Beyond the technological fix

Though poorly developed from a methodological standpoint, stakeholder analysis (SA) now belongs to the long list of virtues and catchwords reigning over the field of development. Much to its credit the method travels well across disciplinary and theoretical boundaries. It is so eclectic as to stretch across the political spectrum and fit in with most of what it encounters, be it informed by participatory methodology or not (Burgoyne 1994: 205, Grimble and Wellard 1997: 182). SA is currently used in fields ranging from political science to policy development and international relations. The concept and related methodology have made significant inroads into poverty reduction studies and applied research pertaining to issues of sustainable livelihood, community-based natural resource and conflict management (Ramírez 1999). It is also part of World Bank thinking on participation methodology since about 1993 (MacArthur 1997a: 5).

Source: Oudman et al 1998

The origins of SA, however, belong to the history of business and managerial science. This is reflected in the term "stakeholder" itself, apparently first recorded in 1708, to mean a bet or a deposit. The word now refers to anyone significantly affecting or affected by someone else's decision-making activity. Economic theory centered on notions of stakeholder relations goes back to the beginnings of industrialism and is embedded in ideals of 19^th century cooperative movement and mutuality (Clarke and Clegg 1998: 295). Stakeholder theory reappears in business and management discussions of the 1930s (cf. Brugha and Varvasovszky 2000: 239-40). The approach was then designed and continues to be used nowadays by firms and organizations to factor in stakeholder interests in order to enhance the enterprise's relationship with society and secure better prospects of financial success. With the help of SA firm decisions can profit from views that go beyond the narrow interests of stockholders and shareholders investing in a business.

But there is more to SA than new wine emerging from old bottles. In recent decades SA has been significantly transformed by developments principally in political economy, decision theory, environmental studies, and also RRA-PRA-PAR methods of project design (rapid rural appraisal, participatory rural appraisal, participatory action research). Dispute resolution practices and the social actor perspective in the social sciences are also kindred spirits of SA (Grimble and Wellard 1997: 185, Oudman et al 1998).

The idea of SA is catching on for several reasons. For one thing the methodology involves a recognition of the fact that obstacles to peace, equity, sustainability or growth cannot be dealt with through technological means alone. When tackling issues of poverty and environmental degradation, power relations and conflicting interests must be addressed. Social relations involving all "interested parties" must be examined and alternative practices explored if blueprints for technological change are to be grounded in reality and add up to more than pie in the sky.

SA also has the advantage of being a flexible, context-specific paradigm that helps focus attention on specific problems, actors and opportunities for change. This is particularly helpful in the context of NRM (natural resource management) issues where complex and interdependent relationships of groups relying on common resources such as land, water and forests typically prevail. Agro-export producers, small-scale farmers or fishers, government agents, conservation groups and ethnic minorities may all have a stake, and conflicting interests, in the management of particular resources. Multistakeholder analysis and involvement is all the more needed where resources crosscut different administrative, social, economic andpolitical systems operating at micro and macro levels.

Stakeholder matrix

Social assessment is almost always a must where management practices generate negative externalities, i.e., production costs not borne entirely by the producer and not factored into resource management decisions. A detailed and realistic understanding of these multistakeholder relationships is critical to developing equitable and sustainable management practices in situations of actual or latent conflict and competing interests. Other widespread problems calling for SA include stakeholder under-representation, disputes generated by unclear access and property rights, or problems of incompatible stakeholder uses and agendas. Stakeholder considerations are equally relevant in situations of opportunity costs and trade-offs that must be addressed at the policy level -- e.g., choosing between short-term and long-term horizons, or balancing conflicting objectives such as conservation, development, equity and peace (Chevalier and Buckles 1999, Grimble et al 1995: 11-16, Ravnborg and del Pilar Guerrero 1999).

It should be stressed that NRM systems featuring some combination of these characteristics tend to be the rule rather than the exception. It is doubtful therefore that the indices listed above can be used as "the basis for early screening of projects, policies and situations, to be followed by a full SA where necessary" (Grimble and Wellard 1997: 179).

