What Does Transformation Look Like? Post-Disaster Politics and the Case for Progressive Rehabilitation

: This paper responds to the ‘deliberate transformation’ discourse within climate change and disaster scholarship. It calls for a cautious approach to deliberate transformation as a practice space for non-governmental organisations (NGO), arguing that greater clarity is still needed on precisely what form these transformations may take, how and where they might be stimulated, and—most importantly—who decides. Drawing on a case study of post-disaster NGO interventions in the Andaman Islands following the 2004 Indian Ocean tsunami, the paper analyses the role of community-based, rights-oriented education and advocacy programmes in opening space for communities to critically reﬂect on state responsibilities, capacities, and weaknesses. It argues that these shifts were potentially transformative for the ways local people interact with the state in Little Andaman; however, it also surfaces pragmatic and ethical challenges to the notion of ‘deliberate transformation’ in practice. The data reveal transformations to be complex, spontaneous, and fundamentally political, with simultaneously divergent pathways of change and re-entrenchment; this presents a deep challenge to external agencies seeking to instigate transformation in planned, linear ways over known timescales. A further challenge relates to deliberate transformation as a normative endeavour. The paper argues that any attempt to actively instigate deliberate transformations according to pre-determined visions held by external actors is a direct contradiction to the principle that progressive transformation—as deﬁned in the literature—should be shaped deliberatively by the values and priorities of citizens themselves. To avoid diluting the radical power of transformation as a principle, the paper proposes progressive rehabilitation as an alternative approach in post-disaster contexts, requiring a transformation of the NGO itself from ‘doing to’ to ‘doing with’ citizens, with an emphasis on supporting locally-deﬁned futures. The paper thickens the conceptualisation and evidence base for transformation pathways, with implications for research and practice.


Introduction
This paper builds on the deliberate transformation discourse in climate and disaster literature, which is increasingly emerging in international disaster risk reduction, adaptation, and humanitarian practice. The paper questions deliberate transformation as a practice space for non-governmental organisations (NGOs) working in these sectors, arguing from a critical development perspective that as a normative project, transformation must be approached with caution. This is not to say the paper disagrees with the idea of transformation in principle: quite the opposite. The paper adopts a social transformations approach, defining transformations as critical junctures wherein spaces of possibility for emancipatory social-political change are opened. In line with Manuel-Navarrete [1], Pelling [2], and other critical scholars working at the adaptation-development nexus [3], the paper takes as Moving forward, the paper suggests that rather than aiming for any single, pre-defined vision of deliberate transformation, it may be better for non-governmental agencies in post-disaster contexts to plan for progressive rehabilitation-requiring a transformation of the NGO itself from 'doing to' to 'doing with' local people. The emphasis in this approach is not to impose set visions of change, but rather to support the emergence of locally-defined futures. This shift in focus may help prevent transformation becoming a vehicle for the further subjugation of communities, and protect the language of transformation from ideological or political co-option.

The Case for Transformation
The literature on transformation has burgeoned over the past five to 10 years, driven by the argument that global environmental change constitutes a drawn-out, protracted environmental crisis which necessitates-yet also opens opportunities for-fundamental shifts in the ways that we practice development and/or govern our relationship to the environment [10][11][12]. Definitions of transformation are diverse; for a comprehensive review, see [11]. Table 1 summarises four dominant approaches to transformation, building on O'Brien and Sygna [12]: transformational adaptation, transformations to sustainability, transforming behaviours, and social transformations. Several frameworks have distinguished the various sites in which transformative change can occur, including O'Brien and Sygna's [12] three 'spheres' (practical, political, and personal), and Pelling et al.'s [13] seven 'activity spaces' (the individual, technology, livelihoods, discourse, behaviour, the environment, and institutions). Table 1. Differing approaches to transformation (building on O'Brien and Sygna [12]).

