefutable way, or reliably
predict future states of the system. Perhaps one of the attractions (and the
solaces) of anthropology is that its deductive models are neither very
compelling nor particularly intrusive, and that one is almost sure to learn
something interesting and new by fieldwork (even if this is no longer
always "exotica and trips"). The formal structuring of problems and hypotheses in the research proposal, and the (sometimes very different) relationships of data, argument, and theory in the finished product seldom
overtly reflect the subjective experience or the sentimental journey that
led the student in that direction. At the risk of ruminations and other
evidences that I may be entering my anecdotage, let me ask how it was
that I came to study agriculture, households, and land tenure, rather
than, say, kinship terminology, the peasant view of the good life, or caste
in India, as my esteemed teachers had. It might also be useful to try to
reconstruct why, at this point, I should leave those ethnographic cases
that I know at first hand for the much more hazardous terrain of global
ethnological comparison and synthesis.
As a graduate student at the University of Chicago in the late 1950s, I
read Julian Steward's Tlieory of Culture Change in Fred Eggan's course on
ethnological theory and method, and I did a source paper on Fulani ecology for a seminar at Northwestern University with Jim Bohannan. But
my proposal for my first field research, begun in 1960 on theJos Plateau
in Northern Nigeria, did not make such interests explicit. As I once admitted, " I did not have a carefully thought out plan of ecological study
when I entered the field, and my findings came piece-meal in response to
the elementary questions of why people lived where they do, what they
did with their time, and how they got enough to eat. Many of my conclusions came in the ana lysis of quantitative material after leaving the field"
(Netting i968: 23). A generous interpretation of this choice of scholarly
direction might be that I admired the apparent self-sufficiency of the isolated Kofyar, and their cultural vitality, and that I wanted to communicate
Prologue
5
to others some appreciation of the economy and "material culture" that
supported this African society. Perhaps it is more accurate to say that I
and my assistants had collected a lot of data in a fairly standardized form
on household membership by name, gender, age, and kin relationship,
and on the crops and domestic animals the household produced and consumed. I was impatient to begin my study before fully mastering the
language, and I found that a household census, with its repetitive questions, relatively straightforward answers, and generally nonsensitive content, was a good way for me to get acquainted with people and practice
my Kofyar. Perhaps a household survey also reflects a certain lack of
imagination. I remembered the advice of Sol Tax, one of my professors
at Chicago: "When you can't think of anything else to do, you can always
census."
But judging from my field notes, I spent as much time attending divinations, recording folktales, exploring witchcraft beliefs, and drinking
beer as I did talking about farming and observing work groups. Although Kofyar cultural concepts and behavior in such areas as gender
relationships, politics, warfare, and contacts with the British colonial
government have not been neglected (Netting 1964, 1969a, 1969b, 1972,
1974b, 1987), my core concerns have remained stubbornly centered on
issues of work, agriculture, households, and rights to the means of production. One can count on the existence of activities and things that can
be counted. The mundane, petty facts about residence, kinship, and
crops that individuals can tell the interviewer with reasonable accuracy
can be transmuted through numbers into statistics. From what people
know and see can come approximations of mean and central tendency,
classifications by age and sex and village of origin, information about
differences (with, one hopes, some measure of significance) between subsistence cultivators and cash-croppers, correlations of household size and
wealth. These are things we might guess at, but no one knows the answers accurately until you do the numbers. Moreover, unlike norms or
ethical principles or aesthetic judgments, quantitative measures of behavior are not part of people's collective consciousness. Though individuals
can and assuredly do make economic decisions about market exchanges,
stored food, and labor expenditures, they generally do so without bookkeeping and exact calculations. They have little way of estimating
changes in social behavior at the group level; indeed, there may be a
vested interest in asserting a somewhat spurious cultural continuity and
the strength of tradition (Murphy 1971). Statistical representation of a
decline in fallow, an increase in age-specific female fertility, or a process
of polarization in household incomes is not information that is available
~or people to apprehend or incorporate into their stock of cultural meanmgs. But these trends and changing relationships affect systems of farm-
6
Prolog11e
ing, labor, and landholding, and they can be analyzed by the observer
using the quantitative methods of practical reason.
It was in examining graphs plotting a regular association between field
area and crop production in intensive farming at Kofyar homesteads, in
contrast to the direct relationship between labor input and production on
cash-crop farms worked by migrants, that I first became aware of differently patterned agricultural systems (Netting 1968: 13 5, 205). It appeared
that farmers in the densely settled Kofyar homeland practiced permanent
cultivation of small fields , with high yields per unit of land, as opposed
to the same people's shifting cultiva,tion of abundant land on the frontier,
where fields were large and yields per unit of land were low. Moreover,
traditional homestead cultivators had small households and nuclear or extended families, whereas migrant farmers had statistically larger households, achieved by increased rates of polygyny, and more multiple-family
households. Household size appeared to correlate closely with land availability, and it va ried with different labor needs (Netting r965, r968).
The household was not a static traditional grouping generated by fixed
cultural rules of postmarital residence or the practice of polygyny (Goodenough 1956; Wilk and Netting 1984). Nor was it a predictable precipitate of a stage in the household developmental cycle of a social structure
at equilibrium (Goody 1958). Because farm labor was largely mobilized
and consumption organized in the family household, processes of household formation and fission might alter appreciably and quickly. The composition and structure of the household group, as it emerged from the
figures of hundreds of household censuses, varied with changed circumstances of production. And chis rapid adjustment was unaccompanied, as
far as I could see, by changed cultural standards or expectations about
household membership, marriage, socialization of children, or rights to
land . Kofyar customary systems of meanings remained intact and did not
constrain substantial nonrandom changes in social behavior.
Quantitative evidence of change was for me the genesis of recognition
that the smallholder household had readily distinguishable characteristics
related to a particular type ofland use under a specific population regime.
It also suggested certain limlted ethnological comparisons co test the posited functional relationships. The Chokfem Sura, who lived near the
Kofyar ancestral homeland in a similar plateau escarpment environment,
practiced shifting cultivation and had a lower population density, with
large, mulciple-family households. One could also predict chat as the
Kofyar filled up their frontier land, they would revert to more intensive
agriculture, with smaller fields, and that their recently augmented household size would begin to decline. These projections for change over rime
were in fact supported by a restudy of the Kofyar in 1984-85 (Netting et
al. 1989; G. D. Stone et al. 1990).
Prologue
7
Comparing and Generalizing: How to Recognize Smallholders
Like most people most of the time, the Kofyar have no means of reducing behavior to statistical terms, but they were quite ready to explain the
patterned actions that the enthusiastic anthropologist had "discovered."
Why shouldn't a young adult man remain in the parental household when
his father provided bridewealth and a motorcycle from the family's new
cash-crop earnings? Women recognized the value of an extra pair of hands
and pressured their husbands to marry co-wives (M. P. Stone 1988). Successful farmers argued with me over the costs and benefits of cooperative
beer-party work groups as opposed to wage labor. It may be that there is
a strong streak of practical reason in certain areas of Kofyar life, just as
there seems to be among other smallholders. It is also possible that the
economically minded investigator asks questions that elicit pragmatic responses. But just as systems of meaning and behavior are not exhausted
by a materialistic, ecological approach, so the utilitarian activities of production and reproduction are not solely culturally constituted or
changed.
Could the Kofyar be nothing more than an interesting, but perhaps
anomalous or idiosyncratic, ethnographic case? Smallholders are usually
thought of as peasants with an intermediate technology of the plow and
draft animals, living in a state, and subject to demands for tax or tribute
from other elite groups in the complex society. The Kofyar practiced hoe
cultivation in a rugged escarpment area of the Jos Plateau, where they had
remained largely outside the political system and market economy of
northern Nigeria's Hausa-Fulani kingdoms. Kofyar country was only
made part of the British colonial state in this century, and they have retained a large measure of control over their own land and production system down to the present. They did not fit easily into the standard peasant
mold.
The smallholder adaptation only became a generalizable category for
me inasmuch as it appeared to encompass other examples of peoples
practicing intensive agriculture and resisted conformity to the older and
more conventional typologies to which these groups had been consigned.
