Designing agroecological transitions

22 Pages·2017·1.86 MB·English

Designing agroecological transitions; A review
Michel Duru, Olivier Therond, M’hand Fares
To cite this version:
Michel Duru, Olivier Therond, M’hand Fares. Designing agroecological transitions; A review. Agron-
omy for Sustainable Development, 2015, 35 (4), pp.0. ￿10.1007/s13593-015-0318-x￿. ￿hal-01340332￿
HAL Id: hal-01340332
https://hal.archives-ouvertes.fr/hal-01340332
Submitted on 30 Jun 2016
HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est
archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents
entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non,
lished or not. The documents may come from émanant des établissements d’enseignement et de
teaching and research institutions in France or recherche français ou étrangers, des laboratoires
abroad, or from public or private research centers. publics ou privés.
Distributed under a Creative Commons Attribution| 4.0 International License
Agron.Sustain.Dev.(2015)35:1237–1257
DOI10.1007/s13593-015-0318-x
REVIEWARTICLE
Designing agroecological transitions; A review
MichelDuru1,2&OlivierTherond1,2&M’handFares1,2
Accepted:28May2015/Publishedonline:1July2015
#INRAandSpringer-VerlagFrance2015.ThisarticleispublishedwithopenaccessatSpringerlink.com
Abstract Concerns about the negative impacts of agriculture, (2) the pathway of the transition and (3) the re-
productivist agriculture have led to the emergence of two quiredadaptivegovernancestructuresandmanagementstrat-
forms of ecological modernisation of agriculture. The first, egies.Weconcludebyanalysingkeychallengesofdesigning
efficiency-substitutionagriculture,aimstoimproveinputuse suchacomplextransition,developingmulti-actorandmulti-
efficiencyandtominimiseenvironmentalimpactsofmodern domainapproachesbasedonacombinationofscientificand
farming systems. It is currently the dominant modernisation experiential knowledge and on building suitable boundary
pathway.Thesecond,biodiversity-basedagriculture,aimsto objects (computer-based and conceptual models, indicators,
developecosystemservicesprovidedbybiologicaldiversity. etc.)toassessinnovativesystemsdesignedbystakeholders.
Itcurrentlyexistsonlyasaniche.Herewereviewchallenges
ofimplementingbiodiversity-basedagriculture:managing,at Keywords Adaptivemanagement .Agroecology.
thelocallevel,aconsistenttransitionwithinandamongfarm- Ecosystemservices .Farmingsystem .Social-ecological
ingsystems,supplychainsandnaturalresourcemanagement. system .Sociallearning .Socio-technicalsystem .
Wediscussthestrengthsandweaknessesofexistingconcep-
Transdisciplinarity
tual frameworks developed to analyse farming, social-
ecological and socio-technical systems. Then we present an
integrativeframeworktailoredforstructuringanalysisofag- Contents
riculture from the perspective of developing a territorial 1.Introduction
biodiversity-basedagriculture.Inaddition,weproposeapar- 2.Biodiversity-basedagriculture:foundationsandchallenges
ticipatory methodology to design this agroecological transi- 2.1.Foundationsofbiodiversity-basedagriculture
tion at the local level. This design methodology was devel- 2.2.Challenges ofthe transitionto biodiversity-basedag-
oped to support a multi-stakeholder arena in analysing the ricultureanditsdesign
current situation, identifying future exogenous changes and 3.Anintegrativeconceptualframeworkforanalysingagricul-
designing (1) targeted territorial biodiversity-based turalsystemsatthelocallevel
3.1.Strengthsandweaknessesofthreeexistingconceptual
frameworks
MDuruandOTherondaretwofirstco-authors 3.2.1. Farming systems and associated innovation
systems
* MichelDuru
3.2.2.Social-ecologicalsystems
[email protected]
3.2.3.Socio-technicalsystems
* OlivierTherond
3.2. An integrative analytical framework of the local
[email protected]
agriculture
3.3. A local polycentric system of actors for promoting
1 INRA,UMR1248AGIR,31326CastanetTolosan,France biodiversity-basedagriculture
2 INPT,UMR1248AGIR,UniversitéToulouse, 4.Amethodologicalframeworkfordesigningtheagroecolog-
31029Toulouse,France icaltransitionofagriculture
1238 M.Duruetal.
4.1.The strongecologicalmodernisation ofagriculture:a etal.2011)orgeneticallymodifiedorganisms(Godfrayetal.
co-innovationprocess 2010).Itsmainobjectivesaretoreducenegativeenvironmen-
4.2. A participatory methodology for designing the agro- talimpactsandraiseproductionlimitsofproduction-oriented
ecologicaltransitionofagriculture agriculture.Duetotheleversofactionitisbasedon,wecallit
4.3.Afive-stepmethodology “efficiency/substitution-basedagriculture”,whileotherscallit
5.Methodologicalissuesandchallenges “ecologicalintensification”(e.g.Hochmanetal.2011).
5.1.Aniterativemulti-level,multi-domain,andmulti-actor Thesecondformaimstoenhanceecosystemservicespro-
approach videdbybiodiversity(LeRouxetal.2008;Zhangetal.2007).
5.2.Developmentofusefulscientificartefacts These ecosystem services depend on the degree of
5.3.Governancestructurestosupportsociallearning (agro)biodiversityatfield,farmandlandscapelevels(Kremen
6.Conclusion et al. 2012). Still focusing on the main lever of action, we
called it biodiversity-based agriculture, while other authors
call it “ecologically intensive agriculture” (Kremen et al.
1Introduction 2012) or“ eco-functional intensification” (Levidow et al.
