Ecology and Society10 - PDF document

Ecology and Society10 http://www.ecologyandsociety.org/vol10/iss2/art8/ Synthesis Strengthening adaptive capacity Foghorns to the Future: Using Knowledge andTransdisciplinarity to Navigate Complex Systems Georgina N. R. Cundill , Christo Fabricius , and Neus Marti 2 ABSTRACT. Complex systems are shaped by cross-scale interactions, nonlinear feedbacks, anduncertainty, among other factors. Transdisciplinary approaches that combine participatory and conventionalmethods and democratize knowledge to enable diverse inputs, including those from local, informal experts,are essential tools in understanding such systems. The metaphor of a “bridge” to overcome the dividebetween different disciplines and knowledge systems is often used to advocate for more inclusiveapproaches. However, there is a shortage of information and consensus on the process, methodologies, andtechniques that are appropriate to achieve this. This paper compares two case studies from Peru and SouthAfrica in which community-level assessments were conducted as part of the Millennium EcosystemAssessment, and explores the different conceptual models used to deal with scale and complexity, themethods adopted to deal with epistemology, and the different means of dealing with uncertainty in eachassessment. The paper highlights the conceptual and practical challenges encountered by each assessmentand discusses some of the conceptual and practical trade-offs involved in the selection of particularapproaches. We argue that a boat navigating between unknown shores may be a more appropriate metaphor Key Words: ecological assessment; community-based assessment; complexity; scale; epistemology;methodology; Millennium Ecosystem Assessment; complex systems; uncertainty; Peru; South Africa; casestudies; transdisciplinary research INTRODUCTION Understanding the relationship between people andthe environment requires that researchers on theground simultaneously navigate multiple worldviews (Gadgil et al. 2003) and complex social-ecological systems (Scheffer et al. 2001, Berkes etal. 2003) characterized by cross-scale interactions,nonlinear feedback, and uncertainty (Gundersonand Holling 2002). Transdisciplinarity is oftentouted as the answer to the difficulties involved inunderstanding such systems. The metaphor of a“bridge” is frequently used to typify the crossing ofthese disciplinary and knowledge divides, and thisconcept is often reflected in conference andpublication titles, e.g., Gunderson et al. (1995), aSociety for Economic Botanists conference held in http://www.atbio.org) , and a more recent Millennium Ecosystem Assessment http://www.maweb.org) There is, however, a lack of guidance andexperience in adopting integrated approachesinvolving different world views, and few academiccurricula address these challenges. In reality, thistype of complex systems research allows fordifferent conceptual and practical approaches,which can be confusing and disconcerting toresearchers. The recent Millennium EcosystemAssessment (MA) illustrated this point poignantly,as the papers submitted to the MA conference inEgypt bore testament. This paper is intended to gosome way toward dispelling the confusion socommon in this type of research, at least incommunity-level studies. By comparing two casestudies from Peru and South Africa, where 1Rhodes University, 2Autonomous University 10 http://www.ecologyandsociety.org/vol10/iss2/art8/ community-level assessments were conducted aspart of the MA, this paper explores the differentconceptual models used to deal with scale andcomplexity, the different methods adopted to dealwith epistemology, and the different means ofIn so doing, the paper highlights the conceptual andpractical challenges encountered by eachassessment, and the ways in which each team dealtwith those challenges. Throughout, the emphasis ison the multiple pathways that present themselves incomplex systems research, and the added challengeposed by the need to compare findings. The finalsection uses the two case studies to highlight someof the conceptual and practical trade-offs involvedin the selection of particular approaches. We arguethat the “bridge” metaphor might be too rigid andpredetermined for this poorly defined terrain; thispaper is intended to serve as a foghorn amid theconfusion that calls for integrated studies of The variable theoretical landscape of complexsystems It is useful at the outset to outline what is meant bycomplex systems research and some of the majorfactors that may lead to confusion. Theories aboutthe relationship between people and theenvironment influence the ways in which naturalresource management is understood and applied(Janssen 2002). Whereas early theories reliedheavily on a dichotomy between people and theenvironment (Malthus 1798, Meadows et al. 1972),more inclusive approaches emphasizing systemsthinking and human adaptation to environmentaland social processes have gained currency duringthe past century. This has brought with it principlesand ideas that emphasize complex system dynamics(Kay et al. 1999), linked social-ecological systems(Berkes and Folke 1998), nonlinear feedback atmultiple scales (Gunderson and Holling 2002), andresilience to change in social and ecological systemsEcosystem assessments such as the MA compelresearchers to deal with complex system dynamics,including but not restricted to nonlinear processes,uncertainty, emergence, cross-scale interactions,self-organization, novelty, slow- and fast-changingvariables, and a nested hierarchical structure(Walker and Abel 2002, Berkes et al. 2003, du Toit et al. 2004). Both natural and human systems exhibitcharacteristics of complex systems, and linkedsocial and ecological systems are increasinglyconsidered to be self-organizing, with a loosehierarchical structure (Gunderson and Holling2002) and various emergent processes. They arefrom one state to another (Scheffer et al. 2001).Natural resource managers and systems researchersface enormous challenges when confronting thisMany fields of research have contributed to therecognition of complex system dynamics in bothhuman and natural systems, which led to someconfusion and much debate within and between theMA assessment teams during working groupmeetings. Some of this confusion resulted from thevarying emphases placed on the