The most usual way of pigeonholing farmers is by contrasting technologies, often along an implicit evolutionary scale. Primitive farmers or horticulturalists use the hoe, the axe, the digging stick, and perhaps the
sickle, and agriculturalists add animal draft power and the plow to these
manual implements. The more developed technology captures nonhuman energy, presumably lowering human labor inputs and increasing agricultural production over larger land areas. Mechanization, energy from
fossil fuels , and scientific methods of fertilization, plant breeding, and
8
Prologue
crop protection (using pesticides and herbicides) carry the same evolutionary process further. Smallholders with relatively simple tools farming small, often fragmented fields, and relying on traditional "prescientific" understandings of agriculture, are automatically relegated to a
lower, and presumably earlier, farming type.
Scale and productivity are, however, slippery concepts. The bigger
fields made possible by the use of nonhuman sources of energy do indeed
save labor, and production per hour rises. But productivity as reflecting
production per unit of land may in fact be lower under more extensive,
technologically advanced systems. Because intensification refers to
achieving and maintaining relatively high land productivity over time, it
can be applied to farming systems with varying dependence on nonhuman energy. The Kofyar first claimed my attention because, with nothing more than iron-bladed hoes, digging sticks, and sickles, they
achieved relatively high and reliable yields from small land areas, using
compost manuring, intercropping, stall-feeding of animals, arboriculture, ridging for water retention and drainage, and terracing. The high
labor inputs of intensive agriculture increase yields and reduce variability
by conserving and enhancing soil nutrients and diversifying production.
If we include under the rubric technology the repertoire of skills, the folk
knowledge and ethnoscience brought to the task, and the building and
maintenance of intricate systems of irrigation, flood control, and drainage by means of hand tools, the evaluation of technology along the single
axis of "labor-saving" becomes inadequate. Intensive techniques applied
with care, and frequent monitoring of the field, garden, or orchard, also
imply a sustainable agriculture that prevents the erosion and degradation
that frequently accompany large-scale, extensive land use. Part of my
reason for beginning research in alpine Switzerland during the 1970s was
to see a system that had persisted for centuries in an easily damaged environment of steep slopes, short growing seasons, and low rainfall. Historical documentation attested to continuous use of irrigated mountain
meadows, terraced vineyards and grainfields, forests, and high-altitude
pastures with no evidence of erosion, declining soil fertility, waste of irrigation water, overgrazing, or deforestation. As in the Kofyar case, techniques of crop rotation, manuring, and controlling the tendency of
worked soil to creep downhill were practical rather than based on a
"scientific" understanding of hydrology and soil chemistry. Yet low-tech,
highly effective methods maintained relatively dense, permanently settled
local populations in a manner that both conserved and enhanced the production of existing natural resources. The modernist cant that traditional
intensive cultivators must be taught how to farm with machines, purchased inputs, and scientific knowledge is direccly contradicted by the
Prolog11e
9
land productivity, the reliability, the ecological sustainability, and the
adaptability of these systems.
The fact that the Nigerian Kofyar and the mountain Swiss are both
geographically and economically peripheral to the concerns of modern
industrial nation-states stimulated my interest in smallholders who not
only persist but play a domjnant role in market production as well as
subsistence. Though peasant smallholders have had an abiding presence
in the north of Portugal, the Netherlands, parts of Germany, and Denmark, it is the ancient wet-rice societies of Asia where the type is most
clearly and pervasively exemplified. There, with long-term, high- density
populations in China, Japan, and Java, skill replaces scale (Bray 1986),
renewable energy competes successfully with imported and mechanical
energy, and household management demonstrates its superiority to both
hjred labor and collective farming. The great Chinese river valleys have
proved more hjghly productive and more agriculturally sustainable,
through cycles of intensification, than any comparable region on earth.
Although the historic form taken by labor-intensive smallholder enterprises in the Chinese market economy (P. C. C. Huang 1990) may not be
a model for emulation elsewhere, it does suggest the durability and amazing resilience of the smallholder techno-social type. As an ethnological
comparison, the richly documented Asian cases best call into question the
reigning hegemonic ideal of large-scale, energy-expensive, mechanized,
specialized, scientific, capital-intensive, labor-saving agriculture enshrined by the West. Under certain circumstances of hjgh population
density and market economy, there is a viable smallholder alternative.
Myths of Modernization: Evolutionary Mystificatio11s and
Smallholder Persistence
Why have smallholders been ignored or regularly stigmatized as oldfashioned, resistant of innovation, inefficient, and a barrier to modernization? Almost from the beginning, my field experience tended to collide
with and contradict conventional views of a unilineal development in ag- '
riculture and a static subsistence segment. No outsider had recently introduced the Kofyar to concepts of composting green vegetation with goat
manure or preventing erosion by making rectangular ridges on top of
bench-terraces. When the Kofyar summarily discontinued these practices
in favor of slash-and-burn farming on the frontier, it was not a sign of
some evolutionary regression but a reasonable reaction to abundant land
and the desire to make labor more productive. Kofyar later bought and
used chemical fertilizer as well as motorbikes and trucks for transport,
but refused both the ox plow and the tractor, which got in the way of
IO
Prologi1e
intercropping. They responded to growing land scarcity by reintensifying agriculture on smallholdings rather than taking the path of cultivating
large-scale farms with hired labor. And the options they chose were fitted
by trial and experiment to a savanna environment with seasonal, soil, and
cropping differences that were in part new to them.
The Swiss smallholders had modestly revolutionized their alpine farming system twice, once with the adoption of the potato as a greater and
more dependable source of calories in the eighteenth century, and again
when garden tractor-mowers for cutting hay replaced the scythe a few
decades ago. This latter-day technology allowed agriculture to continue
as a part-time activity along with employment in industry and the tourist
trade. But the ancient peasant subsistence system had always coexisted
with and mutually supported households whose income came in part
from off-farm employment as everything from mercenary soldiers to
chambermaids. The security of a diversified and intensive farming system maintained an astonishing proportion of village family lines from
before i700 through IO to r3 recorded generations to the present (Netting 1981: 70-89), yet necessary cash and manufactured goods always
came from outside the community (Netting 1984).
Perhaps the most stubborn and pervasive myth about smallholders is
that their physical isolation in rural areas, their simple technology, and
their modicum of self-sufficiency remove them from dependency on a
market and the mentality of maximization, greed, private property, and
inequality that is thought to be the market's inevitable accompaniment.
Again, the evolutionary construct of the peasant mode of production, or
of a precapitalist social formation where labor and resources are shared
and reciprocity is unreckoned, did not seem to fit the intensive cultivators
I knew. Scarcity was not an artificial, arbitrary creation of some elite but
a condition of the ratio of population to land. Resources like irrigated
fields or terraced vineyards, where the investment of labor and capital
over years had built up and buttressed the productivity of the land, could
not readily be loaned to others or periodically reallocated among village
families. Ethnological comparison cross-culturally and through time assured me that intensive agriculture under circumstances of real, objective
limitations on arable land makes primitive communism impossible.
Where money, legal titles, notaries, and courts exist, as in medieval
Switzerland, land is bought and sold, and its price seems remarkably
high. But even where market relationships are economically insignificant
and no state legal system intrudes, as among the precolonial Kofyar and
the Philippine highland lfugao, households have clearly defined, very valuable rights in real property, and land is heritable. With the assertion of
continuing use, occupation, temporary exchange by loan or lease, and
public litigation over disputed rights, an institution very close to private
Prologue
11
property comes to exist, even if permanent alienation by sale seldom occurs. Individualized, socially recognized rights to scarce, highly productive resources and the improvements that increase and maintain their
yields are inherited along lines of close kinship or transferred in exchange
for other valuable goods. At the same time and place, land with low or
temporary production with little potential for intensi£cation, as in marginal, long-fallow bush fields or rough grazing areas, may remain in
communal tenure with occasional redistribution or shared, controlled access (Netting 1969a, 1982a). The documentary evidence that the resident
families of the Swiss village had exercised private property rights in irrigated meadows, grainfields, gardens, and vineyards since the thirteenth
century while maintaining legally instituted common property in the
community alp and forest convinced me that there was no evolutionary
watershed separating an earlier stage of communal rights from a later
period of private property emerging with the market and the state (Netting 1976). Smallholder intensive cultivators hold land, and, all other
things being equal, it is the ecological factor of land use that most
strongly determines land tenure.
