Typhoon Yolanda and Post-disaster Resilience: problems and challenges

Typhoon Yolanda hit the Visayas region of the Philippines on 8 November 2013. With wind speeds reaching 315 km/h, the storm remains one of the strongest and most destructive typhoons ever to make landfall (Santos, 2013). This article focuses on how the concept of ‘resilience’ was articulated and experienced post-Typhoon Yolanda (international name Haiyan) with reference to the rebuilding and rehabilitation efforts in Tacloban City. Tacloban is the administrative center of the Eastern Visayas (Region VIII) in the Philippines and one of the most heavily urbanized areas to be devastated by Yolanda. This city was chosen as the focus for this article as it became the ‘exemplary center’ of the relief and rehabilitation effort. It stands as a ‘microcosm’ (Geertz, 1980: 13) of the institutional and logistical practices common in urban lower and middle-income post-disaster scenarios. Arguably, these practices perpetuate social inequality, operate against a backdrop of ‘status pride’ (Geertz, 1980: 13) and re-materialize power in ways that consolidate the status quo and profit seeking norms.


Introduction
Typhoon Yolanda hit the Visayas region of the Philippines on 8 November 2013.
With wind speeds reaching 315 km/h, the storm remains one of the strongest and most destructive typhoons ever to make landfall (Santos, 2013). This article focuses on how the concept of 'resilience' was articulated and experienced post-Typhoon Yolanda (international name Haiyan) with reference to the rebuilding and rehabilitation efforts in Tacloban City. Tacloban is the administrative center of the Eastern Visayas (Region VIII) in the Philippines and one of the most heavily urbanized areas to be devastated by Yolanda. This city was chosen as the focus for this article as it became the 'exemplary center' of the relief and rehabilitation effort. It stands as a 'microcosm' (Geertz, 1980: 13) of the institutional and logistical practices common in urban lower and middle-income post-disaster scenarios. Arguably, these practices perpetuate social inequality, operate against a backdrop of 'status pride' (Geertz, 1980: 13) and re-materialize power in ways that consolidate the status quo and profit seeking norms.
Disaster resilience is often related to the management of change or disturbance with the aim of sustainable material wellbeing, for example DFID refers to resilience as 'maintaining or transforming living standards in the face of shocks or stresses ' (2011: 6). The analysis to follow suggests resilience can be understood in terms of the effective management of a set of practices, rather than a condition. Disasters inform resilience as they are the moment when 'the public and its problems are brought into clear view' (Chandler, 2014: 147) and the 'natural' order of things is tested. This case shows that resilience-building strategies must be crafted with reference to specific local socio-economic contexts if they are to be sustainable, effective and properly understood. The experience in Tacloban matters for understanding post-disaster resilience because it is illustrative of the increasingly frequent and intense climate change related threats to urban coastal areas in South-East Asia.
Resilient was a term frequently used by the media, survivors, government officials and various other stakeholders to describe the response to the disaster and those affected by it. I argue that the term is overused, complex and contested, not least because it is socially situated and culturally determined. Resilience building involves examining 'the "reality" of processes and relations rather than […] a top-down imposition seeking to direct, manage or assert control over things' (Chandler, 2014: 13). This article examines the ways that post-disaster resilience is presented as an admirable quality and positive process whilst at the same time shoring up pre-existing power structures and inequalities. I argue that framings of resilience that focus on strength and endurance mask the limitations of post-disaster recovery and marginalise intangible factors such as trust, equity and community cohesion. The rhetoric of resilience matters for survivors as it (re)constructs recovery in ways that limit meaningful change and shuts down dissent. This article investigates the following questions: Firstly, how was resilience articulated and experienced post-typhoon Yolanda? And secondly, why do we need a better understanding of resilience for disaster relief and rehabilitation?
Resilience may be driven from the bottom up in ways that are hard to measure.
Survivors may take pride in strength through adversity and state that they are resilient when they are not. Or they may simply equate resilience with survival, even under severely diminished conditions of existence. Government and non-governmental agencies and personalities may designate survivors as resilient in order to justify their recovery strategies, without due consideration of what it actually means to be resilient. This interrogation of resilience is important as it offers a way to help unpack the social and material protection of the status quo from above and explains why resilience from below must include social variables such as trust and equity if it is to be sustainable. Nevertheless bottom-up and top-down resilience-building strategies are seldom linear and often context dependent. In order to make sense of these complexities and contradictions, this article considers how resilience was articulated and understood in governmental and non-governmental policy terms and at the community level.
The first section of this paper explains Tacloban's political and geographical vulnerability in relation to Typhoon Yolanda. The second section outlines some of the main themes, issues and complications identified in top-down and bottom-up analyses of disaster resilience. The third section outlines the primary data gathering strategies for this article with an explanation of the area under investigation. The way in which 'resilience' was understood by various individuals and organisations in relation to Typhoon Yolanda is explained in the fourth section. The fifth section outlines how Typhoon Yolanda survivors themselves articulated resilience during fieldwork investigations and draws on survey findings. In this section I focus on community, as a form of social capital, and housing and infrastructure, as forms of urban material capital. The sixth section is a discussion that outlines the problems and challenges in resilience-building post-Typhoon Yolanda. Resilience is not just about satisfying material need, it is also about successful social adaptation. I aim to show that the rhetoric of resilience did not match up to the reality after Typhoon Yolanda. In conclusion, I argue that more critical and nuanced investigations are needed on what it really means to be resilient in the urban context post-disaster.