Another attractive feature of SA lies in its postmodern, pluralist and constructivist leanings (Burgoyne 1994: 188-89, Calton and Kurland 1996: 154, Ravnborg et al 1999). The approach constitutes a middle-range "social actor" alternative to the positivist methodologies often prevailing in studies of resource management practices (Pretty 1994, Röling 1994, Rubiano). Given its focus on people’s intentions and self-identified interests or stakes, the method emphasizes processes of social construction; biophysical properties and economic interests are assigned not to objective systems but rather to agents and socially-positioned perspectives on social and natural reality. SA also goes beyond participatory methods and practices that emphasize popular involvement and that pay little attention to inherent structural problems and multilevel conflicts plaguing 'local peoples." By the same token SA represents a challenge to conventional economic analysis, an approach that "does not adequately consider the distribution of costs and benefits among different stakeholders: the winners and losers. It ignores the fact that different stakeholders do not perceive environmental problems in exactly the same way and will therefore seek different solutions and use different criteria to assess the desirability or worth of an intervention. Ways for better anticipating and dealing with stakeholder opposition and conflict, and better incorporating various interests, especially those of weaker groups in society, are therefore crucial for improving policy design and project implementation" (ILEIA 1999; see also Grimble and Wellard 1997: 183).

At the same time stakeholder theory is a response to conventional research methods in the social sciences such as stratification theory or political economy, class-centered perspectives that are still widely used in research on the social aspects of NRM issues (with recently-gained concessions to gender and ethnicity). Given their emphasis on macro-level studies of deep-seated power structures that social scientists alone can apprehend and rethink, there is a widespread dissatisfaction with the use of these methods alone. Class analyses often fail to capture the specificity of agent-constructed problems and foundations of local NRM conflicts and stakeholder-driven options for resolving them. They also require ready-made definitions of class membership and dynamics fashioning the course of history. By contrast, SA performed through participatory methods (not always the case) precludes a priori conceptions of stakeholder categories and relations applicable to an area or population as a whole.

In short, SA brings something new to participatory methods, formal economics and political economy. It highlights local actor perspectives on conflicting interests and alternative strategies aimed at promoting equity and sustainability in NRM systems.

But what is a stakeholder and what does SA actually do when applied to NRM problems or conflicts? Briefly, stakeholders are groups, constituencies, social actors or institutions of any size or aggregation that act at various levels (domestic, local, regional, national, international, private and public), have a significant and specific stake in a given set of resources, and can affect or be affected by resource management problems or interventions. When applied to NRM issues, SA thus serves to identify:

1/ the stakeholders involved in a competition or conflict over natural resources;

2/ stakeholder values and views on NRM problems and conflict-management strategies;

3/ the multiple interests and objectives of stakeholders in relation to particular NRM systems;

4/ the actual resources, influence, authority or power that stakeholders can bring to bear on particular NRM initiatives;

5/ the networks that stakeholders belong to and patterns and contexts of interaction between them, be they collaborative or conflictive;

6/ the distributional and social impacts of NRM policies and projects (winners and losers, potential trade-offs and conflicts), hence the risks and viability of particular NRM interventions;

7/ the appropriate type or degree of participation by primary and secondary stakeholders (internal, external), at successive stages of a project cycle (cf. MacArthur 1997a: 6-7, MacArthur 1997b: 255);

8/ feasible coalitions of project sponsorship and ownership aimed at efficient, equitable and sustainable livelihood strategies (based on compromises between public goals and divergent stakeholder interests).

SA is thus an invitation to examine the power relations among groups and individuals and their respective interests in a resource or situation. The method helps identify key differences among groups and areas of potential common ground and feasible interventions aimed at a better management of natural resources and related conflicts. What is at stake here is a commitment to stakeholder participation in the realm of political economy and NRM science, towards a critical social research agenda. Stakeholder theory and practice hold great promise in the sense of promoting systematic actor involvement and a pragmatic focus on problems to solve.

A method rough and not-so-ready

Participatory methods in the social sciences have taught us the virtues of the rough-and-ready. They substitute knowledgeable actors for non-analytical informants, producing worthwhile results without being scientifically pretentious or technically spurious. SA aspires to achieving comparable goals but is not entirely successful at it. On the whole it tends to be rough and not-so-ready. As argued below, many refinements have yet to be brought to the method before it can actually deliver the goods.

