Understanding disaster (in)justice: Spatializing the production of vulnerabilities of indigenous people in Taiwan

Abstract

Existing disaster studies scholarship tends to uncritically privilege official institutional responses to disasters over bottom-up, community-based reactions and adaptations in the longer term. Meanwhile, the state-recognized disasters mostly exclude socioeconomic and environmental contradictions that generate disasters by making people vulnerable in the first place. Discussion of disaster justice, then, is limited to the immediate state responses to understanding disasters as natural episodes, often incorporating distributive justice into its policy responses. A more spatialized understanding of disaster justice should move beyond these dual limitations—constraints in defining disasters as isolated episodes and in planning for recovery as only emergency responses—and pay attention to the socio-spatial production of risk. This approach would better attend to underlying societal vulnerabilities created through urbanization, in this case, of the indigenous peoples of Taiwan. A spatialized understanding of disaster justice guides this research to attend to both the reoccurring displacements of indigenous people from where their socioeconomic–ecological relations were embedded and the systematic production of risk that has been reinforced and exacerbated by inadequate land use planning. Based on a study of two indigenous groups in Southern Taiwan, Kucapungane and Makazayazaya, this paper explains disaster justice from the indigenous context wrought by multiple expropriations and displacements from the 1930s onwards. This paper plans to do two things: First, it will investigate bottom-up responses to a series of historical disaster-driven migration/urbanization through theories of community resilience. Secondly, it will expand understanding of “disasters” by taking into account the state-led relocation projects as human-made disasters and examine how they have contributed to the vulnerability of the indigenous communities under climate change. It will contribute to a better understanding of disaster justice that overcomes the dual limitations and opens up opportunities for alternative research and practices.

Keywords

Indigenous communities disaster justice Taiwan recovery planning Anthropocene

Introduction

Today, human beings fail to understand—if ever there was a willingness to do so—the combined effects of human actions that may not each present an issue on an individual level, but become one when they together form an imponderable and disastrous event that is capable of altering the basic ecological cycles of the planet. The crisis of scale and agency underlies a disjunction of the increasingly serious effects of environmental change and people's daily sense of responsibility, which can be exemplified by the anger and confusion of those who lost their homes and/or families after Typhoon Morakot hit Southern Taiwan in 2009. But this article looks beyond what happened in 2009 by recontextualizing a case of recovery planning to offer a historicized understanding of disaster (in)justice. This research examines justice in the indigenous context wrought by multiple expropriations and displacements from the 1930s onwards in order to overcome the crisis of scale and agency in understanding disasters today.

With the Anthropocene becoming a buzzword lately, it is not enough to just recognize the increasing power of humankind as part of the biosphere (Crutzen and Stoermer, 2000) without activiely engaging with more interdisciplinary research to better understand and address its consequences. The Anthropocene disorder—the scalar gap between “what can I see and what is really happening”—that Timothy Clark (2015) identifies is a serious issue, especially in Asia because population growth and rapid urbanization have brought people together with conflicting senses of norms and responsibility. Indeed, increased extreme weather due to climate change are gradually causing more casualties than other catastrophes (Bouwer, 2013). Some of them have particularly changing the timing, intensity, and extent of the South Asian monsoon and thus affecting rainfall over East Asia (Lau et al., 2006; Lewis et al., 2011). The cost of living in the Anthropocene can be particularly high in the region.

With most of the study of the Anthropocene remaining largely a topic in science and environmental studies, the emerging interest in the crisis of the Anthropocene in Area Studies with a focus on Asia (Hudson, 2014; Oakes, 2017) is worthy of attention. It is argued that case studies of long-term strategies of the locally contextualized systems of adaptive learning in Asia, which area studies specialists have been doing but sometimes appears like rather parochial scholarship (Oakes, 2017), can be adjusted and then contribute to scholarship of the Anthropocene. Meanwhile, it is important not to repeat the earlier Orientalist description of Asia that assumed there was Asian harmony with nature and ignored the reality “that the region has a very long history of ecosystem transformation by humans. Agriculture (especially wet-rice farming), deforestation, urbanization, advanced structures of economic production and exchange, and coal use in premodern China all had major impacts on the environment” (Hudson, 2014: 954). Others also point out that anthropogenic landscapes have long and substantial histories in many parts of Asia (Aikens and Lee, 2013).

Importantly, there is significant discrepancy within Asia as a broad territory. For example, the indigenous communities in Hokkaido and Taiwan had remained their living depending on a mix of hunting and small-scale slash-and-burn until recently, which differed from the more sedentary settlements characterized by intensive rice farming in the more populated areas in Asia. Thus, the argument about rice farming being a major preindustrial source of atmospheric methane, even more than carbon dioxide (Fuller et al., 2011; Mitchell et al., 2013) are not applicable to those indigenous communities in Asia, whose less sedentary lifestyles contributed less to the environmental transformations yet the distribution of risk somehow reached them. The many ways in which the indigenous communities within Asia have been involved in the Anthropocene both as a global and local process particularly demand careful scrutiny, especially from the indigenous perspective in understanding the Anthropocene (Castree et al., 2014; Yeh, 2016). In this light, a discussion of disasters should be pushed well beyond the management and policy focus to treat “catastrophe as a topic capable of transcending entrenched epistemological divides” (Oakes, 2017: 405).

Asia is also a region where the colonial legacy of resource extraction and forced relocation of indigenous communities remains at work though colonizers may appear long gone; this makes all the accumulated results of Anthropocene disorder even more indiscernible. Yet policymakers and laypeople prefer to believe that there is a clear boundary of disasters, demanding certain responses and responsibilities on a certain scale. While it is understandable that politicians have to cope with urgency, the political opportunism arising from action as such, however, is not friendly to long-term policy and more careful planning. Such a tension between long-term policies and coping with political pressure, as argued below, is unjust when it comes to understanding disasters as processes in Asia.

Actions for disaster justice must include an understanding of how human beings affect one another on multiple scales across time and space, with a recognition that vulnerability and uncertainty are socially and politically produced rather than naturalized preconditions (O'Brien et al., 2006; Perry and Quarantelli, 2005). It was not until the 1990s that disaster research slowly moved from a “(hazards) agent centered” approach to an increasing focus on vulnerability (Perry, 2007) and argued that disasters are “social in nature” (Mileti, 1999: 3). The shifts in disaster scholarship have gradually allowed for a long-awaited discussion on disaster and justice that the Anthropocene disorder underlies. But in the world of practices outside of academia, it has been largely ignored, as much as the multiscalar nature of disasters, in disaster governance (Miller and Douglass, 2016). More work has been done in public health, especially in relation to the spatial and racial character of the environmental riskscape (Morello-Frosch et al., 2001; Morello-Frosch and Shenassa, 2006) but not as much in post-disaster planning until recently (see one example in Tierney and Oliver-Smith, 2012). Discussion of justice and disaster, then, is often limited to the immediate state responses, and to limited understanding of disasters as natural episodes, often involving the adoption of distributive justice in state policy implementation. The bias in disaster governance contributes to a particular kind of disaster scholarship, which tends to uncritically privilege official institutional responses to state-recognized disasters over bottom-up and informal reactions and adaptations in the long term. Acting toward disaster justice requires us to overcome such a bias. To be more precise, disaster justice, as understood here, is about acting against injustice as resulting from Anthropocene disorder without theorizing justice as decontextualized normative claims, which urges us to address the crisis of agency and scale by understanding disasters in adequate temporal/historical and spatial contexts.

