article

doi:10.1111/j.0361-3666.2009.01124.x

In the aftermath of the Qa’yamat:1 the Kashmir earthquake disaster in northern Pakistan

Sarah J. Halvorson Associate Professor and Jennifer Parker Hamilton Graduate Researcher, Department of Geography, The University of Montana, United States

This article explores the local impact of the catastrophic earthquake in northern Pakistan on 8 October 2005. Drawing on field research, including interviews with 40 earthquake survivors, the post-disaster analysis presented here focuses upon risk awareness and the reactions of re- spondents to the 7.6 magnitude earthquake that devastated areas of Azad Jammu and Kashmir State, and North-West Frontier Province. The analysis provides insights into local perceptions of seismic hazard and exposure as well as survivors’ priorities with regard to rebuilding and reconstruction. The article suggests that the tragedy of the devastating earthquake is entangled in a deeper knot of causal factors that are social, economic and political in nature. Rapid popula- tion growth, urbanisation, changing building styles, environmental degradation and lack of preparedness and mitigation are associated with the circumstances that place the population at risk. Remarks concerning present and future risk reduction efforts are included.

Keywords: earthquake disaster, local perceptions, Pakistan, reconstruction, risk

Introduction

Earthquakes, particularly in developing countries, often have catastrophic impacts. In the Islamic Republic of Pakistan the earthquake that occurred on 8 October 2005 at 8.50 a.m. resulted in more than 73,000 deaths, at least 69,000 severe injuries and more than 3.5 million people left homeless (ERRA, 2006). In terms of death and destruction of property, the Kashmir earthquake is the greatest single tragedy the country of Pakistan has faced. Similar to earthquakes elsewhere, the Kashmir earth- quake was the result of high energy geological processes that ‘occur irrespective and independently of social action and any modification of the environment’ (Wisner et al., 2004, p. 274). Progress has been made in the scientific understanding of the seismic threat in the Himalayas (Bilham, 2004; Bilham et al., 2001). Nevertheless, social processes have played a fundamental role in determining the vulnerabilities and resilience of local residents to this earthquake disaster.

The epicentre of the earthquake, magnitude 7.6 on the Richter scale, was located at 34.493°N, 73.629°E, on the western end of the Himalayan Arc created by the con- vergence of the Indian and Eurasian tectonic plates (see Figure 1) (Bendick et al., 2007; GSP, 2006). The epicentre was north-east of Muzaffarabad, the capital of the state of Azad Jammu and Kashmir (AJK), located along the contentious Line of Control separating Pakistan and India. The earthquake affected a very large area,

Disasters, 2010, 34(1): 184−204. © 2010 The Author(s). Journal compilation © Overseas Development Institute, 2010 Published by Blackwell Publishing, 9600 Garsington Road, Oxford, OX4 2DQ, UKand 350 Main Street, Malden, MA 02148, USA

Figure 1 The location of the research setting

Source: Map created by B. Growley, Department of Geography, The University of Montana (2007).

including neighbouring areas of North-West Frontier Province (NWFP) and the capital, Islamabad (105 kilometres or 65 miles from the epicentre), where the tremors resulted in the collapse of a major residential/office building that caused over 70 deaths. Seiches were observed as far away as Uttar Pradesh and Bangladesh (USGS, 2005). The environmental consequences were tremendous as numerous landslides, rock avalanches and debris flows ensued (Dunning et al., 2007; Owen et al., 2008).

Some general assessments of the impacts of the earthquake have been undertaken (CRPRID, 2006; EERI, 2007; Khan and Mustafa, 2007; Mahmood, 2007; Özerdem, 2006). Additional studies have focused attention upon the experience of women and children (Hamilton and Halvorson, 2007; IFRC, 2006; IUCN, 2006; Shirkat Gah Women’s Resource Centre, 2006), coordination of relief operations and humanitarian aid (Hicks and Pappas, 2006; Wilder, 2008), building safety (Wisner, 2005), recon- struction (ERRA, 2006) and biogeophysical consequences (IUCN, 2005; Sudmeier- Rieux et al., 2008). These analyses help to underscore the enormous challenges ahead in reconstruction and earthquake risk reduction in northern Pakistan.

