Participatory Technology Development With Farmer Research Groups

A|{D NATI.IRAL REsoUcEs MINISTRY or AGRICULTURF,,LIVESTOCK ZA|%IBARcAsHcRoPsTARMINGSYSTEMSPROJECT

RwiewMecting Paperpreprcd for theZARCfunual Ressarch Nocnrber14fi & 15,1995

PARTICIPATORY TNCHNOI,OGYDEVELOPMENT WITH FARMER RESEARCHGROUPS

By: Ismatt MgPni tvlartin Walsh RuPert woods

zccFsP PO Bor 228it Zanzlbor Tel./far: (05Q33121

CONTENTS Introduction

Pageno. I

Localsocial,economic andinstitutional factorswhichled to theadoptionofthe FRGapproach

I

WhyformFarmerResearch Groups?

2

Initid orperiences

2

Currentmethodfor selectingandforminggoups

4

Methodsfor workingwith groups

6

Problemareas

9

Otherapproaches beingusedby ZCCFSP andMALNR

l0

Theinstitutional constraints to furtherdevelopment of the FRGapproach in Zanabuandprospects for thefuture

t2

Lessons

t2

Summary of methodology

l4

ABBREVIATIONS FRG MALNR PRA

zccFsP

FarmerResearch group Ministryof Agriculture, Livestockandnaturalresources Participatory Ruralappraisal ZanabuCashCropFarmingSystems Project

INTRODUCTION The ZanzibarCashCrop FarmingSystemsProject(ZCCFSP)startedin l99l with the cashcrop farmingsystems,increasing principleobjectivesof developingsustainable foreignorchangeearnings,reducingexpenditureon importsandraisingthe incomesof rural householdsonZanzibar(PembaandUngujaislands). It is fundedby Overseas Administration(ODA) of the UK andmanagedby the NaturalResources Devetopment Institute (NRI). It wasoriginallyfoundedon the principlethat new exportcropscouldbe developedto the existingexportcrop of cloves,for whichthe world replace,or at leastto supplement, marketdemandhasgreatlyreduced. At the outs€f,therefore,the approachtakenwasto selectindividualcropswhichwere to Zarvibari thoughtto havea good market,whichwould grow well andbe acceptable resultedin the investigation agtonomic process and PRAs, market surveys of farmers. A compilationof a list of candidatecropsfor furtherresearchanddevelopment. andthe decision approachmetwith little success Thisessentialty top-downcrop-oriented moreon the was soonmadeto adopta moreparticipatoryapproach,to concentrate farmingsystemspart of the projecttitle. of FarmerResearchGroups with the development our experiences This paperdescribes researchandextension, farmer-participatory (FRGs)asa key methodfor enhancing LOCAL SOCIAI,, ECONOMIC AND INSTITUTIONAL FACTORS WHICH LEI) TO THE AI}OPTION OF FRG APPROACH Thc choiceof mostsuitablemethodsfor workingwith farmersdependsto a largeextenton socialandeconomicfactors,Theconditionsthathavebeenexiatingwithin institutional, asfollows; Zulzibarcanbe summarised (MALNR) Within Ministry of Agriculture,LivestockandNaturalResources o A top-downepproachhasbeenused,usingthe principlethat the researchsectionsdo 'taught' to farmers' the researchandthenthe resultinginformationis o Researchhasmainlybeenfocusedon theoreticaltechnicalfixes:fertilizers,spacing,time of planting. It is seldomproblemorientatedor takesinto accountfarmerknowledgeor conditions.On-stationtrialshavebeenusedandsomeon-farmtrials socio-economic with scatteredindMduals. r Lack of coherentpoticy. The policyis relatedto countryobjectives(eg productionof rice) andseldomrelateto farmerproblemsor conditions. projectsattrastthe bestof the staff r Projectisation andunderfunding.Donor-supported in decisionmaking. Very little co-ordinationbetween andareoftenautonomous limitedto payingstaffcostsand sections, Fundingfrom the governmentis essentially little is givenfor runningcosts. tractor r Long historyof freeor subsidised inputs(fertilizers,seeds,agro-chemicals, services).Thisis now reduceddueto lackof funds,

o Organisation by crop or discipline:makesit difficult to respondto moregeneral problemsfacedby farmers(eg declineof soil fertility). Rural areas r High populationdensity,smallfarm size,considerable landbonowing in manyareas, r Significantvariationbetweenare&s:agro-ecological; oropsgrown;importanceof ofl farm income;accessto markets;involvementof womenin agriculture. o Historyof governmentcontrolhasresultedin a reductionin the sens,e of responsibility of farmersfor solvingtheir own problemsanda lack of communitystructurewithin villages,outsideof familygroups. Also an expectationof directbenefitswhenworking with government. r Low input-low output agriculturepractisedin most il€as.