2012a, b). Extreme forms of these two forms of agriculture
In developed countries, the production-oriented or (Table1,top)mustbeconsideredastwoendsofacontinuum.
productivist model of agriculture developed greatly after Importantly, in biodiversity-based agriculture, practices in-
WorldWarII.Itisbasedontheuseof“off-the-shelf”technol- creasingresourceuseefficiencyandrecyclingarealsoimple-
ogies(e.g.syntheticinputs,genetics)totrytoovercomeenvi- mented,ifnecessary.
ronmentalheterogeneityand,moreparticularly,effectsoflim- Some authors address production-oriented agriculture is-
iting and reducing production factors (van Ittersum and sueswithoutdistinguishingbetweenthetwoformsofecolog-
Rabbinge 1997; Caron et al. 2014). This model contributed icalmodernisation(Godfrayetal.2010;Wezeletal.2013).In
to a specialisation of territories based on their suitability for linewithHorlingsandMarsden(2011),wearguethatthisis
specific land uses (Foley et al. 2005). It also led to problematicforatleasttwomainreasons.First,biodiversity-
standardisationofproductionmethodsand,asaconsequence, basedagricultureintroducesaparadigmshiftinthevisionof
adecreaseintheplace-basedcognitiveresourcesnecessaryto agriculturalinnovationsandsystems,especiallyintheirobjec-
implement them. The desire to increase the health safety of tives and expected performances of agriculture (Caron et al.
agricultural production and its standardisation, even normal- 2014) and inhowtoapprehend and manageinteractionsbe-
isation,prolongedandstrengthenedthisprocess(Horlingsand tweentheenvironmentandproduction(Levidowetal.2012a,
Marsden2011;Lamine2011).Sincethe1980s,realisationhas b). Efficiency/substitution-based agriculture usually consists
emerged about the negativeeffects ofthisproduction model of incrementally modifying practices in specialised systems
onbiodiversity,ecosystemfunctioningandclimatechange,as tocomplywithenvironmentalregulations,e.g.EUdirectives
well as on product quality, human health and the increasing (DuruandTherond2014).Incontrast,biodiversity-basedag-
scarcity of fossil resources, water and natural phosphate de- riculture seeks to develop diversified cropping and farming
posits(MAE2005,IAASTD2009).Atthesametime,devel- systems(Fig.1a,b)andlandscapes(Fig.1c)todevelopeco-
opment of the concepts of sustainability, multi-functionality system services and, in turn, drastically reduce the use of
andecosystemserviceschallengedthemonolithiclogicofthis anthropogenic inputs (Duru et al. 2015). Second,
model(Huangetal.2015).Evaluationofenvironmentalim- biodiversity-based agriculture requires “adopt[ing] a more
pactsofagriculture,socialawarenessoftheseissueslinkedto creative eco-economy paradigm which replaces, and indeed
mediacoverage,redefinitionoftheobjectivesofagriculturein relocates,agricultureanditspoliciesintotheheartofregional
agriculturalpoliciesand,morerecently,theworseningofthe and local systems of ecological, economic and community
foodsecurityissuehavepromotednewwaystoaddresslim- development”(Marsden2012).Accordingly,developingthis
itationsofthismodelofagricultureandofthe“greenrevolu- type of agriculture requires implementing multi-level and
tion”(Godfrayetal.2010). multi-domain approaches at the local level (Caron et al.
Horlings and Marsden (2011) distinguish two forms of 2014; Horlings and Marsden 2011; Lin 2011; Marsden
ecologicalmodernisationofagriculturethataddresstheselim- 2012;Stassartetal.2012).
itations. The first,incontinuitywithproduction-oriented ag- Inrecentdecades,researchhasgeneratedknowledgeabout
riculture,isbased onincreasingresourceuse efficiency(e.g. key interactions between biotic and abiotic components of
fertiliser, pesticides and water), recycling waste or by- agriculturalecosystems(e.g.Lin2011)andbetweenbiodiver-
products of one subsystem in another (Kuisma et al. 2013) sity,ecosystemfunctionsandservices(Cardinaleetal.2012).
