Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

States and societies in the digital arena : ICT, state capacity, and political change in Asia Wand, Itay 2012

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2012_fall_wand_itay.pdf [ 1.53MB ]
Metadata
JSON: 24-1.0073175.json
JSON-LD: 24-1.0073175-ld.json
RDF/XML (Pretty): 24-1.0073175-rdf.xml
RDF/JSON: 24-1.0073175-rdf.json
Turtle: 24-1.0073175-turtle.txt
N-Triples: 24-1.0073175-rdf-ntriples.txt
Original Record: 24-1.0073175-source.json
Full Text
24-1.0073175-fulltext.txt
Citation
24-1.0073175.ris

Full Text

STATES	
  AND	
  SOCIETIES	
  IN	
  THE	
  DIGITAL	
  ARENA:	
   	
   ICT,	
  STATE	
  CAPACITY,	
  AND	
  POLITICAL	
  CHANGE	
  IN	
  ASIA	
  	
  	
  	
  by	
  	
  	
  ITAY	
  WAND	
  	
  B.Sc.	
  (Hons.)	
  Electrical	
  Engineering,	
  Queen’s	
  University,	
  1996	
  	
  	
  	
  	
  	
  A	
  THESIS	
  SUBMITTED	
  IN	
  PARTIAL	
  FULFILLMENT	
  OF	
  THE	
  REQUIREMENTS	
  FOR	
  THE	
  DEGREE	
  OF	
  	
  MASTER	
  OF	
  ARTS	
  	
  	
  in	
  	
  	
  THE	
  FACULTY	
  OF	
  GRADUATE	
  STUDIES	
  	
  (ASIA	
  PACIFIC	
  POLICY	
  STUDIES)	
  	
  	
  	
  	
  THE	
  UNIVERSITY	
  OF	
  BRITISH	
  COLUMBIA	
  	
  (Vancouver)	
  	
  	
  	
  	
  September	
  2012	
  	
  	
  ©	
  Itay	
  Wand,	
  2012	
   	
   ii	
   Abstract	
   	
   How does adoption of information communication technology (ICT) alter the balance of power between state and society in Asia? There is no question that these tools – the Internet, mobile phones, and social media – are transforming the political communications landscape across the region. Since political science views power as zero-sum, a central question is how its distribution is altered between digitally- strengthened states and digitally-empowered societal actors. On the one hand, societal actors are empowered through increased information access and dissemination, as well as decreased costs of mobilization and organization. At the same time, the state's digital capacity is greatly enhanced through increased information collection, monitoring, and control. This study hypothesizes that adoption of ICT in Asian states empowers societal actors over time enhancing non-electoral democratic processes subject to regime legitimacy and the digital state capacity governments build and apply. It first develops a theory for how ICT empowers both societal actors and states before testing this across Asian states through a quantitative analysis. The results suggest that Asian state policy determines whether and how ICT empowers societal actors and net political change. It then develops this policy concept through the lens of digital state capacity - how states control and manage digital information. Finally it conducts a qualitative analysis for China on the interaction of ICT adoption, regime legitimacy, and digital state capacity policy to determine net political change. The results demonstrate that while ICT adoption has strengthened the Chinese state through digital state capacity this has come at the loss of state control over a range of political issues. For these issues, the net result in China has been empowered-societal actors and enhanced transparency and accountability. 	
   iii	
   Table	
  of	
  Contents	
   	
   Abstract	
  ....................................................................................................................................................	
  ii	
   Table	
  of	
  Contents	
  ................................................................................................................................	
  iii	
   List	
  of	
  Tables	
  ..........................................................................................................................................	
  v	
   List	
  of	
  Figures	
  ........................................................................................................................................	
  vi	
   List	
  of	
  Abbreviations	
  ........................................................................................................................	
  vii	
   Acknowledgements	
  .............................................................................................................................	
  ix	
   Chapter	
  1	
  	
   Introduction:	
  The	
  Digital	
  in	
  Asian	
  State	
  and	
  Society	
  .......................................	
  1	
  1.1	
  Context:	
  Political	
  Change	
  and	
  Information	
  Communication	
  Technology	
  .....................................	
  1	
  1.2	
  Central	
  Puzzle	
  and	
  Argument	
  ..........................................................................................................................	
  2	
  1.3	
  Study	
  Roadmap	
  ......................................................................................................................................................	
  3	
   Chapter	
  2	
  	
   Alternative	
  Arguments:	
  Democracy,	
  Determinism,	
  and	
  Dictators	
  ..............	
  5	
  2.1	
  Introduction:	
  Snapshots	
  of	
  New	
  Technology	
  and	
  Political	
  Change	
  .................................................	
  5	
  2.2	
  Cyberutopianism:	
  Tweets	
  Over	
  Bullets	
  .......................................................................................................	
  6	
  2.3	
  Cyberskepticism:	
  Bullets	
  Through	
  Tweets	
  ................................................................................................	
  8	
  2.4	
  Dictators	
  and	
  Their	
  Dilemmas	
  .......................................................................................................................	
  10	
   Chapter	
  3	
  	
   Theory:	
  The	
  Digital	
  Dimension	
  of	
  Democratization	
  .....................................	
  12	
  3.1	
  From	
  Theory	
  to	
  Praxis	
  ......................................................................................................................................	
  12	
  3.2	
  Bias,	
  Assumption,	
  and	
  Norms:	
  Or,	
  Which	
  Facts	
  Matter	
  ......................................................................	
  13	
  3.3	
  The	
  Dependent	
  Variable:	
  Democracy	
  .........................................................................................................	
  16	
  3.4	
  The	
  Independent	
  Variable:	
  Information	
  Communication	
  Technology	
  ........................................	
  17	
  3.5	
  The	
  Intervening	
  Variables:	
  Digital	
  State	
  Capacity	
  and	
  Regime	
  Legitimacy	
  ...............................	
  18	
  3.6	
  Core	
  Theory	
  and	
  State	
  Typology	
  ..................................................................................................................	
  20	
  3.7	
  Social	
  Movement	
  Political	
  Opportunities:	
  Government	
  Under	
  an	
  Unblinking	
  Eye	
  ................	
  23	
  3.8	
  Social	
  Movement	
  Mobilization:	
  Social	
  Ties	
  Online	
  and	
  Off	
  ...............................................................	
  25	
  3.9	
  Social	
  Movement	
  Issues	
  Framing:	
  Contention	
  in	
  the	
  Public	
  Sphere	
  .............................................	
  26	
  3.10	
  Conclusion:	
  Framing,	
  Mobilization,	
  and	
  Political	
  Opportunities	
  in	
  Action	
  .............................	
  27	
   Chapter	
  4	
  	
   Quantitative	
  Analysis:	
  ICT	
  and	
  Democratization	
  in	
  Asia	
  .............................	
  28	
  4.1	
  Why	
  Model	
  ICT	
  and	
  Democracy?	
  ..................................................................................................................	
  28	
  4.2	
  Hypotheses,	
  Data,	
  and	
  Methodology	
  ...........................................................................................................	
  29	
  4.3	
  ICT:	
  Empty	
  Promise	
  or	
  Democratic	
  Indicator?	
  ......................................................................................	
  32	
  4.4	
  ICT	
  and	
  Asian	
  Democratization	
  Over	
  Time	
  ..............................................................................................	
  33	
  4.5	
  The	
  Missing	
  Piece?	
  Online	
  Censorship	
  as	
  a	
  Proxy	
  for	
  Digital	
  State	
  Capacity	
  ............................	
  37	
  4.6	
  Completing	
  the	
  Puzzle:	
  Communications	
  and	
  State	
  Capacity	
  Combined	
  ....................................	
  40	
  4.7	
  Results:	
  The	
  Digital	
  Asian	
  State	
  Strikes	
  Back	
  ..........................................................................................	
  43	
   Chapter	
  5	
  	
   Digital	
  State	
  Capacity:	
  Bringing	
  the	
  Asian	
  State	
  Back	
  In	
  ..............................	
  44	
  5.1	
  Whither	
  the	
  Digital	
  Asian	
  State?	
  ...................................................................................................................	
  44	
  5.2	
  Conceptualizing	
  DSC:	
  One	
  Part	
  Orientation,	
  Two	
  Parts	
  Ability,	
  and	
  Mix	
  ....................................	
  45	
  5.3	
  Locating	
  Agency:	
  the	
  State	
  and	
  Its	
  Intermediaries	
  ...............................................................................	
  47	
  5.4	
  The	
  Sum	
  of	
  Its	
  Parts:	
  A	
  Taxonomy	
  of	
  Digital	
  State	
  Capacity	
  ............................................................	
  49	
  5.5	
  Looking	
  Under	
  the	
  Hood:	
  The	
  Infrastructure	
  Dimension	
  ..................................................................	
  51	
  5.6	
  Government	
  Knows	
  Best:	
  The	
  Censorship	
  Dimension	
  .......................................................................	
  52	
   	
   iv	
   5.7	
  Regulating	
  Cyberspace	
  and	
  Real	
  Space:	
  The	
  Legal	
  Dimension	
  .......................................................	
  55	
  5.8	
  Squeezing	
  Online	
  Dissent:	
  The	
  Coercion	
  Dimension	
  ...........................................................................	
  56	
  5.9	
  DSC	
  Policy	
  Suites:	
  More	
  Than	
  The	
  Sum	
  of	
  Its	
  Parts	
  .............................................................................	
  57	
   Chapter	
  6	
  	
   Qualitative	
  Analysis:	
  Chinese	
  Political	
  Space	
  a	
  Click	
  at	
  a	
  Time	
  .................	
  59	
  6.1	
  Introduction	
  –	
  Why	
  China	
  and	
  Why	
  Now?	
  ...............................................................................................	
  59	
  6.2	
  Mobile	
  Nation	
  Wired	
  Dragon	
  .........................................................................................................................	
  61	
  6.3	
  Legitimacy	
  and	
  Policy	
  in	
  China	
  ......................................................................................................................	
  62	
  6.4	
  The	
  Tiger’s	
  Teeth:	
  Digital	
  State	
  Capacity	
  in	
  China	
  ................................................................................	
  64	
  6.5	
  Falun	
  Gong:	
  State	
  Capacity	
  Eclipses	
  a	
  Social	
  Movement	
  ....................................................................	
  67	
  6.6	
  ICT,	
  DSC,	
  and	
  Societal	
  Actors	
  in	
  Action:	
  Wukan	
  2011	
  and	
  Other	
  Cases	
  ......................................	
  69	
  6.7	
  Epilogue:	
  Claiming	
  Political	
  Rights	
  One	
  Click	
  at	
  A	
  Time	
  ....................................................................	
  75	
   Chapter	
  7	
   Conclusion:	
  Research	
  Significance	
  and	
  Implications	
  .....................................	
  78	
  7.1	
  Significance	
  and	
  Overview	
  ..............................................................................................................................	
  78	
  7.2	
  Research	
  Applications:	
  What	
  This	
  All	
  Means	
  ..........................................................................................	
  79	
  7.2	
  Future	
  Research	
  ...................................................................................................................................................	
  81	
   Bibliography	
  ........................................................................................................................................	
  82	
   Appendices	
  ........................................................................................................................................	
  104	
   	
   	
   	
   	
   v	
   List	
  of	
  Tables	
  	
  Table	
  3.1:	
  Ideal	
  State	
  Types………………………………………………………….…………………………....….21	
  	
  Table	
  4.1:	
  Asian	
  Democratization	
  Models	
  -­‐	
  Time	
  Series	
  Cross	
  Section	
  Analysis	
  Results……………………………………………………………………………………………………………...…….…...34	
  	
  Table	
  4.2:	
  Asian	
  Democracy,	
  ICT,	
  and	
  Online	
  Censorship	
  (Multivariate	
  Analysis)……...…….38	
  	
  Table	
  4.3:	
  Asian	
  Democracy,	
  ICT,	
  and	
  Online	
  Censorship	
  Over	
  Time	
  (Time	
  Series	
  Cross	
  Sectional	
  Analysis)…………………………………………………………………………………………...…………..40	
  	
  Table	
  5.1:	
  Digital	
  State	
  Capacity	
  Typology…………………………………………………..…………………50	
   	
   	
  	
   	
   	
   vi	
   List	
  of	
  Figures	
  	
  Figure	
  3.1:	
  Conditions	
  for	
  Maximal	
  Poltical	
  Change…………………………………………………....….21	
  	
  Figure	
  4.1:	
  Communication	
  vs.	
  Economic/Social	
  Factors	
  as	
  Democratic	
  Indicators…………32	
  	
  Figure	
  4.2:	
  Correlation	
  between	
  ICT	
  and	
  Democracy	
  in	
  Asia:	
  Regional	
  (OLS)	
  vs.	
  State-­‐level	
  (fixed	
  effects)………………………………………………………………………………………………………...…….36	
  	
  Figure	
  4.3:	
  Non-­‐Censoring	
  States:	
  ICT	
  and	
  Democracy	
  in	
  Asia………………….………...…………..41	
  	
  Figure	
  4.4:	
  Censoring	
  States:	
  ICT	
  and	
  Democracy	
  in	
  Asia……………...……………..…………………42	
  	
  Figure	
  6.1:	
  ICT	
  /	
  Legitimacy	
  /	
  DSC	
  Levels	
  in	
  China……………………………………………………..….59	
  	
  Figure	
  6.2:	
  Digital	
  Contention	
  Space	
  in	
  China…………………………………………………………...……73	
  	
   	
   	
   vii	
   List	
  of	
  Abbreviations	
  	
    BBS – Bulletin Board Service  CCP – Chinese Communist Party  CDP – Chinese Democratic Party  CDT – China Digital Times  CINIC – China Internet Network Information Center  CPJ – The Committee to Protect Journalists  DDOS – Distributed Denial of Service Attack  DNS – Domain Name System  DSC – Digital State Capacity  FHI – Freedom House Index  ICP – Internet Content Provider  ICT – Information Communication Technology  IP – Internet Protocol  ISP – Internet Service Provider  ITU – The International Telecommunications Union  IXP – Internet Exchange Point  MII – Ministry of Information Industry  NYT – New York Times  PLA – People’s Liberation Army  RSS – Really Simple Syndication  SARS – Severe Acute Respiratory Syndrome  SOE – State Owned Enterprise  	
  viii	
   UGC – User Generated Content  UNDP – United Nations Development Program  UNESCO – United Nations Educational, Scientific and Cultural Organization  URL – Universal Resources Locator  VOIP – Voice Over IP  WB – World Bank  WHO – World Health Organization  WTO – World Trade Organization 	
   	
   	
   ix	
   Acknowledgements	
  	
   This thesis is the culmination of a multi-year personal journey. The transition from professional to academic pursuits has not been easy. But it has been a path of challenge, learning, and growth for which I am grateful. It would not have been possible without the help, support, and understanding of mentors, colleagues, and family.  I could not have been more fortunate to have Dr. Paul Evans, Dr. Yves Tiberghien, and Dr. Julian Dierkes as my thesis committee. In addition to crucial insights from a political scientist's lens, Dr. Tiberghien has consistently offered invaluable support and advise which I greatly value and deeply appreciate. Dr. Dierkes' encouragement over the past years, his feedback, and ever-open door have meant a great deal to me. I would not have done this thesis nor been in the UBC MAAPPS program were it not for these two individuals. Dr. Evans' time and mentorship throughout this program have proven formative in my professional, academic, and personal directions. Their support has been pivotal and I thank them.  Several other professors have generously offered time and input into this thesis. Dr. Pitman Potter helped conceptualize important parts and I have gained tremendously from his periodic guidance. Dr. Fred Cutler's patience in explaining quantitative methods was a crucial building block for the analysis performed. Dr. Jessica Main's stimulating exposition of media's effects was fundamental for the theory developed.  Friends and colleagues, some co-navigating the waters of communications in Asia, have been extremely helpful throughout this project. Josh Rudolph's finely honed review skills were put to the test editing this thesis; Byron Hauck's sharp critique greatly strengthened the work; and Benjamin Tipton's consistently-solid feedback and support proved invaluable. I benefited considerably from their generous help. 	
   x	
    Neither this thesis nor a return to university would have happened nor been possible had it not been for Noa. She has pushed, cajoled, advocated, and encouraged me to pursue academic studies and supported me unfailingly throughout. There was a race between this thesis and the arrival of our daughter, Maya. The thesis lost and I could not have been happier for it. This work is fundamentally about my pursuit of deeper challenge and fulfillment. In this, I dedicate it to her. 	
   	
   1	
   Chapter	
  1	
  	
   Introduction:	
  The	
  Digital	
  in	
  Asian	
  State	
  and	
  Society	
  	
    	
   1.1	
  Context:	
  Political	
  Change	
  and	
  Information	
  Communication	
  Technology	
    The world of communication has changed. Progressive waves of new technology have shortened distances and compressed time. As the tools of information communication technology (ICT) become ubiquitous, information – its access, production, and distribution – is being democratized. 2011 witnessed a wave of social upheavals across the globe difficult to imagine in an antediluvian era of state information monopolies. Borders are no longer hermetic to information. Moreover, ICT has transformed social movement mobilization. The podium is now online. Beyond its social impacts, ICT is driving economic growth and globalization. Manufacturing and service economies thrive on underlying communications infrastructure, tech-savvy workforces, and unrestricted information flow. Here the social and economic become political. States seeking rapid sustained economic development must adopt and even promote ICT usage despite the resulting diffusion in voice and control. For non-democratic Asian states this represents a tectonic shift in the political communications landscape, and states have responded by developing a new capacity to control digital information. ICT in Asia is not a story of revolution, regime transition, nor Western democratization. Both societal actors seeking political change and the regimes they contend with have been empowered by these technologies. The key question is which side of the roster – state or society – has gained the most? If power is zero sum, how has its distribution changed?  No region demonstrates the conditions conducive to this transformation – rapid economic growth, massive social change, high ICT adoption – as does Asia. The shift in economic and political power from the Atlantic to the Pacific has coincided with a period 	
   	
   2	
   of global hegemonic transition, from U.S.-dominated to a multi-polar world. But economic development is not easily decoupled from social and political change, and the future political forms of Asia’s growing states – both democratic and non – remains an open question. Some governments, notably the U.S., Canada, and other G8 states,1 see ICT as a democratizing force favouring societal actors, and its promotion a policy objective. Conversely other states, among them China, Vietnam, and Burma, see ICT as a new policy pillar of information control. Both policy sides rest on shaky ground lacking theoretical foundations and evidence. Scholarship in this embryonic field is playing catch-up to current events. In this study, I jump into this fray aiming to offer theory and evidence on the question of ICT and political change in Asia. 1.2	
  Central	
  Puzzle	
  and	
  Argument	
  	
   To do so I pose the following main research question. How does ICT alter the power balance in Asian state-society relations? Two corollary questions emerge. Why does ICT appear to favour societal actors in some states and not others? Finally, how are non- democratic states responding to these changes? In this study, I offer a theoretical framework to address these questions.  Political change is a complex and context-contingent process specific to each state’s unique historical and political trajectory. Nevertheless, this study resides within scholarship including Lipset (1959) and others that aim to discern macro trends in political change and democratization. I argue that ICT is now a central factor in political change in low legitimacy non-democratic states. ICT alters the political opportunities, mobilization potential, and issue framing that societal actors face. But crucially, this is 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   1 United States Congress 2010c, DFAIT 2007, G8 2011. 	
   	
   3	
   contingent on the level of regime legitimacy and how states choose to control information in response. They do so by developing digital state capacity (DSC), or the infrastructural, censorship, regulatory, and coercive capability to manage digital information.2 This is not a story of regime transition or adoption of Western democracy. Rather, any resulting democratization in low-legitimacy states is that of entrenchment and enhancement through underlying non-electoral mechanisms that increase citizen representation, as well as government accountability and transparency. ICT plays a central role in this transformation, and far from the headlines of explosive revolutions, does so through societal and political processes that are time-lagged, slow-moving, and long term. ICT is not a sole factor here, but given high adoption levels and low regime legitimacy, it becomes a necessary, albeit not sufficient condition. I argue that political change and contention now carry digital dimensions. The hypothesis I propose is that ICT plays a role in shifting the balance of state-society relations contingent on the legitimacy of the regime and the level of DSC governments wield in response. 	
   1.3	
  Study	
  Roadmap	
  	
   This study proceeds as follows. Chapter 2 outlines opposing poles in scholarship on the question of ICT and democratization. While analytically simplified, these theories have directly influenced policy. They are the Cyberutopian school seeing direct causality between ICT and empowered societal actors and the Cyberskeptics for which these tools strengthen primarily non-democratic regimes. The theory in chapter 3 builds on the 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   2 This	
  study	
  focuses	
  primarily	
  on	
  how	
  non-­‐democratic	
  states	
  use	
  DSC.	
  Most	
  states,	
  including	
  very	
  democratic	
  ones,	
  are	
  developing	
  and	
  applying	
  DSC	
  to	
  their	
  respective	
  policy	
  needs.	
   	
   	
   4	
   third school that ‘weighs the scales’3 between societal and state empowerment to determine net power shift. A theory is constructed for the effects of ICT on both societal and state actors given regime legitimacy level and using a social movements framework. Chapter 4 tests these contending theories through a quantitative analysis. It constructs a model of democratic change in Asia that includes ICT as an explanatory variable and DSC as an intervening variable. It cannot and does not demonstrate causality nor does it include a legitimacy variable.4 But the results demonstrate that while there is correlation between ICT and democracy level across Asia, individual states do not necessarily follow this relationship. It further demonstrates that DSC level – the information control policy of the state – is a key factor in the resulting ICT/democracy relationship. With the role of the state now demonstrated, chapter 5 investigates and develops the DSC concept. It outlines how Asian states control information through policy suites combining infrastructure, censorship, regulations, and coercion. Chapter 6 now applies the theory developed, the quantitative demonstration of ICT/DSC in action, and the DSC concept to China. China offers an example of high ICT adoption, high DSC, and high regime legitimacy. This chapter describes Chinese DSC power through the example of the Falun Gong movement before demonstrating ICT/DSC interaction at the regional level for the recent protests in Wukan village in Guangdong and other key examples. Chapter 7 concludes with a summary of results, policy implications, and suggestions for further research. 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   3 Drezner 2010.	
   4	
  This was out of scope due to time constraints and is a recommendation for future research.	
   	
   	
   5	
   Chapter	
  2	
  	
   Alternative	
  Arguments:	
  Democracy,	
  Determinism,	
  and	
   Dictators	
   	
  	
    “The 21st-century equivalent of the Berlin wall is a cyberbarrier.”1	
    “Technology is neither good nor bad; nor is it neutral.”2 - Kranzberg’s First Law	
   	
   2.1	
  Introduction:	
  Snapshots	
  of	
  New	
  Technology	
  and	
  Political	
  Change	
    Does ICT empower societal actors or oppressive regimes? Answers to this question rest on where and how agency is assigned. One end sees technology as emancipatory and liberating, while the other as hegemonic and regime-oriented. Nor is this debate a new one. Consider the following quote from the Economist in 2011. “After decades of simmering discontent a new form of media gives opponents of an authoritarian regime a way to express their views, register their solidarity and co-ordinate their actions. The protesters’ message spreads virally through social networks, making it impossible to suppress and highlighting the extent of public support for revolution.” While this technology-as-emancipatory example could easily apply to modern-day social upheavals, it is in fact a description of Martin Luther’s usage of the printing press.3 In contrast, an example of technology as regime tool is the following warning message received by Iranian citizens after the 2009 election protests: 4 “Dear citizen, according to received information, you have been influenced by the destabilizing propaganda which 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   1 Kristof 2009. 2 Kranzberg 1986, p545; Kranzberg discusses several perennial social laws for technology. 3 The printing press proved central to the success of Luther’s movement in its production, distribution, and publication speed (Anderson 2006). 4 Quoted from Morozov (2012, p.11), but see Aday et al. (2010) for a detailed account. 	
   	
   6	
   the media affiliated with foreign countries has been disseminating. In case of any illegal action and contact with the foreign media, you will be charged as a criminal.”  This chapter explores the state of the debate on ICT and democratization. This interdisciplinary question represents an emerging academic field with as yet insufficient theory and empirical evidence.5 Three schools of thought are discernable. This chapter first explores the ‘Cyberutopian camp’, so called because it sees direct causality between ICT and democratization, before crossing the aisle to the ‘Cyberskeptics’, who question this link and, in some cases, reverse it towards counter-democratization. A discussion of the Dictator’s Dilemma theory (Kedzie 1997) spanning both concludes the chapter.6 I discuss the third school, on which the theory for this study rests, in the following chapter. 2.2	
  Cyberutopianism:	
  Tweets	
  Over	
  Bullets7	
  	
   Cyberutopianism sees causality from ICT to societal over state empowerment.8 Its two- pronged argument is premised on the instrumental benefits that communication tools offer in organizing and mobilizing citizens, in addition to the diffusion in the political economy of media that these tools entail. In this lens, Twitter, Facebook, and other ICTs were catalytic in the events of the 2011 Arab Spring, as well as other recent social 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   5 Aday et al. (2010); But deeper research is emerging: Diebert et al. (2008, 2010, 2011); Howard (2011). 6	
  Originally coined by Kedzie (1997), the Dictator’s Dilemma concept is central to recent US foreign policy rhetoric (Quinn 2011; Sheridan,2011). 7 This borrows from the ‘Twitter Revolution’ popular, yet over-simplistic, 2011/2012 media meme; see Synovitz (2009) for Moldova, Kristof’s (2009) ‘protesters firing tweets’ in Iran, and Gladwell’s (2010) ‘Why the Revolution Will Not Be Tweeted.’ 8	
  Diamond, 2010; Kahn and Kellner 2005; Shirky 2008; Groshek 2009; Hermans 2008; Said (2003, p.xxix): “We are today abetted by the enormously encouraging democratic field of cyberspace, open to all users in ways undreamed of by earlier generations either of tyrants or of orthodoxies.” 	
   	
   7	
   upheavals.9 Indeed, transaction costs for social mobilization have plummeted with recent ICT tools. It is now faster, cheaper, and easier to communicate and coordinate large numbers of individuals towards social, economic, and political ends.10 Information monopolies and asymmetries have been reduced as aggregate information availability has increased. This is epitomized by ‘information cascades’, smart mobs, and China’s ‘human flesh search engine’ phenomena.11 ICT diffusion here favours distributed tech- savvy citizens over centralized, lumbering regimes.12  Second, Cyberutopians argue that, in addition to its immediate instrumental benefits, ICT rewires the media ecosystem, bypassing traditional gatekeepers in favour of “the people formerly known as the audience.”13 In this view, the monopoly of power to reach a mass audience, once reserved for governments and large corporations (and predicated on communication economies of scale), has been shattered. Diffusion of media power has placed a metaphorical printing press in each wired citizen’s hands.14 But Cyberutopianism is not without its critics.  Few dismiss the technology-induced structural transformations media is undergoing and ICT has, in fact, facilitated social organization and mobilization. But 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   9	
  For media examples, see Kristof (2009) and Synovitz (2009). 10	
  Drezner 2010; Diebert 2000; The seminal work on transaction costs and organization is Coase (1937). 11	
  ‘Iinformation cascades’ occur when information creates tipping points towards mass mobilization. See Lohmann (1994); See also Bikhchandani et al. (1998) for an econometric approach. Smart mobs are “large, loose coalitions of citizens … able to use the Internet and related technologies to organize themselves with breathtaking speed” (Hindman 2009, p.10); also Rheingold (2002). The dark side of this mobilization (see also Hermans, 2008) is exemplified by China’s human flesh search engine (rén ròu sōusuǒ - 人肉搜索), is an online collaborative (or vigilante) approach to identifying and flaming (punishing) individuals who have committed social or moral transgressions that has led to several deaths and new legislation (Liang and Lu 2010; Chao 2011). 12	
  Kahn and Kellner (2005), for example, offer a succinct empanicpatory lens on digital communications. 13	
  This phrase is credited to Rosen (2006) to indicate the shift to participant media consumers. It refers to a transformation from passive information consumption to active production. See also Benkler (2006). 14 See Shirky (2008). But also Chomsky and Herman (2002) for a critique of this perspective. 	
   	