WHY FORM FARMER RESEARCH GROUPS? The reasonsfor usingthe FRG approachcanbe summarised asfollows.

!; ;.': ' '-. :1.i,'.:

o Relevanlresearch.With farmerresearchgroupsthe researchprioritiesarisefrom the constraintsandopportunitiesfacedby the farmerswithin the group. If the group is well whicharedevelopedby the groupwill be relevant chosenthe newideasor technologies to a wide sectionof farmingfamilieswithin that farmingsystemzone. o Organistion. From anorganisational point of view it is easierto work with 15 farmers in a group,ratherthan l0 individuals. o Contirurry. If farmersdrop out otherscanjoin, the groupwill havea'memory' of what hasbeendoneandthereis morechanceof activitiescontinuingif the directproject assistance ends, Long-termcontact. o Exchangeof informationwithin thegroup. The learningprocessis muchquicker as farmerscanlearnfrom eachothers' experiences. t Responsibility.Farmersaremorelikely to takeresponsibilityfor their own problemsby talkingtluoughthernandsearchingfor possiblesolutions. t Collaborationhelweenfarmers. Thereareadditionalbenefitsfor the farmersfrom comarketing;andattractingattentionfrom other operationin a group:sharingresources; organisations. t Collaborationwithin lult4l.ilR. It was felt that the FRG approachwould be a good methodfor encouraging co-operationwithin the MALNR, Differentsectionscan by the farmers, collaborate canbe demand-led on differentproblems.Thecollaboration 'ownership' (the problemof one sectionbeingaskedto whichreducesthe problemsof helpwith anothersection'strials). o Trainlng. Good training experiencefor ministry staff

,5:t.':.:.'' 'il{iil{ , .::

..

,l

r:::i'iRi; ' '"1,1

"]j

INTIIAL EXPERIENCES ZCCFSPis a so-called'processproject', Thismeansthatthe preciseend-pointandthe methodsfor gettingtherecannotbe clearlydefinedfrom the outset. To this enda flexible approachhasbeenusedto try to developmethodsbestsuitedto Zanabu conditions.

':;.", ' "' i '1i:'lr'

"_. .1.:

'.,:'

'i'.';:,il . ' l;1,';.,;iili

3

:.'i.::\1.I

. . : : -:

.,r..,;::.' l.

The farmerresearchgroupprogrammehasbeencautiouslybuilt up. Formingonegroupat a time anddevelopingthe approachovertime. In this way it is hopedto avoidmakingtoo manymistakes.This is particularlyimportantwith farmerresearchgroupsbecause considerable investmentis madeat the outsetto starta group. Cunentlytherearea total of six FRGs,threeon eachisland(PembaandUnguja). A rangeof differentmethodshavebeenusedfor startingandworking with thesegroups. The experiences gainedduringthe formationof the first few groupsaredescribedbelow, of the currentmethods, beforediscussion Farming SystemZones As a first step,beforeforminganyFRG,the islandswere sub-dMdedinto FarmingSystem Zones. The objectivesof this exercisewereto describewhat washappeningin agriculture within the differentzonesandthen,ttrough a seriesof meetingswith othersectionsof MALNR' to identify the mainagriculturalconstraintsand opportunitiesfor eachzone. I sv es regarding locati on

:

It was decidedthat formationof FRGswould relateto the FarmingSystemZones. The key issuesregardingthe locationwerewhetherto targetareas:whichhavethe mostcashcrop opportunities;or whichhavethe mostproblems,or wherechangeis happeningfastest;or to try andcoverall areas? Also whetherto selecta villagerepresentetive of a certainzone or a villagewheretheyhaveaccessto two zones. /ssaesregwding granp seleclion Therearotluee mainfactorsto consider e Size o Composition o Method of forming a group In the pastfarmerswereselectedmainlyon the basisof willingnessto co-operatein doing research.Whenforminga researchgroupthereareadditionalrequiranents,suchas wealth,familygroupings,race,gender,ageandwillingnessto do research.Thesefactors haveto be considered for selectionof a groupwhichis representative of a wider population andwhishwill be ableto work together. For our purposeswe wanteda goup which wasmostrepresentative of the peoplemost involvedin agriculturewithin that area. It is easyto get homogenous goups (eg womenor family-based groups),whicharethe sort of groupswhichthe extensionserviceusuallyworks with. A mixedgoup, however,is moredifficultto achiwebecauseit maynot representa naturalgrouping. The first groupwasformedby invitinga contactfarmerto selecta $oup. The resulting grouphasa reasonable mix of ages,but too manyof the groupmembersarerelated. There is only onewomanin the group,but womenarenot muchengagedin agriculturein this area.

