Diversity of forest institutions around the world

Some progress has been made to move forward from the overreliance on panaceas in efforts to improve environmental conservation. An important component of this progress is the extensive empirical research now conducted on the diversity of institutions found around the world that are involved in environmental conservation (Berkes & Turner Reference Berkes and Turner2006; Ostrom & Nagendra Reference Ostrom and Nagendra2006; Bray Reference Bray2010). Empirical studies are uncovering a diversity of institutions that achieve sustainable development, as well as those that do not. Several factors beyond the very generalized labels associated with the idealized property-rights regimes discussed above are associated with successes in the field.

Many of these empirical studies have been conducted within the International Forestry Resources and Institutions (IFRI) research programme, active since the early 1990s. At that time, Marilyn Hoskins of the FAO asked Elinor Ostrom whether the Workshop in Political Theory and Policy Analysis would apply their prior theory and methods to forestry resources and begin a long-term collaborative research network of centres located around the world (Gibson et al. Reference Gibson, McKean and Ostrom2000; Poteete & Ostrom Reference Poteete and Ostrom2004; Wollenberg et al. Reference Wollenberg, Merino, Agrawal and Ostrom2007). IFRI programme collaborating centres are now located in Bolivia, Colombia, Guatemala, India, Kenya, Mexico, Nepal, Tanzania, Thailand, Uganda and the USA, with new centres being established in Ethiopia and China. IFRI is the only interdisciplinary long-term monitoring and research programme studying forests owned by governments, private organizations and communities in multiple countries.

Forests are a particularly important form of land cover given their role in climate change related emissions and carbon sequestration (Canadell & Raupach Reference Canadell and Raupach2008), the biodiversity they contain (Dallmeier & Comiskey Reference Dallmeier and Comiskey1998), and their contribution to rural livelihoods in developing countries and regions (Shackleton & Shackleton Reference Shackleton and Shackleton2004). As mentioned already, government-owned protected areas are frequently recommended as ‘the’ way of preserving the ecosystem services generated by forests (Terborgh Reference Terborgh1999).

In an effort to examine whether government ownership of protected areas is a necessary condition for improving forest density, Hayes (Reference Hayes2006) used IFRI data to compare different forest governance types via a rating of forest density (on a five-point scale) assigned to a forest by the forester or ecologist who had supervised the mensuration of trees, shrubs and ground cover in a random sample of forest plots. Of the 163 forests included in the analysis, 76 were government-owned forests legally designated as protected forests, and 87 were public, private or communally owned forested lands used for a wide diversity of purposes. No statistical difference was found between the forest densities in officially designated protected areas and all other forested areas.

Conversely, other IFRI studies have found that lower-level (more specific) variables can be very significant. Gibson et al. (Reference Gibson, Williams and Ostrom2005), for example, examined the monitoring behaviour of 178 forest user groups and, controlling for many other variables, found a strong positive correlation between the level of monitoring and a forester's assessment of forest density. Recent studies by Coleman (Reference Coleman2009) and Coleman and Steed (Reference Coleman and Steed2009) also found that a major variable affecting forest conditions was investment in monitoring by local users. Further, when local users are given harvesting rights, they are more likely to monitor illegal uses themselves. Other studies also stress the relationship between local monitoring and better forest conditions (Banana & Gombya-Ssembajjwe Reference Banana, Gombya-Ssembajjwe, Gibson, McKean and Ostrom2000; Ghate & Nagendra Reference Ghate and Nagendra2005; Ostrom & Nagendra Reference Ostrom and Nagendra2006; Webb & Shivakoti Reference Webb and Shivakoti2008).

Chhatre and Agrawal (Reference Chhatre and Agrawal2008) examined the changes in the condition of 152 forests under diverse governance arrangements as affected by the size of the forest, collective action around forests related to improvement activities, size of the user group and the dependence of local users on a forest. They found that ‘forests with a higher probability of regeneration are likely to be small to medium in size with low levels of subsistence dependence, low commercial value, high levels of local enforcement, and strong collective action for improving the quality of the forest’ (Chhatre & Agrawal Reference Chhatre and Agrawal2008, p. 1327).