dynamics involved.For example, although general systems theoryargues for an emphasis on connectedness, context,and feedback (von Bertalanffy 1968), chaos andcomplexity theory highlights the recognition of self-organizing behaviors in social and ecologicalsystems (Casti 1994, Kay et al. 1999). Evolutionarytheorists, on the other hand, might argue for anemphasis on feedback to avoid simple dichotomiesbetween human and natural systems (Wicken 1987,Adger 1999), whereas historical ecologistsemphasize history (Balee 1998), and postnormalscientists call for an emphasis on uncertainty andmethods to ensure the validity of conclusions ininherently complex systems (Functowicz andTherefore, researchers who take on the challenge ofa complex systems approach with Quixotean zealmay quickly become confused and frustrated by themany directions in which their analyses are pulled.This confusion arises not only from the existence ofvarious approaches and understandings betweendisciplines but also from attempts to communicatefindings between assessments and individuals whendifferent approaches have been used. The resultingchallenges are discussed later in this paper. Thereare two other major factors that cause confusion incomplex systems research: scale and epistemology. Scale Scale refers to the spatial, temporal, quantitative, oranalytical dimensions used by scientists to studyobjects and processes (O’Neil and King 1998, 10 http://www.ecologyandsociety.org/vol10/iss2/art8/ Gibson et al. 2000). Ecological and social systemstend to organize into strongly interacting clusters ofprocesses operating at similar spatial or temporalscales (Allen and Holling 2002). Consequently, anunderstanding of how a selected scale of analysismay influence the patterns observed, and thereforeinferences regarding causality, is essential inunderstanding interactions between human andnatural systems (Gibson et al. 2000, Munda 2000).However, despite recent comprehensive reviews ofscale (see, for example, Schulze 2000), the disparatetreatment that scale has received from the variousdisciplines makes it one of the most fundamentalmethodological challenges confronting researchers.For example, whereas systems ecologists mightargue that scale is an explicit consideration whenassessing any system (Levin 1992), geographerswould place the emphasis on spatial scale (Woodand Lakshmi 1993), historical ecologists ontemporal scale (Balee 1998), economists on1991), sociologists on interactions between scales(Coleman 1990, Scheffer et al. 2002), and politicalscientists on institutional and conceptual aspects ofscale (Ostrom and Hess 2000). This makes for aninconsistent theoretical landscape for researcherswho seek to become transdisciplinary in theirendeavor to come to terms with scale in complexConcomitantly, community-based assessmentsinevitably involve peer review by localcommunities, making the process even moreenter the equally varied theoretical landscape ofepistemology while still grappling with scale and Epistemology Epistemology is the philosophy of knowledge.More specifically, it is a field of research that seeksto come to terms with what we can know, and thestatus of knowledge about a particular reality (Jones2002). There is much disagreement about whetheror not reality can be divorced from socialexperience, and therefore whether it can beobjectively accessed by a particular knowledgesystem (Jones 2002). For this reason, debates aboutknowledge are often centered on power (Healy2003), because logically the system of knowledgethat is recognized as being able to tap into the “objective reality” holds greater sway than otherknowledge systems. This has lead to tension aboutthe validity of science vs. that of informal,There are various approaches and rationales bothfor and against the integration of scientificknowledge and informal or traditional knowledgein natural resource management. Whereas some,such as the social constructivists, argue fromontological perspectives (Milton 1996, Macnaghtenand Urry 1998, Jones 2002), others argue fromethical and even management standpoints (Gadgilet al. 2000, Berkes et al. 2003). Still others rejectthe very idea of integration and argue thatcommunicating between knowledge systems maylead to further marginalization of the nondominantknowledge systems concerned (Latour 1987,However, community-level projects are alreadyunderway worldwide (Barrett et al. 2001,Chakraborty 2001, Shackleton and Campbell 2001),and therefore knowledge systems are coming toheads, regardless of the arguments behind thesevaried perspectives. For this reason, methods andapproaches need to be found to conduct community-level research and assessments that pay attention tothe challenges outlined in this section. Comparative CASE STUDIES FROM SOUTH AFRICAAND PERU: MULTIPLE PATHWAYS Different navigational tools such as conceptualmodels, methods, and techniques were used in local-level assessments conducted in Peru and South http://www.maweb.org/ ). The Peruvian casestudy aimed to test and adjust tools andmethodologies for the Vilcanota MillenniumEcosystem Assessment, which was still under wayat the time of writing. The South African caserepresents work conducted in 2003 as part of theSouthern African Millennium Assessment. In thissection, we highlight the trade-offs that had to bemade when researchers endeavored to study human-ecosystem interactions in the current theoreticallandscape of complex systems. The MA case studieswere similar in terms of the conceptual frameworksused, the involvement of local people, and theincorporation of informal knowledge in information 10 http://www.ecologyandsociety.org/vol10/iss2/art8/ gathering. In all case studies, there was a directconnection between local people and ecosystemservices: all the communities needed ecosystemservices in their everyday lives. However, theydiffered widely in terms of their ecological, tenural, Both the South African and Peruvian case studiessought to answer the overarching question posed byAssessment 2003): what are the current conditionsand trends of ecosystems and the associatedimplications for human well-being? To answer this,the South African and Peruvian studies branchedoff in different directions (Table 2), payingtestament to the multiple paths available in complex Dealing simultaneously with scale andcomplexity As indicated in Table 