A rudimentary tool

One problem with SA carried out in a participatory manner is that it is often confused with techniques to facilitate stakeholder involvement or input in managing NR projects or conflicts. SA may be part of the "stakeholder approach to management" (Roy 1999) but is not synonymous with it. While one is a contribution to the other, caution must be taken not to lose sight of the precise goals of SA:the method is an exercise in what might be called "ethnopolitics," a participatory analysis that seeks an actor-driven assessment of the power structure affecting NRM systems. Many organizations acknowledge the importance of including stakeholders in their policy development (e.g., Harrison and Burgess 2000) but only a few have done so on the basis of a stakeholder-driven analysis of social obstacles to change and strategies to overcome them. Office-based stakeholder snapshots aimed at a better management of stakeholders are often done with sensitivity shown to actor representations but without a commitment to active stakeholder representation -- enhanced participation and empowerment in NRM systems (e.g., Burgoyne 1994, Brugha and Varvasovszky 2000, Kogo 2000, Dick 1997, REC 1996; cf. Grimble and Chan 1995, Grimble and Wellard 1997: 185-87).

Some training materials have been developed to enhance skills in stakeholder analysis among non-government organizations and others working with local communities (DFID 1995, Engel and Salomon 1997, FAO, Grimble and Wellard 1997, Grimble et al 1995, Herweg et al, Horelli et al 2000, IADB, IIRR 1998, MacArthur 1997a, 1997b, ODA 1995, SEAGA 1999, Selener et al 1996, Team Technologies Inc 1994, World Bank Group PRS). The most articulate version of SA can be found on IIED's web site.

These tools emphasize the facilitation of discussion among local groups who are in one way or another involved or affected by a particular set of resources or management initiatives. By and large, however, booklets, guidelines or checklists designed for fieldworkers generally focus on gathering social information that is strictly necessary for managing active or potential conflicts over natural resources. Few if any offer more advanced analytic tools towards a better understanding of the interests and power differentials affecting NRM practices in site-specific contexts. If SA happens to be carried out at some length, case studies will emphasize results and findings as opposed to explaining how researchers went about doing participatory SA and drawing lessons from the exercise (including merits and weaknesses of the method).

Source: Grimble, Chan, Aglioby and Quan 1995

SA is still a new kid on the block, "one of the main ideas in development thinking that have been introduced in recent years" (MacArthur 1997b: 251). It has not yet become a standard procedure and draws on concepts and methods that vary considerably in rigor, ranging from a few methodological instructions to qualitative matrices of stakeholder interests, relations, impact potential and levels of desired participation (see MacArthur 1997a). On the whole SA is a pragmatic tool used at the beginning of a process geared essentially to the actual business of managing a project or a conflict. As a result, SA procedures are not always spelled out, methodological guidelines tend to be sketchy, and detailed accounts of explicit uses of SA are very few. Save perhaps for a manual on participatory conflict management techniques about to be published by FAO, instructions needed to convert SA techniques into a methodology that does justice to the complexity of social factors and obstacles to change are generally lacking.

SA methodology can be improved in at least three ways. First, SA tools should permit adjustments to contextual assumptions, circumstances and points of entry. The method should take into consideration not only the sectorial and cultural context of the analysis but also the time limits and resources that can be reasonably allocated to this activity. It should also take into account the actual purpose of the exercise, i.e., whether it is done for NR area, project, policy, problem or conflict management purposes. In some cases the analysis may be designed to explore how pre-established projects or policies will impact on target beneficiaries and other actors possibly affected by the proposed activities, including those who have the power to influence or determine their outcome. SA can thus feed into a conservation program (e.g., Borrini-Feyerabend and Brown), a health care policy (Brugha and Varvasovszky 2000: 244), a poverty reduction strategy (MacArthur 1997b), a food security project (World Food Programme 2001) or any aid-related activity (Foell et al, ODA 1995, REC 1996). In other cases, a manageable area is first selected and SA is then used to secure stakeholder input into problem identification and project design and management activities (Ravnborg et a 1999; see Grimble and Wellard 1997: 177, 186, Grimble et al 1995: 6, MacArthur 1997b: 253). While these various strategies are an integral part of the development scene, the context in which SA is undertaken will have a direct bearing on how the analysis is to be designed and carried out.