   8	
   Cyberutopianism has been attacked as technologically deterministic,15 seeing new media as inherently and teleologically democratizing. In this critique, these tools are but the latest fetish in a diachronic lineage of ostensibly liberating technologies from the telegraph to the telephone to the television.16  In Gladwell’s (2010) words, “Social media can’t provide what social change has always required.”  Perhaps a sharper arrow in the criticism quiver, however, is assignment of agency. The outcome of technology usage - by whom and for what – is not inherent. In fact, states have been ready consumers, adopters, and adaptors of ICT to their own, and often non-democratic, ends.17 But Cyberutopians foreground and assign agency to only societal, and not state, actors. Finally, this school fails to explain the absence of democratization in states with high adoption levels, such as Singapore. The Cyberskeptic camp offers a response.	
   2.3	
  Cyberskepticism18:	
  Bullets	
  Through	
  Tweets	
  	
   In contrast to the emancipatory visions of the Cyberutopians, Cyberskeptics do not view ICT as inherently liberalizing, with some emphasizing instead regime empowerment.19 This theory rests on the primarily weak social ties that ICT promotes, as well as the growing array of non-democratic regimes’ information control capabilities. First, ICT, in its ability to span time and space, does promote an aggregate increase in social ties. Cyberskeptics point out that these are primarily weak ties and not those necessary for 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   15	
  Technological determinism is defined as “the belief that new technologies have an intrinsic, autonomous power to shape and transform society” (Jenkins and Thorburn  2004, p.5).	
   16	
  Morozov 2012. 17 Chapter 5 demonstrates this, but see Diebert et al. (2011), Stockman and Gallagher (2011), Kalathil and Boas (2003), and Keen (2009). 18	
  The terms ‘Cyberrealists’ or ‘instrumentalists’ have also been applied. 19	
  Morozov (2012, 2011, 2011b, 2011c); Bueno de Mesquita & Downs (2005); Kalathil and Boas (2003); 	
   	
   9	
   the high-cost, high-risk protest and activism that political change necessitates. 20 Furthermore, weak-tie online support for political action - or slactivism - tends to be a mile wide but an inch deep in an online environment saturated with distraction and entertainment.21But the second dimension of cyberskepticism is a dystopian view of how authoritarian states are strengthened through ICT. Non-democratic states have indeed leveraged these tools to develop information control bureaucracies. These allow monitoring dissidents, attacking opposition groups, and digitally manipulating public opinion, among other capabilities. Underlying this is the non-democratic state’s evolving ability to separate provision of ‘coordination goods’ - those necessary for political mobilization - from other public goods (railways and hospitals).22 However, as with cyberutopiansim, cyberskepticism is not without detractors.23  While ICT may indeed facilitate primarily weak social ties, an aggregate increase in tie density can be transformative in itself. The quantitative, beyond a threshold, can become qualitative.24 This has not only been the lesson of large-scale collaborative efforts such as Wikipedia, but also of political processes, such as campaign mobilization 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   20 Gladwell 2010. The argument is based on McAdam’s (1986) model of low vs. high risk/cost activism and the latter’s basis on ‘pull’ factors (i.e. the connection of a potential participant to a group) over ‘attitudinal’ factors (i.e. how close a participant’s values match the group. In other words, strong links trump overlap in values and identity and is thus the crucial factor for organized activism. 21 Morozov (2012) coined the term slactivism, but goes further in asserting that some states, such as Russia, actually promote superficial entertainment to counter possible political online activities. 22 Bueno de Mesquita and Downs (2005) and Bueno de Mesquita and Smith (2010). This model connects state revenue source to public good provision. Some non-democratic states need to provide coordination goods (and other public goods) for economic development. But these states are learning to decouple the two types of goods netting economic development without the mobilization risk. 23 A good overview of detractors from a recent media perspective is Rosen (2012). 24 See Shirky’s (2008) on the transformative nature of large, wired populations. The argument, reminiscent of McLuhan (1968), is that a large initial increase in social tie density inexorably leads to new patterns of information flow that carry potential (and later) political consequences. 	
   	
   10	
   and financing.25 Furthermore, large participant numbers spread not only action cost, but in mass mobilization, also risk, crucial to political contention.26 Finally, while non- democratic states have created powerful digital information control regimes, cyberskeptics repeat the agency fallacy of the cyberutopians. Namely, they place agency solely on the side of the state. This thus fails to explain cases of political change and democratization where ICT proved central, such as that ending the rule of Estrada in the Philippines in 2001, as well as more recent examples from the Arab Spring. 27 Astride the two camps just outlined sits the Dictator’s Dilemma theory. 2.4	
  Dictators	
  and	
  Their	
  Dilemmas	
  	
   The Dictator’s Dilemma theory explains why non-democratic states choose to adopt ICT despite its putative liberalizing effects. 28  It is premised on theories that trace democratization from economic development.29 In this view, economic development and globalization necessitate adoption of information technology infrastructure and ICT.30 But since these diffuse control of communications and media – both coordination goods – states incur a corresponding loss of control over information flow. Government 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   25	
  For an example of transformative aggregate low-cost impact in the 2008 Obama presidential campaign see Cogburn and Espinoza-Vasquez (2011) 26	
  Example from the 2011 Jasmine Revolutions include mobilization in Tunisia and Egypt, where tipping points were partly reached through pervasive mass participation. 27	
  See Rheingold (2002), McCargo (2003), Kahn and Kellner (2005). But other notable examples connecting ICT and political change include the South Korean elections of 2002 and Spain in 2004. See also Zuckerman (2007) and The Economist (2006). 28	
  Kedzie (1997) originated the term; See Hachigan (2002) on application to East Asian one-party states. 29	
  There is a long history of scholarship connecting economic to political development, from Lipset’s (1959) modernization theory and Boix and Stokes’ (2003) update on economic growth leading to democracy; to Przeworski et al.’s (2000) view that economic growth does not lead to democratization but rather prevents de-democratization; to Acemoglu and Robinson’s (2006) view of inequality as the key mediating factor. The Dictator’s Dilemma theory is a child of this scholarship. 30	
  Not all economic growth is the same. Valuable models here are Bueno de Mesquita’s (1995, 2010) theories linking state revenue to public good type provision, as well as his (2000) ‘selectorate’ model on regime type and revenue sources. A well-known example is Friedman’s (2006) First Law of Petropolitics. Nevertheless, the Dictator’s Dilemma fits well with the developmental state model (Evans 1995). 	
   	
   11	
   monopolies of media and propaganda are shattered, as is exclusive ownership of the political communications agenda. These states are thus trapped in a Dictator’s Dilemma: economic development gains and subsequent regime legitimacy come at ever- increasing cost (and decreasing success) of information control.  Both schools outlined here offer one-sided views in a field lacking nuanced theory and empirical evidence. The following chapter outlines a third school that assigns agency at both societal and state levels. It develops this further towards an alternative theory and hypothesis on the question of ICT and political change.	
   	
   	
   12	
   Chapter	
  3	
  	
   Theory:	
  The	
  Digital	
  Dimension	
  of	
  Democratization	
   	
   “On Facebook and Twitter, journalists posted on-the-spot reports… protesters coordinated their next moves…millions answered in real time… Then the government pulled the plug.”1   - US Secretary of State Hillary Clinton, on Egypt’s 2011 revolution	
   	
   3.1	
  From	
  Theory	
  to	
  Praxis	
    A tsunami of digital communications is washing over Asian states, transforming their social, economic, and political fabric in its wake. Will these technologies empower societal actors or non-democratic states? Are they liberating, hegemonic, or both? As per the previous chapter, the debate is vociferous, and in in the past this question may have remained in literature. But it is no longer limited to scholarship. How governments answer this theoretical question has direct policy implications. States have taken sides, with some actively promoting ICT for its purported democratizing impact and others building tomorrow’s information dykes and dams. I believe the very right to freedom of expression, increasingly manifested online,2 hangs in the balance.  This chapter develops the theoretical bricks and mortar for the subsequent quantitative and qualitative analyses. While the previous chapter outlined two poles of the debate on ICT and state vs. societal empowerment, this chapter builds on a third approach to construct a theoretical framework. This middle road is inhabited by scholars who see agency on both sides of the state-society arena with net political change a function of the interplay of the two (Drezner 2010, Howard 2011, Meier 2007, 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   1	
  Excerpt from Clinton’s speech during the Egyptian Arab Spring (Memmort 2011). 2	
  Some states (Finland, Spain, France) have redefined online freedom of expression as a positive/active right incumbent on the state to provide, premised on a definition of online access as a requirement for full citizen participation (Dutton 2010). 	
   	
   13	
   Meier and Stodden 2009). Howard (2011) offers perhaps the most comprehensive study of ICT and democracy, focused on Muslim states. He sees societal ICT empowerment through civil society, new media actors, and transnational information diffusion, and state empowerment through digital surveillance and censorship tools, as well as control of the public sphere. For Drezner (2010), societal actors may be the main beneficiaries of ICT, but non-democratic states can control political communications albeit at increasing overhead and not hermetically.  Building on this and other literature, I argue that ICT empowers societal actors dependent on regime legitimacy, penetration rates, and the level of information control the state applies. ICT is never a sole factor in political change, but given high adoption rates and low regime legitimacy, it plays a central role by altering the communications landscape. Where states lack legitimacy and capacity, ICT empowers social movements.3  After positioning this model in its metatheoretical place, this chapter outlines the main variables for the study: democracy, ICT, legitimacy, and digital state capacity. Finally, it develops a theory for their interaction against a social movement framework. 3.2	
  Bias,	
  Assumption,	
  and	
  Norms:	
  Or,	
  Which	
  Facts	
  Matter4	
  	
   Prior to delving into theory, the underlying normative and metatheoretical foundations and their limitations are exposed. To be explicit, the animus fuelling this study is the normative role of information flow in modern societies. Freedom to access and produce 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   3 I focus in this study on societal actors with interest towards political change. Of course, societal actors may use ICT for many ends and not all societal actors are interested in political change. See for example Broadhurst and Chang (2012) for criminal usage in Asia.	
   4	
  “Theories do not simply explain or predict, they tell us what possibilities exist for human action and intervention…they define…our ethical and practical horizon (Smith 1996, p.13). 	
   	
   14	
   information carries both inherent and instrumental benefits.5 The former underlies, for example, the UN Bill of Rights article on Freedom of Expression as intrinsic and inalienable to human beings. The latter drives the ability to make informed choices, whether in a market or the political sphere.6 While freedom of expression is never absolute, its limitations come at the cost of both its inherent benefit and instrumental role in reducing information asymmetries and transaction costs.7  At the extreme, centralized control of information can have catastrophic results. Limitations on freedom of expression and information distortion lie at the heart of modern famines8 and failed ‘high-modernist’ attempts at large-scale societal transformations.9 If information is a public good,10 this study’s normative bias sides with its widest possible dissemination.  This study aims for a general model of ICT and democratization in Asia and thus rests on scholarship going back to Lipset (1959) on the relationship between economic, social, and political development. Any such macro theory has limitations in application to individual state cases. State society relations and democratic change are complex, time-sensitive, and context-dependent processes. Generalizable theory, however, offers a starting point for case analysis. In this respect, the approach here is post-positivist.11 In tackling both quantitative variables, such as literacy levels, and subjective concepts, such as legitimacy and public sphere boundaries, the ontological framework is material 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   5	
  This twin role is based on Sen’s (1999, 1999b) thesis. 6	
  The article is part of the International Covenant on Economic, Social and Cultural Rights (ICESR) and reads, “Everyone has the right to freedom of opinion and expression; this right includes freedom to hold opinions without interference and to seek, receive and impart information and ideas through any media and regardless of frontiers” (United Nations 1966, art.19; 1948). 7 See, for example Stiglitz 2001; Sykes 2006; and Cottier and Khorana 2009. 8 This is central to Sen’s (1982) argument on exchange entitlement relations and their role in famine. 9 During China’s Great Leap Forward, the purposeful inflation of production figures up the chain of command partly led, with other factors, to the ensuing famine (Fenby 2009). 10	
  Rather, an inherent and instrumental “double public good” (Farber 1991, p.563). 11	
  Smith 1996; Marsh and Furlong 2002. 	
   	
   15	
   first and social second. Levels of ICT penetration and GDP matter, as do citizen views of regime legitimacy and state capacity. 12  Further, a focus on contestation between states and social movements, and the state as an arena of public politics, belies the complex interconnections and mutually-constituted nature of these actors. 13  A reductionist ontology and empirical epistemology in the quantitative analysis is later expanded to an holistic ontology, and interpretive epistemology in the qualitative section. Empiricism is sufficient for the former, but socially constructed and subjectively negotiated concepts such as legitimacy and representation14 require an interpretive lens.15  Agency here is at two primary levels: individual states and societal actors within their territories. This analytical simplification excludes myriad actors with roles in the political communications arena. The mass media of newspapers, television, and radio, as well as private corporations, are not prime movers here. They are addressed primarily as intermediaries in the ecosystem of state communication policy.16 The state here is a Weberian “organization that monopolizes legitimate means of violence within a territory.”17 But while early definitions focus on the state as a war-making machine, the concept here is expanded to include the state’s role in economic transformation, as well 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   12 Cummings and Nørgaard (2004, p.687), for example, introduce a concept of ideational capacity as “the degree to which the state – its actors, role, and policies – is legitimated and embedded in state institutions.” Material power is subject to its ideational counterpart. Some scholars go further to assert that the material only matters through its subjective interpretation. See Wendt (2003) for the ideational composition of the material. See also Price and Reus-Smit (1998). 13	
  Checkel (1998) outlines in more detail the implications of such an ontological view. 14	
  Li and O’Brien (2008), for example present the role of authority perceptions in Chinese rural protests. 15	
  Understanding social structures requires an interpretive epistemology. Bevir and Rhodes 2002, Jackson and Sørensen 2003, Marsh and Furlong 2002. 16	
  An alternative is an approach that foregrounds the media (Zhao 2008, Howard 2011, McCargo 2003); For an example of the motivations of corporations, see Mueller (2011). 17	
  Weber 1965, p.1. For a superset definition, see Tilly (2007, p.11). 	
   	
   16	
   as rule making and its enforcement.18 States in this study are both corporatist actors and arenas of political contestation inhabited by social movements, or “groups of challengers…[engaged] in mutual claim-making with powerholders.”19 3.3	
  The	
  Dependent	
  Variable:	
  Democracy	
  	
   This section defines the four main variables: democracy, ICT, legitimacy, and digital state capacity (DSC), dependent, independent, and intervening variables, respectively. Since DSC is a new concept, however, it is developed fully in chapter 5. Democracy is a semantically-overloaded term, with a descriptive definition often eclipsed by its normative association; 20  “even dictators appear to believe that an indispensible ingredient for their legitimacy is a dash or two of the language of democracy.”21 Definitions of democracy fall into classification that includes procedural, focusing primarily on electoral mechanisms; constitutional, or regime type; substantive, addressing issues such as welfare provision; and process-oriented approaches that emphasize social mechanisms and are prioritized here.22 A democracy, in Tilly’s (2001, p.31) definition, is a “regime of protected consultation,”23 with “relations between agents and subjects of a government in which (a) different categories of subjects enjoy relatively broad and equal access to agents, (b) governmental disposition of persons, activities, and resources within the government’s purview responds to binding consultation of subjects, and (c) subjects receive protection against arbitrary action by 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   18	
  See Tilly (1992) for the state as a war-making machine. For economic transformation and rule making see, respectively, Evans (1995) and Levi (1989). 19 Giugni (1998, p.xiii); See Krasner (1984) and Skopcol (1985) on the state as corporatist actor and arena. 20	
  Hindman 2009, Przeworski et al. 2000. 21 Dahl (1998, p.2). 22	
  See Almond and Verba (1989) and Tilly (2007) for democracy scholarship classification. 23 McAdam, Tarrow, and Tilly 2001, p.78. 	
   	
   17	
   governmental agents.”24  Democracy is thus a “process for reaching collective and binding decisions,”25 and democratization is movement towards democracy.  Democracy in this study focuses beyond the electoral and primarily to underlying processes of transparency, accountability, and representation. My focus is not regime transition.26 Most of the study focuses on the democratization processes just outlined, but chapter 4 measures democracy using the Freedom House index.27 3.4	
  The	
  Independent	
  Variable:	
  Information	
  Communication	
  Technology	
  	
   The main explanatory variable is the level of ICT adoption in a state. This includes both underlying telecommunications infrastructure and the digital communication consumer and commercial products, services, and platforms this supports. ICT is “the hardware, software, networks, and media for the collection, storage, processing, transmission and presentation of information (voice, data, text, images), as well as related services.”28  ICT here refers to the Internet and mobile phones, as well as hybrid platforms such as smart phones.29 Whether a protester uses a smartphone to create a video or an online application to text a message is beside the point. Functionality is converging and so too is ICT’s impact.30 Several recent transformative technologies have widened the scope of ICT, most notably social networks, like Facebook and Renren, and microblogs 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   24	
  But see also Tilly (2000). 25	
  Dahl 1989, p.5; See also Dahl 1998. 26 This is the scope used by Howard (2011), but many studies focus exclusively on transitions despite their rarity (Przeworski et al. 2000; Acemoglu and Robinson 2006). Transitions are dramatic, but rarity limits their generalizability and analytical relevancy. 27	
  This includes the underlying mechanisms of interest here (Freedom House 2011). 28	
  World Bank 2011. 29 See Cortimiglia et al. (2011) for a mobile/internet 2.0 classification scheme. 30	
  Technically, this is known as ‘digital convergence’ (PC Magazine 2012). 	
   	
   18	
   such as Twitter and Sina Weibo. ‘Web 2.0’ 31 signals this shift towards user-generated content (UGC) ICT for many-to-many communication central to the narrative to follow. 3.5	
  The	
  Intervening	
  Variables:	
  Digital	
  State	
  Capacity	
  and	
  Regime	
  Legitimacy	
  	
   State control of digital information is conceptualized here as a new form of state capacity: digital state capacity (DSC). Chapter 5 develops this concept but this section offers a foundation. State capacity is the ability “to penetrate society, regulate social relations, extract resources, and appropriate or use resources.”32 DSC is primarily an ‘infrastructural’ or ‘instrumentational’ capacity necessitating the government to work through intermediaries to achieve its policy aims. 33  This capacity is central to developmental states with close government-private actor relations that direct market policy. It is a set of state functions34 dependent on the quality – the professionalism, meritocratic nature, lack of corruption – and the corporate cohesion of the underlying institutional bureaucracy.35 Building this capacity is difficult; “there is no inexorable tendency for the supply of bureaucracy to meet the demands … put on it.”36 DSC lends itself to cross-case comparison37 and it is possible to identify divergent ideal types of 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   31	
  See OECD (2007) and O’Reilly (2005) for differences between web 2.0 and previous ICT generations.	
   32	
  Migdal 1998, p.4. 33	
  Mann (1984) divides state capacity into infrastructural versus despotic, with the key difference matching that of McAdam, Tarrow, and Tilly’s (2001) instrumentation concept and Evans (1995) binding the state “to society for continual negotiation and renegotiation of goals and policies” (Evans 1995, p.12). 34	
  See Fukuyama (2004, p.13) on analytical separation of state capacity and function. 35 Evans (1995); But see also Howell (2006) for extension and application to China. 36 Evans 1995, p.70. 37	
  India and China, by analogy, have been compared on their relative levels of extractive (read: taxation) and redistributive (read: welfare transfer) capacity. See Winters and Yusuf (2007) and Bardhan (2009). 	
   	
   19	
   DSC states (below). Quantitative operationalization would necessitate an approach similar to measuring concepts such as democracy and rule of law.38  The second intervening variable is regime legitimacy. Lipset (1959, p.86) defines legitimacy as “the capacity of a political system to engender and maintain the belief that existing political institutions are the most appropriate or proper ones for the society.” Legitimacy depends primarily on how citizens perceive the role and performance of the state “as rightfully holding and exercising political power” (Gilley 2006, p.48). But as with democracy, legitimacy’s descriptive and normative dimensions of are often conflated.39 Weber (1965) detailed three legitimacy types, but political scientists have used many factors to account for state legitimacy. Among these, governance quality, rights, and welfare have the greatest explanatory power.40 Electoral democracy in itself is not an empirical prerequisite for legitimacy and many non-democratic regimes enjoy high legitimacy premised on their performance.41 Whether ICT plays a role in political change depends more on the legitimacy of the regime than its democracy level. This plays heavily into the theory that follows and its later application to China in chapter 6. 	
  	
   	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   38	
  Operationalization and measurement of DSC is beyond the scope of this study. The conceptual waters are further muddied by the possible connection of regime type to state capacity level. Several studies address this. Back and Hadenius (2008) hypothesize a J-shaped relationship between capacity and regime type and separating capacity and democracy is central to Tilly’s (2007) state comparative method. 39	
  Hardin 2009.	
   40	
  Gilley (2006) analyzed a multitude of factors across 72 states to determine which had the strongest explanatory power.	
   41Rothstein (2009) refers to this as output-based performance (i.e. governance) rather than input-based (i.e. electoral); Gilley (2006, p.58), “It is notable that democratic rights, while certainly qualifying as one of the most important causes of legitimacy, turn out to be roughly on par with welfare gains, and both of these are far less important than good governance.” See also Zhu (2011). 	
   	
   20	
   3.6	
  Core	
  Theory	
  and	
  State	
  Typology	
  	
   In this study, I argue that large-scale adoption of ICT in Asian states with low regime legitimacy can, over time, be a factor in democratization. ICT, however, is never a sole factor. Non-electoral democratization processes involve complex social, political, and economic dimensions contingent on a state’s unique context. In addition, these changes occur within a larger media, including old mass media, ecosystem.42 Nonetheless, high ICT penetration rates are transformative and lead to fundamental changes in the political communications landscape. Beyond a certain adoption level, ICT becomes a necessary, if not sufficient, factor in political change.43 While ICT can empower societal actors and, in some cases, act as a force towards democratic change, governments enjoying high public support and legitimacy face lower pressure for political change. Conversely, the regime’s very right to govern becomes a focus of discontent and protest in states with low government legitimacy. ICT here acts as both a medium and amplifier for political pressure from societal actors. States, however, play a role in countering some of these ICT effects. They can do so by developing capacity to control and manage digital communications. The catch, as per the Dictator’s Dilemma, is that as ICT adoption rises so too do costs for controlling its effects. This forces states to prioritize information control policy objectives, surrendering the agenda on many issues. In effect, states are forced over time to narrow their scope of information control. Issues previously controlled by the state enter the public domain of societal actors shifting the boundaries between the allowed and disallowed. For the prudent state, this is a rear- 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   42 The case of Tunisia 2011 is instructive. Social media was not a sole factor, but worked in concert with Al Jazeera broadcasting protester videos back to Tunisian living rooms (Zuckerman 2011; 2011b). 43 Howard 2011. 	
   	
   21	
   guard action and not a rout. Nevertheless, into this gap enter societal actors, empowered by ICT, staking claims to political ground newly relinquished by the retreating state.44  The interplay between the levels of ICT and countering state capacity is shown in table 3.1. Ideal types are provided for states with combinations of low and high ICT diffusion rates and low and high DSC.45 The case of low / low is excluded. 46 Dynamics of political contention are expected to differ based on the ideal type closest to the state investigated. Maximal pressure for political change is expected where ICT adoption is high, and DSC and legitimacy are low (see figure 3.1). Low legitimacy in each case increases the chance and risk of political change as societal actors are empowered. Table	
  3.1	
  Ideal	
  State	
  Types	
   	
   	
   	
   High	
  Digital	
  State	
  Capacity	
   Low	
   Digital	
  State	
  Capacity	
   High	
  ICT	
  Adoption	
   Ideal	
  Type	
  I	
  (ex:	
  China)47	
   Ideal	
  Type	
  II	
  (ex:	
  Indonesia)	
   Low	
  ICT	
  Adoption	
   Ideal	
  Type	
  III	
  (ex:	
  Burma)	
   Excluded	
    	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   44	
  For digital communication changes in S.E. Asia, Woodier states, “The pursuit of economic development was seen to have dissipated the once centralized ownership and control of the communications media…as a result, the ruling elites were forced to compete with other groups” (Woodier 2008, p.22). 45 Application to the ideal types would necessitate normalization of these two variables within the range for the states under investigation, yielding relative ordinal values for state classification. 46	
  This is because this study aims to understand the effects (not absence of) digital communications.	
   47 See chapter 4 and appendices for empirical ICT values and chapter 5 for examples of DSC levels.	
   Figure	
  3.1	
  –	
  Conditions	
  for	
  Maximal	
  Political	
  Change 	
   	
   22	
    But if communication technology diffusion results in loss of political power by the state, why would non-democratic governments willfully choose and promote adoption of ICT? Why do they enter the Dictator’s Dilemma? 48  The answer lies in Asian globalization49 and the ‘golden bargain’ this offers. Tiberghien (2007) introduces the golden bargain concept as a trade-off national politicians face between liberalizing market controls for long-term economic gain partly at the cost of short-term social disruption. The lessons of developmental states – Japan, the Asian Tigers, and China – is that high rates of economic growth are possible through selective, gradual but yet inexorable, integration into the global economy.50 The lodestar for the developmental Asian state is an economy with large segments atop the global value chain, driving innovation, pushing the envelope of tomorrow’s products and services. Information technology and the educated workforce to design, engineer, and apply it, are central organs in this globalized vanguard economy.51  Seeing this, political entrepreneurs arbitrage between the global and local system,52  and actively promote technology infrastructure development, and with it ICT products and services. However, unlike just-in-time supply chain infrastructure, these are targeted primarily at consumers and citizens, not corporations. As with other consumer goods, economies of scale, often propelled by newly privatized companies, drive prices down and adoption rates towards 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   48	
  As discussed in the previous chapter, the Dictator’s Dilemma applies primarily to states that depend on ICT adoption for economic development. This may not be the case, for example for states with other large revenue sources, such as petro-states.	
   49	
  Globalization is the international integration of trade, capital flows, migration, and culture (Garrett 2000), but here is used primarily as economic integration (Rodrick 2000). 50 Globalization is but one facet of these states’ success (Schuman 2010), as is their reticence towards full integration (Stiglitz 2002, Rodrick 1999). See Branstetter and Lardy (2008) and Dollar (2008) for details on China’s economic success. 51	
  See Schuman (2010) in general, and Winters and Yusuf (2007) for Indian and Chinese examples. 52 Tiberghien (2007, p.20).	
   	
   	
   23	
   saturation.53 But unlike other consumer and public goods, ICT are ‘coordinating goods’54 allowing citizens to independently produce and consume information. Tools adopted for economic reasons rewire communication with political consequences. The end result for non-democratic regimes may be a Faustian pact rather than not a golden bargain. 55 3.7	
  Social	
  Movement	
  Political	
  Opportunities:	
  Government	
  Under	
  an	
  Unblinking	
  Eye	
  	
   Detecting these changes necessitates an analytical framework that delves into underlying causal mechanisms. The quantitative analysis of chapter 4 only serves to demonstrate macro regional patterns and correlations. To address this limitation and glean possible causality, the McAdam, McCarthy, and Zald (1996) Social Movements Framework is adopted. This divides contention analysis into changes in political opportunities, social movement mobilization, and issue framing. 56 As an analytical lens to peer into the effects of ICT, it acts as an heuristic for enumerating and categorizing disparate processes. Not every social movement empowered by ICT is state-wide, transformative, or successful in achieving its aims. Nor is the state versus society binary dichotomy representative of the ‘fog of contention’, the shifting constellation of actors and interests that is political contention.57 But its use allows assigning agency to societal actors and its DSC counter, explored in chapter 5, develops a lens for the opposing capacity of non-democratic states. 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   53	
  Singh and Raja (2010), but also Howard and Mazaheri (2009) for governments’ role in ICT promotion. 54 Bueno de Mesquita and Downs (2005).	
   55	
  Tiberghien (2007); Unlike Friedman’s (2000) ‘golden straightjacket’ globalization metaphor (Rodrick 1999), here states choose ICT development for economic benefits at the cost of loss in policy control. 56	
  McAdam (1996) adds movement form to the framework: from institutional reform to full revolution. 57	
  Yang (2009) criticizes usage of this framework exactly on these grounds, but a response is Migdal’s (2001) assertion that states and their societies are mutually constituted and delimiting. Furthermore, the historical basis for this lens goes back to de Tocqueville’s (1839, 1856) comparison of state-society relations between pre-revolutionary France and post-revolutionary America (McAdam 1996). 	
   	
   24	
    ICT adoption alters the political opportunities58  that societal actors face by increasing the visibility of government actions (and mis-actions), facilitating the reading of public opinion, and creating maneuver space for political entrepreneurs. First, ICT increases the sources for and aggregate information on government policy and activity. Limitations (regulatory and functional) on traditional media are circumvented by expansion in citizen recording and dissemination of information, compounded by alternative media. This nets higher policy visibility and more eyes on government actions,59 and has been termed a ‘reverse panopticon’, flipping Foucault’s famous metaphor.60 This also incentivizes governments to become more information savvy and proactive in promoting positive policy achievements,61 and is part of Asian states’ race towards e- and m-government and sundry other digital government initiatives.62  Second, greater aggregate citizen-generated information offers policy makers a barometer on public opinion, and thus a policy prioritization tool. 63 This pulse-taking is a central factor in the resilience of authoritarian Asian regimes. Finally, social movements become attuned to possible gaps between government levels and ministries,64 but the expanded ICT realm has created additional avenues for political entrepreneurs. These individuals, especially in regimes sensitive to populist sentiments, position themselves as tribunes of the people carrying the banners of social movement 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   58	
  Political opportunity changes alter institutional features, political alignments, or repressive capacity that “reduce the power disparity between a given challenging group and the state” (McAdam 1996, p.32). 59	
  There is also debate as to how much (and whether) greater attention leads to less corruption. Diamond 2010; Grönlund 2010; Faris and Etling 2008; Mahmood 2004; and Yang 2009. 60	
  Garrett 2006; Foucault 1995. 61 Non-democratic propaganda bureaucracies, of varying levels of path-dependency, are responding with layering and realigning focus and functionality (Brady 2008; Thelen 2003). 62	
  Howard 2011. 63 Nathan (2003) emphasizes public input mechanisms different from new media. But see also Kalathil and Boas (2003) on pre-emptive liberalization. 64 McAdam 1998. 	
   	
