FSR Background Papers

THE FARMING SYSTEMS APPROACH TO RESEARCH

by David W. Norman

FARMING SYSTEMS RESEARCH AND THE LAND-GRANTSYSTEM: TRANSFERRING ASSUMPTIONS OVERSEAS

FSR BACKGROUND PAPERS

by Cornelia Butler Flora

Paper No. 3 October, 1982

THE FARMING SYSTEMS APPROACH TO RESEARCH

by David W. Norman

FARMING SYSTEMS RESEARCH AND THE LAND-GRANTSYSTEM: TRANSFERRING ASSUMPTIONS OVERSEAS

FSR BACKGROUND PAPERS

by Cornelia Butler Flora

Paper No. 3 October, 1982

FARMING SYSTEMS RESEARCH PAPER SERIES

Kansas State University’s Farming Systems Research (FSR) Paper Series issupported by the U.S. Agency for International Development Title XIIStrengthening Grant. The goal of the Strengthening Grant is to increase theUniversity’s capacity for, and focus its commitment to, implementing Title XIIagricultural and nutritional development assistance programs in less-developedcountries. This series is maintained by the FSR Program Associates— a multi-discipl inary team of professors who are focusing their act ivi t ies aroundapplied research on farming from a systems perspective.

Those faculty involved in the FSR Group (past and present) includeUlysses Acasio (Grain Science and Industry), Charles Bussing and Duane Nellis(Geography), James Converse, Cornelia Butler Flora, Barry Michie, andMartin Ottenheimer (Sociology, Anthropology, and Social Work), J. Arthur Hobbsand L. Van Withee (Agronomy), Berl Koch and John Wheat (Animal Sciences andIndustry), George Larson (Agricultural Engineering), David Norman (Economicsand Agricultural Economics), Meredith Smith (Foods and Nutrition), andDwight Wiebe (Adult and Occupational Education).

The purpose of the FSR Paper Series is to provide an avenue for dissem-inating information on FSR to those interested. Publications to be includedfall into the following categories: updated bibliographies from KSU’s FSR database; proceedings from KSU’s annual Farming Systems Symposium; selected paperspresented in KSU’s FSR Seminar Series; selected papers prepared by KSU’sProgram Associates.

Copies of these papers are available from the Farming Systems Research and Extension Collection, KSU Libraries.

Vernon C. LarsonDirector

Wendy J. SheppardEd i t o rInternational Agricultural Programs

NOTICE OF NON-DISCRIMINATION

Kansas State University is committed to a policy of non-discrimination on thebasis of race, sex, nat ional or igin, handicap, or other non-merit reasons, inadmissions, educational programs or activities, and employment, all as requiredby applicable laws and regulations. Inquiries may be addressed to: Director,Affirmative Action Office, 214 Anderson Hall, Kansas State University,Manhattan, KS 66506 (913-532-6220), or Regional Director, Office of CivilRights, Department of Education, 1150 Grand Avenue, Kansas City, MO 64106(816-376-2356).

THE FARMING SYSTEMS APPROACH TO RESEARCH

by David W. Norman*

1. Topics

(a) Introduction, object ives, and layout (sect ion 2)(b) Farming system determinants (section 3)(c) Defining the farming systems approach to research (FSAR) (section 4)(d) Attributes of the FSAR (section 5)(e) Methodological problems of the FSAR (section 6)( f ) Functions (section 7)(g) Components including funding (section 8)(h) Boundaries (section 9)

2. Introduction, objectives, and layout

2.1 The farming systems approach to research is a product of the 1970s.It developed due to frustrat ion over part ial or complete fai lure ofother approaches in developing technology relevant for farmingfamil ies located in relat ively unfavorable environments.

2.2 There is considerable confusion over how a farming system isdefined and what the farming systems approach to research (FSAR) is.It is important to have a concensus on what these are in order toaddress the central objectives of this paper. At the moment, theterm FSAR is used too loosely.

2.3 The FSAR is still evolving and, therefore, conventional wisdom forsolving methodological and implementation problems has still notdeveloped. However, some general guidelines are emerging, but theseare l ikely to be modif ied in the l ight of further experience.

2.4 After defining what a farming system is and describing what constitutesthe FSAR and its attributes, the foundation is laid for a considerationof the various other topics assigned to this paper: methodologicalproblems, functions, components and funding, and boundaries. Thesetopics, however, are not mutually exclusive, therefore, there is someoverlap in the discussion.

3. Farming system determinants

3.1 In developing countries, there is considerable overlap between theunit of production and the unit of consumption. Thus, the means oflivelihood and household are intimately linked and cannot beseparated (Figure 1).

3.2 A farming system adopted by a given farming household results from itsmembers, with their managerial know-how, allocating the three factorsof production, i .e., land, labor, and capital , to which they haveaccess, to three processes (crops, livestock, and off-farm enterprises)in a manner which, within the knowledge they possess, will maximize theattainment of the goal(s) for which they are str iving.

*Prepared for the Farming Systems Research Symposium "Farming Systems in theField," Kansas State University, Manhattan, Kansas, November 21-23, 1982.

- 2 -

3.3

3.4

3.5

3.6

The farming system is determined by the environment in which thefarming family operates. The "total" environment in which it operatescan be divided into the technical (natural) and human elements.

The technical element reflects what the potential farming system canbe and, therefore, provides the necessary condition for its presence.The technical element can be divided into:

(a) Physical factors— water, soil, solar radiation, temperatures, etc.(b) Biological factors— crop and animal physiology, disease, insect

attack, etc.

Technical scientists have been able to modify the technical elementto some extent.

The human element has often been neglected in traditional researchapproaches to development of improved technologies, which accountsfor the i r o f ten be ing re jec ted or , a t best , be ing d i f fe rent ia l lyadopted, thereby result ing in an inequitable distr ibution of benefi ts.The human element, providing the sufficient condition for the presenceof a farming system, determines what the actual farming system willbe— being a subset of the potential defined by the technical element.

The human element can be divided into two components or groups offactors. The exogenous factors— the social milieu in which thefarming household operates— are largely out of the control of theindividual farming household, but will influence what its members areable to do. They can be divided into three broad groups:

(a) Community structures, norms, and beliefs.(b) External institutions or support systems. This is often provided

by government, both on the input (extension, input distribution)and product (direct and indirect intervention) sides.

(c) Miscellaneous influences— location, population density, etc.

On the other hand, endogenous factors— land, labor, and capital,along with management— which are under the control of the individualfarming household, can be used by them to derive a farming systemconsistent with their goal(s) subject to the boundary condit ions l a i ddown by the technical element and exogenous factors. The endogenousfactors can, under certain circumstances, be complemented andsupplemented in quantitative and qualitative terms through theinfluence of exogenous factors— such as capital through a creditprogram, management via extension, etc.

4. Defining the farming systems approach to research (FSAR)

4.1 The primary aim of the FSAR is to increase the overall productivityof the farming system— therefore, hopefully the welfare ofindividual farming famil ies— in the context of the entire range ofprivate and societal goals— given the constraints and potentialsimposed by the determinants of the existing farming systems.

4.2 Increased productivity is achieved through two types of developmentals t ra teg ies :

(a) Farming systems research (FSR)— involving the development anddissemination of relevant improved practices (technologies).

(b) Farming systems perspective (FSP)— involving influencing thedevelopment of relevant policies and support systems (externalinstitutions).

4.3 Both developmental strategies have a "micro to macro" or "bottom-up"orientation compared with the more "top-down" or "macro to micro"orientat ion of research work that starts at the experiment stat ionor in the upper echelons of planning ministries.

4.4 Given the right institutional setting and linkages, both FSR andFSP are possible. However, because FSR programs have usuallybeen located in agricultural research inst i tutes— primarily croporiented— often with poor linkages to planning or policy-makingagencies, the FSP has usually not been operative. Thus, the supportsystems have been considered parameters (implying a submissiveapproach to them on the part of the FSR team) rather than variablesamenable to manipulation (implying an interventionist approach on thepart of the FSR team). An interventionist approach permits a widerrange of possible improved technologies to be considered in theresearch process.