Tacloban City: Geography and Politics
Official figures indicate 6,293 individuals reported dead, 1,061 missing and 28,689 injured as a result of Typhoon Yolanda, while 591 municipalities were affected and the total damage was estimated at US$904,680,000 (National Disaster Risk Reduction and Management Council, 2014). An estimated minimum of 31,478 (Lagmay, 2014) informal settlers were living in areas vulnerable to a storm surge at the time of Typhoon Yolanda. Many of these settlers lived on the coastline, or even over the sea, in makeshift wooden structures. During Typhoon Yolanda the storm surge reached five meters high in Tacloban (Nakamura, 2015: 376). The effect on these settlements was catastrophic. The Philippines experiences around 20 typhoons per year and is used to coping with flooding, however, Typhoon Yolanda was exceptional in terms of the damage wrought. At the time of writing rehabilitation work is still ongoing.
Tacloban is a rapidly growing 'highly urbanised city' 1 . In 2010 its population was 221,174 with numbers swelling to 246,115 in 2015 (Philippines Statistics Authority, 5 2016: 6). Overall, the economy of the Eastern Visayas is booming with 12.4 percent growth (Recuerdo, 2017) being recorded in 2016, the highest rate of any region in the Philippines. Due to its strategic position, Tacloban is at the center of that boom.
However, in common with many other Asian cities (UNESCAP, 2013), Tacloban's growth has resulted in intense competition for urban space.
Tacloban is the bailiwick of the Romualdez family. The current Mayor is Christina Gonzales Romualdez, wife of former Mayor Alfred Romualdez. Former First Lady Imelda Romualdez Marcos is Alfred's aunt and Tacloban is her hometown.
Historically the Marcos connection has meant that Tacloban benefitted from the largesse of the Marcoses allowing the Romualdez family to consolidate their economic and political influence (Roces, 1998: 300). As such, there is a special interest in (re)constructing Tacloban as an exemplary centre of the rehabilitation effort as the fortunes of Tacloban are indelibly intertwined with the political credibility and status pride of the Marcos/Romualdez family.
However, the Romualdez/Marcos connection meant that the relationship between Tacloban and the national government was extremely poor under the administration of former President Benigno 'Noynoy ' Aquino III (2010. The Marcos family is widely assumed to have been behind the 1983 assassination of Aquino's father, Benigno 'Ninoy' Aquino Jr., a former Senator and leading opponent of Marcos. Corazon 'Cory' Aquino, wife of Ninoy, assumed the presidency in 1986 when Marcos was ousted by the peaceful EDSA revolution. Noynoy Aquino, who rode a groundswell of support triggered by his mother's death in August 2009, was elected to the presidency in 2010. When President Aquino left office in 2016 it was found that 20.7 billion PHP of a 25.6 billion PHP national housing assistant budget and 18.4 billion PHP of a 18.9 billion PHP National Disaster Risk Reduction and Management Fund for Typhoon Yolanda survivors lay undisbursed by the Aquino government (Ager, 2016).
With reference to the overarching focus of this special issue, Clifford Geertz argued that in pre-colonial Bali 'court ceremonialism was the driving force of court politics' (1980: 13) and served to legitimate social inequality and status pride in relation to the exemplary center. It could be argued that comparable rituals legitimize the notion of the 'exemplary family' in the Philippines. Dynastic governance is consolidated through various means including patronage, monuments, rituals such as fiestas and religious ceremonies, the presence of the family at commemorations, celebrations and the visible sponsorship of civic events. Thus legitimizing dynastic rule as the natural order of things. It is also common for politicians to describe themselves as the mother or father of their constituents, therefore invoking notions of benevolent paternalism.
Legitimate hierarchy and good leadership matter for resilience, and relief efforts that ignore or undermine legitimately established structures of command and control risk further disruption to already shattered communities. However 'legitimacy' is not just mandated for by elections; it is also a cultural construction that is reinforced by the issues highlighted above. Disaster rehabilitation practitioners that fail to appreciate the socio-economic and cultural nuances of particular localities, as they relate to resilience, may waste resources and even undermine the very communities that they are trying to help. The complexities of this issue are discussed below.