   25	
   objectives.65  Substantial digital state capacity can counteract these political changes, but their impact is compounded when combined with mobilization potential.	
   3.8	
  Social	
  Movement	
  Mobilization:	
  Social	
  Ties	
  Online	
  and	
  Off	
  	
   As discussed in the previous chapter, ICT enhances social movement mobilization by reducing transaction costs and information asymmetries. But there is a stronger, time- lagged, force at play in the large-scale adoption of ICT: restructuring of the social communication network topology.66 All human beings are enmeshed in communication networks characterized by the number of ties to others, the tie strength (whether family or acquaintance), and the level of similarity between individuals. Counterintuitively, for information diffusion – including political information and its interpretation – weak and disparate ties matter more than strong and similar ties.67  Topographically, clusters of similar individuals are connected through dissimilar and often distant ties.68 ICT adoption increases overall aggregate ties and tie density, and consequently, over time, leads to acceleration and growth in information diffusion across society. This makes it more difficult to control and limit information.69 In highly wired states, the very act of blocking information becomes exceedingly conspicuous.70 In sum, ICT mobilization 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   65	
  China, for example, has served as a fecund environment for this in its divisions between government levels (Lieberthal 2004) and populism of top leaders (Stockmann and Gallagher 2011). 66 See Shirky (2008, 2011) for this social transformation that becomes political over time. 67	
  Technically, similarity level (homophily) and the number of issue for which an individual is influential (polymorphism) bear directly on network topology. Trust is proportional to homophilousness and diffusion inversely proportional. So there is a balance between weak ties, that promote information diffusion, and homophily, which determines adoption of information and ideas (Rodgers 2003). 68	
  This model of communication network analysis is based on Granovetter’s (1973) seminal discovery that most jobs are found through weak ties. See Fischer (1994) for why weak and disparate ties were crucial to Paul Revere’s successful ride. See also Rodgers (2003) and Gladwell (2000). 69 See, for example, Asian political contention examples in Hermans (2008) and McCargo (2003). 70 Examples discussed in chapter 5 are the shut down of regional communication in Xinjiang in 2009 (Xinhua 2009) and national shut down in Egypt in 2011 (Glanz and Markoff 2011). 	
   	
   26	
   impacts exceed better organization71 to the transformation of information flow across society. DSC can counter some mobilization changes, but not all. As per McLuhan’s dictum, in altering communication topology, the media here is the message.72 3.9	
  Social	
  Movement	
  Issues	
  Framing:	
  Contention	
  in	
  the	
  Public	
  Sphere	
  	
   Beyond changes in political opportunities and mobilization, ICT alters the speed, control, and nature of social movement issue framing, often in newly emergent online public spheres. 73 Issue framing is a function of the public environment for debate and discussion. With diffusion in the political economy of traditional media, ICT adoption enables more and disparate voices to enter the fray of public contention. Subject to the limitations the state chooses to apply, this can create and entrench a public sphere distinct from its offline counterpart.74 Issues discussed and debated depend on the shared interests and goals of participants. These are amplified through group consolidation as offline groups, for example those fighting for specific rights, find power in numbers online, and new online groups form around shared norms and values.75 New collective online identities, or ‘imagined communities’, 76 have, in fact, proven catalytic in social movement empowerment. With a growing number of voices populating the 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   71 For examples of fast-mobilizing wired ‘smart mobs’ that rely on ICT see Rheingold (2002). 72	
  McLuhan 1965. 73	
  Although McAdam (1998) sees a shared public space as instrumental to all mobilization. 74	
  With less choke-points, more voices participate. Habermas (1989, 2006) emphasizes the role of media in creating a public sphere and this has been applied to its online counterpart (Yang 2009). 75	
  Yang (2009) offers a poignant example of Chinese Hepatitis-B sufferers grouping online for support and resulting in legislation changes on employment discrimination towards carriers. 76	
  Anderson’s (2006) nationalism thesis applies. As with nations, these groups are communities since members share norms and values, but are imagined since members never meet and associate. A prime 2011 example is that of online social network communities in Egypt and Tunisia where members found common cause that translated to offline protest actions (Giglio 2011). 	
   	
   27	
   public sphere, issues at the borders of the allowable attract debate.77 Overtime, these boundaries expand. Finally, ICT strengthens early framing of issues.78 The advantage lies with distributed nimble societal actors over centralized governmental bureaucracies. The first frames to resonate can be echoed widely through an online multiplier effect.79 Digital state capacity can attenuate this, but in wired states, the government no longer decides alone which issues are debated, enter the national agenda, or become viral. 3.10	
  Conclusion:	
  Framing,	
  Mobilization,	
  and	
  Political	
  Opportunities	
  in	
  Action	
  	
   This chapter has laid out the study’s theory. Social movements are empowered by ICT adoption and, where state legitimacy is low, this can be a central factor towards non- electoral democratic mechanisms. Governments are placed under unblinking digital eyes and often respond with preemptive information liberalization. Accountability increases in a media environment of rapid news cycles and competition from citizen journalists. Public interest can be better read and possibly represented through emerging public spheres. But these political changes are contingent on states’ ability to control digital information, their DSC. Prior to developing this concept in chapter 5, however, chapter 4 quantitatively analyzes this and the alterative theories of chapter 2. It demonstrates that ICT is correlated to democracy level in Asia over time contingent on DSC. 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   77 In some states, discussions of sensitive social topics, such as gender relations, take on political overtones. Howard (2011) offers examples from Islamic states where discussing changes in gender relations is a political act tolerated only because it occurs at the margins online. 78	
  Livingston and Bennett (2003) see ICT’s speed as increasing early issue framing significance. Examples include framing after the 9/11 attacks by Jihadists (Kellner 2003) and US government (Said 2001). 79 Frames can be enduring, for example for religion (Said 1985, 1997; Tweed 2008) and race (Ng 2010). 	
   	
   28	
   Chapter	
  4	
  	
   Quantitative	
  Analysis:	
  ICT	
  and	
  Democratization	
  in	
  Asia 	
   4.1	
  Why	
  Model	
  ICT	
  and	
  Democracy?	
    This chapter sets out to test the theory in the previous chapter and contending alternative theories through a quantitative analysis. This approach rests within the Lipset (1959) tradition to understand democratic change based on underlying economic and social factors.1 But my approach introduces digital communication factors as well. I will not be able to demonstrate causality and the analysis will exclude legitimacy.2 But if the Cyberutopians are correct and ICT empowers societal actors over states, we expect to find a correlation between ICT and democratization. If the Cyberskeptics are correct, we expect to find an opposite relationship. The results will demonstrate that, while ICT adoption is correlated to democracy level in Asia over time, this is subject to state policy. ICT does not carry inexorable democratizing or counter-democratizing effects. In other words, societal actors are empowered but any net political change is a function of both societal empowerment and state action. Asian state policy matters in countering ICT’s political effects. Chapter 5 will develop this role of the Asian state in detail.  The analysis in this chapter is divided into four parts. After assessing ICT as an explanatory factor for democracy level, a comprehensive model of Asian democratization over time is built with communication at the centre and accounting for economic and social factors. This demonstrates that while regional correlation exists, individual state context trumps macro trends. Second, a proxy variable (online 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   1 See also Przeworski et al. (2000), Boix and Stokes (2006), and Acemoglu and Robinson (2006).	
   2	
  It was not possible to include legitimacy measures due to time constraints.	
   	
   	
   29	
   censorship) is tested as a stand-in for DSC. Finally, this proxy is applied to the larger model to assess how ICT varies with democracy given state DSC level. 4.2	
  Hypotheses,	
  Data,	
  and	
  Methodology3	
    This section defines the hypotheses to be tested, outlines data used, and discusses the methodology employed across models. Three hypotheses are tested. H1 and H2 test for the alternative arguments in chapter 2.4 But if, as postulated, DSC is an intervening variable for the effects of ICT then H3 offers the best explanation. H1  CYBERUTOPIANISM H2 CYBERSKEPTICISM H3 ICT AND DSC ICT	
  is	
  consistently	
  and	
   strongly	
  positively	
  correlated	
   with	
  democracy	
  level	
  in	
  Asia	
   ICT	
  is	
  consistently	
  and	
   strongly	
  negatively	
  correlated	
   with	
  democracy	
  level	
  in	
  Asia	
   Democracy	
  level	
  in	
  Asia	
   depends	
  on	
  relative	
  level	
  of	
   ICT	
  and	
  DSC	
     See the appendix for details on data and sources. In scope are all states in the regions of East Asia, South East Asia, South Asia, and Central Asia for a total of 26 countries, between 1994 and 2009. For a comprehensive view of the region, this includes states with consistently high levels of democracy, such as Japan and South Korea. Only East Timor is excluded completely.5 1994 is the starting point as this represents initial ICT measurement in many states and the launch of the first Internet browser (Mosaic).6 2009 is the last year for which data was comprehensively available. 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   3 I would like to thank Dr. Fred Cutler (UBC) for extensive assistance with the data, methods, and especially methodological challenges presented in this chapter.	
   4	
  Disproving H1 puts in question considerable foreign policy, such as that of the Canada, the US, and several G8 statements. See DFAIT (2007), US Congress (2010c), and G8 (2011). 5 North Korea is excluded for 1994 to 2008 due to lack of data. 6	
  This also allows for comparison with Kedzie (1997), Best and Wade (2009), and Howard (2011). 	
   	
   30	
    The dependent variable, democracy, is measured using the Freedom House index (FHI) which offers an annual measure of political rights and civil liberties. FHI is unique in incorporating factors that point to underlying democratization mechanisms independent of electoral processes, and directly pertinent to the conceptual understanding of democracy outlined in the theory above. 78 FHI scores for political rights and civil liberties are combined to yield a scale ranging from 2 (most democratic) to 14 (least democratic). ICT, the primary explanatory variable, is measured as a combination of the average country Internet and mobile subscriber penetration rate per 100 population.9 Control variables representing economic (GNI per capita PPP) and social development (literacy and educational enrolment) are included.10 Finally, the DSC proxy (2009 online censorship) uses data from the OpenNet Initiative (ONI).11  In total, 15 models are tested and presented in this chapter.12 These include bivariate correlation (models I to III), multivariate correlation using ordinary least squares (OLS) (models X to XIII), as well as time series cross sectional models (models IV to IX and XIV to XV). 13 The time series models, while invaluable in understanding the variables under investigation, pose specific methodological challenges. Asian states are a diverse set, varying considerably in economic, social, and educational context. 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  7	
  Freedom House 2011. Other measures, such as the Polity IV index, are available. 8	
  Democratization models often rely on a sliding scale democracy index: Przeworski et al. 2000, Persson and Tabellini 2003, Acemoglu and Robinson 2006. 9 These are available from the World Bank and International Telecommunications Union and, while self- reported, offer the most-widely used measures (WB 2011; ITU 2011). Combining mobile and Internet offers a wider and more realistic understanding of digital communication, although a more comprehensive approach would include Internet hosts and bandwidth measures (Howard 2011). 10	
  Educational enrolment combines average primary and secondary enrolment. Literacy level is the literate percentage of population above the age of 15. These are sourced from the World Bank 2011, UNESCO 2011, and UNDP 2011; as well as several national statistics agencies.  11	
  ONI 2011a, 2011c. 12	
  Please see the appendix for all models in mathematical notation. 13 This builds and expands on previous research. See Best and Wade (2009) and Groshek (2010). 	
   	
   31	
   Studying these states as cases over time incurs autocorrelation and heteroskedasticity. 14 The concern is that relationships detected across the data set (regional) may not apply to individual cases (states). To address this, two models (V and VI) control for sub-regional variation. This offers some correction, but at the cost of coarse groupings that mask intra-group differences.15 Geography here is a blunt proxy for myriad factors affecting democracy level. Fixed effects models (VII, VIII, and XV) address this (and offer contrasting results) at the cost of loss of generalizability.16  The difficulty is exacerbated by the nature of the dependent variable. Democracy, for the most part, is a slow-moving variable.17 Time-tested methods for addressing this challenge, including lagging the dependent variable, are of limited use here.18 There is no simple solution. So the above models are used in parallel for a more nuanced, but more complex, view of the data. It is also possible that relationships investigated suffer from reverse causality.19 Unfortunately, fully addressing this is not within the scope of this study and would require more advanced techniques.20 A weaker alternative (not fully accounting for reverse causality) used here lags the main independent variable. 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  14	
  Autocorrelation means that patterns in the data may vary consistently across time independent of and clouding results for the relationship under investigation. Heteroskedasticity refers here to variance between individual states that overwhelms and possibly mask the main relationships under investigation. Best and Wade (2009) encountered similar challenges studying the Internet and democracy. 15	
  Regional groups work best with large numbers of states and underlying similarities (Kedzie 1997). 16	
  Fixed effects isolate the variable relationship at the unit level by removing differences between states. 17	
  Time series analysis has specific difficulties with slow-moving variables. This is particularly so when entrenchment and non-electoral mechanisms are of interest. Countries do not normally exhibit large, sharp changes in democracy level. The exception is regime transitions - rare events are of limited use here. See Wilson and Butler (2007) for a deeper discussion. 18	
  Beck and Katz (1995) suggest several methods, including lagging the dependent variable and differencing the dependent variable. All of these were tired here with no success. Democracy was lagged from 1 to 6 years but all attempts removed too much explanatory power from the independent variables. 19 This would mean that the level of democracy affects the adoption of ICT.	
   20	
  For example, using instrumental variables that affect ICT and not democracy. One option may be a measure of competitiveness in the digital communications sector as an instrument for digital communications adoption exclusive from the level of democracy. See Clarke and Wallsten (2006). 	
   	
   32	
   4.3	
  ICT:	
  Empty	
  Promise	
  or	
  Democratic	
  Indicator?	
    If democracy is indeed directly correlated with ICT adoption, as per the claim of the Cyberutopians, this relationship should be observable. 21  This is now evaluated in comparison to economic and social development indicators. If H1 is true, a consistent and positive correlation is expected between ICT and democracy. 22  	
   Figure	
  4.1	
  -­‐	
  Communication	
  vs.	
  Economic/Social	
  Factors	
  as	
  Democratic	
  Indicators	
   Figure 4.1 displays the results of annual bivariate correlations between democracy and respective economic, social, and ICT factors.23 Although ICT demonstrates a positive correlation to democracy, this relationship is decreasing in magnitude over time.24 Furthermore, this relationship is unidirectional unlike the other two variables. 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   21	
  Democratization models have generally included economic and social explanatory factors but not communications. See Przeworski et al. (2000), Acemoglu and Robinson (2006), and Boix (2003). This is done here through a bivariate regression of democracy against measures of economic, social, and ICT, respectively, and builds on previous studies (Kedzie 1997, Best and Wade 2009, Howard 2011).	
   22	
  It is important to look at annual correlations and their change over time and not a correlation across all observations. Doing the latter offers a misleading view of the data. See Appendix Figure A.1 on this.	
   23 Each value is the sum of the independent variable coefficient and annual interaction term, with 1994 as the reference. Only the values for ICT are statistically significant at the 10% level, excepting 1994-1996. This	
  figure	
  shows	
  the	
  annual	
  correlation	
  between	
  ICT	
  (IntMob),	
  educational	
   enrolment	
  (EduEnrol),	
  and	
  GNI	
  p.c.	
  PPP	
  against	
  democracy	
  level	
  across	
  all	
  states.	
   	
   	
   33	
   Thus H1 is rejected: although ICT is correlated with democracy and offers a strong indicator for democracy in comparison with economic and social measures, this relationship is not consistent over time in Asia. Two contending explanations emerge.  It may be that ICT adoption demonstrates diminishing returns against democracy where ICT increases correspond to ever decreasing increments in democratization; the bang for the digital buck is greatest for lower and initial adoption. But this does not match the behaviour exemplified by innovation diffusion, technological or otherwise.25 An alternative explanation is that while both democratic and non-democratic Asian states have been adopting ICT, non-democratic states have employed measures to counter democratization effects. This would explain increasing penetration rates, decreasing correlation with democracy, and the policy choices of many Asian states.26 To investigate this, the next section looks at this relationship over time. 4.4	
  ICT	
  and	
  Asian	
  Democratization	
  Over	
  Time	
    Understanding how economic, social, and communication factors are correlated with and affect democracy over time necessitates a different method. A time series cross sectional analysis is now conducted with multiple explanatory variables.27 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   24 A 1-point more democratic score correlates to an additional 2-3 Internet/mobile users (per 100 pop.) for 1994-1996. For 2006-2009 an equivalent democratic change correlates to 17-23 additional users.	
   25	
  Rogers (2003) provides a model for how diffusion of innovations carries social effects that is based on a cumulative distribution function (an s-shaped graph). Beyond certain penetration rates, the effects are only reversible at substantial cost. Diminishing returns are only apparent at very high penetration rates, and this does not represent the majority of Asian state cases analyzed. 26	
  See, for example, ONI Asia (2007) and Kalathil and Boas (2003). 27	
  Models include: a baseline model that controls for economic development, social development, and population for all states (IV). This uses OLS and panel corrected standard errors; Models to account for sub-regional variance through regional dummy variables and their interactions with the digital 	
   	
   	
   34	
    The baseline model (IV) yields a strong and statistically significant (at the 1% level) correlation between ICT and democracy controlling for economic development, population, and educational enrolment. Increasing ICT adoption from the lowest in the Table	
  4.1	
  –	
  Asian	
  Democratization	
  Models	
  -­‐	
  Time	
  Series	
  Cross	
  Section	
  Analysis	
  Results	
   Dependent	
  Variable:	
  Democracy	
  Level	
  (2	
  =	
  very	
  democratic,	
  14	
  =	
  least	
  democratic)	
   Explanatory	
   Variable	
   (IV)	
   	
   Base	
   (V)	
   	
   With	
   Regions	
   (VI)	
   	
   Regions&	
   Interactions	
   (VII)	
   	
   Country	
   Fixed	
   Effects	
   (VIII)	
   	
   Country&Yr	
   Fixed	
   Effects	
   (IX)	
   	
   Panel-­‐ specific	
   AR(1)	
   Internet/	
   Mobile	
  (t-­‐3)	
   -­‐0.0544***	
   (0.0147)	
   -­‐0.0505***	
  	
  	
   (0.0136)	
   0.0230	
   (0.0278)	
   0.0044	
   (0.0053)	
   -­‐0.0024	
   (0.0023)	
   0.0112	
   (0.0072)	
   GNI	
  per	
  capita	
   PPP	
  x	
  10,000	
   0.0311	
   (0.161)	
   0.531***	
   (0.175)	
   0.564***	
  	
  	
  	
   (0.156)	
   -­‐0.814***	
  	
   (0.196)	
   -­‐0.981***	
  	
  	
  	
   (0.108)	
   -­‐1.875***	
  	
  	
  	
   0.363	
   Log	
  of	
  Population	
   -­‐1.2343***	
  (0.0844)	
   -­‐0.4178***	
  	
  	
   (0.0760)	
   -­‐.32948***	
   (0.0603)	
   5.9767**	
   (2.9115)	
   -­‐2.6256	
   (6.0468)	
   -­‐0.8281	
   (0.6206)	
   Education	
   Enrolment	
   -­‐0.0358***	
   (0.0133)	
   -­‐0.1049***	
  	
  	
   (0.0111)	
   -­‐0.1012***	
   (0.0086)	
   -­‐0.0739***	
   (0.0114)	
   -­‐0.0928***	
   (0.0105)	
   -­‐0.0264	
   (0.0304)	
   East	
  Asia	
   	
   0.5871*	
  	
  	
  (0.3482)	
   2.4902*	
   (1.3008)	
   	
   	
   	
   South	
  East	
  Asia	
   	
   2.8749***	
  	
  	
  (0.2855)	
   2.9919***	
   (0.6039)	
   	
   	
   	
   Central	
  Asia	
   	
   5.1809***	
  	
  	
  (0.2734	
   5.3215***	
   (0.4043)	
   	
   	
   	
   South	
  Asia	
   	
   (omitted)	
   (omitted)	
   	
   	
   	
   IntMob*EAsia	
   	
   	
   -­‐0.0099**	
  (0.0392)	
   	
   	
   	
   IntMob*SEAsia	
   	
   	
   -­‐0.0414	
  (0.0296)	
   	
   	
   	
   IntMob*CntAsia	
   	
   	
   -­‐0.0196	
  (0.0174)	
   	
   	
   	
   IntMob*SAsia	
   	
   	
   (omitted)	
   	
   	
   	
   Constant	
   22.1236***	
  (1.4792)	
   19.2763***	
  	
  	
   (0.7428)	
   18.1924***	
   (0.6513)	
   -­‐35.6312	
   (22.9287)	
   151.48***	
   (22.5939)	
   18.4906***	
   (6.1896)	
   R2	
   0.21	
   0.43	
   0.46	
   0.94	
   0.94	
   ρ=0.88	
   N	
   326	
   326	
   326	
   326	
   326	
   326	
   *p	
  <	
  .10	
  **p	
  <	
  .5	
  ***p	
  <	
  .01	
   	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   communications (V & VI), all using OLS; and fixed effects models that control for individual states (VII) and year effects (VIII).	
   	
   	
   35	
   set (North Korea) to the highest (Singapore) 28 corresponds to an increase in democracy level from that of Brunei to that of Indonesia. This hints at a positive correlation between democracy and ICT over time.  Dialling granularity up to the regional level in models V and VI yields a better explanatory model (with R2 rising to 0.43). 29 The ICT variable maintains its strength and most variables, including regional dummies, are significant at the 1% level. With South Asia as the reference region, East Asia, South East Asia, and Central Asia have less democratic scores at 1, 3, and 5 points, respectively, on the combined FHI. 30 Adding regional interaction terms in model VI comes at the cost of statistical significance for the ICT variable and most of the interaction terms.31 Given the limited number of cases per region, the data is being asked to do too much.32 An OLS approach, even in attempting to account for regional variance, can yield no more and a change of tack is in order.  Two fixed effects models are consequently tested to control for individual state democracy levels and inter-temporal trends, in effect isolating the ICT/democracy relationship from variance between states and trend ‘waves’ across years.33 Model VII thus includes dummy variables for state and model VIII for individual years as well. Both offer considerably more explanatory power (R2 of 0.94) but render the ICT variable not statistically significant (plus reducing it in magnitude and changing its sign). 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   28	
  Holding all other variables at their mean, a 1-point more democratic score on the FHI corresponds to 20 additional Internet/mobile users per 100 population.	
   29	
  This relaxes the assumption of common state intercepts by allowing regions to differ in respective average democracy levels.	
   30 With the exception of East Asia at a significance of 10%	
   31	
  With East Asia as the exception. The previous model allows for different regional democracy level. The interaction terms accounts for how much ICT affects each region. 32	
  The sole salient result is for East Asian states, where, with all variables held at their mean, a 1-point more democratic score corresponds to 13 additional internet/mobile users per 100 population.	
   33	
  This relaxes the assumption of a common intercept for all states and then expands to include individual intercepts for each year in the data set.	
   	
   	
   36	
   Crucially, this demonstrates that unlike the relationships previously described – both the significant but decreasing magnitude and the overall positive OLS correlations – individual state context trumps overriding regional relationships. It is not macro-trends, but rather the state level that determines how ICT interacts with democracy. This divergence between a holistic and individual- state relationship is displayed graphically in Figure 4.3. Each coloured line indicates the movement in democracy score and Internet/mobile adoption for a country from 1994 to 2009. The thick dark lines show the correlation between ICT and democracy using OLS BGD BTN BRNKHM CHN IND IDN JPN KAZ KOR KGZ LAO MYS MNG MMR NPL PAK PHL SGP LKA TJK THA TKM UZB VNM Fixed Effects OLS 0 5 10 15 De mo cr ac y Sc or e 0 20 40 60 80 100 Internet/Mobile per 100 pop. Figure	
  4.2	
  –	
  Correlation	
  between	
  ICT	
  and	
  Democracy	
  in	
  Asia:	
  Regional	
  (OLS)	
  vs.	
   State-­‐level	
  (fixed	
  effects)	
   Coloured	
  lines	
  show	
  the	
  change	
  in	
  an	
  individual	
  state’s	
  democracy	
  and	
  ICT	
  level	
  between	
   1994-­‐2009	
  (see	
  appendix	
  for	
  state	
  name	
  codes).	
  The	
  bottom	
  dark	
  line	
  represents	
  the	
  overall	
   correlation	
  for	
  all	
  states	
  (OLS).	
  The	
  top	
  line	
  shows	
  correlation	
  within	
  states	
  (fixed	
  effects).	
   Importantly	
  the	
  different	
  angle	
  shows	
  differing	
  strength	
  of	
  ICT	
  on	
  democracy	
  level.	
   	
   	
   37	
   - one relationship for all states - and fixed effects that control for variance between states.34  The purported effects of ICT on democracy in Asia (the OLS line) are partly a function of the different adoption rates of the very democratic states (such as Japan and South Korea towards the bottom right) compared to the very non-democratic states (such as Turkmenistan and Laos) at the top left. Both H2 (negative ICT/democracy correlation) and H1 (positive correlation) can now be safely refuted. An additional, and as yet unobserved, factor is mitigating and offsetting potential democratizing effects that ICT may entail. The missing ingredient may well be the effect of state capacity hypothesized above. Namely, while all Asian states are adopting ICT at varying rates, non-democratic states are concurrently controlling the flow and thus the effects of ICT. H3 now addresses this possibility. 4.5	
  The	
  Missing	
  Piece?	
  Online	
  Censorship	
  as	
  a	
  Proxy	
  for	
  Digital	
  State	
  Capacity	
    H3 postulates that DSC – the ability of the state to control digital information – mitigates against the democratizing effects that ICT may entail. Non-democratic Asian states, as elsewhere, are indeed becoming increasingly sophisticated in building and applying DSC.35 However, as there is yet no measure for this multi-faceted capacity, this analysis leverages a partial but useful proxy in the form of online censorship.36 Online censorship is but one weapon in the state’s arsenal to manage and control digital information, and thus offers only a partial measure of DSC. Data is further limited solely 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   34 See appendix for country codes. The relationship graphed here includes only the Internet/mobile variable and does not control for economic and social development factors. Doing so, as per the models above, would yield a stronger divergence in results with a possible mildly positive slope for the FE line. 35	
  See Chapter 5 for detail, also Diebert et al. (2011) and Kalathil and Boas (2003). 36 The OpenNet Initiative (ONI) (2011a, 2011c) has been compiling data on censorship online. 	
   	
   38	
   to 2009. Nevertheless, this serves as an indicator of the state’s role contra ICT effects and a first step in testing H3. Multivariate regression is now performed for models of ICT against democracy with and without economic and social controls. These are then compared with two models including online censorship.37  Table 4.2 displays results. 38 Model X acts as a baseline and comparable result to previous OLS (non fixed effects) regressions. ICT is a strong and statistically significant (at the 5% level) predictor of democracy for 2009.39 . Adding control variables in model XI strengthens this relationship in magnitude and statistical significance, increasing explanatory power (to an R2 of 0.45). The plot thickens when censorship is added.40  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   37	
  The data set now includes North Korea.	
   38	
  Literacy (% literate pop. above 15) replaces enrolment for 2009 due to greater data availability.	
   39	
  A 1 point more democratic score correlates to around 17 additional Internet/mobile users per 100 pop.	
   40 Setting all variables at their mean, an ICT increase from that of Nepal to that of Malaysia in 2009 (a 70 users per 100 pop. delta) corresponds to a 10-point jump in democracy score, from that of China to Japan.	
   Table	
  4.2	
  -­‐	
  Asian	
  Democracy,	
  ICT,	
  and	
  Online	
  Censorship	
   (multivariate	
  analysis)	
   Dependent	
  Variable:	
  Democracy	
  Level	
  (2=very	
  democratic,	
  14=non)	
  	
   Explanatory	
   Variable	
   (X)	
   	
   Base	
   (XI)	
   	
   With	
   Controls	
   (XII)	
   	
   Censorship&	
   Interactions	
   (XIII)	
   	
   Censorship &	
  Controls	
   Internet/	
   Mobile	
  (t-­‐3)	
   -­‐0.0614**	
   (0.0249)	
   -­‐0.1527***	
   (0.0499)	
   -­‐0.0047	
   (0.0342)	
   -­‐0.0205	
   (0.0646)	
   GNI	
  per	
  capita	
   PPP	
  x	
  10,000	
   	
   1.43	
   (0.906)	
   	
   -­‐0.	
  248	
   (1.088)	
   Log	
  of	
   Population	
   	
   -­‐0.5345	
   (0.7365)	
   	
   -­‐2.9619**	
   (1.0393)	
   Literacy	
   	
   0.0833**	
  (0.0385)	
   	
   0.0057	
   (0.0452)	
   Online	
   Censorship	
   	
   	
   0.5827**	
   (0.2353)	
   0.6626**	
   (0.2329)	
   Interaction:	
   Censorship*	
   IntMob(t-­‐3)	
   	
   	
   -­‐0.0185*	
  (0.0102)	
   -­‐0.0072	
   (0.0095)	
   Constant	
   10.8840***	
  (0.8762)	
   8.5285	
   (6.6594)	
   8.6297***	
   (1.3162)	
   30.2289**	
   (9.9936)	
   R2	
   0.20	
   0.45	
   0.40	
   0.70	
   N	
   26	
   26	
   20	
   20	
   *p	
  <	
  .10	
  **p	
  <	
  .5	
  ***p	
  <	
  .01	
   	
   	
   39	
   First, the ICT term loses magnitude and statistical significance. Second, online censorship becomes a strong and statistically significant (at the 5% level) indicator of less democracy. Third, the interaction of ICT and online censorship is strong and statistically significant (at the 10% level). Finally, the overall explanatory power of the model increases considerably (to R2 of 0.40). When accounting	
   for online censorship, ICT does not indicate more democracy outright. Rather, digital state capacity – online censorship – appears to counteract the effects of ICT.41 Model XIII now includes controls for economic and social development. Both ICT and its interaction with online censorship lose statistical significant (perhaps due to data set size limitations). Online censorship on its own, however, strengthens in magnitude and retains statistical significance (at the 10% level). This is now the dominant indicator of democracy level in a model with considerable explanatory power (R2 of 0.70). Decreasing censorship from the level of Myanmar to that of Bangladesh corresponds to an increased democracy level from that of Lao to Sri Lanka.42  The preceding analyses offers clues on the interplay of ICT and democracy in Asia. It is not that ICT is not correlated with democracy. Rather, the interaction of DSC with ICT limits and modifies any possible democratizing effects that ICT may carry on its own. This is not surprising. Indeed, for many non-democratic Asian states this is the underlying justification for an online censorship policy. Since the full battery of DSC counts infrastructural, legal, and coercion elements as well, online censorship provides 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   41 For Pakistan or China in 2009, for example, removing online censorship is correlated with a 2 point more democratic score.	
   42	
  Online censorship at the highest data level corresponds to a 5-point less democratic score compared to a state without censorship.	
   	