The present article aims to contribute to this body of work by reporting the results of a qualitative study of earthquake survivors’ views of risk and reconstruc- tion. The geographical setting of the study included the three heavily impacted valleys of Neelum, Jhelum and Khaghan. The objectives guiding this study were:

In the aftermath of the Qa’yamat 185

186 Sarah J. Halvorson and Jennifer Parker Hamilton

1) to document earthquake survivors’ perceptions of seismic hazard and exposure, the impacts of the earthquake and reconstruction priorities; 2) to develop a deeper understanding of the root causes of social vulnerability and the conditions that gave rise to this catastrophe; and 3) to identify practical and meaningful ways to enhance local resilience through learning and adjustment in the face of this destructive event. The overall goal of this paper is to inform the post-disaster reconstruction process in order to move toward measures of disaster risk reduction that are practical and realistic in the research area.

In the following section, we provide background information to underscore northern Pakistan’s seismicity and vulnerability to earthquake disasters. This section is followed by a discussion of the data and methods used in the study. After the presentation of the major findings, a synthesis and discussion section follows. The article concludes with several remarks regarding the implications of this study for policy and future research.

Geographical setting of the earthquake disaster

Owing to its location on the western edge of the Himalayas, the topography of AJK and the northern part of NWFP is extremely mountainous with very limited low- lands or arable land. Seismically, Pakistan is an extremely active area. In the past 75 years three earthquakes—Quetta (1935), Makran (1945) and Kashmir (2005)—have exceeded magnitude 7.5 and together caused more than 120,000 deaths (Khan, 2007). Since 2002, two earthquakes of magnitude 6 have struck northern Pakistan: Astor (2002) and Upper Hazara (2003) (Khan, 2007). The rugged terrain is charac- terised by significant elevation gradients ranging from 360 metres to 6,325 metres. The climate is marked by a hot season, heavy precipitation associated with the South Asian monsoon during the months of June to August, cooling temperatures in the autumn, and cold and snow between November and March. In addition to seismic activity, other mountain hazards include floods, landslides, rock falls, avalanches and extreme cold (Prime Minister Secretariat, 2007).

The population of AJK is around 3,271,000, with a high population density of 252 individuals per kilometre squared (AJK, 2005a). Similar to the rest of Pakistan, the population is relatively young; 36 per cent is under the age of 15 (Yin and Lalasz, 2005). The municipality of Muzaffarabad, one of the most historic towns in AJK, rests at the confluence of the Neelum and Jhelum rivers. Prior to October 2005, 80,000 people lived in the city. Another large municipality in the area is Balakot, located on a fault line near the north-eastern edge of NWFP. Once a popular rest stop along the roaring Kunhar river, Balakot was 80–90 per cent destroyed by the earth- quake. Until recently, both settlements were clustered around a town centre and market districts. Over the past few decades, dwellings and urban infrastructure expanded on hillsides and steep slopes of the surrounding mountains, thereby creating highly vulnerable situations. Today, extensive terraces and road networks are evidence of the intensive human use of this unstable mountainous environment (see Figure 2).

In the aftermath of the Qa’yamat 187 Figure 2 Many structures located on or near slopes were destroyed by landslides

triggered by the 8 October earthquake

Photo: J.P. Hamilton (November 2005).

The majority of the population depends upon agriculture, forestry and animal husbandry for its livelihood. The major crops are maize, wheat and rice, along with vegetables, pulses and oil seeds. Fruit and nut production are also important to the local economy. With an average annual income of PKR 96,000 (USD 1,600), 32 per cent of the population is at or below the poverty line. In response to the high un- employment rate of 35–50 per cent, more than 700,000 Kashmiri men leave the region to work in the Middle East, the United Kingdom or elsewhere (AJK, 2005a; Khan and Mustafa, 2007).

The tectonically active region lacked a disaster plan at the time of the earthquake. The lack of disaster preparedness was in part related to the fact that AJK is a highly inaccessible militarised zone owing to its status as a territorial battleground between Pakistan and India. This disputed zone has been at the heart of three wars between the two countries and continues to be riddled with political uncertainty.

The impacts of the earthquake were far-reaching. Approximately 600,000 houses were destroyed (ERRA, 2006). Many government buildings, mosques, hospitals and health centres were devastated. UNICEF estimates that at least 17,000 children were killed when more than 7,000 shoddily constructed schools collapsed (see Figure 3) (Wisner, 2005). Approximately 50–70 per cent of the sanitation, water supply, telecom- munication and power infrastructure were destroyed, along with over six thousand

188 Sarah J. Halvorson and Jennifer Parker Hamilton Figure 3 Government schools were among the most devastated buildings. Only a few

rows of seats remained in this school auditorium

Photo: J.P. Hamilton (November 2005). Figure 4 Destruction of critical infrastructure such as this bridge rendered some areas

entirely cut off from relief and aid supplies

Photo: J.P. Hamilton (June 2006).

kilometres of road networks (ERRA, 2006). The infrastructure impairments along with the collapse of bridges greatly hampered the relief effort (see Figure 4). For many villages, the only way to access relief was by helicopter. The total direct economic impact for Pakistan was estimated at USD 5.2 billion.