2, both the Peruvian and SouthAfrican studies used the MA framework as a2003). The MA framework assumes a dynamicrelationship between people and ecosystems.Human and ecological systems are seen asinterconnected, with ecosystem change affectinghuman well-being and vice versa. The frameworkassumes that the relationship between ecosystemsand human well-being cannot be understoodwithout a consideration of multiple spatial andtemporal scales; it also recognizes cross-scaleinteractions. The mismatch between the scale ofecosystem processes and the scale of decisionmaking is considered to be a key reason for manyenvironmental problems. The model also introducesthe ethical problems encountered by researcherswho conduct local-level investigations into thesekinds of linkages, and acknowledges that differentknowledge systems may be more important whendealing with different scales of analysis. However,the MA framework alone does not do justice to thedynamism of the interactions between human andnatural systems at the local level (MillenniumTo overcome this shortfall in the framework, theSouth African local-level assessments used theadaptive renewal or “panarchy” model (Holling1986, Berkes and Folke 1998, Gunderson andHolling 2002) as a conceptual guide to deal withscale and complexity simultaneously and to address the shortcomings of the MA framework for local-level purposes. This model integrates the ideas (1)of fast- and slow-moving emergent features ofcomplex systems, borrowed from ecologicaleconomics; (2) of temporal scale, borrowed fromgeography and environmental history; (3) of verticalscale, borrowed from the political sciences; and (4)that micro-level phenomena affect macro-levelprocesses just as much as the macro affects the local,The premise of the model is that both natural andhuman systems undergo cycles of organization,collapse, and renewal. The adaptive renewal cycleemerged from earlier discussions around multiplestable states (Holling 1973) and incorporates keyprocesses underpinning resilience (Walker et al.2002), institutional memory (Berkes and Folke2002), disturbance (Gunderson 1999), adaptation,and novelty (Berkes et al. 2003). Thus, the modelprovides a useful navigational tool (Berkes et al.2003) for conceptualizing and assessing the self-organizing characteristics of complex adaptivesystems (Kay et al. 1999), historical processes(Balee 1998), context and feedback (vonBertalanffy 1968), and the evolutionary linkbetween institutions, culture, resources, and thephysical environment (Adger 1999). The model alsoacknowledges the adaptive capabilities of localIn the Peruvian case study, the challenge was toincorporate the traditional Andean cosmologyframework and principles (TACFP) and use themalongside the MA framework so that the assessmentcould be based essentially on indigenousunderstandings of ecosystem change. Complexadaptive hierarchical system (CAHS) theory (Allenand Starr 1982, Lowrance et al. 1986, Giampietro1994) was used as a starting point to assess thefeasibility of using TACFP to assess multiscaleprocesses. CAHS theory has three discerning1. Hierarchy as a system of filters. Society andits rules act as a system of constraints thatbuffers the intensity and frequency of changes2. Holons and the dual nature of hierarchicalsystems. A component of the hierarchy, theholon consists of smaller parts that are lowerin the hierarchy. It maintains its own integrity 10 http://www.ecologyandsociety.org/vol10/iss2/art8/ Table 1 Case study Land tenure Livelihoods Relationship with naturalresources Biome/vegetationtype South Africa, EasternCape Province.Community names:Qongqota andMachibi(27° 28’ E, 33° 00’ S) Communal, mosthave been subjectedto resettlement atsome point in thepast Animal husbandry isvery important, arablefield cultivation is on thedecline, most familieskeep home gardens.Other sources of incomeinclude collection ofwild resources, statepensions and welfaregrants, migrant labor. isiXhosa identity is stronglyfounded on interaction withancestors. There are stronglinks between the spiritualworld and environmentalfeatures such as pools, intactforests, medicinal plants andancestors' graves. Thisrelationship has becomeincreasingly strained throughinteraction in the formaleconomy and associated land-use change. Valley Bushveld/xeric Succulentthicket South Africa,Richtersveld NationalPark( 28º15’ S, 17º 10’ E) Communal tenure,those who liveoutside the NationalPark hold a 30-yrlease allowingaccess for grazing.The originalinhabitants remaininside the park. Seminomadic pastoralists(mainly goats andsheep); other sources ofincome includecollection of wildresources, state pensionsand welfare grants,employment in localmines, migrant labor. Fuel wood is the primaryenergy source. Bushmeat, fish,and wild fruits supplementdiets. Natural streams andwatering points are central toall pastoralist activities. Succulent Karroo Peru, South AndeanMountain Chain,Cusco region.Community names:Sacaca, Amaru, Paru-Paru, Cuyo Grande,Chawaytiri, andPampallaqta(between 13º30’ E,70°31’ S and 14º20’E, 71°21’ S) Communal afteragrarian reform inthe 1960s Polyagriculture athousehold, community,and landscape levels;animal husbandry;collection of wildresources; barterinterchanges; handicraftsand fabrics; migrantlabor Close relationship with naturalresources through traditionalAndean cosmology, whichlinks natural resources,processes, and services withspiritual beliefs and humanlandscape management andlocal practices Mountain ecosystems,high diversity ofecological conditionsfollowing analtitudinal pattern.Puna, Suni, andYunga bioculturalzones. while simultaneously supporting the otherparts of the whole, on which it depends for its3. Arbitrariness. Investigators can arbitrarilyselect a particular window of observation toisolate, describe, and simplify a part of asystem as an independent entity. In assessinga social-ecological system and predicting itsselect the windows of observation carefullyTACFP has many similarities to CAHS. TACFPidentifies the existence of three main hierarchical systems containing all of the ecological, social, andcultural processes of life. These systems areKaypacha or the real world, HananPacha or theworld of sacred features and divinities, andUkuPacha, the world of dead people or ancestors(Milla 1983). Each of these worlds could be seen asa holon of