LGBT ACTIVISM IN CHINA giving them tremendous control over many LGBT NGOs.As a result,activists'most important relationship is with the government agent,not the international donor. Domestic political conditions and transnational advocacy networks Having described the domestic context for LGBT activism,I now show how this helps us understand the nature of its linkages to the international community.To do so,it is instructive to revisit the concept of TANs and subject the boomerang model to the realities of the political economy of LGBT activism in China. The boomerang first anticipates significant blockage of domestic civil society. While the state is neither uniformly nor fully responsive in China,it has not blocked LGBT organizations completely.In fact,when working on issues like HIV/AIDS many have become an important government partner.Narrow channels of political opportunity have opened,giving activists reason to believe the state is fulfilling at least some of their needs (and perhaps hope of further opening in the future).As a result,an unexpected relationship emerges between state and society.To ensure sustainability and preserve gains (however limited and contingent they might be), activists avoid taking actions that might sour this relationship.This includes keeping ties with certain international actors (e.g.Open Society Institute,National Endowment for Democracy)and domestic NGOs to a minimum (especially those whose activities are seen as more antagonistic).Although LGBT activists foresee some benefit in building linkages,it does not necessarily behoove them to do so; benefits are likely outweighed by other costs,namely those that threaten their political opportunities.In essence,by giving activists limited opportunities to operate openly,the state places them in a 'bubble',insulating them from and limiting their ability to network with international actors.In addition,as I show in the next section, NGOs also exist in individual 'bubbles',which isolate them from other domestic groups. One important critique of the transnational advocacy network literature,which is highlighted by the articles in this collection,is its tendency to treat the state as a unitary actor.35 Domestic political structures,the complex relationships between levels of government and within agencies,are crucial for understanding the impact of transnational actors in any given polity.36 As shown above,the existence of LGBT activism in China is highly contingent upon local factors.This insight is not limited to China or even authoritarian contexts:Jalali shows that particular legislative practices in democratic states like India can affect the influence of transnational actors.57 Because local context matters,attention must be drawn to the diversity of interests within the state.Not all governments are similarly non-responsive or open.In Sichuan and Yunnan,activists report closer relationships with local government,largely because of the role that gay men's groups play in combating HIV/AIDS;the situation is 54.In some cases,LGBT activists mistakenly believe funding is from the government alone. 55.Wilson.'Introduction':and Wilson,'Seeking one's day in court'. 56.Stephen Krasner,'Power politics,institutions,and transnational relations',in T.Risse-Kappen,ed.,Bringing Transnational Relations Back In:Non-state Actors,Domestic Structures and International Institutions (New York: Cambridge University Press,1995).p.258:R.Lipschutz.Global Civil Sociery and Global Environmental Governance:The Politics of Nature From Place to Planet (Albany.NY:SUNY Press,1996).p.57. 57.Jalali.'International funding of NGOs in India',p.165. 855giving them tremendous control over many LGBT NGOs. As a result, activists’ most important relationship is with the government agent, not the international donor.54 Domestic political conditions and transnational advocacy networks Having described the domestic context for LGBT activism, I now show how this helps us understand the nature of its linkages to the international community. To do so, it is instructive to revisit the concept of TANs and subject the boomerang model to the realities of the political economy of LGBT activism in China. The boomerang first anticipates significant blockage of domestic civil society. While the state is neither uniformly nor fully responsive in China, it has not blocked LGBT organizations completely. In fact, when working on issues like HIV/AIDS, many have become an important government partner. Narrow channels of political opportunity have opened, giving activists reason to believe the state is fulfilling at least some of their needs (and perhaps hope of further opening in the future). As a result, an unexpected relationship emerges between state and society. To ensure sustainability and preserve gains (however limited and contingent they might be), activists avoid taking actions that might sour this relationship. This includes keeping ties with certain international actors (e.g. Open Society Institute, National Endowment for Democracy) and domestic NGOs to a minimum (especially those whose activities are seen as more antagonistic). Although LGBT activists foresee some benefit in building linkages, it does not necessarily behoove them to do so; benefits are likely outweighed by other costs, namely those that threaten their political opportunities. In essence, by giving activists limited opportunities to operate openly, the state places them in a ‘bubble’, insulating them from and limiting their ability to network with international actors. In addition, as I show in the next section, NGOs also exist in individual ‘bubbles’, which isolate them from other domestic groups. One important critique of the transnational advocacy network literature, which is highlighted by the articles in this collection, is its tendency to treat the state as a unitary actor.55 Domestic political structures, the complex relationships between levels of government and within agencies, are crucial for understanding the impact of transnational actors in any given polity.56 As shown above, the existence of LGBT activism in China is highly contingent upon local factors. This insight is not limited to China or even authoritarian contexts: Jalali shows that particular legislative practices in democratic states like India can affect the influence of transnational actors.57 Because local context matters, attention must be drawn to the diversity of interests within the state. Not all governments are similarly non-responsive or open. In Sichuan and Yunnan, activists report closer relationships with local government, largely because of the role that gay men’s groups play in combating HIV/AIDS; the situation is 54. In some cases, LGBT activists mistakenly believe funding is from the government alone. 55. Wilson, ‘Introduction’; and Wilson, ‘Seeking one’s day in court’. 56. Stephen Krasner, ‘Power politics, institutions, and transnational relations’, in T. Risse-Kappen, ed., Bringing Transnational Relations Back In: Non-state Actors, Domestic Structures and International Institutions (New York: Cambridge University Press, 1995), p. 258; R. Lipschutz, Global Civil Society and Global Environmental Governance: The Politics of Nature From Place to Planet (Albany, NY: SUNY Press, 1996), p. 57. 57. Jalali, ‘International funding of NGOs in India’, p. 165. LGBT ACTIVISM IN CHINA 855