4.5 The term FSR has often been used very loosely. There are programscalled FSR that are not FSR, and there are programs not called FSR thatare indeed FSR. The following characterizes FSR:

(a) The farm, as a whole, is viewed in a comprehensive manner.(b) The choice of priorities for research reflects the initial

study of the whole farm.(c) Research on a farm sub-system is legitimate FSR, provided the

connections with other sub-systems are recognized and takeninto account.

(d) Evaluation of research results explicitly takes into accountlinkages between sub-systems.

Using the above characterization, as long as the concept of the wholefarm and its environment are preserved, not all factors determiningthe farming system need to be considered as variables— some may betreated as parameters. Therefore, FSR may be called FSR "in thesmall" ( low rat io of variables to parameters) or FSR "in the large"(high ratio of variables to parameters). Incorporation of FSP intoFSR increases the ratio of variables to parameters. However,methodological and implementation issues become more complex as therat io of variables to parameters increases. In addit ion to themethodological issue, the scope of the FSR program will be partiallydetermined by the mandate of the institution in which it is located,the effect iveness of l inkages with other inst i tut ions and agencies,resources available, i.e., time, skill, finances, etc.

-3 -

- 4 -

4.6 As well as FSR programs being differentiated on the basis of the ratioof variables to parameters, they can also be classified as follows:

(a) "Upstream" types of FSR programs have a developmental orientationand usually do not provide results for immediate adoption byfarming families. Perhaps more aptly called resource managementresearch, "upstream" FSR programs involve using a systems approachto provide prototype solutions on experiment stations to majorconstraints to crop or agricultural improvement, e.g., watershedmanagement, intercropping, etc. Along, with results, from commodityresearch programs— reductionist research— they contribute to thebody of knowledge (Figure 2) and are available for feeding intothe "‘downstream" FSR programs.

(b) "Downstream" types of FSR programs, which are the main concern ofthis paper, have an applied orientation and aim at developing andintroducing strategies that wi l l improve the productivi ty offarming systems for target groups of farming families now and inthe short-run. This requires select ively drawing upon avai lableinformation, i.e., body of knowledge in Figure 2, in the processof designing, practices or recommendations for a particular farmingsystem on the basis of an analysis of the constraints of thatsystem. Therefore, recommendations are produced which are suitedto a spec i f ic loca l s i tuat ion. Th is invo lves work ing d i rec t lywith farmers, i .e., on-farm research, and, as a result , reducingto a minimum work on the experiment station.

4.7 There are four stages in applied or "downstream" FSR (Figure 2):

(a) The descriptive or diagnostic stage in which the actual farmingsystem is examined in the context of the "total" environment— toidenti fy constraints farmers face and to ascertain the potentialf lexibi l i ty in the farming system in terms of t iming, slackresources, etc. An effort is also made to understand goals andmotivat ion of farmers that may affect their efforts to improvethe farming system.

(b) The design stage in which a range of strategies is identifiedthat is thought to be relevant in deal ing with the constraintsde l ineated in the descr ip t ive or d iagnost ic s tage. Heavyrel iance at this stage is placed on obtaining information fromthe "body of knowledge."

(c) The testing stage in which a few promising strategies arisingfrom the design stage are examined and evaluated under farmcond i t i ons t o asce r ta i n t he i r su i t ab i l i t y f o r p roduc ingdesirable and acceptable changes in the existing farming system.Th is s tage cons is ts o f two par ts : in i t ia l t r ia ls a t the farmlevel with joint researcher and farmer part icipat ion (researchermanaged), then farmer’s testing with total control by farmersthemse l ves ( f a rmer managed ) .

(d) The extension stage in which the strategies that were identifiedand screened during the design and testing stages are implemented.

In practice, there are no clear boundaries between the various stages.Design activities, for example, may begin before the descriptive anddiagnostic stages end and may continue into the testing stage, as

-5 -

promising alternatives emerge during the trials at the farm level— where farmers and researchers interact direct ly. Similar ly, test ingby farmers may mark the beginning of extension activities.

5. Attributes of the FSAR

5.1 The object ives of the farmer (farming family) are direct ly incorporatedinto the research process. The farmer is the central unit in theresearch process, being direct ly involved in the descript ion, test ingand extension stages. Involvement of farmers gives them a "voice"in the research process and ensures the use of evaluation criteriare levant to them. For the farming fami ly , eva luat ion cr i te r ia forthe adoption of improved practices can be divided into the followinggroups, although it should be emphasized they are not mutuallyexc lus ive:

(a) Necessary conditions determine whether the farmer would be ableto adopt the improved practices. Such conditions would includetechnical feasibi l i ty, social acceptabi l i ty, and compatibi l i tywith external inst i tut ions— that is support systems.

(b) Sufficient conditions determine whether the farmer wouldbe wil l ing to adopt the improved pract ices. Obviously thenecessary condit ions wil l be inf luential in determining thiswi l l ingness. Suf f ic ient cond i t ions wi l l inc lude compat ib i l i t yof the improved pract ices with the goal(s)— self-suff iciency,prof i t maximizat ion, etc.— of the farming family and of thefarming system they currently practice.

5.2 Efforts are made to incorporate community and societal needs intothe FSR process by trying to ensure a convergence between private(usually short-run) and societal (usually longer run) interests.Examples of possible conflicts would be where satisfying short-runneeds of individual farming famil ies would result in long-runsocietal costs in terms of degradation of the natural resource base,increased inequal i t ies in welfare distr ibut ion, etc. I t is necessaryto develop improved strategies that wi l l avoid such confl icts.

5.3 The FSR approach, by including farmers, taps the pool of knowledge inthe society and enables research and hence developmental strategiesto build upon the good points of the present farming systems, whileat the same time minimizing the time spent in "rediscovering thewheel"— for example, the value of intercropping.

5.4 FSR recognizes the locational specificity of the technical and human(exogenous and endogenous factors) elements. This requiresdisaggregating farming families into homogenous subgroups (recom-mendation domains) and developing strategies appropriate to each.Farming famil ies in a part icular subgroup wil l tend to have similarfarming act iv i t ies and to include similar social customs, simi laraccess to support systems, comparable marketing opportunities, andsimilar present technology and resource endowment.

5.5 The whole farm perspective of FSR compels the adoption of an integrativefunction which increases the potential for exploiting complementary and

-6 -

supplementary relationships between resources and enterprises, andthe derivation of solutions compatible with the needs and capacitiesof farming families. The farming systems farmers practicedtraditionally recognized such relationships— for example crops andlivestock, staggered planting dates, etc. To ensure that theintegrative and beneficial relationships are adequately consideredand exploited requires a multidisciplinary team— both technicaland social scientists— working together at all four stages of theresearch progress.

5.6 The process of FSR is reorganized as being dynamic and iterativewith linkages in both directions between farmers, research workers,and funding agencies, rather than simply the presence of forwardlinkages characteristic of the "top-down" approach. The iterativecharacteristic can improve the efficiency of the research processthrough providing a means to fine-tune improved technologies to aspecific locale.

5.7 Finally, FSR complements, and does not compete with, other researchapproaches. The complementarity is illustrated through FSRcontributing in two ways:

(a) Fine-tuning through adaptive testing at the farm level thosetechnologies developed on experiment stations through "upstream"FSR and the more conventional reductionist approach (bothcontribute to the "body of knowledge"). Successful testinggives rise to successful dissemination (all other things beingequal), resulting in improvement of farming families’ welfare.

(b) Failure under adaptive testing at the farm level results incloser specification of requirements for improved technologydevelopment that can be fed back to experiment station basedresearch programs outside the FSR program itself. Hopefully,this will contribute to the development of improved technologiesthat wi l l improve the welfare of farming famil ies in the future.

