MICHIGAN STATE UNIVERSITY
The Farming Systems Research Group at Michigan State University is drawn from
the departments of Agricultural Economics, Agricultural Engineering, Animal
Science, Crop and Soil Science, Food Science and Human Nutrition, Sociology,
Veterinary Medicine, and supported by the International Agriculture Institute of
M.S.U. and the U.S. Agency for International Development through a matching
strengthening grant under the Title XII program.
Farming Systems Research Group
Michigan State University
The Farming Systems Research Group at Michigan State University, supported
by Title XII Strengthening Grant Funds from the U.S. Agency for International
Development, and administered by the Institute of International Agriculture,
has included Dr. Jay Artis, Department of Sociology; Dr. Robert J. Deans,
Department of Animal Science; Dr. Merle Esmay (and Dr. Robert Wilkinson),
Department of Agricultural Engineering; Dr. Eric Crawford, Department of
Agricultural Economics; Dr. Russell Freed, Department of Crop and Soil
Sciences (also representing Horticulture); Dr. Al Pearson, Department of
Food Science and Human Nutrition; Dr. Tjaart Schillhorn van Veen, Department
of Veterinary Medicine; with Dr. George Axinn, International Studies and
Programs and Agricultural Economics, Chair, and Ms. Beverly Fleisher,
graduate research assistant.
ISSUES IN FARMING SYSTEMS RESEARCH-
a multidisciplinary behavioral science perspective
by George H. Axinn
Working Paper No. 8
WORKING PAPER #8
Issues in Farming Systems Research -- A Multidisciplinary Behavioral
by George H. Axinn*
1. What is Farming Systems Research?
A farming system has been defined by the MSU Farming Systems Research
Group as a unit consisting of a human group and the resources it manages in
its environment, involving direct production of plant and/or animal products,
and possibly other products, as well as consumption of those products.
From this perspective, a farming system is a result of interactions
among several interdependent components. Although no two farms are identical,
each farming system may include such components as land, people, crops, and
animals. Like other systems, such a system may be considered, at another
level of analysis, to be merely one component of a larger system. Similarly,
each of the components in such a system may also be considered as systems
themselves, with smaller components within each of them.
Farming systems research can be defined as the application of the
systematic approach to the study of whatever is defined as a farm. The
purposes of such research may vary from the attempt to understand such a
system to a programmatic attempt to make changes within certain aspects of
such a system.
The original or traditional type of farming systems research, is that
which is done by farm families themselves. Those who actually tend the
livestock and till the soil of this world also practice a continuous process
of learning from their environment, solving the problems which face them, and
making appropriate adjustments from what they learn. This type of farming
*Professor of Agricultural Economics and Assistant Dean of International
Studies and Programs, Michigan State University
systems research is not confined to the narrow parameters of any of the
academic disciplines of professional scholars. Instead, it applies the
"science of the farming family" to the "real world problems" of that
family, in a continuous series of interactions. It may be labelled "non-
Without this non-formal type of farming systems research, humanity
would not have survived over the centuries. On the other hand, since the
researchers themselves must be highly generalized in such an effort, it
does not take advantage of the fruits of application of the highly special-
ized "science of the academic" which could be applied to the problems of the
The "non-formal type" farming systems research carried on by each in-
dividual farming family also lacks the ability to access the world's ac-
cumulated knowledge which the highly specialized system of "formal" agri-
cultural research has evolved over the last dozen or so decades.
Both the "non-formal type" and the "formal type," while necessary and
functional for their own purposes, have serious problems if they operate in-
dependently of each other.
From time to time the international professional agricultural research
community has been the resource from which others have attempted to deliver
inappropriate technologies to farming systems in several parts of the world.
This tends to happen when the formal system lacks sufficient information about
the nature of particular farming systems, or when formal agricultural re-
search system purposes are different from those of the farming families. At
the same time, small farming systems, trying to adjust to a rapidly changing
economic, social, political, and technological outside environment have been
placed under increasing stress conditions -- particularly during the last
Growing populations in some areas have pressed the ecosystem to its
productive capacity, becoming a force toward technological change. In-
creased speed and capacity of the world's communication and transportation
systems have influenced national policy affecting agricultural prices, for
example, altering relationships between farming systems and marketing systems.