4

For the secondgroupit wasdecidedto targetwomenonly. This decisionwasbasedon the observationthat womendo mostof the agriculturalactivitiesin the selectedfarmingsystem zone,whereasmenaremostlyinvolvedin earningoFfarm income. Followinga more detailedstudy,howwer, it wasdiscoveredthat althoughwomendo mostof the agriculturalworlg theywerenot muchinvolvedin farmingon the coralrag, an important farmingareafor the village,becausethey arenot ableto do the necessary bushclearance, A discussion washeldwith the womenaboutthis issueandtheydecidedto includefour menin the group.This groupnow consistsof eightwomenandfour men.

CURRENT METHOD FOR SELECTING AND FORMING GROUPS, Vlllage selection Therewasmuchdebateaboutwhereto form the mostrecentgroup. Therewere somethat favouredstartingin a FS zonewherethe projecthadnot hithertobeeninvolved,but where thereareperhapsfew cashcrop opportunities;othersfavouredhavinga secondgroup in the largestzone,wheretherearethe mostconstraintsandopportunitiesfor agricultural dcvelopment.The compromise that wasreachedwasto selecta villagewithin the largest zone,whichwasfar from the existinggroupwithin that zoneandwhichalsohadaccessto the adjoiningzone. It wasfelt that anytechnologies developedwould therebyhave relevance to two zones. Severalvillageswere chosenwhich met this generalselectioncriteria. Thesevillageswere visited over a period of two daysandinformationwas collectedon agriculturein each village. Finallyonevillagewaschosen,whereit wasconsidered that the mostnew (independent developments happening were of outsideassistance). Group selectlon

' ' . 1

:;;:'i,-rt.i\i . ,r':i1..i ' . t : : I

Havingdecidedon the village,a teamof turo peoplewere sentto the villagewith the task of identiffingpossiblegoup members.Threeseparateinformantswere identifiedwithin the village,who wereselectedon their abilityto be ableto identify possiblegroup members. Theseinformantsweretold aboutthe aimsof forminga researchgroupand wereaskedto suggesta mix of people-difFerentages,gorder,wealth,familygrouping. The importantcharacteristic for all groupmembersbeinga willingnessto participateand try newtechniques. Followingthesediscussions andbasedon a comparisonof the threeseparatelists a final list of I 5 nameswasdrawnup. Our experience is that the idealgroupsizeis from 12to 15 people,anylargerthanthis andthe groupbecomestoo difficultto managewith no extra benefits. Creatinga researchprogrqmme We havefoundthat the importantfirst stepwith farmerresearchgroupsis to gaina thoroughunderstanding of their farmingpracticesandsocio-economic conditionsbefore

'ij

".:, '- !

:..r;i;.':,ii \'l',!ii i.: 'ij t iri:*

'..',:'',',.'} '... ..'.:,..,.

. i'l.i.'; .:r:: '','.;'.il.