Second, SA should be a permanent feature of NR project or conflict management processes. SA tends to be used at the beginning of a cycle, as in a project Concept Note, an ex ante appraisal of a project or policy proposal, or the initial identification of stakeholders involved in a conflict or problem domain (Grimble and Chan 1995). The technique then serves as a strategic entry point for a development or conflict management process, not a critical methodology that supports the process throughout its entire duration (Ramírez 1999). If anything SA should be an iterative, action-oriented exercise in social analysis (Foell et al). If not revised during the project or conflict management cycle, a SA matrix may become obsolete; i.e., stakeholders and their interests and views may evolve, new actors may appear on the scene, or central issues and stakes may shift over time (Brugha and Varvasovszky 2000: 244-45, Burgoyne 1994: 196, Frost 1995: 657, Grimble and Chan 1995, MacArthur 1997b: 253). More importantly, assessments of stakeholder relations should feed into plans of immediate action, thereby generating activities that will require stakeholder relations to be monitored through on-going analysis. The notion that SA is a one-shot, quick-and-dirty exercise to be followed immediately by alternative social and NR management activities constitutes a disservice to the field as whole.

Third, SA should not be pitted against scientific approaches to social reality. Otherwise misinformation and superficiality may result. While remaining accessible to non-experts, a full-fledge SA methodology should indicate how tools can be organized in sequence and with some cumulative effect, moving beyond the rapid snapshots of social reality frequently advocated in the SA literature. Also, organizing SA tools on a sliding scale of complexity should provide users with various methodological options that can adjust to variable NRM circumstances. Another strategy consists in exploring complementarities between expert and non-expert analyses of stakeholder relationships, an option to which we now turn.

Synergies of knowledge systems

By and large there is some confusion and no standard view in the SA literature as to how much weight should be granted to "emic" stakeholder views obtained through participatory methods compared to "etic" analyses of NRM systems obtained from office-based studies performed by project teams or social scientists (e.g., Frost 1995, MacArthur 1997b: 261, Warner 2000). One striking tendency, however, is for both strategies to generally ignore one another. While social scientists have a long tradition of seeking truth through disciplinary means alone, advocates of participatory methods (emphasizing the insider look at reality) tend to view their research strategy as a radical challenge to scientific expertise. Synergies between knowledge systems are rarely considered.

Eliciting local perspectives on the social and power relations governing NRM systems is vital to understanding obstacles to change and promoting dialogue on alternative practices. At the same time it is important to recognize that some answers to important SA questions -- e.g., how will the market or the environment impact on different stakeholders following particular NRM interventions — may not be readily available to stakeholders. This is so true that it is not uncommon for stakeholders themselves to recognize gaps in their own knowledge base and to seek means to secure more information. Practitioners of SA should thus avoid the populist assumption built into some participatory methods (and studies of indigenous knowledge systems as well): the notion that knowledge is equally distributed and freely accessible to all parties involved and that much of it can be captured through rough-and-ready methods such as RRA, PRA or PAR. Social and natural analyses must avoid habits of spurious precision and exclusionary access. But this is not to say that technique, rigor and theory should be dismissed altogether.

When carried out through participatory methods, SA may require critical information and analysis obtained through methods developed in the natural and social sciences. While this need for multiple sources and methods of analysis is recognized in practice, concrete guidelines regarding methodological synergies should be further explored. Advanced tools should suggest ways in which SA can inform and be informed by other useful methodologies, including natural (biological, agronomic, etc.) and social scientific perspectives (market studies, policy analyses, legal expertise, etc.) on the NRM problems or conflicts at hand. These synergies can take one of two forms: exchanges of information and findings between knowledge systems, or adaptations of methods used and tested on both sides of the epistemic divide.

Participatory methods applied to SA would stand to gain from appropriately-scaled adaptations of some of the methodologies currently used by social scientists to make sense of stakeholder relations (e.g., Mitchell et al 1997, discussed below). As it now stands participatory SA is confined to a few power-structure matrices using prefabricated binarisms (influence/importance, direct/indirect impact, positive/negative relations, strong/weak connections, etc.) that can hardly do justice to the social knowledge possessed by stakeholders let alone the social reality they live in. While easy to use, general distinctions such as between primary (targeted), secondary (intermediary) and external stakeholders (people and groups not formally involved but who may have an impact or be impacted by an activity) may be an invitation to gloss over the complexity, dynamics and self-constructions of social reality (cf. Clayton et al 1998, MacArthur 1997a: 3, 7-9, MacArthur 1997b).