   	
   40	
   but a single tool in this arsenal. The true counter-democratizing force of DSC may be far reaching. With H3 appearing promising, the final analysis aims for a fuller validation. 4.6	
  Completing	
  the	
  Puzzle:	
  Communications	
  and	
  State	
  Capacity	
  Combined	
    The goal now is construction of a model that incorporates lessons from the previous three analyses. This aims to demonstrate the interaction of ICT and state capacity over time in Asia again leveraging censorship data. But since this is available only for 2009 two assumptions are introduced. First, it is assumed that states that employ online censorship in 2009 did so, at some level, in previous years.43 Second, since censorship is but a proxy for a multifaceted capacity, it is further assumed that states with online censorship in 2009 were more likely to use other means of controlling digital information as well. These assumptions are incorporated into a dummy variable that is populated for all states and all years. Model XIV applies OLS time series cross sectional analysis, while XV uses state fixed effects. 44 Results are in Table 4.3.  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   43	
  In other words, censorship levels are assumed not to have increased from zero in 2008.	
   Table	
   4.3	
   -­‐	
   Asian	
   Democracy,	
   ICT,	
   and	
   Online	
   Censorship	
  Over	
  Time	
   (Time	
  Series	
  Cross	
  Sectional	
  Analysis)	
   Dependent	
   Variable:	
   Democracy	
   Level	
   (2=very	
   democratic,	
   14=non)	
   Explanatory	
   Variable	
   (IV)	
   	
   Base	
   (XIV)	
   	
   Online	
   Censorship	
   (XV)	
   	
   Censorship	
   Fixed	
  Effects	
   Internet/	
   Mobile	
  (t-­‐3)	
   -­‐0.0544***	
   (0.0147)	
   -­‐0.0157***	
  	
  	
   (0.0060)	
   -­‐0.0091	
  	
  	
   (0.0101)	
   GNI	
  per	
  capita	
   PPP	
  x	
  10,000	
   0.0311	
   (0.161)	
   0.056	
   (0.081)	
   -­‐0.0535	
   (0.457)	
   Log	
   of	
   Population	
   -­‐1.2343***	
   (0.0844)	
   -­‐1.8259***	
  	
  	
   (0.0412)	
   6.6491**	
  	
  	
   (3.2301)	
   Education	
   Enrolment	
   -­‐0.0358***	
   (0.0133)	
   -­‐0.0591***	
  	
  	
   (0.0108)	
   -­‐0.0774***	
  	
  	
   (0.0146)	
   Online	
   Censorship	
   	
   2.9683***	
  	
  	
   (0.3970)	
   1.1993**	
  	
  	
   (0.5309)	
   IntMob*	
   Censorship	
   	
   -­‐0.0710***	
  	
  	
   (0.0109)	
   0.0223**	
  	
  	
   (0.0086)	
   Constant	
   22.1236***	
  (1.4792)	
   26.675***	
  	
  	
   (1.0926)	
   -­‐40.8598	
  	
  	
   (25.5578)	
   R2	
   0.21	
   0.43	
   0.94	
   N	
   326	
   326	
   326	
   *p	
  <	
  .10	
  **p	
  <	
  .5	
  ***p	
  <	
  .01.	
   	
   	
   41	
   Model IV is included as a baseline comparison. In model XIV all variables (with the exception of economic development) are statistically significant at the 1% level. While both ICT and its interaction with online censorship correlate with more democracy, online censorship correlates with less democracy. This demonstrates the bifurcation predicted by theory and hinted at in the previous analysis. While DSC dominates as a (non) democratic predictor for low levels of ICT, this relationship flips for higher levels.45 This offers an initial validation of H3 on the interplay between capacity and communications. However, as above, OLS may be demonstrating inter-country differences over of intra-country effects.46  Model XV, using fixed effects, addresses this limitation. DSC and its interaction 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   44	
  As discussed above, this offers both a holistic inter-country and a reductionist intra-country view.	
   45	
  For Internet/mobile rates of 0-34 users per 100 pop., DSC correlates to democracy level. Above this, however, ICT is correlated with more democracy at the rate of 1 point for each additional 12% of users.	
   46 Given variation between states in the data set, this may not be applicable at the individual state level.	
   BGD BTN BRNKHM JPN KAZ PRK LAO MYS MNG NPL LKA TKM UZB Fixed Effects OLS 0 5 10 15 De mo cr ac y Sc or e 0 20 40 60 80 100 Internet/Mobile per 100 pop. Figure	
  4.3	
  –	
  Non-­‐Censoring	
  States:	
  ICT	
  and	
  Democracy	
  in	
  Asia	
   Coloured	
  lines	
  show	
  change	
  in	
  individual	
  state’s	
  democracy	
  and	
  ICT	
  levels	
  between	
   1994-­‐2009	
  (see	
  appendix	
  for	
  state	
  name	
  codes).	
  The	
  dark	
  line	
  represents	
  overall	
   correlation	
  for	
  states	
  (OLS).	
  The	
  yellow	
  line	
  shows	
  correlation	
  within	
  states	
  (fixed	
   effects).	
  Different	
  angles	
  show	
  differing	
  strength	
  of	
  ICT	
  on	
  democracy	
  level.	
   	
   	
   42	
   with ICT (both at the 5% level) are now the predictors of less democracy. 47 As expected, explanatory power increases (to an R2 of 0.94), but at the cost of statistical significance on the ICT term.  Although insufficient data may be at fault, this offers a cautionary note to over-simplistic interpretation of the holistic model above. For individual states, ICT is not necessarily an indicator of democracy level. Rather, DSC offsets and counteracts possible democratizing effects. Figures 4.3 and 4.4 display this graphically.48 Figure 4.3 shows OLS, fixed effects, and individual country regressions for  states without online censorship, while Figure 4.4 offers a comparable view for states with online censorship. OLS offers a similar ICT/democracy correlation for both: the inter- country variation effect discussed above. For states without censorship, fixed effects demonstrates a similar, albeit weaker relationship. ICT here is correlated with more democracy within 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   47	
  The assumption of common intercepts is relaxed to account for disparate democracy starting points.	
   48	
  As with similar graphs above, these display only ICT against democracy without any control variables. CHN IND IDN KOR KGZ MMR PAK PHL TJK THA VNM Fixed Effects OLS 0 5 10 15 De mo cr ac y Sc or e 0 20 40 60 80 100 Internet/Mobile per 100 pop. Coloured	
  lines	
  show	
  change	
  in	
  individual	
  state’s	
  democracy	
  and	
  ICT	
  levels	
  between	
   1994-­‐2009	
  (see	
  appendix	
  for	
  state	
  name	
  codes).	
  The	
  dark	
  line	
  represents	
  overall	
   correlation	
  for	
  states	
  (OLS).	
  The	
  yellow	
  line	
  shows	
  correlation	
  within	
  states	
  (fixed	
   effects).	
  Different	
  angles	
  show	
  differing	
  strength	
  of	
  ICT	
  on	
  democracy	
  level.	
   Figure	
  4.4	
  –	
  Censoring	
  States:	
  ICT	
  and	
  Democracy	
  in	
  Asia	
   	
   	
   43	
   states49 For censoring states, however, this relationship is reversed: ICT is increasing as democracy level is decreasing. This confirms H3: the putative effects of ICT on democracy are subject to DSC. 	
  4.7	
  Results:	
  The	
  Digital	
  Asian	
  State	
  Strikes	
  Back	
  	
   The analyses in this chapter demonstrate that there is no simplistic uni-directional relationship between ICT and democracy in Asia. This dispels alternative theories that see ICT’s effects as either inherently democratizing or non. As predicted in the theory, results here suggest50  that although ICT may have democratizing effects these are subject to whether and how the state employs DSC. Although these are tentative results, clearly Asian state policy plays a role in the ICT/democracy relationship. But what is this role? How do Asian states control and manage digital information and what does this digital state capacity look like? The next chapter describes this in detail. 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   49 I.e. controlling for differences between counties.	
   50 H3 can only be tentatively accepted. Full validation requires a comprehensive measure of DSC – not simply 2009 online censorship. Also this analysis points only to correlations postulated and not causality.	
   	
   	
   44	
   Chapter	
  5	
  	
   Digital	
  State	
  Capacity:	
  Bringing	
  the	
  Asian	
  State	
  Back	
  In	
  	
   “Government jurisdictions are geographic. The Internet knows few boundaries. The clash between the two will reduce what individual countries can do. Government sovereignty, already eroded by forces such as trade liberalization, will diminish further.”1	
  	
   “Instead of a World Wide Web … it is more accurate to say we have a …an Uzbek Wide Web, a Pakistani Wide Web, a Thai Wide Web, and so forth. The theory of ‘‘unregulability’’ no longer has currency, if ever it did.”2	
   	
   	
   5.1	
  Whither	
  the	
  Digital	
  Asian	
  State?	
    The previous chapter demonstrated the role of the state in countering the possible democratizing effects of ICT over time in Asia. This chapter conceptualizes and constructs this facet of state capacity – digital state capacity (DSC) in Asia. I hypothesize that ICT adoption empowers societal actors at the expense of centralized states and, given low government legitimacy, can prove a factor in political change. But states mitigate these ICT effects by building and applying DSC. Asian non-democratic states, as this chapter shows, have been at this forefront of utilizing DSC.  In the early days of mass digital communications and especially the Internet, a bottom-up and distributed architecture melded easily with teleological views of Western democratic liberalism as an end of history. 34 If democratic capitalism was the end of history, globalization - driven in part by the information revolution - was its vanguard chipping away at the power of states. But history did not end. Non-democratic states 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   1 Cairncross 1997, p.177. 2	
  Zittrain and Palfrey 2008, p.31. 3	
  Fukuyama 2006. 4	
  The Internet was seen as counter to the nation-state. Lessig (2004, p.5) summed up this utopian vision as follows: “Cyberspace is unavoidable, and yet cyberspace is unregulable [sic.]. No nation can live without it, yet no nation will be able to control behaviour in it. Cyberspace is that place with individuals are, inherently, free from the control of real space sovereigns.” 	
   	
   45	
   have not lain supine in the face of this onslaught but rather have reasserted power over political communication. State political communication costs have shifted from production and distribution to management and control. 5 Asian states have created a new policy arsenal including laws and regulations, technology, and online social norms, all seasoned with coercion.6 Individual state policy is a unique combination of these along a spectrum from prioritizing self-censorship to arrests of cyberdissidents. Nor are Asian state policy objectives homogenous, but rather span from promotion of national identity and social morals to countering opposition movements and cybercrime. 7  This chapter unfolds as follows. After defining and conceptualizing DSC and the actors involved in its creation and implementation, the focus shifts to categorizing and enumerating its dimensions. These include infrastructure, censorship, law and regulations, and coercion. In combination these form distinct national information control policy packages. 5.2	
  Conceptualizing	
  DSC:	
  One	
  Part	
  Orientation,	
  Two	
  Parts	
  Ability,	
  and	
  Mix8	
  	
   The concept of DSC combines power and will in the political communications realm. It is the institutional ability and regime ideological orientation to control and manage the production, dissemination, and consumption of digital information in line with state policy objectives. 9  DSC is separated conceptually into potential ability resting on underlying institutions and application premised on policy objectives and their driving 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   5	
  Previously, mass media was expensive to produce and distribute, but relatively easy and cheap  - as in contraband films and books – to destroy. ICT has flipped this relationship and costs for states have shifted from production and distribution to management and control (Howard 2011). 6	
  Zittrain and Palfrey (2008) focus on the first three, but I argue that DSC also includes coercion as well. 7	
  Dieber et al. 2011 8	
  I would like to thank Professor Pitman Potter (UBC) for his advice and suggestions in framing DSC by separating it into ability and orientation and the analogy to institutional capacity (Potter 2004). 9	
  See chapter 3 for a discussion of how DSC is positioned within the broader state capacity literature. 	
   	
   46	
   ideology.	
   Orientation, defined as “the priorities and habitual practices that inform institutional performance,”10 normally drives accumulation of ability. For example, Vietnam prioritizes and has built technical mechanisms in order to censor foreign-based opposition movements.11 But this relationship may be reversed in cases where ability makes application increasingly tempting. Centralized communications infrastructure in Egypt became a target for government shut down during the 2011 Jasmine revolution.12 Contingent on context, orientation and ability vary considerably between states.  Orientation ranges in degree of codification and formality, with some states clearly defining DSC policy objectives to others preferring the (often arbitrary) power that ambiguity provides.13 Ability is further subdivided into four institutional aspects: financial, since information control bureaucracies are expensive to build and maintain; technical, as monitoring and filtering tools require development and customization;14	
   and human resources, in both quantity, or numbers of censors, and bureaucratic quality.15	
  The fourth factor is institutional corporate cohesion, or “the willingness of individuals within institutions to comply with edicts from organizational and extraorganizational [sic] leaders and to enforce institutional goals.”16 This is particularly germane to non-democratic states with (often purposefully) ambiguous information 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   10	
  Potter 2004, p.473 11 Diebert et al. 2011 12 Glanz and Markoff 2011; This specific capabilty is discussed below. 13	
  Thailand’s codification of lèse-majesté laws (McCargo,2006) is juxtaposed here against China’s often- ambiguous information censorship guidelines for corporations (MacKinnon 2011), a product of balancing ideological orthodoxy and loyalty requirements against legal formality (Potter 2004). 	
   14 These can be purchased (MacKinnon 2011), but extensive usage necessitates local expertise. 15	
  It takes high-quality institutions to properly operate extensive DSC regimes. An analogy for bureaucratic quality is made, for example, between China and India, in redistributive state capacity towards welfare policy. India’s bureaucratic nepotism, low level of professionalism, and lack of corporate cohesion, in comparison to China, have sabotaged this ability (Evans 1995, Das 2006, Economist 2012c). 16 Potter 2004, p.474 	
   	
   47	
   control regulations leading to disparate policy interpretation. Even with clear policy and high corporate cohesion, the state is not a sole actor in DSC application. 5.3	
  Locating	
  Agency:	
  the	
  State	
  and	
  Its	
  Intermediaries	
  	
   DSC rests on an ecosystem of actors spanning beyond the state to intermediaries and citizens. These three levels - state, intermediary, and citizen - exist in all cases, but their roles and the resulting agency topology depend on contextual factors such as ICT penetration level, degree of telecommunications privatization, and social norms. In effect, this determines loci of pressure: where information is monitored, filtered, controlled, or blocked. 17 The diffusion in production and dissemination of information that ICT entails increases possible choke-points, and while the state sets the control policy agenda (in boundaries and ideology) topology dictates which actors become key proxies. The mechanism at play bears considerable similarity to that of infrastructural capacity, describing how Asian developmental states implement economic policy with and through key non-state intermediaries.18  In DSC, the state never acts alone.  An intermediary is any non-state actor located between end users and the information transactions they are trying to conduct.19 The list of possible intermediaries extends to hardware, software, and network infrastructure providers as well as those necessary for digital transactions, such as encryption providers and financial companies. But some of the most active DSC intermediaries, however, are service (ISP) and content 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   17 This typology is partly based on Goldsmith and Wu (2006) who describe a range of conceptual censorship cases based on which information intermediaries are under the state’s purview. 18 See Evans’ (1995) Embedded Autonomy thesis; Mann (1984, p.113) defines infrastructural capacity as “the capacity of the state to penetrate civil society, and to implement logistically political decisions.” 19 While this discussion focuses primarily on DSC mechanisms in non-democratic states, the argument of the role of intermediaries in information control has been applied to democratic states and traditional media. A powerful example is Chomsky and Herman’s (2002) Manufacturing Consent thesis. 	
   	
   48	
   (ICP) providers.20 The type and degree of proxy pressure depends on the intermediary- government relationship and is predicated on licensing, registration, liability, and self- censorship requirements. States modify the strength of this infrastructural capacity by varying formality of intermediary ties and regulatory clarity.21 China, for example, mixes industry-level self-censorship pledges, liability threats for user-generated content, and regulatory ambiguity for a cocktail of ‘private-public filtering’.22 Jurisdiction is key, with some intermediaries changing jurisdiction in response to government demands and others complying despite being beyond the reach of the law.23  Intermediaries are central DSC proxy actors, but citizens themselves are often equally active participants.  Citizen participation in DSC implementation can be broadly divided into self- censorship, driven by a combination of social norms and fear of coercion, and ‘online commentators’24 operating in line with state policy objectives. All states set boundaries on allowable and acceptable public information.25 In Asian states with considerable DSC, however, while intermediaries may receive information control guidelines, netizens often need to deduce where the digital limits of the allowable lie. Amorphous 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   20	
  Zittrain and Palfrey 2008, Goldsmith and Wu 2006; An interesting extension to this definition consists of criminal organizations that surreptitiously act as DSC proxies, through DDOS and other targeted attacks. They do so in return for governments turning a blind eye to other activities (Dieber et al. 2011). This carries the added benefit of plausible deniability for the government involved. 21 Deibert and Rohozinski (2008) coin this proxy level ‘second-generation’ controls since they offer more nuanced and powerful capability compared to earlier wholesale information blocking. 22	
  MacKinnon 2011; Details on China’s censorship requirements were leaked in 2009 by a Baidu employee offering a rare glimpse of DSC’s operation through a large intermediary (Xiao 2009). 23	
  A famous case of the former is Google’s decision to leave the Chinese market (partly) in response to information monitoring and filtering demands made by the Chinese government (Stone and Xin 2010); An example of the latter includes Youtube’s decision to filter content violating Thailand’s lèse-majesté laws for users in Thailand (OpenNet Initiative 2008); Jurisdiction was a central question also for Research In Motion’s (RIM) encryption operating in the Middle East and South Asia (Bremmer 2010). 24	
  ‘Online commentators’ are individuals recruited by the state or its institutions to produce information that appears organically generated and non-directed. An analogy is made to ‘crowdsourcing’ which operates (primarily) on a voluntary basis (Estelles-Arolas and Gonzalez-Ladron-de-Guevara 2012). 25 This is as true of democratic as non-democratic states. France, for example, prohibits propagation of Nazi ideology. This has led to litigation over online sale of Nazi paraphernalia (Delta and Matsuura 2011). 	
   	
   49	
   boundaries can be combined with selective and public coercion to strengthen the state’s hand. Anti-anonymity regulations, such as real-name registration requirements, further enhance the perceived and actual risk netizens face. 26 The net result is that while the few and intrepid may risk crossing from the allowed to the disallowed, the majority errs comfortably on the side of caution. The second, newer and growing, facet of citizen-level DSC involves surreptitious government-sponsored information production and dissemination agents – or online commentators. 27  States recruit citizens to post information in line with policy ideology and objectives, producing a perception of public support. Both Russia and China, for example, have enhanced their DSC through recruitment of online citizen proxies.28 The state drives DSC orientation and ability, but operation requires the parts and pistons of intermediaries and proxies. 5.4	
  The	
  Sum	
  of	
  Its	
  Parts:	
  A	
  Taxonomy	
  of	
  Digital	
  State	
  Capacity	
  	
   When the political will to obstruct certain information and views is coupled with … an efficient and technically competent bureaucracy, an established regime of political intimidation and surveillance, and embedded corporatist structures facilitating cooperation between state officials and administrators across the public and private sectors, you have a formidable mix.”29  This describes one of Asia’s earliest and most-emulated information control regimes: Singapore. 30  DSC is indeed always multifaceted. In Singapore DSC combines infrastructure design that routes traffic through government servers, legal instruments 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   26 The state’s objectives can often readily be achieved by targeting few and select individuals. As Goldsmith and Wu (2006) argue, the objective here is to raise the perception of cost associated with illicit activities beyond the individual calculated threshold benefit. See Bueno de Mequita and Cohen (1995). Since 2008, for example, South Korea has required real-name registration for blog posting (Fish 2009). 27	
  Diebert et al. (2011) see this as part of the progressive evolution of state online information control from wholesale blocking, to nuanced filtering, to crowdsourcing and opinion management. 28 Calingaert 2010, Bandurski 2008; Also  the next chapter for China’s Internet commentators. 29 Rodan 1998, p.88. 30	
  Ibid. Both China and Vietnam, for example, studied Singapore’s information control regime in the late 1990s, borrowing and applying select elements. 	
   	
   50	
   with manoeuvre room for government interpretation, a high level of litigiousness, and purposefully infrequent coercion resulting in a climate of self-censorship over political views and information.31 Each state is unique, and not all Asian non-democratic states have so thorough an approach. Nevertheless a common set of DSC policy tools has emerged consisting of infrastructure modifications, censorship mechanisms and bureaucracies, legal and regulatory instruments, and coercion (see Table 5.1).32  Table	
  5.1	
  –	
  Digital	
  State	
  Capacity	
  Typology DSC DIMENSION MEANING EXAMPLES INFRASTRUCTURE Underlying	
  telecommunications	
  systems,	
   including	
  network	
  topology	
  and	
  external	
   connections	
   IXP	
  Choke	
  Points	
   Kill	
  Switches	
   CENSORSHIP Technical	
  and	
  manual	
  tools	
  for	
  controlling	
   and	
  monitoring	
  digital	
  information,	
  as	
  well	
   as	
  the	
  resources	
  to	
  operate	
  these	
   Filtering	
  Technology	
   Information	
  Control	
   Bureaucracy REGULATORY Any	
  legal	
  or	
  regulatory	
  instrument,	
   including	
  industry-­‐specific	
   Laws	
   Industry	
  Regulations	
   Industry	
  Pledges	
   COERCION Compelling	
  individuals	
  to	
  act	
  or	
  avoid	
   acting	
  in	
  specific	
  ways	
  concerning	
  digital	
   communications	
  activity	
   Arrests	
  /	
  Intimidation	
   	
   	
  	
   	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   31	
  Rodan 1998; ONI 2007; Singapore combines several legal instruments, industry registration requirements, and codes of conduct to control unwanted political (and social information). 32	
  Self-censorship based on societal norms could be considered an additional dimension of DSC. But since the objective here is to conceptualize DSC in a way that lends to future operationalization and measurement, social norms are excluded. 	
   	
   51	
   5.5	
  Looking	
  Under	
  the	
  Hood:	
  The	
  Infrastructure	
  Dimension	
  	
   The world of information is flat: 33  millions of nodes and connections, infinitely distributed, and completely redundant; so goes the popular perception of the underlying network infrastructure of ICT. In fact, these networks have highly uneven traffic flow distributions similar to that of highway commuter traffic.34 States have a hierarchy of information hubs and a limited set of external connections to the wider world. It is at these points that large volumes of information can be monitored, filtered, blocked, and controlled, and where a primary infrastructural dimension of DSC is located. 35 “Information architecture is politics,”36 and in many states, these connections remain firmly under government control.37  But some states go further. North Korea has centralized all external information access through a government-run provider for a select few, and established a parallel, disconnected, and isolated national intranet, known as Kwangmyong, for ordinary citizens.38 Controlling key geographic hubs offers the added advantage of regional or national ICT shut down capability, otherwise known as a ‘kill switch’.39 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   33	
  Taken from Friedman (2007), who used this phrase for globalization, partly resting on the global availability of information communication technology. 34	
  Technically, these are scale-free networks (following a decreasing power law and not bell curve), where a small number of hubs takes the lion’s share of information (Homer-Dixon 2006). 35 See Blum (2011); Nor is this a capacity limited to non-democratic states. The U.S. NSA, for example, monitors digital communication through mirroring facilities at several information hubs (US Congress 2007; Singel 2007). 36 Howard et al. 2011, p.9. 37	
  In China, Internet exchange points are operated by a select number of SOEs (Harwitt 2008). 38	
  Economist 2007; ONI 2007b. In effect national intranets rely on white lists of allowed not black lists of disallowed destinations. Other states are following in North Korea’s footsteps. Iran is in the process of designing a sanitized internal Internet known as the Halal Network (Fasshi, 2012). 39 Infrastructure centralization also offers bandwidth throttling. Burma has used this to cap upload bandwidth for large media to minimize uploading sensitive videos (Reporters Without Borders 2011). 	
   	
   52	
    Since infrastructure vulnerabilities can be exploited, they are increasingly being designed-in. The most infamous application case is perhaps the Egyptian regime’s shut down of external communications during the January 2011 Arab Spring Revolution,40 but several Asian examples are equally telling. National-level shut down occurred, for example, in Burma in September 200741 and in the 2005 Nepal coup42 following declaration of marital law by King Gyanendra.43 China has implemented a tiered shut down capability. Riots in Xinjiang Province in 2009 triggered shut down for mobile and Internet services throughout the province.44 Kill switch usage, while powerful, carries considerable costs since it not only disables a communication and economic resource, but draws unwanted attention to contentious areas.45 It is reserved as a last resort. Rather, the bread and butter of DSC lies in censorship. 5.6	
  Government	
  Knows	
  Best:	
  The	
  Censorship	
  Dimension	
  	
   While infrastructure design centralizes information flow, filtering and blocking capability more granular than a kill switch relies on censorship. Asian states vary considerably in their digital censorship capacity, allocation of human resources, scope of issues filtered, and the depth of coverage.46 DSC censorship is the state’s ability to pervasively filter and limit information based on defined (and often dynamic) criteria in 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   40 External access had been purposefully centralized in one facility in Cairo. The regime forced this facility off line disconnecting most Egyptians from global communications for 5 days (Glanz and Markoff 2011). 41 Wang 2007, Chowdhury 2008. 42	
  Jones 2005. 43	
  An interesting analogy occurred in Pakistan in 2007 as President Musharraf shut off broadcast media capabilities, driving several media outlets to provide news online (Committee to Protect Journalists 2009). The lesson being perhaps that weak DSC can be worse for the state’s objectives than none. 44	
  Xinhua, 2009 45	
  For an example of the costs of kill switch usage, in this case on academic research, see Stone (2009). 46 This section relies heavily on the OpenNet Initiative’s methodology for measuring and classifying online censorship regimes (ONI 2011a). 	
   	
   53	
   line with policy. 47  This normally relies on blocking of IP addresses, URLs, or DNS entries for websites and platforms, as well as key words and search results filtering.48 An additional, offensive censorship mechanism favoured, for example, by Vietnam, uses DDOS, or “censorship by information overload,”49 to render an offending website or service inoperable. Policy dictates technique50 based on whether the state chooses to inform users that the information is forbidden or simply non-existent. The choice of physical location varies from backbone, IXP to ISP/ICP, institutional, or individual computer levels, depending on a state’s communications environment and resources.51  Granular censorship regimes beyond wholesale blocking used to require considerable development and customization above the technical know-how of many developing states. The growing participation of private companies, however, has greatly increased scope of usage in Asia.52 But technology is always limited. To be successful, the shifting, dynamic, and ever-more-complex nature of filtering necessitates investment in human resources to wield the tools of censorship. Thailand, for example, 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   47	
  Censorship also extends to silent monitoring of digital communication, such as emails, that becomes an input to coercion effectively silencing individuals. In a controversial 2005 case in China, for example, editor Shi Tao was charged with antistate activity based on private emails provided by Yahoo (CPJ 2010). 48	
  A technical discussion of methods is beyond scope, but briefly IP and URL filtering is based on blacklisting a specific website address or domain name, and this can extend to individual services (such as email) through port blocking. DNS blocking eliminates the mapping between a website or domain and its corresponding IP address altogether; A platform blocking example is Thailand’s 2007 full ban on Youtube following posting of several videos deemed insulting to King Bhumibol Adulyadej (Jackson 2007). 49	
  CPJ 2010, p.8; DDOS, or distributed denial of service attacks, were used successfully by Vietnam in 2010 against sites that opposed Vietnam’s bauxite mining policies (Ruwitch 2010). 50	
  MacKinnon (2009) offers examples in a study of censorship across Chinese blogging services. When posting and searching for disallowed information, responses ranged from closing the user’s connection, to indicating a non-existent destination, to indicating that the post contained forbidden content. 51	
  ISP-level censorship is common if a limited number of ISPs are licensed and controlled. Institutional censorship is applied, for example, at corporate or university level. Individual computer censorship is rare, but was attempted, in China’s Green Dam PC-based software (Diebert et al. 2011). 52	
  The role - or complicity (Cellan-Jones 2009)– of Western-based companies in Asian censorship regimes has led to the formation of an as-of-yet limited effectiveness industry organization known as the Global Network Initiative. Zittrain and Palfrey 2008, Maclay 2010, MacKinnon 2011. 	
   	