The legal designation of a forest as a protected area is not by itself related to forest density. Our research shows that forests under different property regimes (government, private or communal) sometimes meet enhanced social goals such as biodiversity protection, carbon storage or improved livelihoods, but sometimes these property regimes fail to provide such goals. Indeed, when governments adopt top-down decentralization policies, leaving local officials and users in the dark, stable forests may become subject to deforestation (Banana et al. Reference Banana, Vogt, Bahati and Gombya-Ssembajjwe2007). Detailed field studies of monitoring and enforcement illustrate the challenge of achieving high levels of forest regrowth without active involvement of local forest users (see Batistella et al. Reference Batistella, Robeson and Moran2003; Agrawal Reference Agrawal2005; Andersson et al. Reference Andersson, Gibson and Lehoucq2006; Tucker Reference Tucker2008). Thus, it is not the general type of forest governance that is crucial in explaining forest conditions; rather, it is how a particular governance arrangement fits the local ecology and social context, how specific rules are developed and adapted over time, and whether users consider the system to be legitimate and equitable (for a more detailed overview of the IFRI research programme, see Poteete et al. Reference Poteete, Janssen and Ostrom2010, chapter 5).

Co-management and polycentricity

Progress has also been made horizontally by considering multiple idealized governance types at once (as opposed to vertically, digging down into the importance of lower-level institutional variables). It has been increasingly recognized that, even within the three ideal governance types, a mix of these idealized property regimes is likely to be more appropriate in many situations than any single one (Lubell Reference Lubell, Scholz and Stiftel2005; Scholz & Wang Reference Scholz and Wang2006). Indeed, many communities associated with common-property arrangements actually employ a mix of common and private property (Ostrom Reference Ostrom1990). For example, it is typical in community-based irrigation systems for agricultural parcels of land to be privately owned, while the water and pastures are held as common property (Trawick Reference Trawick2001).

Two or three of the idealized governance types can interact in environmental conservation contexts. For example, recent cap-and-trade programmes effectively combine government management and private ownership of the traded rights (Stavins Reference Stavins2008). One important and now firmly established branch of research explores the outcomes that result from co-management of a natural resource between user groups and governments (Carlsson & Berkes Reference Carlsson and Berkes2005; Borrini-Feyerabend et al. Reference Borrini-Feyerabend, Pimbert, Farvar, Kothari and Renard2007; Brunckhorst et al. Reference Brunckhorst, Reeve, Morley, Bock, Pettit, Cartwright, Bishop, Lowell, Pullar and Duncan2008). There is a close relationship between the literature on co-management and the study of institutions as an organizing concept for analysing social systems (Sandström Reference Sandström2009).

Several justifications exist for adopting co-management arrangements to address environmental problems. First, scholars focusing on one level frequently find that processes at another level affect outcomes. Local systems are increasingly affected by external economic events and public policies. This argues for the inclusion of larger governance units when undertaking analysis of small-scale governance arrangements (Agrawal Reference Agrawal2003). Meanwhile, others have noted the difficulties that top-down government or international development agency interventions have faced when they do not take local user groups into account (Lam Reference Lam1998). This finding is also reflected in the IFRI programme. When local users are not involved in any aspect of the planning of a new project, they have no vested interest in the success of the project, and worse, they can directly or indirectly act to undermine it. In contrast, when users are involved, they can potentially use local knowledge to make a governance regime more adaptive, if collaborative arrangements are used to facilitate systematic learning (Armitage et al. Reference Armitage, Berkes and Doubleday2007).

Additionally, particular governance arrangements may have comparative advantages in resolving different types of problems. Local user groups frequently have comparative advantages in gathering and maintaining knowledge of local ecological complexity that would be costly for governments to collect (Moller et al. Reference Moller, Berkes, Lyver and Kislalioglu2004). However, communities may have comparative disadvantages in managing large-scale natural resources and environmental pollution problems (Rose Reference Rose, Ostrom, Dietz, Dolšak, Stern, Stonich and Weber2002).

Berkes (Reference Berkes, Ostrom, Dietz, Dolšak, Stern, Stonich and Weber2002) noted that co-management is not merely a task of assigning components of an environmental problem to various groups of participants and their associated property regimes. Because of cross-scale social and biophysical linkages, the interplay between local and more central governance structures must also be taken into account (Young Reference Young2002; Hill et al. Reference Hill, Williams, Pert, Robinson, Dale, Westcott, Grace and O'Malley2010). Additionally, co-management is not necessarily between a monolithic central government and one coherent community. Rather, it is a more complex arrangement between multiple sources of governance, or what has been referred to as polycentricity (McGinnis Reference McGinnis1999; Ostrom Reference Ostrom and McGinnis1999a, Reference Ostrom and McGinnisb).