What is disaster resilience?
The term natural disaster should be treated with caution. The scale and impact of natural environmental disasters is mediated by human activity to the extent that there may be little that is 'natural' about their effects. People living in typhoon affected, low lying, densely populated areas in makeshift housing are extremely exposed, and therefore vulnerable to, the vagaries of the weather. This was the case in Tacloban and is a trend that can be more widely identified across the Philippines and elsewhere.
Reports suggest the frequency and intensity of typhoons is increasing (Mei and Xie, 2016). Consequently environmental hazards have an increased likelihood of turning into disasters with the poor being least able to mitigate or adapt in the face of such hazards.
The word resilience is derived from the Latin word resilio meaning to leap or spring back. It is a hybrid concept that has been debated across disciplines (Cote and Nightingale, 2012;Folke, 2006;Holling, 1973) in physical/material and social terms, and in relation to governance (Chandler, 2014); community (Aldrich and Meyer, 2015;Cutter et al., 2008), faith (Wilkinson, 2015) and culture (Bankoff, 2003;Kruger et al. 2015). Resilience has relevance for complex post-disaster interventions and for how people and communities recover or 'bounce back'. It is about physical and social evolution that reduces vulnerability to future disasters. Human resilience is not distinct from ecological resilience and any 'delineation between social and ecological systems is artificial and arbitrary' (Folke, 2006: 262). Therefore the institutions and various stakeholders tasked with devising mechanisms for resilience after environmental disasters must consider how social and environmental adaptations impact upon each other. argues that Filipinos have become 'primarily resilient with great capacities to organize, resist, learn, change and adapt' (Curato and Ong, 2015a, 216). Bankoff and Aldrich both focus on how collective action comes about and the difference that it can make. Under disaster conditions social capital acts as a kind of informal insurance involving trust, norms and networks. This is closely linked to social networks and notions of community and mutual labour, known in the Philippines as bayanihan (Gaillard, 2015: 78).

Resilience From Below
However, post Typhoon Yolanda the social capital-based notions of resilience outlined above were revealed to be overly optimistic. Pauline Eadie and Yvonne Su offer a more negative reading of social capital arguing that people only '"invest" in social capital/relationships because they have the expectation of some sort of return' (2018: 341). After Typhoon Yolanda that 'return' became uncertain due to perceived inequalities in the allocation of relief goods . Meanwhile those whose homes were destroyed suffered a series of displacements from one temporary home to another whilst also vying with other similarly placed families for limited resources thus creating a 'culture of mistrust' (Curato, 2018). Thus the trust, norms and networks that could have nurtured bottom-up resilience were disrupted.
Nevertheless social capital is one of the few sources of capital available to the poor and normatively thought of as a good thing. The potential for effective civic engagement should remain a central issue for resilience building.

Resilience From Above
Trust, norms and networks also work for the powerful. Consequently, resilience strategies may 'maintain a system in an undesirable state, making recovery or transformation difficult' (Weichselgartner and Kelman, 2014: 5). This argument is informed by Colin Walch who argues that technical, apolitical approaches to resilience are 'often favoured by governments' (2018: 131) as rights-based approaches are too politically loaded. Whilst on the other hand, NGOs may 'consider disasters an opportunity to raise people's awareness and mobilise them for social change' (Bankoff and Hilhorst, 2009: 698). Both sides of this debate highlight how resilience can mean different things to different stakeholders and how important social context is to resilience building. This debate matters because there are dangers in resilience strategies that rely on 'command and control styles that risk preserving the status quo, and which might entrench exclusion, and take attention away from the inequality, oppression and entitlement loss that results in cases of proneness to insecurity and disaster (Manyena, 2006: 438). Resilience strategies that resurrect predisaster inequalities and vulnerabilities, or even create new ones, are not resilient at all. Nevertheless an unequal political culture may be so embedded that the status quo is near impossible to shift.