In the hours, days and weeks of damaging aftershocks and landslides, survivors fled from mountain villages into larger towns and urban centres in search of aid, safety, shelter and protection from the freezing temperatures. In response, a massive relief effort was organised by the Pakistani military and assisted by foreign govern- ments, international non-governmental organisations (INGOs) and an unprecedented amount of volunteerism by individuals and civil society groups in Pakistan and the Himalayan region. At a mid-November 2005 donor conference, donors pledged USD 6 billion in aid. Thousands sought shelter in government and INGO tent camps or the homes of friends and relatives. Three months after the earthquake 466 tent camps provided shelter to over 252,000 individuals (Mahmood, 2007).

Data sources and methods

Fieldwork for this study was undertaken during two field campaigns: four weeks in November 2005 and four weeks in May–June 2006. The timing of the field inves- tigations allowed for some comparisons to be made between the initial weeks after the earthquake and six months post-event. Semi-structured interviews (from 30 to 60 minutes in length) were conducted during the second field campaign with 40 residents (26 men, 14 women) ranging in age from 13 to 75. The interviews consisted of 18 open-ended questions. Questions were focused upon the perceived impacts of the earthquake, attitudes toward the relief efforts, priorities for rebuilding and percep- tions of the rebuilding process. An effort was made to include an equal number of male and female interview participants in order to capture their unique and differing perspectives. However, owing to issues around gender relations, cultural practices and social mobility, an equal ratio of female to male study participants was not possible.

All respondents were from the heavily impacted Khaghan, Jhelum and Neelum valleys, and specifically from ten communities:

16); 3); 2); 1); 2); 1); 2); 1); 8); and 3).

In the aftermath of the Qa’yamat 189

190 Sarah J. Halvorson and Jennifer Parker Hamilton

This was a convenience sample that was identified with the help of Pakistani field assistants. Nineteen of the respondents were living in their own homes, while 20 were living in tent villages.2 Of those in tent villages, four respondents had been renting prior to the earthquake and 16 had been living in their own homes. In addition to interviews, informal conversations were held with residents and relief workers. Field methods also included the following: quick visual surveys in temporary shel- ters and eight tent camps; walking tours of impacted structures and dwellings; and photo-documentation of damage to buildings and infrastructure.

Sensitivity to the socio-cultural setting was maintained throughout data collec- tion. Interviews were conducted in Urdu and translated into English. In keeping with the gender norms and widespread practice of purdah (female seclusion from the public), the field team sought permission from male heads of households to interview women. Women were interviewed in the private spaces of houses, tents or courtyards. Male study participants were interviewed in their homes or in public spaces, such as tea houses, businesses and markets. To protect the confidentiality of the study partici- pants, all subject names employed in this article are pseudonyms.

Local views on risk and reconstruction

The analysis of the field data reveals many important aspects of the earthquake experience and local residents’ initial impressions of recovery and rebuilding efforts. Of the 40 interview respondents, none had ever experienced an earthquake of this magnitude. For the most part, people were unaware that earthquakes of such incred- ible power were possible. Interview data and field observations indicated that residents generally did not know the level of hazard exposure before the earthquake or how to prepare for seismic hazard, nor were they informed of what to do during or after the event. Several indicators of these patterns include:

·  minimal earthquake awareness and knowledge among local populations;

·  widespread rumours about the cause(s) of the earthquake;

·  inappropriate building construction and a general lack of enforcement of building

codes, which was evident throughout the research setting;

·  very low public participation in preparedness and earthquake hazard mitigation

processes;

·  a lack of regional and national-level preparedness, information dissemination and

emergency services.

Perceptions of seismic threat and exposure

Responses to the question, Why did the earthquake happen here?, can be grouped into two categories: religious/metaphysical and geophysical. The majority of respondents (68 per cent, n=27) felt that the earthquake occurred because ‘it was God’s will’. Of the remaining respondents, ten (25 per cent) indicated that they did not know

the cause of the earthquake, and three (8 per cent) suggested geological reasons. Many held firmly to the idea that God’s supernatural powers were responsible for sending an earthquake to punish perceived societal wrongdoings, such as watching films or dressing inappropriately. Ahmid3 from Muzaffarabad commented, ‘The earthquake happened because of injustice in society and God’s orders not being followed, so he sent a curse in the form of an earthquake’. There was a widespread notion that only when community members followed the writings of the Qur’an more diligently ‘would God be pleased and stop punishing them’. Among the sug- gested geophysical causes of the earthquake was the response given by a 30-year-old teacher: ‘I have two guesses [as to the cause], one is hot springs and one is plates col- liding’. Ahktar, a 17-year-old student, suggested that ‘there is lava in the earth, and when it comes out there is an earthquake’.

