Framing the Material Global: The Grounded Politics of HIV Testing Scale-Up

Abstract Drawing on the framework of transnational materiality (Gille 2014), a conceptual contribution from the Framing the Global project, this article examines how HIV testing and counselling services became a site of political contestation in Taiwan since expanding service delivery was proposed as a response to the HIV epidemic by global health institutions in the 2000s. This global scheme, the study argues, not only reshaped the organization and practices of HIV service delivery, but also generated vulnerability as these practices connected to HIV-positive people’s lives were largely governed by global models and national programs that may not fully reflect local concerns. In response, local NGO workers politicize service delivery in part by transnationally adopting and promoting practices that attend more carefully to localized forms of vulnerability. Using qualitative data collected in Taiwan, the analysis reveals a grounded politics of HIV service delivery that highlights the materialization of vulnerability and its management.


Introduction
HIV politics has been a central concern for global scholars since the outbreak of the HIV epidemic. The ever-richer scholarship in the field has demonstrated various ways in which global politics is approached by different scholars. For example, in a study which seeks to answer why former South African President Thabo Mbeki's controversial thesis about AIDS denialism resonated with a large part of the South African society, Fassin (2007) argues that such a politicized response to HIV and biomedical authority should be understood in the global and South African history of colonialism, racism, apartheid and inequalities. Meanwhile, Susser (2009) has contended through a gender lens that transnational aid programs and national policies alike overlook African women's particular vulnerabilities and collective action in fighting the epidemic. Chan (2015) has analyzed the politics of AIDS activism around the world by looking at how they not only challenge the rules of pure science and the perfect market, but could also be penetrated and transformed by the global funding resources they challenge. Swidler and Watkins (2017) have explored the hierarchical "aid chain" of the transnational AIDS enterprise by investigating the gap and discordance between foreign altruists and local brokers who work together in Malawi. With three solid case studies, Harris (2017) has examined the ways in which the global prospect of universal health coverage is endeavored by professionals in the developing world who are empowered by democratization. Such examples of recent scholarship on HIV have pointed to multiple analytic angles for understanding global politics: historical, institutional, discursive, race, gender, among others.
The following study of HIV service delivery in Taiwan seeks to contribute to this large body of scholarship on global HIV politics from a perspective of the material. Drawing on the framework of transnational materiality (Gille 2014) from the Framing the Global project (Kahn 2014), it looks at how material practices of HIV testing and counselling delivery on the ground are governed and contested by actors at multiple scales beyond the sites of service delivery themselves. As Gille (2012; suggests in her conceptualization of transnational materiality, material practices and technologies as constitutive of social life are not just locally contained, but are shaped by power exercised far beyond the local. In other words, even in the absence of a direct institutional force, social life could be altered by transnational power and politics when such a politics becomes materialized in the substances and practices that constitute social life, which Gille calls the materialization of politics. This framework is particularly useful in analyzing the material practices of HIV service delivery in Taiwan, since HIV service programs and practices in Taiwan are often influenced by actors beyond the national border. Moreover, as will be illustrated in this article, the material practices of HIV service delivery are informed by service clients' lives; this is because in a time when the social exclusion of HIV positive individuals prevails, service practices may expose clients to social harms. In fact, there have been an increasing number of studies on HIV that attend to the relations between politics and materiality in transnational contexts. For instance, Biehl (2007) finds that an HIV treatment policy in Brazil has unintendedly contributed to what he calls the pharmaceuticalization of public health, a form of governance that centers on pharmaceutical solutions and overlooks structural inequalities. Nguyen's (2010) study in West Africa reveals an emerging therapeutic form of surviving and citizenship under triage and HIV treatment trials. Moreover, Crane (2011) describes how the genetic maps of HIV developed primarily from the Global North, as well as the knowledge claim about treatment and drug resistance based on the maps, reflect global inequalities and have clinical and political implications for patients in the Global South. Decoteau (2013) illustrates the ways in which HIV biomedicine constitutes new technologies of the self that privatize blame and patient responsibility in South Africa. These studies illuminate how HIV treatment is embedded in larger national, transnational, and colonial webs. My study seeks to contribute to the scholarship on HIV biotechnology and politics by focusing on a set of material practices that are less studied -HIV testing and counsellingand by examining how the material practices of HIV testing and counselling are not only a site of governance, but also one of dispute and contestation.