The realization of disaster (in)justice in a way extends the scholarship of risk and vulnerability (Blaikie et al., 2006 [1994]; Burton and Cutter, 2008; Schmidtlein et al., 2008). It guides research to attend to the long-term production of vulnerability that leads to disaster and thereof unequal distribution of socioecological conflicts caused by a modern economic system driven by the industrial revolution and colonization (Escobar, 2008). In Leichenko and O'Brien's (2008) words, this unequal distribution of socioecological conflicts results from “double exposure” to global environmental change and market-driven processes of globalization; these often hide themselves in forms of spatialization, such as urbanization and relocation after disasters, which demand a spatialized reading of the production of vulnerabilities. Indigenous people, who are often the populations suffering the most from this naturalized unequal distribution of conflicts, are the focus of this article. I attend to the recurring displacements of indigenous people from where their socioeconomic–ecological relations were embedded, and from there, to trace the systematic production of risk.

Based on a study of long-term displacements of two indigenous tribes in Southern Taiwan, Kucapungane and Makazayazaya, this research examines justice in the indigenous context wrought by multiple expropriations and displacements from the 1930s onwards; this context has been entangled with a global assemblage of development initiatives, variegated governance through the export of socio-natural resources (Ogden et al., 2013; Tsing, 2005), and a landscape framed as left-behind dangerous areas. I conducted research from October 2010 through January 2017 by employing participatory observation, paying 13 site visits, varying from 2 days to 14 days each time. I attended meetings with policymakers, public hearings, community meetings, informal discussions, ceremonies, etc. The research was complemented by interviews with key actors such as affected villagers, community leaders and organizers, officials from different levels of government. Data anonymization is adopted except with few that the informants agreed to go public. In the following, this article aims to do two things. First, it expands the current understanding of “disasters” by taking into account state-led relocation projects as human-made disasters and examine how they have contributed to the vulnerability of indigenous communities under climate change. Second, it investigates bottom-up responses to a series of historical disaster-driven migration/urbanization episodes to understand the complex, interactive, and reactive process of bottom-up responses. It ends with a reflection on the idea of disaster injustice in the Anthropocene.

Theoretical groundings: Connecting vulnerability with disaster justice

Disasters are a complex mix of natural hazards and human action. […] A disaster is not a single, discrete event. (Blaikie et al., 2006 [1994]: 5)

In At Risk, Piers Blaikie et al. (2006 [1994]) set out the Pressure and Release model (PAR) to explain, at the micro level, the production and trajectory of vulnerability and its variation between individuals and households. Blaikie et al. point out that so-called “natural disasters” are often only “trigger events” that bring about a transition to disaster when there is an unsafe condition in the first place. Existing social and environmental issues such as poverty, deforestation, or low capacity for self-protection may contribute to the greater vulnerable situation. It places emphasis on “the various ways in which social systems operate to generate disasters by making people vulnerable” (Blaikie et al., 2006 [1994]: 10). More importantly, it challenges the mainstream disaster research that tend to neglect how the actions of states and the agenda of development and globalization produce disasters (Tierney, 2007: 509–510).

Along these lines, it is also important to attend to the role that “vulnerability” plays in the making of disasters. In At Risk, vulnerability is defined as “the characteristics of a person or a group and their situation that influence their capacity to anticipate, cope with, resist and recover from the impact of a natural hazard” (Blaikie et al., 2006 [1994]: 11) Moreover, vulnerability is constituted by social networks that encompass the individual, household, kinship network, and larger collectivities, which may all develop implicit or explicit strategies to manage risk and provide the basis for action when vulnerability is made a reality by the disaster event (Blaikie et al., 2006 [1994]: 15).

The recent research trend that sees the causes of disasters as inevitably socially constructed rather than purely natural (Dynes, 2000; Perry and Quarantelli, 2005; Quarantelli, 1997) does not deny the disastrous effects but asks for a more careful consideration of the role of interpretations, politics, and institutional practices in defining disasters and necessary responses to particular disasters. Meanwhile, this view traces the origin of disasters in a longer time frame and therefore brings up issues of intergenerational equity and responsibilities. Yet despite progress in disaster research, we mostly fail to account for how long-term policy and politics produce vulnerable conditions to precipitate natural disasters, and therefore are complicit in reinforcing the problem of “disaster gerrymandering” (Platt, 1999), or disaster (in)justice. Disaster scholarship is yet to focus more on existing socioeconomic and environmental contradictions that generate and exacerbate disasters by making people and cities vulnerable in the first place (Blaikie, 2006 [1994]; Gotham and Greenberg, 2014; Tierney, 2007).

To study the ways in which state-led urbanization of indigenous communities contributes to the making of vulnerable conditions under which Typhoon Morakot triggered a disaster in 2009, I adapt the PAR model developed by Blaikie et al. (2006 [1994]: 50–51) to study the making of Rinari, a resettlement project as part of the recovery effort in response to the disaster. The model divides the production of vulnerability into three phases in progression: root causes, dynamic pressures, and the production of unsafe conditions that eventually lead to disasters as trigger events when there is hazard (Figure 1). More concrete exemplary details will be presented in the third section.

Figure 1.

The progression/production of vulnerability in the case of Rinari based on the PAR model.

The approach adopted here challenges the kind of discussion of justice following disaster that is limited to immediate state responses as well as to a limited understanding of disasters as natural episodes; such responses and understandings often adopt distributive justice in states' policy implementation. In contrast, the present approach calls for a more spatialized understanding of the processes leading to disasters, that is, the socio-spatial production of risk and vulnerability, to move beyond the dual limitation of limits in defining disasters and in planning for recovery. As Parthasarathy (2018) notes, referencing the case of informal settlement in Mumbai, “[u]ncertainty is exacerbated by ill-conceived adaptation strategies that are themselves the outcomes of unequal power relations”. In this light, it can be unjust to apply existing fair and just models of distribution to deal with disaster effects if the long-term production of vulnerability is not taken into account. Similar reflection can be found in a recent criticism to resilience thinking by Evans and Reid (2015a) in their book Resilient Life, with which they politicize the discourse of resilience by interrogating its implications for political subjectivity. They call for attention to how resilience thinking has become “a resilience doctrine” that operates as “a new form of political nihilism that forces us to accept the inevitability of the liberal politics of catastrophe” (Evans and Reid, 2015b). The new doctrine of resilience prevents us from questioning the rule of markets as a principal driver for human interactions and social affairs in its successfully naturalizing all the human failures to uncertain, inevitable disasters. Meanwhile, with the planetary (Rockström et al., 2009: 32) emerging as a new geographical imagery for human as one community living in the Anthropocene, change expected means the adaptation of personal behaviors by particular populations, especially indigenous populations in resource-rich areas, who mostly have contributed the least to catastrophic events but ironically have been encouraged to conform to changes in the interest of “planetary stewardship” (Evans and Reid, 2015a: 9). In other words, there is a problem in the way in which the neoliberal doctrine of resilience makes the poor even more vulnerable by turning them into agent of their own change. Worse still, the poor are seen as the most “ecologically ignorant” (Evans and Reid, 2015a: 35) and considered as those who should be adapting to a new system where economic notions like property rights or ecosystem services are brought in so that markets can function in them.

Following this line of argument, it is important not to confuse historicizing disaster (in)justice with “impress(ing) the past upon the present to provide a moral reasoning to future governmental activities as carried out in the name of a collective, which is defined on account of its radical endangerment” (Evans and Reid, 2015a: 25). Instead, it is necessary to ensure that the complex memories of social changes in relation to catastrophic events are not purged of their political traces or reduced to arbitrary and inescapable violence of the Anthropocene. In other words, it is necessary not to allow political responsibilities to escape from public scrutiny and instead replaces with manipulated resilient thinking that would reinforce “vulnerability” as the subject's unquestionable ontological status (Evans and Reid, 2015a: 29), which largely differs from the way vulnerability is discussed in this paper.