When asked why some areas were more heavily impacted than others, nearly half of those interviewed (45 per cent, n=18) believed that the level of impact was controlled by God, and that he decided which villages were destined to be punished more than others. Responses included, ‘the good people suffer with the bad’, and, ‘if God likes you, he’ll keep you safe’. This fatalistic outlook was emphasised by another elder who indicated that the lessons brought by the October 2005 earth- quake would go unlearned. He stated, ‘Bad things bring earthquakes—this was a good example. I expect another [earthquake] soon based on people’s behaviours’.

A few respondents suggested additional factors that contributed to the massive destruction, for example: the inadequate structural capacity of buildings; the number of levels in buildings; the soil on which the structure was built; the surrounding topography; proximity to fault lines; subterranean gas; and the speed with which the earthquake passed through the area. In general, however, study participants lacked very basic knowledge of earthquake science and fundamental geophysical processes.

The interview data supports the observation that no warning system or earthquake preparedness plan was in place. Interestingly, nine of the respondents observed what they believed to be an earthquake precursor. The perceived precursors included changes in the behaviour of birds, dogs, cows and mentally disabled people, as well as the inability of individuals to sleep the night before. For example, a man from one village indicated that all of his animals were anxious and that his ‘cow wouldn’t give milk from one week before the earthquake’. Such forms of awareness of natural precursors have been well-documented elsewhere and found to be an indispensible component of cultural adaptations to seismic risk (Rikitake et al., 1993; Tributsch, 1984).

Only six respondents (15 per cent) felt that prior to the earthquake there was available information within the community to learn about earthquakes. Personal experience, teachers, newspapers and booklets from unspecified NGOs were all listed as sources of information about earthquakes. It is important to note that written forms of information dissemination are largely inaccessible to much of the population. The adult literacy rate in Pakistan is 62 per cent of males and 35 per cent of females (UNICEF, 2006). However, literacy rates are lower for the mountain- based population; the literacy rate among males in NWFP, for example, is 40 per cent while that of females is 18 per cent (AJK, 2005b).

In the aftermath of the Qa’yamat 191

192 Sarah J. Halvorson and Jennifer Parker Hamilton

Tent camps, torn communities

The overwhelming sentiment among Pakistanis seeking shelter and assistance in the tent camps was a desire to return home. During the first phase of fieldwork, the individuals with whom we spoke were primarily focused on the loss of many loved ones and the devastating collapse of schools, neighbourhoods and entire communi- ties. By May 2006, the general attitude of tent village residents was despondency and discontent with the relief effort. Typically, these individuals were waiting to hear from the government whether or not they would be allowed to return to the former locations of their homes. However, given the fact that they possessed few material and financial resources, many wondered how they would survive the pro- cess of rebuilding their devastated homes.

Respondents identified a range of issues with regard to the immediate and long- term impacts of the earthquake disaster upon their lives and communities. Nearly half of the respondents (47.5 per cent, n=19) felt that the high mortality rate was the most negative aspect of the earthquake. Mohammed summarised his feelings emotionally in this way, ‘Money can be made, buildings can be rebuilt, but I can’t get my family members back’. Another respondent commented, ‘I lost everything— there is nothing left’. Other significant impacts that were mentioned included the loss of peace of mind, mental anguish, interruptions in services such as water supplies (see Figure 5), and disruptions in the school system (see Figure 6).

Figure 5 Following the earthquake, tent village residents, particularly women and children, faced the challenge of accessing adequate amounts of clean drinking water

Photo: J.P. Hamilton (June 2006).

Figure 6 Classes were held outdoors in the summer heat while new school facilities were constructed. The school under construction here is designed to incorporate the lightweight government-recommended materials

Photo: J.P. Hamilton (June 2006). Figure 7 Squalor and destitution in a Muzaffarabad tent village

In the aftermath of the Qa’yamat 193

Photo: J.P. Hamilton (November 2005).