This study draws on both secondary and primary qualitative data to trace the transnational trajectory of HIV testing practices and its politics. I examine various reports and guidelines in response to the HIV epidemic issued by international institutions, in particular the World Health Organization (WHO) and the Joint United Nations Programme on HIV/AIDS (UNAIDS). The two institutions and their policy recommendations are often cited in Taiwan HIV policy documents, despite Taiwan not being a member of the WHO. I trace when policy recommendations on HIV testing practices emerge and how they unfold through an analysis of those reports and guidelines. At the national level, I look into HIV program documents and journal articles regarding HIV testing programs compiled and published by government health officials. Attention is paid to how national programs on HIV testing are informed by ongoing shifts in international guidelines and evolve accordingly.
To unveil a grounded view of a transnational project of HIV testing, I examine the documents, petition letters, and public messages concerning HIV testing programs and practices issued by local NGOs who closely work with HIV positive communities and marginalized communities (for example, injecting drug users and sexual minorities, who are often considered as target populations for HIV testing by health authorities). Moreover, during my fieldwork in Taiwan in 2019, I conducted in-depth interviews with 18 NGO workers who offer HIV testing services, focusing on the knowledges mobilized and practices employed in their services. The aim of this study is not to compare the practices of HIV testing services among various sectors and agencies. Instead, the aim is to scrutinize how politics are implicated in the practices of HIV testing and related service delivery, that is, how the materiality of service delivery is simultaneously associated with clients' potential vulnerability and with providers' efforts to reduce and navigate clients' vulnerability.

The Global Scheme of Scaling Up HIV Testing
In response to the global epidemic of HIV, international institutions provide financial support to national governments and local actors, but also promote knowledge about "best practices" as a way to guide their intervention strategies and practices. Yet what constitutes those best practices has been shifting over time.
In the late 1980s and early 1990s, the Global Programme on AIDS, a major program established by the WHO, focused primarily on condom use and behavioral change as a key strategy (Packard 2016, 281). Later in a collection titled Summary Booklet of Best Practices, UNAIDS suggested a model of flexible provision of HIV testing services in accordance with the client's specific needs: "Voluntary counselling and testing is usually carried out in STD clinics, hospital outpatient departments, and hospital wards, but can also occur in centers dedicated to this specific purpose" (UNAIDS 1999, 105).
However, with the advancement of diagnostic technologies, a new approach stressing the expansion of HIV testing services emerged. In 2001, the United Nations General Assembly Special Session on HIV/AIDS included "expanded access to voluntary and confidential counselling and testing" as a target in its Declaration of Commitment on HIV/AIDS (2001,20). Perhaps most notable was WHO's call to scale up (i.e., expand) HIV testing provisions. In a report titled Increasing Access to HIV Testing and Counselling issued in 2002, WHO elaborates on this model as "best practice": New approaches to HIV testing and counselling must now be implemented in more settings, and on a much larger scale than has so far been the case. WHO is advocating that health-care workers should offer testing and counselling to all those who might benefit from knowing their HIV status, and then benefit from advances in the prevention of HIV infection and treatment of HIV-related diseases (WHO 2002, 3).
In this instance WHO simultaneously redefines who ought to have access to testing, how testing could be delivered, and the role of testing in HIV care. It suggests that not only those who seek medical services but also all those who want to know their HIV status should have access to testing; that testing services should go beyond a clinic-based approach and be offered widely; and that testing services could link individuals to further services such as prevention and care. With this model, the problem of HIV is redefined, as well: "In many parts of the world most severely affected by HIV/AIDS, fewer than one in 10 people with HIV know that they are infected. This widespread ignorance of HIV status is the direct result of people's poor access to HIV testing" (WHO 2002, 3). Whereas in earlier years a lack of knowledge about safer sexual practices among sexually active people was seen as a main challenge to HIV prevention (WHO 1997, 6), five years later, HIV-infected individuals' "ignorance of HIV status" was presented as a key problem to be solved. A new organization of practices of scaling up HIV testing and counselling is proposed as the solution.