And the strategies which many national governments have employed, with help
from the international development assistance community, have been designed
to convert subsistence farming systems into market-oriented surplus pro-
ducing farming systems. This pressure to change from relatively self-
contained, small, unspecialized, independent, low-energy farming systems
which consume most of what they produce and produce only what they consume
to relatively market-oriented, larger, specialized, dependent, higher-energy
farming systems which purchase inputs from outside with cash or credit, and
sell outputs to others for cash ... this pressure is perhaps the greatest
stress encountered by millions of small farming systems in today's world.
An opportunity, then, for the MSU farming systems research group, is
to evolve innovative approaches to bridging the gap between the farm-family-
conducted "non-formal" farming systems research and the academic-community-
conducted "formal" agricultural research. If this succeeds, it will weaken
neither and strengthen both. Its goal is to combine the wisdom of the farming
and herding families with the wisdom of the academic scientists, and address
both knowledge building and problem solving activities.
The strategies for accomplishing this goal are evolving among many pro-
fessional agriculturalists who are devising innovative approaches to "farmer's
field" application trials, multidisciplinary research teams, involvement of
farmers themselves in evaluation of new technologies, and participation
with farmers in establishing research goals. (Collisen, 1979; Harwood, 1979;
Hildebrand, 1977; McDowell and Hildebrand, 1980; Norman, 1978; Zandstra,
1979). A major objective is to improve the information flow from what
has been labelled as "indigenous knowledge systems" to those who have
been trained by and are in communication with the international scientific
knowledge systems. (Brokensha, Warren, and Oswald, 1980)
In places where farmers themselves have banded together and pooled
their resources to establish agricultural experiment stations, those em-
ployed as scientists at the experiment stations (or research farms, as they
were often called) had to listen to their farmer clientele. Farmers paid
their salaries, supported their research, and identified the most pressing
problems of local farming systems for research attention. The early ex-
periment stations in Scotland, Germany, and the Northeast part of the U.S.A.
were of this type. (Knoblauch, et. al., 1962; Kuhn, 1955; Rasmussen, 1975;
As the numbers and sizes of agricultural research organizations grew,
they were "forced" to decentralize to stay in tune with the different types
of farming systems which supported their work. Thus "branch stations"
characterize the organizational structure of several national agricultural
However, where agricultural research organizations are parts of large
national government ministries, where the main source of financial support
is in the urban centers of political strength, and where the scientific staff
tends to be persons with urban academic backgrounds, rather than any on-
the-farm work experience, problems of communication with those who till the
soil and tend the livestock have arisen. This is particularly significant
in those systems which have grown out of "colonial" agriculture research
organization focused on export crops.
One approach to overcoming this problem, and increasing the ability
and willingness of those conducting "formal" agricultural research to
learn from farmers, is the organizational approach described in the next
section of this paper.
2. Organization and Administration for Farming Systems Research
From the mixed perspective of a multidisciplinary behavioral science
approach, governments and educational institutions of various types have
attempted to supplement the kind of farming systems research which is con-
ducted within the farming system itself in various ways over the years. To
do this they have set up formal agriculture research systems. There seem
to be two polar "ideal-type" approaches to formal agricultural research
The term "polar ideal type" is taken from sociology. There it re-
fers to a category of human organization which is quite different from an
alternative category of human organization. Thus, "polar ideal types" of
schools might be the "teacher-centered-school" and the "student-centered-
school." The polar ideal type teacher-centered school is absolutely dom-
inated by the teacher. The polar ideal-type student centered school is
absolutely dominated by the student. These are merely categories which can
be used for comparison. In the "real world," there may not be any schools
which actually fit the polar ideal-type. However, all schools may have
some characteristics of one polar type, and some characteristics of the
other polar type. Using such a typology, any two schools can be compared
with each other. (Weber, 1947; Parsons, 1949)
With this approach to the study of agricultural research systems, it
is possible to create and define polar types, and then compare "real" cases
with these "polar ideal types," and with each other.