and applying sound agricultural practices (Ingram 2008) or However,thisknowledgeistoogeneraltobeuseddirectlyto
precision-agriculturetechnologies(Rainsetal.2011).Itisalso support design and management of diversified farming sys-
basedonreplacingchemicalinputswithorganicinputs(Singh temsandlandscapespromotingecosystemservices.Methods
Designingagroecologicaltransitions.Areview 1239
Table1 Featuresofthetwoparadigmsofecologicalmodernisationofagriculturethatrepresentthetwoextremesofacontinuum
Feature Efficiency/substitution-basedagriculture Biodiversity-basedagriculture
Mainaim Reducingnegativeenvironmentalimpact;“ecological Producingecosystemservicesforconservingresources;
intensification”ofagriculture “ecologicallyintensiveagriculture”
Paradigms Incontinuitywiththeproductivistagricultureparadigm: Breakswiththeproductivistagricultureparadigm:eco-
bio-economyandeconomyofscale economyandeconomyofscope
Agricultureisconsideredasaseparateandindependentsector Agricultureisconsideredashighlyinterdependentand
integratedinthelocalhuman,culturalandecological
ruralsystem
Environmentisconsideredthroughconcernsaboutresource Environmentisconsideredthroughitsnaturalandcultural
scarcity,wasteandpollution dimension(e.g.,craftsmanship,stewardship,farmingstyle)
Competitivenessisintheglobalmarket Competitivenessthroughsustainabilityandvalorisationof
naturalresources
Innovationnature Generictechno-sciencesolutions(“one-size-fits-all”)to Place-andspace-baseddiversifiedpracticesandfarming
improveefficiencyofinputsbasedongenetics,organic systemsbasedonadhocspatialandtemporal“planned”
inputs,mechanisation,precisionfarmingandrecycling and“associated”biodiversityandlocalknowledgesystems
(industrialecology)
Publicpolicy Top-downsteeringandregulation Adaptivegovernancebasedonlocalstakeholderparticipation
andfacilitatinglocalnetwork/consortiadevelopment,
knowledgesharingandcollaboration
Analysisfromtheangle Standardisedpracticesandspecialisedfarmingsystemswith Place-andspace-basedfarmingpracticesandsystemsbased
offarmingsystems asmallnumberofcropsbasedonuseofexternalinputs; onagroecologicalprinciples,withdiversifiedcropor
lineartop-downtransferofstandardisedtechnologies livestockinteractions,allowinggreatlyreduceduseof
externalinputs;farmpracticesalsodefinedaccordingto
objectivesatthelandscapelevel;collaborativeinnovation
Analysisfromtheangle Farmingsystemsconsumemanynaturalresources(NR); Farmingsystemsmanagenaturalresources(NR);adaptive
ofsocial-ecological decoupledmanagementofagriculturalsystemsandNR governanceandmanagementofNRincluding“associated”
systems leadingtoconflictsbetweenagricultureandotherresource biodiversityforimprovingbiologicalregulationatthe
usersandprotectorsoftheenvironment(including landscapelevel
institutions)
Analysisfromtheangle Globalisedandexport-andcomponent-basedmarket; Local-scalefoodsystembasedon“tightfeedbackloops”
ofsocio-technical industrialisation;privateandpublicfoodsafetyregulations linkingproducers,consumersandecologicaleffects,
systems andglobalisedstandardscreatinglock-ins(e.g.regulatory resilienttoexogenouschanges,withhighsovereigntyand
barriers)toplace-basedproducts;highlysensitiveto autonomy;powerinlocalagri-foodnetworksallowing
exogenouschanges;powerconcentratedinlargeretailers; creationofspaceforagentstobuildalternativeproduction
“free-trade” (promotionofniches);“fairtrade”andproductionof
regionalspecialities
are still needed to managethe context-dependent features of This paper proposes a transdisciplinary and multi-level
biodiversity-basedmanagementpractices (Duru etal. 2015). conceptualand methodologicalframework toanalyse condi-
Inparallel,socialsciences,focusingmainlyoneconomicsec- tions ofand designthe transitiontowards biodiversity-based
tors(e.g.transport,energy),providedframeworkstoanalyse, agricultureatthelocallevel.InSection2,wedefinethemain
a posteriori, socio-technical transitions (e.g. Geels 2002, characteristicsofbiodiversity-basedagricultureandkeychal-
2005) and methodological frameworks for “transition man- lengesoftheagroecologicaltransitionandexplainwhyitmust
agement”, i.e. to steer transition towards a normative goal, be managed at the local level. In Section 3, we analyse the
suchassustainability(Loorbach2010;Rotmansetal.2001). strengthsandweaknessesofconceptualframeworksoffarm-
These frameworks were not developed, however, to address ing,social-ecologicalandsocio-technicalsystemstostructure
characteristicsoftheagroecologicaltransitiononfarmsandto analysiswhenimplementingtheagroecologicaltransition.We
reconnectagriculturetolocalecologicalandsocio-economic presentanintegrativeanalyticalframeworkthatconnectsand
systems (Sutherlan et al. 2015; see Sections 2 and 3 for enrichesthepreviousthreeframeworksanddescribesthena-
details). Accordingly, research is still needed to develop sa- tureofthelocal(territorial)systemconcernedbythisagroeco-
lientandlegitimatemethodologiesthatallowstakeholdersto logical transition. In Section 4, we present a participatory
designthe“agroecologicaltransition”thatis,here,thetransi- methodologythatusesthisconceptualframeworktosupport
tion from productivist or efficiency/substitution-based to stakeholdersdesigningtheagroecologicaltransitionatthelo-
biodiversity-based agriculture (Caron et al. 2014; Kremen callevel.InSection5,weconcludebyexaminingkeyissues
andMiles2012). for researchers and stakeholders when implementing our
1240 M.Duruetal.
Fig.1 Developingaadiversifiedcroppingsystem,bfarmingsystem local processing chains for wood-based products. c Hedgerows in an
andclandscapethatstronglyincreaseprovisionofecosystemservices agriculturallandscapecanhelplimitsoilerosion,improvewaterquality,
requires adaptation in supply chains or in local natural resource and provide habitats for natural pest enemies and pollinators. It is
management.a Intercroppingcan beproblematic forseparatinggrains estimated that nearly 70 % of the 2 million km of hedgerows likely
ofcultivatedspeciesatharvestorinthefactory;thislogisticalproblem present in France in the early twentiethcentury have been destroyed.
limits adoption of such cropping systems. b Agroforestry, based on Coordination amongstakeholders is necessary to ensure development
growingwoodyvegetationwithinand/oraroundfields,requiresspecific oflandscapesadapted to deliverexpected ecosystemservices, such as
managementpracticesandmaterials;itsdevelopmentcanrequireusing biologicalregulationandwaterprovision
methodology and offer suggestions for further research. managed with the intention of producing, distributing and
Throughoutthetext,efficiency/substitutionandbiodiversity- consuming food, fuel and fibre and of the resources, infra-
basedagriculturesarecomparedtohighlightthechangesnec- structure, markets, institutions and people involved in these
essarytoimplementthelatter. functions. It thus “encompasses all the complexity a social-
ecological system can have” (Cabell and Oelofse 2012;
Koohafkanetal.2011;Marsden2012;Pretty2008).