the whole system made of smallerspace management principles were widely studiedby anthropologists in the 1970s (Mayer and de laCadena 1989, Murra 1975). These principles, whichtake place and are implemented by each social unit,i.e., person, household, community, ethnic group,region, at each scale. These cultural conceptions of 10 http://www.ecologyandsociety.org/vol10/iss2/art8/ Table 2. Summary table of approaches followed by the South African and Peruvian case studies. MA stands Case study Conceptual models usedto deal with scale andcomplexity Methods for bridgingepistemologies Approaches for dealing with uncertainty South Africa MA frameworkAdaptive renewalNested institutional and ecosystem change Summaries of literatureForum theatrePRA workshopsCombination of local and GIS mappingVegetation surveysWater quality testingHousehold surveysKey informant interviews Triangulation through:Historical recordsReview of existing literatureCombination of various qualitative and quantitative methodsCommunity validation of scientific knowledgeCommunity validation of its own knowledge through feedback meetingsScientific validation of local knowledge through surveys and literature Peru Complex adaptive hierarchical systemsTraditional Andean cosmologyMA framework Arariwas Acceptance of uncertainty and variability as inherent property of the Andean system and the research processApplication of traditional space management principles to methodologies and toolsTriangulation through:Review of existing literatureHistorical researchAnalysis and use of customary practices and normsIntegration of traditional and occidental taxonomic systems for space and resources characterizationOral traditional registration of knowledgeCombination of quantitative and qualitative information space, processes, and endogenous principlesconstituted the roots of the assessment strategy in Divergent methods for dealing withepistemology Both the Peruvian and South African assessmentssought explicitly to include different knowledgesystems and world views in the assessment process.The reasons behind this were, however, different inthe two cases. Whereas the South African local-level assessments came predominantly from theecosystem management school (see, for example,Berkes et al. 2000, Gadgil et al. 2003), the Peruvian study emphasized the ontological and ethicalaspects of systems assessments (see, for example,Callicot 1994, Milton 1996, Macnaghten and Urry1998), highlighting the need to respect and empowerConceptual models help researchers to navigatetransdisciplinary research in complex systems, butlocal assessment practitioners require innovativemethods and techniques if they hope to bridgeepistemologies on the ground. The probleminvolves not only researchers communicating withand understanding informal knowledge, but also theadditional difficulty of communicating theinformation thus received back to other scientists ina way that makes sense and does not further 10 http://www.ecologyandsociety.org/vol10/iss2/art8/ marginalize the less powerful informal knowledgeIn the South African assessment, learning andmemory were considered to occur and to be storedat the level of the group, i.e., a social constructivistapproach, and therefore the techniques and methodsused to bridge knowledge systems were consensusbased. Realizing that the methods used during aninvestigation also have ethical implications (Munda1999), researchers used a combination ofparticipatory research techniques that incorporateda range of visual, verbal, and interactive techniques.These included forum theatre, focus groupworkshops, and interviews (Borrini-Feyerabend1997), semistructured interviews with keyinformants (Pretty et al. 1995), and a range ofparticipatory rural appraisal techniques (Chambersbetween knowledge systems at the village level butwere less helpful in communicating findings backto decision makers. For this reason, and also toimprove confidence in the data, qualitative findingswere validated through more conventional methodssuch as household surveys, vegetation surveys,mapping exercises were also conducted with the aidof geo-referenced orthographic photographs ratherthan free-hand mapping. The maps were thendigitized, and land-use maps based on informalknowledge could thus be presented in aA final aspect of the South African local-levelassessments involved the dissemination of thecombined informal and scientific knowledge backto the communities involved. This was achievedthrough scenarios (van der Heijden 1996, Petersonet al. 2003) or story lines representing a range ofplausible futures. These scenarios were aninterpretation of information already gathered at thelocal level, combined with national-level data onScenarios were first developed at broader regionallevels (Bohensky et al. 2004) and then interpretedby the researchers for the communities in question.These interpretations were based on the researchers’understanding of the local-level processes in eachcommunity. A development acting group whospoke the same language as the communities thenturned these interpretations into simple story lines and later into dramas. These dramas were performedfor the community and then amended throughfeedback from participants to demonstrate howbroader economic and political changes might playout at the village level. Through a videodocumentary and written reports, this informationwas then presented to other scientists. Scenariostherefore provide just one example of howinformation generated at broader scales can betranslated to local-level actors in a way that makessense to them, and how local responses can betranslated back to scientists working at coarserIn Peru, on the other hand, the ethical componentof complex systems research underlay the entireprocess, emphasizing the need to respect the localand traditional rights of the communities involvedand to empower them. The inherent complexity ofthe system forced the assessment to select the mostrelevant services and processes to be assessed. Forwhom these services and processes were relevantwas the key question in this task. Therefore, localpeople identified the processes and services to beassessed through debates in which scientific andtraditional information was cross-checked. Anoverview committee was established with localauthorities to control the entire process. Once