A more contextualized reading of the uncertainty faced by indigenous people and the way they adapt to disasters can prevent us from essentializing and positively or negatively racializing qualities and characteristics associated with resilience. Thus, we can reject the tendency to describe the poor, minorities, or people of the global south as being more vulnerable and resilient at the same time, as if their exposure to and experience of dealing with a multiplicity of hazards and risks can be justified or, worse still, naturalized to some degree (Parthasarathy, 2018). This tendency, as Parthasarathy (2015) reminds us, may focus on social characteristics, structural aspects, and forms of social relations that engender resilience and lead to a misunderstanding of the community's great effort in pursuing recovery as a spontaneous ability to bounce back from a disaster's aftermath. Resilience thinking, as it oftentimes draws on a bounded system—a representational form of socio-ecological systems (SESs)—tends to privilege proximate rather than underlying structural and historical processes, which has some shortcomings (Yeh, 2017). Yeh potently reminds us that “[r]elational power, interest, multiple social identities, and the interplay of structure and agency, are difficult to represent and thus easy to lose sight of…The dynamics of capitalist accumulation and class struggle are, however, strikingly absent from SES framework studies of resilience and adaptation” (2017: 37). Meanwhile, studies of vulnerability and adaptation to climate change easily fails to address the colonial past and present and then risks delimiting indigenous perspectives and reinforcing dispossession and accumulation (Cameron, 2012; Yeh, 2017). In other words, vague and generic descriptions of a population that may render the indigenous as essentially resilient can be unjust in themselves.

Notably, from an Science, Technology and Society (STS) perspective, science and technology might simultaneously produce risk in modern industrial societies and provide tools to assess and manage it (Fortun et al., 2016: 1003). Thus, STS studies calls for critical reflection on established methods, theoretical frameworks, and “ways of representing and sharing research results” (Fortun et al., 2016: 1004), which echoes Yeh's concern for SES. Fortun et al. (2016: 1005) argue for more expansive time frames for understanding disaster and for denaturalizing disaster disparities, which resonates with Bankoff (2003), who calls for deeper historical analysis of vulnerability earlier. This is particularly critical reminder for understanding the connection between disaster justice and vulnerability—to trace the complicated production of vulnerability across time and scales to know where to begin to enable disaster justice, instead of expecting a particular affected population to perform resilience as an essentialized reading might assume.

Another useful discussion to engage with here is “crisis-driven urbanization” and its arguably disastrous effects. Crisis-driven urbanization, as Gotham and Greenberg (2014: xi) argue, “refers to the general approach employed by urban growth coalitions, aided by government and private interests at local, state, and federal levels, to interpret and use ‘crises’ to advance and legitimize radical policy reforms and redevelopment projects that would be far more difficult to implement in normal times.” In the name of recovery, post-disaster urban redevelopment projects are often more easily justified without enough attention paid to the likely processes of crisis-driven urbanization as intensifiers of urban instability, vulnerability, and hazard risk. In the case of post-disaster recovery of the Kucapungane and Makazayazaya tribes in Taiwan, it may be fair to say that their vulnerability to natural disaster makes it easier for the state-led relocation/urbanization of the indigenous communities to advance, which would be far more difficult to implement in normal times. In bringing attention to these crisis-driven spatial processes, this research intends to illustrate how the state takes disasters out of the historical context of crisis-driven urbanization and the uneven redevelopment as a result, without taking into account how historical cycles of urbanization may have profound implications in terms of producing vulnerable situations before the disastrous event.

Recovery policy after Typhoon Morakot

Typhoon Morakot, a Category 2 typhoon, arrived in Taiwan on 7 August 2009. The heavy precipitation brought by Morakot—2500 mm of precipitation in certain regions, equivalent to half a year's rainfall on the island in three days—was unprecedented and caused serious landslides in Southern Taiwan. It resulted in 681 deaths and 18 people missing. More than 10,000 people were temporarily or permanently displaced. Morakot also struck other states in the region, most notably China's southern provinces, the Philippines, and Japan. The total cost of damage caused by Morakot to infrastructure and agriculture was approximately US$912 million.

To facilitate recovery, Taiwan's cabinet approved a USD$2.9 billion special budget and formed the Morakot Post-Disaster Reconstruction Council (MPDR Council) on 15 August 2009, pursuant to the Disaster Prevention and Protection Act. The Morakot Post-Disaster Reconstruction Special Act (MPDRS Act) was enacted on 28 August 2009, based on a decree by the president, to be effective for five years (29 August 2009 to 29 August 2014). Later, more than USD$1.1 billion was added to the recovery fund. In addition to the public fund, civilian donations were crucial in establishing resettlement housing for those who were displaced. A total of nearly USD$722 million was collected from the Tzu Chi Foundation, the Red Cross Society, World Vision, and other donors. The MPDRS Act allowed for recovery work and also provided a legal foundation for designating “hazard areas” and thereby necessary relocation actions, such as limiting land use or forcing publicly funded relocation, which has controversially caused displacement—the publicly funded relocation brought together different tribal communities in 32 collective resettlement sites in the interest of land efficiency.

Taken together, the principles for recovery planning were set up within a very short time despite continuing discussions that reflected different opinions. The principles for siting the resettlement projects in recovery planning were laid out as a preference hierarchy: (1) moving away from disasters but staying within their original villages; (2) moving away from the original villages but staying within their townships; and (3) moving collectively to a location that is most adequately distanced from their original township. Thus, upon these principles many tribal villages were asked to relocate and co-reside with other villages. What underlay this recovery planning was a strong belief by the government that it could draw a clear line between dangerous and safe areas.

When it comes to the construction and management of the “permanent houses” provided in resettlements, the fact that the displaced only “own” the houses under strict conditions was rarely made clear to the public. As “recipients” who were granted the right to live in the houses, they cannot sell, let, or sublease the houses to others. They also do not have an entitlement to lands. Therefore, they cannot deal with the houses as properties or investments, such as securing a housing equity loan. While these limitations attached to the “free permanent houses” may appear reasonable to the public, the recipients are mostly disappointed because they lost their right to stay in their old homes as soon as they received the free gift. Considering the lands and properties they gave up, the gifts are not really free.

In general, the recovery under MPDR Council was considered an efficient and successful experience by the state and the society. Earlier, former president Ying-Jeou Ma once left this notable comment after spending a night at Rinari, one of the resettlements: “This is great! It's almost like the Provence in France!” (8 August 2011). A number of Taiwanese followed Ma's words and were attracted to the resettlement site. It is not uncommon to hear comments expressing how lucky the indigenous communities were in receiving such nice houses for free. For instance, one visitor said, “This is much nicer than my house in the city” (fieldnote, 2 December 2011). This type of reaction reflects how the disaster and post-disaster recovery were misunderstood by the public in Taiwan. Other comments questioned the necessity for some of the recipients to obtain new houses while their old homes appeared intact: “It's like doubling one's property! It's unfair. Why can the village get rehoused when their old homes remain in place?” (fieldnote, 18 December 2014). It is worth noting that some affected communities shared this view, which is, nevertheless, far from reality. It is necessary to point out that many of these comments were apparently based on a limited understanding of disaster and therefore a misunderstanding of “disaster justice.” As such, discussion of justice and disaster is often limited to immediate state responses as well as to a limited understanding of disasters as natural episodes; such reactions often lead to distributive justice in the state's policy implementation.

