Engaging Hashima: Memory Work, Site-Based Affects, and the Possibilities of Interruption

How is memory embodied, narrated, interrupted, and reworked? Here, we take a postphenomenological approach to memory work that is attentive to how site-based affects prompt and ossify, but also transmogrify, memory of place. With reference to an intensely traumatized, but also domesticated and entropied, environment—the island of Hashima, off the coast from Nagasaki City in Japan—we demonstrate the relevance and explanatory reach of culturally specific accounts of memory, time, and place; how an attentiveness to cultural context in the making of meaning helps mark out the epistemological violences that accrue around sites such as Hashima as objects of analysis in and of themselves; and the affective capacities of the materialities and forces that compose such sites, which can present a welter of surfaces and interiorities that are sensuously “felt” as memory.

The figurative heart of this article is the island of Hashima, which, devoid of human residents, is nevertheless a focal site for all manner of memory work from a wide range of groups and individuals both inside of and outside of Japan. The outcrop, lying around 15 km from Nagasaki City, was mined under the auspices of the Mitsubishi Corporation between 1890 and 1974, and handed to public authorities in 2002. The coal produced helped animate Japanese modernization and imperialist expansion in the early twentieth century, and the island itself became the location for new high-rise, body-dense living arrangements, as well as a holding place for Korean forced labor during World War II. These high-rise constructions led to an architectural dominance over the rocky landscape that altered the island's profile. The distinctive shape resembled a floating battleship and led it to be dubbed Gunkanjima (or Battleship Island), a nickname recorded in press reports as early as 1911 (Nagasaki Press 1911). Abandoned and left to the elements, Hashima has in recent years become a source of images of ruination within a specifically Japanese postindustrial genre, but also as part of an Anglo-American visual economy of "ruin porn" composed of abandoned buildings haunted by the remains of past lives. In this it perhaps resembles the depoliticized smokestack nostalgia criticized by figures such as (High and Lewis 2007, 9) in his concept of the industrial sublime. Yet, there is more to the formal memorialization of Hashima than this: The island has been enrolled as part of a larger industrial and political ensemble of late-nineteenth-century facilities that overcame objections from the Korean government to receive UNESCO World Heritage Status for its embodiment of the Meiji revolution (S. Kim and Reynolds 2015).
Our engagement with Hashima, although acknowledging the contentious politics of memorialization that have taken place, nevertheless did not take this formal political process as its substantive object of analysis; hence, in what follows we do not dwell on the manner in which such formal efforts at commemoration were manifest via a choreography of people and things. Rather, we were concerned with an accounting of the conditions within which the memory work that has helped prompt, sustain, legitimize, trouble, and protest such a memorialization takes place. These conditions, for us, revolve around Hashima as a place recalled-brought into being, one might say-by its former inhabitants. As such, we wanted to gain insight into how these recollections were precipitated, as Tolia-Kelly (2004) put it, through a sensuous engagement with place, and iterated as memory. In our work, we emphasized two key points. First, it is important in our academic analyses to be cautious of the reach and extent of European framings of time (and specifically regarding issues such as progress and ruination). Relatedly, an attentiveness to cultural context in the making of meaning helps draw attention to the epistemological violences that accrue around sites such as Hashima as objects of analysis in and of themselves. These violences emerge not only in relation to official demarcations of Hashima, but also in more intimate interactions between guide and visitor, witness and audience that can certainly facilitate such formal commemorative practices. Second, in paying attention to the precipitation of memory, we need to be aware of how these shapings and disruptions are allowed for by the affective capacities of locales such as Hashima. These sites are more than a repository for past decisions and activities by the subject; and they are more than a node in the distributed network of socially constructed meanings. They are also of the inhuman and, as such, can frustrate and interrupt efforts to have a place make sense.
For this project, we focused our attention on the memories associated with one particular person, Sakamoto Doutoku, who spent several years as a child on the island in the 1960s and 1970s, and who led the nonprofit heritage campaign group Gunkanjima o sekai isan ni suru kai, translated by the group as The Way to World Heritage Gunkanjima, although perhaps a better transliteration would be The Association to Make Gunkanjima a World Heritage Site. Sakamoto fronts commercial and other tours to the island, and has written and contributed to publications on the heritage value of the site (D. Sakamoto and Gotō 2009; Gunkanjima o sekai isan ni suru kai 2011; D. Sakamoto 2014; D. Sakamoto and Takagi 2014). Sakamoto's stories also permeate the volume of other material produced in recent years. As head of the campaign group, he has been the primary facilitator of interaction with the island for former residents, academic researchers, and documentary and feature filmmakers. Insofar as our own presence on site would clearly depend on the mediating role of both city officials and most prominently Sakamoto, we were provided the opportunity to critically reflect on how memory work on Hashima was undertaken in situ, and, moreover, the possibilities afforded for a reshaping of this.