Approaches to Transformation Object of Change; Key Goals Example References
Transformational adaptation Going beyond incremental adaptations to climate change; managing pathways of change. [14,15] Transformations to sustainability Deep changes to the structure, function, and governance of energy, transport, agriculture and other systems; focus on social-ecological systems analysis.
[ [16][17][18] Transforming behaviours Concerned with psychological and cognitive barriers; also the role of held values, self-efficacy and individual capacity to become an agent of change. [19,20] Social transformations "Transforming the political, economic, and social structures that maintain the systems associated with increasing risk and vulnerability intact" [12] (p. 4); political economy grounding. [12,[21][22][23] The most radical of the above approaches is social transformations, which positions social, economic and political structures as the necessary site of change for fairer, lower risk, more inclusive and progressive development futures. It is now well-recognised that the root causes of disaster vulnerability and recovery lie in the social-political reproduction of inequitable and uneven development [24][25][26][27]. This has led critical disasters scholars to argue that in order to meaningfully reduce vulnerability to hazard, enhance recovery capacity, and improve human well-being and equality, a fundamental renegotiation of development trajectories is necessary, wherein not only the practices of development are re-negotiated but also the values and power relations that underlie and reproduce them (for example [2,5,25,28]). Of the above approaches, social transformations are arguably the most urgent (yet the most challenging), since the perpetuity of unequal social structures may well undermine progressive change in other spheres.
The social transformations perspective is in part a reaction against the conservatism of the resilience discourse that now dominates disasters, climate, and sustainability literatures. Resilience sceptics have argued that it focuses disproportionately on preserving the status quo rather than critical reflection on the inequalities, governance weaknesses, and gaps in provision that led to disaster risk in the first place [2]. Those with a political economy leaning argue that resilience-based approaches often serve to protect (not challenge) dominant modes of production-the same modes of production that simultaneously produce risk (via uneven development and inequality) and profit from it (via the vast aid and relief agency, or what Naomi Klein [4] calls 'disaster capitalism') [5,29]. This is particularly pertinent in contexts where Mustafa [30] argues, "the pre-disaster 'normal' conditions may be characterised by such levels of injustice, poverty, and environmental degradation", wherein the most pertinent question to be answered in designing post-disaster interventions is: "recovery to what?" (p. 71).
Pathways of transformation are a topic of growing research interest across multiple scholarships, particularly in adaptation (e.g., [11,31]), social-ecological sustainability (e.g., [23,32]), disaster risk reduction (e.g., [33]), and development geography (e.g., [34,35]). This crossover is welcome, since reducing social-ecological risks, adapting to dangerous climate change, and meeting human development objectives "can only be accomplished if they are undertaken in an integrated manner" [36] (p. 19). There's been an emphasis on the role of disasters as 'tipping points' for transformative regime shifts [9], and the role of disaster risk management as a vehicle for altering the speed and trajectory of development-what Gibson et al. [37] termed transformative disaster risk management. O'Brien [11] amongst others has called for shifts in social contracts as a way to pro-actively reconstitute governance frameworks in ways that benefit the poor and vulnerable-what she terms 'deliberate transformation'.

Transformation in Practice
Whilst O'Brien [11] makes a compelling case for the need for deliberate transformation, she also raises a key challenge in terms of precisely "what exactly do we mean by transformation? What types of transformations are considered necessary . . . [and] who decides?" (p. 4). As has been argued in the adaptation literature, while few would disagree that altogether fairer, more prosperous, and less precarious futures are generally desirable, there are strongly divergent views on how we get there, and a contested politics around whose voices and priorities count [38][39][40][41]. Some scholars have explicitly sought to distinguish between transformations as a normative goal (i.e., a particular change viewed as necessary by a particular stakeholder), and transformation as a process signified by the magnitude of change rather than its direction per se (see [2,37]). A magnitude-focused framing can help facilitate discussion across leftist/rightist political lines. However, in practice, it seems difficult to imagine how a process of deliberate transformation-implying a conscious, intentional activity-could be anything but normatively driven. As O'Brien highlights, this raises acutely the question of: deliberate transformations by, and for, whom?
Despite the answers to these questions being far from clear, the language of transformation is already gaining ground in the disaster risk-relief-recovery practitioner sphere. INGOs that are embedding the language of transformation into their practice already include Oxfam [42], ActionAid [43,44], and CARE [45]. These organisations' ostensive commitment to addressing root causes of vulnerability is exciting. However, they also render grounded research on what works and what doesn't, how the politics of transformation are negotiated, and the ethics of transformation as a normative agenda, as increasingly urgent.