2Biodiversity-basedagriculture:foundations Biggs et al. (2012), through their thorough review of sci-
andchallenges entific literatureand expert knowledge about conditions that
increase production and resilience of ecosystem services,
2.1Foundationsofbiodiversity-basedagriculture identify three properties of the social-ecological system to
manageandfourprinciplesforitsgovernance.Thethreekey
Biodiversity-based agriculture seeks to develop diversity in propertiesarediversity-redundancy,connectivityandstateof
cultivatedspeciesandgenotypesatmultiplespatialandtem- slowvariables.
poral levels to favour functional complementarities in re- & Diversity (taxonomic and functional) of biological (e.g.
sourceuseandbiologicalregulation(Caronetal.2014;Duru genes,species,ecosystems,spatialheterogeneity)andso-
etal.2015; Kremenetal.2012;Ostergardetal.2001).This cial(e.g.individual,socialgroups,networks,institutions)
“planned” diversity and the landscape-matrix heterogeneity entitiesandtheirlevelsofredundancydefinethepotential
promotes the beneficial “associated” diversity (Altieri 1999; for ecosystem services provision and adaptation. Func-
Duru et al. 2015; Kassam et al. 2012; Power 2010). These tionalredundancydeterminesthedegreetowhichreplac-
threeformsofbiodiversitydeterminethelevelsatwhicheco- ing one set of components with another can meet a bio-
system services are provided to the agroecosystem (e.g. soil logical or social function. Even though diversity and re-
fertility, pollination, biological regulations) and to society, dundancy are required to provide ecosystem services, a
whether marketed (plant and animal production) or non- threshold existsabove which diversity can leadtoa sys-
marketed(carbonstorage,controlofthewatercycle)(Zhang tem whose functioning is cumbersome, complicated and
etal.2007).Thelevelofagroecosystem’sinternalregulations lessefficientandhaslowadaptationcapacity.
depends on maintaining biologicaldiversity and thus the in- & Connectivitybetweenbiophysicalentitiesaswellassocial
tegrityofagroecosystems(Drinkwater2009;Koohafkanetal. entities determines circulation of materials, energies and
2011). Accordingly, when implementing biodiversity-based informationandthusthesystem’sperformance.Toomuch
agriculture, the challenge for farmers lies in designing, connectivity can favour massive propagation of initially
implementingandmanagingconsistentlydiversifiedcropping local perturbations or individualist behaviour harmful to
and farming systems and landscape structures that promote the system. For the biophysical dimension of
highlevelsofecosystemservices. agroecosystems,connectivitymainlydescribesthespatial
Suchchangesinagriculturalpracticesandincreaseddiver- relationships between landscape elements (e.g. patches).
sity in agricultural production involve multiple and interde- Theydeterminespeciesdispersionabilitiesbetweenhab-
pendent adaptations to managing the whole agroecosystem itats (Tscharntke et al. 2005). For the social dimension,
(Caron et al. 2014; Lin 2011). Here, like other authors, we connectivityencompassesthemultipledimensionsofso-
consideranagroecosystem tobecomposedofanecosystem cialnetworks.
Designingagroecologicaltransitions.Areview 1241
& Thestateofslowvariables(e.g.soilorganicmatter,water and farming systems that increase provision of ecosystem
resources, management agencies, social values) deter- services must be strongly adapted to a wide diversity of
minesdynamicsoffastvariables(e.g.fieldmanagement, production situations (soil-climate-biodiversity at field
water withdrawals, authorisation to access resources) in and landscape levels, constraints of natural resource man-
complex systems. The manner of middle- or long-term agement). In other words, they must be site-, space- and
management of slow variables thus determines day-to- time-specific (Caporali 2011; Caron et al. 2014; Duru
day,year-to-yearandlong-termsystemfunctioning. et al. 2015; Douthwaite et al. 2002; Godfray et al. 2010;
Koohafkan et al. 2011; Power 2010). Uncertainties about
Four governance principles may favour adapted manage- thenatureandperformancesofagroecologicalpracticesin
mentofthesethreesystemproperties: each farming system, even within each production situa-
& Understand the social-ecological system as a “complex tion of farming systems, may lead to strong risk aversion
adaptive system” characterised by emergent and non- by farmers (Fig. 1a, b). To support farmers in reducing
linear behaviour, a high capacity for self-organisation and managing these uncertainties, complementary re-
andadaptationbasedonpastexperiences,distributedcon- search is required to fill in the gap between generic sci-
trol and ontological uncertainties linked to incomplete entific knowledge and local knowledge. This need renews
knowledge ofmanagers. Suchenvisioning ofthesystem the way researchers can contribute to locally adapted ag-
to manage may help stakeholders to consider adaptive ricultural innovations (Caron et al. 2014; Duru et al.
governanceandmanagement(nextsection)asstructurally 2015).
necessary. Atthelandscapelevel,naturalresourcemanagement,e.g.
& Encouragelearningandexperimentationasaprocessfor managementofthelandscapestructuretopromotebeneficial
acquiring new knowledge, behaviour, skills, values or associated diversity (Fig. 1c), requires coordination between
preferences at the individual or collective levels, which stakeholders with different interests and adapted institutions
ultimately determines decisions and actions in situations (Brewer and Goodell 2010; Caron et al. 2014; Darnhofer
ofuncertaintyandthusmethodsformanagingthesystem. 2015; Duru et al. 2015). The need to anticipate and manage
Experimentation, particularly within the framework of cascade effects between organisational levels makes these
adaptive management, is a powerful tool for generating landscape-level practices complex to implement (Duru et al.
suchlearning. 2015;Gallowayetal.2008;WalkerandMeyers2004).
