There are two bases for evaluating between adaptive choices: economic costs and human rights. At the global scale both approaches have already been used to argue for mitigation (Stern, 2006). Lack of agreement on global responsibilities for the distribution of costs of adaptation (which have not been fully calculated but likely far outweigh those of mitigation) mean the case for adaptation has been less forcefully argued using either approach, although human rights has been used to frame accounts of climate change impacts as unjust, for example by the UN Human Rights Commission in its resolution 7/23 (UN Human Rights Commission, 2009).

At the regional level and within countries there is some experience in the use of costbenefit a.n.a.lysis (CBA) as a tool for adaptation decision-making (Splash, 2007). CBA tries to establish the costs of alternative adaptive measures and how much damage can be averted by increasing the adaptation effort given a specific climate change scenario. CBA works for individual sectors where costs and benefits can be derived from market prices; it is harder when multiple sectors are included and when market prices are unavailable for example, in placing a value on human health or wellbeing and where the items being compared are incommensurable (Adger et al., 2009c). Despite such limitations, some sophisticated methods are emerging which can at least show clearly what is known and provide a logical framework for political judgement. For example, it has been suggested that the range of choices for adapting to heat stress in the UK (though not their social and environmental costs, including potential for maladaptation) is likely to be maximised in future global contexts characterised by active free markets and entrepreneurialism, but more limited if strong environmental regulation becomes the norm (Boyd and Hunt, 2006). CBA has also been used effectively to argue for proactive adaptation through investment in disaster risk reduction as an alternative to managing disaster risk through emergency response and reconstruction. The World Bank and US Geological Survey calculate that an investment in risk management of US$40 billion could have prevented US$280 billion in losses during the 1990s alone, a CBA ratio of 7:1. In high risk locations advantages of proactive risk reduction are even higher, Oxfam calculates that construction of flood shelters costing US$4,300 saved as much as US$75,000 a ration of 17:1 (DFID, 2004a). These are compelling ratios but do not allow estimation of costs for specific investments before disaster strikes and in this respect their weight in decision-making is limited.

Given the methodological constraints on economic a.s.sessment for the costs and benefits of adaptation options can ethics help? Caney (2006) argues that people have a moral right not to suffer from the adverse effects of climate change. However, a central dilemma for investing in adaptation based on human rights when resources are scare is whose rights to prioritise. What is the basis on which to decide? Is it fairer to target interventions to reduce risk of climate change impacts and aid adaptation amongst the most vulnerable (as Rawls would argue), or aim to generate the maximum collective good (following the utilitarian philosophy of Bentham). The latter approach may well target those who are only marginally vulnerable. It is justified by the a.s.sumption that the overall increase in wellbeing would provide a resource for compensating those negatively impacted by this decision. The utilitarian approach is one origin of economic costbenefit a.n.a.lysis.

There are many strands to systematic thinking on justice that could inform decision-making for adaptation. The dominance of OECD countries in international policy and the academic literature positions the Western philosophical tradition closer to the existing intellectual core, and the relative potency of justice arguments thus framed. This is not to deny that non-Western philosophies, many perhaps not formalised, will shape local decisions and actions. Indeed their interaction with top-down policy based on Western ideas of justice may be a source of tension or misunderstandings. There are also inspiring and profound differences that can inform questions of sustainability and adaptation from non-Western sources. For example, the Buddhist aim to decrease suffering (including unmet desires) through individual control of the birth of desires (Kolm, 1996) presents a radical departure from dominant Western logics which aim to address perceived need not through individual self-knowledge, chosen restraint and a revelation of happiness, but through the social rights of access, distribution and procedure; or worse through imposed coping and restraint in the worst forms of adaptation. Meeting these Western elements of justice has further been constrained by a framing of the solution in dominant liberal political-economies that a.s.sumes needs must be met through increasing material wealth and energy consumption an error identified by many Green philosophers and lying at the heart of Norgaard's (1995) observation of the lack of sustainability and risk produced by humanity's dangerous coevolution with hydrocarbons.