Empowerment and participatory action research

One frequently stated purpose of participatory SA is to highlight the interests of marginalized groups, giving them voice and representation in situations of high power imbalance. SA is particularly important in situations where stakeholders lack agency in the sense of being unorganized constituencies with limited awareness of their interest in a given NRM system (see Borrini-Feyerabend 2000). By itself, however, SA is not necessarily designed to guarantee these groups stronger representation or empowerment during the SA research process let alone after (through full involvement in project decision-making or conflict management activities). This is especially the case where stakeholders are ranked according to influence and importance, a strategy that can lead to stakeholder information playing into the hands of the more powerful groups and an even greater under-representation of lower-ranked groups (Calton and Kurland 1996: 159, Grimble and Chan 1995, MacArthur 1997a:14). Rather than eliciting the participation of the most visible and powerful and pursuing project or policy effectiveness above all (see Foell et al), a fully-developed SA methodology should be clearly committed to principles of empowerment and related measures aimed at "leveling the playing field" -- giving equal voice to the perspectives and the priorities of less powerful stakeholders.

The question of empowerment brings us back to the issue of participatory methodology. SA is usually committed to enhancing stakeholder involvement in NR management processes. Yet not all stakeholder analyses are carried out through participatory methods. As in much of the management literature devoted to this topic (with some exceptions, see Oudman et al 1998), SA is frequently done independently from the actors, prior to their actual involvement in decision-making activities. No stakeholder participation is sought when answering a critical question -- i.e., "who decides on the purpose of the analysis and who counts most?" (Ramírez 1999). Since stakeholder identification is a consequential matter, analyses done without participation are likely to reflect the interests and agenda of the agency directing the exercise in social assessment.

In point of fact, SA's commitment to using participatory methodology and "leveling the playing field" varies considerably. Even when done in a participatory fashion, SA may be used primarily as a tool to extract information to be processed by "independent" stakeholder analysts (as in REC 1996; see Schmink 1999). SA is thus "a particularly good example of a tool that can be used in a participatory way, or in a top-down way that only pays lip service to participation" (World Bank Group statement on "Stakeholder Analysis"). Actually the same can be said of participatory methodologies in general; they can take different forms that vary greatly in quality and levels of empowerment (Karl 2000).

This is not to say that SA requires neutrality on the part of those initiating or carrying out the exercise. Rather the issue is one of transparency through self-analysis and disclosure. SA practitioners committed to principles of empowerment must not omit to incorporate the starting-up team into the SA exercise. They should put those initiating the analysis (sometimes bent on feigning neutrality) in the picture of power relations and related interests from the start (Burgoyne 1994: 192).

Issues of power and social complexity

Actors doing SA should not be so naïve as to think that their analysis will be immune to the social problems they are attempting to address, those of power differentials and conflicts of interests. As with any research activity, SA is subject to variable agendas, some of which may prefer concealment to transparency. Participatory views expressed on the interests and assets ascribed to stakeholders are a case in point. They may be understated or overstated and may have to be checked through independent means if transparency and reliability are to be achieved.

Not that full truth and nothing but the truth is always desirable. Putting everyone's cards on the table may be neither possible nor advisable and individual interviews collected and analyzed by independent stakeholder analysts may be the best option, to be disclosed to all parties involved with diplomacy and circumspection (Brocklesby et al, ODI 1996, Ravnborg et al. 1999). "Stakeholder analysis often involves sensitive and undiplomatic information. Many interests are covert, and agendas are partially hidden. In many situations there will be few benefits in trying to uncover such agendas in public" (ODA 1995). SA enthusiasts should bear in mind that undiplomatic analyses may exacerbate or generate conflict. In cases of serious conflicts of interest and drastic power differentials, shuttle diplomacy or bilateral negotiations may be more appropriate than roundtable analyses and negotiations. Better guidelines regarding diplomatic adaptations and related issues of information disclosure (cf. Sinclair-Desgagné and Gozlan 2001 and Pelle-Culpi's thesis) are thus in order.