Transformations of, by, and for Whom?
Much of existing literature on transformation pathways is systems-oriented, strongly influenced by social-ecological systems (SES) thinking in adaptation and sustainability scholarship [2]. However, such a priority can obscure the power relations, politics, and vested interests that either present obstacles to, or favour pre-determined pathways to, transformation [21]. A systems focus can tend towards abstraction and obscure the micro-politics operating at an everyday scale. As Loftus [46] argues from an urban political ecology perspective, "sensing [an] alternative world" occurs in "shared acts of making the world, [when] people hear, feel, and begin to touch the possibilities for making things differently" (p.x). This paper does not engage with the literature on everyday political ecologies in depth, but its scale of analysis is inspired by Loftus' suggestion that the everyday is an essential scale for understanding transformative pathways. Analytically, this leads to an emphasis on the intricate, fluid, and mutually constituted relationship between an individual's sense of self (internal) and their social interactions and practices (external), and the political realities that shape and are shaped by this dynamic.
Participatory development geographers have long argued that local action can be an engine for progressive social-political change, when situated within a wider commitment to a radical politics of development and an evolved, responsive governance framework [34,47]. Participatory, plural, and deliberative governance is posited to have multiple benefits, including promoting societal equality and fairness through plural deliberation and representation, taking account of local and traditional forms of knowledge, and increasing the legitimacy of decision-making. From a social justice perspective, there is intrinsic value in local people having the capacity to self-determine (or at least participate in defining) the development pathways that govern their future. Although there has been a strong backlash against tokenistic and superficial attempts at participatory development (particularly when undertaken by NGOs), active forms of citizenship remain central to visions of more locally-empowered and emancipatory development futures [48,49]. Growing attention to the role of subjectivity [50] and Paulo Freire's concept of conscientisation [51] (drawn on by [2,11,21]), reflect a recognition that the transformation of the political self is a necessary seed for broader structural change. However, to date, grounded empirical research on local pathways to transformation has been limited. As O'Brien observes, "the relationship between consciousness and individual and collective transformative action has been largely ignored" [11] (p. 672).
This is not to say that climate change adaptation (CCA) and disaster risk reduction (DRR) scholars have ignored the individual and local scale. To the contrary, many have emphasised the importance of recasting vulnerable individuals from 'victims' to "actors with the capacity to contribute to the management of their vulnerability levels" [52] (p. 175) (see also [53,54]. Wisner et al. [25] were early to argue that "any recovery programme that shifts power and control back into the hands of households and localities" is more effective in addressing structural "root causes" of vulnerability than projects conceived and implemented solely from the top-down (p. 365). However, research on participatory DRR and CCA to date has often focused on localised studies of NGO project interventions and their challenges. The stifling effect of centralised governance, rigid or unresponsive institutions, a lack of political will, and poor channels of accountability on community-based DRR/CCA action, are also well documented (e.g., [25,52]). This work has been essential in demonstrating the fundamentally political nature of disaster risk; however, a gap remains around how local people gain and exert agency to resist or renegotiate those stifling and risk-producing governance and development norms. The challenge ahead is highlighted by Scoones [23] and Stirling [22], who each argue transformation pathways will be political, messy and "unruly" (p. 293 and p. 1, respectively), involving the active and embodied contestation of existing power relations through local action. Inspired by Williams' [55] approach to participation, this paper calls for relational and politically-oriented research on pathways to transformation from the bottom-up.
Taking this agenda forward, the following case study follows Blackburn's [56] attempt to situate participatory DRR spaces within a wider politics of scale. The case study examines the outcomes of disaster relief and recovery interventions by NGOs following the 2004 Boxing Day tsunami, responding to the suggestion in disasters scholarship that disasters can open a 'window of opportunity' for transformation [57]. This research asks what role there is for post-disaster community-based NGO interventions in deliberate transformation (if at all), and how transformation from the bottom-up might best be supported-thickening understanding of what Cretney [58] terms the "politics of hope and possibility" (p. 4) in the post-disaster space.

Data Collection
The paper comes out of an ESRC-funded research project on pathways of social-political change through post-tsunami relief and rehabilitation in the Andaman and Nicobar Islands (South India). Little Andaman was selected as case study because: (i) the island was severely affected by the tsunami yet minimally documented; and (ii) the island's governance landscape was previously devoid of NGO intervention, making the impact of those agencies clearer to observe. Qualitative fieldwork, in the form of interviews with tsunami-affected persons and key stakeholders (political and administrative staff, NGO staff, community leaders) and observation at public NGO meetings, was conducted over nine months in South India in 2014. Interviews explored imagined and performed political identities, everyday state-society interactions, forms of local associational life and activism, and channels of local leadership and representation. The analysis focused on changes post-tsunami and identifying specific local pathways of change. Throughout the analysis, names and identities are protected for confidentiality. The notation Rx (where x is a number) is used in-text to refer to a specific interview number (or to distinguish between respondents, where used within a quotation). The notation POx denotes observational data.

Post-Tsunami Relief and Rehabilitation: the Andaman Islands
The 2004 Boxing Day tsunami brought an unprecedented number of non-governmental aid agencies to the Andaman and Nicobar Islands [59], which historically have been closed to foreign organisations on account of the region's inaccessibility (over 1000 km east of the Indian mainland), paternalistic governance (the islands are home to six heavily protected indigenous tribes), and geopolitical sensitivity (located in a strategic military and trading position) [60]. The Andaman and Nicobar Islands are a Union Territory of the Government of India, governed by an appointed Lieutenant Governor and Secretariat, and under them three District Commissioners (one each for North Andaman, South Andaman, and Nicobar districts), and a Tehsildar for each island or other territorial sub-division or Tehsil. In terms of political (elected) governance, the territory has one Member of Parliament (MP) and mirrors the mainland's local Panchayati Raj Institutions (PRI) system, which is the lowest tier of elected government in India (in this paper, referred to in local terms as the panchayat). Little Andaman has one Tehsildar and four panchayats.