Resilience: Tensions and Contradictions
David Alexander notes that historically the use of the term resilience has been 'potentially contradictory, such as restoring equilibrium and getting away from it by moving to a new system state ' (2013: 2710). Therefore resilience is about both established and evolving social and ecological systems. This contradiction is evident in the definition of resilience: 'the ability of a system and its component parts to anticipate, absorb and accommodate, or recover from the effects of a hazardous event in a timely and efficient manner, including through ensuring the preservation, restoration, or improvement of its basic structures and functions' (IPCC, 2012: 5).
The reference above to both preservation and improvement is indicative of both continuity and change.
Local voices may also have conflicting ideas of how resilience capacity-building should be built. Some may have a vested interest in maintaining the status quo, some may favour different adaptation strategies than others, and some may champion adaptation strategies that may compromise others with effects that are difficult to disentangle. Strategies that are designed to be equitable and inclusive, but are poorly thought out, may result in the already vulnerable being further exposed to harm or in one vulnerability being replaced for another.
Curato and Ong argue that local cultural norms dictate how communities construct the difference between 'acceptable sacrifice and traumatic suffering' (2015b: 16).
Meaning that comparable disasters could generate quite different articulations of resilience from similarly affected communities. It is also useful to consider the argument of Bènè et al. who argue that adaptation, rather than being a purely positive notion, involves 'unacceptable trade-offs' (2014: 601) between resilience and wellbeing. Household coping strategies that involve missing meals, neglecting health care (including mental health) or education, or living in inadequate shelter can hardly be considered resilient. Bènè et. al. also argue that disaster rehabilitation strategies that are premised on resilience-building may simply entrench pre-existing vulnerabilities and inequalities. Meanwhile Rigg et al. discuss post-disaster resilience-building with reference to how various aid agencies can disturb social hierarchies and patterns of accumulation meaning that 'resilience is not a quality that can always be computed from pre-existing patterns of wealth and poverty' (Rigg et al, 2008, 150), in other words, disasters can, depending on the social and physical context, disturb the status quo. Overall, real resilience is likely to involve some element of redistribution and social justice and this is likely to suit some more than others. As such, the crafting of resilience is also about power and control.

Data Gathering
Without 'community participation, disaster relief often inadvertently rebuilds […] structures of vulnerability' (Bhatt and Reynolds, 2012: 74) consequently, this article is based on the premise that local communities should be consulted on how to best build resilient and sustainable disaster relief strategies. The primary data for this article was gathered in seven coastal barangays 2 in Tacloban. All the barangays were inundated by the storm surge and were populated by informal settlers. They all have areas now designated as 'no-dwell' by the Philippine government (Gardner, 2016).
This policy means that no resident is allowed to live within 40 meters of the shoreline.
The multi-dimensional nature of resilience means that it is difficult to set a benchmark against which one is deemed to be resilient or not. Recovery from disasters is often measured in terms of economic growth, the reduction of poverty, or the volume of goods and services supplied. However, these economic variables do little to reflect socio-cultural aspects of resilience. Consequently we asked a range of survey questions relating to community cohesion and material welfare. We also asked whether survivors self-assessed themselves as resilient. The surveys ran in 2015 and 2016. Survey data was complemented by 16 focus group discussions (FGDs), involving specific respondents such as women, the elderly, young people and people with disabilities and their carers. Survey work and FGDs were carried out and transcribed by local Waray speakers in consultation with English speaking partners working on this project. Special attention was paid to keywords such as resilience, the nuances of which could have been lost in translation. For resilience we used mabaskog, which translates to strong or robust. Resilience is not an indigenous term and there was no word that also clearly encapsulated the notion of recovery. We also ran small pilot studies to ensure that the meaning of the questions was clear. Our questions were designed to provide us with information on the technical and social issues that helped or hindered resilience-building.
We also interviewed a range of governmental and non-governmental stakeholders in 'the task immediately before us lies in ensuring that the communities that rise again do so stronger, better, and more resilient than before' (Aquino, 2013)  However these settlements have a number of problems including a lack of utilities and inadequate livelihood provision (Ong, et. al. 2015). Many of these homes were flooded during Typhoon Urduja in December 2017 due to poor drainage and inadequate construction.