One type might be labelled the decentralized system, while the other
could be called the centralized system. There is a tendency for the first
to serve the needs of rural people, and to enhance rural life. There is a
tendency for the second to serve the needs of urban people, and to indust-
rialize, commercialize, and depopulate the countryside. Several aspects of
these two "ideal types" are characterized below.
The decentralized agricultural research system tends to be "owned and
operated" by local rural organizations,and control is in the hands of farmers.
It is an extension of what farm families can do within their own family farm
ecosystems; an effort to bring together their knowledge and resources, and
to focus somewhat more specialized effort on problem solving for and on be-
half of their day-to-day interests. The centralized agricultural research
systems, on the other hand, tend to be "owned and operated" by national
governments, with control in the hands of scientists and administrators
in the central capitol cities.
In the decentralized systems, information exchange tends to be part of
the broad function and responsibility of the staff of scientists. Their
task is to learn from farmers, do research appropriate for farmers, and in-
form farmers of what they find. In the centralized systems, scientists can
specialize in the discovery of new knowledge, with little responsibility for
communication with family farm ecosystems.
In the decentralized agricultural research systems, the staff tends to
be recruited from among the local people. In the centralized systems, the
staff tends to be recruited from outside the local situation, because the
base of selection is specialized skills.
Funding for research activity in decentralized systems tend to come from
local sources. In the centralized systems, funding tends to come from central
In the decentralized systems, there is a tendency for research and ex-
tension functions to be merged. In the centralized systems, there tends to
be a specialization of research and extension functions, with centralized
agricultural research systems focusing on their knowledge discovery functions.
There are also differences in methodologies between the two kinds of
systems. A basic research methodology in the decentralized systems tends to
be continuous interaction with rural people in an effort to learn from them.
In the centralized system, methodology tends to be based on replicated field
experiments, with interaction wherever possible with the international agri-
cultural research community (through publications, meetings, correspondence,
The decentralized systems tend to emphasize local applications, and en-
courage communication with farmers. The centralized systems tend to empha-
size basic scientific research, and encourage communication with the inter-
national scientific community.
In the "real world" there are not many cases of a "polar ideal-type" of
either the decentralized or the centralized kind. However, there are many
more agricultural research systems which are closer to the centralized --
looking at the various nation states and international agencies in the world
in 1981 -- than there are decentralized systems. One reason for the renewed
interest in the so-called "farming systems research" in the last few years
is an effort to achieve some of the benefits to farming families of the de-
centralized system while also maintaining the strength of the centralized
3. Research in the social and behavioral sciences relating to farming systems.
There has been a great deal of anthropological, sociological, psychological,
economic, and political research which is tangential. What follows is an attempt
to reflect on multidisciplinary social science scholarship on this subject.
A first observation is that there has not been much serious scholar-
ship relating to the phenomenon discussed above. Neither political scientists
nor anthropologists nor sociologists nor economists have successfully re-
lated the types of agricultural research systems found around the world with
patterns of political power and control and patterns of farming systems
themselves. This seems to be an area of opportunity for future
The climate for such research has improved, both within the disciplines
involved and among those agencies and organizations which support agricul-
tural research. Further, the conceptual and methodological issues can now
be worked out because international collaboration makes comparative ap-
There is a growing body of study relating to the appropriateness or
inappropriateness of technologies; the attempts to transfer technologies
from one social system or culture to another; and some back-tracking re-
lating to the rationale for the development of particular types of tech-
nology. (Dunn, 1978; Eckaus, 1977; Goldschmidt, 1978; Long and Oleson,
1980; Lovins,1977, 1978; Morrison, 1980a, 1980b; Schumacher, 1973).
All of this can contribute to scholarship addressed to comparative
effectiveness of different approaches to the organization, the planning,
the staffing, and the directing of agricultural research organizations.
Second, with respect to the farming systems themselves, it seems that
the social sciences, again, have failed to study them from a systems per-
spective. Rural sociologists have been more concerned with urbanization,
the rural community, and the diffusion of innovations from a central research
source to rural people, than they have with the nature of the farming systems,
perhaps seen as farming family ecosystems. There is a great opportunity
for research in which the conceptual tools of the various social sciences
can be applied to the farming system qua system.