In looking to memory and Hashima, we eschewed the notion that personal narratives of place were simply being upscaled to help form social discourses on heritage-a framing that hinges on the notion of the sovereign human subject as both source and director of memory. Instead, we were inspired by a postphenomenological framing of memory that dwells on this as emerging from the relations between the living subject and the physical as well as social world within which that life is lived. The post in postphenomenological thus refers to the fact that although we remained attentive to the preconscious nature of many of these relations-what some have called a more-than-representational being in the world-as well as knowledge as an embodied capacity, we nevertheless rejected the Husserlian idea that there is a universal humanness that can thereby be revealed. We looked to the contingent, dynamic role of geographically distributed, site-specific affects in prompting, but also interrupting, such memory work. Affects can certainly be considered as a felt manifestation within the emotional dispositions and tensions of the lived body; these do not thereby indicate an interiorized and pure self, however, but emerge and circulate as the specific capacities of bodies to affect one another in the form of pressures or intensities. Such bodies are by no means confined to the corporeal, insofar as the lived body is enabled by a host of material and energy transfers that we tend to sort into categories such as the organic and the elemental. What is more, and depending on their singular capacities for action, bodies are differentially able to recognize this same process, and to critically reflect on it. From a postphenomenological perspective, it is thus productive to consider memory work as an ensemble of materials and associated capacities inherently tied to, but also thoroughly open to, a complex environment that enables memory to be produced, and to be recognized as such.
In methodological terms, as we followed our guide one hot, humid day in July 2013, we were thus mindful of the manner in which bodily rhythms of walking, and the sensuous engagements this precipitates, have become learned and deployed as a means of gathering knowledge about the world, and of making sense of our own and others' place in that world. What is more, as a cross-disciplinary team composed of a geographer, earth scientist, and cultural historian, we were also sensitive while on site to the differential field work techniques in which we were trained. Collectively, we were attentive to how our own and others' bodies-at rest and in motion, listening and talking, observing and being observedfelt the environs as a prompt for critical reflection; these reflections, as we go on to note later, engendered their own affects, such as the facilitation, but also interruption, of Sakamoto's narratives of place. In the following, which recalls something of our walk that day, we dwell on the ongoing relationship among Sakamoto, Hashima, and ourselves, wherein the past is called into the present in the context of a landscape that, far from being a mere backdrop, is crucial to the prompting and shaping of an embodied memory work. We are attentive to how moving through, and with, this site prompted specific memory acts that can be manifest in anecdotes, certainly, but also in a bodily activity-or "muscular consciousness" (Cole 2013)-that allows the site to be knowable in particular ways. We also forefront how changes to this landscape, whether in the form of physical transformations or the activities of others, often frustrated such efforts. Before we detail some of our findings, however, we turn in the next section to an expanded note on our conceptual framing of the site of memory work, and then provide a contextualization of Hashima.

THE SITE OF MEMORY WORK
At the heart of our analysis lies an acknowledgment of the complex geographies of memory, and in particular the role of the site in prompting, embodying, and interrupting memory work. Memories here are not so much rounded narratives as imaginative encounters with objects that are both manifest in, and exceed, representational forms such as the anecdote (Philo 2003;Edensor 2005;Lorimer 2006). For Till (2012), "wounded cities" such as Berlin, struggling to reconstruct the notion of heritage, can thus provide for a traumatized memory work, as residents daily encounter densely settled locales that have been harmed and structured by particular histories of physical destruction, displacement, and individual and social trauma resulting from state-perpetrated violence … these forms of violence often work over a period of many years-often decades-and continue to structure current social and spatial relations. (6) Certainly in our engagement with Hashima, we have strived to attend to how this immersive being-in-the-world takes place, wherein memory work is a mode of bodily laboring that stutters and starts, fashions and refashions not only the relations among self, other, and world, but their constitution. Although the memories evoked can be understood as deeply felt, singular expressions of an uncanniness that haunts both remembering and forgetting (see Hill 2013), they are also, as Till (2012) intimated, very much shaped, disrupted, and shaped again by widely circulating, or social, denotations as to the nature of the self, the other, and the distance between the two. As Rothberg (2009) suggested, "Not strictly separable from either history or representation, memory nonetheless captures simultaneously the individual, embodied and lived side and the collective, social and constructed side of our relations to the past" (4).
To be sure, the ongoing bid to memorialize Hashima is a personalized project on behalf of Sakamoto that serves to bolster his felt experience of life on the island, but it is also very much part and parcel of a state-sponsored effort to provide an ad hoc narrativization of events that, it is asserted, helped to define modern Japan. In this, Sakamoto's efforts are akin to a range of other memory projects in the context of modern Japan. As with the field of memory studies more broadly, the emergence of which has been traced most prominently to the Holocaust and its aftermath, the founding traumas of Hiroshima and Nagasaki helped shape this project in Japan (see, e.g., Treat 1995;Yoneyama 1999;Seaton 2007). As Gluck (2003) argued, "If … the Germans remembered the Holocaust to forget the Third Reich, it may be said that Japan remembered the Bomb to forget Manchuria … [turning] victimizers into victims and atomic memory into imperial denial" (294). In recent years, however, this narrow focus on atomic memory and imperial denial has broadened to a much wider engagement with memory beyond the war, and beyond these sites of national trauma. The East Asian Lieux de Memoire project, for example, has built on the work of Pierre Nora in beginning to fashion a set of sites, objects, and ideals that together shape an "East Asian memory" (Itagaki, Chon, and Iwasaki 2011).