makinganydecisionsabouta possibleresearchprogramme.Thereforewe conducta tlpe of mini-PRA The methodologyusedis asfollows. l. Soonafterthe selectionof a researchgroup,we hold a discussionwith the groupabout the conceptof researchandthe aimsof formingsucha group. A dateis thensetfor doing the PRA. 2, ThePRA teamareselected,Theaimis for 8-10teammembers, drawnfrom the group. The sectionswithinMALNR thathavean interestin collaborating with the research teammeeta few daysbeforethe PRA anddrawup a checklist for datacollection.Topics that arecoveredinclude:Landuseandownership;labourissues;soil fertility and practices;croppingsystems(food andcashcrops);forestry;livestock; associated marketing;incomesources;household decision-making; andgenderissues. 3. Informationcollectionis baseduponsemi-structured interviewsandvisitsto the individualplotsof eachFRG member.A teamof two or threeinterviewersspendoneday with a singlefarmer,the morningvisitingthe farmer'splotsandthe afternooncompleting the interviewat the farmer'shome. 4. Eachday'sfindingsarediscussed collestivelyeacheveningandanynecessary modificationsaremadeto the checklistof the followingday. 5. A meetingwith all the membersof the FRG is heldon the final morningto providean opportunityto pres€ntanddiscusspreliminaryresultsof the surveyandoutlinesubsequent stagesof the researchprocesswith farmersbeforethankingthemfor their co-operation, 6. Backin theoffice,a dayis givento the qollectiveanalysis of the data. Minutesof the analysisaretakento form the basisof a reportsummarising the findingsof the study. Also notesarecompiledon eachfarmer,whichgenerallytakesabouta week. Responsibility for drawingup the reportis givento two teammembers. 7. Theteamthenget togetherto drawup a tist of possibleareasfor researchandany furtheractionneeded,basedon the collectiveanalysis. 8. Finallythe researchproposalsandcommentsof the teamarediscussed with the FRG to modifr andrefinethe proposals,togetherwith anyadditionalsuggestion. :

The advantages anddisadvantages of this mini-PRAprocessareasfollows:

'.r.!

Advantages r importanttrainingfunctionfor projectandMALNR staff Improvesinterview,report writing andanalyticalskills. o More rapidthanconventiondPRA whichinvolvesmatrixandwealthranking. r The informationgatheredcanbe directlyused. I

Disadvantages o Sometypesof informationarediffictrltto obtainwith zucha rapidprocess,for example informationaboutincome.

o Farmersexpectquickrespomefrom the researchteam. o Time consumingfor farmer.Takesroughlylr/2 daysfor eachfarmerand 4 - 5 daysfor the wholegroup.

METIIODS FOR WORKING WTTH GROUPS The methodsusedfor workingwith groupshasdevelopedover time. The historyof researchwith the oldestgroup(DayaFRG) is givenaBan exampleof this development. wasput on doingsimpletrialswith new cashcrops(chillies,ginger, Initially the ernphasis furmeric). Simpletreatmentswereused,suchas* or - shade,r or - mulch, Monitoring asthe farmersexpecteda lot wasdonebi-monthly.This approachwasnot very successful from the newcropsandtwo of the threecropsprovedunsuitablefor the localconditions. Subsequently, the groupwasgiventhe choiceof a wide rangeof tree crop seedlings, rate, were soldat a subsidised availablein the projectandforestrynurseries,The seedlings plant left farmer. where was the numbers and to of species, to choice This was still not very satisfactoryasit wasdifficult to dealwith majoriszues,suchas soil fertility andredevelopment of the cloveplantationares. A detailedstudywasthen undertakenof eachfarmer(asdescribedfor PR.lrabove). Fromthis a muchwider rangeof researchthemesarose.The themesareactuallyquite similarfor all the FRGs,although therearesomedifferences.The cunentthernesfor DayaFRG are: suchaserosioncontrol,compost This cov€rsinvestigations Soilfertility mointerwnce. making,agroforestrymixedcropping,economicuseof fertilizers,mulchapplication,useof legumes,rotatioq greenmanure,covercroppingandfallowing. Also the adviceto concentrate effortson a smallerarea. Farn outptttdivvrsilicatioz. This is particularlyaimedat developmentof the clove areas, whereonly cloves,cassava andbananas aregrown. Oneobjectiveis to ortendthe home gardenideainto the cloveplantation,to havemulti-storeycroppingof a rangeof crops. Main emphasis is on trylng newspecies,extendingcultivationof someexistingspeciesand trylng improvedfood crop varieties,CIher activitiesin this themeincludenursery establishment, monkeycontrolandbreadfnritdrying, Development of rice areqs. It is felt that the ricevalleysareunder-utilised,despitethe fact that manyhavewaterall the year. The objectiveis to improveproductivitythroughcrop insteadof rice) diversification,diversificationinto newuses(eg sugarcaneor vegetables anduseof bundsto improvewaterholdingcapacity. with the groupandthe individual Followingthe PRA in Daya,the findingswerediscussed farmersdecidedwhat sort of investigation theywould do. Now thereis a mix of : t Semi-Iormaltrials. Examples:comparisonof artificialfertilizer,compostandcattle designed- farmer asresearcher manureon bananas.Thesetrialscouldbe categorised implemented, e Inlormal investigalions.Examples:compostmaking,monkeycontrol,trying new crops/varieties.Couldbe describedasfarmerdesigned- farmerimplemented.