SA methods should also pay attention to levels of social and cultural complexity. For one thing, participatory SA must be careful not to assume "clear definitions" of problem domains. Nor should group interests and boundaries and related mechanisms of representation be taken for granted. These assumptions impose excessive rigidity where flexibility is sought. Also they are constantly contradicted by muddles in the models -- e.g., the multiple hats that stakeholders wear and the complex networks they belong to (Borrini-Feyerabend et al 2000, Foell et al, Grimble and Chan 1995, MacArthur 1997a: 261-62, 1997b: 258, Ramírez 1999). The issue of group boundaries is especially tricky, which means that all exercises in SA should raise two questions: when to disaggregate a particular group into various stakeholders, and when to lump various actors into one stakeholder group (MacArthur 1997b: 262). Other advances in SA include better guidelines as to what should be done in situations of greater cultural complexity. For instance, what happens in situations where "stakeholder" concepts and related semantics are antithetical to local conflict-management values and practices? Participatory techniques designed to unravel the cultural semantics of actor identification, rules of communications and engagement, natural resource management practices, conflict resolution activities, accounts of power relations and conflicts, and so on, would be particularly critical in this regard (see Mäkelä 1999, Beckley et al 1999, Chevalier and Buckles1999).

While committed to using participatory methods and resolving "manageable" problems, SA should not shy away from addressing broader power structures and inherent structural and institutional problems affecting most NRM situations (Grimble and Chan 1995, Hatzius 1997). In the end, the usefulness of SA will hinge on its ability to factor in the complexity of social forces and related factors (economic, political, cultural) governing NRM systems. SA is by no means a product of comprehensive world-system or political-ecology thinking (see Schmink 1999). But this is no excuse for the method to succumb to the naïveté of "stakeholder neocorporatism" -- trading off expedient studies and settlements for critical analyses and radical challenges to local or global structures governing NRM activities.

Given the inevitable tension between the requirements of pragmatism and those of critical thinking, one basic question for any multistakeholder management strategy is thus the following: to what extent should compromises be sought between the interests of dominant and subordinate groups? That is, under what circumstances will consensus building work against principles of equity and sustainability? Should concessions be made to stakeholders whose interests cannot be reconciled with end-goals of redistribution and empowerment (MacArthur 1997a: 12)?

Source: Warner 2000

On these issues, a quick reading of the SA literature gives the impression that the deeper the problem is and the larger its scale, the less useful the methodology is. As as result macro-level adaptations of SA are few and far between. This may reflect the issue of the optimum group size or number normally required for SA (see MacArthur 1997b: 258). While SA should not be carried out with so few groups as to leave out some key players, the method can also present problems if undertaken with so many groups as to exceed participation levels deemed feasible, sensible and cost effective. As Burgoyne (1994: 194) remarks, "stakeholder situations have no natural outer boundaries," with the implication that methodological choices must be made. Stakeholder participation may be easier to achieve in micro-level NR project/conflict management activities (local or regional) compared to meso- and macro-level situations involving complex information systems and policy interventions. This question of flexible macro-level adaptation needs to be closely examined for SA to maximize its full methodological potential. For SA and participatory processes at the macro-policy level, see Foell et al, PRECOD, REC (1996), SEAGA (1999),Whyte (2000) and the World Bank Group (PRS).

Adaptations to situations of intense conflict or extreme power imbalances are also in order. Most practitioners of SA are of the view that SA should not be contemplated when stakeholders are embroiled in highly reactive conflict situations. Nor is the exercise feasible where key actors have few assets (e.g., time, money, authority, knowledge, skills) that they are able or willing to invest in roundtable problem assessment activities and related management alternatives (Ramírez 1999). For that matter SA will not appeal to dominant groups incurring limited costs from a given NRM system or conflict. These stakeholders may have little to gain from ADR and NRM strategies based on principles of stakeholder empowerment and negotiation.

When the latter conditions prevail, a distinction should perhaps be drawn between two forms of SA: the general and the restricted. The general form mobilizes all relevant parties and excludes none. Note that this general stakeholder involvement does not presuppose equality between actors. Nor does it require an infusion of "good will" that is so massive as to override all considerations based on crass stakeholder interest. Rather the general form of stakeholder engagement, which includes SA among other activities, presupposes two things: (a) a decision on the part of all concerned parties to reduce the costs incurred in maintaining a given NR conflict or mismanagement situation; and (b) a general "interest" in exploring multistakeholder negotiation strategies in order to achieve individual ends. Admittedly not all NRM problems will lend themselves to this form of SA. But some will and may be conducive to activities aimed at reconciling the four strategic E's: efficiency, environmental sustainability, equity and empowerment (Grimble and Wellard 1997: 174).