A polycentric governance structure may at first seem chaotic. It was a fear of this supposed disorder resulting from the interactions among multiple decision-makers that for decades led many scholars and practitioners to advocate either extreme centralization or extreme decentralization and privatization in order to make decisions the function of one coherent decision-maker (Marshall Reference Marshall2005). However, extensive study of communities of users has established that polycentric arrangements that enable users to develop rules and organizations at multiple levels can work effectively (Ostrom et al. Reference Ostrom, Parks and Whitaker1978; Bromley & Feeny Reference Bromley, Feeny, McKean, Peters, Gilles, Oakerson, Runge and Thomson1992; Ostrom & Parks Reference Ostrom, Parks and McGinnis1999). Coping with all our present environmental problems will likely require similar polycentric governance arrangements across geographic scales beyond those considered by many community case studies (see Brunckhorst et al. Reference Brunckhorst, Coop and Reeve2006).

Institutional frameworks

Over the past several decades, there has been a gradual shift in the literature from an exclusive focus on general property-rights regimes to a more inclusive emphasis on the institutions that affect environmental conditions. Institutions are defined as ‘potentially linguistic entities . . . that refer to prescriptions commonly known and used by a set of participants to order repetitive, interdependent relationships’ where ‘prescriptions refer to which actions (or states of the world) are required, prohibited, or permitted’ (Ostrom Reference Ostrom1986, p. 5). Institutions may be seen as commonly understood codes of behaviour that potentially reduce uncertainty, mediate self-interest and facilitate collective action.

Property-rights arrangements or regimes are one important result of institutional processes that affect outcomes. The three broad property-rights regime categories discussed earlier are a function of a myriad of specific rules that govern a particular setting. A functional common-property regime for a grazing pasture, for example, results from many rules, including those that specify rights and conditions for resource access, monitoring and leadership positions, possible sanctions and demands for contributions (Netting Reference Netting1972). A property right held by an actor results from these rules and enables the actor to take particular actions (Schlager & Ostrom Reference Schlager and Ostrom1992). In this context, the study of property rights becomes part of a more expansive study of the institutions that affect environmental conservation and natural resource management.

Following this transition, the task remains to construct the theories that relate institutional processes, including property rights, to outcomes. We have identified the two extremes in our discussion of the panacea problem: excessively general theories and excessively precise models. Costs are associated with each of these two problems, the first being a lack of meaningful content of overly general theories, and the second being a lack of applicability or accuracy with overly precise models (Cox Reference Cox2008). Young (Reference Young2002, p. 175) stated the resulting task for institutional analysts as follows: ‘What is necessary, at least at this stage, is an intermediate approach that avoids both pitfalls of excessive generalization and limitations arising from the treatment of each environmental problem as unique.’

Several frameworks have been developed to facilitate the construction of such theories and the institutional analysis of environmental management and conservation. Schlager (Reference Schlager and Sabatier2007, p. 294) described the importance of frameworks in scientific progress: ‘Frameworks provide theories with the general classes of variables that are necessary to explain phenomena. As theory development proceeds, frameworks may be revised to provide additional content and specificity to general classes of variables.’ Thus the relationship between framework and theory development is complementary and reciprocal. One institutional framework with which we are quite familiar is known as the institutional analysis and development (IAD) framework.

The IAD framework is best thought of as a metatheoretical conceptual map that identifies an action situation, patterns of interactions, outcomes and an evaluation of these outcomes (Fig. 1). Efforts to explain collective action in field settings with diverse structures, particularly the complex public economies of USA metropolitan areas and common-property regimes around the world, were the stimuli leading to the development of the IAD framework. The IAD framework is consistent with multiple theories and models, but its closest relationship is to the language of game theory, since the working parts of an action situation and a game are intended to be the same.

Figure 1 A framework for institutional analysis. Adapted from Ostrom (Reference Ostrom2005, p. 15).