7 o Discassionon topical issaes.Examples:developmentof clove areas,managernent of mayarisefrom the disqussions, forestarcas,discussion aboutcrops. New investigations that otherwiseanydevelopments arenoted. Oftenthe point is madein thesediscussions the farmersmusttakeresponsibilitythemselves andthat they shouldco-operatebecause if only onepersonactson their own thuyarelikely to fail. t Tloining. Studytours, visits Meaflrements and recording All trialsarekept very simple. The majorobjectiveis to get the farmers'opinionsabout the viabilityof a technology(technicalandsocio-economic).In sometrialsthe farmer the yield,in othersno measurements aretaken, measures An exampleof a trial whereno measurements aretakenis an erosioncontroltrial. This is a completelynewconceptfor the farmersanda wide rangeof specieshavebeenplantedon methodson a single the contour,togetherwith a comparisonofvegetativeand mechanical farm.The ideais to get a first response of the farmersaboutwhatt)?e of methodsmay plot. havea chanceto work. This will alsoserveasa demonstration Thereis not a strongtraditionof recordinginformationor writing paperswithin the MALNR andwhereaseffortsaremadeto improvethe analyticalandreportwriting skills, paperworkis alsokept to a minimum.The currentthinkingis to havea shortannualreport on the progr€ssofeach group. Inprts Materialinputsshouldbe kept to the minimumpossible. Therearecases,however,when the outcomeof a trial is completelyuncertainandthe costsof materialsor labourarehigh; or wherethe benefitsto the projest,in termsof the knowledgegained,maybe potentially greaterthanthosefor the farmer;or whena largeplot hasto be plantedto get realistic results. In thesecaseswe feelit is justifiableto give a great€rlorel of support. In a recentexample,a farmerwaspaidfor someof his labourfor implementinga large erosioncontroltrial on his farm.Someof the plantingmaterialwasalsoprovided. Before agreeingthe rateto be paid an independent labourerwasaskedto give a quotefor carrying out the work andthe paymentof the farmerwasbasedon this. Erosioncontroltechniques arenew to the islandsandthe aimwasto comparea largenumberof techniqueswithirta singlefarm, Farmer training and studytours The useof trainingor studytoursis an importantcomponentof the work. Sometimes farmersaretakento otherareas(eg from one islandto the other)to get first-hand experience,This maybe to se€a specifictechnology(eg compostmaking,vanilla (eg cattlecropintegration). cultivation,erosioncontrol),or to seenewfarmingsystems Trainingin the villageis alsogrvenfor somenewtechnologies for whichthe farmershave hadlinle experience,

8 Meetings Meetingswith the grouptakeplaceat irregularintervals,dependingon the time of yearand the researchactivitiestakingplace,the intervalcanrangefrom 2 to l0 weeks.In between the groupmeetingsindividualfarmersarefollowedup. Group Ieader ship/organisation Someof the groupshaveappointeda chairmanandsecretary,othershavenot. Decisions of this kind areleft to the groupsto decide.For the groupswhich haveseveralmembers from a singlefamilygroupit is perhapsbetternot to haveanychairmanasthis maycause rifts. Topical studies issuesor researchpossibilitiesarisefrom the work with the FRGsfor which Sometimes moredetailedstudyis required.An example(plannedfor October1995)is a studyon the agroforestrypossibilitiesfor the coralrag (soil type). The objectiveof this studyis to identifyindigenousandimportedtreespecieswhicharecurrentlygrowingon the coralrag, to investigatethc performance of eachspeciesandthe way in whichtheyareused. From this studywe expectto be ableto designagroforestryinterventionsfor adaptationby the KangaganiFRG. Reyalts It takesat leasta yearbeforeanyrenrltsof co-operationwith FRGscanbe seenandmay the takelonger. It needstime to fosterganuineparticipation;for the farmersto understand conceptof researchandto startto overcomethe problernsof unrealisticogectationsfrom the farmersgroups. of a group:the Therearetwo areasto considerwhentrying to evaluatethe effectiveness changeswithin the group andthe effect on farmersoutsidethe group. The changesthat haveoccurredwithin the groupsare significant. To take the example againofthe first goup, Daya: o All farmersarestartingto useorganicmatterrnanagement methodswhichtheywere not manure. use of mulches, compost and cattle usingbeforer Farmersaretakingresponsibility for reducingthe monkeypopulatioqwhichis allowing themto plantlargerrangeof cropsoutsidethe immediatelocalityof the village. o A largenumberof newcropshavebeenassessed, somecontinueto be grown others wererejected. . All farmersarestartingto diversifytheir cloveplantationareas,plantingsomecropsthat theyhadformerlylaughedat the ideaof (eg timberspecies,cinnamon,othertree crops). r Somefarmersarestartingto raisetheir own tree plantingmaterial. o Somefarmersareconcentrating their effortsin a singlearea(beforetheywereusing scatteredplots). r Startingto try som€new cropsin ricevalleys(eg sugarcane). o The farmersarenow 'researchminded'.