194 Sarah J. Halvorson and Jennifer Parker Hamilton

The tent villages were never meant to provide permanent shelter (see Figure 7). Fearful of the continued costs, the government set a deadline of May 2007 for earthquake victims to return to their villages. This directive provided additional stress for individuals already waiting to see if they would be able to access resources to support their return and the costs of rebuilding their homes. A compensation scheme for those who lost their homes was set up by the Government of Pakistan in late autumn 2005.4 Nevertheless, two years after the deadly earthquake, approximately 30,000 individuals faced their third winter in a tent, while 3.5 million continued to live in temporary or semi-permanent shelters.

Recovery, rebuilding and resettlement

Most of the government, military, civil society and international effort in the early stages of recovery focused on establishing temporary shelters, providing aid and med- ical care to victims, rebuilding damaged houses and businesses, the resettlement of the affected population and infrastructure reconstruction, especially the repair of the road network, power supplies and water supplies. In response to the question, Who has helped you the most since the earthquake?, almost half (42.5 per cent, n=17) replied that an NGO (either local or foreign) had provided the most help. A quarter of the respondents (25 per cent, n=10) replied that everyone provided aid and assist- ance. The assistance of the Pakistani government, local people and the Jamaat-i- Islami political party was also mentioned. The majority of respondents felt that it would take up to five years for the earthquake-devastated areas to fully recover, assuming that the government remained committed to providing financial and tech- nical assistance. For some, full recovery seemed a near impossibility, as one woman exclaimed, ‘We have to make a decision between shelter and food right now. [Full recovery] could take 1,000 years’.

The role of the government in providing financial support to rebuild houses and in identifying appropriate sites for reconstruction drew mixed impressions and critique. Saeed, a local school teacher, said, ‘The government did inspections and made a report. They promised to pay but haven’t yet. I expect the government will build homes for those that can’t build their own, and build primary buildings like hospitals and schools’. An elderly man from Patika elaborated:

The government should tell [us] whether it is safe or not and should provide drinking water because the old springs are dried up. Continuing rock falls scare us, and we want to know when they will stop. When to rebuild or when to move needs to be determined by the government.

In terms of respondents’ priorities in reconstruction, houses were listed as the top priority by 60 per cent of the respondents. Mosques (17.5 per cent, n=7), schools (15 per cent, n=6) and hospitals (7.5 per cent, n=3) were also given a high priority for initial rebuilding. When asked if buildings should be rebuilt in the same location rather

than being moved, 53.8 per cent (n=14) of men and 7.5 per cent (n=3) of women felt that structures should not be built in the same location.

For the rebuilding effort, different architectural styles and earthquake-resistant dwelling types were proposed by a variety of international organisations. The gov- ernment adopted a building scheme that incorporated the use of a wood frame with one-metre-high brick or concrete lower walls and corrugated galvanised iron (CGI) sheets for upper walls and roofs. From the interview data, it is apparent that public education regarding the importance of structural integrity and low roof weight had been conducted over the months following the earthquake. Eighty per cent of re- spondents felt that the government-recommended method offered the best approach for rebuilding dwellings. Another ten per cent (n=4) did not specifically mention the CGI building type, but did indicate that a lightweight wall and roofing material was imperative.

In spite of the broad support for rebuilding dwellings in a lightweight manner in order to reduce the risk of collapse posed by future earthquakes, accessing resources and specific information regarding the rebuilding scheme was complicated for res- idents to negotiate. Earthquake retrofitting and rebuilding with the recommended

Figure 8 The post-disaster landscape is marked by a patchwork of dwelling types— traditional mud-brick, the government-recommended CGI design and concrete cinderblock—reflecting earthquake survivors’ relative access to income, resources and financial support

Photo: J.P. Hamilton (June 2006).

In the aftermath of the Qa’yamat 195

196 Sarah J. Halvorson and Jennifer Parker Hamilton

lightweight alternatives to heavy mud, rock and concrete structures was beyond the means of most families (see Figure 8). More than half of the respondents indi- cated that they had no place to go to find information about specific construction methods or available building resources. Abdul from Muzaffarad said, ‘No one is listening. No one cares. Everyone is taking care of their own, buying perhaps one CGI sheet per month. I don’t expect much more to come’. Another respondent summarised the situation in this manner: ‘There is nothing. We can only ask God’.