Established as a short-term entity, the MPDR Council was disbanded on 8 August 2014. Nevertheless, the long-term effect left by the disaster, and arguably new issues caused by the short-sighted recovery, have not ended with the disbanding of the council. Among other goals, the central government intended to take the recovery as an opportunity to force a relocation of indigenous communities away from the mountainous areas (Interview with one of the experts working for the National Science and Technology Center for Disaster Reduction in Taiwan, 15 January 2017). The MPDR Council invited experts to evaluate the indigenous homelands and designated some of them as “hazard area” and some as “potential risky area,” which would then require the residents to accept the relocation arrangement provided by the government and the charities (Interview with the former chair of the MPDR, 13 January 2017). The policy attracted significant criticism and remains under suspension to date. Among others, Fan's (2015) examination of disaster governance after Typhoon Morakot reflects misrecognition problems and residents' lack of empowerment and exclusion from the decision making of recovery planning. There was no adequate interim housing provided to accommodate the waiting and planning process, while the displaces were crowding in the shelters because the MPDR Council thought it would be more efficient to provide “permanent housing” in one shot. The reconstruction was carried out without taking into account cultural issues either (Lin and Lin, 2016). Moreover, there was a lack of reflection on changing human–nature relations. The government believes that the dangerous areas, if left alone for enough time, will be restored to their previous form. There is little discussion about the “end of Nature,” i.e., the end of the idea of nature as a pure place untouched by human hands that has been so central to modern environmentalism (Lorimer, 2015, 2017).

How Makazayazaya and Kucapungane were brought together

Of regions affected by Typhoon Morakot, Pingtung was the most hard-hit county, especially the mountainous areas inhabited by indigenous people. As of December 2014, about 5000 people were relocated to six resettlement sites in Pingtung. Most cases involved cohabitation of more than one tribe in the relocation site (Table 1). One such site, Rinari, is the largest resettlement project. It used to be the former Maka Farm owned by the Taiwan Sugar Corporation. The Kucapungane, Makazayazaya and Tavalan communities, despite their different ethnic backgrounds and even historical rivalries, were brought together in Rinari after 25 December 2010 in the context of post-disaster recovery (see the location of the old settlements in Figure 2). According to population data (updated on February 2016), there are 589 households with 1747 residents. The largest population group is Tavalan, followed by Makazayazaya and then Kucapungane.

Table 1.

Resettlement sites in Pingtung county.

Relocation sites	Charity partner	Tribes relocated	No. of applicants (household)	No. of allocations (houses)
Rinari	World Vision	Kucapungane	186	177	483
		Makazayazaya	141	132
		Tavalan	183	174
Changchi	Tzuchi	6 tribes in all	408	254
Ulaluz	Red Cross	Taiwu	197	117
Shinglaiyi	Red Cross	4 tribes in all		239
Old Kausi	World Vision	Kausi	45	22
Central Rd.	World Vision	Central Rd.	51	31
			1624	1146

Figure 2.

The map of the old indigenous settlements and Rinari, Pingtung. The old settlements of the indigenous communities and the resettlement Rinari where they were relocated to from December 2010 onwards. The numbers in the brackets represent the time period of settlement history.

It is important to note that the Makazayazaya and the Tavalan belong to Paiwan People, the most important indigenous ethnic group in the south in terms of population (98,606 people, the fourth largest indigenous group in Taiwan as of September 2016). Due to the fact that the incumbent president is allegedly a descendant of a Paiwan great grandmother, the Paiwan people are considered more prominent than ever in this island country, which has 16 officially recognized indigenous groups. Kucaungang belongs to Rukai People, which has only 13,081 people, the 10th largest indigenous group (as of September 2016). The population discrepancy has political implications. In general, Paiwan tribes are more powerful than Rukai tribes within local politics, which is believed to have affected the planning of Rinari Resettlement and its development in the following years.

Spatializing the production of vulnerability over the past century

In this section, following the phases shown in Figure 1, I illustrate how vulnerability was produced historically, and how it exposed the indigenous tribes to unsafe conditions before Typhoon Morakot triggered the disaster. The vulnerability is twofold. First, the constant displacement of indigenous communities from the deep mountains toward the floodplain exposed them to flooding. Second, the indigenous communities have been forcibly urbanized/modernized as they went through the loss of traditional livelihood and self-governance. In a way, they suffer from a localized “double exposure” to environmental changes and the globalization process that Taiwan was forced through since the late 19th century.

Root causes

With the Dutch arrival in Taiwan, the earliest documentation of the settlement of indigenous people throughout the island dates to the 17th century (Mabuchi, 2014). Migration took place occasionally, mainly due to war or population growth, but the movement mostly remained within their traditional territories. For example, Rukai tribes mostly remained within Rukai Proper, the region west of the Central Range, especially upstream of I-Liou River. In the 20th century, unlike the administrations of the Dutch and the Qing, who concentrated their colonization of Taiwan on the coastal area, the Japanese adopted a tighter rule, considering it politically dangerous for indigenous comunities to stay behind, especially after the Wushe Incident in 1930, which sent a warning of rebellion to the colonial government and even to the central administration in Tokyo. The indigenous communities were encouraged or forced to move out of the mountains by a series of policy incentives. Even if they managed to stay in the relatively hilly areas to maintain their traditional livelihood practices, new settlements were often formed by merging several tribes into one so that the colonial government could manage their control more easily. Some were encouraged to relocate into a whole new settlement on reclamation grounds, such as was the case with Sanhe Village (formed on a floodplain downstream of the I-Liou River), which attracted lower class populations from Old Majia and Kucapungane in the 1930s. Before Sanhe was fully occupied, the Pacific War intervened and the policy was suspended (see where Sanhe is in Figure 2).

Dynamic pressures

After the Second World War, the postwar nationalist Kuomingtang (KMT) administration continued the resettlement policy. Throughout the 1950s, a series of programs were promulgated to relocate indigenous people to the outskirts of cities in the name of improving their quality of life (Kuan, 2014). Indigenous people were also encouraged to move outside of their tribal society and to give up their migratory farming practices. Sanhe Resettlement gradually became a popular destination, in the 1960s and 1970s, for those who would give up their traditional lifestyles. Over time, three sub-settlements were formed along the river, with the north and the central one occupied by Paiwan people and the south occupied by Rukai peoples.¹ In 1967, Sanhe Resettlement officially became Sanhe Village under Majia Township. The fact that Sanhe is very close to where Rinari is situated—a contingent geographical proximity—interestingly connects Sanhe with disaster, which I discuss later.

At about the same time, most old indigenous settlements witnessed serious out-migration under the government's policy intervention and as a result of rapid urban expansion of the coastal cities in the postwar years. Both Old Majia and Kucapungane experienced great loss of population in the 1960s and 1970s. While Old Majia managed to retain its presence until 2009, the Kucapungane tribe in 1978–1979 decided to resettle into a new place on the floodplain (named “Hoacha,” which literally means “good tea” and is imposed by the postwar government or sometimes named “New Hoacha” as compared to the old settlement) between midstream and downstream of I-Liou River. In between Kucapungane and Hoacha, a distance of 10 km, no motorized transportation is available even now. The planned resettlement (one of the very last cases in Taiwan) was facilitated by the state, which secured the lands and offered some cash incentives for building costs. At that time, some senior tribal members strongly resisted the resettlement proposal because of legends that places near the section of the river used to be an ancient war field; this was deemed dangerous due to hovering ghosts (interview with M, June 2010). Some elders recalled memories of flooding in the 1920s and 1930s (Chen, 2016). Despite these concerns, the younger generation at the time dominated the discussion and eventually pursuaded most of the community members to move. The improved access to modern amenities promised by relocation, such as education, medical care, and transportation, did not stop the out-migration but in fact accelerated the trend. A large amount of out-migration beginning in the 1950s did not slow down after the planned relocation, as the issue of insufficient job opportunities remained (see Figure 3).

Figure 3.

A sectional view of how Kucapungane people were moved toward and away from risk.