The Dayafarmersgroupis now startingto repaythe effort requiredto startit. Noneof the farrrershavedroppedout. Someadditionalfarmerswantto join, but the grouphavenot wantedto expand.Thefarmersreportthat otherpeoplein the villageusedto makefun of the farmerswhentheyhavetried new cropsor techniques, but the group farmersthink that the lastlaughwill be on them, We haveyet to evaluatethe impactof the farmerresearchgroupwork on farmersoutside of the groups. For mostofthe gfoupsthis wouldbe premature,but a studywill be done within the localareaof the first grouplaterthis yearto determinethe extentof the dispersal of new ideasandadoptionof newtechnologies.

PROBI.EM AREAS Expectation The dominantproblemwhentry to work with groups,or individualsfarmersfor that matter,is that of expectation.The expectationof what the farmerwill get out of participatingwith the projectand expectationof the relativerole of farmerandprojectin problemsolving. This issuemustbe tackledbeforoanyrealparticipationcanbe realised. The historyof strongstatecontrol,subsidies andtop-downapproachhasreinforced farmersunderstandingthat: . farmerswill be giveninputsor otherincentiveswhenco-operatingwith the government. r all problemsoutsidethe immediatecontrolof an individualwill be tackledby the governmentandnot by the community. . anYtrialsareseen&sbelongingor beingfor the benefitof the Ministry andnot for the farmer. Onefarmersaidthat he thoughtof the governmentasbeingmotherandfather. Therehasbeenlittle historyof realparticipatoryschemesinZamibu. Althoughtherehas beena moveby variousdepartments within the governmenttowardshelpingg.upr, this is oftencounterproductive groups asthe havebeenformingsolelyto attracttangUte resourees (tools,inputs)andnot througha desireto try andalleviateproblemsthrough joint action.Thetypesof groupsthathavebeenencouraged arethosefor joint production, whichoftenfail, Thereareno localNGOsworkingwithin the agriculturalsector. Conceptof research It takessometirnefor the farmersgroupsto reallyappreciate the conceptof research.The Dayagroup,for example,expectedthat theywould be ableto quicklypiantlargeareasof introducedcashcropsandtherebymakemoney, Whentheyweregivenplantingmaterials of newgropstheywantedto plantthemin a largeareabeforetheyknewhow thlesecrops would perform. The projectobjectiveswereto experimentwith a rangeof options, plantingsmallareasonlS andaomeup with improvedsolutions,

l0 WorHngin malti4isciplinary mode The project'sremitis development of cashcrops. The ideatwhenworking with FRGsis to havea multi-disciplinary teamof researcherVextensionists, whereeachmemberof the team cancontributedifferentareasof expertise.Cunentlythereis no suchgroupwithin Zanzibar,so therearea numberof alternatives: o involveothersections; . cover all areas; e limit the group exclusivelyto cashcrop problems, In practicewe areusinga mixtureof the three. We aredoingsomejoint work with other sections(particularlyforestry),we aredealingwith someproblemsoutsideof our direct remit andwe arealsorejectingsuggestions from the FRGswhicharecompletelyout of our capabilities(or suggestotheragencies whichmightbe ableto helpthem). Workingoutsidethe remit of the projectis not idealasit mayrezultin duplicationof effort, but on the otherhandthe divisionsbetweensectionsaresomewhatartificialandarenot a logicaldividebut relateto donoractivity. Most MALNR staffreceivegeneraltrainingin agriculture.Someget specialised trainingbut thenmaybe postedto anothersectionon returnfrom training. The structureof the Ministry haschangedmanytimeswithin the last l0 yearsandwill probablychangeagainin the future, Someproblemsarediffisultto classify,anexamplebeingthe problemof damageby monkeys.We felt that this problemis onethat relatesto the PlantProtectionDivisioq they saidthat monkeysareunderthejurisdictionof the districtgovernmentoffice,who in turn saythat it is an issuefor the Departmentof Environment,andso on. In reality,monkeys area majorconstraintin thearea,because thefarmersareunwillingto plantanycrop in the areasawayfrom the villageapartfrom cassava, rice andthe existingclovetreesor wild trees. Thereusedto be a governmentprogramrne for shootingmonkeys,but this progranrmeceasedmanyyearsego. Our input hasbeento: givetrainingon trappingandpoisoningmethods;to persuadethe villagersthat it is their problemandonly theycansolveit throughcommunityaction;andto advisethat if thereis largescaleplantingof cropswhicharesusceptible to monkeydamage, thendamagesufferedby eachindividualwillbe reducedandalsomonkeyhabitatwill be reduced.