An action situation is a central part of the IAD framework, and is internally structured by seven working parts, including (1) the set of actors, (2) the sets of positions actors fill in the context of this situation, (3) the set of allowable actions for actors in each position, (4) the level of control that an individual or group has over an action, (5) the potential outcomes associated with each possible combination of actions, (6) the amount of information available to actors, and (7) the costs and benefits associated with each possible action and outcome. These seven attributes of an action situation can be thought of as core micro variables that affect the preferences, information, strategies and actions of participants. Each attribute can take multiple forms that jointly affect the decisions made by actors. The internal structure of an action situation is itself affected by the relevant biophysical world, the community within which it is located, as well as by the specific rules in use. In light of fieldwork and further analysis, seven types of rules are posited to be most important in affecting outcomes in an action situation, as these are the rules that directly affect each of the internal working parts (Fig. 2).

Figure 2 Rules as external variables directly affecting the elements of an action situation. Adapted from Ostrom (Reference Ostrom2005, p. 189).

The IAD framework has been used as the foundation for creating coding forms to be used in an extensive meta analysis of irrigation and fishery cases around the world (see Schlager Reference Schlager1990; Tang Reference Tang1992), for irrigation systems in Nepal (Shivakoti & Ostrom Reference Shivakoti and Ostrom2002) and for the extensive studies of forests undertaken by the IFRI network. As such, the framework has proven to be quite useful, and the body of theory produced, particularly in common-pool resource (CPR) settings, is now extensive (Ostrom Reference Ostrom and Sabatier2007a). However, there have been increasing numbers of calls for a more biophysically sophisticated approach to these settings (Berkes & Folke Reference Berkes and Folke1998; Young Reference Young2002; Agrawal Reference Agrawal2003). It has become increasingly accepted that in order to make progress in environmental conservation theory and practice, an interdisciplinary approach needs to be taken that recognizes the equal importance of social and biophysical variables (Walker & Salt Reference Walker and Salt2006; Brunckhorst Reference Brunckhorst2010). We now turn our discussion to a more recent framework that addresses this issue, among others.

Social-ecological frameworks

Elinor Ostrom (Reference Ostrom2007b, Reference Ostrom2009) has introduced a diagnostic framework for the study of complex social-ecological systems (SESs). SESs are defined by Anderies et al. (Reference Anderies, Janssen and Ostrom2004) as social systems ‘in which some of the interdependent relationships among humans are mediated through interactions with biophysical and non-human biological units.’ A primary interest of scholars focusing on SESs has been examining their ability to sustain themselves in the face of disturbances over time, a feature which has been referred to by a wide range of concepts, including adaptive capacity, resilience, robustness, stability and transformability. These terms are used by various communities of scholars for different purposes, leading to some confusion in their meanings. At times they are used interchangeably, and other times different authors use a term in contradictory ways. An important common element, however, is their focus on dynamics of systems over time. This is an important step forward from the traditional ‘comparative statics’ (see Marshall Reference Marshall2005) approach, and is complementary to the progress in resolving the panacea problem discussed earlier.

McGinnis (Reference McGinnis2010) presented a revised version of the SES framework, composing a SES of several primary classes of entities, which are in turn embedded in a social, economic and political setting, and in related ecosystems (Fig. 3). Resource systems, resource units, governance systems and social systems composed of actors are the primary components of a SES. A SES can have multiple instances of each of these primary components. For example, within a SES there could be multiple resource systems, such as a forest, a lake and a river system. Each of these entities, in turn, has a set of attributes that can take on various values.

Figure 3 Revised SES framework combining the IAD and SES frameworks (Source: McGinnis Reference McGinnis2010).

McGinnis (Reference McGinnis2010) also worked to combine the IAD and the SES frameworks; the primary source of integration is in the component labelled focal action situations (see Fig. 3). McGinnis (Reference McGinnis2010) conceptualized this as a network of interrelated action situations. To illustrate how this concept can organize the study of a SES, we briefly discuss here the case of the Taos valley acequias, a network of community-governed irrigation systems in northern New Mexico (Cox Reference Cox2010).

Illustrating action situations in a social-ecological system

The approximately 51 acequias in the Taos valley (Fig. 4) have an average of 40 members, and have maintained agricultural productivity over a long period of time in a high-desert environment by maintaining high levels of collective action. This cooperation is required in order to build and sustain the irrigation infrastructure that steers snowmelt-derived surface water from the rivers flowing out of the eastward mountains out onto their irrigation fields. Each acequia is organized by a common-property arrangement to distribute its water and maintain its infrastructure, and is governed by an executive mayordomo and a set of three administrative commissioners who carry out these arrangements.