Returning to the question of how to prioritise resources to support adaptation, a review of Western philosophical traditions suggests there is no simple or single answer. Justice theories distinguish between logics of equality, priority, sufficiency and desert. Egalitarian principles demand that justice be concerned with equality of some relevant distributable elements. Prioritarian principles claim the importance of supporting adaptation for the least advantaged subjects. Sufficientism holds that every subject must have a sufficient, yet not equal, share of support in adaptation. The justness of a society depends on its capacity to give people the support they deserve (Gra.s.so, 2008). Theories are further differentiated by feminist and communitarian arguments that justice is contextual (Konow, 2003) and over the nature of equality. With respect to egalitarian principles, Sen (1987) differentiates between equality in outcomes (equal post-adaptation vulnerability), the meeting of needs (some basic level of security for all) and command over resources (equality in adaptive capacity). Individual principles can be reinforcing strengthening arguments. For example, prioritarian logic is supported by Shue's 'guaranteed minimum' principle of equity (Shue, 1999) which, from a sufficientarian standpoint, states that those who have less than enough for a decent human life be given enough. This general principle of justice has been applied to climate change adaptation to support the argument that interventions prioritise the most socially vulnerable first (Paavola and Adger, 2006; Paavola et al., 2006, Adger et al., 2009c).

Of the approaches outlined above, it is worth spending some more time with Rawls who helps add clarity to the different realms within which justice for climate change adaptation is manifest. Rawls argues that for any social system justice requires both the application of distributional and procedural justice. Rawls made these two elements of justice the cornerstones of his Theory of Justice (1971). Procedural justice talks to the inst.i.tutions and behaviours that frame decision-making, distributional justice talks to the outcomes of these decisions. Under Rawls, a just society is one where procedural justice is embodied in an egalitarian social contract based on reciprocity, so that individual or sectional interests are given les weight than the overriding drive for distributive justice (Chapters 5 and 8 develop the importance of the social contract for establishing justice in adaptation). This understanding of procedural justice places with individual citizens the responsibility for producing specific declinations of equality and defining the basic structures for their society. With this responsibility comes the right to craft and argue for alternative development and adaptation visions. The climate change literature highlights three aspects of procedural justice that it is argued determine the quality of procedural justice (Paavola, 2005; Paavola et al., 2006): * Recognition demands acceptance of minority perspectives in planning and decision-making processes, implying that the views and aspirations of the most marginalised and vulnerable be acknowledged.

* Partic.i.p.ation requires access to knowledge so that all affected parties can formulate informed viewpoints and be involved in the decision-making process with engagement ranging from consultation to local autonomy.

* Distribution relates to whom holds and uses power to ensure equal partic.i.p.ation and recognition of the weakest in decision-making.

This triad can be applied across scales from global negotiations on adaptation regimes to local planning for adaptation in development and together with distributional justice is necessary to underpin legitimacy and popular consent for international, national and local adaptation strategies (Adger et al., 2006).

As with economic a.n.a.lysis, ethics does not provide an easy answer but rather a logic around which options can be discussed with more transparency. Experience from the disaster risk reduction community suggests that while ethical arguments may be useful in the shaping of priorities, once political attention is gained economic based arguments are more persuasive in advocacy.

Three visions of adaptation: resilience, transition and transformation.

Adaptation offers a unique lens for understanding and influencing development, and operates at different levels of engagement with specific social systems. Table 2.4 identifies three levels at which adaptation can intervene in development through enabling resilience, transition or transformation. These three levels are introduced below to provide a framework for a.s.sessing adaptation aims and outcomes and then developed in the following chapters. No level of adaptation is intrinsically more desirable than the others; everything depends on context and viewpoint. Very little in social life is uncontested, so it is unlikely there will be many cases where there is an easy consensus on which form of adaptation is required. Indeed different actors may be working to build capacity and action for adaptation at different levels simultaneously; for example, when local community actors organise to challenge local power asymmetries as part of an agenda for transformative adaptation in a locale which is also the target of government sponsored technical reforms to livelihood or infrastructure provision seeking to build resilience (and possibly mollify local acts of transformation).

Adaptation to build resilience acts at the most contained level, seeking only change that can allow existing functions and practices to persist and in this way not questioning underlying a.s.sumptions or power asymmetries in society. Transformation is the deepest form of adaptation indicated by reform in over-arching political-economy regimes and a.s.sociated cultural discourses on development, security and risk. Transition acts at an intermediary level of engagement, focusing on the governance regime but through acts that seek to a.s.sert full rights and responsibilities rather than make changes in the regime. In a.s.serting rights or undertaking responsibilities that might previously have been neglected or disallowed incremental transformation is a possibility. Each form of adaptation can include changes to values, inst.i.tutions, behaviour and a.s.sets so that it is the scope and range, rather than depth of change that distinguishes each adaptive form.