theservices and resources had been selected,methodologies and tools were designed and adaptedwith local technicians who were identified andlegitimated by the local communal assemblies thatThe Peruvian assessment dealt explicitly with acrucial question that arises from the issue ofparticipation: who has the power to impose theresearch process? Within this question, are thequestions of who decides on the key goals,methodologies, and tools to be applied and whoidentifies the stakeholders and social groups toparticipate? To deal with these questions, thePeruvian assessment treated local experts notsimply as stakeholders who were asked toparticipate, but as leaders and specializedresearchers with the right to raise relevant researchissues and to suggest appropriate methodologies andThe main research strategies used with localtechnicians included resource surveys, participatorymapping, local debates within learning groups,endogenous video reports, arariwas or traditionalcitizens’ juries, traditional resources registers for 10 http://www.ecologyandsociety.org/vol10/iss2/art8/ priority resources such as potatoes and medicinalplants, household surveys, and in-depth interviewswith people who had specialized knowledge aboutkey resources. The information thus generated wasshared and validated with the communities at twodifferent levels: (1) learning groups composed ofvolunteers from all age and sex groups and (2) localcommunal assemblies that involved the entire TRADE-OFFS INVOLVED IN A COMPLEXSYSTEMS APPROACH The inadequacy of the literature dealing withresearch processes in complex systems research(Campbell 2002) means that researchers inevitablyenter uncharted waters. Two key sets of trade-offswere identified in the Peruvian and South Africanassessments. The first set related to the advantagesand disadvantages of selecting a particular set ofconceptual models and research approaches.Although this provided a useful organizingframework for researchers working in differentcontexts, it necessarily influenced the questionsasked, the selection of issues to be addressed, andthe interpretation of results. The second set of trade-offs related to the advantages and disadvantages ofdemocratized expertise, the confrontation ofuncertainty, and the resultant dynamic and Trade-offs in the selection of approaches andconceptual frameworks In terms of research approaches, the three mostimportant trade-offs were related to the followingissues: (1) predesigned frameworks are convenient,but they eliminate alternative perspectives; (2)transdisciplinary assessments are inclusive andcomprehensive, but their research outcomes areoften superficial; and (3) predesigned questionsmake for more rigorous and comparableassessments, but they place constraints on reflexivelearning processes. These trade-offs are discussedPredesigned frameworks vs. the loss of alternativeperspectivesThe South African case study incorporated informalknowledge, predominantly from a natural resourcemanagement perspective, as opposed to the ethical and ontological approach adopted by the Peruvianstudy. The latter approach proved very useful inidentifying underlying causes of change, adaptiveprocesses at the local level, and nonlinearrelationships between processes and outcomes atdifferent spatial and temporal scales. The use ofthese frameworks also improved the legitimacy andvalidity of the local assessments in the eyes ofscientists and most policy makers. However, thesemodels and relationships represented particularworld views developed outside of the local contextto identify processes deemed important byscientists. Therefore, the research team had tocompromise between the use of local cosmologiesto understand changing human-environmentrelationships and the a priori identification of issuesrelevant to scientists. The negative trade-off wasthat the process was less participatory than thatadvocated by the proponents of community-basednatural resource management (Fabricius et al. 2004)and possibly, in the eyes of the local people, lessInclusiveness vs. superficialityWorking across disciplines is indispensable whendealing with complex multiscale systems(Campbell et al. 2001). Local management systemsand resource use patterns know no disciplinaryboundaries, and the drivers of social-ecologicalsystems are ecological, biophysical, geographical,climatological, historical, political, and economic.A transdisciplinary, inclusive approach allowed usto appreciate and record the many factors thatinfluence such systems. The negative trade-off ofthis, however, was the sacrifice of a more detailedunderstanding of individual key processes. Thistrade-off was linked to the tight time frame thatcomes hand in hand with multiscale assessmentssuch as the Millennium Ecosystem Assessment(MA). In the South African case study, researchersbiological surveys, and historical analysis.Although this allowed for a broad and inclusiveanalysis of key processes and the linkages betweenthem, it was impossible to attain an in-depthunderstanding of the respective processes in thetime frame involved. Some of these processes, suchas the relationship between diversity andproductivity in natural and anthropogeniclandscapes (e.g., Salmon 2000), the social andinstitutional impacts of large-scale economic 10 http://www.ecologyandsociety.org/vol10/iss2/art8/ remain poorly understood but are probably criticalRigor vs. reflexive learningTime constraints introduced some urgency into theassessment to ensure prompt and rigorous deliveryof results, and this affected our ability to facilitatereflexive learning in participatory processes. Thepredesigned nature of the MA helped us to get ourassessments off the ground rapidly. Althoughresearchers were given a great deal of freedom interms of their approach to the assessments as wellas the techniques used, there was little time in theassessments for participatory learning. Forexample, when dealing with scenarios, the timeconstraints allowed only for community responsesto the possible futures. No time was available toexplore the community’s own scenarios, to assesshow feasible or appropriate the communityresponses were, or to evaluate responses to allowfor critical thought. Although all of the methodswere, to some degree, useful to both researchers andlocal participants, none of them promoted in-depth Trade-offs in