OTHER APPROACHES BEING USED BY ZCCFSPAND MALNR The FRG programmeis the mainmethodusedby the projectfor doinglong term in{epth researchandderrelopm€nt of newtechnologies in a participativenunner,but it is by no meansthe only approachbeingusedby the projector the widerMinistry. Other ZCCFSPapprmches Differentapproaches havebeentried,manyof whichwerezubsequently modifiedor rejected,Onemethodwhichhasbeendroppedis the useof so-called'pilot trials'. This approachwasbasedon commercialdevelopment of singlecrop. Within a givenarea

ll farmerswereidentified(opento anyfarmer),potentialtradersfoundandplantingmaterials weredistributed. The mainproblemswith this approachwere: the difficultyto correctlyidentifya crop whichhadreal scopefor expansionto a commercialscale,the usualproblemsof €xpectationof the farmers,aswell asthe physicaldifficultyof providingsufFcientplanting materialof good qualityat the right time. The Forestrysub-Commission hadmoresuccess with this typeof approachwith development of Casarina production.This speciesfilled a realnichefor buildingpolesand couldalsobe grown on'free land'. Plantingmaterialwasgivenfreeto farmersfor many years,but is now soldbecausea realdernandhasdevelopedfor this species.

'

The currentproject progrunmeis betterintegratedandlesstop-down. o Indivi&ral crops. Somework continueswith individualcrops,for which there is considered to be realcommercialpotentialandsomeresearchquestionsremain.' Examplesarethe agronomicassessment of vanilla;top-workingandvarietyselectionfor mangoe3. o Networking.Of individualsin differentareasfor agroforestryinvestigations, t Topicalstudies.Whenmoredetailedinformationis needed.An exampleis the recent asses$ment of blackpepperproduction. o Policy andplanning. Promotionof farmingsystemsapproachwithin MALNR. Developmentof Crop Fact Sheets t Germplasmdevelopmenl.Introductionof new propagationpractices;identificationand selectionof superiorgermplasm. Dwelopmentof capacityfor villageplantingmaterial production. o Detailedsocio+conomicstudies.On rural households, their economicstrategiesand relatedmatters. t Marketing. Particularlyresearchanddevelopment of regionaltradingopportunities; facilitationof export(reductionof bureaucracy). Apprucheswithin other sectionsof M4LNR In the pastthe emphasis within the Ministry hasbeento do traditionalon-stationandonfarm researchusingresearcher designed- researcher managedtrials. More recentlyit has beenrecognised that this methodwasachievinglittle. Subsequently manyof the different sectionswithin the Ministry aremovingtowardsa groupapproach. Unfornrnately,the problemsof poor co-ordinationwithin MALNR hasresultedin different approaches beingusedby differentsections. Participatorygroupscanwork well if the contactpurposeor objectiveis clearandsimple (eg treeplantinggroups)or is left openandflexible(eg FRGs). The problemoccurswhen groupsareformedwith objectivesthat don't respondto a clearlyidentifiableneed, Thus the PlantProtectionSectionencountered problemswhenthey formedgroupsfor pest controlin certainfood crops. It soonbecameclearthatthe mainproblemsfacedby the farmerswereproblernsof marketinganddecliningsoil fertility andnot pestsper se,