While it is possible to distinguish individual ideal types theoretically and empirically, for a specific policy domain or social group different levels of adaptation may not be clearly bounded and can influence one another. Transformative adaptation will at a minimum include a critical reflection on existing inst.i.tutions Table 2.4 Attributes of adaptation for resilience, transition and transformation Resilience Transition Transformation Goal Functional persistence in a changing environment Realise full potential through the exercise of rights within the established regime Reconfigure the structures of development Scope Change in technology, management practice and organisation Change in practices of governance to secure procedural justice; this can in turn lead to incremental change in the governance system Change overarching political-economy regime Policy focus Resilient building practice Use of new seed varieties Implementation of legal responsibilities by private and public sector actors and exercise of legal rights by citizens New political discourses redefine the basis for distributing security and opportunity in society and socialecological relationships Dominant a.n.a.lytical perspectives Socio-ecological systems and adaptive management Governance and regime a.n.a.lysis Discourse, ethics and political-economy and practices working at the levels of transition and resilience. Over time, resilient and transitional adaptations may highlight wider challenges, build capacities and weaken barriers for reform and so feed the adaptive transformation of regimes. It is also possible that apparent success at one level of adaptation may hide problems at other levels so that resilience can inhibit transition or transformation. The power of resilience to suppress deeper changes in the inst.i.tutions and values that shape development and risk management is reinforced by its attractiveness as a solution to climate change risks for donors and government precisely because it does not challenge the wider status quo. The technical and organisational innovations required by resilient adaptation are less politically challenging, often more visible and quicker to implement than transitional and transformative adaptations.

Part II

The resiliencetransitiontransformation framework.

3.

Adaptation as resilience.

Social learning and self-organisation.

The ability of a social or ecological system to absorb disturbances while retaining the same basic structure and ways of functioning, the capacity for self-organization, and the capacity to adapt to stress and change.

(IPCC, 2008:880).

The IPCC definition of resilience, presented above, is forward looking, placing emphasis on capacities rather than outcomes of self-organisation and social learning. Within this, adaptation is positioned as a sub-set of resilience (along with functional persistence and self-organisation). Following from this definition, the framework suggested uses the idea of resilience to capture the first kind of adaptation to be discussed in detail in this book. In our use, adaptation as resilience is a form that seeks to secure the continuation of desired systems functions into the future in the face of changing context, through enabling alteration in inst.i.tutions and organisational form.

Elsewhere (Olsson et al., 2006; Nelson et al., 2007) the need to recognise adaptation as including more fundamental shifts has led authors to include the areas of transition (Chapter 4) and transformation (Chapter 5) as sub-sets of resilience. These are not problematic arguments, but the framework presented in this book finds the distinctions so central to the nature of adaptation that separate ident.i.ties are proposed for these three forms of adaptation. This conviction comes from empirical work where imposing resilience in the face of great social inequality is very problematic (see Chapters 7 and 8).

The IPCC definition, and ours, both point at the influence of socio-ecological systems (SES) theory on the understanding of resilience. The three cornerstones of the SES construction of resilience are included: functional persistence, self-organisation and adaptation (if seen as an outcome of social learning) (Folke, 2006). The contribution of SES theory to understanding resilience will be reviewed here and also in following chapters where the elements of resilience described in SES theory contribute to understanding transitional and transformative adaptation. The defining quality of resilience that distinguishes it from transition and transformation is a desire to maintain functional integrity.

This chapter begins by presenting a vision of adaptation as resilience. The contribution of SES theory to this construction of resilience is then examined with a detailed a.s.sessment of social learning and self-organisation. This framework is then combined with organisational management theory to build a framework for examining adaptation as resilience.

A vision of adaptation as resilience.

Resilience seeks to protect those activities perceived by an actor to be beneficial for human wellbeing and ecological sustainability but threatened by contemporary or future pressures a.s.sociated with climate change. The vision of adaptation as resilience is to support the continuation of desired systems functions into the future through enabling changes in social organisation and the application of technology. Such changes are facilitated through social learning and self-organisation (see below) to enable technological evolution, new information exchange or decision-making procedures. More than this, and within the limits of bounded systems, such as development policy for a single watershed or a dairy farming business, achieving resilience may require change in values and inst.i.tutions within managing organisations, and this can include the challenging of established priorities and power relations and potentially lead to a redistribution of goods and bads (Eakin and Wehbe, 2009). In this way, adaptation as resilience has the potential to contribute to incremental progressive change in distributive and procedural justice within organisational structures. When individual cases that build resilience through internal value shifts are upscaled through government action or replicated horizontally, real opportunities can open for contributing to transitional or transformative change in society (see Chapters 4 and 5), though outcomes can be regressive as well as progressive for sustainable development.