Like UNAIDS, WHO positions itself as a knowledge translator and policy advocate by offering technical guidance to national health authorities. In Global Health-Sector Strategy for HIV/AIDS 2003-2007: Providing a Framework for Partnership and Action, a document addressing national health ministers and policymakers, WHO advocates for the model of scaling up HIV testing and counselling: [P]eople have a right to know their HIV status, and testing and counselling should be widely accessible through innovative, ethical and practical models of delivery. HIV testing and counselling are entry points to HIV-related care and prevention services and provide opportunities for clients to reduce their risk of acquiring or transmitting HIV. (WHO 2003a, 12) The term "entry point" maps even more clearly out the material deployment envisioned in the model, that is, clients are not only being provided with testing services but are also linked to prevention and care recourses through the testing services. In that regard, WHO makes it explicit that knowing one's HIV status enables them to "[g]ain early access to HIV-specific care, treatment and support" (WHO 2003b, 1) and could stem the transmission of HIV, which constitutes the ground for an expansion of HIV services.
At the same time, WHO also elaborates on the principles that should be implemented along with a scaling-up of HIV testing and counselling. One is that testing should be voluntary and that pre-test information on the purpose of testing should be offered. Another is that post-test support and services, such as counselling and referral to care, should be offered according to the result. Still another is that confidentiality must be protected (WHO 2002, 5). In subsequent years, the model of scaling up HIV testing and counselling have been reiterated and promoted in various documents and technical guidance issued by WHO and UNAIDS (e.g., UNAIDS/WHO 2004;WHO 2005;WHO 2006;UNAIDS 2010), thereby gaining hegemony in global response to the epidemic.

The Transformation of HIV Testing Delivery in Taiwan
The scale of HIV testing services has been growing in Taiwan. During the 1980s and early 1990s, most HIV testing services were only offered at governmental health departments and a few major hospitals. For individuals who test for HIV, their identities would by law be reported to the government, and the government would keep track of their health conditions by getting in touch with these patients. In 1997, however, in an attempt to increase people's willingness to take a test, the government established additional anonymous HIV testing programs at six major hospitals located in cities (Liu et al. 2008, 8), allowing one to get a test without disclosing their identity.
Since the 2000s the Taiwanese government has expanded HIV testing delivery statewide by adopting WHO's model of scaling up HIV testing (Liu et al. 2010, 186). In a 2001 policy document compiled for HIV program implementation for 2002 to 2006, the Taiwan Centers for Disease Control (TCDC) stated that a main challenge for HIV prevention and care is that "at-risk groups are unwilling to get a test, which prevents HIV-infected individuals from early detection and early treatment and entails continued transmission of the virus" (TCDC 2001, 28). It also explicitly states that an expansion of HIV testing delivery should be a policy priority (ibid., 25), and sets goals for numbers of testing for target populations. Similarly, increasing HIV testing services was a policy aim from 2007 to 2011 as well (TCDC 2006, 22).
In health care settings, TCDC has implemented HIV testing programs for sexually transmitted diseases patients since 2003, and those for drug addiction patients since 2004. HIV testing has become mandatory for pregnant women since 2005 (Liu et al. 2010, 186). Additionally, anonymous HIV testing programs have also been established at governmental health departments in 2005. The numbers of hospitals offering anonymous HIV tests increased from six to 10 in 200710 in , to 19 in 200810 in , and to 22 in 201010 in (TCDC 2015. Since 2008, TCDC has initiated a "national testing week" campaign annually to encourage the public to get an HIV test. Moreover, TCDC has asked local health departments to offer outreach testing and counselling services at venues such as bars and parks in order to recruit target populations who were rarely reachable in health care settings. For instance, one of the tasks assigned to health departments in 2008 and 2009 was to "collaborate with the gay community and for-profit sector to conduct health education campaign and testing for HIV prevention" (DOH 2008, 5). 1 As a result, the number of HIV tests performed increased significantly. The number of anonymous tests in a year rose from 2,169 in the year 2000, to 5,354 for 2005, 11,194 for 2008, and 19,895 for 2010(TCDC 2015. These numbers increased almost tenfold in a decade. Individuals who tested for HIV with their identities would be recorded in an electronic system managed by TCDC, and within just a few days they would be contacted by local health departments and asked to see a doctor. If they need to take medication for HIV from a doctor, the payment would be covered either by National Health Insurance or the state budget, depending on the policies in place. In other words, the Taiwanese government has striven to put in place the global model of scaling up HIV testing and link-to-care services.