   54	
   maintains a ‘war-room’ for fighting its lèse-majesté battles,53 and China has built the world’s largest censorship bureaucracy. 54  As with other facets of state capacity, successful application of DSC depends on a professional, highly trained, and corporatist bureaucracy. But what are Asian states actually filtering and censoring?  The answer varies between states in topics and censorship breadth, but the OpenNet Initiative offers an instructive classification. This divides censored topics into: political, such as regime opposition; social, including religious and moral issues; conflict, such as secession movements; and access to specific communication tools, such as foreign-based social networks.55 Thailand’s narrow focus on royalty protection is contrasted, for example, with censorship across all categories in Burma, China, and North Korea; most categories in Vietnam; social and political categories throughout Central Asia; conflict (between the Koreas) in South Korea, and social categories in Singapore.56 Asian states also vary in the consistency of application from persistent to sporadic, with the latter often leveraged as policy signalling.57 In this case, the frequent bark of censorship is used instead of its infrequent bite. With censorship mechanisms in place, however, temptation exists for mission creep widening censorship scope over time.58  Censorship exists in parallel with a DSC legal and regulatory environment. 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   53	
  Fuller 2011. 54	
  Diebert et al. 2011 55	
  OpenNet Initiative 2011a. 56	
  Diebert et al. 2011. 57	
  Contrast South Korea’s pervasive filtering of North Korean propaganda and unification information (Diebert et al. 2011) with censorship in Malaysia and Singapore. In Singapore, for example, occasional and public censorship acts as a warning on the direction and emphasis of policy (Rodan 1998). 58	
  Accusations of mission creep, for example, have been leveled against Thailand in political application of lèse-majesté (The Nation 2008), and Pakistan, which censored access to Facebook in response to a page encouraging visitors to submit drawings of Mohammed (Imtiaz 2010, Diebert et al. 2011). 	
   	
   55	
   	
   5.7	
  Regulating	
  Cyberspace	
  and	
  Real	
  Space:	
  The	
  Legal	
  Dimension	
  	
   The legal dimension of DSC extends beyond legislation and regulation governing information control to the transparency and consistency of its application.59 Legal DSC is as much about perception of risk and coercion as about detailed laws. A salient theme in this chapter is that lack of clarity over boundaries of allowable information can, in fact, be a key policy tool.60 Perception of risk and ambiguity over boundaries, together, make an effective silencer. Even with a degree of transparency in legislation, consistency depends on the existence of an administrative or judicial appeal facility for those wrongly or incorrectly censored. In fact, there is considerable state variance here. While Thailand provides for some appeal, China, for example, offers no official recourse resulting in an environment of heightened ambiguity.61 But boundaries can be gleaned by looking at how legal instruments join with intermediary regulations in empirical application. Digital information legislation focuses state policy while locating specific action – direct censorship or punitive consequences for its failure - at the citizen or service provider level. Consequently application can be located in Burmese Internet café regulations or Singapore’s political/religious content provider registration requirements.62  At the citizen level, DSC regulations, and punitive consequences for 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   59	
  Transparency and application consistency are factors in degree of Rule of Law, defined by as “the extent to which agents have confidence in and abide by the rules of society…the quality of contract enforcement, property rights, the police, and the courts …the likelihood of crime and violence” (World Bank 2011). 60 This ‘ambiguity principle’ offers government interpretation scope and strengthens corporate intermediary and citizen self-censorship. 61 In Thailand censoring a website requires a court order (OpenNet Initiative 2007a). In China, on the other hand, there is no judicial procedure, and this has been flagged as a violation of WTO market access commitments over cross-border digital services (WTO 1994; Hindley and Lee-Makiyama 2009). 62	
  Burma’s Internet café regulations focus on operator proxy DSC (Xinhua 2010); In Singapore, the Class License Scheme governs registration of political /religious content providers (OpenNet Initiative 2007a). 	
   	
   56	
   their violation, rest on user identification. South Korea has implemented anti-online- anonymity legislation originally touted to protect netizens from cybercrime, but since used for politically-expedient prosecution.63 As in Burma, China has long required user identification and monitoring in Internet cafes, but more recently has been pursuing comprehensive anti-anonymity regulations for popular microblogging services. 64 Finally, governments also use dual-tiered pricing strategies to limit foreign information access in favour of national intranets.65 For those intrepid (or foolhardy) who exceed allowable regulatory information boundaries, coercion presents an additional barrier. 5.8	
  Squeezing	
  Online	
  Dissent:	
  The	
  Coercion	
  Dimension	
  	
   The final facet of DSC is the level of coercion – intimidation, arrest, torture, execution - the state is willing to apply to achieve its information control ends. Several organizations are focused on monitoring this, extending the erstwhile definition of journalist to include bloggers, netizens, and those who support them.66 Given the role of ICT in non- democratic Asian states, it is perhaps not surprising that the greatest number of arrests is now for online publication and expression than for any other medium.67 Individuals are often singled-out, cases well-publicized, and timing honed to maximize perception of risk, self-censorship, and increase intermediary pressure. Pakistan’s 2007 election, for example, saw 120 journalists arrested and 30 injured, in addition to a swath of 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   63	
  A famous case is the prosecution of Park Dae-Sung, a vocal critic of government policy who blogged under a pseudonym, but was exposed under government pressure by his ISP (Wall Street Journal 2009). 64	
  Internet cafes are required to monitor and maintain a digital trail of customer activity (Branigan 2011); For recent Weibo real-name registration changes, see China Daily (2011) and Zhu and Zhang (2012). 65	
  Howard 2011, p.270; For example, price controls are part of Vietnam’s DSC, see ITU (2002). 66	
  Committee to Protect Journalists (2010); Reporters Without Borders (2011). 67 Committee to Protect Journalists 2010. 	
   	
   57	
   fraudulent government lawsuits against reporters and editors.68  Excessive litigation and its associated intimidation have also feature prominently in Singapore against critiques of government officials.69 Although coercion is infrequently used in comparison to the other DSC dimensions, its power lies in the perception of punitive and liability risk online contributors incur. Through this it is an effective control. 5.9	
  DSC	
  Policy	
  Suites:	
  More	
  Than	
  The	
  Sum	
  of	
  Its	
  Parts	
  	
   This chapter has laid out the constituent parts of information control in Asian states through the lens of DSC.70 It provides an analytical prism for separating what in application is a continuum of policy instruments acting in concert. In Vietnam, for example, infrastructure, such as telecommunications architecture through government proxies, is combined with censorship, such as elimination of opposition movement content, legislation, through registration and pricing for external access, and aggressive litigation for dissenting publishers.71 This policy suite is enforced through an ecosystem of government and private intermediaries, as well as direct pressure on citizens. The result is a chilled communications climate and a contained public sphere.  While the previous chapter could only detect the role of DSC in opposing ICT effects, this chapter has solidified this emerging state capacity. It has outlined an ecosystem of participating actors and peeled back the constituent DSC layers: infrastructure modifications, censorship regimes, regulatory framework, and coercion. 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   68	
  Reporters Without Borders 2008. 69	
  Rodan 1998.	
   70 Operationalization of DSC is not within scope, but the conceptualization of components provided – infrastructure, censorship, legal environment, and coercion – are measurable, although these have not yet been combined into a single indicator. 71	
  Boas and Kalathil 2003. 	
   	
   58	
   The chapters thus far have outlined the underlying theory on ICT and its political effects in states with low legitimacy, have demonstrated that both ICT and DSC play a role in democratic Asian change over time, and have detailed the nature of DSC. The next chapter analyzes how ICT, legitimacy, and DSC interact for the exemplar of China.  	
   	
   59	
   Chapter	
  6	
  	
   Qualitative	
  Analysis:	
  Chinese	
  Political	
  Space	
  a	
  Click	
  at	
  a	
  Time1	
   	
   “There’s no question China has been trying to crackdown on the Internet. Good Luck! [Laughter] That’s sort of like trying to nail Jell-O to the wall.”2   - US President Bill Clinton, March 8, 2000  By linking with the Internet, we don’t mean absolute freedom of informa- tion. I think there is a general understanding about this. If you go through customs, you have to show your passport. It’s the same with management of information. There is no contradiction at all between the development of telecommunications infrastructure and the exercise of state sovereignty.3   - Wu Jichuan, China Minister of Post and Telecommunications,   November 25, 1995	
   	
   6.1	
  Introduction	
  –	
  Why	
  China	
  and	
  Why	
  Now?	
    With evidence for the role of DSC in the interplay between ICT and political change in Asia (chapter 4) and an understanding of DSC (chapter 5), this chapter now turns to an analysis of the theory outlined (chapter 3) for China. It offers a preliminary qualitative assessment of the interaction of ICT and state capacity. China presents a unique case as a state combining high (and increasing) ICT penetration, high regime legitimacy, and considerable (and growing) degree of DSC. These frame a country undergoing tectonic shifts - and occasional ruptures - in its economic and political power 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   1	
  From an Economist issue on China communications, “The Chinese is being dragged click by click out of its cone of silence” (Economist 2012). 2	
  US National Archives and Records Administration 2005, p.404 3	
  Quoted from Wu 1997, p.647. Originally quoted in The Straits Times, November 25, 1995. Figure	
  6.1	
  –	
  ICT	
  /	
  Legitimacy	
  /	
  DSC	
  Levels	
  in	
  China	
   	
   	
   60	
   distribution. China’s political trajectory carries global ramifications, and digital contention lies at its heart.  The central argument in applying the theory outlined to China runs as follows. The CCP has promoted the adoption of ICT as part of China’s overall economic development. Consequently, ICT adoption levels in China are high and rising, and this has inexorably led to diffusion in the political economy of communications. The CCP continues to enjoy a high level of legitimacy. To mitigate against possible political ICT effects, the CCP has built considerable DSC, delimiting permitted public debate and narrowing the boundaries of disallowed political communications in the process. Issues falling outside these have become contested terrain to which societal actors have laid claim. China does not exhibit the High ICT/Low Legitimacy/Low DSC combination necessary for political change at the national level. However, these conditions exist in some cases at local and provincial levels. I argue that ICT adoption in China has resulted in a more transparent, accountable, and representative Chinese government.  It is beyond scope to provide a full case study of ICT and DSC in China, but this chapter glimpses their interaction. I first outline China’s communications and bureaucratic context, the CCP’s legitimacy context, and the policy lens through which these are applied. An exposition of China’s considerable DSC is then followed by two examples demonstrating differing outcomes of its application: the first a case of overwhelming DSC in the face of the Falun Gong social movement, and the second the case of the 2011 Wukan protests where societal actors gained net political ground.	
  	
   	
  	
   	
   	
   61	
   6.2	
  Mobile	
  Nation	
  Wired	
  Dragon	
    China is the world’s most wired country measured in aggregate mobile and Internet users4 and China’s telecommunications bureaucracy is correspondingly large and multi- faceted. From its first external online connections in 1994, China’s current ICT numbers are staggering: 513 million Internet users (a 16 fold per decade increase),5 over 9 million domains and 15 million websites (under the national .cn level);6 over 1 billion mobile phones representing penetration of 73.5% (300 million Internet-enabled);7 a vibrant blogosphere and rapidly expanding microblogging reaching 250 million users and over 150 million daily tweets. 8  Despite an acute urban-rural digital divide, 9  China’s communication landscape is rapidly digitizing. Beneath lies an ecosystem of government agencies, state-run telecoms, and private providers under extensive regulations.  Centralized media control during the Mao era gave way to tentative liberalization during the early reform period culminating in the present complicated matrix of control. 10  For Internet alone, there are 12 responsible government agencies. 11 Exemplifying Lieberthal’s (2004) fragmented authoritarianism model, divisions in responsibility also extend to the vertical, with often-disparate implementation at local 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   4	
  ITU 2011a, 2011b. 5	
  CINIC 2012, OpenNet Initiative 2011. 6	
  Numbers are from Zhao (2009) for 2007. 7	
  OpenNet Initiative 2011; ITU (2011) quotes 859M (reported). 8	
  ManKinnon 2009; Wines 2012. 9	
  Qiu 2009; OpenNet Initiative 2011. 10	
  Ambivalence in government signaling to the telecommunications/media sectors characterizes dual objectives of control and commercialization. In 2001 Premier Zhu Rongji exhorted the media to both act as watchdogs for the state and actively compete for audience share (Kalathil and Boas 2003). 11	
  These range from the Public Security Bureau (PSB) policing online activity, to the Central Propaganda Department (CPD) monitoring (and guiding) public opinion, to the Ministry of Information Industry (MII) focusing on commercialization (Brady 2007, Harwitt 2008); In the early days of the 1990s bureaucratic struggles were limited to a handful of ministries (Kalathil and Boas 2003). 	
   	
   62	
   levels.12  Liberalization of ownership restrictions with China’s accession to the WTO13 and a multitude of private service providers belie the fact that core infrastructure remains under state-control14 and (official) media under the CCP’s thumb.15 A striking example is the 2004 forced reshuffling of executives among China’s largest telecom firms. China’s telecom landscape and ICT levels are functions of state ideology and policy. 6.3	
  Legitimacy	
  and	
  Policy	
  in	
  China	
   The CCP enjoys a high level of legitimacy within China. At first glance, this may be surprising when compared to other Asian non-democratic states, such as Burma. 16 But this is not without justification. The CCP’s legitimacy originally stemmed from its revolutionary credentials and “hegemonic domination of ideology”17 in the Maoist post- 1949 period. In the reform era, however, and especially with Jiang’s Three Represents policy, CCP legitimacy was reoriented from these historical roots to a performance- based approach. Legitimacy became predicated on “accomplishing concrete goals such as economic growth, social stability, and national unity.”18 China’s globalization, rapid economic development, and concomitant ICT adoption, are parcel to this narrative of 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   12	
  Vertical division operates in both directions. In addition to disparate local enforcement (Qiu 2009), new communication policies are often piloted locally and then promoted nationally. This was so in testing early internet restrictions in Anhui and more recently use of online characters Jingjing (警警) and Chacha (察察) (from Chinese for police:警察 jingcha) to remind that activities are monitored (Harwitt 2008). 13	
  China allows up to 49% foreign ownership in the media sector (Zhao 2009). 14 There are 8 SOE ISPs licensed under MII, with China Telecom having a 70% market share (Qiu 2009). The party owns 51% of all registered media providers (Stockman and Gallagher 2011). 15	
  Known as the ‘night of the Telco long knives’ the CCP’s Central Organization Department (COD), overseeing personnel, shuffled executives of China Telecom, China Mobile, and China Unicom without prior notice and to the dismay of investors and executives alike (McGregor 2010, Harwitt 2008). 16 In Gilley’s (2008) comparison of 72 states, Chinese national legitimacy placed a high 13th.	
   17 Zhu 2011, p.126.	
   18	
  Ibid., p124; Gilley (2008) ascribes institutional change in China primarily to the pursuit of legitimacy by the CCP; Legitimacy reorients institutions to higher-performance results.	
   	
   	
   63	
   performance-driven CCP policy.19 To many Chinese, there is no functional separation between the CCP and the Chinese government as long as performance levels persist.20 But crucially, legitimacy in China decreases with government level. High national-level legitimacy contrasts with lower legitimacy at local levels. 21  Since low legitimacy increases potential ICT effects, it is at the local level where these are most apparent.  Policy in China extends beyond law to “authoritative pronouncements… party documents…state council regulations, leadership speeches, [and] editorials in prominent newspapers.”22 Scholarship is replete with models of Chinese policy.23  A simplified lens used here focuses on Chinese policy as a function of the strengths and weaknesses of the Leninist political model, prioritization of regime legitimacy, and challenges unique to the transition from low to middle-income status. 24 Shambaugh (2008) outlines how Chinese Leninism offers the advantage of high state capacity – with DSC a prime example – while suffering from a lack of mechanisms for “aggregating 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   19 Jiang 2010; Digital communications is also a core component of China’s modernization and development. For example, China sees the Internet as “accelerating the development of the national economy, pushing forward scientific and technological advancement, and expediting the informational transformation of social services” (State Council Information Office 2010, foreword, para 2).	
   20	
  Gilley 2008. 21	
  In one study, support for government decreased from 50%, 25%, 5%, 2%, to 1% when moving from central, provincial, city, county, township (Gilley 2008).	
   22	
  O'Brien and Li 2006, p.5. 23 Several models include: Authoritarian Informationalism (Jiang 2010) that combines aspects of authoritarianism, capitalism, and Confucianism emphasizing state-driven development with individual responsibilities; Authoritarian Populism (Stockmann and Gallagher 2011) that foregrounds the Chinese state’s attention to the lower economic tiers of society as mitigation for social unrest; MacKinnon’s (2010) Networked Authoritarianism; Cyber-Reactionism (Mueller 2011) that classifies states based on their perception of the Internet as either hierarchical or networked, and transnational or national. Cyber- reactionary states (like China) are hierarchical and national in orientation; and Dot-Communism (Kalathil 2001) as the promotion of global commerce while limiting import of foreign ideas. 24	
  This is perhaps more in line with Market-Leninism that describes a market economy driven by a centralized and developmentally-driven party state (Kristof 1993). For McGregor (2010, 2011) communication is one pillar in the CCP’s ‘3Ps’: Party, PLA, and Propaganda. 	
   	
   64	
   disparate demand.”25 To compensate, the CCP leverages ICT as a channel for reading public opinion, and this plays a central role in the CCP’s quest for legitimacy and regime resilience. CCP promotion and control of ICT serves these goals. In public opinion polls, Chinese support government communications control, 26  viewing it as a necessary counter to new media’s inherent risks.27 Given these policy priorities and the CCP’s legitimacy focus, what does the Chinese state’s DSC look like? 6.4	
  The	
  Tiger’s	
  Teeth:	
  Digital	
  State	
  Capacity	
  in	
  China	
    To strengthen the CCP’s legitimacy, manage ideology and nationalism, monopolize sensitive political topics, and quell organized opposition, the Chinese state has developed a powerful DSC arsenal. This rests on designed infrastructure, censorship mechanisms, legal instruments, and coercion. But in implementation, DSC is not the sole purview of the Chinese state. Rather it is joined by an array of private-sector intermediaries, an army of Internet commentators, and self-censorship driven by social norms. Intermediaries must enforce government-defined regulations, resulting in ‘private-public information control partnerships’.28 Service providers, for example, must sign an industry self-regulation pledge of compliance with censorship standards.29 Cross-border service provision by foreign entities is stickier and has resulted in 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   25 Shambaugh (2008, p.8) refers here to a lack of free press, public sphere, and elections – all mechanisms for measuring citizen wants and needs. 26	
  Nathan (2003) ascribes the CCP’s resilience to a stabilized leadership succession, increased meritocracy and institutional specialization, and greater room for public appeal and political participation. 27 A Pew poll on perceptions of the Internet in China found that Chinese fear the Internet’s effects on morality, children, and society and trust online government information sources. This is partly explained by government and media portrayal of the Internet as inherently risky and corrupting (Liang 2007). 28	
  Universities, for example, must limit access to primarily China-based sources (Brady 2008). 29 One pledge includes “[r]efraining from producing, posting or disseminating pernicious information that may jeopardize state security and disrupt social stability, contravene laws and regulations and spread superstition and obscenity” (Internet Society of China 2002, art. 9.1; Goldsmith and Wu 2006). 	
   	
   65	
   negotiations with the Chinese state.30 These providers must offer geolocational filtering or risk the ‘nuclear option’ - wholesale blocking and market exclusion.31 DSC application varies across intermediaries,32 but control need not be perfect to be effective.33  Beyond corporations, China also relies on both Internet commentators and individual self-censorship to control digital information. Internet commentators, pejoratively known as the ‘50 cent party’, are often university students trained to post online comments in line with CCP interests.34 The result is a policy tool for modulating public outrage with a nationalist bend.35 The effect of self-censorship is difficult to estimate, but netizens are well aware of both monitoring levels and their boundaries.36  China’s DSC is balanced on an infrastructure designed for control. The ‘Great Firewall’ combines limited external access, centralized and region-level management, extensive monitoring, and real-time surveillance. All consumer and commercial access is at the tier-2 level through tier-1 government-controlled external access points.37 Kill- switch ability - communication shut down – exists at the regional level and has been 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   30 Faris et al. (2008) term this ‘private-public negotiated transnational filtering’. 31 Ibid. China’s wholesale filtering of foreign service providers has been the subject of calls for WTO legal challenges on grounds of violating GATS market access (Hindley and Lee-Makiyama 2009). Both the US Congress (US Congress 2010a,  2010b) and Google following its 2010 altercation with China (Google Corporation Inc. 2010) have been active here. 32 Pressure depends, for example, on company location and the local government (MacKinnon 2009. 33	
  For Goldsmith and Wu (2006) the key is managing debate and marginalizing opposing voices in China. 34	
  These are known as wǎngluò pínglùn yuán (網絡評論員). The pejorative wǔ máo dǎng (五毛黨) comes from the purported 50 cents paid per comment (Yang 2009). Diamond (2010) places the number at 250,000, but most estimates are considerably lower. 35 Notable events where the CCP ‘dialled up’ the level of online outrage include the American bombing of China’s Bosnian embassy in 1999, the American spy plane incident of 2001 (Goldsmith and Wu 2006), and more recently anti-Japanese outrage following the Japanese arrest of a Chinese sailing vessel captain in contested waters (Bao 2010). But the CCP also fears extreme nationalism (Zhao 2008). 36	
  Yang 2009; All major sites have designated moderators, or bǎnzhǔ (版主), to ensure discussions remain within the permissible (Brady 2008); For exmple, freedom of association or strike issues are often deemed off-limits (Stockmann and Gallagher 2011) 37	
  This was originally designed by Cisco for the Chinese government in 1996. External access is solely through SOEs and all other ISP/ICPs sit within the network (Kalathil and Boas 2003, Goldsmith and Wu 2006). 	
   	
   66	
   used in Tibet and Xinjiang following large protests.38 While initial DSC focus was the Internet, viral SMS information during the SARS 2003 crisis shifted attention to SMS monitoring, now occurring in real-time.39 In sum, this has effectively created a near- national intranet: controlled internal and restricted external communication. Foreign media focuses on the external, but internal censorship is far more significant.  Indeed it is filtering of information internal to China that dominates China’s DSC. This is implemented primarily through service provider intermediaries acting on instructions from regional and central government agencies. Directions include updated lists of disallowed keywords, topics, URLs, and IP addresses, and applies to sensitive topics, such as ‘Taiwan independence’ and sites such as www.falundafa.org.40 An army of human Internet censors complements technical filtering. While extensive, this DSC is not designed to be hermetic; rather intermediaries are expected to identify and remove salient and sensitive emerging issues. There is thus considerable intermediary interpretation and netizens are adept at playing mouse to the censor’s cat.41  Finally, China’s DSC also relies on an evolving web of regulations, laws, and selective use of coercion. From 1997 onwards, a slew of regulations has emerged governing all major digital services in China, their providers, and user registration.42 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   38 Xinhua 2009. 39	
  Telecom intermediaries are tasked with filtering (and storing) SMS traffic and there are over 2800 SMS surveillance centres in China (Brady 2008, Qiu 2009).	
   40	
  The State Council News Office circulates keywords (Yang 2009). There are an estimated 500,000 blocked websites (Brady 2008). 41	
  MacKinnon 2009. For example, netizens creative use of ‘grass mud horse’ (cǎo ní mǎ 草泥馬:a homonym for an expletive) and ‘river crab’ (hé xiè河蟹:also to harmonize, or shut down a site) terms (Diamond 2010, NYT 2009). 42 Examples include regulations on external access (Abbot 2001), industry pledges (Kalathil and Boas 2003), Internet café regulations, BBS services (Yang 2009), blog service providers, and individual real- name weibo (microblog) registration (CINIC 2011). 	
   	
   67	
   China’s 2010 Internet White Paper offers a framework for much of these.43 But it is coercion and its threat that give DSC teeth. In 2010 China was holding 34 journalists in prison and there were 68 imprisoned cyberdissidents in 2012.44 Indeed, China has built an arsenal of DSC. How has DSC combined with digital-empowered societal actors? 6.5	
  Falun	
  Gong:	
  State	
  Capacity	
  Eclipses	
  a	
  Social	
  Movement	
    The interaction of empowered societal actors and DSC does not necessarily have predetermined outcomes. Prior to discussing how ICT has led to political change in China, an example of DSC eclipsing a social movement is presented. Falun Gong is as an exemplar of a Chinese social movement45 that leveraged ICT heavily for mobilization and was then crushed by the state. Caught off guard, the CCP never again underestimated the political power of ICT. The Falun Gong case proved crucial in delineating boundaries of political contestation in China and paradoxically instructing future societal actors on tactics for staking political claims. Defining the disallowed clarified the allowed. Falun Gong’s 1990s growth and escalating struggle with the CCP culminated in 1999 with a 10,000-member protest outside Zhongnanhai, the epicentre of Chinese politics. Falun Gong’s mobilization had relied heavily on ICT, especially the Internet. This combined distributed service hosting across jurisdictions and online organization. A “virtual and real…transnational and local”46 structure had rhizomatic 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   43	
  State Council Information Office 2010. 44	
  Committee to Protect Journalists 2010, Reporters Without Border 2012. In 2012 China had the highest number of cyberdissidents behind bars. 45	
  It is beyond scope to delve into the nature of China’s social movements in general. But it is instructive to view the Falun Gong case in the context of a shift in Chinese movement objectives from the ideational, as part of the Chinese May 4th enlightenment project, to its aborted termination in 1989 in Tiananmen, to its current material focus. Yang (2009) sees an analogy to a Polanyi counter-movement against a market- oriented corporatist state inspired by Western New Social Movements’ focus on the self (Salter 2003) 46	
  Zhao 2003, p.219. 	
   	
   68	
   resilience reminiscent of the CCP’s revoluationary-era organization. This, membership size, and penetration into PLA and party ranks were perceived as ideological challenges netting the CCP’s counter-attack. China continues to point its DSC at Falun Gong.  Filtering and censoring of Falun Gong information happens at all DSC levels, from external connections up the provider chain. Following the 1999 protest, the government temporarily shut down selected ISPs, and surveillance of group activities since has prioritized digital monitoring.47 The state has engaged in DDOS attacks on foreign-hosted group services.48 In the censorship dimension, black lists have included Falun Gong sites, publications, and keywords, and monitoring responds quickly to violations.49 Issue framing, in new and old media, is central to the Falun Gong debate.  The group presents itself “as a virtuous form of self-cultivation and spiritual enlightenment,” 50  an ‘imagined community’ 51  responding to China’s material transformation. The CCP originally portrayed the group as an antithetical ‘evil cult’, echoing Maoist anti-communist and earlier imperial anti-sectarian campaigns.52 This shifted to a ‘superstition’ label contra the CCP’s ideological prioritization of material development and modernization.53  Legal and coercive DSC followed: by prohibiting the promotion of ‘evil cults’ in communications regulations,54 and though mass arrests, 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   47	
  Bell and Boas 2003. 48 Chase and Mulvenon 2002. 49	
  Bell and Boas 2003. 50	
  Zhao 2003, p.210. 51	
  To borrow Anderson’s (2006) term on groups that share values and identity but never actually meet. 52	
  Perry 2002. This harkens back to a long history of Chinese state framing of campaigns against heterodox sects, often termed xiéjiào (邪教), or evil cults, such as the Taiping Rebellion, with a dash of millenarianism in the mix. See Liu and Shek (2004) and (Spence 1996). 53	
  Media interviews shifted to scientists and psychologists who coined a new disease - qìgōng piānchā (氣 功偏差), or qigong deviation, for ‘diagnosed’ members (Perry 2002). 54	
  See for example Kalathil (2001). 	
   	
   69	
   high-profile trials, and group member re-education.55 Falun Gong offers an example of a digitally-empowered movement eclipsed completely by the Chinese state’s DSC. 6.6	
  ICT,	
  DSC,	
  and	
  Societal	
  Actors	
  in	
  Action:	
  Wukan	
  2011	
  and	
  Other	
  Cases	
  	
   The Falun Gong case is important in helping demonstrate the bounds of allowable contestation in China. The CCP does not tolerate independent mass movements nor those challenging its legitimacy or ideological supremacy. But this leaves a spacious terrain of issues where societal actors make tactical, incremental claims winning greater government transparency, representation, and accountability in the process. ICT is central to this process. Successful contestation “operates near the boundary of authorized channels, employs the rhetoric and commitments of the powerful to curb the exercise of power, hinges on locating and exploiting divisions within the state, and relies on mobilizing support from the wider public.”56	
  This section demonstrates this dynamic with an emphasis on the interplay between ICT and DSC. It focuses on the events in the Guangdong village of Wukan in 2011, and supplements with other examples.  In 2011 residents of Wukan village protested illegal selling of farmland by local officials to private interests without due consultation or compensation. There are a large number of land conflicts in China annually.57 But the Wukan incident demonstrates the dynamics of Chinese contention where digital information, DSC, and low levels of government legitimacy - in this case for local leaders - combine. Villagers in Wukan eventually achieved their aims winning limited representation and accountability. Issues 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   55	
  Bell and Boas 2003.	
   56 This ‘rightful resistance’ concept for rural Chinese protests (O’Brien and Li 2006, p.2) applies here. 57 43% of famers have been subject to ‘land grabs’ in the past decade; compensation averages 1/40th of land resale value; each year land is appropriated from 4 million Chinese (Economy 2012).	
   	