Adaptation as resilience can also allow unsustainable or socially unjust practices to persist (Jerneck and Olsson, 2008). This is perhaps easiest to understand in social contexts where entrenched power asymmetries and exploitative economies are manipulated by the elite to maintain power, even when this undermines sustainability. Such outcomes are less likely when local or national decision-making is held to account, but resilience can still undermine long-term sustainability while appearing to meet the demands of adapting to climate change. This can happen when sustainability challenges are recognised but the transactions costs (including political costs) of change are perceived to be higher than doing nothing, with the least bad option being to adapt within available constraints until perceived thresholds of sustainability are breached, forcing change. For example, in the use of desalination plants to compensate for water demand, the proximate need is met but at a cost of high energy use and pollution of the marine environment. The dynamism of climate change and the unpredictability of local impacts provide the additional rationale of uncertainty to justify resilience as the preferred form of adaptation.

The SES science base that has come to influence thinking about resilience in the climate change literature (Gunderson and Holling, 2002) is closely connected to the adaptive management literature outlined in Chapter 2. SES offers a rich and elegant theoretical landscape and one that continues to expand (Liu et al., 2007). Some have pushed resilience theory towards a recognition of transitional adaptation (for example, Olsson et al., 2006) but in this chapter we focus on SES resilience theory contributions to understanding how valued functions can be helped to persist. SES theory emphasises that ecological and social systems are inextricably linked and that their long-term health is dependent upon change, including periods of growth, collapse and reorganisation (Walker et al., 2006b). In addition to s.p.a.ce and time, sociological conceptions of scale also consider how humans symbolise and make sense of reality at different organisational levels (Pritchard and Sanderson, 2002; c.u.mming et al. 2006).

Both a strength and weakness of SES is its presentation as an apparently value neutral, realist epistemology, a product of its origins in systems theory. This has produced a rational and structured framework for understanding human action, one that is particularly attractive to climate change research in offering an approach for integrating human and environmental elements into quant.i.tative modelling of futures scenarios under climate change (Jannsen et al., 2006). A parallel literature that has more recently been brought into an understanding of resilience is that from organisational theory which shares a realist and apparently value neutral epistemology, but is otherwise a much looser body of work sometimes reflecting individual views without being explicitly grounded in a philosophical tradition of enquiry. Organisational theory is reviewed at the end of this chapter, and both literatures are combined in Chapter 6 to a.n.a.lyse the production of adaptive capacity within two contrasting organisational forms.

In thinking through a framework for examining adaptation as resilience built from SES and organisational management theory two limitations inherent in the epistemologies of both approaches must be considered. First, while power is acknowledged, in particular by SES, both literatures are infused with a sense of technical optimism that can downplay the contested character of social life and socio-nature relations. The messiness of decision-making (O'Brien, 2009) is not easily captured. Apparent value neutrality in both cases conspires with technical optimism to emphasise technological innovation and efficiency over critical a.n.a.lysis that might place more weight on the political-economy and cultural root causes of risk and its perception. In this way SES theory has been criticised for a weak integration of social science theory and a tendency to allow for an oversimplification of complex social phenomena (Harrison, 2003; Jannsen et al., 2006). Second, and related, both approaches focus on relational social s.p.a.ce but limit a.n.a.lysis to the outer world of interactions between individuals, groups and inst.i.tutions. Inner worlds of emotion and affect value, ident.i.ty, desire, fear that give shape or meaning to, as well as being drivers for, public actions including adaptation choices (Grothmann and Patt, 2005) are difficult to include.

Framing of resilience.

Thinking on resilience within climate change has been influenced by two schools: disaster risk and SES. Disaster risk itself includes varied interpretations of resilience including as a capacity for absorbing disturbances and shocks (Birkmann, 2006) and as the opposite of vulnerability, capturing all those acts and capacities that seek to reduce vulnerability to risk (Adger et al., 2005c). More recently both disaster risk and climate change have been influenced by SES theory so that an additional reading of resilience in the face of natural disasters and climate change has become a.s.sociated with systems regenerative abilities and capacity to maintain desired functions in the face of shocks and stress (Birkmann, 2006), the meaning used here. In this way SES has acted as a bridge between climate change adaptation and disaster risk theory (and with wider literature on natural resource management). Both interpret adaptation as a process as well as a product of social relations and as a dynamic property such that adaptive capacity can change over time in response to shifting risks and capacities (Pelling, 2003b; Young et al., 2006). Arguably another commonality is a failure to question the framing values and political context of decision-making and fall short of addressing adaptation as transformation (Manuel-Navarrete et al., 2009).