The legacies of Japanese imperialism play a significant role in this attempted regional construction, and yet, for sites like Hashima, centrally implicated in that same project, a continued carving out of certain uncomfortable histories from heritage campaigns endures. As Smith, an industrial heritage consultant on the Hashima UNESCO bid argued: The Pacific War was the unfortunate outcome of the Imperial designs of the Japanese armed forces, resulting in the Allied use of atomic weapons. For this reason many people chose to avoid discussing the period from the Meiji restoration onwards but there is very little to be ashamed of in the incredible pace of technological and social development which took place during the 40 years between 1868 and1908. (National Congress of Industrial Heritage, Japan n.d.) This narrative also carried through in the official bid documents, which sought to limit the scope of the bid to the years leading up to 1910. The International Council on Monuments and Sites report to the World Heritage Committee meeting in June and July of 2015 supported this temporal demarcation, arguing: After 1910, the cut-off date for this nomination, Japanese industrial development continued to grow, relying more and more on imported raw materials, but its concentrated period of technological innovation associated with the blending of western and Japanese technologies had come to an end: the Japanese industrial system was established. (International Council on Monuments and Sites 2015, 92) In contrast, the South Korean government expressed strong reservations about the bid, focusing particularly on the role of Japan's industrialization in the development of empire and its later documented exploitation of imperial subjects through processes of forced labor. For the Koreans, the Japanese industrial system might have been established by 1910, but it was sustained only through the exploitation of unpaid labor. Under threat of formal opposition, the Japanese side retreated, announcing that they were, "prepared to take measures that allow an understanding that there were a large number of Koreans and others who were brought against their will and forced to work under harsh conditions in the 1940s at some of the sites" (S. Kim and Reynolds 2015). This included plans to build an information center about the victims.
Bearing in mind this ongoing, fraught memorialization of Hashima as a UNESCO Heritage Site, but also the postphenomenological impetus outlined earlier, we would elaborate on two points. First, we need to be attentive to the relevance and explanatory reach of culturally specific accounts of time and place. As Hoskins (2015) pointed out, in his interrogation of how extractive sites are commemorated via grand narratives of environmental conquest, these heroic frames are very much a product of concrete social relations of power that marginalize the lived grind of mining. We would add that the modern ruinenlust obsession is largely framed (more often than not implicitly) as a European phenomenon, emerging as it does from the Romanticism of the eighteenth and nineteenth centuries (Zucker 1961, 119). Although this is often understood as a global universal-that is, as part of a reaction against the emergence of modernizing processes of industrialization, urbanization, and the rationalization of nature-it is important to also recall, as Wu (2012, 7-8) argued, that there are specific historical and cultural contexts for the interpretation of and engagement with these sites and representations. Wu (2012) suggested that an examination of ruins in the Chinese context requires a necessary examination of "the complex historical interactions between China and the West, which have dominated the development of Chinese society, politics and art since the nineteenth century" (8). Others have also begun a similar process of historical contextualization in Japan (Hladik 2008;Pendleton 2011).
Following on from the preceding, an attentiveness to cultural context in the making of meaning marks out the epistemological violences that accrue around sites such as Hashima as objects of analysis in and of themselves. These violences emerge not only in relation to official demarcations of Hashima as an imperialist triumph and a futuristic domesticated space, as outlined later-narratives that have sidelined Korean and Chinese experiences of the islandbut also in the more intimate interactions between guide and visitor, witness and audience. As noted earlier, the work of memory is constructed through the embodied interactions both with the site and with its interlocutors. In the context of this project, those interactions also enrolled us, a group of researchers trained in diverse U.S. and European academic traditions, functioning as participants, witnesses, or spectators in what Taylor (2003, 32) described as necessary parts of embodied performance, "part of the act of transfer" of meaning. We are placed within the frame of the embodied mnemonic performance, "implicating us in its ethics and politics" (Taylor 2003, 33). Elsewhere, Taylor (2014, 44) described her walks through the Villa Grimaldi in Santiago de Chile with a survivor of the torture that took place there. She argued that for guides at traumatic sites like this, "routine serves a mnemonic function-people can remember certain events by associating them with place. … Through the … act of walking, the body remembers." Yet, routine is also about containment, about "keeping the past alive yet under control." Interruptions to that routine can cause that mnemonic control to shift, to slip, to potentially open up the present to unexpected affect. In the case of the traumatic sites of torture that Taylor described, the work of interrupting routine runs the real risk of causing emotional harm. In our case, the outcomes are perhaps less personally charged, yet are no less historically loaded, as we see later.
Second, we emphasize that these shapings and disruptions of memory work also ensue from the affective capacities of the materialities and forces that compose sites. Such sites are more than a repository for past decisions and activities by the subject; and they are more than a node in the distributed network of socially constructed meanings. To be sure, we acknowledge the wealth of literature (e.g., Crang 1994) that demonstrates how memory work is facilitated by the choreographing of sites such as museums that, in turn, rely on the narrativizing capacities of its visitors: On Hashima, we ourselves followed such pathways. We also want to acknowledge accounts of how a sensuous engagement with the environment precipitates rememories of the self in and alongside others in place (Tolia-Kelly 2004). What we want to emphasize here, though, is that these engagements are dynamic, complex assemblages (of flesh, viscera, and bone, water, air, and stone) that allow for particular human and inhuman entities to be realized and transmogrified, and that present a welter of surfaces and interiorities that can be sensuously "felt." Such materialities and forces underpin the shifting morphologies of site, but they also express a series of temporalities, from the rapid pace of storm surge and chapped lips, for example, to the slow time of sedimentation and cellular mutation (Dixon, Hawkins, and Straughan 2012). As such, they can also be configured as time-rich accumulations of matter that perform their own ruination, via entropy, and that express a continual becoming of new configurations (Dixon, Lavery, and Hassall 2015). The human bodies that inhabit such sites are singular entities, to be sure, but they are also thoroughly and affectively entangled with a host of other, nonhuman entities via a series of shifting, asymmetric relations. As Edensor (2010) remarked, the walking body, "demolishes any sense of a distanced, romantic conception of landscape, of any visual imperialism. The lively moving body beholds not some passive inert scene but a pulsing space" (75). Here, we do not undertake a creative response to Hashima as a means of capturing something of its inhuman becomings, although we have participated in such experimentations elsewhere. Instead, we want to intimate how a critical awareness of the affective capacities of this site often paused our musings, pauses that carved their own spaces into our sense of self as interlocutors, as walkers and as human beings.