One approach to the "whole systems" study of farms has been demon-
strated by several social scientists who have traced interactions among
the various components of farming systems. Among them, Odum, 1971, 1976;
Thomas, 1974; Rambo, 1979; Axinns, 1978, 1979, 1980; and Cox and Adkins
(1979) have used energy as a proxy for materials flow. They can clearly
demonstrate the effects of the size of the farming system, the extent of
specialization of the farming system, and the eco-system relationships of
the farming system to its relationships with other farming systems, marketing
systems, and the larger political-economic-social-cultural systems of which
they are a part.
Another approach, in which money values serve as a proxy for all types
of flows, has been developed by those who simulate the farming system and
its larger economic system with computer simulations. See, for example,
Rossmiller, 1978; Heady, 1949; Johnson and Rossmiller, 1978; Abkin et. al.
1980; Manetsch, 1971; Crawford, 1980; McRea, 1980; and FAO, 1980.
4. Basic Principles, Concepts, and Generalizations
From the above, it seems apparent that there are some generalizations
about the way in which agricultural research systems are organized and con-
trolled which relate to the type of research they are likely to do, and the
extent to which it will be appropriate for various types of farming systems.
Within the family farm ecosystems themselves, there are social phen-
omena which have not been sufficiently studied, and which offer great op-
portunity. One of these is differentiaion. Farming systems range from the
least differentiated to the most differentiated, with many gradations.
The least differentiated family farm ecosystems are quite unspecialized.
They feature a mixture of many different crops and different classes of
livestock with some "other" production, some work off the farm by family
members, plus a kitchen garden. Women and children do a major portion of
the work on such systems, and in many portions of the world the adult men
of the system exchange labor outside for some of the internal requirements
of the system.
The most differentiated are highly specialized farming systems, which
produce only one crop or type of livestock, and the farm is the major source
of the family income. They tend to be capital intensive, labor efficient,
and highly specialized. (See Axinn & Axinn, 1980).
The least differentiated types of farming systems are also the least
dependent on the outside world. The most differentiated farming systems
are also the most dependent upon the outside world. Thus the variable of
dependence/independence tends tovary directly with the variable differentiated/
Dependency upon the physical world is a characteristic of almost all
types of farming systems, with weather, soil characteristics, and other
geological features, along with the risks and the uncertainties of the
biological processes which farming systems manipulate as constant concerns.
Therefore, there are some trade-offs between the dependence of specialization
and the independence of unspecialization. It is an area which has promising
opportunities for research.
Also, there is a relationship between the quantities of energy trans-
formed and both the differentiation variable and the dependency variable. The
least differentiated and the most independent farming systems also tend to
transform smaller quantities of energy, vis-a-vis what their ecosystems
can sustain over time, than do the most differentiated and most dependent
types of farming systems.
Farming systems also range from almost 100 percent subsistence
farming systems to almost zero percent subsistence farming systems. The
latter type are highly market oriented. Subsistence agriculture can be
understood in terms of the three variables mentioned above: differentiation,
dependency, and energy transformation. Understanding of farming systems
from this perspective can lead to applied and problem solving research which will be
more appropriate to the interests of members of such family farm ecosystems
than the types of farming systems research which ignores these phenomena.
(Axinn & Axinn, 1980).
To generalize again, highly subsistence farming systems tend to con-
sume most of what is produced; tend to supply whatever inputs are needed for
production; have fewer linkages with outside systems; and fewer transactions
on whatever linkages there maybe; and thus are relatively independent. In-
novative technology which requires credit, (or which assumes production for
sale on a market) is not likely to be readily adopted by such family farming
On the other hand, highly market oriented farming systems tend to sell
most of what is produced to others; tend to purchase inputs needed for pro-
duction; tend to have more linkages with outside systems and carry on more
transactions on those linkages; and are thus relatively dependent upon those
outside systems. Because of that dependence upon the market, and susceptibility
to changes in price and price policy, such highly market-oriented farming
systems are also dependent upon innovative agricultural technology -- partic-
ularly technology which will allow them to produce more for the outside market
at lower cost of the inputs they must purchase from the outside supply system.