The government made a decision to resettle residents of Balakot rather than rebuild the city, a decision that presented a difficult challenge owing to limited available land and resources and to human safety concerns. The Earthquake Reconstruction and Rehabilitation Authority (ERRA)5 determined that Balakot would be relocated away from the fault zone to an area approximately 20 kilometres south at a place called Bakrial. Construction of the resettled Balakot is expected to take three years (IFRC, 2007). Just over half of the respondents felt that the plan to relocate the entire city of Balakot was a good one. Some responded more fatalistically, such as Nazia, a 30-year-old mother, who said, ‘No, death will follow. It will come, no matter where you go, when it is your time’. When asked why Balakot was the only city to be moved (which was the case as of May 2006), there was little agreement, but the high- est percentage of respondents (17.5 per cent, n=7) indicated that they did not know.

The majority of the respondents believed at the time of interviewing that their settlement was unsafe for human habitation. As one respondent put it, ‘The area is not safe. People are tense and tremors make people worry. Areas should be planned to be safer’. Samina, a housewife from Muzaffarabad, said the area is ‘not safe, but there is nowhere to go’. Many individuals expressed interest in moving, but cited land and business ownership as reasons for remaining in the area. Ten per cent of respondents indicated that it was impossible to assess the safety and stability of the area because ‘only God knows’. These respondents found comfort in the idea that God was in control, as one respondent put it, ‘God will tell me. All is in God’s hands’. Nadeem, an agriculture department worker, said, ‘It doesn’t matter, red zone or green zone, only faith in God [matters]’.6

Root causes of the Kashmir disaster

As indicated above, the interview data paints a bleak picture of the lack of knowl- edge about the risk environment among the public, the challenges of recovery at the micro-level, and the resulting mismatch between residents’ needs and the re- construction schemes provided by the government and INGOs. We argue that this is only a partial picture of the catastrophe, one that is entangled in a deeper knot of factors that contributed to social vulnerability. These factors include: rapid popula- tion growth and concomitant urbanisation; the propagation of structures that were incompatible with the level of seismic hazard; environmental degradation and in particular deforestation; and a lack of seismic risk perception and planning at the national level.

First, uncontrolled population growth in combination with urbanisation con- tributed to the disaster in several ways. As Khan (2007, p. xi) states, ‘. . . Pakistan has one of the highest growth rates, resulting in its being the third fastest growing population in the world, and this is resulting in urban as well as rural expansion into hazard-prone areas’. The population growth rate in AJK from 1998 to 2002 was 2.41 per cent, a statistic attributable to a combination of natural increase and im- migration from outlying areas (AJK, 2005a). Hughes (1984) observed in the 1980s that a shortage of land in the valleys of northern Pakistan was pushing the population to build structures on steep slopes. The trend towards increasing population pressures in this mountainous environment has continued since that time, thereby drawing people into locations that are highly susceptible to seismic hazards. In an attempt to preserve scarce arable land for field crops and orchards, many families have no other option than to live on steep slopes that are geologically unstable. Owing to long- term denudation rates and extensive road cutting and road building, these steep slopes have over time become increasingly prone to landsliding and slope failure (Owen et al., 2008).

Second, the phenomenon of building structures incompatible with the level of seismic hazard can be partly linked to population pressures discussed above. Many of these buildings were erected quickly under minimal standards and using shoddy construction techniques and low quality materials. The failure of structures during the earthquake resulted from both the use of unreinforced and brittle building materials, and the prevalence of heavy roofs. Prior to the earthquake, many houses had thick, heavy, poorly attached concrete roofs. Our observations indicated that, typically, the walls of such houses collapsed entirely and the roofs came down in one massive slab, crushing everything between the roof and the floor. An additional concern was that approximately ten times as many governmental structures collapsed than did family dwellings during the earthquake (Khan, 2007). Our field research suggests that corruption among contractors and the lack of craftsmen trained in earthquake-resistant building techniques contributed to this depressing trend.

Third, the fundamental problems associated with structural failures in the area also stem from significant transformations in construction methods, architectural styles and local preferences (Davis, 1984; D’Souza, 1984; Halvorson and Hamilton, 2007; Hughes, 1984). Historically, populations inhabiting these hazard-prone areas of northern Pakistan integrated an awareness of high levels of seismicity into aspects of vernacular architecture using their own resources, technologies and knowledge of the physical surroundings. Studies have shown that traditional methods of house construction utilised timber and combinations of stone and clay, thereby achieving earthquake resistant properties (Dekens, 2007; Nash and Spence, 1984; Spence and Coburn, 1984). The elaborate use of wood to lace joining walls—a technique known as cator and cribbage—is one strategy that has been practiced in northern Pakistan for over a thousand years (Hughes, 2000).