& Developparticipation:theparticipationofstakeholdersin At the supply chain level, development of new farming
governanceandmanagementprocessesfacilitatescollec- systems based on crop and animal diversity (e.g. crop
tiveaction,asdoestherelevance,transparency,legitimacy associations or intercropping, multi-animal species and
and,ultimately,acceptabilityofsocialorganisations,deci- breeds) and a decrease in farming system inputs may
sions and actions. It also allows actors to respond more cause economic, technological and organisational prob-
quicklytointernalorexternalchangesandpromotescon- lemsforsupplychains,particularlyduringcollection,pro-
structionofsharedrepresentationsandstrategies. cessing and marketing phases (Fig. 1a; Fares et al. 2011).
& Promotepolycentricsubsystemsofgovernancethatstruc- Furthermore, developing the four principles of gover-
turedebateanddecision-makingamongdifferenttypesof nance suggested by Biggs et al. (2012) may require pro-
stakeholders, at different levels of organisation, and in found changes in local institutions, i.e. rules of the social
different forms (e.g. bureaucratic, collective, associative, game, such as formal rules (including public policies),
informal). The basic principle of polycentric governance informal agreements and ways of doing things that struc-
is to organise governance systems at the spatial scale at ture human interactions and activity (North 2005 in
whichthe problems tomanage, orobjectivestoachieve, Darnhofer 2015). Fostering biodiversity-based agriculture
emerge.Thiscanincreasethesystem’sabilitytoproduce will require dealing better with the intricacy between ag-
the expected ecosystem services as well as its flexibility riculture and local socio-economical dynamics (Caron
and responsiveness. This organisation promotes et al. 2014; Marsden 2012). Accordingly, to develop
organisational diversity, redundancy and connectivity of biodiversity-based agriculture, innovations cannot be only
decisionandactioncentres. technological and technical, but also must be social, eco-
nomic and institutional. They cannot exist only at the
2.2Challengesofthetransitiontobiodiversity-based farm level but also at the levels of local supply chains
agricultureanditsdesign and natural resource management institutions. Thus,
implementing the agroecological transition requires con-
Key challenges at different levels arise when applying the sidering and integrating interconnected processes and
above principles to promote development of biodiversity- organisational levels in ecological systems, for example,
based agriculture. At the farm level, diversified cropping from populations and communities to the landscape
1242 M.Duruetal.
(Rabbinge and de Wit 1989), as well as in entire human- agroecological transitions, the challenge is then to design
technology-environment (or social-ecological) systems multi-domainandmulti-levelandmiddle-tolong-termchang-
(Pahl-Wostl et al. 2010). esinlocalagriculture.
Due tothe natureand level ofnecessary changes,the de- Oneofthegreatchallengesforresearchistodevelopop-
velopment of biodiversity-based agriculture cannot resort to erationalknowledgethatsupportsstakeholdersinstructuring
simpleincrementalagronomicinnovationssuchasincreasing thedesignofanagroecologicaltransitionfromproductivistor
the efficiency of production factors. It requires extensive re- efficiency/substitution-basedtobiodiversity-basedagriculture
definition of agricultural performances leading to break- atthelocallevel.Thisknowledgeshouldenablethedevelop-
through (radical) innovations and extensive redesign of the ment of capacities of local heterogeneous stakeholders for
agroecosystem (Caron et al. 2014; Duru et al. 2015; Hill designing and steering such transitions (Caron et al. 2014;
1998;Meynardetal.2012).Tomeetthechallengeofdevel- Darnhofer2015;Duruetal.2015).
oping an agroecosystem adapted to and interconnected with To reach this objective, research should develop transdis-
the local ecological, socio-economical and institutional sys- ciplinary and multi-stakeholder approaches integrating
tems,theprocessofinnovationmustbeconductedina“local contextualised scientific knowledge and local knowledge to
agricultural system of innovation” including a network of analyse the current organisation of local agriculture and to
interacting institutions, businesses and individuals (Klerkx designtheonethatisanticipated(Caronetal.2014;Darnhofer
and Leeuwis 2008). This “territorial” implementation of the 2015). To participate in this scientific challenge, this paper
innovationprocessisalsonecessaryforittobelegitimateto proposes both an integrative conceptual framework that al-
thesocialandculturalnetworks’valuesandtraditions(Caron lowsstakeholderstoanalysethemulti-dimensionalcharacter-
etal.2014;Darnhofer2015). isticsandissuesoflocalagriculture(Section3)andapartici-
Asefficiency-substitutionagricultureisstronglysupported patorymethodologytosupportlocalstakeholdersindesigning
bycurrent institutions and dominant socio-technical regimes a local agroecological transition and its governance
(next section), biodiversity-based agriculture has few oppor- (Section4).
tunities to develop strongly through emergent transition, i.e.