HASHIMA/GUNKANJIMA
Sakamoto's memories of Hashima stem from its postwar "Golden Period," when the island was run by Mitsubishi as a productive site that also displayed cutting-edge domestic technologies and a benevolent social order predicated on community awareness. This postwar history of social interaction has in recent years become the focus of large-scale memory work by former residents. Most prominent among these is, undoubtedly, Sakamoto, but he is by no means alone. Former residents have published collections of photography documenting life on the island (Minagawa 2013); volumes that bring together photography, written testimony, or oral history (Itō and Akui 1995;Ōhashi 2010;M. Sakamoto 2011); poetry compilations (Hyūga 2005); and even children's picture books (Murakami 1999). Almost uniformly, these collections paint a picture of the island as an engaged and highly connected community, often positioned either explicitly or implicitly as being at odds with the perceived alienation of contemporary Japanese urban life.
The family and community-oriented routines and living arrangements that Sakamoto and these authors reference are in sharp contrast with an earlier past. Bought by Mitsubishi in 1890, and ultimately excavated to a depth of over a kilometer for coal mining, Hashima was to play a vital role in Japan's industrial and political revolutions. By 1907 the small rock outcrop had doubled in size; this physical extension, formed from the mine's spoil, was leveled and made ready for the construction of dormitories for the island's workers, brought from all areas of Japan, including women as well as men. Whereas the former dealt with communal cooking and laundry, as well as the transport of coal onto the ships and other necessary support roles, the latter largely worked in shifts deep underground. The work was dangerous and life on the island was tough. During the period between 1925 and 1945, for example, 1,299 people died of a range of causes (Nagasaki Zainichi Kankokujin no jinken o mamoru kai 1986, 63-64).
Through the end of World War II, Chinese and Korean forced laborers were also sent to this concrete, labyrinth-like site. In 1944, Hashima and its neighboring island Takashima housed a recorded 1,355 Korean workers, or about 25 percent of the population. Confined to tiny rooms at the southern end of Hashima, and kept separate from Japanese residents, these laborers worked 1,000 m below sea level, where methane gas accumulated in the cramped shafts. They were underpaid (or unpaid), malnourished, and overworked. Of the 1,299 recorded deaths, 122 Koreans and 15 Chinese died in a range of accidents and of illnesses including juvenile dysentery, cerebral hemorrhages, typhoid, and other afflictions (Nagasaki Zainichi Kankokujin no jinken o mamoru kai 1986, 64, 69-71). The blast of the atomic bomb that devastated Nagasaki Prefecture in 1945 smashed windows on Hashima; afterward, both prisoners and miners were sent ashore to help in the cleanup operation. At the time of Nagasaki's atomic bomb, more than 75,000 Koreans were resident in the city, and some 10,000 were killed. The extent of Mitsubishi's exploitation of Korean and Chinese, not to mention Japanese, workers is still clouded by the corporation's long-running efforts to deny compensation and to restrict access to historical records, should they still exist in company archives (see, e.g., Underwood 2006aUnderwood , 2006b; H. S. Kim and Kil 2010). 1 For the course of World War II, as well as the postwar boom, the mine's output was to prove a prime means of production for Japan's industrial growth. In 1959, the island reached a peak population of more than 5,000, a population density of around 835 people per hectare. This necessitated the accelerated construction of a range of social infrastructure that had begun in the prewar period. High-rise reinforced apartment towers were first constructed in 1916; the Shōwakan Cinema arrived in 1927; the massive U-shaped apartment complex housing 388 apartments was built between 1945 and 1958; and the hospital and school both followed, also in 1958 (Akui 1985;O Project 2008). These locations all loom large in Sakamoto's recollections, as we see later. As coal gave way to petroleum, however, as a primary source of energy, the workforce at Hashima was quickly moved on. The mine was closed in January 1974 and the evacuation of the entire population was completed by 20 April.
The rapidity of the departure process resulted in a host of everyday and specialist objects being abandoned in homes, school classrooms, and businesses. The vast machinery of the minehead was also largely left behind. Devoid of residents and exposed to the elements, the site was made officially off limits to visitors. In the late 1980s, the site came to be popular with haikyo (ruins) tourists, who documented their interactions with the site through extensive photography, characterized by the absence of human presence that reinforced the popular image of the island as a mujintō (uninhabited island; e.g., Kobayashi 2004). After visiting what was at that point still a relatively well-preserved site, Burke-Gaffney (2002) published what was to become a widely circulated and influential essay on Hashima as a microcosm of an exploited Earth. Commercial tourism began in 2009, allowing once more for an in situ memory in the making. This corresponds with campaigns to have the site listed on the UNESCO World Heritage register, despite ongoing controversies over the occlusion of its role in Japan's imperial past (Ministry of Foreign Affairs 2013).