6 . Methodological problems of the FSAR

6.1 Due to the fact that the methodology for undertaking "downstream" FSRis st i l l going through a period of evolut ion, a large variety ofmethodological issues require resolut ion. Not surprisingly perhaps,there are often considerable dif ferences in opinion as to how severethey are and how they should be dealt with. Some of the most frequentlymentioned methodological issues follow.

6 . 2 How holistic should FSR be? As mentioned earlier, the methodologicalproblems increase as the FSR program becomes more holistic, i.e., theratio of variables to parameters becomes higher. Also stressedearl ier was the fact that the present state of the arts of undertakingFSR means that most current work is on the crop process and is largelyconfined to development of improved technologies. Pract ical problemsalso restricting the scope of "downstream" FSR are the mandates ofins t i tu t ions in which they are located, i .e . , usua l ly techn ica l c ropresearch institutes, poor or weak linkages with other researchinstitutions, and policy-making and farmer contact agencies. Related to

- 7 -

the question of how holistic "downstream" FSR should be is the issueof whether the pol icy-inst i tut ional environments should be treatedas parameters or variables. Increasingly in FSP, It is being suggestedthat these might be treated as variables subject to manipulation, assuggested earl ier. This micro-macro l ink is important in maintainingthe viability of "downstream" FSR in the long-run through the addeddimension it gives to creating conditions conducive to improving theproductivity of farming systems and therefore hopefully the welfareof farming famil ies.

6.3 What needs or constraints are to receive focus in the research process?Should they be those art iculated by farming famil ies, i .e., fel t needs,those scientifically ascertained by research workers, or thoseref lect ing the needs of society? As discussed earl ier, cr i ter ia usedin developing improved strategies should reflect the felt needs offarming families, providing they are not incompatible with the needs ofsociety, e.g., there is not a decline in soil fertility, nutritionalleve ls , increas ing ly inequ i tab le income d is t r ibut ion, e tc . ) .Strategies developed need to ensure convergence between short-runprivate interests and those of the society in the long-run. Althoughthere is, in principle, agreement with the above, there is oftendisagreement as to how societal interests can be incorporatedpractically into "downstream" FSR. The problem of doing this relatesto the methodological complexity of their incorporation and the timethat would be required in deriving societal impact evaluations.

6.4 The needs or constraints that are identified may be technical, economicor socio-cultural in nature. What approach should be used in dealingwith them? The two approaches generally used are:

(a) Accepting the constraint and developing strategies that exploitthe f lexibi l i ty that exists in the current farming system whileat the same time not further exacerbating the constraint.Socio-cultural constraints should not generally be broken.

(b) Developing strategies that will overcome the constraint. Thedecision as to which approach to use usually depends on thecons t ra i n t seve r i t y , f l ex i b i l i t y t ha t ex i s t s i n t he cu r ren tfarming system, avai labi l i ty of potential improved strategiese i ther to break the const ra in t or to exp lo i t the f lex ib i l i ty ,compat ib i l i t y w i th soc ie ta l goa ls , e tc .

6.5 Is it necessary for "downstream" FSR to be expensive? It is viewedby some to be expensive because of its locational specificity and,therefore, the need to focus on limited numbers of farmers. Theexpensive nature is emphasized because of the opportunity costs ofneglecting other farmers. Thus, the quest for minimizing costsin the research process is a major issue. Considerable controversyexists concerning the degree to which costs can and should be reduced,and the ways in which they should be reduced. In general threeapproaches are being used to try to minimize costs:

(a) Seeking ways to reduce time and resources required for movingthrough the four research stages— methods used should be basedon the degree of understanding that is necessary. Can this be

-8-

done with base data analysis plus an informal exploratory(sondeo) survey and a one-shot formal survey? Or is a detailedtwice weekly formal survey required for a period of one year?Can modeling techniques help improve understanding— or doesthis come at too high a cost? In the testing stage, shouldfarmers be selected that are the better farmers, most cooperativefarmers, or simply representative farmers? Representativefarmers may not, for example, be so cooperative, thereby reducingthe efficiency and effectiveness of dialogue and the timelyconclusion of the test ing stage. Considerable controversy st i l lexists concerning the way in which these and other questionsshould be resolved in the interests of minimizing costs and time.

(b) Finding ways to maximize the return from the location specificnature of "downstream" FSR by determining the transferability ofthe resu l ts to o ther s imi lar " to ta l " env i ronments . In t roduc ingsome f lexibi l i ty into the improved pract ices increases thepotential of transferability, but this may come at some cost interms o f the potent ia l leve l o f re turn . Is th is , o r is th is not ,desirable? Controversy exists with respect to this.

( c ) Seeking best of readi ly avai lable solut ions— that is "better, butnot necessari ly best" or "non-perfectabi l i tar ian." How muchfine-tuning should there be, thereby extending the testing stage?

6.6 In terms of developing improved practices (technologies), shouldemphasis be placed on single trait innovations, which may preclude theexploitation of possible complementary or synergistic effects betweenthe various components in packages of improved practices? In theorythe former would be desirable, but in practice the latter are muchmore common. A possible compromise is to design and develop packagesof improved practices that permit, in an explicit manner, a stepwiseapproach to the adoption of the various components of the package.

7 . Functions

7.1 The location and linkages of the FSAR program will determine whether:

(a) FSR, or helping to generate relevant improved technology, is theprimary focus.

(b) FSP, or helping to develop relevant agricultural pol icies/supportsystems, is the primary focus.

(c) If linkages are good between research and planning/implementationinstitutions, some mix between the two may be possible, i.e., FSRplus FSP = FSAR.

At the moment, the major thrust is on developing relevantimproved technologies for the crop sub-system. This is, in part,because most FSR-type programs are located in agricultural researchinstitutes whose mandates revolve a round crops.

7.2 FSAR-type programs help generate and communicate information in thefollowing ways:

(a) Through providing a means for farmers to communicate their needsto researchers (FSR) and planning/implementation agencies (FSP)

-9 -

— which has often been lacking in the more conventional "top-down"approaches.

(b) Through bringing about linkages between farmers, extensionpersonnel, and researchers (FSR), and farmers, extension personnel,and planning/implementation inst i tut ions (FSP).

7.3 FSAR-type programs provide a "practical component" in the on-the-jobt ra in ing o f research, ex tens ion, and p lann ing personnel . In fac t ,much of the skill obtained to date in FSAR-type activities has beenderived from longevity in the field rather than through formaltraining programs. Although formal degree training is not avai lablein this general area, short courses in the interdiscipl inarycharacteristics of FSAR-type activities are being developed at anumber of international (CIMMYT, IRRI), regional (CATIE), andnat iona l ( ICTA) ins t i tu tes .

7.4 In these various funct ions, FSAR-type act ivi t ies perform a faci l i tat ingor in tegrat ing ro le ra ther than in i t ia t ing ro le . Therefore , thesetypes of activities should not come right at the beginning of thedevelopment process unless funding agencies are willing to accept along gesta t ion per iod in terms of ach iev ing resu l ts . For example,FSR programs are likely to have low immediate returns if the "bodyof knowledge" (Figure 2) is poorly developed. FSR activities are nota substitute for experiment station based commodity research. Suchnational research programs can benefit from linkages with CRSPs andIARCS (provided understandable inhibitions are overcome)— which canprovide expertise in deficient areas and provide critical masses ofstaff and resources for looking at complex problems ("upstream" ordevelopmental FSR).

8. Components

8.1 Because of the locational specif ic i ty of FSAR-type act ivi t ies, thefuture— particularly of "downstream" or applied activities— must liewithin national programs.

8.2 Problems within national programs that make introduction of FSRactivities difficult include:

(a) Staff constraints:-relat ive immobil i ty of staff within national sett ings sometimesdiscourages changes and encourages maintenance of the status quo,therefore resulting in opposition to "new" FSR programs.

- interdiscipl inary FSR activi t ies require the interact ion of bothtechnical and social science discipl ines— the latter are usual lylack ing in techn ica l agr icu l tura l research ins t i tu tes in Af r ica .