democratizing expertise Participation vs. the ability to plan and predictThe Peruvian case study provides a useful exampleof trade-offs related to the inability of scientists toplan and predict community-based researchprocesses. In line with the ethical considerationsthroughout the Peruvian research process, a greatdeal of control over the research questions andmethods was devolved to local participants. Thestudy therefore gained considerably by achievingdesired levels of participation and therebyintegrating indigenous cosmologies into theHowever, the devolution of control over researchgoals and processes raised the expectations of allthe parties involved (see, for example, Fabricius etal. 2001) and hampered the ability of the scientiststo plan the research process. During the Peruvianresearch process, three major expectations arose atthe outset: (1) from the research team, expectationsrelated to the relevance, content, and consistency ofthe final results; (2) from the local communities,expectations related to immediate short-termbenefits; and (3) from politicians and authorities, expectations related to the legitimacy of their role,The researchers decided on the goals andmethodology at the outset of the process. This isnormal in scientific assessments, and even aprerequisite when dealing with externally fundedresearch. Therefore, the researchers had certainexpectations with regard to the results and theirvalidity. However, despite the shared understandingamong researchers of the sui generis nature of thework, uncompleted activities or information thatwas perceived to be less precise provoked fuzzyuncertainty and even distrust on the part of someresearchers. For example, it led to questions aboutthe influence of the facilitators, the methodologiesapplied, and even the usefulness of relying oninformal knowledge. In the end, the frequency ofthe situation just described forced researchers toredefine the process itself so that the data could bevalidated in the scientific arena. Fortunately, theadaptive nature of their approach gave them theExpectations from the community relatedpredominantly to the short-term benefits that wouldaccrue to key individuals. Local participantsexpected to become more respected in theircommunities and to win socio-political powerthrough their participation in the project because ofAs the assessment progressed, doubts emergedregarding the true interests of local participants.Some tried to satisfy personal interests, leading tolocal conflicts among local participants, betweenlocal participants and the rest of the community, andA similar situation emerged with the politicians andentered the process with expectations aboutlegitimizing their competencies and control at thelocal level. This led some of them to intervene inthe assessment process, whereas others presentedthe process as their own. In both cases the aim wasto maintain control over local processes. Thisrepresented a major trade-off in the Peruvian study.Expectations had to be confronted to achieve thedesired levels of participation. Each groupembarked on the assessment process with existingexpectations, and the final outcome differed forevery group. Therefore, the actual research processwas only identifiable in hindsight, despite efforts byall parties to shape the process at the outset. The 10 http://www.ecologyandsociety.org/vol10/iss2/art8/ adaptive approach established at the beginning wasessential to allow for an experimental process thatmade it possible for those involved to navigate theConfronting uncertainty vs. simplificationBoth assessments confronted uncertainty as aninherent property of complex systems and ofknowledge systems that cannot be tested usingtraditional scientific techniques. Thus, a significanttrade-off was made between, on the one hand,simple data that lent itself to validation and, on theother, information that was more difficult to teaseapart but which provided a more realistic reflectionof the relationships between drivers of change atbroader spatial and temporal scales and realities onTo deal with the uncertainty this approachgenerated, the teams sought to validate bothscientific and informal knowledge. In the SouthAfrican assessment, informal knowledge andscience were treated as equally powerful sources ofknowledge. Informal knowledge was thereforesubjected to scientific cross-validation bycomparing it to the results of quantitative analysesof, e.g., landscape change and to relevant literature.To deal with uncertainty and fuzzy data in Peru, theconcepts of traditional space management wereapplied to methodologies and tools. Specialemphasis was placed on methods and tools thatencouraged a diversity of responses, rather than ontrying to identify a few methods that wouldeliminate uncertainty (Martínez-Alier et al. 1998,O’Neill and King 1998). Literature reviews andhistorical research were integrated with theinterpretation of customary practices and norms,with occidental taxonomic systems. Finally, oralinformal knowledge was registered through the useof video and then analyzed and validated by thecommunities concerned. In this way, the uncertaintyresulting from a systems approach anddemocratized expertise was confronted and dealtAlthough scientific rigor is a significant trade-off inlocal-level assessments, both studies sought variousmethods to deal with the uncertainty thus created(Table 2). This process of validation also had thepositive effect of encouraging deliberative andreflexive learning as local participants were forcedto debate responses and opinions. This does, to some extent, make up for the trade-offs discussed CONCLUSION The metaphor of bridging different knowledgesystems assumes that there are two known shores:science and informal knowledge. This paper hasdemonstrated that this is scarcely the case and hashopefully sounded a foghorn or an early warning toassessment teams and researchers alike. Multipleshores exist, both within and between the sciencesand informal knowledge. We argue that themetaphor of a boat navigating between unknownshores is a more appropriate metaphor for local-level assessments that adopt a systems approach inAs the Peruvian and South African cases havedemonstrated, complex systems research allows fordifferent conceptual and practical approaches, andthere are various trade-offs involved in the selectionof these