.
One implication of the scenario depicting small cultivators as low producers with poor technology, little market participation, and communal
tenure is that they are homogeneous in property and wealth (Redfield
1941, 1955). Even as smallholding peasants within a state, they are perceived as economically stagnant and politically inert, forming a mass "of
homologous magnitudes, much as potatoes in a sack form a sack of potatoes," as Marx notoriously put it. 1 The closed corporate community of
peasants systematically cuts back emerging inequalities of wealth by diI. Marx 1971 [1852): 230. Marx found the French peasants of the nineteenth century a
~onservative, amircvolutionary group who cared only for their selfish and narro~ int~res.ts
1 ~ property: "The smallholding peasants form a vast mass, the members of which hve 1_11
similar condioons but without entering mto manifold relations with one another. ThCJr
mode of production isolates them from one another mstead of bringing them mto mutual
intercourse. The 1solatton is increased by France's bad means of commurucaoon and by the
poverty of the peasants. Their field of producuon, the smaUholdmg, admits of no division
oflabor in its cultivation, no application of science and, therefore, no diversity of develo.pment, no variety of talent, no wealth of social relationships. Each individual peasant family
is almost self-sufficient; 1t itselfdirectly produces the ma;or part of its consumption ~nd th~
acquires Its means of life more through exchange with narure than in intercourse with society. A smallholding, a peasant and his family; alongside them another smallholdmg, another
peasant and another family. A few score of these make up a village, and a few score of
villages make up a department. In this way, the great mass of the French nation is formed
by simple addition of homologous magmtudes, much as potatoes in a sack form a sac~ of
potatoes" (ibid.). It is noteworthy that the same characteristics of isolation, homogeneity,
and self-sufficiency that represent the srrcngth and cultural mtegnty of the folk soo~ry for
Robert Redfield conveyed to Marx only stagnation, ignorance, and a bu to evolunonary
progress.
12
Prologue
recting gossip and envy agamst the rich through an idiom oflimited good
(Foster 1965), redistribunng use rights in the commons, requiring leaders
to sponsor fiestas and host feasts (Wolf 1957), and relieving subsistence
cnses through fo rced generosity (J. C. Scott 1976).
Alchough such " leveling mechanisms" assuredly do exist, it is my impression that they by no means equalize access to resources within the
rural community. What smallholders have, they hold on to with a tight
grip, and they compete with vigor and craft for a scrap of garden, a larger
herd of goats, or a new granary. A single family household may grow
from relative poverty when an adulc couple supports many young dependents to a large, prosperous group with several productive workers. Because fa milies are at different po ints in their domestic developmental
cycles, and because they do not all follow the same trajectory, inequality
in wealth is the rule rather than the exception. There are few mechanisms
of gifts am ong kin, reli gious charity, or communal sharing that effectively redistribute such important resou rces as res1dent1al buildings, livestock, and land.
Inequality among smallho lders, as opposed to the quite profound differences among farmer-owners and merchants, government officials,
professionals, and landless laborers, is present and measurable but not
rigidly stratified. Over time, Swiss villagers showed considerable mobility, bo th up and down the economic scale, but they did not polarize into
a class of landowning, wage-labor-employing m anagerial farmers, or
kulaks, and an impoverished group of minifundistas and rural proletarians (M cGuire and Netting 1982). Even without substantial charity, public
redistribution, or institutional checks on accumulation, there seem to be
economic factors acting to circulate wealth. The Chinese case shows rich
farmers incurring transactio n costs for recruiting and supervising paid
labo r while households provide more skilled and dependable workers
willing to accept lower m arginal returns on their work. The combinatio n
of higher costs, lower production per unit of land, and high land prices
m eans that rich farmers can get better returns o n their capital in commercial or o ther urban occupations outside of agriculture. Partible inheritance may also divide a big estate among many sons of the owner. On the
other hand, intensive agriculture rewa rds managem ent skills, conscientious work, knowledge of resources, and careful, long-term planning.
Thus the m o re clever and industrious smallholders can potentially increase the size and wealth of their enterprises over time.
Population Parameters and the Smallholder
There is something of a paradox in the particular cast of thought and
theory that I bring to the problem of the smallholder as an enduring so-
Prologue
13
cial and economic type. Consideration, some of it quantitative, of systematic interrelations among factors of demography, technology, environment, economy, and social institutions, in search of cross-cultural
associations and regularities in processes of change is an exercise in practical reason, but it does not fall neatly into place with the major paradigms of materialism. The most general orientation toward the functional interactions of what I have called effective environment, productive
and protective technology and knowledge, and social instrumentalities
(Netting 1965) was that ofJulian Steward's cultural ecology. But despite
his theoretical emphasis on causal change and evolutionary patterns,
Steward (1938) was most persuasive m outlining a relatively simple
hunter-gatherer ecosystem from his own superb Great Basm ethnography. The dynamic roles potentially played by change in population, technology, or environment appear only obliquely in comparisons between
Owens Valley and other Paiute peoples or between Basket Maker and
Pueblo settlements (Steward 1955). Where intensive cultivators appear in
Steward's writings, they are pawns in the schematic play of hydraulic
power politics of Karl Wittfogel's irrigation civilizations.
For me, a more precise and better-articulated model of agricultural
change came from Ester Boserup. Reading her book The Co11ditio11s of
Agrirnltural Growth ( 1965) only a few years after completing my dissertation on Kofyar farming gave me an electrifying sense of an inclusive and
consistent pattern that logically accounted for both the regularities and
the processes of contemporary change reflected in my data. The intensive, highly productive, permanent agriculture of the Kofyar homestead
farm that I had described as occurring with dense local population, high
labor inputs, and individualized land-tenure rights (Netting 1963, 1965)
represented a subsistence type that apparently occurred worldwide as an
adaptive response to population pressure. Boserup also asserted that if
land increased in abundance, people would save labor by reverting to
more efficient shifting cultivation-which was just what the Kofyar were
doing on the Benuc plains frontier.
My own variation on the Boserup theme was to emphasize the role of
the small, nuclear or polygynous family household as the social unit that
typically mobilized labor, pooled consumption, and exercised tenure
over the intensively tilled farm. I saw household size and composition as
correlated with and responsive to farm area, cultivation techniques, and
especially the labor needs of the agricultural operation. It was easy for me
to postulate that a dense local population, drawn to the Jos Plateau escarpment both for its desirable rainfall and oil-palm vegetation and for
the protection it offered from slave-raiding neighbors, would create the
higher demand for subsistence food that gave a selective advantage to intensive methods of agriculture. The numerically preponderant presence
14
Prologue
in the census of relatively small family households as the units of farm
production and consumption, landholding, and residence suggested that
a social dimension could be added to the original Boserup formulation
that population pressure caused or made highly probable a more permanent and intensive system of cultivation.
The brilliant reductionism of the Boserup hypothesis seemed, however, to treat population growth as an exogenous factor rather than as a
variable element in a local ecosystem. Under what circumstances of
changing fertility, mortality, and migration did population increase? Were
there environmental limits to agricultural intensification beyond which
population could not grow, and could stability be achieved by social
means or only through the harsh imposition of Malthusian checks? My
attempt to reconstruct the demographic history of the Swiss peasant village of Torbel was an effort to examine the dynamics of a smallholder
system in which documented population change could be seen as both
caused and causal. The record reflected an alpine community continuously occupying a fixed agrarian territory, and a medieval population
dense enough to require impressive irrigation works for intensive dairy I
grain subsistence pursuits (Netting 1974a).