   	
   70	
   were kept local and within CCP rhetoric, and there was no challenge to top leadership or its legitimacy, unlike Falun Gong in 1999 or Tiananmen in 1989.5859  Illegal land sales had been occurring in Wukan for many years, and villagers had petitioned various government levels with no redress.60 In September 2011, a protest by 200 villagers led to confrontation with police. The following day 2000 villagers blockaded the police station demanding the release of several villagers. The village Party Secretary Xue Chang and Village Head Chen Shunyi, both implicated in the illegal land sales, fled the village. After receiving no compensation, in November 5000 villagers marched on Lufeng City, the local party headquarters, where the party secretary promised villagers compensation and full redress.61 A December press release, largely hostile to the protesters, indicated that demands had been met and several villager committees would be made illegal. Leaders of earlier protests were then arrested and one, Xue Jinbo, would later died in police custody. His death, the lack of compensation, and hostile media coverage combined to produce a firestorm of protest. Police cordoned off Wukan preventing supplies, including food and fuel, from entering the village.62 Now, with constant international media attention and intermittent domestic online information, Guangdong Party Secretary Wang Yang became involved ensuring villagers 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   58 See also the still-born China Democracy Party (CDP) organized online in 1998 and terminated through DSC (Kalathil and Boas 2003, Goldsmith and Wu 2006), and the World Uyghur Congress that resulted in in shutting off internet to Xinjiang in 2009 (Xinhua 2009).	
   59	
  These act within and not against the system and lessons of past challenges to the system are in the DNA of today’s Chinese protests. These often act as political pressure release valves (Fisher 2012).	
   60	
  Petitioning the central government has a long history in China. The CCP has set up modern channels and received 10 million petitions in 2005. But redress hovers around 1% (Starr 2010).	
   61	
  Fewsmith 2011. 62 Moore 2011, Caixing 2011. 	
   	
   71	
   their demands will be met and a transparent local election process would be set up. This would later result in the election of one of the protest leaders as Village Head. 63  What were the political opportunities driving the Wukan protests? While the legitimacy of local leaders was attacked, villagers explicitly differentiated this from an attack on the CCP in general, its central or provincial leadership.64 China’s fragmented authoritarian structure gives local leaders considerable power, but also creates potential rifts with central leadership. Rifts become chasms as top leaders adopt populist governance goals that prioritize the disenfranchised resonating with protesters. 65 Dynamics amplify with political uncertainty, such as during leadership transitions. ICT here acts to signal societal actors and top leadership issue positioning.   An example occurred when Wang Yang prioritized and met villagers’ demands. This was signalled through a change in media tone and online coverage.66 Successful movement goals are those that resonate with top leaders, such as environmental protection and anti-corruption. In Wukan villagers immediately and repeatedly emphasized allegiance to CCP leaders and the local nature of the issue, despite land grabs being a pervasive Chinese problem. 67  This strategy was crucial in Wukan given fast information diffusion online and non-hermetic cross-border information flow that resulted in international media coverage and thus increased political pressure. Similar 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   63	
  Beitarie 2011, Meilian 2012. 64	
  One banner in English read: "We are not a revolt. We support the Communist Party. We love our country" (Fisher 2012, para.1). 65	
  Lieberthal 2004; This is central to Stockmann and Gallagher’s (2011) Authoritarian Populism model. 66	
  Wang Yang also leveraged Wukan for coverage to enhance his position in line for a politburo standing committee spot (Fewsmith 2012) in the 2012 leadership transition.	
   67	
  Economy 2011, Fisher 2012. 	
   	
   72	
   dynamics have played out in other cases, including working condition protests68 and anti-pollution campaigns, such as in Xiamen in 2007.69 For example, in the 2007 Shanxi child labour kilns case, online postings by parents in neighbouring Henan forums were cross-posted nationally, went viral, and resulted in President Hu’s direct involvement.70 Division in governmental levels on a legitimacy basis thus leaves space for mid and top- level leadership to signal their support for protesters.71 In Wukan this played out through what was censored, to what extent, and the tone of media coverage. The initial pejorative tone towards the protesters changed radically with the involvement of Wang Yang, and DSC pressure was dialled down with increased national leadership support.  Wukan also serves as an example of the impact of DSC on societal actor mobilization.72 Wukan’s proximity to Hong Kong enmeshed reports about the protests in a wider media ecosystem including traditional Chinese media, ICT channels, and the international press. Traditional Chinese media coverage, modulated through DSC, signalled leader intentions throughout the events. Censorship varied through scope and changes in keyword filtering as a function of leadership positions. Online filtering for ‘Wukan’ later expanded to code words for Wukan and searches for the cities of Shanwei 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   68	
  Economist 2012. 69	
  Sills 2010, Economist 2007a. 70 This resulted in the arrest of local officials and police within days (Yang 2009, Dongfang 2007). 71	
  Perry 2002; Opportunties also exist for politial entrepreneurs to position themselves accordingly.	
   72 Although other notable protests relied directly on ICT for mobilization before the state could respond with DSC. This was the case in a 2006 Beijing taxi driver strike, organized over email and websites (Harwitt 2008), as it was in the 2007 Xiamen anti-paraxylene plant protests initiated over text messages and further circulated on blogs (Economist 2007a).	
   	
   	
   73	
   and Lufeng.73 Nevertheless, coverage from Hong Kong re-entered the Chinese web and was cross-posted in a continual cat-and-mouse game between netizens and censors.74 Figure	
  6.2	
  –	
  Digital	
  Contention	
  Space	
  in	
  China	
    Indeed media restrictions often serve to direct controversial news to unofficial sources (and unofficial journalists75) where events can resonate generating public pressure. This was the case during the 2003 SARS crisis76 and the recent Wenzhou high-speed rail crash, where first netizens and then journalists suspected a cover up eventually forcing a government backtrack.77  Since DSC and ICT are rarely immediately decisive, this interplay necessitates proactive public positioning. Aware of the power and speed of ICT-fuelled media, villagers in Wukan publicly positioned themselves within CCP rhetoric, distanced themselves from offers of connections to dissidents, and 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   73 See BBC (2011) and CDT (2011b) for filtering details.	
   74	
  Bandurski 2011a.	
   75	
  Yang (2009, p.94) documents an example of the tools used by China’s expanding legion of citizen journalists, gōngmín jìzhě (公民記者), that include: mobile phone, digital camera, laptop, and platforms for email, blogging, RSS, VOIP, chat, and video/photo sharing. 76	
  See the appendix figure A.2 for a schematic of both cases. For SARS in 2003, Beijing and Guangdong failed to counter leaked information that went viral. The WHO later regulated to allow including unofficial media reports in future epidemic investigations (J. Zhao 2003). 77 China Daily 2011, Bandurski 2011b, CDT 2011a. 	
   	
   74	
   continuously emphasized fealty to China and the CCP. 78 China’s leaders, for their part, supported Wang Yang’s proactive role through public policy statements. 79  Indeed Chinese protesters are well aware of the potential for digital information to go viral and the power of fast escalation.80 ICT and DSC here directly alter the actions, perceptions, and tactics of societal and state actors in Chinese political contention.  How issues are framed is also crucial in China’s political contestation space (see Figure 6.2), and this proved central to the Wukan protests. ‘Sudden incidents’, or breaking news, are framed early and rapidly in digital contestation often catching less- nimble government agencies off guard.81 Villagers in Wukan emphasized from the beginning that their targets were local corrupt leaders and their goals well in line with CCP rhetoric. Rather than framing orthogonal to disallowed issues, this approach defines “claims with reference to protections implied in ideologies or conferred by policy makers” (O'Brien and Li 2006, p.2). Villagers in Wukan leveraged the official Chinese narrative on corruption, framed as ‘bad apples happy ending’82 foregrounding citizen rights undergirded by the rule of law. Rapid framing dynamics also played out in the 2011 Wenzhou train crash mentioned above and the 2001 Jiangxi school firework explosion.	
  83 In the former, initial governmental framing of ‘tragedy’ and ‘personnel 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   78	
  Beitarie 2011. 79	
  Fewsmith 2011.	
   80	
  Protesters are aware of how potentially fast escalation pressures the state (Fisher 2012).	
   81	
  Negative news often appears first on unofficial websites, setting the tone and timbre of coverage and forcing a now straightjacketed government response (J. Zhao 2009). 82	
  Stockmann and Gallagher 2011. 83	
  Nathan 2003; This case involved a schoolhouse where students made fireworks (instead of studying) that resulted in an explosion killing 42. The government narrative focused on the ‘tragic’ aspect blaming a local mentally-ill villager who set off the blast (People's Daily 2001a, 2001b). The alternate ‘unofficial’ media narrative asked why students’ were making the fireworks at all (Y. Zhao 2008, Saich 2001). Interestingly, it is often humour or satire online that first tests the limits of allowable debate by poking fun at icons, heroes, and myths (J. Zhao 2009). 	
   	
   75	
   heroism’ were trumped by the media’s insistence on a narrative of ‘negligence’ and ‘corruption’. This later frame would triumph online later resulting in removal of accused officials.84 As with the Wenzhou and Jiangxi cases, famous incidents such as Wukan serve to delineate allowed boundaries of contestation and expand the envelope of permissible debate over time. Wukan offers an example of how ICT has altered the political opportunities, mobilization potential, and issue framing that societal actors face. Full DSC can counteract these effects, but this is only wielded in China if the CCP perceives a direct threat. For other cases, a more nuanced ICT / DSC interplay results. 6.7	
  Epilogue:	
  Claiming	
  Political	
  Rights	
  One	
  Click	
  at	
  A	
  Time	
    ICT and its role in China’s contestation terrain is not about to lead to regime change nor a democratic transition. And this is not the intent of the vast majority of Chinese protests, nor the Chinese population at large. But this is not where ICT’s power lies. Rather, it has offered societal actors channels to contend with and sometimes best the Chinese state. Chinese governance is more transparent, representative, and accountable for it, political debate and participation has widened, and these are tenuous yet undeniable shoots of non-electoral democratic mechanisms. This diffusion of digital voices is part of a macro process of state withdrawal from a Maoist-era totalitarian control over citizens’ lives to a central governance core. This core remains off limits to challenge: both direct (as in the CDP) and oblique, if organized by mass movements independent of the state (as in Falun Gong). Outside this core is the contestable terrain of citizen, worker, consumer, and environmentalist fighting issues that resonate with 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   84 China Daily 2011, Bandurski 2011b. 	
   	
   76	
   CCP rhetoric and policy. In response, the state is reactively and proactively conceding to greater political participation. The result enhances its regime resilience and legitimacy.  Peeling back the layers of this process, progressive devolution of power from the centre throughout the reform era85 has increased government accountability. New tactics available to societal actors can check local corruption and abuse of power.86 Issue resonance is a function of framing, and non-hermetic DSC information control means that a local issue can become national. Several examples above – including the 2007 Jiangxi child labour case and 2011 Wenzhou train crash – demonstrate this. In both, the media, often led (and chased) by citizen journalists and online sites, have adopted the role of people’s tribune.87 A poignant case is that of Sun Zhigang, a university student beaten to death by migrant detention centre police in 2003 for failing to carry the proper residence permit. Online letters and petitions on local sites picked up nationally forced the arrest of local officials and eventual closure of all similar detention centres.88  Transparency in Chinese governance has increased with a multitude of digital eyes and voices observing government actions.89 The dialectic between citizen demands and government liberalization is exemplified in ‘sunshine’ transparency policies.90  The central government is pushing e-government, ministry websites, online chat, and email 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   85	
  Saich 2001. 86 In an online public chat, Premier Wen Jiabao stated, “He who knows the leakage of a house lives under the roof. He who knows the mishandling of a state is among the populace” (Jiang 2010, p.80). 87	
  This process is also driven by increased media professionalism and competition (Nathan 2003, Stockmann and Gallagher 2011). 88	
  Diamond 2010; Y. Zhao (2008) exaplains how contending online identities propelled this issue. 89	
  Brady 2008. 90	
  Xinhua 2002. For a critique, see Wang (2009); Kalathil and Boas (2003) offer an example in the form of a sunshine purchasing policy in online coal procurement. 	
   	
   77	
   complaint systems, 91 as well selective online publication of draft regulations for public input.92 This widening of government information and citizen input, albeit contingent and controlled, the increased professionalism of media, and the empowerment of individual voices is arguably a nascent, state-sanctioned and bounded, public sphere.93 This increases representation of citizen views, not through direct electoral mechanisms, but through the indirect dynamics of digital contestation.94 The gains for the CCP leadership are enhanced resilience, legitimacy, and a ready, if noisy, source of public opinion. For Chinese citizens and societal actors focused on issue-specific political change, the wide-scale adoption of ICT has fundamentally altered how claims are made against the state. Their view of government actions is clearer, their voices are now louder, and their views en mass are now more powerful.	
   	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   91	
  Jiang 2010. 92	
  Chongqing municipality is a leader here (Zhou 2007). 93 Jiang 2010. 94 Nathan (2003) sees all of these as part of a long-term process of greater political participation driven by the regime’s ongoing quest for legitimacy, that started at the work-unit level, expanded to village ellections, and now manifests in new media dynamics. 	
   	
   78	
   Chapter	
  7	
   Conclusion:	
  Research	
  Significance	
  and	
  Implications	
  	
   7.1	
  Significance	
  and	
  Overview	
  	
   This study has investigated how ICT is playing a role in political change, and specifically state-society relations, in Asia. With economic, political, and social transformation throughout the region and fast-evolving communications technology, much is unclear about the dynamics of digital contention. There are unknown unknowns. But given ICT’s continued adoption in globalizing Asian states, a known unknown is the full impact this will have on state and society relations. What is known is that ICT has modified the policy choices of states and strategies of societal actors. As key instruments of political communications, these tools are increasingly central to democratization processes in Asian states, whether these be incremental, entrenchment, transition, or counter-democratization mechanisms. Political contention now has a digital dimension.  Two schools of scholarship – those seeing digital empowerment of societal actors and those seeing a strengthening of non-democratic states – have contributed to a seemingly binary debate and simplistic policy interpretation. The theory developed here, however, provides for agency both at the societal actor and state levels, and does so through a social movement framework and the new lens of digital state capacity (DSC). The resulting plausible hypothesis asserts that societal actor usage of ICT can have democratizing effects in non-democratic states with low legitimacy and subject to the application of DSC. Democratization here is not primarily electoral, but rather the underlying mechanisms of transparency, accountability, and representation. Nor are these outcomes forgone conclusions, but rather depend on the interface of these two forces over specific areas of political contention. A quantitative analysis on Asian 	
   	
   79	
   democracy levels, ICT penetration, and DSC put this and contending theories to the test. It found that while a regional correlation exists between ICT and democracy, this is decreasing over time. A deeper investigation demonstrated a bifurcation in this relationship based on the role of the state: the correlation between democracy and ICT being reversed for states that employ online censorship. To understand this, the state’s ability to control digital communications – DSC – was then constructed. This new concept and the dual-agency theory were applied to China. This process has both strengthened the Chinese government’s capacity and opened up previously off-limits political terrain, netting opportunities for increased transparency, accountability, and representation in China. 7.2	
  Research	
  Applications:	
  What	
  This	
  All	
  Means	
  	
   Outside of scholarship, this study carries policy implications for international relations, trade, media, and human rights. The foreign policy of some Western states, including Canada and the United States, has at times prioritized democracy promotion in Asia.1 ICT, and notably the Internet, have been seen as instruments towards economic and political liberalization and, subsequently, democratization. Initiatives such as the US 2010 Global One Internet Act are premised on this assumption.2 This study, however, demonstrates that the truth is far more nuanced and that non-democratic regimes are also empowered through ICT. Oversimplified theory risks oversimplified policy, and a key lesson is that ICT penetration is not a linear stepping-stone to democratization. Policy lessons also abound for non-democratic Asian regimes, foremost being the 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   1	
  DFAIT 2007, United States Congress 2010a, G8 2011. 2	
  United States Congress 2010b, Shirky 2011. 	
   	
   80	
   inability to combine high ICT penetration with a political communications monopoly. DSC allows control of information, but at ever-increasing cost, and it is not absolute. Non-democratic regimes in wired Asian states consequently need to pick their information control battles, delimiting which political areas are off and which within limits. Borders can be segregated to information, but in globalizing states this is never hermetic.  This leads to implications both in the realm of trade policy and media. An expanding proportion of cross-border services trade now occurs online.3 This is the bread and butter not only of e-commerce, but also the financial services, digital media, and telecommunication sectors, among others. States filtering these services for political reasons are limiting trade and, in the case of WTO members, possibly violating GATS rules. This is especially so for wholesale service blocking and for states relying on regulatory ambiguity. Challenging these aspects of DSC is possible within the global trade regime rules.4 For media, the non-hermetic nature of cross-border information means they no longer operate in an isolated environment. News crosses between the domestic and foreign as between journalist and citizen blogger. The playing field is wider, competition fiercer, and news cycles faster. Media must respond accordingly.  Finally, ICT and its role in democratization is altering human rights priorities. Several organizations, including the Committee to Protect Journalists, Reporters Without Borders, and Amnesty International are placing the right to digital information 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   3	
  WTO 1994, GATS, art. VI. 4	
  GATS calls for the existence of administrative or judicial appeal mechanisms for market access limitations on cross border trade in services. I.e. for cases where a state chooses to block a complete platform, as with Thailand’s Youtube filtering, this can be challenged on grounds of appeal. This consequently also counters regulatory ambiguity (Wu 2006, Hindley and Lee-Makiyama 2009). 	
   	
   81	
   on their agendas.5 Freedom of expression is increasingly being manifested online and tomorrow’s prerequisite for citizenship may well entail a digital component. Over time, this may shift freedom of expression from a passive right, shielded from the state to an active right, incumbent on the state to provide.6 7.2	
  Future	
  Research	
  	
   This study is necessarily limited in scope and depth and has been unable to address key aspects of Asian digital communications and democratization. Suggestions for further research are offered. First, the concept of DSC introduced attempts to encapsulate and model the state’s control of information. In order to be of use in quantitative analysis and case comparison, this would need to be operationalized, measured across states, and normalized to states under investigation. Second, while current quantitative analysis is limited to detecting correlations, more powerful testing can include causality. Finally, a fuller understanding of the interplay of Asian ICT and DSC requires case comparisons for states with variance in these variables. This would offer invaluable testing for this theory, with cases selected based on proximity to normalized ideal types.  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   5	
  Committee to Protect Journalists 2010, Reporters Without Borders 2011, Amnesty International 2012. 6	
  See Abbot (2006) on the implications of active versus passive human rights; This has already occurred in several European states that have guaranteed digital access as a right of citizenship (Finish Ministry of Transport and Communications 2010, Morris 2009, French Republic Constitutional Council 2009). 	
   	
   82	
   Bibliography	
  	
   Abbot, Frederick. "TRIPS and Human Rights: Preliminary Reflections." In International Trade and Human Rights: Foundation and Conceptual Issues, by Frederick Abbot, Christine Breining-Kaufmann, and Thomas Cottier. Ann Arbor: The University of Michigan Press, 2006. 	
   Abbot, Jason. "Democracy@Internet.Asia? The Challenges to the Emancipatory Potential of the Net: Lessons from China and Malaysia." Third World Quarterly 22, no. 1 (February 2001): 99-114.  Acemoglu, Daron, and James Robinson. Economic Origins of Dictatorship and Democracy. New York, NY: Cambridge University Press, 2006. 	
   Aday, Sean, Henry Farrell, Marc Lynch, John Sides, John Kelly, and Ethan Zuckerman. Blogs and Bullets: New Media in Contentious Politics. Washington, DC: United States Institute of Peace, 2010. 	
   Almond, Gabriel, and Sidney Verba. Civic Culture: Political Attitudes and Democracy in Five Nations. Newbury Park, CA: SAGE Publications, 1989. 	
   Amnesty International. Freedom of Expression. http://www.amnesty.org/en/freedom- of-expression (accessed March 20, 2012).  Anderson, Benedict. Imagined Communities. London: Verso, 2006. 	
   Back, Hanna, and Axel Hadenius. "Democracy and State Capacity: Exploring a J-Shaped Relationship." Governance: An International Journal of Policy, Administration, and Institutions 21, no. 1 (January 2008): 1-24. 	
   Bandurski, David. "China’s Guerrilla War for the Web." Far Eastern Economic Review, July 2008. 	
   —. "History of High-Speed Propaganda Tells All." China Media Project, July 25, 2011. http://cmp.hku.hk/2011/07/25/14036/ (accessed August 1, 2012). 	
   —. "Chinese Language Coverage of Wukan." China Media Project, December 19, 2011.  Bao, Daozu. "Diaoyu Protests Intensify." China Daily, September 11, 2010.  Bardhan, Pranab. "India and China: Governance Issues and Development." The Journal of Asian Studies 68, no. 2 (May 2009). 	
   BBC. "China protest in Guangdong's Wukan 'vanishes from web'." December 15, 2011.  Beck, Nathaniel, and Jonathan Katz. "What to do (and not to do) with time-series cross- section data." American Political Science Review 89 (1995): 634–47. 	
   	
   83	
   	
   Beitare, Rachel. "The Sprit of Wukan." Foreign Policy, December 23, 2011. 	
   Bell, Mark, and Taylor Boas. "Falun Gong and the Internet: Evangelism, Community and the Struggle for Survival." Nova Religio: The Journal of Alternative and Emergent Religions 6, no. 2 (April 2003): 277-293. 	
   Benkler, Yochai. The Wealth of Networks. New Haven: Yale Univeristy Press, 2006. 	
   Best, Michael L., and Keegan W. Wade. "The Internet and Democracy Global Catalyst or Democratic Dud?" Bulletin of Science, Technology & Society 29, no. 4 (August 2009): 255-271. 	
   Bevir, Mark, and R. Rhodes. "Interpretive Theory." In Theory and Methods in Political Science, by David Marsh and Gerry Stoker, Eds. Houndmills: Palgrave Macmillan, 2002. 	
   Bikhchandani, Sushil, David Hirshleifer, and Ivo Welch. "Learning from the Behavior of Others: Conformity, Fads, and Informational Cascades." The Journal of Economic Perspectives 12, no. 3 (Summer 1998): 151-170. 	
   Blum, Andrew. "Tunisia, Egypt, Miami: The Importance of Internet Choke Points." The Atlantic, January 2011. 	
   Boix, Charles. Democracy and Redistribution. New York, NY: Cambridge University Press, 2003. 	
   Boix, Carles, and Susan Stokes. "Endogenous Democratization." World Politics 55, no. 4 (July 2003): 517-549. 	
   Brady, Anne-Marie. Marketing Dictatorship: Propaganda and Thought Work in Contemporary China. Plymouth: Rowan and Littlefield, 2008. 	
   Branigan, Tanya. "China boosts internet surveillance." The Guardian, July 26, 2011. 	
   Branstetter, Lee, and Nicholas Lardy. "China’s Embrace of Globalization." In China’s Great Economic Transformation, by Loren Brandt and Thomas Rawski. Cambridge: Cambridge University Press, 2008. 	
   Bremmer, Ian. "Democracy in Cyberspace." Foreign Affairs, November/December 2010. 	
   Broadhurst, Roderic, and Yao-Chung Chang. "Cybercrime in Asia: Trends and Challenges." In Asian Handbook of Criminology, J. Liu and B. Hebenton, Eds. Springer, 2012. 	
   	
   	
   84	
   Bueno de Mesquita, Bruce. Logic of Political Survival. Cambridge, MA: The MIT Press, 2000. 	
   Bueno de Mesquita, Bruce, and Lawrence Cohen. "Self-Interest, Equity, and Crime Control: A Game-Theoretic Analysis of Criminal Decision Making." Criminology 33, no. 4 (1995): 483-518. 	
   Bueno de Mesquita, Bruce and George Downs. “Development and Democracy.” Foreign Affairs 84, no. 5 (Sept. – Oct. , 2005): 77-86.  Bueno de Mesquita, Bruce, and Alastair Smith. "Leader Survival, Revolutions, and the Nature of Government Finance." American Journal of Political Science 54, no. 4 (July 2010): 936-950. 	
   Cairncross, Frances. The Death of Distance: How the Communications Revolution Will Change Our Lives. London: Orion Publishing Group, 1997. 	
   Caixing. "Wukan Village Under Police Seige." December 15, 2011.  	
   Calingaert, Daniel. "Authoritariansim vs. the Internet." Policy Review, April/May 2010. 	
   Cellan-Jones, Rory. "Hi-tech helps Iranian monitoring ." BBC News, June 22, 2009. 	
   Chao, Chian-Hsueng. "Reconceptualizing the Mechanism of Internet Human Flesh Search: A Review of the Literature." 2011 International Conference on Advances in Social Networks Analysis and Mining. IEEE Computer Society, 2011. 	
   Chase, Michael, and James Mulvenon. You've Got Dissent! Chinese Dissident Use of the Internet and Beijing's Counter-Strategies. Arlington, VA: RAND Corporation, 2002. 	
   Checkel, Jeffrey. "The Constructivist Turn in International Relations Theory." World Politics 50, no. 2 (January 1998): 324-348. 	
   China Daily. "Weibo asking users to register identities." December 3, 2011. 	
   —. "Hasty Burial of Wreckage Sparks Suspicion." July 26, 2011. 	
   China Digital Times. "Directives from the Ministry of Truth: Wenzhou High-Speed Train Crash." July 25, 2011. chinadigitaltimes.net/2011/07/directives-from-the- ministry-of-truth-wenzhou-high-speed-train-crash/ (accessed August 1, 2012). 	
   —. "'Wukan' on Sina Weibo: Unblocked as Protest Postponed [Update]."December 21, 2011. 	
   China Internet Information Center. "Shanghai to require real-name weibo registration." China Internet Information Center, December 26, 2011. 	
   	
   85	
    Chomsky, Noam, and Edward Herman. Manufacturing Consent: The Political Economy of the Mass Media. New York, NY: Pantheon Books, 2002.  Chowdhury, Mridul. "The Role of the Internet in Burma’s Saffron Revolution." Internet & Democracy Case Study Series. Berkman Center for Internet and Society, 2008. 	
   CINIC. China Internet Network Information Center. National Statistics. 2012.  Clarke, George, and Scott Wallsten. "Has the Internet Increased Trade? Developed and Developing Country Evidence." Economic Inquiry 44, no. 3 (July 2006). 	
   Coase, R.H. "The Nature of the Firm." Economica, November 1937: 386-405. 	
   Cogburn, Derrick, and Fatima Espinoza-Vasquez. "From Networked Nominee to Networked Nation: Examining the Impact of Web 2.0 and Social Media on Political Participation and Civic Engagement in the 2008 Obama Campaign." Journal of Political Marketing 10 (2011): 189–213. 	
   Committee to Protect Journalists. "Attacks on the Press in 2008: Pakistan." Committee to Protect Journalists, 2009. 	
   —.  Attacks on the Press 2010. New York: United Book Press, 2010.  Cortimiglia, Marcelo, Filippo Renga, and Andrea Rangone. "A Classification Schema for Mobile-Internet 2.0 Applications." In ICETE 2009, CCIS 130, 126-138. Springer- Verlag Berlin Heidelberg , 2011. 	
   Cottier, Thomas, and Khorana Sangeeta. "Freedom of Expression and Unfair Competition." In Human Rights and International Trade, by Thomas Cottier, Joost Pauwelyn and Elisabeth Burgi. Oxford: Oxford University Press, 2005. 	
   Cummings, Sally, and Ole Nørgaard. "Conceptualizing State Capacity: Comparing Kazakhstan and Kyrgyzstan." Political Studies 52 (2004). 	
   Dahl, Robert. Democracy and Its Critics. Yale University, 1989. 	
   —. On Democracy. New Haven: Yale University Press, 1998. 	
   Das, Gurcharan. "The India Model." Foreign Affairs 85, no. 4 (Jul-Aug 2006). 	
   Delta, George, and Jeffrey Matsuura. Law of the Internet, Volume 1. New York, NY: Aspen Publishers, 2011. 	
   DFAIT, Canada Department of Foreign Affairs and International Trade. "A New Focus on Democracy Support: Government Response to the Eighth Report of the Standing Committee on Foreign Affairs and International Development: 	
   	
   86	
   “Advancing Canada’s Role in International Support for Democratic Development”." DFAIT. November 2, 2007. http://www.international.gc.ca/international/democracy_support.aspx?view=d. (accessed September 24, 2011). 	
   Diamond, Larry. "Liberation Technology." Journal of Democracy 21, no. 3 (July 2010). 	
   Diebert, Ronald. "International Plug 'n Play? Citizen Activism, the Internet, and Global Public Policy." Policy and International Studies 1 (2000): 255-272. 	
   Diebert, Ronald, and Rafal Rohozinski. "Good for Liberty, Bad for Security? Global Civil Society and the Securitization of the Internet." In Access Denied: The Practise and Policy of Global Internet Filtering, by Ronald Diebert, John Palfrey, Rafal Rohozinski and Jonathan Zittrain. Cambridge, MA: MIT Press, 2008.  Diebert, Ronald, John Palfrey, Rafal Rohozinski, and Jonathan Zittrain. Access Contested: Security, Identity, and Resistance in Asian Cyberspace. Cambridge, MA: MIT Press, 2011. 	
   —. Access Controlled: The Shaping of Power, Rights, and Rule in Cyberspace . Cambridge, MA: MIT Press, 2010. 	
   —. Access Denied: The Practise and Policy of Global Internet Filtering. Cambridge, MA: MIT Press, 2008. 	
   Dollar, David. "Lessons Other Nations Can Learn from China." China Daily, November 20, 2008. 	
   Dongfang, Han. The Shanxi Brickyard Scandal and Child Labour in China. China Labour Bulletin, 2007. 	
   Drezner, Daniel. "Weighing the Scales: The Internet's Effects On State-Society Relations." Brown Journal of World Affairs 16, no. 2 (Spring/Summer 2010): 31- 44. 	
   Dutton, William, Anna Dopatka, Michael Hills, Ginette Law, and Victoria Nash. Freedom of Connection - Freedom of Expression:  The Changing Legal and Regulatory Ecology Shaping the Internet, A Report Prepared for UNESCO’s Division for Freedom of Expression, Democracy and Peace. Oxford: Oxford Internet Institute, 2010. 	
   Economist. "Liberation Technology: Mobiles, protests and pundits." Economist, October 26, 2006.  —. "Mobilised by Mobile: Organised by text messages and internet chats, China's middle classes are daring to protest, and giving the government a fright." Economist, June 21, 2007. 	
   	