The Materialization of Vulnerability
Perhaps unexpectedly, the transnational scheme of scaling up the delivery of HIV services has generated various discontents from local communities in Taiwan. In 2008, two non-governmental organizations (NGOs) working on HIV prevention and rights advocacy together organized a press conference to contest how HIV testing was delivered by health departments in public venues. This was the very first public protest specifically on HIV testing policy. Moreover, in 2009, three NGOs jointly initiated a protest petition to TCDC concerning the national HIV testing policy, which was endorsed by more than 30 local NGOs (including Taiwan AIDS Society, a medical professional organization) and more than 25 scholars and experts in the field of HIV (see TTHA, PRAA, and TLA 2009). This was one of the first protests jointly mounted by numerous HIV-focused NGOs in Taiwan.
The 2008 and 2009 protests shared some similarities in their views on HIV service delivery scale-up, that the lives of people who were enrolled in those expanding testing programs could be adversely impacted by the ways in which testing was delivered. In other words, the materialization of the global model on the ground has increased patients' vulnerability. I analyze how these specificities were materialized in the following three categories.
First, in their protests local communities expressed concerns about the physical space and environments in which HIV testing was carried out. Based on their investigation in venues, NGO members found that the space for testing was often not organized in a way that was able to secure clients' privacy. As described in an NGO's press release, the space in many cases "was an open space with little separation, and everything said during the testing procedure could be overheard" (TTHA 2008). The consequence could be serious if a HIV positive rapid test result was disclosed to individuals other than the client who took the test, given the Far from being occasional, a breach of clients' confidentiality, I argue, was a product of systematic extension of HIV testing delivery into non-health-setting public space without an appropriate reorganization of the space. One of my respondents who provides testing services in venues describes how a tension emerges in space selection and utilization: Take sauna as an example, we usually offer services at the social hall, because there is light and because everyone would see us and come to us. If we sit at a dark corner, no one would know.… But when people come to us, it is possible that other people would be around, so confidentiality could be an issue (Interview 04/10/2019).
In order to gain people's attention, service providers are likely to locate their services at an eye-catching spot in venues; however, that could also jeopardize clients' privacy if additional spatial arrangements are not in place to separate each client from other people in the vicinity. In that regard, how HIV care services are provided with particular material arrangements of the space becomes consequential for clients' vulnerability.
The second category of a materialized specificity appears in how advocates' protests voiced concern that in many cases HIV testing was provided at the expense of counselling. Counselling services, in their views, include offering HIV-related knowledge, risk evaluation, providing emotional support and posttest support. They are particularly important in HIV care because through counselling, clients would gain knowledge for self-protection, would understand the meaning of testing, could be emotionally and socially supported, and could be linked with needed resources and care (PRAA 2008;TTHA 2008). However, advocates pointed out that in many testing services, "counselling provision is neglected, whereas many [government] resources were used to increase testing sessions and recruit people [for testing]" (TTHA 2008). Such an imbalance between testing and counselling provisions has produced a new form of vulnerability, as an NGO report observed: Many callers have expressed their doubt and resistance toward HIV testing. Even when some of them summon their courage to get a test, they could be anxious or would panic about the test result due to a lack of counselling provision. This could lead to the public's distrust toward the HIV testing system and lower their willingness to get a test in the future (PRAA 2008).