Adaptive capacity then is best indicated not by goodness of fit to current or predicted future threats but by flexibility in the face of unexpected as well as predicted hazards, vulnerabilities and their impacts (Janssen et al., 2007). This opens questions about the trade-offs to be made between flexibility, adaptation and welfare (Nelson et al., 2007). Walker et al. (2006a) argue that adaptation can undermine net resiliency by shifting resources and so decreasing capacity or increasing risk in another place or sector, and through over-adaptation and lock-in such that a system becomes unable to adapt to novel threats. For example, in southeastern Australia rounds of engineering based solutions have been used by government to respond to a rising water table and salination. This has created a state of lock-in, making it increasingly difficult for the management system to conceive or invest in a non-engineering response. A highly adapted but fragile system is the result one that is vulnerable to collapse through dependent coevolution (Anderies et al., 2006), an example of Handmer and Dovers' (1996) account of resilience as resistance and maintenance.

SES theory on resilience applies thresholds to describe movement from one systems state to another (see Chapter 2). This helps theorise what it is that leads one system to respond to the local impacts of climate change risk through resilience and another through transition or even transformation. Empirical work shows that identifying the location of thresholds before change is difficult because of the multiple and non-linear feedback mechanisms active within SES, so that the ways discourse, inst.i.tutions and practical action interact are not always transparent or predictable (Nelson et al., 2007). However, evidence does indicate that to activate adaptive capacity requires a social or environmental trigger (a change in att.i.tudes, policy, market conditions or environmental risk and impact) and the appropriate inst.i.tutional framework.

Nelson et al. (2007) contrast deliberate and inadvertent crossing of thresholds from resilience into transition. They argue that deliberate crossing is an indication of both greater adaptive capacity and higher levels of resilience. Two case studies are compared to reach this conclusion. Deliberate transition from agriculture to tourism is exemplified through the actions taken by a local authority in Arizona, USA, in changing its development strategy and support from local agriculture to tourism base. Inadvertent transition is noted in the abandonment of an agricultural economy in Jordan precipitated by unsustainable resource use. The Arizona case shows an actor overcoming the inertia inherent in an established system to move into a more advantageous economic position. No clear point of movement is identified, however, to mark the change from resilience to transitional adaptation, although it is suggested that while both resilience and transitional adaptation rely on the same kind of adaptive capacities it is social systems with greater intensity of vertical organisation (such as a functioning system for information exchange and partic.i.p.ation in development planning from local to regional and national levels of government) that are more likely to be able to cross thresholds into transitional adaptation.

Two elements of SES resilience theory that deserve closer attention are social learning and self-organisation. These ideas have been paralleled in other literatures for example, self-organisation in social movements, partic.i.p.atory and communicative planning (Pugh and Potter, 2003) and much of the emphasis on trust and relationships that underlies social learning echoes work on social capital which has also been applied to adaptation (Adger, 2003; Pelling and High, 2005). To this extent these ideas represent widely accepted social phenomena key to the understanding of any collective dynamic. They are at work within transitional and transformative as well as resilient adaptations; the distinction between these levels being the subject and context rather than the object of a.n.a.lysis.

Social learning.

Social learning is a property of social collectives. It describes the capacity and processes through which new values, ideas and practices are disseminated, popularised and become dominant in society or a sub-set such as an organisation or local community. The outcomes of rounds of social learning are the common values, beliefs and behavioural norms that shape the inst.i.tutional architecture of social life (Wenger, 1999). Social learning is also ascribed to the socialised process of learning and a.s.sociated change. This is clearly seen in differences over scale where local worldviews or value systems fit uneasily within dominant discourses of development or culture (Argyris and Schon, 1996). Such diversity can be a resource when alternative behaviour is well suited to meet the challenge of changing environments, but also a compounding factor in inst.i.tutional inertia and potential barrier to the flexibility needed for resilience (Olsson et al., 2004). At the heart of the contribution of social learning to studies of adaptation lies a tension between dominant and alternative or novel ways of seeing and being, and the potential this opens for individual social actors to shape the trajectory and content of collective learning (challenges for the use of social learning as an a.n.a.lytical tool are discussed in the case studies presented in Chapter 6).

Much of the literature on social learning is interested in improving the efficiency of established practices rather than seeking new practices to resolve underlying sustainability challenges, and in this way it meets the goals of resilience (Armitage et al., 2008). A smaller literature examines the role of social learning in enabling transitional and transformational adaptation. According to Diduck et al. (2005) such changes that focus on reform of inst.i.tutions and organisational frameworks are characterised by: * high levels of trust, * willingness to take risks in order to extend learning opportunities, * transparency required to test and challenge embedded values, * active engagement with civil society, and * high citizen partic.i.p.ation.