not planned and managed (Darnhofer 2015; Horlings and
Marsden 2011; Kremen and Miles 2012; Levidow et al. 3Anintegrativeanalyticalframeworkforanalysing
2012a,b;VanloquerenandBaret2009).Accordingly,ifstake- agricultureatthelocallevel
holders seek to develop this form of agriculture at the local
level,theymayhavetodevelop“transitionmanagement”,i.e. 3.1Strengthsandweaknessesofthreeexistinganalytical
a process of governance that seeks to steer or modulate the frameworks
dynamicsoftransitionsthroughinteractiveanditerativepro-
cesses among networks of stakeholders (Darnhofer 2015; Consideringthekeychallengesoftheagroecologicaltran-
Foxon 2011; Foxon et al. 2009). The challenge is then to sition presented above, we identified three conceptual
manage a “purposive transition”, i.e. “deliberately intended frameworks that consider the organisational levels and
andpursued…toreflectanexplicitsetofsocietalexpectations domains in which the necessary changes must occur: (i)
or interests” (Geels and Schot 2007). A key element of this thefarmingsystemframework,tostructureanalysisofthe
process is the “transition arena”: a relatively small group of organisation and dynamics of farm production systems;
innovation-oriented stakeholders that reached consensus (ii) the social-ecological system framework, to analyse
abouttheneedandopportunityforsystemicchangesanden- natural resource management; and (iii) the socio-
gageinaprocessofsociallearningaboutfuturepossibilities technical system framework, to analyse the dynamics of
and opportunities (Foxon et al. 2009). Here, mutual under- agriculturalinnovations.Inthissubsection,we presentthe
standing and collective development of sharedgoals and vi- main characteristics of these three frameworks and iden-
sionsoftheexpectedfutureandpotentialpathwaystoreachit tify their shortcomings in dealing with agroecological
are particularly at stake (Kemp and Rotmans 2005 in transition issues.
Darnhofer 2015; Loorbach2010).Transition managementis
seenasaformofparticipatorypolicy-makingbasedoncom- 3.1.1Farmingsystemsandassociatedinnovationsystems
plexsystemsthinking(Foxonetal.2009).Inagriculture,the
term“transition”isusedtoindicateareconfigurationofactiv- Inrecentdecades,manyanalyticalapproachesoffarmingsys-
itieswithinthefarm(Wilson2008;Lamine2011)butalsoasa temsweredeveloped(Darnhoferetal.2012).Approachesof
radicalchangeinagricultureorsub-sectorsmostoftenatna- Hendricksonetal.(2008)andDarnhoferetal.(2010)describe
tionalorhigherlevelsand,morerecently,atlocaltoregional wellthe two polar forms offarming systems thatimplement
levels. Such transitions may takeplace ina middle- tolong- efficiency/substitution-based and biodiversity-based agricul-
term time span (Darnhofer 2015). When dealing with ture according to the number and level of integration of
Designingagroecologicaltransitions.Areview 1243
productionsubsystemsaswellastheirperformancelevelsand technologiesandinfrastructurestomanageartificialandnat-
capacitiestoadapttochangesincontextsorobjectives. uralresources,aswellasacomplexecologicalsystemgener-
Mostfarmingsystemswithproduction-orientedagriculture ating these natural resources. Conceptual frameworks pro-
that implement efficiency/substitution-based practices are duced to analyse or model such social-ecological systems
specialisedandtendtohavefewcropsandpre-plannedman- (Anderies et al. 2004; Ostrom 2009; Sibertin-Blanc et al.
agement, which in some situations generates environmental 2011) allow the complexity of social, ecological and social-
issues(e.g.erosion,nitrateleaching,highpesticideuse).Their ecological interactions occurring in these systems to be
dynamics are based on genetically improved plants and ani- encompassed.Analysingthedynamicsofthesecomplexsys-
malsandtheacquisitionofhigh-performanceequipment(ma- tems is based on the concepts of resilience, adaptation and
chinery,buildings).Innovationonsuchfarmsismostlylinear transformation of system structure and/or functions (Folke
andtop-down,fromresearchtofarmers.Thistypeofinnova- et al. 2002, 2011; Folke 2006; Holling 2001; Walker et al.
tionprocess,mostpopularinthe1960s,isstilladaptedtothe 2006).Inmanysituations,naturalresourcemanagementprob-
lineartransferofstandardisedtechnologiessuchasprecision lems are associated with a failure in governance (Table 1).
agricultureortheimplementationofgeneticallymodifiedor- This failure is often linked to underestimating the changing
ganisms(Table1).Incontrast,productionsystemsthatimple- nature and complexity of the social-ecological system con-
ment biodiversity-based agriculture are based on multiple cerned(Folkeetal.2011;Pahl-Wostl2009;Pahl-Wostletal.
crops and/or enterprises that interact dynamically in space 2010).Thechallengeisthereforetwofold.Ontheonehand,it
andtime(e.g.cropdiversification,crop-livestockinteraction). consists of strengthening the adaptive capacities of gover-
Itallowsthemtobenefitfrommultiplesynergiesmadepossi- nancesystems,i.e.theirabilitytochangetheirmodesofaction
ble by interactions between components (Sanderson et al. and,ifnecessary,toimplementstructuralmodificationstobest
2013). Their high level of diversity can reduce impacts of address past or expected environmental or social changes.