   87	
    —. "Weird but wired Online dating in Pyongyang? Surely not ." Economist, February 1, 2007.  —. "Social Media in the 16th Century: How Luther Went Viral." Economist, December 17, 2011.  —. "India’s identity scheme: The magic number." Economist, January 12, 2012.  —. "A Dangerous Year: Economic conditions and social media are making protests more common in China at a delicate time for the country’s rulers." Economist, January 28, 2012. 	
   —. "Reporting Chinese Politics: Hidden News." Economist, February 11, 2012. 	
   Economy, Elizabeth. "China's Land Grab Epidemic Is Causing More Wukan-Style Protests." Council on Foreign Relations, February 8, 2012. 	
   Estelles-Arolas, Enrique, and Gonzalez-Ladron-de-Guevara. "Towards and Integrated Crowdsourcing Definition." Journal of Information Science XX, no. X (2012): 1- 14.  Evans, Peter. Embedded Autonomy: States & Industrial Transformation . Princeton, NJ: Princeton University Press, 1995. 	
   Farber, Daniel. "Free Speech Without Romance: Public Choice and the First Amendment." Harvard Law Review 554 (1991). 	
   Faris, Robert, and Bruce Etling. "Madison and the Smart Mob: The Promise and Limitations of the Internet for Democracy." The Fletcher Forum of World Affairs 32, no. 2 (Summer 2008). 	
   Faris, Robert, Stephanie Wang, and John Palfrey. "Censorship 2.0." Innovations, Spring (2008): 165-187. 	
   Fasshi, Farnaz. "Iran Mounts New Web Crackdown." Wall Street Journal, January 6, 2012. 	
   Fenby, Jonathan. The Penguin History of Modern China. London: Penguin Books, 2009. 	
   Fewsmith, Joseph. "Guangdong Leads Call to Break Up 'Vested Interests' and Revive Reform." China Leadership Monitor, no. 37 (2012). 	
   Finish Ministry of Transport and Communications. "1 Mbit Internet Access a Universal Service in Finland from the Beginning of July: Press Release." June 29, 2010. 	
   	
   88	
   http://www.valtioneuvosto.fi/ajankohtaista/tiedotteet/tiedote/en.jsp?oid=30197 9 (accessed April 16, 2011). 	
   Fischer, David H. Paul Revere's Ride. New York, NY: Oxford University Press, 1994. 	
   Fish, Eric. "Is Internet Censorship Compatible with Democracy? Legal Restrictions of Online Speech in South Korea." Asia-Pacific Journal on Human Rights and the Law 2 (2009): 43-96. 	
   Fisher, Max. " How China Stays Stable Despite 500 Protests Every Day." The Atlantic, January 5, 2012. 	
   Foucault, Michel. Discipline and Punish: The Birth of the Prison. New York, NY: Vintage Books, 1995. 	
   Freedom House. Freedom in the World 2011. Freedom House, Washington: Freedom House, 2011. 	
   French Republic Constitutional Council. "Act furthering the diffusion and protection of creation on the Internet. Decision n° 2009-580 of June 10th 2009."  Friedman, Thomas. The Lexus and The Olive Tree. New York: Random House, 2000.  —. "The First Law of Petropolitics." Foreign Policy, May/June 2006: 28.  —. The World is Flat: A Brief History of the Twenty-First Century. New York, NY: Farrar, Straus and Giroux, 2007. 	
   Fukuyama, Francis. State Building: Governance and World Order in the 21st Century. New York, NY: Cornell University Press, 2004. 	
   —. The End of History and the Last Man. New York, NY: Free Press, 2006. 	
   Fuller, Thomas. "A High-Tech War Against Slights to a Centuries-Old Monarchy." New York Times, October 2, 2011. 	
   G8. "Deauville G8 Declaration: Renewed Commitment for Freedom and Democracy." G8 Summit, May 26-27, 2011, G8 France 2011. http://www.g20-g8.com/g8- g20/g8/english/the-2011-summit/declarations-and- reports/declarations/renewed-commitment-for- freedom-and- democracy.1314.html (accessed September 24, 2011). 	
   Garrett, Geoffrey. "The Causes of Globalization." Comparative Political Studies 33, no. 6/7 (August/September 2000): 941-92. 	
   	
   	
   89	
   Garrett, Kelly. "Protest in an Information Society: A Review of Literature on Social Movements and New ICTs." Information, Communications & Society 9, no. 2 (2006): 202-224. 	
   Giglio, Mike. "The Facebook Freedom Fighter." Newsweek, February 13, 2011. 	
   Gilley, Bruce. "The Determinants of State Legitimacy: Results for 72 Countries." Internaional Political Science Review 27, no. 1 (2006): 47-71.  	
   —. "Legitimacy and Institutional Change: The Case of China." Comparative Political Studies 41, no. 3 (2008): 259-284.  Giugni, Marco. "Introduction: Social Movements and Change: Incorporation, Transformation, and Democratization." In From Contention to Democracy, by Marco Giugni, Doug McAdam and Charles Tilly. Lanham, MD: Rowman & Littlefield, 1998. 	
   Gladwell, Malcolm. "Small Change: Why the Revolution Will Not Be Tweeted." The New Yorker, October 4, 2010. 	
   —. The Tipping Point: How Little Things Can Make a Big Difference. New York, NY: Little, Brown and Company, 2000. 	
   Glanz, James, and John Markoff. "Egypt Leaders Found 'Off' Switch for Internet." New York Times, February 16, 2011. 	
   Goldsmith, Jack, and Tim Wu. Who Controls the Internet: Illusions of a Borderless World. New York, NY: Oxford University Press, 2006. 	
   Google Corporation Inc. "Promoting Free Trade for the Internet Economy: Google Public Policy White Paper." November 15, 2010. http://googlepublicpolicy.blogspot.com/2010/11/promoting-free-trade-for- internet.html (accessed April 18, 2011). 	
   Granovetter, Mark. "The Strength of Weak Ties." American Journal of Sociology 78 (1973): 1360-1380. 	
   Grönlund, Åke. "Using ICT to Combat Corruption." In Increasing Transparency and Fighting Corruption through ICT Empowering People and Communities, by Cecilia Strand. Stockholm: The Swedish Program for ICT in Developing Regions, 2010. 	
   Groshek, Jacob. "The Democratic Effects of the Internet, 1994–2003: A Cross-National Inquiry of 152 Countries." The International Communication Gazette 71, no. 3 (2009): 115-136. 	
   	
   	
   90	
   —. "A Time–Series, Multinational Analysis of Democratic Forecasts and Internet Diffusion." International Journal of Communication, 4 2010: 142-174. 	
   Habermas, Jürgen. The Structural Transformation of the Public Sphere. Cambridge, MA: The MIT Press, 1989.  —. "The Public Sphere: An Encyclopaedia Article." In Media and Cultural Studies, by Meenakshi Durham and Douglas Kellner. Malden, MA: Blackwell Publishing, 2006. 	
   Hachigan, Nina. "The Internet and Power in One-Party East Asian States." The Washington Quarterly 25, no. 3 (2002): 41-58. 	
   Hardin, Russell. "Compliance, Consent, and Legitimacy." In The Oxford Handbook of Comparative Politics, by Susan Stokes and Charles Boix, Eds. New York, NY: Oxford University Press, 2009. 	
   Harwitt, Eric. China's Telecommunications Revolution. New York, NY: Oxford University Press, 2008. 	
   Hermanns, Heike. "Mobile Democracy: Mobile Phones as Democratic Tools." Politics 28, no. 2 (2008): 74-82. 	
   Hindley, Brian, and Hosuk Lee-Makiyama. Protectionism Online: Internet Censorship and Internet Trade Law. Working Paper No.12/2009, European Centre for International Political Economy, Brussels: ECIPE, 2009. 	
   Hindman, Matthew. The Myth of Digital Democracy. Princeton, New Jersey: Princeton University Press, 2009. 	
   Homer-Dixon, Thomas. The Upside of Down: Catastrophe, Creativity, and the Renewal of Civilization. Washington, DC: Island Press, 2006. 	
   Howard, Philip. The Digital Origins of Dictatorship and Democracy: Information Technology and Political Islam. New York, NY: Oxford University Press, 2011.  Howard, Philip, and Nimah Mazaheri. "Telecommunications Reform, Internet Use and Mobile Phone Adoption in the Developing World." World Development 37, no. 7 (2009): 1159-1169. 	
   Howard, Philip, Sheetal Agarwal, and Muzammil Husain. "The Dictators’ Digital Dilemma: When Do States Disconnect Their Digital Networks?" The Center for Technology Innovation, The Brookings Institute, Washington, 2011. 	
   Howell, Jude. "Reflections on the Chinese State." Development and Change 37, no. 2 (2006): 273-297. 	
   	
   	
   91	
   Imtiaz, Saba. "Pakistan unfriends Facebook." Foreign Policy, May 19, 2010.  International Telecommunication Union (ITU). Vietnam Internet Case Study. Geneva, 2002.  —. World Telecommunication Indicators. Geneva, 2011. 	
   —. Mobile Cellular Subscriptions. Geneva, 2011.  Internet Society of China. “Public Pledge of Self-Regulation and Professional Ethics for China Internet Industry.” Revised July 19, 2002. Retrieved April 25, 2011 from http://www.isc.org.cn/20020417/ca102762.htm. 	
   Jackson, Robert, and Georg Sørensen. Introduction to International Relations: Theories and Approaches, 2nd Edition. Oxford: Oxford University Press, 2003.  Jackson, Steve. "Thai Ban on Youtube Website Ends." BBC News, August 31, 2007. 	
   Jenkins, Henry, and David Thorburn. Democracy and New Media. Cambridge, MA: MIT Press, 2004. 	
   Jiang, Min. "Authoritarian Informationalism: China's Approach to Internet Sovereignty." SAIS Review 30, no. 2 (2010): 71-89. 	
   Jones, Kristin. Nepal: Out of the Silence. Committee to Protect Journalists, 2005. 	
   Kahn, Richard, and Douglas Kellner. "Oppositional Politics and the Internet: A Critical/ Reconstructive Approach." Cultural Politics 1, no. 1 (2005). 	
   Kalathil, Shanthi. "China's Dot-Communism." Foreign Policy 122 (Jan-Feb 2001): 74- 75. 	
   Kalathil, Shanthi, and Taylor Boas. Open Networks Closed Regimes: The Impact of the Internet on Authoritarian Rule. Washington, DC: The Brookings Institute Press, 2003. 	
   Kedzie, Christopher. Communication and Democracy: Coincident Revolutions and the Emergent Dictator’s Dilemma. Santa Monica, CA: RAND, 1997. 	
   Keen, Andrew. "Social media and the internet do not spread democracy." The Telegraph, August 18, 2009. 	
   Kellner, Douglas. "September 11, Spectacles of Terror, and Media Manipulation: A Critique of Jihadist and Bush Media Politics." Logos 2, no. 1 (Winter 2003). 	
   Kranzberg, Melvin. "Kranzberg's Laws." Technology and Culture 27, no. 3 (LJuly 1986): 544-560. 	
   	
   92	
   	
   Krasner, Stephen. "Approaches to the State: Alternative Conceptions and Historical Dynamics." Comparative Politics 16, no. 2 (1984): 223-246. 	
   Kristof, Nicholas. "China Sees 'Market-Leninism' as Way to Future." The New York Times, September 6, 1993. 	
   —. "Tear Down This Cyberwall!" The New York Times, June 17, 2009. 	
   Lessig, Lawrence. "The Laws of Cyberspace." In Readings in Cyberethics, by Richard Spinello and Herman Tavani. Sudbury, MA: Jones and Bartlet, 2004. 	
   Levi, Margaret. Of Rule and Revenue. Berkeley, CA: University of California Press, 1989. 	
   Li, Lianjiang, and Kevin O’Brien. "Protest Leadership in Rural China." The China Quarterly 193 (March 2008): 1–23. 	
   Liang, Bin, and Hong Lu. "Internet Development, Censorship, and Cyber Crimes in China." Journal of Contemporary Criminal Justice 26, no. 1 (2010): 103-120. 	
   Liang, Guo. Surveying Internet Usage and its Impact in Seven Chinese Cities. Center for Social Development Chinese Academy of Social Sciences. Beijing: Markle Foundation, 2007. 	
   Lieberthal, Kenneth. Governing China. New York, NY: W.W.Norton, 2004. 	
   Lipset, Seymour. "Some Social Requisites of Democracy: Economic Development and Political Legitimacy." The American Political Science Review 53, no. 1 (March 1959): 69-105. 	
   Liu, Kwang-Ching, and Richard Shek. Heterodoxy in Late Imperial China. Honolulu: University of Hawai'i Press, 2004. 	
   Livingston, Stephen, and Lance Bennett. "Gatekeeping, Indexing, and Live-Event News: Is Technology Altering the Construction of News?" Political Communication 20, no. 4 (2003): 363–380. 	
   Lohmann, Susanne. "The Dynamics of Informational Cascades: The Monday Demonstrations in Leipzig, East Germany, 1989-91." World Politics, October 1994: 42-101. 	
   McAdam, Doug. "The Future of Social Movements." In From Contention to Democracy, by Marco Giugni, Doug McAdam and Charles Tilly, 229-246. Lanham, MD: Rowman & Littlefield, 1998. 	
   —. "Recruitment to High-Risk Activism: The Case of Freedom Summer." The American Journal of Sociology 92, no. 1 (July 1986): 64-90. 	
   	
   93	
   	
   McAdam, Doug. “Clarifying the Concept of Political Opportunities.” In Comparative Perspectives on Social Movements: Political Opportunities, Mobilizing Structures, and Cultural Framings, by Doug McAdam, John McCarthy, and Mayer Zald. New York, NY: Cambridge University Press, 1996.  McAdam, Doug, John McCarthy, and Mayer Zald. Comparative Perspectives on Social Movements: Political Opportunities, Mobilizing Structures, and Cultural Framings. New York, NY: Cambridge University Press, 1996. 	
   McAdam, Doug, Sidney Tarrow, and Charles Tilly. Dynamics of Contention. New York, NY: Cambridge University Press, 2001. 	
   McCargo, Duncan. Media and Politics in Pacific Asia. New York: RoutledgeCurzon, 2003.  —. "Network monarchy and legitimacy crises in Thailand." The Pacific Review, August 2006. 	
   McGregor, Richard. The Party: The Secret World of China's Communist Rulers. New York, NY: HarperCollins, 2010.  —. "5 Myths About the Chinese Communist Party: Market-Leninism Lives." Foreign Policy, January/February 2011. 	
   McLuhan, Marshal. Understanding Media: The Extensions of Man. Boston, MA: MIT Press, 1965. 	
   MacKinnon, Rebecca. "China's Censorship 2.0: How Companies Censor Bloggers." First Monday 14, no. 2 (2009).  —. "Networked Authoritarianism in China and Beyond: Implications for global Internet freedom." Liberation Technology in Authoritarian Regimes. 2010.  —.  "Corporate Accountability in Networked Asia." In Access Contested: Security, Identity, and Resistance in Asian Cyberspace, by Ronald Dieber, John Palfrey, Rafal Rohozinski and Jonathan Zittrain, 195-215. Cambridge, MA: MIT Press, 2011. 	
   Maclay, Colin. "Protecting Privacy and Expression Online: Can the Global Network Initiative Embrace the Character of the Net?" In Access Controlled: The Shaping of Power, Rights, and Rule in Cyberspace, by Ronald Deibert, John Palfrey, Rafal Rohozinski and Jonathan Zittrain. Cambridge, MA: MIT Press, 2010. 	
   Mahmood, Rumel. "Can Information and Communication Technology Help Reduce Corruption? How So and Why Not: Two Case Studies from South Asia." Perspectives on Global Development and Technology 3, no. 3 (2004). 	
   	
   94	
   	
   Mann, Michael. "The Autonomous Power of the State: Its Origins, Mechanisms, and Results." European Journal of Sociology 25, no. 2 (1984): 185-213. 	
   Marsh, David, and Paul Furlong. "A Skin not a Sweater: Ontology and Epistemology in Political Science." In Theory and Methods in Political Science, by David Marsh and David Stoker. Houndmills: Palgrave Macmillan, 2002. 	
   Meier, Patrick. “The Impact of the Information Revolution on Protest Frequency in Repressive Contexts.” Paper presented at the 50th International Studies Association Convention, New York, February 15th-18th, 2007.  Meier, Patrick, and Victoria Stodden. “A Global Empirical Evaluation of New Communication Technology Use and Democratic Tendency.” Paper presented at the Information and Communication Technologies Development (ICTD) 2009 International Conference, April 17-19, 2009.  Meilian, Lin. "Wukan's Democracy." Global Times, February 8, 2012.  Memmort, Mark. "Clinton: Dictators Will Pay A Price If They Try To Shut Off The Web." Natinoal Public Radio, February 15, 2011. http://www.npr.org/blogs/thetwo- way/2011/02/15/133779438/clinton-dictators-will-pay-a-price-if-they-try-to- shut-off-the-web (accessed February 21, 2012). 	
   Migdal, Joel. Strong Societies and Weak States. Princeton, NJ: Princeton University Press, 1998.  —. State in Society: Studying How States and Societies Transform and Constitute Each Other. New York, NY: Cambridge University Press, 2001. 	
   Moore, Malcolm. "Inside Wukan: The Chinese Village That Fought Back." The Telegraph, December 13, 2011. 	
   Morozov, Evgeny."Facebook and Twitter are just places revolutionaries go." The Guardian, March 7, 2011. 	
   —. "Technology's Role in Revolution: Internet Freedom and Political Oppression." The Futurist, July-August 2011: 18-21. 	
   —."Response to Philip N. Howard’s review of The Net Delusion: The Dark Side of Internet Freedom." Perspectives on Politics 9, no. 4 (December 2011): 897. 	
   —. The Net Delusion: The Dark Side of Internet Freedom. New York, NY: PublicAffairs, 2012. 	
   Morris, Sarah. "Spanish Govt to Guarantee Legal Right to Broadband." Reuters, November 17, 2009. 	
   	
   95	
   	
   Mueller, Milton. "China and Global Internet Governance: A Tiger by the Tail." In Access Contested: Security, Identity, and Resistance in Asian Cyberspace, by Ronald Diebert, John Palfrey, Rafal Rohozinski and Jonathan Zittrain, 177-194. Cambridge, MA: MIT Press, 2011. 	
   Nathan, Andrew. "Authoritarian Resilience." Journal of Democracy 14, no. 1 (January 2003): 6-17. 	
   Ng, Konrad. "Elites and Elitism in American Democracy: Do Model Minorities Drink Wine or Beer? The Representation of Race, Elitism and American Identity." In The Obama Effect: Multidisciplinary Renderings of the 2008 Presidential Election Campaign, by Heather Harris, Kimberly Moffitt and Catherine Squires. New York: State University of New York Press, 2010. 	
   O'Brien, Kevin, and Lianjiang Li. Rightful Resistance in Rural China. New York, NY: Cambridge University Press, 2006. 	
   OpenNet Initiative. "Asia Regional Overview." OpenNet Initiative. 2007. http://opennet.net/research/regions/asia (accessed February 19, 2012).  —. "North Korea Country Report." OpenNet Initiative. May 10, 2007. http://opennet.net/research/profiles/north-korea (accessed February 18, 2012). 	
   —. "Singapore: Country Profile." OpenNet Initiative. May 10, 2007. 	
   —.  "YouTube and the rise of geolocational filtering." OpenNet Initiative. March 13, 2008. http://opennet.net/blog/2008/03/youtube-and-rise-geolocational- filtering (accessed February 18, 2012).  —. Internet Censorship Data. Cambridge, MA: MIT Press, 2011. http://opennet.net/research/profiles/singapore (accessed February 18, 2012).  —. "Country Report: China - 2011." OpenNet Initiative. 2011. 	
   —. "Filtering Data." OpenNet Initiative. November 8, 2011. http://opennet.net/research/data (accessed February 19, 2012).  O'Reilly, Tim. What is Web 2.0. September 30, 2005. http://oreilly.com/web2/archive/what-is-web-20.html (accessed February 22, 2012). 	
   Organization For Economic Cooperation and Development (OECD). Participative Web and User-Created Content: Web 2.0, Wikis, and Social Networking. OECD, OECD Publishing, 2007.   	
   	
   96	
   PC Magazine. PCMAG.COM Encyclopedia. http://www.pcmag.com/encyclopedia_term/0,1237,t=digital+convergence&i=41 316,00.asp (accessed September 12, 2012). 	
   People's Daily. "School Explosion Aftermath in Jiangxi Well Handled." People's Daily, March 2001. 	
   —. "Truth About Jiangxi School Building Collapse Known." People's Daily, March 2001. 	
   People's Republic of China State Council Information Office. "The Internet in China: White Paper." 2010. 	
   Perry, Elizabeth. Challenging the Mandate of Heaven: Social Protest and State Power in China. New York, NY: M.E.Sharpe, 2002. 	
   Persson, Torsten, and Guido Tabellini. The Economic Effects of Constitutions. Boston, MA: MIT Press, 2003. 	
   Potter, Pitman. "Legal Reform in China: Institutions, Culture, and Selective Adaptation." Law & Social Inquiry 29, no. 2 (April 2004): 465-495. 	
   Price, Richard, and Christian Reus-Smit. "Dangerous Liaisons?: Critical International Theory and Constructivism." European Journal of International Relations 4, no. 3 (1998): 259. 	
   Przeworski, Adam, Michael Alvarez, and Jose Antonio, Limogini, Fernando Cheibub. Democracy and Development: Poitical Institutions and Well-Being in the World, 1950-1990. New York, NY: Cambridge University Press, 2000. 	
   Qiu, Jack Linchuan. Working-Class Network Society: Communication Technology and the Information Have-Less in Urban China. Boston, MA: MIT Press, 2009. 	
   Quinn, Andrew. "China faces Internet 'dictator's dilemma': Clinton." Reuters, Feburary 15 2011. 	
   Reporters Without Borders. "Five Key Problems for Media Coverage of February's Legislative Elections." Reporters Without Borders, January 9, 2008. http://www.rsf.org/pakistan-five-key-problemsfor-media-09-01-2008,24976.html/.  —. Internet Enemies. Paris: Reporters Without Borders, 2011.  —. "Arrests, Trials and Sentences Offer No Respite for Chinese Dissidents." Reporters Without Borders, 2012.  Rheingold, Howard. Smart Mobs: the Next Social Revolution. Cambridge, MA: Perseus Books Group, 2002. 	
   	
   	
   97	
   Rodan, Gary. "The Internet and Political Control in Singapore." Political Science Quarterly 113, no. 1 (Spring 1998): 63-89. 	
   Rodrick, Dani. The New Global Economy and Developing Countries: Making Openness Work. Washington, DC: Overseas Development Council, 1999.  —. "How Far will International Economic Integration Go?" Journal of Economic Perspectives 14, no. 1 (Winter 2000): 177. 	
   Rogers, Everett. Diffussion of Innovations, 5th Edition. New York, NY: Free Press, 2003. 	
   Rosen, Jay. The People Formerly Know as the Audience. June 27, 2006. http://archive.pressthink.org/2006/06/27/ppl_frmr.html (accessed February 27, 2012).  —.  The “Twitter Can’t Topple Dictators” Article. February 12, 2012. http://pressthink.org/2011/02/the-twitter-cant-topple-dictators-article/ (accessed February 28, 2012). 	
   Rothstein, Bo. "Creating Political Legitimacy: Electoral Democracy Versus Quality of Government." American Behavioralist Scientist 53, no. 9 (2009): 311-330. 	
   Ruwitch, John. "Web attacks hit Vietnam bauxite activists: Google." Reuters, April 2, 2010. 	
   Saich, Tony. Governance and Politics of China. New York, NY: Palgrave, 2001. 	
   Said, Edward. "Orientalism Reconsidered." Cultural Critique 1 (1985): 89-107.  —. Covering Islam: How the Media and the Experts Determine How We See the Rest of the World. Vintage Books, 1997.  —. "Islam and the West are Inadequate Banners." The Observer, September 16, 2001.  —. Orentalism. Vintage Books, 2003. 	
   Salter, Lee. "Democracy, New Social Movements and the Internet." In Cyberactivism: Critical Practices and Theories of Online Activism, by M. Ayers and M. McCaughey. New York, NY: Routledge, 2003. 	
   Schuman, Michael. The Miracle: The Epic Story of Asia's Quest for Wealth. New York, NY: HarperCollins, 2010. 	
   Sen, Amartya. Poverty and Famines: An Essay on Entitlement and Deprivation. Oxford: Oxford University Press, 1982.  	
   	
   98	
   —. Development as Freedom. New York: Random House, 1999. 	
   —. "Democracy as a Universal Value." Journal of Democracy 10, no. 3 (1999): 1999. 	
   Shambaugh, David. China's Communist Party: Atrophy and Adaptation. Washington, DC: Woodrow Wilson Center Press, 2008. 	
   Sheridan, Mary Beth. "Clinton Warns Governments That Limiting Internet Will Backfire." The Washington Post, February 15, 2011. 	
   Shirky, Clay. Here Comes Everybody: The Power of Organizing Without Organizations. London: Penguin Press, 2008.  —. "The Political Power of Social Media: Technology, the Public Sphere, and Political Change." Foreign Affairs, January/February 2011. 	
   Sills, Jennifer. "China's Environmental Civilian Activism." Science, May 14, 2010: 824. 	
   Singapore Department of Statistics, Ministry of Industry and Trade. Yearbook of Statistics Singapore, 2011. Singapore: Singapore Ministry of Industry and Trade, 2011. 	
   Singel, Ryan. "NSA's Lucky Break: How the U.S. Became Switchboard to the World." Wired, October 10, 2007. 	
   Singh, Rajendra, and Siddhartha Raja. Convergence in Information and Communication Technology: Strategic and Regulatory Considerations. Washington, DC: The World Bank, 2010. 	
   Skopcol, Theda. "Bringing the State Back In: Strategies for Analysis in Current Research." In Bringing the State Back In, by Peter Evans, Dietrich Rueschemeyer, and Theda Skocpol. New York, NY: Cambridge University Press, 1985. 	
   Smith, Steve. "Positivism and Beyond." In International Theory: Positivism and Beyond, by Steve Smith, Ken Booth and and Marysia Zalewski, 11-46. Cambridge, MA: Cambridge University Press, 1996. 	
   Spence, Jonathan. God's Chinese Son: The Taiping Heavenly Kingdom of Hong Xiuquan. New York, NY: W.W.Norton & Company, 1996. 	
   Starr, John Bryan. Understanding China: A Guide to Chinese Economy, History, and Political Culture. New York, NY: Hill and Wang, 2010. 	
   State Council Information Office of China. China Internet Whitepaper, June 2010. http://www.china.org.cn/government/whitepaper/2010- 06/08/content_20208007.htm (accessed September 12, 2012). 	
   	