A 'Language of Rights' Feeding Directed Consciousness
NGO activities were reported to have provided an expanded vocabulary through which to articulate their demands and frustrations: to "speak about the rights" (R2). Many respondents attributed this to the organisations' strongly rights-based approach, particularly when it was related to children and young people. Rights-based NGO activities on the island include a child court (Bal Sabha), child newspaper, providing expert guidance about rights and legal provisions (R113, R122), organised rallies, and NGOs supporting individual or collective visits to the panchayat to raise specific local issues (R8). The NGOs' stated aim for these activities was to increase awareness of the laws existing to protect local interests, and to generate "civic sense" (R122) from the bottom-up.
The data suggest that the language of rights communicated via these activities contributed to a transition from an unfocused feeling of frustration or injustice towards a more clarified, directed, and active articulation of critiques and claims upon the panchayat and Andaman administration. This transition, which is indicative of the emergence of directed consciousness (after Freire [51]), resonates with McFarlane's [61] demonstration of "how the mobilisation of the geographical imaginations of the poor help create spaces of political engagement" (p. 894)-i.e., a translation of a shift in how local people think about their rights, towards altered everyday political actions through which they seek to claim them. The role of NGOs in providing a language of rights, which in turn enabled local people to articulate their claims on the state in more empowered ways, was highlighted by all respondent groups: community members, government, local observers, and NGO staff. Two examples are the following: The NGOs taught about the rights, to speak about the rights; NGOs taught a lot, the kind of education given by the NGOs were to make a person self-dependent, to know what their right is. Even what is child rights, what is women's rights-all these . . . earlier no one knew what the right was. The specific complaints being vocalised varied from case to case (common themes including unemployment, unequal relief distribution, lack of government responsiveness). More significant than the nature of the claims, however, was the consistent message that these were made with greater force and intentionality-indicating a reduced willingness to settle for an unsatisfactory status quo.

Sense of Political Self
A second pathway of NGO impact was in shaping individuals' sense of themselves individually as a 'person with rights', and of having civic responsibility for their wider community. Two Little Andaman NGOs in particular have strong emphases in their messaging and programming on "helping people help themselves" (R99, R122). Their livelihood, education, child welfare, and rights-based activities all seek to generate a culture of personal responsibility and ownership, not only over livelihood generation or access to services, but over the political processes that govern development futures. One, for example, holds parents meetings that emphasise parental responsibility for children attending school and completing homework, and the implications of this not only for their child's future, but also for the development of the islands as a whole (R14, R122, PO3). The facilitators emphasise that parents are not "only" citizens of ANI but "also of India", encouraging them to compare the islands' level of development with that on the mainland (generally perceived as distant and aspirational), strengthening parents' sense of themselves as citizens of India (not just of ANI) with equal rights to mainlanders. Facilitators invite parents to reflect on their own responsibility in claiming this development, asking them: "you are the responsible person of all these issues, [so] why you are not taking any initiatives?" (R122).

Self-Confidence and Self-Efficacy
In parallel with this project of responsibilisation, NGOs also sought to build local people's self-confidence as political agents. This is important since, as noted by Zimmerman and Rappaport [62], merely having knowledge of institutional mechanisms and rights is an inadequate gauge of political self-efficacy. NGOs have undertaken a range of activities to promote individual self-reliance and independence that is both financial and emotional (R99, R104). They have emphasised seeking local solutions to local problems, and not simply providing short-term gains but capacity-building through skills-based trainings, which also supported participants' sense of self-worth and achievement (R99, R133, R122). Throughout there has been an emphasis on emotional strength-activities that "lift the people up" (R107) and "get rid of their baggages" (R99)-and by NGO staff maintaining an interpersonal manner that is relatable and sympathetic (R107). Being well-embedded in the community-achieved in one NGO's case by establishing a child centre in every village and mainly employing local people-was identified as contributing to enhanced self-efficacy by promoting trust between the community and the NGO. Participants suggested the emphasis on social-emotional empowerment has helped engender a lasting sense of resilience in the face of disaster: "[now] we can face things" (R32).
Interviews with residents and NGO staff also indicated positive reinforcement between participation in NGO awareness programmes (be it awareness of rights, the world outside ANI, health and sanitation, or any other kind) and self-confidence to approach and enter dialogue with state actors. Respondent R32 explained that their heightened awareness of citizen rights had given him himmath (courage), whilst interview R104 colourfully described a transition from the public being chootiya (meaning: idiots, literally 'dumbfucks') to now being more critically aware of their situation and feeling more able to speak up or participate in political rallies (locally referred to as strikes): Interpreter: So after the tsunami, how did you get this idea of a strike?... how did this idea come to you?
R2: This idea [to join a strike] must have come from saying . . . see the politicians keep saying, 'it's good to do this, good to do that...' they will keep talking to the public . . . [ . . . ] The public is after all chootiya [idiotic/dumbfucks] they don't know anything. Like after the tsunami they at least got some brains about what is good or not. It's good now . . . After the tsunami . . . we learnt how to talk to an officer who's come from outside, how to be respectful and all that.
[ . . . ] R3: . . . Before the tsunami, we would not have known how to talk to you. Like how to interact with people. (Interview with local residents, Little Andaman) Increased familiarity with 'outsiders', meaning people not from the immediate community, as mentioned in the extract above, was another recurrent theme in explanations of why and how political self-efficacy increased. The role of NGOs in this, firstly, was in accompanying local people to government offices or the police station, either youth groups as part of rights-education activities, as part of an organised rally, or when supporting individual families in a specific case. Interviews suggested that whilst enhanced self-efficacy and knowledge of institutional mechanisms were precursors for residents' own inclination to present themselves to state actors and mobilise, supported visits were also in themselves empowering-contributing to positive reinforcement.
Association with NGO staff also had an empowering impact via 'borrowed legitimacy'-meaning the felt support and gravitas that an organisation lends its employees by virtue of their association with it. This phenomenon was described by R130 as follows: They [the panchayat and administration] are scared of us because we have a uniform. We wear a green saree and put on a badge . . . [shows us the badge] When we wear this and go, they respect us a lot. (Community leader, Little Andaman)