Clearly, advocates were not arguing against testing services per se; rather, they argued about how and why a lack of counselling provision in HIV testing delivery engendered vulnerability. In fact, largely owing to the stigma attached to HIV, a number of individuals diagnosed with HIV had committed suicide during the 1990s and 2000s, some instances of which received news coverage (e.g., United News 1990, 1992, 1994. Given the historical background and social climate, service receivers' potential vulnerability is mediated not only by a diagnosis of HIV but also by the sorts of services available to them. A third illustration of a materialized specificity lies in how local communities questioned whether anonymity in allegedly anonymous settings was actually guaranteed. As stated earlier, anonymous HIV testing programs have been implemented in order to encourage people to get a test. However, in practice it was noticed that in an attempt to facilitate linkage-to-care for those who test positive for HIV, testing providers might persuade those who planned to get an anonymous test to get a named one, or might encourage a client to provide personal information even if the client opted for an anonymous test (Chiu 2006, Taiwan Tongzhi Hotline Association Persons with HIV/AIDS Rights Advocacy Association of Taiwan, and Taiwan Lourdes Association 2009). An interviewee of Chiu's (2006) who had taken a HIV test in a hospital observed: I have previously consulted with a friend, so I knew that I could use a pseudonym [for testing]; otherwise, the staff would not tell us that we could take a test anonymously, and in that case we might fill in our name, phone number and address. Most likely the staff would encourage us not to use a pseudonym (Chiu 2006, 91).
What community advocates found in 2008 and 2009 was very similar to what Chiu found earlier, that is, the blurred line between anonymous and named HIV testing in practice. Chiu's finding is clear: [Providers'] encouragement to a named test is to maintain their contact with clients; they could keep track of the clients who test positive. Thus, [providers] told clients that an anonymous test has little use … or that if an anonymous test result is positive, the client needs to do an additional named test in order to proceed to treatment. These wordings implied that anonymous tests are meaningless or do not make one feel safe. Some nurses do not even mention that an anonymous test is an option in the first place (Chiu 2006, 91).
Indeed, HIV treatment programs in Taiwan are only accessible to individuals who test for HIV with their identities and are registered in a state management system; it is also true that one would be contacted by government staff and health providers if they test for HIV and their identities are known.

Framing the Material Global
Nevertheless, how do local communities and ordinary people perceive named testing and HIV registration? In the 1980s and 1990s, the HIV positive status of a number of individuals was disclosed (unintentionally) to others during the processes of state contact tracing (Chiu 2006, 13). Being subject to these contact-tracing procedures and bearing a risk of disclosure of their HIV positive status has constituted one of the main reasons that some people hesitate to take a named HIV test to resist registration. In addition, once registered, having an HIV record in a state data base and in a hospital archival system could entail discrimination in various circumstances. For example, employment and access to health care services were some of the everyday issues confronting HIV positive individuals, and these issues were not new, as a news article reported: Even though the law does not prohibit HIV positive patients from applying for a job at military, police and investigation agencies, these agencies still require a HIV examination. If those enrolled in these agencies are found to be HIV positive, they would be tactfully persuaded to leave (Shih 2009).
Similar situations also applied to many other categories of employment, both public and private (PRAA 2006;PRAA 2009). As for health care, numerous cases have shown that HIV positive individuals' requests for health care services other than HIV care services were likely to be denied by health care institutions (PRAA 2007). Furthermore, if a foreigner tested positive for HIV with their identity prior to 2015, they would by law be deported from Taiwan. In short, a locally-grounded perspective reveals that as named HIV testing services are connected with a state registration system, those services also come with undesirable social risk and vulnerability. This is why in the 2009 protest, NGOs and advocates demanded, "[b]efore conducting a test, testing agencies should disclose full information about the test to clients, including whether the test is named or anonymous … In the case of anonymous testing, clients' personal information should not be collected" (TTHA, PRAA, and TLA 2009).
As demonstrated in this section, an expansion of HIV testing services as well as the material practices (spatial arrangements, the availability of counselling services, and the practices of identification and anonymization) associated with the expansion could and have altered people's lives in significant ways. While these practices of service delivery were determined more by some global models and national policy priorities and less by the people experiencing the delivery, these practices could unexpectedly introduce everyday violence beyond service sites. I would characterize this relation between delivery practices and HIV politics as the materialization of vulnerability.

The Politicization of Service Delivery
Whereas state policy in the 2000s prioritized increasing the number of HIV tests, local NGOs articulated another set of transnational practices in response: confidentiality, supporting counselling, and voluntarism. In fact, NGOs were well aware of some ethical principles and practices of HIV services validated by global actors, which could be seen in their 2008 protest press release: "the necessity of HIV preand post-test counselling is well documented in United Nations' International Guidelines on HIV/AIDS and Human Rights" (PRAA 2008). Some of these NGOs have also been delivering testing services since the late 1990s. In this section, I show how these NGOs constructed themselves as an alternative site of HIV care by promoting a set of practices that responded to the potential vulnerability facing testing service clients.