Unsafe conditions

From the 1960s to the 1980s, indigenous people considered their “village” a place to be left behind. In other words, there had been an identity crisis since most of the tribe members considered individual self-help more practical than addressing the crisis as a collective. In the case of Kucapungane, some families did not even join the planned relocation to (New) Hoacha, including some of the noble families and one of the two chief families. Instead, they either moved to other tribes through kinship networks or settled down into cities in the region, such as Pingtung City or Kaohsiung City. In the early 1990s, the elementary school at Kucapungane was shut down due to population loss. While the community managed to keep a sense of solidarity through religion and shared ritual practices associated with their traditional governance, it became more and more difficult when even the chief family itself moved to the cities for wage labor, which created a political vacuum. Throughout the 1990s, Kucapungane went through a long term, slow community collapse, and a gradual replacement of traditional governance with a new development coalition. On the other hand, small-scale flooding during the summer time had worried the community; the Rukai people had not learned to live by the water over the past centuries. In effect, their historical connection with the mountainous environment was inscribed into the tribe's name, Kucapungane: a cold and beautiful place (Taiban, 2006).

It was not until the mid-1990s that the indigenous people were motivated to express their rights to traditional territories along with the gradual democratization of Taiwan. One by one, following the pioneers including Auvinni Kaluganes and Taiban Sasala, the younger generation of the tribe slowly regained their identity with the old settlement Kucapungane, although this has not reached enough momentum for them to return to the old settlement other than short-term visits. The serious out-migration had handicapped the long-existing self-governance of the tribe. Instead, the emergence of a group of new elites through local elections, a faith-based organization, and mobilization through education has replaced the traditional political system. Conflict between the traditional leadership and the new elites, which was only suspended during the few years following the disaster, has remained a serious issue and reemerged as a more serious issue since 2015 as more and more return-migration has occurred.

Old Majia village, originally called Makazayazaya, was located in the northeast of Pingtung County. Compared to Kucapungane, it had remained relatively intact and did not experience collective resettlement as much as Kucapungane. During the colonial period, two smaller Paiwan tribes (Vacuun and Palur) were incorporated into Makazayazaya. Though Makazayazaya was able to stay at its old settlement with basic amenities and infrastructure, it was not totally free from the social vulnerabilities that Kucapungane experienced. In effect, starting before the war, there were already people who had moved after the Sanhe Resettlement scheme. After the war, more and more community members joined the individual, voluntary out-migration to Sanhe Village or the urban center of Pingtung. Its social crisis might be smaller than Kucapungane in terms of its degree and complexity, but it still presents serious challenges to the community. Natural environmental changes also brought about new risk. Probably due to deforestation and inadequate development of infrastructure, today Old Majia stands on unstable soil and is allegedly threatened by falling rocks during typhoon season. The population's fear was evidenced by a landslide caused by Typhoon Morakot, which became a catalyst for the Makazayazaya tribe to consider their migration to Rinari in order to improve their security.

Disaster as trigger event

It is not an easy task to identify one single factor leading to the unsafe conditions before the disaster happened in 2009. Many people had already forgotten that there were disasters before the most recent one. Among others, Typhoon Herb in 1996 brought about floods and mudslides, causing four deaths. On 12 August 2007, Typhoon Sepat caused a serious mudslide that covered a dozen houses in (New) Hoacha (Interview with the community member, 3 September 2010). The community was forced to leave (New) Hoacha and find shelter in a dilapidated vacant military barracks in rural Pingtung. There were no deaths, but more than a 100 houses were seriously damaged. The disaster brought the tribe into a miserable condition without a clear roadmap to follow. In effect, the level of the I-Liou River bed had increased by 30–40 m over the past four decades due to growing deposits from a serious landslide in the upstream area; the river has expanded in width also (Chen, 2016). Between 2007 and 2009, the community had gradually reached a consensus regarding another collective relocation to Majia Farm. The Pingtung Country Government was responsible for negotiating with the landlord of the then-abandoned farm, Taiwan Sugar Corporation, to obtain the land for the resettlement. Yet the fact that Taiwan Sugar Corporation took over the land from the colonial government, which allegedly took the property forcibly from the Makazayazaya tribe, complicated these land matters. The Makazayazaya had always considered the farmland a part of their traditional territories and had protested against the siting; the government carried out the emergent planning of resettlement anyway. Meanwhile, Kucapungane was faced with internal conflicts while staying in the dilapidated camp. The poor quality of life in transition also brought more stress to the conversation over resettlement. The National Taiwan University Building and Planning Research Foundation (NTUBPRF) was invited to draft a plan for the relocation, given its long-term trustful relationship with the community.

Another disaster after a disaster

Unfortunately, Typhoon Morakot arrived, and soon the post-disaster political climate brought the previous resettlement plan to ground zero as Majia Farm was reconsidered as a potential site for more tribes rather than only for Kucapungane. After a three-week long negotiation under the direction of the MPDR, three tribes decided to settle on the Majia Farm, now referred to as Rinari. The new plan in a way addressed the Makazayazaya's current needs and responded to their historical claims to the Majia Farm, which was taken away by force during the Japanese-occupied era and continued to be properties by the Taiwan Sugar Corporation as mentioned. In this case, either Makazayazaya or Kucapungane could not but share the Majia Farm with others. The density would be tripled, and therefore some public facilities laid out by NTUBPRF's previous plan, such as farmlands and a graveyard, would be compromised. For a short period, the Kucapungane tribe was considering moving to Chang-Jhih (another resettlement site) so that they could at least stay with other Rukai tribes. But to stay slightly closer to Kucapungane, the community decided to stay and joined two other Paiwan tribes, Makazayazaya and Tavalan, at Rinari—which in the Paiwan language means “going there together.” The other two tribes were also faced with debates, but eventually the three tribes settled with the new plan. The compromise that the tribes agreed to make, nevertheless, had created hidden challenges for the tribal members to deal with.

Opportunities or vulnerabilities?

Despite so many failures and frustration, nevertheless, it is not unusual to see bottom-up initiatives that try to address those problems. From these small moves arise conflicts and opportunities. As Tierney (2007: 512) noted, “Disasters generate conflict in part because they open windows of opportunity that competing interests can exploit for their advantage.” In the post-Morakot context, the disaster-driven conflicts manifest spatially in the complicated movements between old settlements and new settlements (some “old settlements” were once “new,” if taking into account the long-term displacement described earlier). With these movements, there are also multiple processes of remaking place identity which, somewhat ironically, may contribute to more movements rather than attaching people to one single settlement. It is worth noting how these post-disaster responses may paradoxically contribute to both resilience and vulnerability in the long term.

Given the long-term out-migration discussed above, several settlements had been formed into a loose network of places, including Rinari, Sanhe, (New) Hoacha, Old Majia, Old Tavalan, and others (see Figure 2). The construction of Rinari from the former Majia Farm, moreover, has brought about more movements among these places, and even attracted some return migration from the cities. The movements, long or short, were often made among the settlements where the tribes had established a relationship. Among others, Sanhe village continues to play a role in the communities' bottom-up responses to the multiple disasters, while Rinari has become a new settlement where welfare resources and media attention have been concentrated. Bottom-up responses are exemplified through self-organized strategies to access farmland through kinship network or migratory farming, return-migration to help and to leverage on recovery effort, and preservation, which I discuss in the following.