Collective Consciousness
Another key trend post-tsunami was the increased number of public protests, which is significant for (at least ostensibly) demonstrating the public's strength of feeling and solidarity. NGOs played an active role in this by organising some protests themselves (e.g., against alcohol or child abuse), but may also have contributed to collective political consciousness through their propagation of self-help groups (SHGs). NGO staff reported that in the early days post-tsunami, it was often difficult to persuade community members to participate in collective relief efforts, and that they also initially had difficulty in recruiting members for SHGs. However, as the number of SHGs grew, NGO staff observed a change in local attitudes towards collective action. Whereas before "we were quite happy with our own family" (R13), now "people cooperate . . . participate in any programmes [that] we organise . . . [and] nowadays they themselves are eager to form groups and come to us to ask for it" (R124). This was emphasised as a specific shift post-tsunami: "this is only since the period of tsunami, the group dynamics have come, working in groups" (R13).
The greatest shift in collective action has been for women, for whom participation in SHGs and NGO activities stimulated increased mobility and social interaction outside the home, including for paid work. Female residents described feeling more engaged with issues affecting the community compared with before ("we minded our own business", R130). Whilst some highlighted the new need for women to leave the home for paid work, the role of NGOs in stimulating female mobility was also highlighted: when asked "why do they [the women] go out more?" one participant replied, "it is the NGOs who brought about this change . . . After continuously listening to all the meetings, the women have gained more confidence that we should also step out and work" (R127).

Institutional Support
Alongside the above activities-and equally important-were the NGOs' indirect impact on the institutional landscape of Little Andaman. During and since the relief phase, NGOs carved a strong niche as boundary organisations between the state and society in Little Andaman in a number of ways: representing communities on government committees; acting as a gatekeeper to state welfare benefits (e.g., giving trainings on government benefits schemes, delivering government-funded upskilling programmes); and acting as a lobby group and bringing collective action suits to the attention of the state. By acting as a spokesperson for local needs and demands, strengthening (and opening) channels of local representation, and highlighting gaps in state welfare provision, NGOs widened the space for state-society interaction and contributed to a migration of existing thresholds of social acceptance. NGOs sought to highlight government failings and call for the proper implementation of the law, largely through negotiation and utilising political pressures (the PRI) to leverage action from the administration. Some NGOs adopted a more adversarial approach, such as leveraging state accountability for rehabilitation failings through the courts.

Lessons for Transformations in Practice
The above findings thicken conceptualisation of transformation by identifying pathways of expanded civic agency at the individual and local scale. In particular, they shed light on the capacity for non-local (albeit well-embedded) actors to support local capacities to articulate and claim their rights, with implications for the way that local people hold powers to account and demand alternative ways-of-doing. Tsunami rehabilitation stimulated the arrival of rights-based NGOs where none existed previously, and the research indicates that their combination of rights-based awareness campaigns, messaging around individual responsibility and empowerment, stimulating collectivisation through group activities, and organising public strikes have each-individually and in combination-strengthened the culture of public politics in Little Andaman. NGO trainings and advocacy have raised local knowledge and expectations around the distribution and adequacy of state services, altering boundaries of social acceptance. Local people emerged more likely-and feeling more able-to raise their voices to the state, make demands, or challenge the status quo. The results indicate that the impacts of different NGO activities do not exist in isolation, but rather co-produce and reinforce each other, and there are lessons here for NGOs that are committed to genuine community empowerment through relief and rehabilitation.
However, despite these promising findings, a highly cautious approach to deliberate transformation remains necessary. This is for both pragmatic and ethical reasons: outlined below.