Figure 4 Map of the Taos valley and acequia irrigated land.

The Taos valley acequias use two primary levels of governance: one within each acequia and another between groups of acequias. We can conceive of these as being two different classes of action situations. The primary function of having these two classes is to create a two-tiered governance structure that minimizes the number of actors involved in any one action situation. With fewer actors, transaction costs are lowered and trust and reciprocity are more easily maintained. This is done in two steps. First, the larger SES that the Taos valley acequias constitute is broken up into the 51 acequias, each of which has its own independent decision-making arrangements developed to resolve internal collective-action problems. This creates a set of action situations within each acequia with regard to the important collective tasks that must be accomplished, such as infrastructure provision as a public good and the distribution of water.

These action situations do not, however, resolve collective-action problems among acequias, as many of them are along the same rivers and have the same upstream-downstream relationship as do their individual members with each other. What is needed is a second set of action situations that take place among sets of acequias. The actors that participate in these settings are a subset of the acequia membership, namely the acequia mayordomos and commissioners. Because it is primarily the acequia officers who participate in these water-sharing agreements (repartimientos), the number of actors, and thereby the magnitude of the transaction costs, is also kept low at this second level of governance. Through these repartimientos, the acequias are able to maintain at least some level of cooperation in the distribution of water along the entire reach of a river.

These two types of action situations do not exhaust the arenas of decision-making that are relevant for the acequias. Local courts, for example, have frequently been an important source of conflict resolution for the acequias. This system has sustained the acequias for several hundred years. However, dramatic changes during the past several decades have altered the traditional incentives and disrupted the functioning of these irrigation systems. These changes include emigration of younger members, the introduction of a tourist economy and wealthy landowners, and an accompanying low-wage labour-market into the area.

Continuing to unpack the SES framework

We now discuss the concept of diagnosis and how it may be facilitated by the SES framework. Diagnosis traditionally means ascertaining the nature or cause of something, usually of an unwanted condition. Tied to a diagnosis is the idea that different types of causes have implications for the efficacy of different types of treatments. The possible amelioration of a condition by successfully applying a particular treatment is the motivation for the process. Our interest in the process of diagnosis here is to explore the possibility of its use in overcoming the panacea problem discussed earlier.

The concept of diagnosis is not novel to its application to environmental conservation and natural resource management. The field of medicine has faced a long struggle in its efforts to overcome the recommendation of panaceas, such as various purges and use of aspirin, to solve an immense variety of physical ailments. Over time, biology has developed a rigorous nested approach to the study of the human body and medicine has adopted a diagnostic approach to understanding the problems related to individual health. Instead of offering one or a few general remedies for most problems (as was common in the nineteenth century), a doctor now asks several initial questions about the potential sources of a health problem and then probes deeper using X-rays and other diagnostic tools after assessing information about higher-level indicators (such as temperature, blood pressure, weight change and reported pain levels). Scholars and practitioners concerned with preserving the environment and maintaining its vital functions still need to develop a rigorous nested approach to analysis.

In the previous section we briefly explored some features of the SES framework. One feature we have not yet discussed is its multi-tiered quality. The primary entities (Fig. 3) are the first level of the framework. These are each associated with a set of attributes, which can, in turn, be decomposed into a set of subattributes to form the second level (or third level) of the framework (Fig. 5). These attributes can work to form particular types of entities that possess them, and it would be possible to arrange these types into hierarchical taxonomies. Depending upon an array of attributes, there may be different types of resource systems, for example, or different types of governance systems. There may also be different types of processes occurring within action situations. For example, social and environmental monitoring are two subtypes of the broader process of monitoring.

Figure 5 Unpacking the SES framework into multiple levels.

This typological decomposition, which we leave for future work, is important for the diagnostic approach, because it enables analysis of different types of environmental problems and systems in order to determine what have been the most effective types of social-institutional responses to various types of natural systems and problems. This point was reinforced by Young (Reference Young2002, p. 176), when he discussed the importance of a diagnostic approach in addressing environmental problems: ‘the diagnostic approach seeks to disaggregate environmental issues, identifying elements of individual problems that are significant from a problem-solving perspective and reaching conclusions about design features necessary to address each element.’