(b) Organizational/operational constraints:-national research programs used to be organized along disciplinelines which have more recently given way to programs organizedlong commodity l ines.

-however, FSR programs involve crossing both discipline andcommodity lines.

(c) Research resource constraints:-financial and manpower constraints are both common in nationalsettings.

-10-

( d ) Locational specificity of applied ("downstream") FSR programsconstraint. This constraint and the fact that FSR programs workwith limited numbers of farming families (hopefully representativeof much larger numbers) add to the perception of some, of theexpensive nature of such research. In arriving at such conclusions,sunk costs involved in developing experiment stations, and lowreturns from other past research endeavors, are likely to beheavily discounted or even ignored. Because of the complementarityof "downstream" FSR activities and other research approaches, thereappears to be little value in comparing the benefit-cost ratios ofdifferent research approaches. However, a challenge does exist infinding ways to minimize the cost of "downstream" FSR activities,a topic that was discussed earl ier (see section 6.5).

8.3 Implications for instituting the components of an FSR program in anational sett ing are, therefore, as fol lows:

(a ) Be minimally disruptive in inst i tut ing an FSR activi ty withinnat iona l programs. I t i s not necessary , for example, to ca l l i tFSR i f th is is po l i t i ca l ly unpala tab le , whi le , in order toencourage intrainstitutional linkages and cooperation, it isprobably not always desirable— init ial ly at least— to have aseparate un i t fo r FSR. Improved poss ib i l i t ies forsupport and commitment are likely to arise if it is grafted ontoan exist ing administrat ive unit, e.g., agronomy.

(b) Staff and funding limitations usually require initial "pumppriming" through:

-support from donor agencies. This will mainly have to come frompublic rather than private sources. Funds and personnel fromthe la t ter source are not l i ke ly to be for thcoming forinst i tut ional izing an FSR-type program where the gestat ionper iod is l i ke ly to be lengthy in terms of get t ing resu l ts .

-support through developing l inks with relevant IARCs that aredeveloping expertise in "downstream" FSR.

-recourse to advice and help from a central core group of expertiseon FSR activi t ies (such as AID is considering providing funds forat present) and Tit le XII Strengthening Grants in the U.S.universit ies which are focusing on FSAR-type activi t ies.

(c ) Although strong arguments can be made for locating FSR teamswithin current inst i tut ions, attent ion needs to be paid todeve lop ing in ter ins t i tu t iona l l inkages, thus improv ing theposs ib i l i t ies for e f fec t ive FSR act iv i t ies . Chances for sucheffectiveness can be increased through, for example:

-obtaining some manpower and financial commitments on the partof extension and planning/implementat ion agencies. Theinvolvement of the former in the research program canpotential ly help mend the r i f t that often exists between researchand extension, whi le the latter provides the opportunity for theinclusion of an FSP and the possible source of agri-cultural economists, often lacking in technical researchinstitutes.

-arranging l inkages, perhaps essential ly of an informal nature,wi th ins t i tu t ions/organ izat ions wi th exper t ise in "downst ream"FSR. Since "downstream" FSR is still evolving, there are many

-11-

methodological issues that st i l l remain to be sat isfactor i lyresolved.

(d) If an FSP is to receive priority in FSAR-type programs— apriority that is currently rare— then the logical institutionallocation of such a program is within a development project,extension program, or planning unit.

9. Boundaries

9.1 As has been emphasized earl ier, the major act ivi t ies of "downstream"FSR programs in the future must be within national programs,although linkages with external agencies and expertise will beimportant in improving the effectiveness of national programs,particularly in solving methodological problems, transferring resultsacross national boundaries in areas with similar "total" (technicaland human) environments, etc.

9.2 Experiment station based research programs are, currently, usuallyorganized along commodity lines and have a national focus.

9.3 Two possible ways exist for organizing "downstream" FSR-typea c t i v i t i e s :

(a) "FSR in the small" emphasizing a couple of products throughoutthe country.

(b) "FSR in the large" where all products (processes) are consideredas in a regionally focused program. (Such a program could stillbe "FSR in the small" where only the major products are focusedon).

Approach (b) is currently being emphasized in preference to approach (a),with regions being either defined ecological ly (more relevant for anFSR program), or administratively (more important where an FSP emphasisis expected), or some compromise between the two.

9.4 In setting up boundaries for FSAR-type programs, it isessential that effective linkages are maintained between:

(a) The various regionally focused FSAR programs.The regionally focused FSAR programs and the experiment(b)station based research programs, extension, implementing andplanning institutions.

This can be facilitated through meetings, visits of staff to otherprograms, etc.

10. Conclusion

FSAR-type activities are not a panacea. They complement and helpintegrate and improve the pay-off of other act ivi t ies through providinga "bottom-up" approach. The decision whether to introduce FSAR activitiesis not an either-or decision, but rather one that can help improve theperformance of other on-going activities.

F I G U R E 1Schematic Representation of Some Farming System Determinants

Elements

Factors

Inputs

Processes

Farming System

Broken lines represent results of farming system

- 1 2 -

- 1 3 -

FIGURE 2Schematic Framework for Farming Systems Research

Farming System Research Stages

1. Description orExternal Institutions

diagnosis of present Current Farming Systemfarming system (Hypothesis Formulation)

2. Design of improvedsystems Experiment Station

TrialsBody ofKnowledge

3. Testing of Trials at Farm Levelimprovedsystems

Farmers' Testing

4. Extension of improvedfarm system

ModifiedFarming System

by Cornelia Butler Flora*

American agriculture owes much of its development and strength to theland-grant system. This system, based on the Jeffersonian ideal of participatorydemocracy, allows for members of the farm households who produce the bulk of ourfood and fiber to relate directly to researchers and extension agents to maketheir needs known. Farmers’ expressed needs then become the basis for developingtechnology and gett ing i t out to the farm, in order to increase productivi ty andra ise the leve l o f l i v ing o f fa rm fami l ies .

That model, so successful in the United States, has been built on a numberof conditions peculiar to the United States that we do not make explicit. Theseconditions, which we take for granted, are both necessary and sufficient for oursystem of research and extension to function well.

FARMING SYSTEMS RESEARCH AND THE LAND-GRANT SYSTEM:

TRANSFERRING ASSUMPTIONS OVERSEAS

Agricultural development has often by-passed the small, marginal farmer.Even major technological breakthroughs in productivity, such as the "miracle"var ie t ies o f wheat and r ice , tended to accrue benef i ts to la rge farmers(Pearse, 1980). Food imports are increasing in many middle-income countries,sapping them of much needed foreign exchange, while in low-income countries,food imports have not decreased and food aid remains a substantial portion oftheir food imports (World Bank, 1981: 102-103). Can the problems of small farmagriculture, which produces the roots, tubers, and grains (wage foods) that arethe basic diet of most people in the world, be met by transferring U.S.institutional arrangements (even more than U.S. technology) to developingcountry settings?

When we begin to try to work with national governments to improve overseasagriculture through research and extension— particularly research and extensionaimed at the disadvantaged small, marginal farmer we— have to go back and look atthe conditions of agricultural development in the United States. We must betterunderstand what went on in the United States in order to know what we need to doto help in developing a system of research and extension relating directly topeople’s needs.

Conditions Contributing to U.S. Success

In the United States, the mandate for research and determination of whowould control that research came through a political process. The laws settingup the land-grant system and structuring the Department of Agriculture came, inpart, from the pressure of the farmers themselves, who understood the need forimproving farming processes in order to better their own lives (Rothstein, 1978).The system was not imposed by colonial powers nor modeled after that already inexistence in a more developed country. An organized and vocal constituency forresearch and extension developed— the first condition contributing to thesuccessful development of U.S. agriculturally-oriented institutions.