In recent years, Hashima has become viral in the sense that it has become iconic-alongside Chernobyl and Detroit-of an Anglo-American discourse on ruin porn that depends on sublime pictures of rotting infrastructure and abandoned objects, all of which speak to an Enlightenment preoccupation with the darkness that lurks at the heart of progress, the gradual fall of empire, and the final triumph of nature over society (High and Lewis 2007;Strangleman 2013; although see Pétursdóttir and Olsen 2014 on photography as a productive engagement with ruins). Yet, Hashima has also become iconic in the context of a specifically Japanese genealogy of ruins manifest in, for example, the extensive body of literature around postwar literary cultures that focuses on the so-called yakeato (or "burned-out ruins") generation of writers shaped by war defeat, the atomic bombings in Hiroshima and Nagasaki, and the broader impact of war and occupation.
In the context of the economic decline of the last couple of decades, what some have described as the Lost Decade(s), members of younger generations have drawn on these narratives of ruination to make sense of their experiences of struggle. Certainly, artificially preserved ruins, like the Atomic Bomb Dome in Hiroshima, take up large spaces in Japanese collective memory. These ruined memorial sites both reify their original destruction and serve as physical contrasts with the perceived peaceful and restored present. In the phenomenon known as haikyo tourism, the abandoned buildings of Japan's modern and industrial pasts became spaces for urban exploration and artistic documentation that circulate more widely through popularly consumed photography books, art exhibitions, and online galleries (e.g., Saiga 1986;Kobayashi 2004). Prominent among these is Hashima. In discussing the work of prominent haikyo photographer Saiga Yūji, curator Kasahara (2003) described what she saw as the strange fascination of how things, that were produced in order to serve humanity and that are then cast away by people and that dissociate with their time with people, begin to be liberated from the meanings, roles, and functions attached to them, and they begin to live a very separate life and in a very different time. (143)

MEMORIES OF HASHIMA
As part of our research, we had initially traveled to the island with a commercial tour, where visitors were restricted to around thirty minutes on the island, traveling along a concrete path within carefully demarcated zones, and with guides closely hovering. Our second trip provided a more expansive encounter with the site, albeit under the careful guidance of Sakamoto. We arranged to hire a fishing boat for transport, obtained permission from the Nagasaki City authorities to land on the island, signed disclaimers in case of injury or death, and were allowed to spend four hours on Hashima. Here, we initially allowed Sakamoto to choose our route around the island, and listened to (and recorded) the stories he told us about his and others' lives. As we walked further on, we interposed more of our own questions to Sakamoto, prompting a memory work that spoke to our interests as well as his own wellversed anecdotes. Finally, we requested that his chosen route be abandoned, such that we could carry out other field work activities particular to our disciplinary training, such as dérives and transect mapping, while continuing to note the interruptive impact of this on Sakamoto's memory work.
As we moved around the island with Sakamoto, we took extensive field notes, took more than 1,000 photographic images, and recorded much of the dialogue that took place. The next day, we sought to recollect the trajectory of the tour using, primarily, the visual cues afforded by our sequenced photographs, and charted the path we took on a map of the island. The difficulty and uncertainty of this process prompted us to reflect on our own memory work, and the framing devices we were using to come to a shared recollection of events, but also that of our guide Sakamoto. As we walked with him, we had listened to his recollections of place, and had contrasted this with other stories we had researched. We had also engaged in his performance of memory, however; that is, the embodied practice of walking a route. To a degree, we had thus shared something of the site's affectivity in regard to the prompting of particular memory acts; that is, we had felt the same work of elements, atmosphere, and infrastructure that reminded Sakamoto of specific experiences and anecdotes. At the same time, we had interrupted Sakamoto's memory work by peering off into corners, asking questions that served our own research purposes, and, most abruptly, by undertaking our own transect analysis of part of the island. In the following, we draw out four moments, each taking place at a different point on our route, that highlight something of this entangled memory work.

Entrance to the School
On landing on Hashima, we entered through a large, locked metal gate and followed Sakamoto across the overgrown school grounds toward the seven-story primary and junior school complex. A vibrant mosaic on the inside of the seawall remains well-preserved among the dilapidation; it was built by Sakamoto and his graduating junior high class. His recollections of the daily ENGAGING HASHIMA activities he undertook here, as well as its organization, are featured time and again in his writings, and as we stood at the entrance he talked through some of these details in Japanese. Two of our team took turns translating his comments into English. Sakamoto stood and pointed out the mosaic (Figure 1), as well as where a baseball net had been located, to protect foul balls from finding their way into the ocean: At the peak there were over 700 children studying here. There was a baseball net over there. When you hit the ball over the fence in baseball, that's usually a home run, but here it was "out." In some cases, the balls would end up drifting to the coast. So children on the coast would never have to buy their own balls. The coast owed the island of Hashima many favours. 2 As we stood with him, and our gaze followed where his hand pointed, we were invited to imagine the din and activity of the school, captured in the minutia of recalled bodily actions, from drinking at the water fountain to pounding up the stairs. The physical materiality of the site lent itself to such a performance, insofar as the basic infrastructure still provided a frame for these activities. It also firmly located these in the past, as only mere traces were now in evidence of particular bodies at play, and these were marked by the touch of wind, rain, and seawater. The light breeze coming up from the water, the sound of the nearby waves, and the dappled light that shone down on us also became a shared, felt experience that connected us with the play of elements on those bodies we heard about. For us, though, these sensuous engagements again served also to distance us from the past, as we became keenly aware of how we sweated in the here and now, and were likely to sweat even more during our time on the island.