Diagnosis and design principles

The SES framework discussed so far is exploratory and tentative. Nevertheless, it can help scholars move beyond the panacea problem by providing multiple levels of analysis. Across levels of the framework, an analyst can determine what is critically important in determining outcomes for one case, or for a whole set of cases. The larger the number of cases under examination, the more general the approach required, using the first or second levels of analysis, in order to produce accurate theories. This observation follows from the principle that at higher levels of aggregation, patterns can be more predictably found across a set of observations (Levin Reference Levin1992, Reference Levin1999). Thus, for a large number of cases, highly specific rules that produce particular outcomes across every case are unlikely to be found. What is more likely is that, when generalizing to a large number of cases, broad social and biophysical attributes associated with general categories of outcomes may be identified.

An example of this process and this principle is the development of a set of institutional design principles (Ostrom Reference Ostrom1990), which characterize long-lasting community-based natural resource management systems. Ostrom (Reference Ostrom1990) had searched for specific rules that might persist across many of these systems, but could not find a specific pattern; however, in agreement with Levin's principle, Ostrom (Reference Ostrom1990) found coarser patterns which were embodied in the design principles. Cox et al. (Reference Cox, Arnold and Villamayor-Tomas2010) have analysed 91 studies produced since Ostrom's (Reference Ostrom1990) original work that evaluate the principles, and found them to be moderately well-supported. We summarize their central findings here:

1. (1) Support for the principles was consistent across the primary sectors examined (forests, fisheries, pasture and irrigation).

2. (2) Support did not differ significantly between empirical studies that explicitly examined the principles and those that examined them implicitly.

3. (3) Conditions stipulated by each individual principle were significantly more likely to be found in successful cases of community-based natural resource management than in unsuccessful cases, and these conditions were significantly more likely to be absent in unsuccessful cases than in successful cases.

4. (4) Much of the criticism and lack of support for the principles came from abstract studies. Empirical studies (case studies, syntheses and large-n statistical analyses), which presented evidence for or against one or more particular principles, were moderately supportive of the principles.

In their review, Cox et al. (Reference Cox, Arnold and Villamayor-Tomas2010) also discussed several prominent critiques of Ostrom's principles that have arisen since their introduction. As implied by the list above, these criticisms were primarily abstract or theoretical. The main empirical criticism was that the principles did not exhaustively list all the relevant factors associated with sustainability, and this is certainly the case. The primary theoretical criticism found in the literature was that the design principles represent just the panacea problem discussed above. The principles are seen by some scholars as an example of the first type of panacea problem: overspecificity. They are seen as representing an overgeneralization to a broad range of cases, abstracting excessively from local context and history of particular cases (Cleaver Reference Cleaver1999, Reference Cleaver2000; Steins & Edwards Reference Steins and Edwards1999; Young Reference Young2002; Blaikie Reference Blaikie2006).

We can use the SES framework and the discussion thus far to better understand this issue. In constructing the principles, Ostrom initially was looking for specific rules that typified successful long-lasting systems. In the absence of such specific patterns, Ostrom resorted to more aggregated conditions that she labelled design principles. In doing so, she traded off specificity for more general applicability.

Several of the principles mentioned by Ostrom are processes that occur within action situations, such as graduated sanctioning and monitoring by accountable monitors. Processes in action arenas result, or emerge, from a configuration of social and biophysical factors, not just one particular rule (Figs. 1–2). This is why it was difficult for Ostrom (Reference Ostrom1990) to find patterns at the more specific level of individual rules; because they are merely one component of a complex configuration producing emergent processes and outcomes (see also fig. 1 in Brunckhorst Reference Brunckhorst2010). When other elements of a configuration change, the effects of a particular rule are likely to change as well. However, the more general conditions stipulated by the principles can be satisfied by a variety of these configurations. The condition of monitoring, for example, encompasses a diversity of configurations of social and biophysical factors that may produce such a condition, and thus can avoid excessive specificity.

Other design principles correspond to important level 2 variables. Along with the other general high-level attributes of the main components of a SES, the design principles can serve as a starting point for a diagnostic approach to the study of complex SESs. These highly aggregated attributes include the idealized property-rights regimes discussed earlier. To conclude, the design principles lack an important degree of specificity. This is a limitation, because they are less precise than a more specific theory could be. It is also a potential strength, in that they may avoid the panacea problem of overspecificity, and serve as part of a basis for a diagnostic approach to analysis.