The restricted form of SA is critical to our enterprise and is universally overlooked in the literature. It is premised on a simple observation: all stakeholders are in the habit of assessing the social conditions under which they operate and will do so through interactive means. Over and beyond the methodology it aspires to be, SA based on exchanges of information is a permanent feature of social life and interaction. Action-oriented SA built into NRM Realpolitik, however, is often restricted to those stakeholders that one actually interacts with. Problem assessment performed under these conditions excludes the active engagement of those who are too distant or whose interests are simply deemed irreconcilable with one's own. The exclusion may also reflect the fact that the exercise is part of a zero-sum game that cannot be avoided; SA is then performed with the aim of reducing the influence of powerful groups and countering their plans (MacArthur 1997a: 12). Indigenous subsistence farmers battling against an international pulp and paper company trying to buy off their lands (in pursuit of a eucalyptus plantation project supported by state politicians) may have nothing to gain from sitting at the same table as the company and trying to settle their differences. Instead they may prefer to explore and develop alliances with other local stakeholders (merchants, cattle ranchers, municipal authorities) and external actors as well (NGOs, some provincial and federal government representatives, etc.), with a view to blocking company plans. Methods to enhance SA performed under such conditions are part and parcel of what SA has to offer and should not be ignored for the sake of a blind commitment to multistakeholder dialogue achieved at all costs.

All three factors must be considered simultaneously in that "power gains authority through legitimacy, and it gains exercise through urgency" (Mitchell et al 1997: 869). The argument is all the more useful as it lends itself to an eightfold stakeholder class typology reflecting variable degrees of salience and types of relationship. "Definitive" stakeholders are those who possess all three attributes and will therefore receive the greatest attention. Three other classes come next in rank: the "dominant" who possess power and are perceived as having legitimate claims; the "dependent" whose claims are deemed legitimate and urgent; and the "dangerous" who possess power and have claims that are urgent though not legitimate. The least salient stakeholders comprise the "dormant" (powerful but with claims that are deemed neither urgent nor legitimate), the "discretionary" (legitimacy without power and urgency), and the "demanding" (urgency without power or legitimacy). Lastly, the analysis lumps all those who possess none of these attributes into a residual "nonstakeholder" category.

Admittedly, this model has been developed in a purely managerial perspective. The firm occupies the center of every stakeholder nexus and a high degree of salience is automatically granted to managers. The latter act as general moderators or mediators and they are responsible for carrying out the analysis. Interestingly, they are never preoccupied with the "attention" they deserve or require; the only question they ask is who they should pay attention to (Mitchell et al 1997: 870-71). This is so because "managers must know about entities in their environment that hold power and have the intent to impose their will upon the firm. Power and urgency must be attended to if managers are to serve the legal and moral interests of legitimate stakeholders" (Mitchell et al 1997: 882). Another problem lies in the concepts of legitimacy, power and urgency, notions that are socially constructed at the concrete level but not problematized at the level of general theory. Constructivism is good at the ground level but goes out the window when doing theory. The interests of theorists are preserved against threats of radical decentering.

All the same, the model is a potentially valuable contribution to SA, provided that it be adapted to NR conflict/project management objectives, actor-centered perspectives and principles of empowerment as well.

Somewhere between pragmatics and utopia

The preceding discussion suggests that SA is an integral part of all social activity and cannot be reduced to mere technique. MacArthur's notion that SA is not essential to all development projects should therefore be qualified: the packaged methodology may not be indispensable but social assessment tactics can never be dispensed with. MacArthur (1997b: 263) goes on to say that "in many kinds of project, who the stakeholders are, and the nature of their stakes, will be readily evident to experienced people in a planning or appraisal team," which means that not too much should be expected from the exercise. This caveat makes sense in a situation where the exercise is carried out by members of a relatively homogeneous group, such as a planning or appraisal team. It is less true of a SA that allows various stakeholders to negotiate amongst themselves an assessment of players and interests involved in a given situation. If performed on the basis of a multistakeholder approach, i.e., for and by all concerned parties, SA is bound to raise critical issues and should be an essential part of all NRM processes.