What Matters for Resilience?
Some of the responses above outline practical understandings of resilience such as effective evacuation and material wellbeing, including housing. Community cohesion, or bayanihan, also features as an aspiration and a practical requirement.
This section details the impact of perceived inequity in the distribution of relief goods and services, and difficulties in the resettlement process as key impediments to community resilience.

Community Resilience
The chart below details answers to the question 'To what extent did your community help itself?' This question was asked in relation to the period immediately after Typhoon Yolanda, two years later (2015) and three years later (2016). It refers to issues such as communal rebuilding, the self-organization of tasks such as clearing up and burying the dead and the sharing of resources.

Table 1
The responses to this question are disappointing given that self-help at the community level is a key factor of the 'bonding' social capital outlined by Aldrich and the notion of bayanihan highlighted by Gaillard and former President Obama.
However, the results do correlate with reports from some barangay captains that not all residents who were given 'cash for work' by organizations including Tzu Chi Foundation (Tzu Chi, 2013) did the work required of them, they were 'free riders' that just turned up at the end of the day to collect their 500 PHP allowance. This result correlates with the findings of Yvonne Su and Ladylyn Lim Mangada (2016) who argue that indigenous notions of bayanihan are outmoded and undermined by free riders and arguments over the allocation of relief goods.
Our 2015 survey results indicated that the primary source of support for survivors was 'themselves or their family' with a score of 42.6 percent. However, it was common for extended families to reside within the same barangay, meaning that the distinction between community and family is somewhat blurred. Nevertheless equity was a key issue in relation to community resilience. Focus group discussion evidence 20 indicated clearly that many people were unhappy with the way that goods and services were allocated after Typhoon Yolanda and that this had an impact on community relations within the barangay. In some cases a lack of trust was evident with respondents feeling that some barangay captains directed resources towards family members and trusted allies (Field, 2016;Ong, 2016;Wilkinson, 2015) an issue that highlights the negative side of social capital.

Housing and Infrastructure
According to the Philippine National Risk Reduction and Management Council It was also decreed that families who lost homes in the 40-meter zone would not be eligible for Emergency Shelter Assistance (ESA) funds in order to discourage them from rebuilding in the zone. Instead they would be rehoused further away from the sea.
However many families did rebuild in the no dwell zone, especially in Tacloban.
This was because the relocation of those that lived in the zone was an extremely protracted process, involving multiple agencies, that is still ongoing in some areas.
With no ESA or affordable alternative building plots available many families had no choice but to rebuild makeshift homes in their original locations. In order to rebuild, people were often forced to rely on the largesse of extended family or loans. The rebuilding process often pushed them deeper into poverty. In 2016, 70 percent of our survey respondents reported that the shelter materials they were given were inadequate and only 43 percent reported an offer of moving to a relocation site. In a June 2017 women-only FGD it was reported that house sharers were unable to access duplicate reconstruction materials and were envious of others, possibly resulting in bad feeling in the community. It was common for extended families to stay in one dwelling. In Tacloban the local government effectively abandoned trying to stop rebuilding in the no dwell zone as they were unable to offer any practical alternative. dispossession in Jakarta) the local government in Tacloban has been relatively inattentive in clearing the no dwell zone. This is potentially because a more forceful approach could trigger a backlash, as many people have still not been offered a viable alternative.

Discussion
It seems that resilience is something to aspire to and it is normatively understood as being a 'good thing' or an admirable characteristic. However, there is little consensus on what resilience is. It is a complex process rather than a quantifiable end point.
Typhoon Yolanda survivors surveyed for this project overwhelmingly confirmed that they considered themselves to be resilient. People generally identified 'strength' as being a positive attribute in relation to resilience. However, the ability to absorb shocks through negative coping strategies does little to reduce inequality or promote resilience. Negative strategies do not improve or adapt the lives of individuals and communities. Reducing food intake, living in substandard housing with inadequate sanitation, increased debt, the increase of excessive working hours in 'own account' (informal) or family work or child labour are not positive adaption strategies. It is particularly worrying that negative coping strategies can be casually equated to positive connotations of resilience by both survivors and aid agencies.
It was evident that the management of some of the rehabilitation strategies used post • Participatory facilitators override existing legitimate decision-making processes.
• Group dynamics lead to participatory decisions that reinforce the 25 interests of the already powerful.
• Participatory methods drive out alternative strategies that may have advantages participation cannot provide (2001: 7-8).
The points raised above are relevant for participatory disaster relief and rehabilitation.
If outside agencies forcibly override existing systems of command and control then this may undermine local hierarchies. Whilst these hierarchies might not be perfect they may be legitimate and undermining them could lead to destabilizing power struggles. Alternatively, 'participatory' approaches may just be hi-jacked by the already powerful leading to resentment in communities or effective leadership may be sacrificed at the altar of participation. The danger of everyone having their say is that nothing gets done. Leadership is important for resilience-building, however, this leadership needs to be legitimate, accountable, self-reflexive and capable of adapting for the greater good.