variabilityinpricesorclimate(Darnhoferetal.2010).These Analytical and methodological frameworks developed thus
productionsystemsaremanageddynamicallybyperforming provideasupportforanalysingexistinggovernancesystems
annualorseasonaladjustmentstomakebestuseoftheoppor- and for leading the transition to “adaptive governance sys-
tunitiesthatpresentthemselves.Theyarebasedonactivities tems”(e.g.Pahl-Wostletal.2010).Ontheotherhand,itcon-
that(i)exploittheexistingpotentialofthefarmingsystemto sists of implementing “adaptive management” that aims for
adjustefficientlytoshort-termobjectivesandhazardsand(ii) continualimprovementinpoliciesandpracticesformanage-
provide the ability to respond to new opportunities in the mentofnaturalresources.Thismanagementstrategyisbased
middle and long terms. This mode of management allows onstructuredlearningabouttheeffectsofmanagementstrat-
systems to adapt more easily to a constantly changing envi- egiesthroughouttheirimplementation(Pahl-Wostl2009).Itis
ronment.Innovationonsuchfarmsisgenerallycollaborative anadaptive,deliberateanditerativedecision-makingprocess
(Klerkx et al. 2012). It is based on coordinated networks of whoseobjectiveistoconsiderandaddress(i)uncertaintiesin
stakeholdersthatseektoco-produceknowledgeandtechnol- thefunctioningofecologicalsystemsandeffectsofmanage-
ogies,possiblyassistedbyparticipatoryandtransdisciplinary mentpractices,(ii)imperfectionsandlimitsindetectingvari-
research(Knickeletal.2009). ationsinthestateoftheenvironmentundertheeffectofeco-
Duetothecentralfocusonthefarmlevel,analyticalframe- logicalprocessesormanagementactions,(iii)theimpossibil-
works of farming systems have two main limits in dealing ity of controlling and monitoring all management actions
withagroecologicaltransitionissues.First,inmostanalyses, within the social-ecological system and (iv) the intrinsically
the socialsystemconsideredisoftenreduced tofarmersand unpredictable character of certain ecological processes (Wil-
sometimestheirdirectadvisors,sothatsocialinteractionsbe- liams 2011).This adaptive management practice isoften as-
tween the farmer and other stakeholders of natural resource sociated with social learning objectives: mutual understand-
managementorsupplychainsaregenerallynotconsidered,or ing,viewpointsharing,collectivedevelopmentofnewadap-
if they are, it is disconnected from ecological processes that tivemanagementstrategiesforresourcesandestablishmentof
sustain management practices (Lamine et al. 2012). Second, “communitiesofpractice”(Armitageetal.2008;Newigetal.
interactionsamongfarmingsystems,thelandscapematrixand 2008; Pahl-Wostl and Hare 2004). Social-ecological system
naturalresourcemanagementarebarelyassessed(Benoîtetal. andadaptivemanagementframeworksareparticularlyapplied
2012). todealwithnaturalresourcemanagementproblems(e.g.wa-
ter:Pahl-Wostl2007)ortocollectivelyaddressenvironmental
3.1.2Social-ecologicalsystems sustainabilityissues(e.g.erosion:Souchèreetal.2010;nitro-
genemissions:Toderietal.2007).Inmostofthesesituations,
The management of natural resources at the local level (e.g. theyareusedatthelocalleveltosupportanalysisorto(co-)-
landscapeorwatershed)impliesasocialsystemcomposedof designsolutionsforreducingtheinvestigatedproblems.The
users, managers and governance institutions using social-ecological framework was also developed to analyse
1244 M.Duruetal.
formalrulesandinformalagreementsusedbyactorsfornat- policyeffectsonvarietyselectionandtrainingandextension
uralresourcemanagementandtheirconsequencesonsustain- servicestofarmers(VanloquerenandBaret2009).Itprovides
abilitymanagementissues(Ostrom 2005).Itwas adapted to an analytical framework for determining ways to strengthen
dealwithagroecosystemissuesinwhicheconomicandsocial regimeswhentheyarethreatenedorforidentifyingtheobsta-
structures depend greatly on agricultural activities (e.g. cles that prevent regime change, even though obvious
Schouten et al. 2012). Beyond its many assets, it has weak- underperformance is observed (Schiere et al. 2012). It also
nesses when dealing with agroecological transition issues. representsausefulframeworkforidentifyingemergenceand
Mainly, it was not developed to account for (i) agronomic stabilisationconditionsfornichesortheiraccesstothestatus
and organisational reasoning and constraints in farming sys- ofaregime(Geels2010).Inotherwords,ithelpsanalysehow
temsor(ii)necessarychangesinagriculturalsupplychains. alternativestothedominantsocio-technicalregimesmayde-
velop.However,whendealingwithagroecologicaltransition
3.1.3Socio-technicalsystems issues, it, like the social-ecological framework, does not ac-
countfor(i)agronomicandorganisationalreasoningandcon-
Thedynamicsoftechnologicalinnovationsandwaysofpro- straints within farming systems or (ii) issues and constraints
ducing goods within economic sectors or supply chains are linked to local natural resource management. Furthermore,
consideredastheresultofinteractionsamongthreelevelsof although the dynamics of niches are considered, in most
organisation:productionniche,socio-technicalregimeandthe socio-technical approaches, their governance is not analysed
landscape (Geels 2002). The former corresponds to formal indepth.
andinformalnetworksofactorsinwhichradicalinnovations
emergeandarenursed.Socio-technicalregimesarerelatively 3.2Anintegrativeanalyticalframeworkofthelocal
stableconfigurationsofinstitutions,techniquesandartefacts, agriculture
as well as regulations, standards and norms of production,
practices and actor networks. They support the evolution of As seenabove,one ofthe challengesofmanagingthe agro-
thedominantformsofproduction.The“landscape”,orglobal ecological transition is to design, in an integrated manner,
context, is the set of factors that frame interactions among technological, organisational and institutional (the rules of
actors,suchasculturalvalues,politicalinstitutionsandenvi- the social game) changes within farming systems, supply
ronmental problems. The socio-technical system integrates chains and natural resource management at the local level.
these three levels. Its dynamics are addressed by analysing Asshown,consideringeachofthethreeframeworkspresent-
the nurturing of new technologies into niches and edaboveseparately(Fig.2a)doesnotdealwithorsupportthe
transforming of the dominant socio-technical regimes under design of necessary multi-domain and multi-level changes.
thepressureofnichedevelopments,nicheincentivesandreg- The integrative conceptualframework presented below aims
ulatory changes coming from the “landscape” (Geels 2002, to structure analysis of the nature of the complex (adaptive)
2005;SmithandStirling2010). systemconcernedbytheagroecologicaltransition:localagri-
Currently in agriculture, the dominant socio-technical re- culture (Section 2.2). We developed this framework by
gimecorrespondstomodelsofhistoricalproduction-oriented hybridising and extending all three frameworks presented
agriculture and its modernisation through the efficiency/ above, considering their strengths and weaknesses (Fig. 2).