Torbel was not, however, a self-regulating ecosystem, delicately balanced in its mountain environment (Netting 1990). 2 In 1532-33, the
Black Death eliminated many local families and opened places for inmigrants, and the Napoleonic invasion of Valais coincided with a dip in
population (Netting 1981: 72, 118). The smallholder pattern, however,
persisted as village population doubled between 1774 and 1867, well before the advent of modern medicine (ibid.: 95-97). These results convinced me that an exogenous technological change-in this case, the introduction of the potato as a food crop-could promote an increase in
female fertility and raise the potential of village territory to support more
people. There was equally good evidence that local institutions of land
tenure, inheritance, marriage, and sexual control had operated to restrict
fertility by encouraging relatively late marriage and frequent celibacy on
the part of villagers while also promoting out-migration. Culturally specific ideals and practices of partible inheritance, monogamous marriage,
chastity, and long lactation figured in the Swiss demographic regime, and
other dense farming populations displayed functional systems different in
operation but similar in effect. The sometimes remarkable persistence
and continuity of the smallholder adaptation in this case appeared to lie
2. The homeostatic ecosystem with deviation-counteracting feedback loops was the favored model of b1ological ecologms in the 196os and 1970s {Odum 1969, 1971; but cf.
Worster 1990). As anchropolog1sts looked to technocnvironmental relationships co expand
their structural-functionalist formulations, the ecosystem became a major heuristic device
{Geertz 1963; Rappaport 1968; Flannery 1968).
Prologue
15
not only in the possibility of raising farm production to feed more people
but also in indirectly controlljng population growth itself. Despite
doomsday scenarios of runaway world population growth, smallholder
farms did not appear to be endlessly fragmented, their resources degraded, or their households impoverished.
The Smallholder Meets the Market
The greatest problem with modeljng a viable system of rural population, land, technology, and labor has been the tendency to treat such systems as self-sufficient and independent. In fact, smallholders do not normally live in isolation from larger networks of economic exchange or
political organization; indeed, the scarcity of their resources and their desire for goods and services they cannot produce at home necessarily involve them in important external relationships. Boserup tends to see a
more complex division of labor, specialization, and trade as stimulated
by the same population increase that fosters agricultural intensification.
But it is also possible that market demand (Turner and Brush 1987) and
the taxes or tribute of political systems that protect and extend the sphere
of market activity may impel cultivators to produce a surplus considerably above their subsistence requirements.
Ifland is abundant, as it was originally on the Kofyar frontier, extensive or shifting methods may be used to raise production most efficiently.
The original motivation for adding bush cash-cropping to exjsting intensive homestead farming was the desire to enter the market. As land availability and fertility declined on the fron tier, the Kofyar reintensified their
agriculture to maintain and even expand the amount of surplus food they
could sell. It is true that population concentration along roads or in periurban areas often coincides with truck gardening or the intensive production of crops of high value, like fruits, dairy products, condiments, and
flowers. But a unicausal model of smallholder intensive household farming systems that neglects either population pressure or market demands
is inadequate to account for the prevalence of the type.
Just as smallholders are seldom solely subsistence cultivators, so the
need to compensate for insufficient resou rces and turn unused agricultural labor to other productive purposes means that household members
will generally pursue a variety of full- and part-time occupations. Processing and selling food; cottage industries like weaving, basketry, pottery, and knitting; and sidelines in trade, transport, and construction may
all be potential sources of income for the farm family. The records of the
Swiss village ofTorbel showed that numbers oflocal men served abroad
as mercenary soldiers in the seventeenth century, and jobs as muleteers,
cheesemakers, herdsmen, mail carriers, cooks, wajtresses, and factory
16
Prologue
workers have helped co support farm families for the past 150 years (Netting 1981: 97-108). Though the household may not be a full-time agrarian unit, the pooling of income from many sources, periodic cooperation
to perform farm tasks, and the protection against risk that comes from a
diversified economic base increase the resilience of the smallholder enterprise.
The salience that I give to the smallholder household is mirrored in the
powerful characterization of the peasant household by the Russian economist A. V. Chayanov (1966 [1925)). The Chayanovian farm household
is a subsistence unit whose workers expend only the effort or labor
"drudgery" necessary to provide for the consumption needs of all household members. Although supported with impressive statistical data, this
radically simplified characterization does not fit the case of intensive cultivators. At the most basic level, in Chayanov's model, the demand of
more household consumers for more subsistence food is met by enlarging farm size or "sown area," an alternative not readily available on the
land-scarce smallholding. Though the peasant household's activities in
allocating labor and leisure conform in broad outline to the assumptions
of neoclassical economics, they take place in an idealized context where
there is no significant production for the market and no wage labor.
Though Chayanov's ideas have been appropriated by many social scientists as a general characterization of peasant economy, they struck me on
first reading as possibly applying only to shifting cultivators farming for
subsistence in a land-abundant environment. The consumer /worker ratio bore little relation to the per capita production of each Kofyar worker,
either on the homestead farm, where field size correlated with crop production, or on the big bush farms, where more effort was made by those
who wanted to participate in the market.
Indeed, Chayanov draws systematic contrasts between his Russian case
and that of peasants in Switzerland, where land is scarce and individually
owned, where labor input per hectare is inversely related to farm size,
and where holdings are unequal. Perhaps the major and still largely unacknowledged reason for the poorness of fit between the Chayanov
model and most intensive cultivators, whether peasant or not, is that the
Russian system was grounded in land-abundant, long-fallow cereal cultivation; a generally sparse rural population; periodic reallocation of
fields in at least some communities; the former system of estate serfdom,
where workers had little direct access to the market; and large multipleor joint-family households under patriarchal direction. In almost every
respect, traditional Russian farmers did not follow a smallholder pattern.
Unpaid household members can indeed produce subsistence when t~e
employment of wage labor would be unprofitable, but Chayanov is
merely specifying the conditions under which economic decision making
Prologue
17
cook place. Workers increased their per capita labor sufficiently to feed
household members, while minimizing the drudgery this entailed. I accept the motivational hypothesis that peasants are rational maximizers of
personal or family welfare, 3 but I would insist that intensive cultivators
calculate their interests over long spans of time, forgoing immediate benefits such as might come from cash-crop specialization in order to lessen
risk in the short term (Cancian 1980). Savings in order to buy land and
investment of effort and capital in land improvement are regularly made
to secure the interests of future generations and of the elderly. A narrow
neoclassical perspective may also deny the ability of peasants to take collective action for shared interests and manage common property at the
viJlage level (Popkm 1979). My own experience suggests that communities of farmers can support cooperative institutions for irrigauon, grazing, and forestry and can protect their resources from some hypothetical
"tragedy of the commons" (Hardin 1961). The dangers of free riders,
theft, and mutual mistrust that economists derive from a postulated individual rationality (Little 1989: 34) can be mitigated by institutions for
communication, monitoring, and sanctioning in active smallholder communities (Ostrom 1990).
Marx Against Smallholding
The tension between what I knew of the ethnographies of smallholder
households and their societies on the one hand and the prevailing schematic, often polemical, categorizations of peasant cultivators on the other
became particularly acute for me when I confronted the dominant concepts of Marxism and political economy. Neither the "lineage mode of
production" nor the various descriptions of precapitalist and peasant social formations coincided with the systems of land use and social organization of the intensive cultivators I knew. The more aggressively materialistic and doctrinaire the political-economic assertions, the more rigidly
e~olutionary and abstract were the generalizations. In "primitive" societies, seen as both technologically rudimentary and representing an ea rJ. Those who, as I do, criucizc a strictly "cultural" approach to undemanding hu~un
behavior and insmunons arc said to be guilty of a simplistic and reductionistic econom~m.
Econom1sm "is the view that the moving forces m md1v1dual behavior (and ihus 10 society,
~hich is taken to be an aggregate of 111d1v1dual behaviors or some straaficauonal arra~g~ent of them) are those of a need-dnven unhty seeker manocuvenng for ad van cage within
a context of material poss1b11iues and normaave constraints" (Geertz 1984: 5i6). Sahlins
makes a similar point when he refers to "the home-bred economizing of the market place
· · · transposed to the exphcauon of human society" (1976: 86). While I have med to understand certain hm1ted kinds ofsocial behavior as they relate to work, household organization,
and
·
f" · "
property rights in economic terms I offer no comprehensive explanaaon o soaecy
~~I
·
'
·
cu cure as "a moving and diversified frame of socially constructed mearungs " (~m
1984 513),
18
Prologue
lier stage in cultural development, agrarian resources were supposedly
not scarce, and neither did population pressure lead to competition.
Rights to land were believed to be held communally, and the inequality
that derives from private property had allegedly not yet emerged. The
household as an important, semiautonomous unit of production and consumption tended to disappear.