approaches. By comparing case studiesfrom South Africa and Peru, this paper has exploredthe different conceptual models that can be used todeal with scale and complexity, the differentmethods adopted to deal with epistemology, and thedifferent means of dealing with uncertainty. In sodoing, the paper has highlighted the conceptual andpractical challenges encountered by eachassessment, the ways in which each team dealt withthese challenges, and the trade-offs involved inIf local-level studies that embrace a complexsystems approach are to add value to analyses takingplace at coarser scales, and if these studies are to bemeaningfully compared, then researchers who seekto become transdisciplinary must identify thevarious directions in which their analyses may bepulled. Both case studies discussed in this paperwere forced to find practical and conceptual toolsthat enabled them to meet the needs of their localcontexts while at the same time meeting the needsof the broader ecosystem assessment. These toolscame from the proverbial toolbox that exists forcommunity-level research. The tools themselvesexist because an agreed-upon framework has not yetIn light of the challenges and trade-offs discussedin this paper, an effective common frameworkwould need to be (1) open enough to be 10 http://www.ecologyandsociety.org/vol10/iss2/art8/ understandable and legitimate to differentdisciplines and world views; (2) flexible enough tointegrate and address different indigenouscosmologies and therefore allow space forknowledge and information from various sources;(3) broad enough to consider multiple spatial anddynamism, adaptability, nonlinearity, lumpiness,uncertainty, and variability; and (4) capable ofdealing with both rigorous and fuzzy data. This is atall order by anyone’s standards, and dauntingenough to make most shy away from the complexityinvolved. Consequently, in the absence even ofconsensus on whether a single framework addsvalue or provides yet another means to extendscientific networks, investigators need to continueexperimenting and comparing their results. Manymore comparative syntheses of divergent casestudies and well-planned adaptive experimentation Responses to this article can be read online at: http://www.ecologyandsociety.org/vol10/iss2/art8/responses/ Acknowledgments: The financial contributions of the MillenniumEcosystem Assessment in sponsoring an exchangebetween South Africa and Peru during 2003 isgratefully acknowledged, as is the generosity andassistance of Alejandro Argumedo and all of the staffat Asociación Los Andes in Cusco, Peru. The viewsexpressed in this paper are those of the authors anddo not necessarily represent the views of AsociaciónLos Andes or the Millennium EcosystemAssessment. LITERATURE CITED Adger, W. N. 1999. Evolution of economy andenvironment: an application to land use in lowlandEcological Economics31Allen, C. R., and C. S. Holling. 2002. Cross-scalestructure and scale breaks in ecosystems and otherEcosystems Allen, T., and T. Starr. 1982. Hierarchy:perspectives for ecological complexity. UniversityBalee, W. 1998. Advances in historical ecology. Columbia University Press, New York, New York,Barrett, C., K. Brandon, C. Gibson, and H.Gjertsen. 2001. Conserving tropical biodiversityBioscience51Berkes, F. and C. Folke, editors. 1998. Linkingsocial and ecological systems: managementpractices and social mechanisms for buildingresilience.Berkes, F., J. Colding, and C. Folke. 2000.Rediscovery of traditional ecological knowledge asadaptive management. Ecological Applications 10:Berkes, F., and C. Folke. 2002. Back to the future:ecosystem dynamics and local knowledge. Pages121-144 in L. Gunderson and C. S. Holling, editors.Panarchy: understanding transformations in humanand natural systems. Island Press, Washington D.Berkes, F., J. Colding, and C. Folke, editors. 2003.Navigating social-ecological systems: buildingresilience for complexity and change. CambridgeBohensky, E., B. Reyers, A. S. van Jaarsveld, C.Fabricius, L. Erasmus, C. Holgate, T. Knowles,L. N. Lebesa, M. Pfab, M. van der Merwe, C.Shackleton, and L. Zondo. 2004. Ecosystemservices in the Gariep Basin: a component of theSouthern African Millennium Ecosystem Assessment(SAfMA).Borrini-Feyerabend, G. 1997. Beyond fences:seeking social sustainability in conservation. Callicot, J. B. 1994. Earths insights: a survey ofecological ethics from the Mediterranean basin tothe Australian outback. University of California 10 http://www.ecologyandsociety.org/vol10/iss2/art8/ Campbell, B. 2002. A critical appraisal ofparticipatory methods in development research.Social Research Methodology Campbell, B., A. Mandondo, B. Sithole, W. DeJong, M. Luckert, and F. Matose. 2001.Challenges to the proponents of common propertyresource systems: despairing voices from the socialWorld Development29:Casti, J. L. 1994. Complexification: explaining aparadoxical world through the science of surprise. Chakraborty, R. N. 2001. Stability and outcomesof common property institutions in forestry:evidence from the Terai region of Nepal. Ecologicaleconomics36:Chambers, R. 1994. Participatory rural appraisalWorld Development22:Coleman, J. S. 1990. The foundations of socialtheory. Harvard University Press, Cambridge,du Toit, J., B. Walker, and B. Campbell. 2004.Conserving tropical nature: current challenges forecologists. Trends in Ecology and Evolution 19Fabricius, C., E. Koch, and H. Magome. 2001.Towards strengthening collaborative ecosystemsmanagement: lessons from environmental conflictand political change in South Africa. Journal of theRoyal Society of New Zealand31:Fabricius, C., E. Koch, H. Magome, and S. D.Turner. 2004. Rights, resources and ruraldevelopment: community-based natural resourcemanagement in southern Africa. Earthscan,Funtowicz, S., and J. R. Ravets. 1990. Uncertaintyand quality in science for policy. Kluwer Academic,Gadgil, M., P. Seshagiri Rao, G. Utkarsh, P.Pramod, A. Chhatre, and the People'sBiodiversity Initiative. 2000. New meanings forold knowledge: the people's biodiversity registersEcological Applications10: Gadgil, M., P. Olsson, F. Berkes, and C. Folke. 