This introduction to life on the island very much set the tone for what was to follow, as Sakamoto proceeded to walk us through successive frames, from hallways to classrooms, and to stop at various points that were meaningful to him, and whose meaningfulness he had clearly articulated time and again to numerous audiences. For example, he stopped in front of a small, rusted elevator entrance cut into the end of a school hallway: This was the only elevator on the island, but it was used only to carry lunches for the students. It wasn't for the people. There was never an elevator on the island that people rode; we were always climbing the stairs.
A narrative of depth and height was introduced here that also structured Sakamoto's tour and therefore our encounter with the site. There were (and are) a multiplicity of gradations in use on Hashima, with staircases being the primary means of moving up and over the island's central rock, between apartments or through the various levels of the high-rise school complex. Our route itself traversed these gradations, moving slowly up through floors of school and apartment buildings, across bridges and paths to the shrine at the top of the island and to views of the expansive ocean and sky, before descending into shadow to what was the social heart of the island where markets and shops were located. Both historical and contemporary social interactions were kept at a material distance from the rocky core island itself, negotiated through the inorganic construction materials that created paths and steps and shaped the routes former residents and contemporary visitors could take.
Importantly, the choreographed route Sakamoto set out on prompted an uneasiness among our group, as we had talked over our own agenda in regard to what we would strive to do while on the island. Sakamoto seemed to be performing a script that appeared repetitively in other contexts-in the books and oral history collections that we had consumed before visiting the site. As we followed Sakamoto onto higher levels of the school building, some of us lingered behind the bigger group to take photographs, caught by the textured surfaces of walls, light fixtures, and so on, and others intermittently followed the lines of sight presented by the building itself, gazing out of windows and peering around corners. The perilous state of the floors, however, also warned us not to roam too far from the pathway Sakamoto had tried and tested so many times.

The School Stairs
As we walked through the school complex, Sakamoto described to us his feelings about the naming of the island-Hashima or Gunkanjima. His wife is also a former Hashima resident, he told us, and the two of them never refer to the island by any other name. By contrast, Sakamoto's ENGAGING HASHIMA nongovernmental organization consciously draws on the Battleship metaphor, both in the group's name and in the iconography on its Web site, flyers, and other materials. We reached the stairs once more, and Sakamoto observed that there was no mention of the Battleship moniker in the school song, although there is an allusion to it with a lyric that refers to an "unmoving ship." For us, one of the themes we had become interested in through prior archival work was the experience of childhood on the island. This resonated with Sakamoto's recollections, of course, but at this point we wanted to take the opportunity to pursue what was a throwaway remark. One of the team asked Sakamoto, "Could you please sing the song?" Sakamoto looked taken aback, but then, as we all stopped to provide an audience for him, he began to sing in a halting, wavering voice the tune that was established as the school song in 1930, based on music by prominent school composer Ikuo Jun and with lyrics penned by Kyushu-based educationalist Yatsunami Norimichi (Figure 2).
Prompted by us, Sakamoto not only brought to mind the tune and the lyrics he had sung as a schoolboy; he also undertook to reiterate what would have been many past performances of singing it. As we listened to him, something of the enduring quality of the past, captured in the disposition of the present body, became apparent. Clearly, this was memory work, but it was memory work that emerged just as much from the desires of the team as it did from Sakamoto's singular experiences. As the tour progressed, more and more questions of this nature were asked, and we could see and hear Sakamoto's excitement in replying to these even as they took him further away from his usual narrative. The walk was becoming an opportunity for a dialogue rather than a speech, such that we all began to become "implicated," in Taylor's words, in the in situ performance of memory.

Sakamoto's Apartment
We had walked through the low-ceilinged stairs and stairwells of a large concrete block of flats to reach Sakamoto's apartment, observing the scratched and eroded walls, and the accumulated litter, some of which, one of our team noted, had not been there on previous visits, and thus indicated that some visitors were arranging items on Hashima as though it were a stage set. Sakamoto lived on the ninth floor of this block from 1969 to 1974. The apartment contained two rooms, one the size of four and a half tatami mats (each mat measures roughly 85.5 cm × 179 cm) and the other six. The broken window looked onto similarly distressed apartments only a stone's throw away (Figure 3). Sakamoto dug out of a cupboard his family's nameplate that would have marked their apartment out to visitors: "The names list my parents, myself and my two sisters." Toilets were shared and there were no bathing facilities other than three communal baths on the island.
We spent around fifteen minutes at the entrance to Sakamoto's apartment; we stood in the hallway, as Sakamoto recalled everyday activities just inside of the main room. The perilous state of the floor here made it too dangerous to peer around as we had at earlier locations. As Sakamoto talked about doing his homework, he pulled out from a small cupboard a notebook with faded English script on it. Sakamoto repeated a line he had used earlier when we were in the school: "I was top of the class in English," he said, "although I could never really speak it that well." The English teacher reportedly only taught grammar, but the notebook also recorded at least one lesson on The Adventures of Huckleberry Finn and The Adventures of Tom Sawyer. Sakamoto recalled another notebook: "I received a love letter from my girlfriend (now wife) and hid it in the back of a notebook. My mother found the letter. 'Who is this from?' she asked." Dating on the island was also structured around gradations of height, this time centered on the rooftop. There was reportedly a beginner, an intermediate, and an advanced "course" for dates, which corresponded to one's physical position on the island. The advanced course participants were able to use the highest points on the island, safely out of view of prying eyes, and at the furthest distance from the island's subterranean raison d'etre. "I progressed through all three courses," laughed Sakamoto.