In line with Lewis Henry Morgan and other nineteenth-century evolutionists, Marx believed that "in the most primitive communities work
is carried out in common, and the common product, apart from that portion set aside for reproduction, is shared out according to current need"
(Engels 1884, quoting Marx's letter to Vera Zasulich, as cited in Meillassoux 1972: 145). French neo-Marxist anthropologists (Meillassoux i981;
Terray 1972) have found in traditional African societies a lineage mode of
production where descent-group work teams farm collectively, store the
produce in communal granaries, and receive food allocated by elders.
This may well typify large multiple-family households of shifting cultivators, and lineages may indeed allocate land and provide for territorial
defense (Johnson and Earle 1987; Netting 1990), but I know of no cases
anywhere where descent groups above the level of the household were
the primary social units of production and consumption. Though intensive cultivators may have a variety of reciprocal exchange and cooperative
labor groups, and though their communities often administer clearly defined rights in common property, households characteristically farm and
eat separately, providing for their own reproduction, and protecting
rights in valuable, heritable property.
If I had cherished any expectation that the Kofyar, like many huntergatherer groups, were primitive communists, because they had only recently been incorporated in a state or a market economy, their firm insistence on property rights in land and livestock and their autonomous corporate households rapidly disabused me. Given the lack of sound
comparative data in evolutionary formulations and the "wistful romanticism" of the nineteenth century, it is perhaps no wonder that Marx
"clearly failed to realize the complexity of rights over property, including
property in land, characteristic of a primitive agricultural community"
(Firth 1973: 36). It is less easy, however, to justify the equation of simple
farming technology with lineage productive groups, communal rights to
resources, and primitive egalitarianism in an anachronistic evolutionism
that is still with us.
It can be argued that I have been self-deluded by ethnographic wilJ-o'the-wisps, projecting functional integration and timeless stability on contemporary groups of numerically insignificant cultivators who are technologically backward, economically undeveloped, and peripheral to the
capitalist world system. For reasons of geographical isolation, folk cul-
Prologue
19
cural conservatism, or political-economic exploitation by the colonialist
state or merchant capital, the Kofyar and the mountain Swiss may be seen
as merely the detribalized or impoverished peasant remnants of previously autonomous, healthy societies. Although I must indeed plead
guilty to consciously seeking out groups that have, or recently possessed,
a modicum of self-sufficiency, a traditional low-energy tool-kit, and few
direct relationships with dominant economic or governmental elites, I
have come to feel chat groups of smallholders with some essentially similar characteristics exist and persist in a wide range of social formations
throughout the world. Their distinctive system of rural population density, intensive land use, household organization of production, and private property rights cannot be consigned to some evolutionary stage.
Smallholders may use hoes or ox plows or tractors and live in rain forests
or oases or temperate savannas. Their mode of agriculture or horticulture
is not regularly associated with one set of political institutions, be it tribe
or chiefdom or state (Sahlins and Service 196o; Netting 1990).
Marxist attempts to place smallholders unequivocally on one side or
the other of some great historical divide separating use values and accompanying lack of accumulation, capital, and private property from exchange values with scarcity, inequality, wage labor, alienation, and capital
(Firth 1975) seem to me to fabricate a Procrustean bed. Though households and village communities may appear inextricably bound up with
the practice of intensive agriculture, more inclusive "relations of production" to absentee landlords or tax collectors or moneylenders seem more
variable and less determinant (Attwood 1992: 42). I have avoided the
"mode of production" designation as well, in part because of the resource
abundance and production limited to basic needs implied by the lineage
mode (Meillassoux 1981), the domestic mode (Sahlins 1972), the kinship
mode (Wolf 1982), and the peasant economy (Chayanov 1966 [1925]), but
also because smallholders flourish in such a variety of ideological and political contexts that links between infrastructure and superstructure become tenuous (Legros 1977; Friedman 1974) .
. It may be foolhardy of me, and it is certainly unfashionable, to question the singular role of capitalism in transforming peasants from 5 ?1allscale, communal subsistence farmers to market-dependent, economically
polarized rural people. But I have trouble finding those intensive. cul~iv~
tors of the Alps, the Low Countries, northern Iberia, or Scandinavia m
Charles Tilly's succmct, magisterial characterization of European peasantry:
~he peasant version of subsistence farming-in which land-controlling houseolds devote a portion of their production to the market-expanded under the
early phases of capitalism and state-making before declining under the later
20
Prologue
phases of the same processes. Capitalism reinforced private appropriation of the
factors of production and gave priority in production decisions to the holders of
capital. Thus capitalism challenged the collective use of the land, resisted the fragmentation of rights to the same land, labor, or commodiues, and worked agamst
the autarky of the household or village. By the same tokens, capitalism provided
farming households with the means and incentives to dispose of a portion of their
products for cash outside the locality. These features of capitalism promoted the
conversion of a large number of peasants mto agricultural wage-workers, pushed
another large portion of the peasantry out of agriculture toward manufacturing
and services, and gave a relatively small number of peasants the opportunity to
become prosperous cash-crop farmers. (Tilly 1978: 408)
The admitted concentration of wealth and power in the factories of the
Industrial Revolution, or even in the extensive agriculture of commercial
East Prussian grain estates or Spanish sheep farmers of the Mesta (Tilly
1978: 410), may not reflect smallholder social processes under structuraJly different systems of organization and land use. We need also to explain why capitaljst landlords were often unable to dispossess an existing
smallholder peasantry. Land use does make a difference. Small-scale, intensive agricultural producers are not synonymous with "peasants,"
"precapitalist subsistence farmers," "petty commodity producers," or
"rural proletarians," and smallholders demand their own explanations.
The literature of social science generally defines peasants not so much
by what they do as by what they don't do, and by what is done to them.
Marx found them a politically inert mass, lacking a consciousness of their
own class status and unwilling to join the industrial proletariat in revolution. Modernization theorists have cJajmed that peasant conservatism and
traditional values prevented the technological innovation necessary for
land consolidation and economic development. A similarly universalistic
dependency theory insisted that the social dynamics of agrarian societies
are everywhere the same, varying only to the degree that production is
oriented to external markets. Where external forces of world capitalism
are overwhelmingly powerful, factors of regional and local ecology and
history can have only negligible explanatory significance (cf Attwood
1992: 12). Anthropologists have viewed peasants as politically and economically subordinate. "It is only when a cultivator is integrated into a
society with a state-that is, when a cultivator becomes subject to the
demands and sanctions of powerholders outside his social stratum-that
we can appropriately speak of peasantry" (Wolf 1966: 11). Peasants "have
very little control over the conditions that govern their lives," and the
basic decisions that keep them poor and powerless are made outside their
communities (Foster r967: 8). Although I find the study of history as a
material social process theoretically trenchant and illuminating, it does
not seem to me that smallholders are adequately encapsulated in the
Prologue
21
"peasant" category of a political economy that primarily analyzes "social
relations based on unequal access to wealth and power" (Roseberry 1989:
44). While not denying the elements of political and economic domination that affect many aspects of smallholder life, I contend that we must
also examine the ecological relationships of population, agricultural technology, household organization, and land tenure that characterize a distinctive smaJlholder adaptation to local environment.
Julian Steward's strength was in part that his cultural ecology never
tried to explain everything, 4 and the cross-culturally recurring elements
of the smallholder pattern smack more of limited, middle-range explan ations than of the technoenvironmental determinism of nomothetic cultural materialism (Marvin Harris 1969). But as I looked beyond my own
field experience to the ethnographies and histories of peoples as different
as the Ifugao, the Aztecs, the Chinese, and the Dutch, recognizable
smallholders emerged from the obscurity of evolutionary stereotypes and
overdetermined categorizations.
Modernization and Evolution: Smallholder Greening
or Withering Away?