One strategy that NGOs took was to inform the public about what they needed to know when planning to take an HIV test. At the end of their press release for the 2008 protest, NGOs included a list of "Advice for Testing Service Clients," which included the following: 1. Informed consent: Before doing a test, we as clients have the right to know and providers have the obligation to provide information about the test. Only after an informed consent is granted can a test be initiated. Some of the key information are: What agency offers the services? What is the testing method and item [the diseases being tested, such as HIV, HPV, or HCV]? Is this a named or anonymous test? … 2. The right to ask: During a test, we as clients have the right and should not hesitate to ask a provider if we have questions regarding sexually transmitted diseases and risk behaviors. Pre-test counselling is a key way to alleviate our doubts. 3. The right to refuse: During a service, if we as clients feel uncomfortable because of unsafe environment or the provider's attitude, or if we are not ready to learn our test result, we have the right to refuse or terminate a test at any stage of the service (TTHA 2008).
The list was circulated on the Internet in 2008 and 2009 and can still be seen on the webpage of some NGOs today. It was obviously responding to the issues NGOs found in many testing service provisions at the time: the safety of a testing environment, the availability of counselling services, and the differentiation between practices of named and anonymous testing. Therefore, the list was not simply a reproduction of global ethical principles; it was rather a product informed by transnational contacts and yet at the same time formulated in response to the Framing the Material Global localized forms of vulnerability potentially faced by clients. HIV testing and counselling service delivery thus became politicized. Beside educating potential testing service clients by circulating knowledge, some NGOs themselves practice these services as well. Since the late 1990s, NGOs offer anonymous testing and counselling services at venues for marginalized communities (Chang 2006;Chang 2007;Chiu 2006). They accumulate and share their knowledge gained by working with communities over the years in a variety of physical spaces. Lin, a social worker who is in charge of an NGO's outreach testing services, describes what she and two of her colleagues do when they are at a sauna to offer services: We say hello to the sauna manager, and then go straight to the top floor of the sauna. We offer services there because people usually do not hang around there. The manager helps advertise our services by posting notices in the space. Those who seek services would come, and we begin with an anonymous questionnaire and counselling. During the counselling I let them know who we are and what our test is about; I check with them whether they have any questions or concerns about HIV and risk behaviors. If they are not familiar with HIV, I would start with the basics. If they express any concerns, I would make sure these are addressed. After taking care of all these, my nurse colleague draws blood for them. We give them a paper tape with our phone number, a serial number and a date and ask them to call back on or after the date.… Often, we have quite a few clients. That is why we hire a part-time employee so that she and I could offer counselling to two clients at two corners at the same time; the top floor is spacious enough to do so (Interview 20 June 2019).
In this case, Lin and colleagues strategically utilize a variety of materials to safeguard clients' privacy and to offer quality counselling in different stages of their services. First, with the NGO's advertising posters circulated in the sauna that inform potential clients where to go for HIV services, NGO workers are able to create a not-too-far distance between the place where their potential clients are and a relatively safe space in which NGO workers provide services, which is the top floor. As such, clients can have access to services without being present physically in the venue and their confidentiality can be maintained. Second, since not every client is comfortable with talking about HIV-related issues to a provider they might not know well, NGO workers offer hard copies of anonymous questionnaires to clients and request clients to fill them out prior to proceeding to blood drawing. These paper-based questionnaires not only help providers to evaluate instantly clients' understandings and perceptions of HIV in a non-verbal way, but also help providers initiate a conversation and counselling service for each client by specific case. Third, when clients are filling out questionnaires, NGO workers stress the anonymity of their services by telling clients that they only need to provide written pseudonyms on the questionnaires and that they will be given paper tapes with serial numbers as later identification. These written pseudonyms and paper tapes with serial numbers on the one hand anonymize clients, and on the other maintain service relationships as they are used for individualized, two-step verification of the relationship in later stages of services. Lastly, NGO workers use both a phone line and a counselling room in their office to optimize post-test services when informing clients of their test results. Usually, when a client whose test result is HIV negative calls back to the NGO, a provider would offer post-test counselling via phone; yet, when a client whose test result is HIV positive calls back, a provider would ask if the client could come to the NGO office counselling room to talk in person, so that a provider could offer further services if needed. Together, it is precisely through the strategic employments of objects and materials (advertising posters, paper-based questionnaires, written pseudonyms, paper tapes with serial numbers, a phone line, and a counselling room) in different stages of the process that counselling services are carried out while confidentiality and anonymity are secured. In other words, global HIV service guidelines concerning confidentiality and counselling are localized through NGO workers' practices engaging material objects.