It was clear to us that here were an array of objects that were crucial to Sakamoto's memories of self. These were not simply testimony to past decisions and expressions of intent, but a potent reservoir for his recollections of living on Hashima. In his former apartment, Sakamoto told us more about the abandoning of the island and objects that had formerly played such a crucial role in people's lives: When people were leaving the island, the objects in their apartments were their property so they were free to take it, but many of the things in other buildings were public [or company] property, so ended up being left behind. In the hospital, x-rays and the like were just left behind.
His referencing of these objects on his tour indicates something of the importance of commodities at this site. During the period when Sakamoto was a child on the island, Mitsubishi was at great pains to "domesticate" the site for its workers and their families by introducing not only a school and hospital, a police station and post office, but also new electrical goods from TVs to a motorized planetary display. They also ensured that Hashima was at the top of the queue for new cinematic releases. For Sand (2013), the centrality of the commodity in popular rememberings of the time reinforces "the sensation created by the mass media that each family's life was a microcosm of the history of the [Japanese] nation" (129). For Sakamoto, the "everydayness" expressed through these material objects of domesticity connects a particular Hashima context to a perceived universal Japanese national experience, which was absent in much of the social infrastructure of the island. There are resonances here with a postwar intellectual phenomenon known as "everyday life studies" (seikatsugaku), which emerged from the work of Waseda University architecture scholar Kon Wajirō and was popularized in the heritage environment by his student Kawazoe Noboru. Kawazoe's work on the everyday was appearing in the years prior to Hashima's closure, and a similar belief that "true culture resided in the most ordinary things" (Sand 2013, 118) appears in Sakamoto's recollections.
Despite the universality of commodity objects, however, Hashima residents had a unique relationship to corporate Japan. Residents experienced a largely benevolent corporate control cocooned from broader social pressures; the company charged nothing for rent or utilities and provided free medical care. The island was seen as a microcosmic capitalist utopia. Yet the utopia had its limitations. Researchers in the mid-1960s attempting to access the island to study its social forms were shadowed by watchers from the corporation, which, according to Sakamoto, led the study to be curtailed. After the island's closure, residents were scattered across the country and struggled to cope with life outside, with Sakamoto reporting several suicides. Although an internal phone network meant the island's residents were well networked, just a single phone line led off the island, reinforcing Hashima's imbrication in, yet separation from, the everyday life of postwar Japan.

Hell's Staircase
At the top of the island there is an open space with views across the rooftops to the watery horizon, and a shrine. Here some of our team talked with Sakamoto about the games he had played on the island, such as hide and seek, rooftop baseball, and other childhood activities. Piqued by the idea of Japanese sociologists in the 1960s seeking to research on Hashima, we asked about broader social relations. Sakamoto recalled how people would regularly leave their babies with neighbors when taking themselves (or their other children) off to the public bath. The physical and social intimacy also created a need for alternative models of dispute resolution, with people filling mediation roles needed to return a sense of social harmony. These were also missing when people were forced off the island. Departing the island was like "leaving your family behind," said Sakamoto.
We were also, however, interested in the changing morphology of Hashima, and the role of various erosional and weathering processes in this, as a means of reflecting on the nonhuman temporalities that helped to constitute this site. Up to this point we had followed, literally, in the footsteps of Sakamoto's planned route. But we needed to carve a reflective space outside of Sakamoto's narrative for these other biographies of place to become manifest. As we walked down to Hell's Staircase (jigokudan) in a physical descent, we requested of Sakamoto that we stop at the bottom to map a partial transect up to the outer sea wall. Relaying and reiterating the request took a minute or so, and while some of the team prepared to map the transect others sat down with Sakamoto to wait until this performance was completed. In reversing the subject of the memory performance, we disrupted the coherence of Sakamoto's route, allowing for a different set of mnemonic implications to precipitate. As the mapping progressed over the course of twenty minutes, only one member of the team remained with Sakamoto; others walked away down the passageways between buildings, watched or filmed the mapping process, and reflected on the sensation of the elements on their skin, ears, and eyes.
A transect consists of a line drawn between two points, along which the dip and slope, as well as the character of all the different materials present, are recorded. In this case, this included rubble, rocks, pebbles, rust fragments, and tiny sea shells, and the erosional marks etched into these, as well as their arrangement. Collectively, these conglomerates and angles told a story of wave action undercutting the nearby sea wall, as well as the everyday action of wind and sea spray on steel-reinforced concrete and wood, intensified during periods when heavy tropical storms drove seawater down the island's narrow passageways ( Figure 4). As currents continued to scour under the wall, we thought, the buildings clustered onto Hashima's rocky outcrop would peel away, slumping down through the streets. The base of Hell's Staircase was clearly an area of great stress in regard to the wind and wave action associated with typhoons and, watching us at work, Sakamoto remarked that a woman had been sucked out through the small sea gate that punctuated the wall here.