This contra t between subsistence farming systems and market farming
systems is not made to suggest that one is, normatively, "better" than the
other, or "worse" than the other. Clearly, as the populations of the
world's urban centers continue to increase, there is need for increasing the
surplus (above farming family needs) of food production from some market
farming systems. Also, clearly, there are in today's world hundreds of
thousands of small farming and pastoralist systems which are primarily sub-
sistence farming systems. As with other aspects of life, that which is
desirable from the perspective of some individual farm families may not be
in the best interest of the larger society. The issues are political,
economic, social, and cultural. How the formal agricultural research
systems approach the application of science to the problems of different
types of farming systems is a sensitive matter. The goals and objectives
of agricultural research organizations are likely to be determined by those
who have the political and economic power to influence them. Perhaps in-
novative and sensitive farming systems research can lead to what will be most
5. What is needed from other disciplines?
A great deal is needed by a multidisciplinary social science perspective
on farming systems from the other disciplines.
In the beginning, it is necessary to approximate and understand the
plant component and the animal component in order to know the extent of dif-
ferentiation in either of those, as well as the extent of energy transfor-
mation through them. Thus, if one wished to study relationships among the
variables listed above, and to use those to assess the appropriateness of
particular biological, mechanical, or economic changes for particular farming
systems, it would be desirable to have collaboration among social and
behavioral scientists with economists, animal scientists, agricultural
engineers, agronomists, horticulturists, human nutritionists, and others.
Abkin, Michael H., Tom W. Carroll, Gary R. Ingvaldson, and Manuel
1980 A Guide to Information and Policy Analysis for Agricultural
Decision Making in Latin America and the Caribbean, Staff
Paper No. 80-31, Department of Agricultural Economics,
Michigan State University, East Lansing.
Axinn, George H.
1978 New Strategies for Rural Development. Kathmandu and East
Lansing, Rural Life Associates.
Axinn, George H., and Nancy W. Axinn
1979 Materials Flow and Energy Transformation on Small Farms of
Nepal: A New Approach to Comparative Analysis of Rural
Family Ecosystems. A contributed paper for the Annual Meeting
of the Rural Sociological Society, Burlington, Vermont, Staff
Paper No. 79-23, Department of Agricultural Economics, Michigan
State University, East Lansing.
1980 The Recycling Ratio: An Energy Approach to Planning Rural
Development. Contributed paper for the 5th World Congress
for Rural Sociology, Mexico City. Staff Paper No. 80-25,
Department of Agricultural Economics, Michigan State University,
Brokenshaw, David W., D. M. Warren, and Oswald Werner (editors)
1980 Indigenous Knowledge Systems and Development. University
Press of America.
1979 A Low Cost Approach to Understanding Small Farmers: Identifying
Target Groups, Evaluating Local Circumstances, The Exploratory
Survey, and the Verification Survey. CIMMYT (Mimeo).
Cox, George W. and Michael D. Atkins
1979 Agricultural Ecology: An Analysis of World Food Production
Systems, San Francisco: W. H. Freeman and Co.
Crawford, E. W.
1980 "Understanding, Quantification, and Modelling in Farming
Systems Research: Results of a Simulation Study in Northern
Nigeria," Agricultural Economics Staff Paper No. 80-19. East
Lansing, Michigan: Michigan State University Department of
Dunn, P. D.
1978 Appropriate Technology: Technology With a Human Face.
New York: Schocken Books.
Eckaus, Richard S.
1977 Appropriate Technologies for Developing Countries.
Washington, D.C.: National Academy of Sciences.
1978 As You Sow: Three Studies in the Social Consequences of
Agribusiness. Montclair: Allanheld, Osmun & Co., Publishers.
Griswold, A. Whitney
1963 Farming and Democracy. Yale University Press, New Haven and
Harwood, R. R.
1979 Small Farm Development: Understanding and Improving Farming
Systems in the Humid Tropics. Boulder, Colorado: Westview
Heady, Earl O.