substitution-based approach. Key resources of the dominant This hybridisation was guided by previous attempts: Smith
regimes, such as infrastructure, production norms and stan- and Stirling (2010) hybridised social-ecological and socio-
dards and main market institutions, are historically adapted technicalapproaches,andDarnhoferetal.(2010)hybridised
tosupporttheseformsofagriculturalproduction.Thissystem farmingsystemsandsocial-ecologicalapproaches.
is dominant due its ability to create technological, Given that in each of the three frameworks the system is
organisationalandinstitutionallock-insthatensureitspersis- consideredtobecomposedoftwomaintypesofcomponents,
tence(Arthur1989;DavidandArthur1985).Thenichesare stakeholdersandresources,we firstconceptualiselocalagri-
alternativeproductionmodelsofvaryingstructure,whichco- culture as a complex system of interactions between them.
existina complementary orcompetitivemanner. Thisisthe After presenting the characteristics of the social system and
case of models promoted by groups of farmers who defend its resources, we examine stakeholder’s strategies and
alternative production methods or a specific ideology (e.g. technologies for managing these resources and address
Diazetal.2013),whichcanbeassociatedwithbiodiversity- the nature of interactions among the stakeholders of
basedagricultureexperiences. local agriculture.
This framework favours identifying learning-by-doing Importantly,weconsiderthatagriculturalstakeholdersin-
constraintsandfarmers’riskaversiontoadoptinginnovations volvedinthemanagementoffarms,agriculturalsupplychains
(Blazy et al. 2011; Cowan and Gunby 1996), supply-chain andnaturalresourcesarestronglyinterconnected.Weconcep-
lock-ineffects(Faresetal.2011;Lamineetal.2012),public tualiselocalagricultureasanentiresystemofactors(thesocial
Designingagroecologicaltransitions.Areview 1245
Fig. 2 From domain-based analytical frameworks to an integrated (another type of cognitive resource) and formal rules. These actors
framework developed to structure analysis of local agriculture and manage material resources of farming systems (e.g. water, soil,
design the agroecological transition at the local level. (a) Three key biodiversity, infrastructure, material, formal rules, workforce), supply
conceptual frameworks developed to analyse characteristics and chains (e.g. products, infrastructure, human and financial resources,
dynamics of farming, social-ecological and socio-technical systems. production standards and contracts, other formal rules) and natural
When applied to agricultural issues,even though domains considered resource management (e.g. water, soil, biodiversity, landscape,
by these frameworks overlap, they focus on different system entities infrastructure, material, formal rules). Actors involved in agricultural
anddynamics:farmresourcemanagement,dynamicsoftechnologiesin teaching,advising,developmentandresearchalsoarepartofthissocial
supplychainsand natural resource management.These three domains system. The tetrahedron reflects that local agricultural development
mustbedesignedandmanagedasawholetosupportdevelopmentofa dependsoninteractionsbetweenitsfourdimensions.Eachedgeofthe
biodiversity-basedagriculture.(b)Localagricultureasasystemofactors tetrahedron(doublearrows) corresponds to a diversity of information
managingthreetypesofmaterialresourcesystemsthroughinformation technologies used to manage material resources and concrete
technologies.Thesystemofactorsconsistsoffarmersandotheractors management processes within a variety of farming systems, supply
involved in supply chains and management of natural resources, with chains and natural resource management institutions. Dotted lines
cognitive resources (e.g. beliefs, values, individual strategies) and connecting the three material resource systems indicate that they are
whose behaviour is determined by informal norms and agreements stronglyconnectedandeveninteract
system)managingmaterialresourcesofthefarmingsystems, thissocialgame(Greenberg1990;KlerkxandLeeuwis2008;
supplychainsandnaturalresourcemanagement(Fig.2b,top). Vanclay 2004). Actors involved in agricultural teaching, ad-
Eachstakeholder can beinvolvedinonlyone ofthese man- vising,developmentandresearchalsotakepartinthissocial
agementprocesstypesor,likemanyfarmers,intwoorthreeof system; for example, this includes agricultural advisory ser-
them (Fig. 2b, top). In social-ecological and socio-technical vices and agricultural development associations. Stake-
systemsapproaches,thesocialsystemisrecognisedasakey holders’ actions depend on their cognitive resources. They
factor.Intheanalysisofsocial-ecologicalsystems,emphasis areintangibleassetsthatcorrespondtoknowledge(orbeliefs),
isplacedonusersofnaturalresourcesandgovernanceorga- values, objectives (or desires), strategies (or intentions) and
nisationsthatregulatetheuseofnaturalresources.Intheanal- informalsocialagreements.
ysisofsocio-technicalsystemsintheagriculturaldomain,itis Managementstrategiesoffarms,agriculturalsupplychains
instead an issue of coordinating production and marketing andnaturalresourcesatthelocallevelaimtoconserve/restore/
systemswithinsupplychains.Infarmingsystems,thefarmer protectorproduce/developcertainmaterialresources.Consid-
representsthesocialdimension(tenNapeletal.2011),buthe/ eringthethreekeysystemsathand,wedistinguishthreemain
sheisrarelyconsideredasasocialisedactorwhosebehaviour systems of material resources (Fig. 2, bottom): (i) material
is determined by his/her social interactions. Here, we claim resourcesofthefarmingsystemusedbythefarmerforagri-
that the farmer, like all other actors, takes part in the social culturalactivities;(ii)materialresourcesusedbystakeholders
gameand that his/her behaviour depends on the outcome of of the supply chain mainly for collecting, processing and