The recent preference for ‘modern’ concrete cinder block houses has further driven the boom in seismically inappropriate construction. Home owners have been un- necessarily replacing earthquake resistant traditional dwellings with ones built in a

In the aftermath of the Qa’yamat 197

198 Sarah J. Halvorson and Jennifer Parker Hamilton

‘modern’ yet more vulnerable manner. This was the unfortunate response during the reconstruction of areas devastated by the 1974 Pattan earthquake in northern Pakistan. Davis (1984) found that the buildings constructed of reinforced concrete construction were likely to be inferior to the traditional structures they were replac- ing. With very minor modifications, traditional structures with non-load-bearing walls and timber baulks within masonry walls could have been made relatively earth- quake resistant. The post-earthquake reconstruction in this case reflected an overall tendency to rebuild structures with less earthquake resistance than was present prior to the earthquake.

A fourth factor contributing to social vulnerability is deforestation. Deforestation has triggered a depletion of available building materials and an erosion of the integ- rity of the built environment during the past three decades. Local access to timber for construction has decreased with the extraction of forest products (Hughes, 2000). Deforestation has largely resulted from the demands of the growing popula- tion and from legal and illegal commercial logging. Since traditional dwellings tend to require a significant amount of timber, residents of lower socio-economic stand- ing have had to seek out more feasible alternatives such as cheap concrete. Only the wealthiest residents have been able to dedicate a portion of their land to the cultiva- tion of the poplar trees used in roof construction. Furthermore, deforestation and concomitant biophysical change have compounded the problems of soil erosion, slope instability and slope failure (Sudmeier-Rieux et al., 2008).

Finally, until this disaster, earthquakes had been on the periphery of national-level disaster mitigation and preparedness initiatives (Wilder, 2008). According to Khan (2007), disaster management prior to October 2005 focused primarily on flooding. Though Pakistan has had five-year disaster management plans in existence since 1957, the plans were generally lacking specific measures, strategies or objectives (Khan, 2007). Pakistan’s earthquake preparedness strategy was restricted to relief, with minimal attention and investment going toward earthquake disaster planning and preparation. As Khan (2007, p. xii) points out,

In Pakistan, the Meteorological Department deals with earthquakes, and this [natural hazard] is not only beyond their capacity in terms of expert manpower and instrumenta- tion, but also is far too much of a burden considering [the Meteorological Department’s] crucial focus on weather-related issues, climatology, and flood forecasting. The lack of a dedicated institution for earthquake hazards has resulted not only in lack of reliable seismicity, ground motion, and neotectonic data, but also has hindered development of appropriate earthquake resilient building codes and their implementation.

The lack of perception of seismic risk at the national level can be seen as a major cause of the disaster associated with the 8 October earthquake. As Wisner (2005), ICIMOD (2008) and others have suggested, ongoing efforts to cultivate an awareness of seismic hazard is critical to generating the political will to ensure appropriate land use, the structural safety of buildings (especially schools and health facilities), and the government’s capacity to address the critical issues surrounding victims’ needs, rebuilding priorities and the resettlement process.

Synthesis and discussion

The Kashmir earthquake should be considered as an extension of an ongoing socio- environmental situation rather than merely the outcome of an extreme event. As the foregoing paragraphs noted, demographic pressure, processes of urbanisation, the propagation of structures that are incompatible with the level of seismic hazard, forest degradation and the overall lack of local and national-level preparedness and mitigation influenced the circumstances that placed the mountain-based population of AJK and NWFP at risk. It seems very unrealistic indeed to delink the everyday struggles highlighted by earthquake victims from the broader issues of ‘normal life’ (for example, Wisner et al., 2004) that have been shaped by underdevelopment, pov- erty and unsafe conditions. Difficulty in accessing arable land and economic resources is now an even greater daily concern of earthquake victims, rendering them more vulnerable as they attempt to rebuild their lives in this tectonically-active zone.

The dominant view of God as the major agent of disaster or survival in this part of the western Himalayas reflects more than the enduring nature of belief systems. Emphasising the metaphysical dimension of human tragedies provides an explana- tion, yet also underscores a sense of being in an inescapable situation. The associa- tion between disaster and fatalism has been analysed elsewhere (see, for example, Grothmann and Reusswig, 2006; Kasapoglu and Ecevit, 2003). These studies sug- gest that with this level of fatalism comes a passive attitude regarding preparedness. As described by Hutton and Haque (2003, p. 416), in areas where ‘individual choice and power are clearly lacking, people are likely to accept the “inevitable as inevi- table” or, in the case of rural poor, to accept it as the “will of Allah”’.