Here, the disruption of Sakamoto's walking route at the most grounded location on the island-towered over by decaying tower complexes and the looming staircase itself, with glimpses of seawater curling under the concrete we walked on-allowed for other social histories of Hashima to emerge. As Sakamoto sat and talked with one of our team, the topic of forced labor on the island came up. Prompted by our team member, Sakamoto observed that the term forced labor is misleading because it fails to account for the fact that many of the Korean workers had signed up to work in factories in Japan (and by chance had ended up on Hashima), that there were Korean families as well as Japanese families living there at the time, and that everyone, whatever the nationality, was faced with severe austerities and brutal working conditions. Obliged to abandon, temporarily, his physical route across the island, and displaced from his otherwise authoritative role as Hashima's biographer, Sakamoto's accounting of what had occurred on the island had become much more wide ranging and inclusive, albeit historically partial. This was an attempt at inclusiveness that is very much absent from official narratives on Hashima's relevance today and in the future-although we note now that our guide's discussion on the meaning of forced labor has been replicated in official discourse on the UNESCO bid. The day after the successful nomination of the bid in 2015, foreign minister Kishida Fumio parsed a careful linguistic distinction between the Japanese terms hatarakasareta (were forced to work) and the more contentious kyōsei rōdō (forced labor), the latter often associated with activist projects aimed at highlighting the exploitative elements of the Japanese Empire (Yoshida 2015). Awkward historicizations and intense battles over language continue to infuse history and heritage sites across Japan, indicating a desire to celebrate an industrial modernization and to somehow separate this from Japan's concurrent imperial expansion.
The transect complete, Sakamoto hurried us back on to his planned route. As we passed the remains of the cinema, and the coal-head, we clambered and slid over large piles of rubble and then balanced along the sea wall back to our landing point. Even as passageways became ever more filled in with brick, stone, wood, and glass, Sakamoto appeared determined to deny the dynamic environment of Hashima in the now and to perform once more the path-finding he had learned as a child.

CONCLUDING REMARKS
For much of the time we spent with Sakamoto on his route around the island we shared with him some of the site-specific affects that he felt as a child, from the drone of the wind up on the rooftops to the dark, sweat-drenched interiors of the stairwells. This corporeally centered, felt experience did much to animate a picture of Hashima's past. At the same time, we acknowledged a distancing from these events as our own, singular responses to the site became ever more apparent. For one member of our team, the site was felt as an artificial one, cut off from the bedrock that we barely saw or touched, and that clearly did not feature as significant in Sakamoto's narratives. For another, Sakamoto's recollections of his and others' games of tag in these same stairwells contrasted with a current sense of claustrophobia in what seemed very cramped environs. For a third, his stories of height and depth felt limited, separated as they were from the dark and cramped undergrounds that were the locations of the back-breaking work of the mine.
As we listened to our guide's recollections, our own singular experiences brought both empathy and alienation. Furthermore, our own experiences incorporated archival work we had undertaken on Hashima, such that we were very much aware while on tour how contentious the planned memorialization of Hashima's physical and social architectures was. Even as we sought Sakamoto's anecdotes about island life during the Golden Age of Hashima, we were aware of how this personalized narrative, alongside the memory work carried out by former residents, had become part and parcel of the heritage bid. Just two days before, for example, we had visited the Gunkanjima Museum, which is affiliated with official attempts to register Gunkanjima as a UNESCO World Heritage site, and noted its displays as structured around a series of nostalgic images of school sports days, women shopping, packed cinemas and community festivals, and the conjuring of Hashima as a place of life and progress, of hardship but community, of work and industry, and, ultimately, of nation and modernity. Similar narratives had infused our earlier commercial tour of the island.
For Sakamoto, the island is important to his memories of childhood, school, and family life, but these memories accrue a wider significance because they are testimony to the particular living arrangements built on Hashima. Avant garde concrete buildings raised on the island enabled particular kinds of urban living to emerge, and the fact that Japan was the first country to experiment with these material and social architectures helps underpin the UNESCO bid. As Sakamoto and Gotō (2009) wrote, "The particular arrangement of buildings on the island and their architectural style are historically and technically of unique value" (208). This focus on the significant singularity of Hashima's architectural form creates a very particular relationship to time and space while simultaneously glossing over other histories and geographies (even of the island itself); Hashima is "placed," in other words, by virtue of its physical infrastructure and associated social life. Clearly such an account sidesteps more problematic aspects of Hashima's social life, and in particular the linkages between the experimental architectures and industrial functions that Sakamoto remembers and a Japanese imperialism. We want to note, however, that it also sidesteps the very material conditions that provide Sakamoto, and other former residents who perform similar memory work, with a wealth of environmental prompts and framings for their narratives.
Over time, Sakamoto's negotiation of this relationship between environment and memory has ossified into a particular path-a tour through which a specific narrative of the past is constructed. This path also weaves its way through the mass of Hashima-related publishing and heritage campaign materials. Yet this process is never entirely stable for two key reasons. The shifting material conditions of Hashima-as its architectural form erodes and its social structures come under increased popular and academic scrutiny-allow spaces for this ossified narrative to be interrupted. The memory work on Hashima always already takes place in a dynamic environment. Our interruptions are, in the scheme of things, marginal and impermanent. Yet the slow reclamation of Hashima by the sea and wind reveals the precarity of mnemonic particularity. The changing materiality of sites-environmental, as well as social and political -can prompt memory work that encompasses more problematic histories and, in so doing, give bodies pause for thought.