Evolutionary schemata combine conjecture about the past with an
evaluative conception of the present and speculation about what represents "progress" in the future. It is intriguing that for both the socialists
and communists of the left and the free-market capitalists of the right, the
agreed-upon path to agricultural development has been the large-scale,
mechanized, energy-dependent, scientific, industrialized farm. Smallholders have been universally stigmatized as unproductive, regardless of
their yields per unit of land, on the grounds that (1) they use too much
labor; (2) they do not produce a large surplus for the market; and (3) they
do not make rational economic and scientific decisions about production
and innovation. For most of my professional life, I have been content to
remain within the conventional anthropological niche, attempting to
understand human behavior in small-scale ethnographically specific societies with preindustrial economies. But the dominant evolutionary paradigms of agricultural change, both within and beyond the Third World,
have increasingly seemed to be contradicted by the practice, and ultimately the logic, of the smallholder pattern.
Modernization theory that laid out apparently obvious stages of global
agricultural development in the flush of optimism after World War II prescribed economic growth through applied science, capital investment,
mechanization, a need-for-achievement mentality, and vastly increased
4. As Geertz pomrs our, culrural ecology "forms an explicitly delimned field of inquiry,
nor a comprehensive narural science" (1963 : 10).
22
Prologue
labor productivity (Roscow 196o). It was believed that foreign aid and
developed technology would inevitably (and rapidly) replace traditional,
stagnant subsistence cultivation, freeing the poor and "underemployed"
rural masses for the urban industrial sector where they belonged. I envision the emblem of this movement as a tractor triumphant on a field of
Iowa corn. Its flaws, at least in Africa, had something to do with the
Western hubris that ignored the existence of working indigenous solutions to the problems of farming an alien environment. Local ethnoscientific knowledge of soils, rainfall, crop varieties, and pests was not appreciated by outsiders and could not readily be duplicated on experimental
farms and in laboratories (Paul Richards r985). No one seemed to consider the fact that bigger fields, even with machine plowing, might require more seasonal weeding and harvesting labor than a typical household could muster (Baldwin 1957), or that shifting cultivators might be
working less and enjoying higher returns per hour than intensive farmers
(G. D. Stone et al. 1990). Furthermore, the costs of high-tech irrigation
systems and manufactured inputs for rice or wheat in West Africa were
far in excess of what the disappointing yields would ever cover (Pearson
et al. 1981; Andrae and Beckman 1985).
Where modern Green Revolution technology of high-yielding rice varieties, chemical fertilizer, and improved irrigation was most effectively
adopted, Asian systems of intensive cultivation were already operating
and the scale-independent innovations were accessible to smaJlholders.
Agricultural economists also began to detail the advantages of household
labor in comparison to the opportunity and transaction costs of hired agricultural workers (Binswanger and Rosenzweig 1986; Pingali et al.
1987). The spontaneous and very effective effort of the Kofyar to produce
a surplus for the market, usmg hoe technology and mobilizing labor by
household and reciprocal means, suggested that a mechanical modernization model of development was not adequate to understanding this
process (Netting et al. 1989; G. D. Stone et al. 1990).
The template of large-scale, monocropped, labor-saving agriculture
dependent on fossil fuels was applied with equally uncritical abandon to
systems of production in the socialist countries. There, however, the vision of radically reformed farming brought with it an ideological stress
on the communal organization of production and the abolition of inequality founded on private property rights. The history ofland reform
after the Russian Revolution, followed by Stalin's forced collectivization
campaigns, the violent seizure of grain, and massive rural starvation suggested the depth of peasant resistance to communist economic policies.
The notorious failure of collective farms to allocate the factors of production efficiently and to provide incentives for responsible skilled-labor in-
Prologue
23
puts has been a major cause of the long-term Soviet agricultural crisis
(Shanin 1990: 188-205).
Even in the midst of those huge spreads of dry wheat, invariably depicted with ranks of combine harvesters under fair summer skies, the
individual household plots of collective members remained significant
suppliers of the nation's food. Tim Bayliss-Smith (1982) describes a
3, 144-hectare collective farm in the Moscow district with only 7 percent
of its I, 700 arable hectares in private plots. Yet over half of all household
labor time was devoted to these fields and gardens, achieving yields per
hectare that were six times those of the collective area. Energy returns on
each calorie of input were esumated at 11 . 2: 1 as opposed to 1.09: 1 on the
heavily mechanized collective farm. Although the private plot was not
technically owned by the user, it resembled a smallholding in being near
the house, receiving constant care and attention, being heavily manured,
and supporting diversified plant and animal production. Decisions concerning these small operations were made by the household, with women
playing a more important role than men. Such management differs decisively from that of the collective, where "any response by farmers to signals from their environment must, in all important respects, be made
with reference to instructions from a remote bureaucracy, instructions
which cannot foresee all the local vagaries of weather, disease, and soil
conditions" (Bayliss-Smith 1982: 97). The inherent contradiction between smallholder efficiency and communist ideology has at last been
recognized by Gorbachev's call for the freeing of Soviet farmers from the
state-run system of collective agriculture. "Comrades, the most important thing today is to stop the process of de-peasantization and to return
the man to the land as its real master," Gorbachev has been quoted as
saying (New York Times, October 14, 1988). The return of private property in some form is a foregone conclusion.
When I began to think tentatively about the extension of some general
characteristics of smallholder farm households to cultures with long traditions as complex civilizations, the case of China was both attractive and
problematic. Even the casual student (and I am no sinologist) is aware
that Chinese agriculture uses methods of double-cropping, controlled irrigation, fertilization, and terracing in a highly productive system of wetrice agriculture, and that China supports an unusually dense rural population. Given a continuous, documented history, I inferred that the tools
and techniques of intensive agriculture and perhaps some indications of
labor organization and land tenure could be investigated over time. Recent work on Chinese agricultural history (Hsu 1980; Chao 1986; Anderson 1989), on sustainable energy input /output measurements (Wen and
Pimentel 1986), and on the distinctive features of comparative Asian wet-
24
Prologue
nee economies (Bray 1986) suggested the exciting possibilities for such
an mvesugation. It was the massive, politically inspired, and centraJly
planned institution of agricultural collecuves m the People's Republic of
China in the 1950s that really piqued my cunosity. If an archetypal system of intensive cultivauon could indeed be carried on by extrahousehold work teams, if harvested goods and income could be shared
equally, and 1f private property could be replaced by communal control
of pooled land and livestock, my smallholder speculations would be just
another romantic anthropological Just-So story. Marx's primmve communism might in fact have a dramatic modern analogue, and there might
already exist a bountiful, egalitarian future at work on the farm. According to many popular and scholarly reports, to propaganda, and to theoretical analysis, collective agriculture had succeeded.
It was only in the mid-198os that I began to hear of an official retreat
from communes, brigades, and work teams and their replacement by a
national agricultural policy of bao,{!an daolm, or the household responsibility system. Households were given use rights on collectively owned land,
and they contracted to fill quotas for delivery of certain products at fixed
prices to the state in retllrn for the right to dispose of their entire surpluses on the free market (Perkins and Yusuf 1984). An administrative
change initiated in 1978 encompassed the majority of C hinese villages by
1982 and was all but universal by 1984 (SmiJ 1985; Hartford 1985). Linking economic rewards directly to output and encouraging household initiative, innovation, investment, efficiency, and risk-taking explosively
raised producuon and the rural standard ofliving, effecting what has been
called "the most far reaching and orderly socioeconomic transformation
of the 20th century" (Smil 1985: 118). An unparalleled experiment in the
social engineering of agriculture had been reversed, and I resolved to follow the smallholder story to China.
Any discussion of smallholder household farming worldwide raises
expectations that it cannot fulfill. The tendency of many Americans is to
categorize the issue as the familiar one of the family farm, whose demise
at the hands of agribusiness, a nanonal government of subsidies and controls, a volatile land market, the agricultural-research establishment, and
international commodity trading has been heralded for decades. Values
ran ging from the agrarian democracy of Thomas Jefferson to the virtues
of raising hard-working and pious children in the salubrious country air
have been enlisted in this debate, and we are now entering a decade in
which questions of the mythic economies of scale (Strange 1988), stewardship of the land (Berry 1987), and sustainable systems (Francis and
Youngberg 1990) wiJI be rightfully (and righteously) publicized.