Given its ethnic connection with both the Rukai and Paiwan tribes, Sanhe village has become an important node that facilitates interchange among migrant settlements (inter- and intra-tribal economic and cultural capital exchange) to facilitate the communities' bottom-up responses to disasters, for example, to address the issue of lacking farmlands at Rinari. In comparison with Rinari, people in Sanhe have coped with living with others for decades and have established a greater cross-tribe cooperation network. In general, the villagers in Sanhe have accumulated more social capital associated with the outside world than their relatives staying behind. Besides, there are farmlands and building plots available for sale or rental in Sanhe, especially after the first-generation migrant farmers retired and the second generation mostly abandoned farming. Some of them have worked as wage laborers in cities or as civil servants, policemen, or school teachers, if they have attained better education. Thus, the vacant farmland became the most available source of land that the displaced can find through either buying or renting. According to Ms. S and Mr. K, both of whom are villagers of Sanhe in association with Makazayazaya, and Mr. L, who is a Kucapungane villager, this arrangement was mostly made through a kinship network (interview, 20 March 2016). In particular, people from Kucapungane have been looking for farmlands in Sanhe through a variety of arrangements. Some allegedly have “borrowed” lands from relatives for seasonal farming. Some have gone a bit further, buying up parcels as a group and planning to build up multiple-family housing for their children's future needs.

To most Makazayazaya community members, moving to Rinari is considered as a village expansion rather than migration or abandoning their homelands (Hu, 2013). In other words, they have expanded their settlement to include both the Old Majia village (where the Makazayazaya community members also identified as Makazayazaya in the past, see Figure 1) and Rinari. In many ways, they sustained their connection to Old Majia. Unlike Kucapungane, who found their ancient farmlands too far away to access after the relocation, Makazayazaya community members still manage their lands there at Old Majia. Elders are used to cultivating their previous farmlands up in the mountains, and young people may offer help during their daily free time or on the weekend. In a way, they managed their responses to the disaster on a daily basis through “migratory farming” as their responses to relocation driven by disaster. Therefore, they are less likely to buy lands at Sanhe as Kucapungane people are. But those who carried out migratory farming would need to figure out effective ways of motorized transportation, which would sometimes require intrafamily or interfamily coordination (fieldnote, May 2016–January 2017).

Meanwhile, there are also people who partially “settle” into Rinari from other places even though they are not supposed to do so according to the resettlement policy. The past two years (2014–2016) witnessed a slow wave of “return migration” from the cities to Rinari. This is not return migration in a traditional sense, as Rinari itself is a whole new resettlement as a result of recovery planning. Some people had already left the old settlements of Makazayazaya and Hoacha for cities from the 1950s to the 1990s. Yet some younger people had been attracted to Rinari for several reasons: connection felt with their relatives and wish to help their communities as they went through difficulties; the concentration of policy incentives after the disaster; and the way in which Rinari has gradually become a destination for “disaster tourism” (Robbie, 2008). During my last two visits to Rinari in 2016, a variety of small businesses were set up by these young returnees, such as souvenir shops, hair salons, eateries, and visual art and photography studios. To be more precise, these individuals are not necessarily “returnees” because some of them only work in Rinari but maintain their residency in nearby villages, such as Sanhe, or in the city center of Pingtung.

The historically complicated, frequent moves between places have changed the traditional tribal politics and to some degree made the communities more politically and socially vulnerable. Due to their continuously changing population, the communities have had difficulties building up the legitimacy of one single leadership. There has been a competition between local politics and traditional inherited leadership, which had already been complicated by long-term out-migration since the 1950s and 1960s. The inadequate relocation and therefore involuntary modernization of the tribal societies brought about out-migration of young people, including those from the chiefs' families, which might have led to the absence of leaders and destabilization of inherited leadership. The collective relocation has witnessed a curious increase in return migration. According to the household registration data from 2009, there were 377 people registered at the village on the eve of moving to Rinari, which was only one-third of the population at prewar Kucapungane (the real population who resided in Hoacha was probably fewer than 300 residents). Today Kucapungane contains 495 people at Rinari (as of July 2014), and the competition had been severe after the relocation in 2010.

The frequent moves between places have also brought up new challenges to the making of a shared place-identity in the context of recovery. The aforementioned processes of long-term migration and recent return migration contribute to this highly political process that challenges both tribes. As mentioned, Majia village has been composed of three tribes—Makazayazaya, Vacuun, and Palur—since the colonial era. While Makazayazaya remains the most significant tribe that governs Majia at Rinari Resettlement, the other two tribes, though they share the same ethnicity of Paiwan with Makazayazaya, still identify themselves as people from Vacuun and Palur. Interestingly, their claims to place identity have increased rather than decreased after their collective relocation to Rinari. Vacuun has identified a place as Tao-Hua yuan (literally meaning “the heaven” in Mandarine), cultivated farmlands around, and opened itself to tourists to promote eco-tourism with a focus on a distinct Vacuun identity. Furthermore, the Vacuun people organize tours and rituals that bring themselves to revisit the historical migration routes. These actions have caused uneasiness within the larger community of Makazayazaya, contributing to emergent identity politics from within (interview with Lily, 30 July 2016).

In Kucapungane, the remaking of place identity manifests itself in a belated preservation as an adaptation strategy (Huang, 2018). Moreover, along with the upcoming state-funded restoration project to repair the damaged structures by long abandonment, some people are aspiring for the next relocation, which may bring them away from Rinari—as a less-than-ideal resettlement only for now (fieldnote, 30 July 2016). The object of the ongoing state-led preservation, operated based on the Act of Cultural Heritage and supervised and funded by Ministry of Culture, is the old settlement of Kucapungane (Old Kucapungane), which has been listed as a national heritage and thus considered as an important piece of shared cultural properties by the nation since 1991—it was actually the very first indigenous case to be inscribed to the list of national heritage. The inscription of the Kucapungane case resonated with the changing national identity in the 1990s to indigenous/local culture away from a more mainland-centered one (Chiang et al., 2017). The settlement comprises more than 160 stone-slab houses, ritual spaces, and farmlands extending to a broader traditional common. Kucapungane's old settlement has remained as a significant cultural reference, a spiritual homeland to the tribal community even after the relocations over the years, which is not exactly the same as how the state conceives it as a shared cultural property by the national citizens. The difference would have had different implications for implementing preservation but there had never been enough momentum for the community or the government to carry out a large-scale restoration as required by the Act of Cultural Heritage. Over the past three decades, there were some attempts at restoration and adaptive reuse, but it was never realized until 2015 (Huang, 2018). Ever since 2007, the Kucapungane tribe has been preoccupied by the post-disaster recovery, leaving no energy to talk about restoration. The disasters, ironically, became a justification for the government to keep postponing the restoration until the World Monument Fund (WMF) Watchlist brought a renewed attention to the historical settlement of Kucapungane. Against this context of uncertainty and conflicts, Kucapungane's old settlement reappeared as a precious place of hope.

Several challenges remain to be addressed before crises can be turned into opportunities. First, the pre-disaster heritage conservation plan requires significant updates, as it certainly does not take into account “disaster” as a recurring issue in the region. Second, the rapidly changing tribal politics in relation to demographic change is a key issue to address in ensuring that historic preservation becomes beneficial rather than making the community more vulnerable to natural hazards. At the moment, the tribe is debating whose houses will be prioritized for fully subsidized restoration. With a community political system endangered for several decades, there is a desperate need to rebuild a mechanism that would enable a consideration of the community. In the course of remapping community boundaries, heritage preservation has become a contested terrain more than ever. Among other issues, it may be particularly important to evaluate the risk of bringing people back to disaster areas, which may contribute to future disaster. While Kucapungane's old settlement is believed to be located in a safe place right below the range, the several roads leading to Kucapungane are exposed to varying degrees of risk, given unstable geological conditions in the past decade caused by extreme environmental change. For example, the road that passes Hoacha—one that the community is most familiar with—is not walkable during monsoon season (from May to August). The heritage preservation that would bring people back to Kucapungane's old settlement, including both residents and tourists, may at the same time become a risk to manage.