* P r o f e s s o r o f S o c i o l o g y a n d P r o g r a m A s s o c i a t e , I n t e r n a t i o n a l A g r i c u l t u r e ,K a n s a s S t a t e U n i v e r s i t y , M a n h a t t a n , K a n s a s 6 6 5 0 6 . T h i s p a p e r i s a d a p t e d f r o mo n e p r e s e n t e d a t t h e 1 8 t h A n n u a l C o n f e r e n c e o f t h e A s s o c i a t i o n o f U . S .U n i v e r s i t y D i r e c t o r s o f I n t e r n a t i o n a l A g r i c u l t u r a l P r o g r a m s , L i n c o l n , N e b r a s k a ,J u n e 8 - 1 0 1 9 8 2 .

-2 -

Further, that farmer constituency was organized both in separate pressuregroups (such as the Grange and the Farm Bureau) and in political parties (theDemocrats in the South and the Republicans in the Midwest). Often, white, malefarmers were elected to the legislature and to Congress in order to articulatetheir needs directly to the fund providers as the land-grant system wasestablished.

This process is very dif ferent from that experienced by the "target"farmers in development projects in Third World countries. There, the smallfarmers are often marginal to the political process. Such farmers control fewfunds and have l i t t le organ izat ion a t a nat iona l leve l— the i r po l i t i ca l c loutis negligible as a result. The mandate for the program and the funds for theresearch and extension aimed at small farmers often come from outside donorsconcerned about macro problems of food production. While such outsiders fromeither international agencies or from the national capital ci ty try to understandthe farmers, their understanding is often incomplete because of the top-downnature of the programs.

Active pol i t ical part icipat ion to inf luence pol icy in one’s own immediateself- interest on an issue by issue basis is not always possible among ThirdWorld small farmers. Indeed, such issue mobi l izat ion, when i t occurs, is oftenblocked directly by governments— such movements then are either radicalized orrepressed, and often both. (The history of the National Peasant Users’Association in Colombia is an example (Bagley and Botero, 1978).) Lack ofeconomic power translates into lack of pol i t ical power. Upsett ing the statusquo by demanding participation in program formulation— which influences resourcedistr ibut ion— is translated often by those in power as revolut ionary andthreatening.

A second condition that we implicitly assume is the unity of goals betweenresearchers and farmers. In the United States, we knew clearly what the goalsand objectives of the farming household were. One of the reasons was that wewere they. Researchers and extension agents— the employees of the land-grantsystem— came from farming backgrounds; their parents were farmers and often theythemselves continued to farm (Busch, Lacy and Sachs, 1980). Goals and objectivesnever needed to be much discussed because they were implicit in the frame ofreference and upbringing of the researchers, the extension agents, and the usersof research— all of whom shared the same social background and the sameexperience in pract ical agriculture. Almost al l of the f irst extension agentswere farmers themselves, and even today, some land-grant faculty are part-timefarmers.

Contrast this now to developing countries. Third World small farmers oftenare not able to send their sons and daughters to grade school, much less tocollege or graduate school to learn agricultural technology. Young men or womenfrom marginal, two-hectare farms are very unlikely to attend the university, andcertainly when they do few of them will wish to return to the rural areas.(A parallel in our country can be drawn in looking at the dearth of blackagricultural scientists. Few black youth who escaped the backbreaking work ofchopping cotton had much desire to pursue agriculture when they had a chance tochoose a dif ferent career.) Instead, those who get into the formal agriculturalresearch and extension system tend to be people of urban backgrounds, often fromupper class or elite origins. These people, if they have been on a farm at all

-3 -

in Latin America, have been on their family finca or hacienda, where they hada chance to ride horses and enjoy a pastoral weekend of relaxation. But, theywere never engaged in the hard, physical labor of agricultural production.Through training in leading U.S. universities, Third World researchers’ goalstend to develop as scientists, not as farmers. Their reference group wi l lmore likely by First World professional journals rather than marginal farmers.Differences in experience in practical agriculture means that the kind ofresearch developed will not necessarily be that which corresponds to the needsof the everyday farmers, particularly those marginal farmers who produce mostof the wage foods— the subsistence crops most of the poor population eat inthe Third World.

A third phenomenon in the United States that does not exist in manydeveloping contexts is valuing the combination of mental and manual labor.This value stemmed from the conditions of production favoring freeholderagriculture. One of the reasons the farming lifestyle has been valued in theUnited States is that the farmer controls the means of production— no onetells the farmer what to do, as one would an employee (Kelly, 1979). That isto say, the farmer owns the land, or at least has t i t le to i t , even i f st i l lpaying off family debts. The farm family provides the work and the farm familyreaps the profits from that work, or shares the losses if there is a bad cropyear. Thus, in the U.S., we see nothing intr insical ly wrong with people gett ingtheir hands dirty and sweating in the fields because it is understood that thesame people who do the dirty manual work also make decisions about planting,harvesting, purchasing inputs, and marketing. They also get to keep any profitsgenerated.

How different this is from many developing countries of the world. Land ismuch less equally divided. (See, for example, Barraclough, 1973, for LatinAmerica; despite land reform, distribution remains much the same— the large farmshave simply become more capital intensive (Barsky and Cosse, 1981:52-56).) A fewlandowners have a lot of land that is often farmed by sharecroppers. Furthermore,even in places in Africa where land is more evenly divided, the people who makethe final decisions about production often are in urban areas, not in the ruralareas doing the manual work, (Clarke, 1980; Bernstein, 1977). The mistakenidea that getting your hands dirty and sweating is inappropriate for gentlemenand ladies was introduced by colonial masters who, because of their greaterrelative wealth, were able to hire the local population to do the manual work— itwas inappropriate for those in command. Education became a tool for getting awayfrom manual work, not for doing it better.

A fourth condition implicit in the development of the U.S. farm economy wasthat of the complete farm family— a husband and a wife sharing management, labor,and land ownership (once women gained property rights). The U.S. farm family isoften a decision-making and management unit, with the wife keeping the farmrecords and discussing purchasing and marketing decisions with the husband.Further, the unit of production was highly intertwined with the unit of repro-duct ion. I t was dif f icult to separate the accounting units of the home andenterprise, al though al l decisions involved al locat ions between them. (SeeWeber, 1947:275-280, on the implications of that lack of separation for becomingpart of a modern economy.) Division of labor by sex existed on U.S. farmscausing the man to be defined as the farmer and the woman as homemaker. But,labor has been important as well and, although rural-urban migration in the

- 4 -

United States has reduced the number of young people on farms, the coupleremains on the land.

What a contrast this complete farm family is to small Third World farms.Women are often active in all aspects of food production, both by custom and bynecessity, as temporary migration by males to participate as wage laborers inconstruct ion, extract ive industr ies and cash crops for export. This results inthe feminization of farming in many areas (Deere and Leon de Leal, 1980). Inthe Third World, division of labor by sex has also been traditional. Even whenwomen were the agriculturalists, men provided complementary labor, such asclearing the f ields or plowing, Temporary male migrat ion upsets the tradit ionalba lance.

A fifth condition that has aided the development of U.S. agriculture has beenthe ability of the larger economy to absorb the non-competitive farmers in otheremployment sectors. Thus, our research and extension efforts have been able tofocus on the more successful large, family farms. This "solut ion" for marginalfarmers is unacceptable in much of the Third World because other opportunit iesto generate family income are not available. Further, because of price and otherincentive structures, food production in the Third World is general ly carr ied outby these marginal farmers, while the modern farmers are large and engaged inexport agriculture.

A sixth condit ion that exists in the United States, but not in ThirdWorld countries, is the close cooperation between private and public sectorsin adapting and distr ibut ing technological innovation. The relat ively highpurchasing power of U.S. farmers has led to the development of a strongprivate sector in rural areas that forms the link between experiment stationand farmer. Private enterprise has had a long history of cooperation withpublic researchers in variety development, for example, focusing on adaptivebreeding, releasing and maintaining breeder seed varieties, and producinghigh-quality plant seed and distributing it to farmers (Grossman, 1982).This division of labor between private and public sectors has led the privatesector to be that most closely responding to (and attempting to form)farmers’ needs. The International Maize and Wheat Improvement Center (CIMMYT)researchers have pointed out how, in developed countries, the agribusinesssector integrates research results into effect ive technologies. "Developingcountries rarely have these mediating entities, and the bureaucracies andincentive systems of public institutions do not encourage researchers toplay this integrative role" (CIMMYT, 1981, p. 10).