I once thought that the plenitude of North American land, its frontier
history of relatively sparse rural population, its marvels of technology
Prologue
25
TABLE P I
Agrimlt11ral 011tp11t
Hectares per
male worker
United States
England
Denmark
France
Germany
Japan
1880
1970
1880
25
17
165
9
18
16
12
2
0.5
1
1
7
6
I
34
111
Six Co1111tries, 1880 and 1970
Output per
hectare
1
Output per
male worker
1970
1880
1970
I
13
16
11
7
8
2
157
88
3
5
4
I
5
3
10
SOURCE: Boserup t98] : • OL Used wuh perm1n ion.
NO TE : Htctarts refe~ to agncultural area- that is , to
94
60
65
16
Ferulizer
(kg/ ha)
(1970)
89
258
223
241
400
386
Workers
per tractor
(1970)
2
3
45
--- -
all land on farms. 0111p111 refer. to output of
both crops and animal products (excluding fodder consumed b) the farm animals) This output has
been recalculated in wheat units, equivalent to one ton of wheat, by YuJorO Hayanu and Vernon
Ruttan ( 197t). Worktrs includes adult male workers, but not women and children Kilograms of
chemical fertilizer arc measured in feruhzer content per hectare of arable land-that 1s, agricultural
area minus pasture and fallow
and science, and its cheap power made most considerations of smallholder intensive agriculture here, and by extension in any modern state,
either trivial or anachronistic. I was wrong. Quantitative comparisons
among leading industrial nations show that the logic of population density and agricultural intensification distinguishes types and defines trajectories through time (sec Table P 1). Since 1980, U.S. male workers have
greatly increased the areas they farm and their output, but output per unit
ofland has not kept up with that of Denmark, France, and Germany, and
is only one-tenth that of Japan. Surprisingly, fertilizer per hectare gives
the same picture of extensive U.S. as compared to intensive European
agriculture. Although productivity grew in all modern nations from
1880 to 1970 (Ruttan 1984), Japan and Germany have raised land output,
while the United States has pushed up labor output (Fig. P 1). Secular
trends in the relationship of population to arable land, the cost of energy,
and the dangers of erosion, chemical pollution, and declining fertility
point to more intensive and sustainable methods of land use on an Asian
or European model as an inescapable necessity even for the United States.
We ignore the proven advan tages of household labor, management, ownership, and inheritance at our peril.
My purpose in this book is not to describe U.S. agriculture or prescribe some illusory smallholder panacea. The adaptations that will
surely come with changing energy prices (Pimentel 1973), government
subsidies (Strange 1988), water and soil depletion, and new models of
alternative agriculture (National Research Council 1989) have legions of
able and passionate expositors. The large class of self-provisioning family
farmers that existed in the United States before World War II has almost
disappeared in a sea of specialized commodity production, and "by the
Prologue
26
-~
c
japan
6.0
1970
::l
~
_g
~
4.0
""...
..c
....
~ 2.0
::l
1880
0..
:;
...""::l
..!:::
::l
u
U.S.
Germany
0
1.0
"5, 0.5
<
1880
1970
I
t88o
France 1880 U.K.
Denmark
188
U.S.
5
10
50
100
200
Agriculrural output per male worker (wheat units)
Fig. P 1. Historical growth paths of agricultural productivity m the United
States, Japan, Germany, Denmark, France, and the United Kingdom,
1880-1970. (From Ruttan 1984: !09)
1980s the farm household had become virtually isolated from the farming
operation" (Adams 1988: 467). Where facets of the o ld smallholder system remain, as among the Amish, in the diversified, family-descended,
"yeoman" operations of Illinois (Salamon 1985), and in part-time and retirement farms in Georgia (Barlett 1987b), I shall note them. But it is in
the Third World, with its sweeping dynamics of population growth,
market expansion, and agricultural change, where an understanding of
the smallholder pattern may be most relevant.
Though I never became the applied anthropologist and secular missionary that I expected to be when I entered graduate school, my concerns with the cultural ecology of cultivators, and more generally with
social change, have brought me into repeated contact with students of
development. From the vantage point of an occasional consultant, I
looked at the economics of rice production in Senegal and the Ivory
Coast, the proposed dams on the Gambia River, and the small farmers in
northern Portugal o n the eve of EEC membership. Perhaps even more
important was my long-term view of the indigenous development of the
Kofyar as they entered a market economy with almost no outside planning, extension, new technology, or credit (Netting et al. 1989). Reports
of the death of the smallholder m a modern high-tech, large-scale world
have proved to be vastly exaggerated. Indeed, scarcity of rural resources
and national demands for food production create just those circumstances
in which agriculture intensifies and the household organization of pro-
Prologue
27
duction demonstrates its comparative advantage. But in the shadow of
proliferating industry, bureaucracy, and education, smallholders often
become invisible or embarrassing.
The fact is that Asian smallholders, with an assist from Green Revolution methods, have astoundingly kept ahead of food demands, sidestepping the sinkhole of involution in Java, and liberating unimagined
effort and productive enterprise in China. Where state policy has allowed
agricultural prices to escape the controls of marketing boards, and where
inputs like fertilizer arc no longer left to the (mal)distribution of parastatal firms in Africa (Bates 1981), rural people have usually responded with
a deluge of food. Where conditions of population density and market
relationships are conducive to intensive, sustainable production, we can
reasonably expect smallholders to rise phoemxlikc from the ashes of
thousands of collective farms. Even the indolent estates and deforested
ranches of Latin American oligarchs may someday become the scene of
much higher yielding gardens and orchards (Anderson 1990) if they are
portioned out to solvent, experienced peasant households by effective
land reform. I believe that inccnsive agriculture by landowning smallholder households is economically efficient, environmentally sustainable,
and socially integrative. This book is an attempt to use the evidence of
social science, the logic of practical reason, and the personal conviction
of a garden-variety ethnographer to show how the smallholder works.
40
Intensive Farm Practices
the production of firewood for warmth and cooking and of timber for
construction. That a single community could encompass and afford access to this diversity of products meant that a relatively dense population
in an area of constrained biotic potential could maintain a remarkable degree of security.
Despite the appearance of difficult, marginal subsistence potential in
the Alps, agricultural change could increase the potential of this area and
allow more intensive exploitation. Grain crops are not well adapted to
high mountains, and even hardy winter rye may yield poorly if spring is
late and an overcast, rainy early summer retards ripening. There is evidence that climatic conditions of this type led to widespread harvest failure in the latter part of the eighteenth century. It was at this time that the
Swiss, along with other European farmers, were able to broaden their
subsistence base by the adoption of the potato, an American tuber crop
domesticated in the Andes, where environmental conditions resemble
those of the Alps. In Torbel, potatoes could produce almost 15 times as
much as rye from the same area by weight and 3. 3 times as many calories
(Netting 1981: 163). Moreover, growing potatoes did not mean that
grainfields had to be sacrificed. Previously, rye had been planted every
other year in a field, but it was found that potatoes could occupy the
fallow period and thus raise the total production of the land. They could
grow in poorer soil and at higher altitudes than rye, could occupy small
patches of land too steep to plow, and resisted the hailstorms that devastated grain crops. Potatoes, which are rich in carbohydrates, could supplement a diet that already had good sources of proteins and fats in its
dairy products. The new food crop, which rapidly became a regular item
of daily consumption, increased the support capacity of the existing land,
and in Torbel and similar European alpine communities, it may have
been instrumental in sustaining a doubling of population over the next
hundred years (Netting 1981: 164-68; Viazzo 1989: 212-14, 269).
Protection of agricultural plants and animals among the Swiss was less
a matter of inhibiting the growth of competing organisms, as in the tropics, than of sheltering livestock from the elements and keeping domestic
animals out of the crops. The lush individual meadow plots, which averaged only about I ,300 square meters each, were too small co divide up
with fences or hedgerows. There were a few wooden fences along heavily
used paths, but animals were usually either closely herded when they
were in transit to other barns or pastures or tethered to a peg and moved
about in a meadow of their owner's. Today, temporary pasturage is defined by portable electric fences, powered by batteries, which are strung
up to enclose small areas. Because of the long winters, all livestock must
be confined to barns or stables for warmth and proximity to their stored
fodder. Substantial log buildings with roofs of local slate ring the Swiss
Intensive Farm Practices
41
villages and hamlets of Valais and dot the meadows. Sometimes a barn
with its haymow will be joined to a cabin for temporary residence when
the farm family must care for cows at some distance from the settlement.