The aforementioned bottom-up responses are actions that the indigenous communities have taken to address issues resulting from a problematic recovery planning. In general, the communities would have liked the resettlement to be located closer to their ancestral lands so they could better maintain their place attachment and access to natural resources. In terms of decision making process, they would have liked to have more participation in selecting the site and designing the houses. Besides, the fact that there was no adequate interim housing provided to facilitate a better process of participation may be a significant factor that rushed the planning of resettlement, which left a lot of internal and external conflicts unresolved. Moreover, the limited options offered for housing different needs and preferences left the communities with few alternatives. The tribal communities were asked to reach a consensus within a relatively short time. Yet the complicated social relations as one of the existing unsafe conditions (as indicated in Figure 1), including competition for leadership among influential families and modern governance structure, and generational differences in perceiving relocation, etc. made it too challenging to make a collective decision as efficient as the government expected. There was no friendly policy environment with flexibility or sensitivity to indigenous culture to facilitate a genuine participation in discussing collective relocation. The people-centered approach to housing recovery in Indonesia after the 2013 eruption of Mount Merapi (Maly, 2018) may be a case to look after, especially the flexibility it offers to the displaced people in building up a variety of houses to meet their different needs and the possibility for them to reutilize their old lands.

Conclusion: Long way toward disaster justice

In this article, I reviewed the long-term production of vulnerability that foregrounded the disaster triggered by Typhoon Morakot. I also presented the way in which the communities have weathered the disastrous outcome arising from state-led post-disaster recovery over the years. Those bottom-up initiatives have played out spatially in the process of communities simultaneously moving out and back to the different settlements within the region. With these observations I argue that adopting a spatialized analysis of disaster (in)justice allows us to understand “disaster” as events occurring in a long-term unjust distribution of outcomes resulting from socioecological crisis, which unfortunately continues to escape social and political accountability. The problematic status quo continues to place blame on indigenous communities for being unable to keep up with scientific disaster prevention and incapable of organizing themselves into a tight community. There is an urgent need to re-articulate social and political responsibilities with an understanding of Anthropocene disorder.

To conclude, I would like to bring attention to three issues in unpacking the long-term production of the uneven landscape of risk and vulnerability that broadly influenced the indigenous people in Taiwan. First, the historical state-led modernization, arguably colonization, has contributed to long-term out-migration and challenges caused by return migration. Over the past century, the urbanization and modernization of indigenous communities has brought them away from where they could access resources for their traditional subsistence strategies without causing the Anthropocene disorder. The effect of the multiple displacements over time, ironically, gives legitimacy to preserving tradition and heritage when indigenous communities are displaced from their traditional territories. Second, the normalization of fixity and sedentary lifestyles in governance, which can be performed in identifying one single settlement associated with household registration and replacing the dynamic cross-tribe negotiation of traditional territories with modern jurisdiction, deprived the indigenous communities of self-governance and land management, with which they adjusted to environmental changes in the past. It caused loss of traditional knowledge and, worse still, loss of the opportunity for them to update their relationship with the changing environment. Even though some populations managed to maintain multiple settlements and a level of migratory farming, their movements and activities are however largely restricted by household registration, modern land use, and property law. The application of resilience in planning as “change management exercise” (Wilkinson et al., 2010: 37) from the indigenous people's perspective rather than the resilience doctrine that serves the state's interest seem to remains a remote goal for us to achieve in recovery planning in Taiwan.

Last but no less important, the degree of modernization of indigenous communities translated into the atomization of the community into taxpayers, election voters, and private property owners, which indirectly threatened the traditional sense of collectivity and leadership. All of them contributed to community disintegration and irregular movements after the disaster, which became more obvious after the recovery officially ended in 2014. As Evans and Reid pointed out, the resilience doctrine has allowed neoliberalism to appropriate the terms of vulnerability and to absorb the critique of sustainable development in order to naturalize the existing inequalities as inevitable fait accompli. Moreover, the resilient subject becomes “the surest embodiment of neoliberal thinking as it conforms to its guiding principles without questioning the political stakes to vulnerability” (Evans and Reid, 2015a: 36–37). The case examined here to some degree become the kind of resilient subject produced by the recovery planning. Meanwhile, the post-disaster period witnessed an emergent interest in remapping community boundaries and, in relation to this, place-identities. While some communities like Makazayazaya grabbed the opportunity to reclaim their historical entitlement to the territory (of Majia Farm), people in Kucapungane continued to experience loss of access to their ancestral territories—they don't even know where to bury the dead now. The belated heritage preservation appears as a timely opportunity but its current disconnection with the post-disaster recovery planning demands policy revision. To have a spatialized understanding of vulnerabilities for disaster justice can be effective strategy for us to expand our conceptualization of risk and disasters to include the long-term global assemblages from where the Anthropocene's changes arise. If we fail to rework our understanding of disaster and continue to regard all conflicts and responses as irrelevant to disaster research, we may reinforce the existing practices of disaster injustice.

Footnotes

Acknowledgements

I thank the three anonymous reviewers very much for their critical comments that help improve the draft. The researcher is grateful to the tribal communities who have generously shared their experiences of disasters. Gratitude is also extended to colleagues from the National Taiwan University Building and Planning Research Foundation, who offered help generously. Part of the data was collected through the University of Illinois Urbana-Champaign (UIUC)–National Taiwan University joint planning studio that I co-instructed with Robert Olshansky at UIUC from 20 March to 2 June 2016. I appreciate all the information and insights shared by the studio participants. Part of earlier versions of this paper was presented at the 2016 Symposium “Disaster Justice in Anthropocene Asia and the Pacific,” the Asia Research Institute, National University of Singapore. I am grateful to the organizers Mike Douglass and Michelle Ann Miller for their effort of extending the conversation into collaborating on a special issue.

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This research is supported by the research grant provided by Ministry of Science and Technology of Taiwan (106-2420-H-002-012-MY3).

Note

References

Aikens

Gyoung-AH

Lee

(2013) Post glacial inception and growth of Anthropogenic landscapes in China, Korea, Japan, and the Russian far east. Anthropocene 4: 46–56.

Bankoff

(2003) Constructing vulnerability: The historical, natural and social generation of flooding in metropolitan Manila. Disasters 27(3): 224–238.

Blaikie

Cannon

Davis

et al. (2006 [1994]) At Risk: Natural Hazards, People's Vulnerability, and Disasters, New York: Routledge.

Bouwer

(2013) Projections of future extreme weather losses under changes in climate and exposure. Risk Analysis 33(5): 915–930.

Burton

Cutter

(2008) Levee failures and social vulnerability in the Sacramento-San Joaquin Delta area, California. Natural Hazards Review 9(3): 136–149.

Cameron

(2012) Securing indigenous politics: A critique of the vulnerability and adaptation, approach to the human dimensions of climate change in the Canadian Arctic. Global Environmental Change 22: 103–114.

Castree

Adams

Barry

et al. (2014) Changing the intellectual climate. Nature Climate Change 4: 763–768.

Chiang

M-C

Huang

S-M

et al. (2017) Policy formation and civil society engagement in heritage-making in Taiwan: A historical examination. In: Hsiao

S-M

Foong

Peycam

(eds) Citizens, Civil Society and Heritage-making in Asia ISEAS, Singapore: Yusof Ishak Institute/IIAS/Institute of Sociology, Academia Sinica, pp. 232–250.

Chen

C-Y

(2016) The place that water keeps coming back. Journal of Soil and Water Conservation 14: 35–37.

10.

Chen TS and Su S-C (陳丁祥, 蘇淑娟) (2004) 國家發展大計下的原住民移住聚落: 屏東縣三和村拼湊的地區內涵 (The creation of a new aboriginal spatial mosaic−A study of the aboriginal re-settlement at San-Ho Village, Ping-Tung County). Jounal of Geographical Science 37: 99–122. [Chinese only].

11.

Clark

(2015) Ecocriticism on the Edge: The Anthropocene as a Threshold Concept, London: Bloomsbury Publishing.