Pragmatic Challenges: Transformation Pathways Are Spontaneous, Fragile, Fragmented, and Political
The first reason why NGO-led deliberate transformation is pragmatically problematic is because pathways of change are complex and multifarious, troubling the idea that deliberate transformation can be instigated according to set, externally-defined parameters. In parallel to the pathways of local empowerment and political action documented in Little Andaman, the data also reveal a counter-narrative of apathy and the entrenchment of inequality, resulting in a patchy, incomplete, and contradictory geography of change. Whilst some people reacted to relief and rehabilitation in a way that sharpened their political engagement, for others, those same resettlement practices were deeply disempowering. In particular, being moved away from friends and family members fragmented supportive social networks, leaving families feeling isolated and unsupported, whilst the need to repair poorly constructed shelters negatively affected household incomes and heightened financial stress. Added to this, several resident and key informant respondents reported a decline in local livelihoods, a rise in alcoholism and joblessness, and an increased dependency on welfare benefits. Comparative analysis between villages and socio-regional groups indicated that these differentiated rehabilitation outcomes were socially and spatially specific. These concurrent trends, of conscientisation for some and relative disempowerment or dependency for others, demonstrate that transformative pathways are not uniform nor innate, but rather contingent and differentiated.
Furthermore, changes were deeply contoured by the Andamans' social-political pre-context, as well as shaped and constrained by prevailing governance norms and power relations. The robustness of ANI's centralised, paternalistic governance model acted as an overarching inhibitor to transformative change, resisting pressures (from citizens and NGOs) to increase institutional transparency and responsiveness. This stemmed in large part from the relative weakness of the political arm of government compared with the unelected administration [63]. In combination, the learnt political apathy of the Little Andaman population-an inherited product of the islands' history of colonisation (see Zehmisch [64,65])-meant that conscientisation did not occur completely nor for all. Dependency on state welfare remains widespread, and many immigrants to the islands exist in a state of precarity, unwilling to raise a dissonant voice to government officials. The islands' resistance to change is both a reason why expanded local agency in Little Andaman is highly significant, as well as why it remains constrained.
Related to this, the second challenge is that the outcomes documented were not the sole outcome of NGO interventions: rather, NGO impacts were multiplied by individuals' own personal reactions to the experience of relief, resettlement, and rehabilitation, which contributed to a subtle repositioning of local identities in relation to the state. For example, uneven and corrupt distribution of relief materials and permanent shelters served to magnify existing inequities between social groups. This reproduced inequality at a time of great personal loss and trauma, which sharpened locals' sensitivity to pre-existing governance norms and lowered thresholds of social acceptance. In addition, new issues such as lack of meaningful consultation during reconstruction decision-making, and the poor quality and unpopular location of permanent shelters, gave further cause for dissatisfaction. Since the state undertook the majority of reconstruction activities, heightened feelings of dissatisfaction became pointed specifically toward the state as the accountable party.
Meanwhile, the influx of external agencies and media coverage exposed Little Andaman communities that were previously accustomed to feeling isolated from 'the world outside' to foreigners, the Indian state, non-state agencies, and the comparative experience of other affected persons. These interactions had three effects: (i) drawing attention to Little Andaman residents' own sense of relative (felt) disadvantage, which raised expectations (realisation of what we want); (ii) increasing local familiarity to state processes and personalities (knowing how to get what we want); and (iii) exposing the (albeit temporarily inflated) financial capacity of the state (knowing that the state can give what we want). This combination served to widen the perceived gap between formal and substantive citizenship (after Holston and Appadurai [66]). By stimulating critical reflection on the role and failings of the state, and at the same time-again-sharpening local people's sense of themselves as citizens with rights, these factors contributed to the conscientisation of the individual (after Freire [51]).
These pathways of individual-scale change (in identities and expectations) served to multiply the effects of NGO activities on local peoples' imagined and felt political agency, and vice versa. Since they feedback and reinforce one another, the combined effects of all of the conscientising processes-NGO-supported or not-should be seen as greater than the sum of their parts, rather than acting separately in time or space. This mutual constitution means that humanitarian organisations should not view themselves as outsiders 'doing' relief or development 'to' the local community in a perceived vacuum of local agency and the concurrent actions of other stakeholders. NGO impacts are not simply cause-effect; rather, they are contoured by concurrent shifts in local capacities, social networks, or preferences occurring in response to multiple drivers (not just the NGO), a myriad of social actors (including the state), and across a range of scales. For example, an NGO becoming embedded in the safeguarding of child safety, or calling for the just application of child protection legislation, is equally the product of national-level legislation as it is of the NGO.
The combination of these factors means that the final outcome of NGO activities are likely to be unanticipated and non-linear, and it is impossible to discern the impacts of one activity or event from that of another. This makes the idea of a planned, pre-meditated deliberate transformation agenda, on a known, budgeted timescale, highly problematic.
A third pragmatic challenge to a set deliberate transformation agenda is that pathways are inherently political, and require careful negotiation of local systems in order to be sustainable. A consistent message from NGO respondents was that their operation in ANI was contingent on careful manoeuvring amid the islands' idiosyncratically state-heavy, paternalistic, and non-plural governance landscape. Framed by some as a necessity and by others as a choice, NGO staff highlighted that working collaboratively with the state was generally more effective in ANI than raising overt challenges. Those remaining operational in Little Andaman 10 years post-tsunami shared a number of characteristics: evolving their project focus to stay relevant (and funded) post-relief phase; employing local staff and networks as a gateway to local embeddedness and generate local buy-in; adopting 'boomerang' tactics to multiply pressures on the administration via the panchayat; operating in an ostensibly politically neutral sector (e.g., education, child welfare); offering financial support in under-funded sectors; and working collaboratively on state-funded programmes e.g., formal partnerships for the roll-out or implementation of government schemes. It is here, in the ability of these NGOs to negotiate a position of some influence-e.g., being permitted on government committees-despite institutional barriers, where the most significant learning points for an NGO operating in such contexts can be gleaned. It is also highly significant (especially for INGO practice) that the most impactful NGOs hired locals as their field staff, which greatly aided local participation in their programmes and their capacity to respond reflexively to local needs.
This evidence suggests that NGOs must act politically in order to negotiate a position of longevity and influence. It indicates that doing so could provide NGOs with greater scope to challenge inequitable and normalised power relations from the 'inside'. Indeed, it is only by engaging in political processes and addressing these root causes that true social-political transformations can occur [2].
However, there are also dangers in arguing that NGOs should be more overt in their consideration of-and interference in-local politics. There is the risk that NGOs would not be allowed to work at all if they declare their political intentions at the outset (indeed, remaining politically neutral has always been a condition for the Red Cross's activities, for example). Furthermore, there is the risk of their political objectives become diluted or subverted in their implementation. Forces of ideological co-option are more likely where change pathways are messy and uncertain, and have already been observed with the language of sustainability (which is only meaningful if global capitalist systems of production and consumption are fundamentally reformed) and resilience (only meaningful if the local scale has economic and institutional flexibility and freedom, requiring a supported decentralisation of power and resources). This is arguably already happening with transformation, the language of which makes a surprisingly radical appearance in the IPCC 5th Assessment report [67], but is already being diluted in its application in NGO programming and global funding calls. Stirling [22] argues that a key challenge for a social transformation agenda is protecting it from co-option by powerful actors to justify the deepening of centralised, non-inclusive forms of governance in the name of sustainability.