1949 "Basic Economics and Welfare Aspects of Farm Technological
Change," Journal of Farm Economics. 31:293-316.
Hildebrand, P. E.
1977 "Generating Small Farm Technology: An Integrated Multi-
disciplinary System," Paper presented at the 12th West
Indian Agricultural Economics Conference of the Caribbean
Agro-Economic Society held in Antigua, 24030 April, 1977.
Johnson, Glenn L. and George E. Rossmiller
1978 "Improving Agricultural Decision Making: A Conceptual
Framework," in Rossmiller (1978).
Knoblauch, H. C., Law, E. M., Meyer, W. P., Beacher, B. F., Nestler, R. B.
and White, Jr. B. S.
1962 State Agricultural Experiment Stations: A History of Research
Policy and Procedure, United States Department of Agriculture.
1955 The First Hundred Years-Michigan State, The MSU Press.
Long, Franklin and Alexandra Oleson (editors)
1980 Appropriate Technology and Social Values A Critical Appraisal.
Cambridge, Massachusetts: Ballinger Publishing Company.
1977 Soft Energy Paths. New York: Harper.
1978 "Soft Energy Technologies" in Jack M. Hollander, Melvin
K. Simmons and David 0. Wood, editors. Annual Review of
Energy. Volume 3. Palo Alto, California: Annual Reviews Inc.
Manetsch, Thomas J., et al.
1971 "A Generalized System Simulation Approach to Agricultural
Sector Analysis with Special Reference to Nigeria."
Department of Agricultural Economics, Michigan State University,
McDowell, R. E. and P. E. Hildebrand
1980 Integrated Crop and Animal Production: Making the Most of
Resources Available to Small Farms in Developing Countries.
New York: The Rockefeller Foundation Working Paper Series.
McRae, Stephen D.
1980 Human Ecological Modeling in the Central Andes. Department
of Resource Development, Michigan State University, East
Morrison, Denton E.
1980a "Should We Follow the Soft or Hard Energy Path: Some Social
Science Considerations," pp. 169-191 in Joel Kameron editor.
The Social Dimensions of Energy Options. Conference Proceedings.
Ramapo, New Jersey: Ramapo Center for Future Studies.
1980b "The Soft, Cutting Edge of Environmentalism: Why and How
the AT Notion is Changing the Movement," Natural Resources
Journal, Special Symposium on "Whither Environmentalism?"
Robert C. Mitchell (ed.), 20, 2 (April): 275-298.
Norman, David W.
1978 "Farming Systems Research to Improve the Livelihood of Small
Farmers." American Journal of Agricultural Economics.
Odum, Howard T.
1971 Environment, Power, and Society. New York: Wiley-Inter-
Odum, Howard T. and Elizabeth C. Odum
1976 Energy Basis for Man and Nature. New York: McGraw-Hill
1949 The Structure of Social Action. Glencoe, Illinois:
Rambo, A. Terry
1979 Development of a Conceptual Framework for Human Ecology.
Working Paper No. 4, Kuala Lumpur, Malaysia: University
Rasmussen, Wayne D. et al
1975 Agriculture in the United States: A Documentary History,
Vol. 2, Random House, New York.
Rossmiller, George E.
1978 Agricultural Sector Planning: A General Systems Simulation
Approach. Department of Agricultural Economics, Michigan
State University, East Lansing.
Schumacher, E. F.
1973 Small is Beautiful: Economics as if People Mattered.
New York: Perennial Library.
Thomas, R. Brooke
1974 "Human Adaptation to Energy Flow in the High Andes: Some
Conceptual and Methodological Considerations." Energy Flow
in Human Communities, Proceedings of a Workshop in New York,
N.Y., University Park, PA: U.S. International Biological
Program and Social Science Research Council.
1947 The Theory of Social and Economic Organization, translated by
A.M. Henderson and Talcott Parsons: Oxford University Press.
Zandstra, H. G.
1979 "A Cropping Systems Research Methodology for Agricultural
Development Projects," Paper presented at a Seminar at the
World Bank headquarters in Washington, D.C., August 1, 1979.