Transformative adaptation that builds on alternative values connects individual to social learning personal beliefs to culture. In thinking through the relationship between learning and political change, Freire (2000 [1969]) calls for critical reasoning. This, Freire argues, is not the default orientation for problem-solving held by the marginalised or powerful both prefer to make adjustments within the confines of established norms and structures. Freire terms this kind of problem-solving adaptive ingenuity finding new ways to fit within and gain advantage from dominant structures without challenging them. Thus critical reasoning is a necessary factor in transformational adaptation but will be absent or marginalised into silence in resilience. There are significant obstacles to be overcome in promoting critical reasoning. The powerful as well as marginalised and vulnerable can be frightened by the uncertainty of change, and change itself can be captured by vested interests. To offset this Freire calls for transformation to come from a dialogue between the marginal and powerful and also between ideas and practice (Freire, 1970). Notions of transformation that have shaped partic.i.p.atory development and specific tools including citizen's fora, citizen's budgets and deliberative decision-making and polycentric governance (Bicknell et al., 2009).

Responding to the novel hazards of climate change requires social learning systems that can respond to the multiple scale and sectors through which risk is felt and adaptations undertaken. Not least to address the challenge of integrating local community level and scientific knowledge and balance strategic thinking with local needs so that decisions are taken at an appropriate level in the organisational hierarchy (Cash and Moser, 2000). Bringing together and making use of local and scientific knowledges is not easy. It is difficult for individuals and organisations to handle both kinds of knowledge and this is exacerbated by inbuilt power imbalances that tend to give greater weight to science over local knowledge (Kristjanson et al., 2009). In response, Wenger (2000) has called for organisations, individuals or tools that can work across this epistemic divide so called boundary objects.

Self-organisation.

Self-organisation refers to the propensity for social collectives to form without direction from the state or other higher-level actors. This can include new canonical (formal) organisational forms such as registered community development groups or trade a.s.sociations, and shadow (informal) organisations such as networks of friends and neighbours that work independently to or cross-cut canonical organisation. Most research on organisations and adaptation focuses on canonical forms which are visible and easy to access, exemplified by the literature on adaptive management outlined in Chapter 2. However, the generation of novel ideas or practices that are in conflict with or undervalued by canonical organisation often first emerges from the unmanaged s.p.a.ce of shadow organisations. Shadow systems are supportive of innovation because they are typically rich in trust, cut across canonical organisational structures and are hidden from formal oversight, allowing experimentation and risk-taking with novel ideas and practices (Shaw, 1997). Successful experiments in shadow systems may in turn become coopted and formalised within the canonical system. This can provide opportunities for the replication of adaptations, but through formalisation of individual roles and relational commitments will limit flexibility and change the social relations which led to the original innovation and potentially undermine long-term sustainability. Alternatively, shadow systems can remain marginalised and informal, operating in parallel with canonical systems. This is especially so under transitional and transformative adaptation where emergent forms are a site for the challenging of established discursive and material power (Pelling et al., 2007).

Self-organisation can evolve slowly in response to changing social values and organisational forms driven by demographic shifts or changes in popular ideology, but also more rapidly. This latter opportunity has been observed following disaster events when new forms of social organisation emerge as dominant forms fail (Pelling and Dill, 2009). Emergent organisation ranges from spontaneous solidarity as neighbours undertake first response, to coordinated networks of NGOs and state agencies in recovery (see Chapters 5 and 8). Capacity to self-organise, like social capital, is particularly difficult to measure in society for this reason much capacity is hidden and latent, its emergence dependent upon wider social and political context and the nature of threats and opportunities presented to society. It is not possible to measure capacity for self-organisation from existing organisational forms alone. Berkes (2007) notes that because social capital can remain latent in society, social relations that might have been used in the past can be reinvigorated as new threats or needs arise. This was the case in Trinidad and Tobago when networks originally established to deal with coral reef management in Trinidad and Tobago have also played a key role in disaster preparedness (Adger et al., 2005b). Existing organisational forms can also serve to hinder the emergence of novel self-organisation through inst.i.tutional inertia so that observed high levels of organisation may not alone indicate high levels of capacity for self-organisation to respond to future climate-change-related pressures.

Social learning and self-organisation reinforce each other so that a social system exhibiting rich capacity for social learning is also likely to have considerable scope for self-organisation. The extent to which social learning can be fixed through self-organisation is tracked through three elements of capacity to change: consciousness, inst.i.tutionalisation and implementation. Consciousness is the capacity to reflect on the outcomes of and alternatives to established norms and practices and sets the limits for subsequent alternative visions or discourses. Bateson (1972) described this as dutero-learning making learning to learn an act of adaptation. There is no normative a.s.sumption on the scope or depth of learning so that this can include adaptive ingenuity and critical consciousness. Inst.i.tutionalisation is the capacity to move from recognition of constraints to affect change in the inst.i.tutional architecture that frames implementation through the reproduction of existing, or insertion of new, values and practices. All three processes can unfold within the canonical and shadow systems. Their interaction provides reinforcing or contradictory realms for experimentation and learning and for novel values and practices to emerge (Pelling et al., 2007).