2003. Exploring the role of local ecologicalknowledge in ecosystem management: three casestudies. Pages 189-209 in F. Berkes, J. Colding, andC. Folke, editors. Navigating social-ecologicalsystems: building resilience for complexity andchange. Cambridge University Press, Cambridge,Giampietro, M. 1994. Using hierarchy theory toexplore the concept of sustainable development.Futures 26:Gibson, C., E. Ostrom, and T. Ahn. 2000. Theconcept of scale and the human dimensions of globalEcological Economics32:Gunderson L. 1999. Resilient management:comments on "Ecological and social dynamics insimple models of ecosystem management" by S. R.Carpenter, W. A. Brock, and P. Hanson.Conservation Ecology http://www.consecol.org/vol3/iss2/art7 Gunderson L., C. S. Holling, and S. Light. 1995.Barriers and bridges to renewal of ecosystems andinstitutions. Columbia University Press, New York,Gunderson, L. H., and C. S. Holling, editors. 2002. Panarchy: understanding transformations inhuman and natural systems. Island Press,Healy, S. 2003. Epistemological pluralism and theFutures35:Holling, C. S. 1973. Resilience and stability ofecological systems. Annual Review of Ecology andSystematics4:Holling, C. S. 1986. The resilience of terrestrialecosystems: local surprise and global change. Pages292-320 in W. C. Clarke, and R. E. Munn, editors.Sustainable development of the biosphere. Janssen, M. A. 2002. A future of surprises. Pages241-260 in L. Gunderson, and C. S. Holling, editors.Panarchy: understanding transformations in humanand natural systems. Island Press, Washington D.Jones, S. 2002. Social constructionism and the 10 http://www.ecologyandsociety.org/vol10/iss2/art8/ environment: through the quagmire. GlobalEnvironmental Change 12Kapoor, I. 2002. The devil's in the theory: a criticalassessment of Robert Chambers' work onparticipatory development. Third World Quarterly 23Kay, J., H. Regier, M. Boyle, and G. Francis. 1999. An ecosystem approach for sustainability:Futures 31Latour, B. 1987. Science in action. HarvardUniversity Press, Cambridge, Massachusetts, USA.Levin, S. A. 1992. The problem of pattern and scaleEcology73Lowrance, R., P. Hendrix, and E. Odum. 1986.A hierarchical approach to sustainable agriculture.American Journal of Alternative Agriculture Macnaghten, P. and J. Urry. 1998. ContestednaturesMalthus, T. 1798. An essay on the principle ofpopulation as it affects the future improvement ofsociety.Martínez-Alier, J., and K. Schlupmann. 1991.Ecological economics: energy, environment,society.Martínez-Alier, J., G. Munda, and J. O’Neill. 1998. Incommensurability of values in ecologicaleconomics. Pages xx-xx in M. O'Connor and C.Spash, editors. Valuation and the environment:theory, method and practice. Edward Elgar,Mayer, E., and M. de la Cadena, editors. 1989.Cooperación y conflicto en la comunidad andina:zonas de producción y organización social. InstitutoMeadows, D., D. Meadows, J. Randers, and W.Behrens. 1972. The limits to growth. UniverseMeentemeyer, V. 2004. Geographical perspectivesof space, time, and scale. Landscape Ecology Milla, C. 1983. Génesis de la cultura Andina. Millennium Ecosystem Assessment (MA). 2003.Ecosystems and human well being: a framework forassessment/Millennium Ecosystem Assessment. Milton, K. 1996. Environmentalism and culturaltheory: exploring the role of anthropology inenvironmental discourse.Munda, G. 2000. Conceptualising and respondingto complexity. Pages 1-17 in C. Spash and C. Carter,editors. Environmental valuation in Europe: policyresearch brief. Cambridge Research for theEnvironment Series. European Commission,Murra, J. V. 1975. Formaciones económicas ypolíticas del mundo andino. Instituto de EstudiosNadasdy, P. 1999. The politics of TEK: power andthe 'integration' of knowledge. Arctic Anthropology 36O’Neill, R. W., and A. W. King. 1998. Homage toSt. Michael: or why are there so many books onscale? Pages 3-15 in D. L. Peterson, and V. T.Parker, editors. Ecological scale: theory andapplications. Columbia University Press, NewOstrom, E., and C. Hess. 2000. Private andcommon property rights. Center for the Study ofInstitutions, Population, and EnvironmentalChange, Indiana University, Bloomington, Indiana,Peterson, G. D., D. Beard, E. Beisner, S. Bennett,G. Carpenter, L. Cumming, and H. T. Dent. 2003.Assessing future ecosystem services: a case studyof the Northern Highland Lake District, Wisconsin.Conservation Ecology7: http://www.consecol.org/vol7/iss3/art1 Pretty, J., I. Guijt, I. Scoones, and J. Thompson. 1995. Participatory learning and action; a trainer'sguide. International Institute for Environment andSalmon, E. 2000. Kincentric ecology: indigenous 10 http://www.ecologyandsociety.org/vol10/iss2/art8/ perceptions of the human-nature relationship.Ecological Applications10:Scheffer, M., S. Carpenter, J. A. Foley, C. Folke,and B. Walker. 2001. Catastrophic shifts inNature413:Scheffer, M., F. Westley, W. Brock, and M.Holmgren. 2002. Dynamic interaction of societiesand ecosystems: linking theories from ecology,economy, and sociology. Pages 195-239 in L. H.Gunderson and C. S. Holling, editors. Panarchy:understanding transformations in human andnatural systems. Island Press, Washington D.C.,Schulze, R. 2000. Transcending scales of space andtime in impact studies of climate and climate changeon agro-hydrological responses. Agriculture,Ecosystems and Environment82:Shackleton, S. and B. Campbell. 2001. Devolutionin natural resources management: institutionalarrangements and power shifts; a synthesis of casestudies from Southern Africa. Report 690-0251.van der Heijden, K. 1996. Scenarios: the art ofstrategic conversation. Wiley, New York, Newvon Bertalanffy, L. 1968. General systems theory. von Kotze, A. 1998. Monologues or dialogues?Missed learning opportunities in participatory ruralConvergence31Walker, B., and N. Abel. 2002. Resilientrangelands: adaptation in complex systems. Pages293-313 in L. Gunderson and C. S. Holling, editors.Panarchy: understanding transformations in humanand natural systems. Island Press, Washington D.Walker, B., J. Carpenter, N. Anderies, G. Adel,M. Cumming, J. Lebel, G. D. Norberg, G. D.Peterson, and R. Pritchard. 2002. Resiliencemanagement in social-ecological systems: aworking hypothesis for a participatory approach.Conservation Ecology 6: http://www.ecologyandsociety.org/vol6/iss1/art14/ Wicken, J. S. 1987. Evolution, thermodynamics, and information: extending the Darwinianprogram.Wood, E. F., and V. Lakshmi. 1993. Scaling waterand energy fluxes in climate systems: three land-atmosphere modelling experiments. Journal ofClimate