The interview data highlights extreme amounts of ‘informational vulnerability’ (Degg and Homan, 2005) present in the earthquake affected area. This was made obvious by the fact that not one of the respondents was aware that an earthquake of such magnitude was possible, or had previously been informed about what to do or how to build to prepare for an earthquake. The finding is somewhat surprising given the history of devastating earthquakes in the area.

Traditional seismic knowledge would need to be relearned in order to support what Gardner and Dekens (2007) refer to as ‘resilience building and enhancement’. Indigenous seismic knowledge was traditionally passed down in northern Pakistan from generation to generation. Owing partially to migration, urbanisation and the ensuing livelihood transformations, the population is now susceptible to a lack of information dissemination and science-based earthquake education. The age struc- ture of the population further plays a role in the constrained capacity to reduce risk and respond effectively during an earthquake. On the other side, the built environ- ment has become increasingly unsafe with changes in building styles and materials, partly owing to long-term rates of deforestation. Traditional buildings require a sig- nificant amount of timber for proper construction. Given the general lack of timber available in the area, using wood for construction is not feasible. Additionally, the preference towards more ‘modern’ looking dwellings drives the market away from

In the aftermath of the Qa’yamat 199

200 Sarah J. Halvorson and Jennifer Parker Hamilton

timber and towards cheap concrete. A hope is that reforestation programmes can be implemented to serve the needs of the population, bolster road cuts and stabilise steep slopes.

One of the outcomes of the disaster was that it set in motion the creation of an earthquake disaster infrastructure at the national level. There are efforts under way in Pakistan that would help to shift institutions and approaches away from reactive, relief-based modes towards more proactive preparedness-based systems. Since 2005, disaster preparedness and planning has been prioritised with the creation of two institutions—the National Disaster Management Commission and the National Disaster Management Authority—to implement, coordinate and monitor disaster management activities in Pakistan. These are positive steps towards what Twigg (2005) describes as ‘institutionalising’ risk reduction.

More earthquake disasters in the future?

While the Kashmir disaster was triggered by a tremendous seismic event, the expe- rience of the earthquake victims emphasises the deeply rooted nature of the causes of social vulnerability in this mountainous region. The perspective provided by local residents illustrates the ways in which this event was perceived and experi- enced in the weeks and months following 8 October 2005. It highlights a set of complex and ongoing issues of local concern, such as safety, access to information and resources, resettlement concerns, reconstruction priorities, ecological degrada- tion and impoverishment.

In addressing the objectives outlined at the outset of this paper, this study brings to light several implications. It suggests that people living in the region were inadequately prepared and lacked fundamental knowledge about seismic hazards. Awareness of the perceptions and adjustments of local residents to the environmental disruption and chaos is essential for moving beyond a static view of this disaster towards re- building and reconstruction in ways that support the long-term goals of inclusive and sustainable disaster risk reduction. Certainly, the men and women who participated in this research had varying views and experiences, which point to the need for fur- ther study of the ways in which gender, age, class, caste, ethnicity, religion and socio- economic standing structure people’s vulnerabilities and response capacities in the research setting.

Moreover, the study sheds light on the deeply rooted factors underlying social vulnerability in the area, including demographic pressure, urbanisation, changes in the built environment, environmental degradation and inadequate earthquake dis- aster planning and preparedness at the national level. As such, a case is made for the need for continued, long-term monitoring of the reconstruction and rebuilding process. The reconstruction process presents a potential opportunity to improve the structural integrity of homes, empower local women and other grassroots organi- sations, educate community members and policy makers about seismic hazards,

and address the very roots of earthquake vulnerability. Currently, very few civic leaders, teachers and children are exposed to a curriculum that supports learning about mountain hazards, disaster preparedness and slope stabilisation techniques. However, great potential exists for applying new approaches for opening the pathways between scientific knowledge and the public in Pakistan (for example, Alexander, 2007; Coburn and Spence, 2002; Hamilton and Halvorson, 2007). It is critical that ‘informational vulnerability’ is overcome by promoting accurate and informative disaster education for this mountain-based population, especially for women, chil- dren and the poor.

The disaster management and planning sector in Pakistan is becoming more cognisant of linking disaster mitigation to daily life. Nevertheless, there is a critical need to identify conceptually and empirically the structural constraints underlying vulnerability—namely poverty, rapid urbanisation, declining environmental quality and the lack of societal protection in the areas of building codes and their enforce- ment. Khan (2007) has argued for a complete rethinking of the contemporary approach to disaster mitigation in Pakistan. As long as the larger reconstruction policies are disconnected from the socio-economic and political context of development chal- lenges in AJK and NWFP, the expectation is for more earthquake disasters to occur in years to come.