The Contributions of Farming Systems Research

In our agricultural development work overseas, we must continually becomemore aware of the conditions we assume exist and compare them to the conditionsthat actually do exist. One tool that has evolved in the past decade to helpin this— a tool that takes us back to some of the early research and extensionin the U.S. land-grant system— is Farming Systems Research and Development (FSR&D).

What is FSR&D, and how can it help us counteract the biases arising fromassuming that the conditions in our own cultural setting exist all over? FSR&Dis aimed at increasing the agricultural welfare of the farm family by under-

-5 -

standing the whole farm in a comprehensive manner. The integrated demands ofthe unit of production/reproduction for alternative sources and uses of land,labor, capital, management, and equipment in the production system are related.The totality of crops and animals, and their by-products, for both subsistence useand for market, as well as temporary off-farm employment, are included. FSR&Dinvolves formal, interdisciplinary problem identification in participation withthe farm family, taking into account the needs of society as a whole. Incollaboration with farm families, appropriate technology is determined (usuallyfrom avai lable technology) and evaluated on their f ields under their constraints.FSR&D implies a two-way flow of knowledge between farm families and researchers.(See Shaner, et al, 1982, for an extensive development of this definition.)

Thus, farming systems research is, in essence, an attempt to use socialand production sciences together to approximate the conditions of research andextension that exist in our country because of the relative economic equality inour society and the relat ive equali ty and similari t ies in background of farmer,researcher, and extension agent in the U.S. — conditions that cannot be assumed toexist in the Third World. FSR&D provides a proxy for pol i t ical mobil izat ion and,pressure, a method of articulating the various goals of the participants (developmentagency, farm family, nation state), a combination of mental and manual work incarrying out the experiments, an awareness of labor and management availabilityby sex and age, and an unwillingness to define a farmer who is disadvantaged inre l a t i on t o l and o r cap i t a l as " i ne f f i c i en t . "

Farming systems research tries to identify the logic of the farming practicesthe farmer actually uses. In the U.S. setting, the logic of why a farmer doessomething is rather clear to the researchers because the researchers themselveshave used those practices, and perhaps grown up with them over time. Theresearcher in the United States intuits the problem and is able to bring about asolution. FSR&D provides explicit problem identification for cooperative problemsolving.

Farming systems research treats motivations and deals explicitly withgoals— factors that we can assume are equivalent between researcher, extensionagent, and farmer in our society, but which may differ radically in developingsoc ie t ies .

The activities farming systems research looks at are broad— perhaps broaderthan they currently are on farms in the United States. They include cropproduction and livestock production— these of course are similar— as well as theprocessing, storage, and marketing of crop and livestock products. Off-farm andnon-agricultural act ivi t ies must be considered as wel l , part icularly as theyimpact on agricultural act ivi t ies. There is a trend in U.S. agriculture towardpart-time farming— another set of factors impacting on goals, production, andprof i t fo r the farm fami ly .

As we have shifted to monoculture in U.S. agriculture, more and moreprocessing and storage has taken place off the farm. This is true despite themove toward building on-farm storage capacity that began in the ’50s. Combinesdo some grain processing, but on Third World small farms, processing is muchless mechanized and much more likely to be carried out on-farm.

FSR&D determines constraints present in the society and tries to work withinthem. If that cannot be accomplished (and generally farmers have been found tobe relat ively eff icient given the inputs avai lable), FSR&D can only be successful

-6 -

i f i t co inc ides wi th the ins t i tu t iona l iza t ion o f prov is ion o f new inputs ormarketing structures. Ready availability of recommended inputs that is assumedin the United States is not taken for granted in an FSR&D project. Because theUnited States is a primary manufacturer of fert i l izer, herbicides, machinery,and petroleum, it does not have the import problems that create blockages forinputs in developing countr ies. Problems of tar i f fs, exchange rates, andprocurement of hard currency here are not the obstacles to getting technologythat they are in the Third World. (Interestingly, there seems to be no researchon risk in farm decision-making when uncertain supply of a factor of productionis also a decision variable (Anderson, et al, 1982:425). Risk avoidance haslooked at the risk involved in natural phenomena— flooding, drought, etc., butnot at fertilizer or improved seed that is not available when needed.)

Similarly, the United States does not have a dual economy separating ruraland urban areas. Because of relatively high purchasing power in U.S. rural areas,distribution systems are in place and functioning, whereas in developing countriesthe state often has to assume that role. The problems Third World farmers havein acquiring input resources in a timely fashion eventually forces FSR&D to lookat institutions as well as contained production systems.

In developing countries, the maintenance, development, and procurement offarm resources involve, for example, the gathering of manure and the making oftools, as well as the gathering of fuel and water that take up endless hours.Fuel and water are a crucial part of the farming system. They are essential tomaintain the family to reproduce the labor necessary to continue the farm work(Hanger and Moris, 1973). This mix of reproductive and productive work, interms of the maintenance, development, and procurement of farm resources, andthe degree to which women perform most of them, is crucial if we are tounderstand fully the constraints under which marginal producers operate(Abdulla and Zeidenstein, 1981).

The farming system is composed of a variety of subsystems: social,biological, technical, and managerial. All of these must be understood as theyinteract with each other. Because of the lack of technology, the relation betweenthe parts is often very sensit ive. In comparing overseas agriculture to U.S.agriculture, we find that, in the United States, technology has overcome many ofthe biological di f ferences that cause dif ferent constraints to be important.Irrigation, of course, is a major technological innovation that has overcome alarge number of constraints in many parts of the world. Irr igat ion tends to bein the hands of large-scale landowners (Pearse, 1980:107). In work such as thatat the International Crops Research Institute for the Semi-Arid Tropics (ICRISAT),the complex social nature of putt ing into pract ice irr igat ion systems that takeinto account the needs and input of small farmers is being attempted (Matlon, 1980).

FSR and Farm Management Research

When we talk about farming systems research, many people who have longexperience in the land-grant universit ies shrug their shoulders and say, "That’snew? Isn’t that what we’ve always been doing in farm management research?"Certainly farm management, as it evolved, was a multidisciplinary attempt to lookat the entire range of factors involved in running a farm enterprise. Gilbert,Norman, and Winch discuss briefly the relation of farming systems research andfarm management research:

- 7 -

Despite important similarities between FSR and the earlyforms of farm management research, differences are apparentin the treatment of motivations and the f lexibi l i ty ofrecommendations emerging from the analysis of existingfarming systems. Farm management research assumed thatsuccessful farmers had to be thri f ty, hard-working, prof i tmaximizers. They would prosper, expand, and should beemulated. As late as 1947, farm management was beingdefined as "the act of judiciously and ski l l ful ly managinga farm" (Boss and Pond, 1947). Further, much of farmmanagement literature tended to be prescriptive in nature,indicating what farmers should do to be successful ratherthan trying to understand the logic of the farming practicesthat the mass of farmers were using (Gilbert, et al, 1980:121-122).

Thus, where prescription might work in the United States because assumptionsabout goals could be taken for granted, this cannot be the case in developingcountries, particularly among marginal farmers. We cannot assume that peoplewant to expand, to get larger to maximize profit. Further, we must realize thatone farmer expanding pushes another off the land, where even fewer chances ofgenerating income can be found. Families on Third World, small, marginal farmsseek to minimize risk. Often people will want to use any surplus generated foroff-farm investments, instead of reinvesting them in the farm, which has beenthe assumption of profit use in the United States. In the Third World, surplusgenerated is used rationally for ceremonial consumption which serves to furtherunite the community— to ward off the possibi l i t ies of divisions that wouldweaken everyone as the whole was weakened. Surplus may be invested to sendchildren to school. These children ultimately work off the farm, where chancesof higher return to the investment are available. Farming systems researchattempts to integrate differing goals into research programs aimed at marginal,small farmers.