Nevertheless, not all venues have a spacious or quiet environment for HIV service delivery; providers thus often need to navigate space or adjust their services to the environments. Hsieh, an NGO worker, related to me how he does this in a park: We usually sit at a corner with a light and our poster shown. Sometimes there could be many people hanging around in the park; in that case, I would request that people around could maintain a distance from us, so that our clients would feel more comfortable talking about their concerns.… Once a client's rapid test result showed positive, but there were quite a few people nearby; so, I did not tell the client about the result immediately, but instead we went to another corner to talk about the result (Interview 5 March 2019).
Although the environment was not ideal, Hsieh strived to carve out better spaces for providing pre-test and post-test counselling services by separating clients and other people around. Additionally, there is a different kind of practice by Lee, another NGO worker: "We only provide blood drawing services but don't conduct rapid tests in a bar. I don't think that is a good place to conduct rapid tests" (Interview 18 July 2019). In Hsieh's case, where the time frame associated with a rapid test is given, he secures clients' privacy by navigating space. In Lee's case, however, where the space is limited, she tweaks the time frame of testing by utilizing a blood drawing method (test results not available right away) instead of a rapid test (test results available right away). In short, through their material practices of spatial and temporal adjustments, providers navigate clients' potential vulnerability by managing types of testing technologies and conversations Framing the Material Global (rapid tests and two-stage counselling, or pre-test counselling and blood drawing) in different settings.
Lastly, NGO workers advocate for voluntariness not only in conducting an anonymous test, but also in respecting HIV positive clients' consideration and resolution regarding when they would like to be registered in the state HIV management system, if their health conditions permit such a delay. While from the state's perspective, early HIV registration could mean early linkage to care and no registration could mean missing out on follow-up, NGOs might have a different viewpoint. As Lee mentions: I seldom worry about missing out on follow-up with my clients who test [anonymously] for HIV. People only disappear when there is no counselling provided, so they don't trust you. Through pre-and post-test counselling, clients know well that I will help them and I am on their side, so they come back to me when needed (Interview 18 July 2019).
For NGO workers, the offering of counselling services is a critical step to build trust with testing service clients, and this is especially true when the client's test result is positive. NGO worker Kuo elaborates on what he does in this circumstance: When a rapid test result is positive, if the client's emotional reaction permits, I would ask them about their concerns. Those concerns could be about family, health, privacy, work, school, life plan, intimate relationship, basically anything. Our work is to assist them in this process, including offering moral support and the information and resources they might need. I would let them know what registration means and discuss with them if there is anything they plan to do before being registered.… When they are ready, I go with them for a named, confirmatory test. This process [till their readiness for registration] could be short or could be long for different people. It took more than a year for one of my clients; during the year we stayed in touch. When he needed advice or when he felt down, we talked (Interview 23 April 2019).
According to Kuo, the time following a notice of a HIV positive test result is often a client's most vulnerable time. In order to make HIV positive clients feel better by accompanying them during the time that they are adjusting to their new status, Kuo would provide them with his personal contact on a mobile application so that they can reach him almost at any time and any place. Even though this act and the technology of mobile application extend the time and load of his counselling services considerably, Kuo understands his services precisely as a response to the vulnerability faced by clients. From my fieldwork, I find that making oneself extensively available to their HIV positive clients via mobile technology is not uncommon among HIV service providers.
For HIV is a stigmatized disease and people with HIV suffer from being discriminated against, knowledge and practices such as confidentiality, voluntarism, care, and support are clearly at the core of the model of scaling up HIV testing and counselling proposed by the WHO: Any scaling up of HIV testing and counselling … should lead to general improvement in the medical ethics and efficacy surrounding current practice relating to HIV and any other conditions where informed consent, confidentiality, counselling and testing are at issue (WHO 2002, 23).