Ethical Challenges: Deliberate Transformation as a Normative Agenda
Moreover, an even more powerful critique is that the involvement of NGOs in deliberate transformation raises important questions around the ethics of transformation as a normative project. Part of the problem is a lack of agreement about precisely what transformation means. However, if local self-determination, inclusion, equality, recognition, justice, valuing local knowledge, culture, and worldviews are important-and these are the cornerstones of many definitions of just adaptation trajectories, including those of Heyward [41] and Paavola and Adger [38]-then an externally-defined idea of transformation cannot deliver this. Seeking to implement deliberate transformation through a top-down transfer model would bypass the opportunity for a genuine renegotiation of unjust power relations. To avoid reproducing neo-colonial power relations, transformation trajectories must be shaped by the worldviews and needs of communities themselves, rather than the priorities of outside actors.
In terms of how this argumentation can feed into practice, there are parallels to be drawn from the critical development literature on the contradictions of INGO action. Cornwall and Coelho [49] and Mohanty and Tandon [68], among others, have noted that although participation as a practice may provide the opportunity for power relations to be challenged and transformed, the effectiveness of this depends on how and why participation occurs. There is a wide and critical literature on the micro-politics of community-based NGO programming-whether in the name of development, CCA, DRR, or conservation-and its often-limited ability to meaningfully alter power relations in favour of local people [49,52,[54][55][56][69][70][71][72][73][74]. Critics have focused particularly on the contradiction between non-local agencies' claims to empowering local people to become more self-determining social-political agents, and the pre-determination of the goals and scope of participatory projects by the implementing agencies. Such activity risks falling into the trap of neo-colonialism when undertaken by international NGOs in low-income settings [75][76][77].
Returning to the case study, the most powerful findings in Little Andaman were: (i) the way that NGOs helped create a space for local people to critically reflect on their political subjectivity and open spaces of state-society interaction, and (ii) that the success of those activities hinged on the organisations being locally embedded well beyond the initial relief period. Moments of transformation occurred equally in the form of spontaneous changes in the self as they did from deliberated, planned interventions. It is a significant finding that many changes were incidental yet supported by sustained interventions by agencies that were well embedded in the community, with real-world implications for how transformation is (or is not) practiced. It indicates a need to move away from the short-termism of most relief and rehabilitation work, and rather for humanitarian interventions to be cautious, considered, and part of a long-term engagement. The experience of NGOs in Little Andaman demonstrates how strategically positioned agencies might go beyond the typical focus on 'doing' relief 'to' target communities, towards supporting communities in claiming development futures of their own choosing. This shift in emphasis is essential in moving towards a post-development humanitarian agenda, since it is well-documented that NGOs that are merely seeking to 'do good' in a political vacuum do little to address the underlying root causes of vulnerability: poverty, social marginalisation, inequality [25].