Current farm management in the United States has as its goal the greatestcontinuing prof i t (Bul ler, 1976:1). This assumption for small farms indeveloping countries cannot be taken for granted. Other goals may be motivatingthe farm family, and these goals must be recognized and taken into account whentechnology is recommended, research is done, and results are assessed.

In farm management, as currently practiced in the United States, theassumpt ion is that pro f i tab i l i ty is the goa l . Prof i tab i l i ty has the advantageof giving us something easy to measure. Weber called it formal rationalitynecessary for a modern economy (Weber, 1947:184-186). But, many Third Worldmarginal farms are not part of the modern economy. Profitability can be measuredby reducing inputs and outputs to dollars and cents, which can be easilysubtracted from one another. This single summary measure makes for easy compar-isons between farms. In developing countries, because we cannot assume that thehighest continuous profitability is the goal, we have to devise other measuresof success. We can no longer reduce all inputs and outputs neatly and carefullyto economic terms.

In the United States, early Farm Management Research (FMR) assumed that abetter cropping mix and more judicious use of resources would improve the farmfamily’s welfare— and maintain soil productivity— through higher sustainable incomes.

-8 -

It took a long-term view, and included costs that could be externalized, such asland erosion and water quality and sustained quantity. FMR allowed for anapproximation of experimental conditions by examining existing practices to seewhat was the most profitable; it then helped farmers to apply the results of thisquasi-experimental research through good record-keeping and problem analysis(McDermott, 1982, personal communication).

In Third World FSR, the researcher works with the farm family to decideupon appropriate innovations that can be introduced to increase farm familywelfare, yet not contradict long-range societal goals. General ly, this involvesincreasing prof i ts, but reducing r isk is even more crucial for the small ThirdWorld farmer. Maintaining a fragile ecosystem, which may be antithetical tosmall farm prof i tabi l i ty, must also be taken into account.

In early FMR in the United States, it was demonstrated that the diversifiedfamily farm was better able than the monocultural farm to withstand theinevitable dips and r ises in the price structure. Yet, because of theconstraints of introducing more efficient machinery to reduce labor costs andtroubles, and because research was commodity oriented and not system oriented,agriculture steadily became more oriented to monoculture— and more susceptibleto market trends (Flora and Rodefeld, 1978). For the small, Third World familyfarm, the purchase of large, eff ic ient equipment is impossible, even i f thefuel and other inputs would continue to be avai lable to run i t . Further, themarginal Third World farmers do not have the ability to withstand several yearsof bad prices i f a single crop is depended on. The family farm diversif ies i tssources of income, even if such a tactic may be inefficient in receiving thehighest return on any single crop. (Research carried on by lnstituto ColombianoAgropecuario (ICA) in Rio Negro, Antioquia, shows the superiority of a croppingsystem using beans, corn, and potatoes over maximizing the yield of any one ofthe crops. )

Farm management, l ike FSR, starts with problem identi f icat ion. Identi fyingthe problem is the only way to go about solving it. However, methods of problemidentification learned and shared, for example, by midwestern farmers, extensionagents, and researchers, may be very different from both the logic and theperceived problems of farmers in developing countries. Farm management research,as farming systems research, stresses that data are necessary in order to makethe proper decisions. Those data involve what people actually do and theimplicat ions of these act ions related to their perceived goals.

has specif ic and definable goals and object ives (Bul ler, 1976). Part of farmAccording to current farm management theory, the successful U.S. farmer

management practice is to help people develop these goals and objectives, whichare primari ly f inancial, general ly within terms that are numerical ly operational,often facilitated by computer.

Less successful U.S. farmers do not have goals and are more passive(Buller, 1976). They wait to see what will happen to them. That is a morefatalistic approach, but one which is more congruent to people who may notcontrol direct ly al l the resources necessary to farm. For example, in thedeveloping countr ies, i f you are a tenant farmer i t is dif f icult to set goalsbecause you cannot control what will happen (Griffin, 1976). If, as in manycases, you are a woman farmer— which is increasingly likely as male migrationto seek off-farm employment increases— your ability to control labor as wellas other resources is greatly diminished vis a vis males (Bourque and Warren,

- 9 -

1981). Setting goals and objectives is outside the farmer’s purviewwhen the farm has to respond to resource scarc i ty . Farming is react ivewhen access to inputs is limited. When access is even more limited because ofgender and differential power, often caused by a colonial history (Etienneand Leacock, 1980), farming becomes even more reactive.

Farm management research in the United States has assumed that the farmeris male, and the successful farmer "involves his wife as a meaningful helpmatein his business and community affairs" (Buller, 1976:4). The assumption behindfarm management research is that farming is a family enterprise and is onlysuccessful when this occurs. This same assumption is held in FSR. Thedifference is that we realize that in the Third World the farmer will sometimes befemale and, in certain situations (particularly when we are talking aboutmarginal farmers producing subsistence crops) the farmer is more likelyto be female than male. The realization of who does the farmingversus who is gett ing the agricultural information is crucial. Thisdisjuncture must be taken into account as we try to overcome our culturalbiases in approaching developing processes.

Farm management in the United States arose as farming became morecommercialized and capitalistic (Case and Williams, 1957:5). This commer-cialization and capitalist nexis is very important to remember when we contrastFSR&D in the Third World with farm management in the First. Farming systemsin the Third World (particularly those of small farmers) very often have asubsistence basis rather than a commercialized basis. Many of the economicstructures into which the farms are placed are pre-capital ist rather thancapitalist.

Third World farms are linked to the capitalist market in three ways: thesale of their products, the sale of their labor, and the inputs they purchase.Linkage through the sale of crops is most common in the United States. Yet,the majority of the crops very often are not sold, but are consumed on thefarm or bartered local ly, part icularly in parts of Afr ica and Latin America.

The more usual way in which the small farm in the Third World is linkedto the market economy— that is to say to the world system— is through the saleof labor. The farm family becomes semi-proleterianized, with one part of thefamily maintaining the farm and production, and another part of the familyselling their labor in the market (Goodman and Redclift, 1982). The people whosell their labor tend to be those who command the highest wages because cashtakes priority. These people tend to be males (Leon, 1982). This leaves womenin charge of production of subsistence crops— the wage foods— that are thebackbone of the economy and, yet, these crops are the least valued within thatsociety. Men wil l take jobs, general ly through temporary migration(Friedman, 1978; Laite, 1977) and will work in the harvest of cash cropsparticularly export crops (Homen de Melo and Zockun, 1977). Export cropsinclude cotton, coffee, cacao, and sugar. The men may also be involved in theconstruction booms that accompany petroleum development, particularly inpetroleum countries. For instance, in Equador and in Nigeria, people willmigrate to the urban centers to provide a low-wage, construction labor force.Employment in extract ive industr ies, part icularly mining, can also be the f irstmajor link with the monetary economy. In much of southern Africa, the movementof males to work in the mines has left farms in the hands of females.

-10-

It is important to understand that in both farm management and farmingsystems research the increased commercialization and increased penetration ofcapitalism has made a sharp impact on the farm family, but the way that hasoccurred is very different in the Third World when compared to the developedworld. Perhaps a good way of seeing this is through understanding where in theUnited States farm management as a science first evolved.

In the United States, farm management research was first developed in theNortheast and the Midwest (Case and Williams, 1957), where independent farmfamilies controlled the land and provided the vast majority of their own labor.It developed much more slowly in the south, where agriculture was dominated bythe sharecropper-tenancy type of organization that t ied the farm family on theland closely to the landowner for direct ion and f inancing. That separation ofthe factors of production in the farming system reduced efficiency. Access toland was divided from ownership. Management was split. Labor was separatedfrom the land and, as a result , product ion credit was dif f icul t to obtain. Whenthe goal of profit maximization was addressed in the South, one had to ask,"Profit maximization for whom?"