Organisations as sites for adaptation.

Organisations operate at scales from the household to firms and national and international bodies. They are often seen as agents in the construction of adaptation for subsidiary actors for example, by enforcing environmental management regimes or regulating land markets but in this and the subsequent chapters we also focus on the ways in which internal social relations shape information flow, agency and the direction an organisation can take in adapting itself to a changing external environment. An important distinction is to be made between organisations and inst.i.tutions.

Following North (1990), inst.i.tutions are defined as the rules of the game (formal and informal) that influence adaptive behaviour. Organisations are the collective units, embodying inst.i.tutions, that are vehicles for adaptation. Organisations are not monolithic; they contain potentially competing agents and interests so that internal adaptive change brings new risks as well as opportunities which are not experienced evenly within organisations even when the stated focus of change is on the external environment. Internal differentiation also means that adaptations undertaken by individuals are not always replicated throughout the organisation with consequences for efficiency as well as equity in adaptation. This is the case for households, firms and public or civil society organisations alike.

If organisations and individuals or social groups within them can learn, how might learning be observed? At a surface level, learning is observed through changes in behaviour (signifying implementation and a.s.suming consciousness and inst.i.tutionalisation). In addition to this behaviouralist Gross (1996) focuses on externally validated, physical behaviour, and following Maturana and Varela (1992) and Ison et al. (2000), internal actions are also interpreted as learning. That is, we can learn in relation to different modes of interacting with the world: emotional and conceptual as well as physical. Our learning corresponds to differences in the way that we act (consciously or unconsciously) within these modes, which in turn arise in response to our ongoing experience. The judgement of what const.i.tutes behaviour lies with the observer in question, but the definition does not rule out internal and tacit activities such as conscious or unconscious cognition, emotional affect or the formation and operation of personal relationships, for example. Identifying different realms of behaviour is important in sharpening our focus on the site(s) where adaptation can be observed; not only in material actions, but in contrasting att.i.tudes or views that have not been allowed translation into action. In this way Pred and Watts (1992) identify the behaviour of marginalised actors who need to keep low visibility in the face of surveillance by more powerful actors, and the potential importance of private language as a mechanism for resistance that could form a potential resource for adaptation when organisational relationships or external contexts change.

Constructing the learner as an individual or social ent.i.ty links individual learning to social processes of change that emerge at the collective level. Thus social adaptation can be seen as collective learning. This is not a claim that individual and collective behaviour are qualitatively the same, but recognises the interaction of learning and adaptive behaviour at these different levels. In this way, adaptation to climate change and variability can be read at different levels of learning operating as a range of system-hierarchic scales the behaviours of components and subsystems of the system, as well as changes to the emergent properties of the system and this can be used to unpack different adaptive trajectories: international, national, local. It may be that adaptive behaviour emerging at one scale say the local is the result of learning that has been ongoing amongst a range of actors networked across a range of scales. Additionally, adaptation at one spatial (or temporal) scale can impose externalities or constrain adaptive capacity at other scales. In short, the system-hierarchic scale where adaptation is or is not enacted is a socio-political construction (Adger et al., 2005a).

Organisations are s.p.a.ces of engagement where learning and adaptive capacity can be constrained as well as enhanced (Tompkins et al., 2002). A useful distinction is between organisations and communities of relationships acting within or across them in supporting, antagonistic or ambivalent ways. Communities describe those collectives through which close relationships reinforce shared values and practices; although reinforcement may not necessarily contribute to the organisation's (or even the community's) adaptive capacity, it can lead to closed thinking and the suppression of questioning established norms a property referred to as groupthink by Janis (1989). In a less formal context a similar phenomenon is described by Abrahamson et al. (2009) who observed how closed social networks amongst the elderly led to the self-reinforcing of myths of personal security regarding vulnerability to heatwaves in the UK as new information or ideas were treated with caution.

For Wenger, learning in a community arises through partic.i.p.ation and reification, the dual modes through which meaning is socially negotiated. Partic.i.p.ation refers to 'the process of taking part and also to the relations with others that reflect this process. It suggests both action and connection' (Wenger, 2000:55). Partic.i.p.ation is thus an active social process, referring to the mutual engagement of actors in social communities, and the recognition of the self in the other. Reification is the process by which 'we project our meanings into the world, and then we perceive them as existing in the world, as having a reality of their own' (Ibid., 58). Thus reification can refer to the social construction of intangible concepts as well as the meanings that members of a community of practice see embedded in physical objects.