The Case for Progressive Rehabilitation
On the basis of these critiques, deliberate transformation as NGO practice (particularly INGO practice) in the post-disaster space needs to be approached with caution. Pelling and Dill [9] and Pelling [2] warn that powerful interests work in the post-disaster space to resist changes to the status quo: pro-poor, pro-justice NGOs must ensure that they are not part of this discourse of inequitable social reproduction. However, as noted earlier, this is not to say that transformations are not needed: far from it! It is merely an argument that 'real' social transformation cannot occur through business-as-usual top-down humanitarianism, with transformation merely inserted into existing ways of working. The need for NGO practice to move beyond a top-down knowledge transfer model is not a new argument; what is new is its urgency, given the particularly normative strength of the transformation agenda and the speed with which it is propagating within policy discourse. Transformation is powerful language and should be reserved to describe deep radical change in the way that development is conceptualised and practiced.
An alternative, this paper suggests, is to plan for progressive rehabilitation-requiring a transformation of the NGO itself from 'doing to' to 'doing with' local people. Progressive rehabilitation means focusing on supporting households and communities to recover in self-determined ways; framing local people not as beneficiaries but rather as citizens; and thinking not of disaster response as a one-time 'fix' (one frozen in time, isolated from local politics), but rather part and parcel of the ongoing governance-development trajectory. In this approach, the emphasis is on creating space for conscientisation, from which structural transformation may or may not follow, without an externally pre-defined vision of the outcome. With progressive rehabilitation as the primary goal, transformation (at any scale) would be incidental and locally-embedded, if at all, rather than according to NGO priorities.
At an international scale, there are nascent indications of this type of thinking; one example is the Christian Aid-led consortium 'Linking Preparedness, Resilience, and Response' (LPRR), which ostensibly seeks to smooth the humanitarianism-development gap through post-disaster interventions that support long-term resilience and are grounded in local priorities and knowledge [78]. Equally exciting is the application of forum theatre approaches in resilience-building initiatives, such as the project 'You, Me, and our Resilience' led by Katrina Brown in collaboration with the NGO S.A.F.E. Kenya [79]. Additional recommendations for progressive rehabilitation practice identified in this research are: protecting social networks and supporting self-efficacy by offering households choice in resettlement location and housing design; maintaining high standards of transparency and accountability in relief and rehabilitation resource distribution, promoting trust between stakeholders; and innovative recovery activities that respond to local people's expressed needs (e.g., livelihood regeneration, information technology (IT) skills training).

Conclusions: What Does Transformation Look Like, and What Next?
The experience of post-tsunami rehabilitation in Little Andaman indicates that pathways of transformative change at the local scale are a combined product of spontaneous, personal reactions to lived experience of rehabilitation, and planned, deliberate, interventions by state and non-state actors. Hence, it is concluded that transformations are not only spatially and socially specific, but also complex, multi-directional, unpredictable, and fundamentally political. These findings build on Pelling and Dill's [9] observation that the potential for post-disaster opportunity spaces to be captured politically is highly contingent and context-specific. They argue it is not one single factor that may induce or constrain transformative change; rather, change (if occurring at all) will be subject to complex, patchy, and unanticipated interactions between a myriad of factors. Amid these 'forking paths' [80], trade-offs are almost inevitable, and transformations will inevitably have winners and losers. Navigating these complexities fairly requires reflexive and sensitive NGO planning that prioritises local voices (however divergent and contradictory those may be) front and centre.
Moving forward, three lessons emerge. First, the evidence indicates that it is possible for NGOs to support emergent local agency through capacity-building for enhanced political self-efficacy at the local scale, and there is scope here for transformation. Conscientisation of the self and an accessible, plural state infrastructure are identified as key conditions for this. Secondly, the data illustrate the inseparability of politics from disaster relief and rehabilitation, not only for determining the geographies of aid distribution and local recovery, but also for shaping the long-term outcomes of rehabilitation on the governance landscape. The paper argues that consideration, navigation, and respect of local politics is essential for pro-poor and pro-local NGO action. Nevertheless, planning for deliberate transformation as a normative project with externally-defined political objectives is neither pragmatic nor ethical for NGOs (nor any outsider stakeholder). Instead (and thirdly), the paper argues in favour of a progressive rehabilitation agenda, wherein post-disaster relief and recovery is reframed as a practice space to embed longer-term social development objectives around enhanced human well-being, livelihood resilience, and effective citizen representation. To meaningfully address root causes of vulnerability [25], humanitarian organisations must target their programming in ways that support communities and institutions to challenge inequitable development trajectories from the inside out, and not seek to impose change from the outside in.
The aim of this paper is not to refute the need for transformation, nor to imply that there is no role for non-governmental agencies in transformation. Rather, it is intended as a call for critical engagement with the realities and ethics of deliberate transformation before it is sought as an explicit programme goal. This will help protect communities from the reproduction of western-centric power relations, and also protect the language of transformation from dilution or cooption. In practice, a progressive rehabilitation agenda should include joined-up, locally-embedded approaches to post-disaster rehabilitation as part of the ongoing development of that place (this much is not new), alongside renewed commitment in the DRR and humanitarian sector to supporting genuine local empowerment and local leadership. Even more key, and what is highlighted here, is the need to recognise that deliberate transformation pathways are not simple nor linear but complex, political, and normative. Improved understanding of just pathways to transformation through local action would benefit from closer cross-fertilisation between climate change adaptation (CCA), disaster risk reduction (DRR), climate justice philosophy, and critical development geography-the latter of which has a long and rich tradition of scholarship on how power and privilege operate through NGO intervention, as well as the politics and pathways of local empowerment. Attention to these overlaps is called for to unpack the normative assumptions embedded in transformative agendas, and to seek more truly emancipatory development futures.