t2 Potentialsolutionsto theseproblemswereto diversifycrops,agroforestryandother practices,ratherthandirectpestcontrolmethods. organicmattermanagement The Extensionsectionhasalsostartedworkingwith groups. The problemsthey havefaced include: the groupstheyareusingareoftenformedfor reasonsdifferentfrom the or for socialreasons);the field staffhave objectives(eg for attractingeconomicassistance, mode; andthereis a little experience of workingwith groupsin a flexibleparticipative whichcanbe extendedto farmers. lack of ready-made technologies THE INSTITUTIONAL CONSTRAINTS TO FURT'HER DEVELOPMENT OF TIIE FRG APPROACH IN ZANZIBAR AND PROSPECTSFOR THE FUTURE. Therearemajorinstitutional,economicandpoliticalchangesin the offing andit is very difEcultto predictexactlywhat futurethis FRG approachmayenjoywithin Zanabu. despitethe changesto the structureof the Thereis a continuedproblernof projectisation, Ministry designedto reducethis problem.Eachprojectimportsdifferentideasand approaches.It difficult for the governmentto controltheseprojectsbecausetheir is minimal, Evenlessresourcesare direstedto contributionto the runningexpenses sectionswhicharenot donor-supported,Also incentivesgivenwithin projectsin the form the projectisation.Allied to this is the lack of a of allowancesandtrainingexacerbates realisticoverallpolicyon agriculture. On the positiveside: r The FRG approachreducesthe dependency of the groupon the institution,through minimisingthe useof inputsandthroughparticipation. r The MALhIR staffbenefitftom prolongedcontactwith farmersandthe flexibletwo-way approachneededfor working with groups. r Changlnginstitutionsis a slow process,oneof the bestinfluenceson positivechangeis to demonstrate effectivemethodsfor agriculturaldevelopment. o FRGsarean effectiveway of ensuringthat farmers'constraintsandoppornrnitiesform the basisfor the work programme. r FRGsareprobablyoneof the cheapestmethodsfor developingappropriateagricultural technologies. Therearea numberof methodological issuesfor the future: r Will the learningcuryedrop oft? Is it mostappropriateto continuewith the same groupsor would it be betterto changeto newgxoupsaftera certainperiod. r Whennewideasor technologies aredeveloped,do you alterthe approachanddo more networkingor startextensiongroups?

LESSONS The mainlessonswhichcanbe drawnfrom the FRG progranrme: r It is importantnot to import methodswholesale.Methodsneedto be adaptedto suit the very differentcircumstances existingwithin differentcountries.

l3 Whendevelopinga newapproachit is bestto startslowly,so that lessonscanbe learnt or farmersareinvolved. This is andsuccesses built uponbeforetoo manyresources oftendifficult whenprojectshaveshortlife spansandunrealistictargets. FarmerResearchGroupsrequirea lot of timeneffort andthereforemoneyto setup. It is essential to selectthe groupandlocationvery carefully. FarmerResearchCrroupsarean excellentmethodfor developingrelevant,useable technologies.

,ji ' r':' '. ).:: ':'..i l.a i ',,:i

. , '. ] '."1

.::.:11

I l,r'i t'..

'.1,

i

::I .:i.; ' '.1.i t '{,-1 :

I

.:t:i.iJ t .;l

-t !,, '

, . !

.'l

.1

'. . :

,i

r

1 ;

t4

STIMMARYOF METHODOLOGY

Locationof FRG

o Basedon farmingsystemzones. o Villagewhichis representative. o Createshortlistof villagesmeetingcriteria,do rapid survey, selectone.

Groupselection andformation

o r r o r

o o

r r

Methodsfor workingwith Sroups

Mix of age,gender,family,wealth, l0 to l5 members. Representative of type of peopleusingland. Must bewilling to participateandexperiment. of possible Threeinformantsin villagemakesuggestions groupmembers, basedon givencriteriafor mix of members, makea combinationof the suggestions. Discussconceptsof researchandaimsof researchgroup Mni-PRA. Mixed groupof MALNR staff(6-8), makecheck list, visit eachfarmerindividually(discussionandvisit plots), meetingat endof field work to giveinitial feedbackto farmers,write reporton mainpointsandon individual farmers. Draw up areasfor research(MALNR statr) Discussresearchproposalswith farmers,modiff plan.

Identifyresearch themes Mix of : designed- Farmerimplemented) Formaltrials(Researcher (Farmerdesignedurd managed) Informalinvestigations Discussion on topicalissues. Trainingandstudytours, Minimisepaperwork. Topicalstudieswhenmoreinformationneeded. Minimalinputs,exceptwhenrisk to farmersis high.