Communities of practice are often not officially recognised by the organisations they permeate (Brown and Duguid, 1991). Their official invisibility in the shadow system can be thought of as being made up of constellations of communities of practice held together by bridging ties of social capital. The link between communities of practice, informal networks and unofficial activity in organisational settings is an important a.s.sociation to make in tracing the workings of the shadow system in building adaptive capacity. Wenger (2000) uses the language of social capital to help define the characteristics of individual communities of practice, which, he argues, can be defined by a shared ident.i.ty and held together by bonding capital. The influence of personality traits and the role of personal and professional sources of trust in bridging across communities within the public sector are discussed by Williams (2002). It is the quality, quant.i.ty and aims of individuals connected together in communities of practice, and their linking of boundary people and objects, that determine the influence of the shadow system on adaptive capacity.

Pathways for organisational adaptation.

Figure 3.1 summarises the preceding discussion by identifying five pathways through which adaptive action can be undertaken by individuals or discrete subgroups within an organisation. These actions are a consequence of interactions with the inst.i.tutional architecture, adaptive capacity and facility for learning held by an organisation. The host organisation can be any collective social unit, from a legally mandated government department or agency to more flexible but nonetheless restrictive private sector organisations or loose a.s.sociations of actors from civil society in organised networks.

Feint arrows indicate the direction of conditioning between these aspects of organisational life, such that the inst.i.tutional architecture conditions the type and scope of learning and adaptive capacity (either directly or through prescribed forms of learning) and adaptive agency through the setting and policing of formal objectives, structures and guidelines for practice in the organisation. Adaptive capacity conditions adaptive actions by setting constraints on what can be thought and done as well as the goodness of fit of existing resources to the identified external pressure. The organisation draws resources, opportunities and threats from the external environment and exerts adaptive actions upon it.

Only two of the five forms of adaptation are visible from outside the organisation. This not only flags the need to examine interior life of organisations but also the overlapping of resilience, transition and transformation with the latter two modes of adaptation potentially being found operating inside an organisation which itself applies only resilience to the external world; indeed transition or transformation might be necessary internal motors for external resilience (see Figure 3.1).

The five adaptive pathways are characterised in Table 3.1. Adaptive agency refers to that held by individuals, sub-divisions or cross-cutting communities of practice below the level of the organisation. Such agency is made reflexive through learning (pathway 1) which is itself an adaptive act. Reflexivity implies strategic decision-making (of an entrepreneurial individual, advocacy coalition and so on within the organisation). This can be focused on symptoms leading to adaptive ingenuity, or consider root causes indicating to critical reasoning. Learning through critical reasoning or adaptive ingenuity can lead to lobbying for change in the inst.i.tutional architecture of the organisation (2). This can in turn force reconsideration of aims and behaviour informing the selection and use of resources that form adaptive capacity (3) (a key concern given a dynamic external environment). Feedback from capacity and inst.i.tutional architecture then potentially reshapes adaptive agency.

One important message from this approach to adaptation is that only two modes of adaptation are observable from outside the organisation. These are efforts to shape the relationship between the organisation and operating environment made either by the agent (4) or organisation (5). This is a strong argument for studies of adaptation to climate change to extend their a.n.a.lysis from the external/public to more internal and private life of organisations of all kinds. The Figure 3.1 Adaptation pathways within an organisation.

Table 3.1 Five adaptive pathways.

Pathway.

Summary.

Example.

1 Agent-centred reflexive adaptation.

Adaptation informed by dutero-learning reflecting on past actions. This can lead to changes in the practice of learning, management of adaptive capacity, the inst.i.tutional architecture or directly on the external environment.

A manager decides that existing work guidelines undermine sustainability and so implements reform.

2 Agent-centred inst.i.tutional modification The agent undertakes to alter the inst.i.tutional context within which it operates so as to shift the inst.i.tutions which control its scope for future adaptive capacity and action.

A scientific advisor lobbies policy-makers to change policy priorities.

3 Agent-centred resource management The agent unilaterally changes the selection or use of resources to undertake predetermined adaptive action.

While no guidelines exist, a manager adjusts work routines to meet a changing environment.

4 Agent-led external action The individual agent undertakes adaptive action oriented to the external environment either in compliance with existing inst.i.tutions or as part of the shadow system; that is, without instruction from the organisation.

Local agents make experimental or spontaneous modifications to physical infrastructure.