Conclusion
The term resilience has been used somewhat carelessly in media and policy circles.
Evidence shows Typhoon Yolanda survivors also referred to themselves as resilient, even when the conditions of their existence were miserable. Since Typhoon Yolanda resilience has been equated, locally and nationally, with strength rather than the effective management of the complicated process of recovery. This framing has been top down as well as bottom up and has served to gloss over the complex challenges of resilience-building.
Tacloban, as the exemplary center of the Typhoon Yolanda relief efforts, was indelibly linked to the political fortunes and status pride of the Romualdez/Marcos family. The recovery of the city was concomitant with that of the family and this link was manifest in rituals, celebrations and commemorations, local government run recovery schemes and endorsements. These rituals lauded the resilience and survival of the people and the city. This narrative served to shore up the political status quo and mask the recreation of material inequalities. The terminology of resilience was potentially used as a 'disclaimer' for the failings of the relief effort. That is: survivors are used to these types of disturbances therefore they will have an innate ability to cope. This effectively reduced resilience to a state of mind. However resilience is more than an attitude or the ability to survive at a basic level of existence. It is about adapting socioeconomic institutions and the natural environment in order to mitigate future disturbances.
Resilience is not achieved through survivors adopting negative coping strategies such as a reduction in food or education. It involves access to material and social resources that bring about an improved ability to absorb, adapt or transform in the face of future disturbances of stressors. However, in societies where the already powerful are well served by existing power configurations transformation, or even certain adaptations, may be resisted. This is not necessarily done overtly but rather by normalizing strategies that support the status quo. In most of the communities that we examined people had adapted, but conditions of life had not necessarily improved. Material adaptations, such as the development of new residential areas, were often poorly planned, crippled by red tape and under-resourced. Nevertheless, survivors tended to designate themselves as resilient. This was potentially because the terminology of 'resilience' does not exist in the local language and interpretations have tended to focus on personal or collective strength rather than a technical process. For disaster relief practitioners and researchers these nuances are important.
Disaster risk reduction has improved in terms of psychological attitudes and material resources in Typhoon Yolanda affected areas. Early warning systems have been enhanced, typhoon resistant evacuation centers have been physically improved and people are psychologically attuned to the need to evacuate when there are typhoon warnings. Nevertheless, disaster relief agencies need to think more carefully about what it means to be resilient and how this can be measured in social and material terms. Indicators of resilience such as safe housing or confidence in community and trust in 'leaders' intersect and impact upon one another. Therefore the relationship between indicators as well as their volume is important.
Resilience has become a policy narrative for disaster relief agencies at international, national and local levels, however, technical and social understandings of what it means to be resilient are complex and sometimes contradictory. The findings in this article reinforce the scholarship that identifies the tension between restoring the status quo and adaptive resilience strategies with particular reference to the dispossession and relocation that many typhoon survivors face. They also highlight the difficulties of equitable and inclusive strategies, that may be desirable, but are in reality inefficient and hard to police.
Non-tangible variables such as faith, trust and community cohesion add to the complexity of what it means to be truly resilient. Optimist narratives of resilience serve to mask sub-standard recovery and the failure to improve the living conditions of many survivors. Policy-makers and relief agencies tasked with disaster recovery should adopt more nuanced approaches to resilience building. They should also reflect on how the narrative of resilience is produced and consumed and to what end. Disaster relief practitioners also need to be realistic about what they can achieve. Disaster rehabilitation efforts are unlikely to 'fix' disaster prone societies characterized by inequality, poverty, corruption and weak or unstable governance even before disasters hit or stressors reach crisis point.