However, as the grounded work of NGO workers has demonstrated, articulating and practicing these principles to counter the vulnerability facing service clients (especially HIV positive individuals) is never a simple task. It involves the nitty-gritty work that strategically utilizes technologies and other materials as well as the management of time and space in a variety of circumstances. These localized HIV services manifest what I would call the materialization of vulnerability management.
The landscape of HIV service delivery in Taiwan did not undergo a differentiation by service practice, nevertheless. Since 2009, TCDC has been supporting HIV Pre-and Post-Test Counselling Workshops organized by NGOs for general HIV service providers' occupational training every year. Unlike some models of knowledge transmission elsewhere, in these workshops, NGO and community workers constitute a significant proportion of trainers, who share their work experiences and knowledge to other providers working at hospitals and governmental health departments. During my fieldwork in one of these workshops, an NGO worker offered me an explanation of this bottom-up institutionalization of knowledge and practices of HIV service delivery; she said that NGOs have long been working with minority and HIV positive communities and are able to accumulate experiences and knowledge by doing so. While this message is illuminating, I would add to that by contending that based on their experiences and knowledge, NGOs have striven to build a "channel" through which a particular set of practices could flow transnationally and materialize on the ground (Sassen 2000;2001).

Conclusion
As illustrated in this study, the material practices of HIV service delivery are rarely just locally determined, but instead are often informed by models of health care delivery proposed by global health institutions. Since the 2000s, WHO and UNAIDS have posited that ignorance of HIV status poses a challenge to HIV prevention and care; at the same time, they have endorsed a massive scale-up of testing and counselling as a model response. This approach has been influential transnationally. Aligning national HIV response with a global one, the Taiwanese government has expanded testing delivery by increasing the number of institutions offering services as well as the target numbers of services. Moreover, individuals tested for HIV with their identities known have been registered in a state management system and recruited into health care programs.
While the global model and national programs might have been designed with good intentions, a perspective from below reveals that the shifting materialities of HIV service delivery is seldom devoid of politics. I have demonstrated three ways in which practices of service delivery could make clients vulnerable. First, as testing services are increasingly being offered in nontraditional health care settings, the physical space in which testing is delivered matters with regard to confidentiality. Reporting a positive rapid test result to a client in a circumstance in which others are likely to overhear can be disrespectful. Second, despite increased testing provisions, limited counselling service delivery deprives clients of access to knowledge and recourses for HIV prevention and care. This could be particularly destructive when a client is informed of a positive test result without social support. Third, if unprepared, a positive and named test result and HIV registration in a state system could unexpectedly affect a client's life, given the prevailing discrimination against HIV positive individuals. In sum, the material practices of service delivery which are steered by particular sets of national and global priorities but are disconnected from local conditions and concerns can generate vulnerability faced by clients.
Being aware of the materialized forms of vulnerability, local NGOs articulate and promote an alternative set of service practices transnationally, yet in ways which are more responsive to local needs. For instance, when offering outreach services in public venues, NGO workers adjust the physical space and utilize testing technologies and material objects in ways that that client's confidentiality could be secured. Moreover, with the aid of mobile technology, they provide clients with extended counselling services, connect them with various local resources, and assist them in navigating potential vulnerability while planning for HIV registration.
As Gille argues, "local materialities [are] consequential for how globalization is experienced" (2014,162). With the case study on HIV service delivery in Taiwan, the article demonstrates how service clients encountering different materialized service delivery experience the globalization of HIV care models in fundamentally different ways. Furthermore, the parallel analysis of what, where, and how transnational practices of service delivery travel reveals that these practices in fact do not travel by themselves; rather, transnational practices only travel through locallyand politicallybuilt "channels" which are in place for them. The mutual constitution of global mobility and local fixity is well theorized in Sassen's (2000;2001) work. In this case, NGOs constitute the local fixity that channels and grounds global practices which could be adopted in response to the potential vulnerability facing service clients. Such channeling and grounding work has made it clear that the transnational material practices of HIV service delivery are not a fixed, monolithic whole, and could no longer be deemed apolitical.