Still, the fact that SA can be put to radically different usages, from manipulative stakeholder relationship management in a business context to stakeholder enabling and empowerment in NRM systems, cautions us against using this methodology as a magic bullet aimed at resolving all problems of inequity and unsustainability. Some observers of "stakeholder capitalism" are too optimistic in this regard. Calton and Kurland (1996: 156) thus suggest that stakeholder enabling practices can resolve the dilemma between social morality and economic rationality, or the paradox that pits ethics without business against business without ethics. In their view the stakeholder approach to doing business is the alternative to management-centered organizations and related hierarchical decision-making structures. It introduces into profit-oriented activities a Habermasian lifeworld of communities of conversation and webs of cooperative, mutually beneficial, trust-based relationships. It brings out pluralistic politics into the open and promotes flexible, adaptive, networklike organizations aimed at solving problems rather than preserving bureaucratic structures for their own sake. Stakeholder capitalism fosters an affirmative "ethic of care," an "institutional capacity for intimacy" that allows stakeholders to share their concerns through decentered voice mechanisms and pluralistic discursive practices. Stakeholders thus become co-authors of their destiny, ends rather than means of development and growth (Calton and Kurland 1996: 160-61, 164-70).

Clarke and Clegg (1998) are equally optimistic. Stakeholder management practices developing mostly in Europe and Asia (especially in Germany and Japan) represent a paradigm shift towards the inclusive company model. The trend is towards stable relationships based on stakeholder accountability rather than a series of fluctuating transactions aimed at reaping short-term profits (Clarke and Clegg 1998: 295-97, 348-49). In this model managers adopt an inclusive concern for the long-term interests of all stakeholders, towards a sense of corporate citizenship and an emphasis on intangible assets (values embodied in human and social capital, including trust, knowledge and skills). Stakeholder capitalism is thus a radical departure from the Anglo-Saxon approach that grants priority to shareholders and value in property and tangible assets. The underlying principles of stakeholder capitalism are so promising as to be echoed in a 1996 UN document entitled Engaging Stakeholders, a statement promoting the systematic and active engagement with stakeholders on the full range of environmental, social and economic questions (Clarke and Clegg 1998: 361).

Source: Warner 2000

Stakeholder theory will do all the great things it is supposed to provided we accept the notion that all conflicting interests can be negotiated and that the well-being of companies and economies hinges on the active participation of all citizens, actors whose material interests ultimately coincide with those of capital. Radical critiques of existing property regimes, managerial systems, market forces and business mechanisms ruling over the economy must be bracketed in the same breath. Critiques founded on class or managerial hegemony theories are rendered obsolete. Management should be granted a leading role in promoting people-centered dialogue across stakeholder boundaries. Last but not least, we are asked to be realistic enough to recognize that multistakeholder conversations and consensus building are merely non-financial means to business ends, those of the company facilitating the operational dialogue (Clarke and Clegg 1998: 367).

Given these assumptions, one might suspect the theory of being an exercise in tactical optimism instigated by business with a view to maintaining "public confidence in the legitimacy of its operations and business conduct; in other words, to maintain a licence to operate" (Clarke and Clegg 1998: 353). It may be in the interest of those powerful groups doing much of the counting, social and managerial, to argue that those who "count the most" should be considered first and foremost. Stakeholder theory would then be recognized for what it is, a thick cloak of bright-eyed "affirmative postmodernism" covering strategies of organizational seduction and manufactured consent (Calton and Kurland 1996: 164, 167, 171).

But why now? What are the current stakes that make stakeholder management (and rhetoric) so attractive to so many? Could it be that the stakeholder concept is becoming as commercial necessity, a public relations or social marketing exercise in a world featuring the "arrival of the professional investor, the sophisticated customer, the empowered employee, the information revolution, a knowledgeable public and government regulation," as Clarke and Clegg wonder (1998: 356, 367; see also Altman 1994)? Could it be that development agencies are also a primary stakeholder in these new social assessment activities -- cognizant as they are that too many conflict/poverty reduction investments are wasted on technical innovations and people-centered projects that ignore issues of power structure and material interests?

There is no automatic answer to these critical questions other than the proverbial "only history will tell." In the end much will depend on how SA is actually developed and used, by whom and for what purpose. If stakeholder theory is to serve multiple interests and permit dialogue across class boundaries, as perhaps it should, caution will nonetheless have to be taken not to forego longer-term challenges to power structures embedded in problems of chronic poverty, inequity and unsustainability.

© 2001 Jacques Chevalier. This page was last updated in June 2001.