A farmer might have access to five different barns, but in many cases only
a fractional share, averaging two-fifths of the structure, would be owned.
In the past, groves of tall trees were left standing around stone-walled
corrals on the alp to give the cows protection from late spring and early
fall storms.
Valuable crops may be subject to specific pests, and today farmers use
a variety of chemical sprays. Potatoes are menaced by the Colorado
beetle, and the grapes must be sprayed periodically against mildew and
insect larvae. In the nineteenth century, the phylloxera epidemic ruined
many of the Torbel vines, and the vineyards had to be replanted with
American root stock. There is a large traditional pharmacopoeia for
treating animal diseases, and in the past, the death of one of the average
family's two adult milk cows would have been a severe economic hardship. Today scientific breeding has produced animals with higher milk
yields, and veterinary medicine is readily available. Though dogs were
too expensive to maintain in former times, many families had cats to keep
down rodent infestation that threatened stored foodstuffs . In the longdomesticated Swiss landscape, there have been no large animal predators
for many years. Theft oflivestock was also unlikely in the small, isolated,
face-to-face community.
Wet-Rice Fam1ing as an Intensive System Par Excellence
Perhaps the epitome of intensive agriculture is represented by the irrigated wet-rice systems of Asia. No brief survey can do justice to the technological achievement, the ecological sophistication, and the variety of
such systems, but a few examples may illustrate the principles and some
of the specific practices of cultivation. The smallholder who cannot
gather the gram of a wild cereal grass from a swamp or a seasonally inundated nver valley, and who lacks the forested tropical uplands to grow
dry nee by slash-and-burn methods, must domesticate and control an
environment specifically adapted to this plant. The underlying soil may
be quite poor in plant nutrients, and sources of organic replacement may
be limited, but managed water can supply these deficiencies and has done
so over centuries of use. Clifford Geertz (1963: 31) likens this means of
converting natural energy into food to "the fabrication of an aquarium,"
a felicitous image that recalls the first model of a balanced ecosystem that
many of us learned about in elementary school science textbooks. On a
wet-rice terrace or pond field, the base of impervious clay, the low banks
or bunds around the edges, the precisely leveled surface, the water inlets
42
Intensive Farm Practices
and the outlets, are all designed to retain water at measured depths for
accurately timed periods in accordance with the needs of the developing
plant. The "rice-growing brew," though it may be from rainfall, serves
its fertilizing function best if it conveys dissolved nutrients and silts from
a river or some other external source (Hanks 1972: 37). Irrigation water
not only restores crop-depleted nutrients to the soil annually but also
promotes the fixation of atmospheric nitrogen {up to 50 kg / ha) through
the symbiotic association of blue-green algae and fern that it supports
(Altieri 1987: 76). In the words ofL. M. Hanks:
A flooded field is short on oxygen .. . , yet oxygen becomes available through
the bacteria chat break up the organic products of fermentation. In chis dank airlessness nitrogen is converted by ocher bacteria into ammonia rather than more
familiar nitrates. In the presence of ammonia the brew tests more nearly neutral
or even [more] aJkaline than acid, so that phosphorus becomes available to plants
as ferrous and manganese phosphates rather than the more familiar phosphorous
acid. UnJike roots in dry soil, the rice roots at the bottom of a flooded field serve
mainly co anchor the plane, while the higher rootlets drink in the necessities of
growth. (Hanks 1972: 33-34)
The chemical and bacterial decomposition of organic material, including
the remains of harvested crops, contributes to this process. 5 Dissolved
nitrogen can be brought in by the water, especially if there is a slow
fl.ow through the pond field (Seavoy 1986: 156), and the rice rizosphere
can also fix considerable amounts of atmospheric nitrogen (Ruthenberg
1976: 184).
Where irrigation is practiced, the high water table prevents the vertical
movement of fluids, thus limiting nutrient leaching (Altieri 198T 76).
Standing water also fills a protective function, shielding the soil surface
from high temperatures and the direct impact of rain and high wind that
could induce erosion. An inundated field restricts the growth of many
weed plants that would compete with the growing rice for space and nutrients. With water as the crucial variable determining rice productivity,
the timing of its application is also important. "Paddy should be planted
in a well-soaked field with little standing water and then the depth of the
water increased gradually up to six to twelve inches as the plant grows
and flowers, after which it should be gradually drawn off until at harvest
the field is dry" {Geertz 1963: 31). Small fields permit the maintenance of
an even depth of water and varying it on a schedule that may stagger crop
maturity so that the critical transplanting and harvesting operations need
5. Under dry-land conditions in the tropics, the nitrogen in organic matter is changed
during decomposition into ammonium, rapidly oxidized to nitrate, and then quickly
leached or dcnitnfied into a gaseous form that is Jost to the atmosphere (Bayliss-Smith 1982:
72). Waterloggmg slows the decomposition rate of organjc matter and the rate at whlch
nitrogen oxidizes, making nitrogen available to the growing wet-nee crop.
Intensive Farm Practices
43
not be performed on all fields at the same time. Even temperature can be
regulated by moving water along a short course, thus retaining warmth,
or over a longer distance that allows it to cool (Hsu 1980: 119). Creating
and precisely controlling a liquid microenvironment for rice not only optimizes and stabilizes crop yields but may in fact improve soil quality
under permanent land use over the long run of 50 to 100 years (Ruthenberg 1976: 184).
On both irrigated and rain-fed fields under a variety of crops, Asian
intensive cultivators used a wide variety of complex and laborious fertilization techniques. Texts from the Han period in China (206 B.C. to A.O.
220) show the existence of an elaborate soil classification of three grades
and fifteen types, along with the knowledge of how their agronomic
qualities could be improved (Hsu 1980: 94). Green manuring by cutting,
burning, and soaking weeds or by turning them under is attested by the
fifth century B.C. Terra-cotta models of Han pigpens with adjoining privies show how pig manure and human feces were collected by pipes and
channeled to a lower clay plate for drying and later distribution to the
fields in powdered form (Hsu 1980: 97). The most careful fertilization
was devoted to the seedbed, as in the instructions of a medieval Chinese
agronomist quoted by Francesca Bray:
In autumn or winter the seedbed should be deeply ploughed two or three times
so that it will be frozen by the snow and frost and the soil will be broken up fine.
Cover it with rotted straw, dead leaves, cut weeds, and dried-out roots and then
bum them so that the soil will be warm and quick. Early in the spring plough
again two or three times, harrowing and turning the soil. Spread manure on the
seedbed. The best manure is hemp waste, but hemp waste is difficult to use. It
must be pounded fine and buried in a pit with burned manure. As when making
yeast, wait for it to give off heat and sprout hairs, then spread it out and put the
hot fertiliser from the centre to the sides and the cold from the sides to the centre,
then heap it back in the pit. Repeat three or four times till it no longer gives off
heat. It will then be ready for use. If it is not treated in this way it will burn and
kill the young plants. Neither should you use night soil, which rots the shoots
and damages human hands and feet, producing sores that are difficult to heal.
Best of all the fertilisers is a mixture of burned compost, singed pigs' bristles and
coarse bran, rotted in a pit. (Bray 1986: 45-46)
The products of other farm activities, such as silkworm excrement,
bonemeal from slaughtered animals, and pond muck containing decomposed plants and waterfowl droppings, were all systematically collected
for manuring. Old straw thatch and cooking-fire ash entered the midden,
and even carbon-impregnated brick stoves and adobe walls were broken
up and returned to the soil (M. C. Yang 1965 [ 1945]: 240). There was also
a thriving market from as early as medieval times for organic materials
from external sources. Night-soil collectors called regularly on urban
Legend
~Pond fields
li!!IiiiiD Drained fields
[[-j]
House terrace
~Woodlots
~
Swiddens
Caneland
~ Main irrigation
I'·.'; ·' l
50
Meters
Fig. 1.2. Land use and scctlcmcnt in che central portion of Bayninan, an lfugao districc communicy m highland Luzon, Philippines. (From Conklin 1967:
11 5)
Intensive Farm Practices
q
45
p
Cross !>«tion of a
pond field (payo) m
a concave-slope valley
a /o/>orrg w ater
b /iiyo'
worlt
Purchase answer to see full
attachment