This is certainly the case in much of Latin America today. The conditionsfor effective farm management research do not exist as we know them. However,farming systems research, using many of the early principles of farm managementresearch (particularly those that closely linked agronomy with the diversegoals of the farm family), can he useful. But, in order to be most effective, wehave to set aside our assumptions about the conditions under which research andextension take place.

First, the small farmer, particularly women farmers, will not have thepolitical organization and clout to influence the programs presumably designedto help them. Second, agricultural researchers and extension workers will nothave the same social, cultural, and economic backgrounds as the small farmers.In particular, they will not have the same gender-related experiences. Theywill not know what it is to carry water, gather wood, cook meals, and nursebabies while also planting, harvesting, and processing crops. That lack ofunderstanding may provide a large constraint to the development of appropriatetechnology and the adoption of innovation to increase productivity by femalefarmers. Third, manual work, particularly women’s work, will be devalued bythose who work with their "heads." Respect and appreciation of the manualaspects of productive and reproductive work have to be developed. Fourth,temporary male migration, in order to increase family cash income, will leavewomen in charge of farming. No longer can we assume, "the farmer, he..."Fifth, men usually have more control than women over the factors of production— the inputs to the farming system (land, labor, capital, technology, andmanagement) both on and off farm. We must understand these separations anddevise mechanisms to overcome the constraints they entail. Sixth, because offarmer’s economic marginality, the integrative function between research andappl icat ion performed by private enterprise in more lucrat ive sett ings wi l lhave to be assumed by other mechanisms, probably with strong public sectorparticipation. Seventh, we must deal with marginal farmers, who are morelikely than large-scale farmers to be women, rather than simply deal withthe modern or potentially modern sector, which is male-dominated. Much ofdomestic food production, as contrasted with export crop production, is doneby Third World women.

-11-

We must recognize and deal with the differential structural conditions ofresearch-extension-production— and the unique roles of women within it, if weare to gain the object ive of increasing the agricultural welfare of Third Worldfarm families— and improve agricultural productivity and the general welfarethroughout the world.

REFERENCES

Abdulla, T. and S. Zeidenstein. 1981. Village Women of Bangladesh: Prospectsfor Change. Pergamon Press, Oxford, England.

Anderson, J.R., K. Dvorak, S. Setboonsarng, J.A. Roumassett, and M.H. Vesenka.1982. BNF and the U.S. rice industry: A review of economic aspects towarda technological assessment of BNF. Interim Report, Resource SystemsIns t i t u t e . Eas t -Wes t Cen te r , Hono lu l u , Hawa i i (Sep t . 22 , 1982 ) .

Bagley, B.M. and F. Botero. 1978. Organizaciones campesinas contemporaneasen Colombia: Un estudio de la Asociacion Nacional de Usuarios Campesinos(ANUC). Estudios Rurales Latinoamericanos. I (Enero-Abri l):59-96.

Barraclough, S.L. 1973. Agrarian Structure in Latin America. D. C. Health,Massachusetts.

Barsky, O. and G. Cosse. 1981. Tecnologia y Cambio Social: Las HaciendasLecheras del Ecuador. FLACSO, Quito, Ecuador.

Bernstein, H. 1977. Notes on capital and peasantry. Review of Afr icanPolit ical Economy. X:60-73.

Bourque, S.C. and D.B. Warren. 1981. Women of the Andes: Patriarchy andSocial Change in Two Peruvian Towns. The University of Michigan Press,Ann Arbor, Michigan.

Buller, O. 1976. Farm Management Laboratory Manual and Supplemental Text.Dept. of Economics, Kansas State University, Manhattan, Kansas.

Busch, L., W.B. Lacy, and C. Sachs. 1980. Research policy and process inthe agricultural sciences: Some results from a national study. RS-66.Dept. of Sociology, Kentucky Agricultural Experiment Station, Lexington,Kentucky.

Case, H.C.M. and D.B. Williams. 1957. Fifty Years of Farm Management.Univers i ty o f I l l ino is Press, Urbana, I l l ino is .

CIMMYT (Centro InternacionaI de Maiz y Trigo). 1981. CIMMYT Economics.CIMMYT, Mexico (April).

Clarke, J. 1980. Peasantization and landholdering: A Nigerian case study,M.A. Klein, ed. Peasants in Africa: Historical and Contemporary Perspective.Sage Publications, Beverly Hil ls, Cali fornia.

-12-

Deere, C.D. and M. Leon de Leal. 1980. Women in Agriculture: PeasantProduction and Proletarianization in Three Andean Regions. Unpublishedmanuscript, International Labor Organization.

Etienne, M. and E. Leacock. 1980. Women and Colonization: AnthropologicalPerspectives. Praeger, Brooklyn, New York.

Flora, J.L. and R.D. Rodefeld. 1978. The nature, magnitude and consequencesof changes in agricultural technology, in Change in Rural America: Causes,Consequences and Alternatives. Rodefeld, Flora, D. Voth, I. Fujimoto, andJ, Converse, eds. C.V. Mosby Co., St. Louis.

Friedmann, H. 1978. World market, state and family farm: Social basis ofhousehold production in the era of wage labour. Comparative Studies inSociety and History. XX(4):545-586.

Gilbert, E.H., D.W. Norman, and F.E. Winch. 1980. Farming systems research:A critical appraisal. MSU Rural Development Paper No. 6. Michigan StateUniversity, East Lansing, Michigan.

Goodman, D. and M. Redclift. 1982. From Peasant to Proletarian: CapitalistDevelopment and Agrarian Transitions. St. Martin’s Press, New York.

Gri f f in, K. 1976. Land Distr ibution and Rural Poverty. MacMil lan PublishingCo., London.

Grossman, R.S. 1982. Control of agricultural germplasm and geneticinformation. Paper presented at the Conference on Biotechnology,Agriculture, and the Public Interest, University of Kentucky, Lexington.

Hanger, J. and J. Moris. 1973. Women and the household economy. Mwea: AnIrrigated Rice Settlement in Kenya. Weltforum Verlag, Munich.

Homen de Melo, F.B. and M.H.G.P. Zockun. 1977. Exportacoes agricolas, balancode pagamentos e abastementos do mercado interno. Estudos Economicos. Vll(2).

Kelly, K.D. 1979. The independent mode of production. The Review of RadicalPo l i t i ca l Economics . 11:38-48.

La i te , J . 1977. Migra t ion, cap i ta l accumulat ion and soc ia l d i f fe rent ia t ionamongst the peasantry. Paper presented to the Social Science ResearchCouncil, Anthropology Workshop, University of Sussex.

Leon, M., ed. 1982. Las Trabajadores del Agro. Vol. II. Debate Sobre laMujer en America Latina y el Caribe-Discusion Acerca de la UnidadProduccion-Reproduccion. ACEP, Bogota, Colombia.

Matlon, P.J. 1980. Local Variet ies, Planting Strategies and EarlySeason Farming Activities in Two Villages of Central Upper Volta. ICRISAT,W.A. Economics Program, Progress Report 2, Ouagadougou, Upper Volta.

Pearse, A. 1980. Seed of Plenty, Seeds of Want: Social and EconomicImplications of the Green Revolution. Clarendon Press, Oxford.

- 1 3 -

Rothstein, M. 1978. Crisis and change in a century of American agriculturalresearch, in Outreach Programs of the Land Grant Universities: WhichPublics Should They Serve? J.L. Flora and J. Converse, eds. Kansas StateUniversity, Manhattan, Kansas.

Shaner, W.W., P.F. Philipp, and W.R. Schmehl. 1982. Farming Systems Researchand Development: Guidelines for Developing Countries. Westview Press,Boulder, Colorado.

Weber, M. 1947. The Theory of Social and Economic Organization. The FreePress, New York.

World Bank. 1981. World Development Report. Oxford University Press,New York.