12.

Crutzen

Stoermer

(2000) The ‘Anthropocene’. Global Change Newsletter 41: 17–18.

13.

Dynes

(2000) The dialogue between Voltaire and Rousseau on the Lisbon Earthquake: The emergence of a social science view. International Journal of Mass Emergencies and Disasters 18(1): 97–115.

14.

Escobar

(2008) Territories of Difference: Place, Movements, Life, Durham, NC: Duke University Press.

15.

Evans

Julian

(2015a) Resilient Life: The Art of Living Dangerously, Cambridge: Polity.

16.

Evans

Julian

(2015b) Exhausted by resilience: Response to the commentaries. Resilience 3(2): 154–159.

17.

Fan

M-F

(2015) Disaster governance and community resilience: Reflections on Typhoon Morakot in Taiwan. Journal of Environmental Planning and Management 58(1): 24–38.

18.

Fortun K, Knowles S, Felipe Murillo L, et al. (2016) Researching disaster from an STS perspective. In: Felt U, Fouché R, Miller CA, et al. (eds) Handbook of Science and Technology Studies. Cambridge, MA: MIT Press. pp.1003–1028.

19.

Fuller

Jacob

Manning

et al. (2011) The contribution of rice agriculture and livestock pastoralism to prehistoric methane levels: An Archaeological assessment. Holocene 21(5): 743–759.

20.

Gotham

Greenberg

(2014) Crisis Cities: Disaster and Redevelopment in New York and New Orleans, Oxford: Oxford University Press.

21.

Hu C-H (胡哲豪) (2013) 禮納里部落災後重建過程中的原住民知識與文化詮釋 (The indigenous knowledge and the interpretation of culture in the reconstruction processes in Rinari). Master thesis, National Taiwan University, Taipei City, Taiwan. [Chinese only].

22.

Huang S-M (2018) Heritage and Postdisaster Recovery: Indigenous Community Resilience. Natural Hazards Review 19(4). DOI: org/10.1061/(ASCE)NH.1527-6996.0000308.

23.

Hudson

(2014) Placing Asia in the anthropocene: Histories, vulnerabilities, responses. The Journal of Asian Studies 73(4): 941–962.

24.

Kuan D-W (官大偉) (2014) 空間秩序, 地理再現與生態政治: 臺灣山地資源利用／保育的歷史地理回顧 (Spatial order, geographical representation and political ecology: A geo-historical review of the resource exploitation/conservation in the mountain indigenous area of Taiwan). Taiwan Journal of Indigenous Studies 7(1): 159–197. [Chinese only].

25.

Lau

Kim

(2006) Asian summer monsoon anomalies induced by aerosol direct forcing: The role of the Tibetan Plateau. Climate Dynamics 26(7–8): 855–864.

26.

Leichenko

O'Brien

(2008) EnvironmentalChange and Globalization: Double Exposures, New York, NY: Oxford University Press.

27.

Lewis

WM Jr

Wurtsbaugh

Paerl

(2011) Rationale for control of anthropogenic nitrogen and phosphorus to reduce eutrophication of inland waters. Environmental Science & Technology 45(24): 10300–10305.

28.

Lin

(2016) Cultural issues in post-disaster reconstruction: The case of Typhoon Morakot in Taiwan. Disasters 40(4): 668–692.

29.

Lorimer

(2015) Wildlife in the Anthropocene: Conservation after Nature, Minneapolis, MN: University of Minnesota Press.

30.

Lorimer

(2017) The Anthropocene: A guide for the perplexed. Social Studies of Science 47(1): 117–142.

31.

Mabuchi T (馬淵東一) (2014) 台灣原住民族移動與分布 (The Migrations and Relocations of the Indigenous Communities in Taiwan) (translated by Iûⁿ Lâm-kūn). Taipei: SMC Books. [Chinese only].

32.

Maly

(2018) Building back better with people centered housing recovery. International Journal of Disaster Risk Reduction 29: 84–93.

33.

Mileti

(1999) Disasters by Design: A Reassessment of Natural Hazards in the United States, Washington, DC: Joseph Henry Press.

34.

Miller

Douglass

(2016) Disaster Governance in Urbanising Asia, Singapore: Springer.

35.

Mitchell

Brook

Lee

et al. (2013) Constraints on the late holocene Anthropogenic contribution to the atmospheric methane budget. Science 342(6161): 964–966.

36.

Morello-Frosch

Pastor

Sadd

(2001) Environmental justice and Southern California's “riskscape”: The distribution of air toxic exposures and health risks among diverse communities. Urban Affairs Review 36(4): 551–578.

37.

Morello-Frosch

Shenassa

(2006) The environmental “riskscape” and social inequality: Implications for explaining maternal and child health disparities. Environmental Health Perspectives 114(8): 1150–1153.

38.

Oakes

(2017) Catastrophic Asia: An introduction. The Journal of Asian Studies 76(2): 401–407.

39.

O'Brien

O'Keefe

Rose

et al. (2006) Climate change and disaster management. Disasters 30(1): 64–80.

40.

Ogden

Heynen

Oslender

et al. (2013) Global assemblages, resilience, and Earth stewardship in the Anthropocene. Frontiers in Ecology and the Environment 11(7): 341–347.

41.

Parthasarathy

(2015) Informality, resilience, and the political implications of disaster governance. Pacific Affairs 88(3): 551–575.

42.

Parthasarathy D (2018) Inequality, uncertainty, and vulnerability: Rethinking governance from a disaster justice perspective. Environment and Planning E: Nature and Space 1(3): 428–448.

43.

Perry

(2007) What is a disaster?. In: Rodriguez

Quarantelli

Dynes

(eds) Handbook of Disaster Research, Singapore: Springer, pp. 1–15.

44.

Perry

Quarantelli

(2005) What is a Disaster?: New Answers to Old Questions, Philadelphia, PA: Xlibris.

45.

Platt

(1999) Disasters and Democracy: The Politics of Extreme Natural Events, Sanfrancisco: Island Press.

46.

Quarantelli

(1997) Ten criteria for evaluating the management of community disasters. Disasters 21(1): 39–56.

47.

Robbie

(2008) Touring Katrina: Authentic identities and disaster tourism in New Orleans. Journal of Heritage Tourism 3(4): 257–266.

48.

Rockström

Steffen

Noone

et al. (2009) Planetary boundaries: Exploring the safe operating space for humanity. Ecology and Society 14(2): 32. Available at: https://www-jstor-org.web.bisu.edu.cn/stable/26268316 (accessed 10 March 2018).

49.

Schmidtlein

Deutsch

Piegorsch

et al. (2008) Building indexes of vulnerability: A sensitivity analysis of the social vulnerability index. Risk Analysis 28(4): 1099–1114.

50.

Taiban S (2006) The lost lily: State, sociocultural change and the decline of hunting culture in Kaochapogan, Taiwan. PhD Dissertation, University of Washington, USA.

51.

Tierney

(2007) From the margins to the mainstream? Disaster research at the crossroads. Sociology 33(1): 503–525.

52.

Tierney

Oliver-Smith

(2012) Social dimensions of disaster recovery. International Journal of Mass Emergencies and Disasters 30(2): 123–146.

53.

Tsing

(2005) Friction: An Ethnography of Global Connection, Princeton, NJ: Princeton University Press.

54.

Wilkinson C, Porter L and Colding J (2010) Metropolitan planning and resilience thinking: A practitioner's perspective. Critical Planning Summer: 25–44.

55.

Yeh

(2016) How can experience of local residents be “knowledge”? Challenges in interdisciplinary climate change research. Area 48(1): 34–40.

56.

Yeh

(2017) Commentary: On the possibilities of transdisciplinary Asian studies. The Journal of Asian Studies 76(2): 499–503.