   99	
   	
   Stiglitz, Joseph. "Information and the Change in the Paradigm of Economics.” Nobel prize lecture delivered at Aula Magna, Stockholm University, December 8, 2001.  —. Globalization and Its Discontents. New York, NY: W.W.Norton, 2002. 	
   Stockmann, Daniela, and Mary Gallagher. "Remote Control: How the Media Sustain Authoritarian Rule in China." Comparative Political Studies 44, no. 4 (February 2011): 436-467.  Stone, Richard. "Internet Blockade in Xinjiang Puts A Strain on Science." Science 326, no. 5959 (December 2009): 1471. 	
   Stone, Richard, and Hao Xin. "Google Plots Exit Strategy As China Shores Up 'Great Firewall'." Science, January 22, 2010: 402-403. 	
   Sykes, Alan. "International Trade and Human Rights: An Economic Perspective." In International Trade and Human Rights: Foundations and Conceptual Issues, by Frederick M. Abbot, Christian Breining-Kaufmann and Thomas Cottier. Ann Arbor: University of Michigan Press, 2006. 	
   Synovitz, Ron. "The Revolution Will Be Tweeted - Moldovan Protesters Exploit Social Networking Sites." Radio Free Europe, April 8, 2009. 	
   Tiberghien, Yves. Entrepreneurial States: Reforming Corporate Governance in France, Japan, and Korea. New York, NY: Cornell University Press, 2007.  Tilly, Charles. Coercion, Capital, and European States, AD 990-1992. Malden, MA: Blackwell Publishing, 1992.  —. "Social Movements and (All Sorts of) Other Political Interations- Local, National, and International - Including Identities." Theory and Society 27, no. 4 (August 1998): 453-480.  —. "Processes and Mechanisms of Democratization." Sociological Theory 18, no. 1 (March 2000): 1-16.  —. "Mechanisms in Political Processes." Annual Review of Political Science 4 (2001): 21-41. 	
   —. Democracy. New York, NY: Cambridge University Press, 2007.  The Nation. "Don't use lese majeste as a political tool." The Nation, November 8, 2008. 	
   Thelen, Kathleen. "How Institutions Evolve: Inisghts from Comparative Historical Analysis." In Comparative Historical Analysis in the Social Sciences, by James 	
   	
  100	
   Mahoney and Dietrich Rueschemeyer. New York, NY: Cambridge University Press, 2003. 	
   Tocqueville, Alexis de. Democracy in America, Vol.1. New York, NY: Scatched and Adams, 1839. 	
   —. The Old Regime and the Revolution. New York, NY: Harper & Brothers, 1856. 	
   Tweed, Thomas A. "Why are Buddhists so Nice? Media Representations of Buddhism and Islam in the United States Since 1945." Material Religion: The Journal of Objects, Art and Belief 4, no. 1 (2008): 91–93. 	
   United Nations. "Universal Declaration of Human Rights." G.A. Res. 217A (III), U.N. Doc. A/810 at 71, 1948.  —. "International Covenant on Economic, Social and Cultural Rights, Dec. 16, 1966." 993 U.N.T.S. 3; S. Exec. Doc. D, 95-2 (1978); S. Treaty Doc. No. 95-19, 6 I.L.M. 360 (1967), 1967. 	
   United Nations Children’s Fund (UNICEF). Education Statistics: Turkmenistan. New York, NY: UNICEF, 2008. 	
   United Nations Development Programme (UNDP). Human Development Report 2011 - Sustainability and Equity: A Better Future for All. New York, NY: Palgrave MacMillan, 2011. 	
   United Nations Educational, Scientific, and Cultural Organization (UNESCO). UNESCO Institute of Statistics Indicators 2011. Montreal, Quebec: UNESCO Institute of Statistics, 2011. 	
   United States Congress. "Restore Act of 2007 - H.R.3773." 110th Congress (2007-2008), 2007.  —. "Google and Internet Conrol in China: A Nexus Between Human Rights and Trade?" Testimony of Edward Black, President and CEO, Computer and Communications Industry Association. Congressional Executive Commission on China, 2010. 	
   —. "The Global One Internet Act of 2010: For the technical integrity of the lnternet, market opportunities for U.S. businesses, and the global free flow of information." 2010. 	
   United States National Archives and Records Administration. Public Papers of the Presidents of the United States: William J. Clinton: Book I, January 1 to June 26, 2000. Washington, DC: National Archives and Records Administration, 2005. 	
   	
   	
  101	
   Wall Street Journal. "South Korean Blogger Beats False-Information Charge." Wall Street Journal, April 21, 2009. 	
   Wang, Stephanie.  "Pulling the Plug: A Technical Review of the Internet Shutdown in Burma." OpenNet Initiative, 2007. 	
   —."Clouds Over China's 'Sunshine' Law." Asia Times, August 1, 2009.  Weber, Max. "Politics as a Vocation." In Social Ethics Series 3, by Sherman Franklin, 1- 7. Philadelphia, PA: Fortress Press, 1965. 	
   Wendt, Alexander. Social Theory of International Politics. Cambridge, MA: Cambridge University Press, 2003. 	
   Wilson, Sven, and Daniel Butler. "A Lot More to Do: The Sensitivity of Time-Series Cross-Section Analyses to Simple Alternative Specifications." Political Analysis 15 (2007): 101-123. 	
   Wines, Michael. "A Dirty Pun Tweaks China's Online Censors." New York Times, March 12, 2009. 	
   —. "China Expands Program Requiring Real-Name Registration Online." The New York Times, January 18, 2012. 	
   Winters, Alan, and Shahid Yusuf. Dancing With Giants: China, India, and the Global Economy. Washington, DC: The Word Bank, 2007. 	
   Woodier, Jonathan. The Media and Political Change in Southeast Asia: Karaoke Culture and the Evolution of Personality Politics. Edward Elgar Publishing, 2008. 	
   World Bank. "Workd Bank Governance Indicators Documentation Introduction." World Governance Indicators. Washington, CD: World Bank, 2011. http://info.worldbank.org/governance/wgi/resources.htm (accessed February 19, 2012). 	
   —. Development Indicators 2011. Washington, CD: World Bank, 2011. 	
   —. ICT Glossary Guide. Washington, CD: World Bank, 2011. http://web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTINFORMATION ANDCOMMUNICATIONANDTECHNOLOGIES/0,,contentMDK:21035032~men uPK:282850~pagePK:210058~piPK:210062~theSitePK:282823~isCURL:Y,00.h tml#I (accessed February 22, 2012). 	
   World Trade Organization. "General Agreement on Trade in Services, Apr. 15, 1994, Marrakesh Agreement Establishing the World Trade Organization, Annex 1B, 1869 U.N.T.S. 183;33 I.L.M. 1167(1994)." 1994. 	
   	
  102	
   	
   Wu, Tim. "The World Trade Law of Censorship and Internet Filtering." Chicago Journal of International Law 263 (2006). 	
   Wu, Timothy. "Cyberspace Sovereignty? The Internet and the International System ." Harvard Journal of Law & Technology 10, no. 3 (Summer 1997): 647-666. 	
   Xiao, Qiang. "Baidu’s Internal Monitoring and Censorship Document Leaked." China Digital Times, August 30, 2009. 	
   Xinhua. "Sunshine Policy to Fight Corruption in China's Capital ." Xinhuanet, March 13, 2002.  —."Official: Internet Cut in Xinjiang to Prevent Riot from Spreading." Xinhuanet, July 7, 2009. 	
   —. "Myanmar tightens security measures with internet cafe running." Xinhuanet, December 1, 2010. 	
   —. "Social Problems Biggest Challenge for Guangdong." Xinhuanet, January 17, 2012. 	
   Yang, Guobin. The Power of the Internet in China. New York, NY: Columbia University Press, 2009. 	
   Zhao, Yuezhi. "Falun Gong and the Struggle Over Meaning Inside and Outside China." In Contesting Media Power: Alternative Media in a Networked World, by Nick Couldry and James Curran, 209-223. Oxford: Rowman & Littlefield Publishers, 2003.  —. Communication in China: Political Economy, Power, and Conflict. Lanham, MD: Rowman & Littlefield, 2008. 	
   —. "Surprises Loom: An uncensored Chinese media might illuminate a citizenry that is both anti-Western and anti-capitalist." The Vancouver Sun, May 21, 2008. 	
   Zhao, Jinqui. "The SARS Epidemic Under China's Media Policy." Media Asia: An Asian Mass Media Communication Quarterly 4, no. 30 (2003): 191-196.  —. "A Snapshot of Internet Regulation in Contemporary China: Censorship, Profitability and Responsibility." In From Early Tang Court Debates to China's Peaceful Rise, by Friederike Assandri and Dora Martins, 141-151. Amsterdam: Amsterdam University Press, 2009. 	
   Zhou, Hanhua. "Open Government in China: Practice and Problems." In Transparency for an Open Word, by Ann Florini, 92-115. New York, NY: Columbia University Press, 2007. 	
   	
   	
  103	
   Zhu, Shanshan, and Hui Zhang. "ID rule to shake up Weibo." People's Daily Online, February 8, 2012. 	
   Zhu, Yuchao. “’Performance Legitimacy’ and China’s Political Adaptation Strategy.” Journal of Chinese Political Science 16 (2011): 123-140. 	
   Zittrain, Jonathan, and John Palfrey. "Internet Filtering: The Politics and Mechanisms of Control." In Access Denied, by Ronald Dieber, John Palfrey, Rafal Rohozinski and Jonathan Zittrain, 29-56. Cambridge, MA: MIT Press, 2008. 	
   —. "Reluctant Gatekeepers: Corporate Ethics on a Filtered Internet." In Access Denied: The Practice and Policy of Global Internet Filtering, by Ronald Dieber, John Palfrey, Rafal Rohozinski and Jonathan Zittrain. Cambridge, MA: MIT Press, 2008. 	
   Zuckerman, Ethan. Draft paper on mobile phones and activism. April 26, 2007. http://www.ethanzuckerman.com/blog/2007/04/09/draft-paper-on-mobile- phones-and-activism/ (accessed February 13, 2012).  —. What if Tunisia Had a Revolution but Nobody Watched. January 12, 2011. http://www.ethanzuckerman.com/blog/2011/01/12/what-if-tunisia-had-a- revolution-but-nobody-watched/ (accessed February 26, 2012). 	
   —. "The First Twitter Revolution?" Foreign Policy, January 14, 2011. 	
    	
   	
  104	
   Appendices	
  	
   Table	
  A.1	
  –	
  Quantitative	
  Analysis:	
  Descriptive	
  Statistics	
  (sorted	
  on	
  democracy	
  level)	
   	
   Case	
   Country	
   Years	
   of	
   Data	
   Democracy	
   Score	
   (mean)	
   Internet/mobile	
   per	
  100	
  pop.	
   (mean)	
   GNI	
  p.c.	
   PPP	
   ($	
  mean)	
   Educational	
   Enrolment	
   (%	
  mean)	
   Log	
  of	
   Population	
   (mean)	
   8	
   Japan	
   16	
   3.1	
   47	
   27851.30	
   101.71	
   8.11	
   11	
   Korea,	
  South	
   16	
   3.6	
   50	
   19001.30	
   100.71	
   7.67	
   15	
   Mongolia	
   16	
   4.5	
   12	
   2380.00	
   86.27	
   6.39	
   6	
   India	
   16	
   5.5	
   5	
   1903.13	
   75.92	
   9.02	
   19	
   Philippines	
   16	
   5.7	
   15	
   2718.75	
   93.25	
   7.90	
   23	
   Thailand	
   16	
   6.7	
   21	
   5578.13	
   79.09	
   7.81	
   1	
   Bangladesh	
   16	
   7.3	
   4	
   1052.50	
   73.32	
   8.12	
   21	
   Sri	
  Lanka	
   16	
   7.4	
   8	
   2985.00	
   95.32	
   7.28	
   17	
   Nepal	
   16	
   8.0	
   2	
   858.75	
   77.64	
   7.40	
   7	
   Indonesia	
   16	
   8.0	
   9	
   2629.38	
   84.32	
   8.34	
   14	
   Malaysia	
   16	
   8.9	
   35	
   9572.50	
   81.53	
   7.38	
   20	
   Singapore	
   16	
   9.4	
   55	
   36131.30	
   100.00	
   6.61	
   12	
   Kyrgyzstan	
   16	
   9.6	
   14	
   1461.88	
   92.26	
   6.70	
   18	
   Pakistan	
   16	
   10.1	
   8	
   1880.63	
   53.88	
   8.17	
   9	
   Kazakhstan	
   16	
   11.0	
   14	
   6161.88	
   98.86	
   7.18	
   4	
   Cambodia	
   16	
   11.3	
   4	
   1150.81	
   70.10	
   7.10	
   3	
   Brunei	
   16	
   11.6	
   36	
   45510.60	
   103.88	
   5.52	
   22	
   Tajikistan	
   16	
   11.9	
   7	
   1152.50	
   88.86	
   6.80	
   2	
   Bhutan	
   16	
   12.0	
   5	
   2918.13	
   65.07	
   5.78	
   13	
   Lao	
  PDR	
   16	
   13.0	
   5	
   1380.63	
   74.82	
   6.73	
   26	
   Vietnam	
   16	
   13.1	
   12	
   1713.13	
   86.89	
   7.90	
   5	
   China	
   16	
   13.3	
   13	
   3281.25	
   89.33	
   9.10	
   25	
   Uzbekistan	
   16	
   13.4	
   6	
   1721.88	
   96.52	
   7.40	
   10	
   Korea,	
  Dem.	
   Rep.	
   1	
   14.0	
   0	
   1005.00	
   101.87	
   7.38	
   16	
   Myanmar	
   16	
   14.0	
   0	
   802.94	
   76.70	
   7.65	
   24	
   Turkmenistan	
   16	
   14.0	
   2	
   3305.00	
   84.55	
   6.66	
   	
   Total	
   401	
   	
   	
   	
   	
   	
  	
   	
  	
   	
   	
   	
  105	
   Table	
  A.2	
  –	
  Quantitative	
  Analysis:	
  Individual	
  States	
  Democracy	
  /	
  ICT	
  Levels	
   	
   Note: This table displays the initial and final democracy and ICT levels for individual states in the data. Initial values represent the first year for which data is available. Democracy (as above) is measured on the combined Freedom House index, ranging from 2 (most democratic) to 14 (least democratic). ICT is measured (as above) as the combination of Internet and mobile users average per 100 population.  Caveat: This table looks only at initial and final values and not at variable movements within the time period. For example, several states experienced drastic changes in democracy level within the time period only to return to earlier levels later. A better representation can be found in figures 4.2, 4.3, and 4.4 in the text. These display the movement in both democracy and ICT over the data period for each individual state as a correlation. 	
   Country	
   Democracy	
   (initial)	
   Democracy	
  	
   (final)	
   Democracy	
  	
   (delta)	
   ICT	
   (initial)	
   ICT	
   (initial)	
   ICT	
   (delta)	
   Japan	
   4	
   3	
   1	
   2	
   84	
   82	
   Korea,	
   South	
   4	
   3	
   1	
   1	
   90	
   89	
   Mongolia	
   5	
   4	
   1	
   0	
   50	
   50	
   India	
   8	
   5	
   3	
   0	
   25	
   25	
   Philippines	
   7	
   7	
   0	
   0	
   44	
   44	
   Thailand	
   8	
   9	
   -­‐1	
   1	
   73	
   72	
   Bangladesh	
   6	
   7	
   -­‐1	
   0	
   17	
   17	
   Sri	
  Lanka	
   9	
   8	
   1	
   0	
   38	
   38	
   Nepal	
   7	
   8	
   -­‐1	
   0	
   14	
   14	
   Indonesia	
   13	
   5	
   8	
   0	
   38	
   28	
   Malaysia	
   9	
   8	
   1	
   1	
   83	
   82	
   Singapore	
   10	
   9	
   1	
   4	
   103	
   99	
   Kyrgyzstan	
   7	
   11	
   -­‐4	
   0	
   63	
   63	
   Pakistan	
   8	
   9	
   -­‐1	
   0	
   36	
   36	
   Kazakhstan	
   11	
   11	
   0	
   0	
   64	
   64	
   Cambodia	
   9	
   11	
   -­‐2	
   0	
   11	
   11	
   Brunei	
   13	
   11	
   2	
   3	
   95	
   92	
   Tajikistan	
   14	
   11	
   3	
   0	
   41	
   41	
   Bhutan	
   14	
   9	
   5	
   0	
   26	
   26	
   Lao	
  PDR	
   13	
   13	
   0	
   0	
   29	
   29	
   Vietnam	
   14	
   12	
   2	
   0	
   65	
   65	
   China	
   14	
   13	
   1	
   0	
   42	
   42	
   Uzbekistan	
   14	
   14	
   0	
   0	
   38	
   38	
   Korea,	
   North	
   14	
   14	
   0	
   0	
   0	
   0	
   Myanmar	
   14	
   14	
   0	
   0	
   1	
   1	
   Turkmenista n	
   14	
   14	
   0	
   0	
   16	
   16	
   	
   	
   	
  106	
   Table	
  A.3	
  –	
  Country	
  Codes	
   	
   Note: This table displays the codes used to identify states in figures 4.2, 4.3, and 4.4 in the text. 	
   COUNTRY CODE Bangladesh	
   BGD	
   Bhutan	
   BTN	
   Brunei	
   BRN	
   Cambodia	
   KHM	
   China	
   CHN	
   India	
   IND	
   Indonesia	
   IDN	
   Japan	
   JPN	
   Kazakhstan	
   KAZ	
   Korea,	
  Dem.	
   Rep.	
   PRK	
   Korea,	
  South	
   KOR	
   Kyrgyzstan	
   KGZ	
   Lao	
  PDR	
   LAO	
   Malaysia	
   MYS	
   Mongolia	
   MNG	
   Myanmar	
   MMR	
   Nepal	
   NPL	
   Pakistan	
   PAK	
   Philippines	
   PHL	
   Singapore	
   SGP	
   Sri	
  Lanka	
   LKA	
   Tajikistan	
   TJK	
   Thailand	
   THA	
   Turkmenistan	
   TKM	
   Uzbekistan	
   UZB	
   Vietnam	
   VNM	
   	
   	
   	
   	
   	
   	
   	
   	
   	
   	
   	
   	
   	
  107	
   	
   Table	
  A.4	
  –	
  Quantitative	
  Analysis:	
  Descriptive	
  Statistics	
  All	
  Countries	
   	
   Note: Literacy and online censorship values are for 2009 only. 	
   	
   	
  	
   Table	
  A.5	
  –	
  Quantitative	
  Analysis:	
  Online	
  Censorship	
  Levels	
  for	
  2009	
   Note: values in this table were calculated by summing the numeric scores for political censorship, conflict-related censorship, and censorship of communication tools. Social issues censorship scores are excluded. Source is the OpenNet Initiative. 	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   VARIABLE MEAN SD MAX MIN DEMOCRACY 9.5	
   3.5	
   14	
   3	
   INTERNET/MOBILE 16	
   23	
   104	
   0	
   INTERNET/MOBILE (T-3) 9.7	
   18	
   86	
   0	
   GNI P.C. PPP 7388	
   11916	
   51980	
   420	
   EDUCATIONAL ENROLMENT 85	
   13	
   110	
   51	
   LITERACY 87	
   16	
   100	
   43	
   LOG OF POPULATION 7.4	
   0.86	
   9.1	
   5.4	
   ONLINE CENSORSHIP 5.5	
   4.6	
   14	
   0	
   COUNTRY CENSORSHIP CHINA 11	
   MYANMAR 10	
   VIETNAM 9	
   TURKMENISTAN 8	
   UZBEKISTAN 8	
   PAKISTAN 7	
   INDIA 6	
   INDONESIA 4	
   SOUTH KOREA 4	
   THAILAND 4	
   KAZAKHSTAN 2	
   KYRGYZSTAN 2	
   PHILIPPINES 2	
   TAJIKISTAN 2	
   	
   	
  108	
   Table	
  A.6	
  –	
  Quantitative	
  Analysis:	
  Data	
  Sources	
  and	
  Descriptions	
   VARIABLE DESCRIPTION SOURCE Democracy	
   Measure	
   (Demolevel)	
   The	
  main	
  dependent	
  variable	
  throughout	
  the	
  paper.	
   Based	
  on	
  the	
  Freedom	
  House	
  democracy	
  index,	
  this	
   measures	
  a	
  combination	
  of	
  political	
  rights	
  and	
  civil	
   liberties,	
  each	
  on	
  a	
  7-­‐point	
  scale.	
  These	
  have	
  been	
   combined	
  for	
  a	
  range	
  of	
  2	
  (most	
  democratic)	
  to	
  14	
   (least	
  democratic)	
  (Freedom	
  House	
  2011).	
   	
   Freedom	
  House	
   Internet	
  and	
   Mobile	
  Usage	
   (IntMob)	
   The	
  main	
  independent	
  variable	
  combines	
  internet	
   and	
  mobile	
  usage	
  to	
  act	
  as	
  a	
  measure	
  of	
  the	
  level	
  of	
   digital	
  communications.	
  It	
  is	
  calculated	
  as	
  the	
   average	
  of	
  internet	
  users	
  per	
  100	
  population	
  and	
   mobile	
  users	
  per	
  100	
  population.	
  This	
  can	
  range	
   from	
  0	
  to	
  above	
  100	
  in	
  cases	
  where	
  multiple	
  mobiles	
   are	
  common.	
  Data	
  is	
  sourced	
  from	
  the	
  World	
  Bank,	
   which	
  in	
  turn,	
  receives	
  this	
  data	
  from	
  the	
   International	
  Telecommunications	
  Union	
  (ITU).	
   Countries	
  provide	
  data	
  directly	
  to	
  the	
  ITU	
  (World	
   Bank	
  2011).	
   International	
   Telecommunications	
   Union	
  (through	
   World	
  Bank)	
   Literacy	
  	
   The	
  percentage	
  of	
  population	
  above	
  age	
  15	
   considered	
  literate.	
  This	
  is	
  used	
  as	
  a	
  measure	
  of	
   social	
  development	
  instead	
  of	
  EduEnrol	
  for	
   multivariate	
  regressions	
  due	
  to	
  greater	
  data	
   availability	
  (World	
  Bank	
  2011).	
   World	
  Bank	
   	
   Educational	
   Enrolment	
   (EduEnrol)	
   For	
  time	
  series	
  data,	
  this	
  is	
  used	
  as	
  a	
  measure	
  of	
   social	
  development.	
  It	
  averages	
  the	
  percent	
  of	
   primary	
  school-­‐age	
  enrolment	
  with	
  secondary	
   school-­‐age	
  enrolment.	
  This	
  can	
  range	
  above	
  100	
  for	
   cases	
  where	
  there	
  is	
  considerable	
  repetition	
  of	
   grades	
  at	
  the	
  age	
  range	
  boundary	
  or	
  when	
  non-­‐ residents	
  are	
  counted.	
  The	
  most	
  recent	
  year	
  for	
   which	
  data	
  is	
  available	
  was	
  used	
  to	
  populate	
   contiguous	
  years.	
  Data	
  for	
  Singapore	
  and	
   Turkmenistan	
  entered	
  by	
  author	
  (Singapore	
   Department	
  of	
  Statistics	
  2011;	
  UNESCO	
  2011;	
   UNICEF	
  2008).	
   UNESCO,	
  UNICEF,	
   Singapore	
  Dept.	
  of	
   Statistics	
   Log	
  of	
   population	
   (PopLog)	
   The	
  log	
  of	
  total	
  population	
  (World	
  Bank	
  2011).	
   World	
  Bank	
   Gross	
  National	
   Income	
   Per	
  Capita	
  and	
  Purchasing	
  Power	
  Parity	
  basis	
  (World	
   Bank	
  2011;	
  UNDP	
  2011).	
   World	
  Bank,	
  UNDP	
   	
   	
  109	
   (GNIPCPPP)	
   Online	
   Censorship	
   (CensorONI)	
   A	
  measure	
  of	
  filtering	
  and	
  censorship	
  of	
  online	
   information	
  used	
  for	
  the	
  multivariate	
  regressions.	
  	
   This	
  is	
  calculated	
  by	
  adding	
  individual	
  scores	
  for	
  four	
   categories	
  of	
  censorship:	
  political,	
  social,	
  and	
   conflict-­‐related	
  filtering,	
  as	
  well	
  as	
  censorship	
  of	
   online	
  communications	
  tools.	
  Scores	
  range	
  from	
  0	
   (no	
  censorship)	
  to	
  16	
  (pervasive	
  censorship	
  in	
  each	
   category)	
  (OpenNet	
  Initiative,	
  2011).	
   OpenNet	
  Initiative	
   Regional	
   Dummies	
   Dummy	
  variables	
  indicating	
  whether	
  the	
  country	
  is	
   in	
  East	
  Asia,	
  South	
  East	
  Asia,	
  South	
  Asia	
  or	
  Central	
   Asia.	
   	
   	
  	
   Table	
  A.7	
  –	
  Quantitative	
  Analysis:	
  All	
  Models	
  (numbers	
  on	
  right	
  referenced	
  in	
  text)	
   	
  (Demolevel)it =	
  	
   	
   β0	
  +	
  β1(IntMob)it	
  +	
  δt(Year)t	
  +	
  γt(IntMob)it	
  *(Year)t	
  +	
  uit	
   (I)	
   β0	
  +	
  β1(GNIPCPPP)it	
  +	
  δt(Year)t	
  +	
  γt(GNIPCPPP)it	
  *(Year)t	
  +uit	
   (II)	
   β0	
  +	
  β1(EduEnrol)it	
  +	
  δt(Year)t	
  +	
  γt(EduEnrol)it	
  *(Year)t	
  +uit	
   (III)	
   β0	
  +	
  β1(IntMob)it	
  +	
  β2(GNIPCPPP)it	
  +	
  β3(EduEnrol)it	
  +	
  β4(PopLog)it	
  +	
  	
   uit	
  +	
  et	
   (IV)	
   β0	
  +	
  β1(IntMob)it	
  +	
  β2(GNIPCPPP)it	
  +	
  β3(EduEnrol)it	
  +	
  β4(PopLog)it	
  +	
  	
   β5(EAsia)i	
  	
  +	
  β6(SEAsia)i	
  	
  +	
  β7(CAsia)i	
  +	
  uit	
  +	
  et	
   (V)	
   β0	
  +	
  β1(IntMob)it	
  +	
  β2(GNIPCPPP)it	
  +	
  β3(EduEnrol)it	
  +	
  β4(PopLog)it	
  +	
  	
   β5(EAsia)i	
  	
  +	
  β6(SEAsia)i	
  	
  +	
  β7(CAsia)i	
  +	
  β8(EAsia)i*(IntMob)it	
  +	
   β9(SEAsia)i*(IntMob)it	
  +	
  β10(CAsia)i*(IntMob)it	
  +	
  uit	
  +	
  et   (VI)	
   β0	
  +	
  β1(IntMob)it	
  +	
  β2(GNIPCPPP)it	
  +	
  β3(EduEnrol)it	
  +	
  β4(PopLog)it	
  +	
  	
   δi(Country)	
  +	
  uit	
  +	
  et	
   (VII)	
   β0	
  +	
  β1(IntMob)it	
  +	
  β2(GNIPCPPP)it	
  +	
  β3(EduEnrol)it	
  +	
  β4(PopLog)it	
  +	
  	
   δi(Country)	
  +	
  γt(Year)	
  +	
  uit	
  +	
  et	
   (VIII)	
   β0	
  +	
  β1(IntMob)it	
  +	
  β2(GNIPCPPP)it	
  +	
  β3(EduEnrol)it	
  +	
  β4(PopLog)it	
  +	
  	
   ρui(t-­‐1)	
  +	
  et 	
   (IX)	
   	
  (Demolevel)i =	
  	
   	
   β0	
  +	
  β1(IntMob)I	
  +	
  u	
   (X)	
   β0	
  +	
  β1(IntMob)i	
  	
  +	
  β2(GNIPCPPP)i	
  +	
  β3(Literacy)i	
  +	
  β4(PopLog)i	
  +	
  u	
   (XI)	
   β0	
  +	
  β1(IntMob)i	
  +	
  β5(CensorONI)I	
  +	
  β6(CensorONI)i*(IntMob)I	
  +	
  u	
   (XII)	
   	
   	
  110	
   β0	
  +	
  β1(IntMob)i	
  	
  +	
  β2(GNIPCPPP)i	
  +	
  β3(Literacy)i	
  +	
  β4(PopLog)i	
  +	
  +	
   β5(CensorONI)I	
  +	
  β6(CensorONI)i*(IntMob)I	
  +	
  u	
   (XIII)	
   	
  (Demolevel)it =	
  	
   β0	
  +	
  β1(IntMob)it	
  	
  +	
  β2(GNIPCPPP)it	
  +	
  β3(EduEnrol)it	
  +	
  β4(PopLog)it	
  +	
   β5(CensorDummy)i	
  +	
  β6(CensorDummy)i*(IntMob)it	
  +	
  uit	
  +	
  et	
   (XIV)	
   β0	
  +	
  β1(IntMob)it	
  +	
  β2(GNIPCPPP)it	
  +	
  β3(EduEnrol)it	
  +	
  β4(PopLog)it	
  +	
  	
   β5(CensorDummy)i	
  +	
  β6(CensorDummy)i*(IntMob)it	
  +	
  δi(Country)	
  +	
  	
   uit	
  +	
  et	
   (XV)	
   	
  	
   Table	
  A.8	
  –	
  Quantitative	
  Analysis:	
  Bivariate	
  Correlation	
   Note: This table displays the respective annual correlations between an economic measure (GNI p.c. PPP), a social measure (primary and secondary enrolment), and a measure of digital communication (ICT). The ICT measure is the average users of Internet and mobile per 100 population. 	
   Year	
   Internet/	
  Mobile	
   GNI	
  p.c.	
   PPP	
   X	
  10,000	
   Enrolment	
   X	
  10	
   1994	
   0.6115***	
   0.309	
   0.4170	
   1995	
   0.4900	
   0.616	
   0.9702	
   1996	
   0.3591	
   0.61	
   1.4730	
   1997	
   0.2318	
   0.513	
   0.5870	
   1998	
   0.1704*	
   0.437	
   1.2073	
   1999	
   0.1263**	
   0.468	
   0.8555	
   2000	
   0.0847**	
   0.447	
   0.7630	
   2001	
   0.0726**	
   0.426	
   0.6976	
   2002	
   0.0705**	
   0.588	
   0.8484	
   2003	
   0.0716**	
   0.562	
   0.7574	
   2004	
   0.0713**	
   0.622	
   0.9666	
   2005	
   0.0656**	
   0.569	
   0.6868	
   2006	
   0.0602**	
   0.525	
   0.7026	
   2007	
   0.0559**	
   0.535	
   0.8092	
   2008	
   0.0438**	
   0.476	
   0.2541	
   2009	
   0.0418**	
   0.564	
   0.3597	
   *p  <  .10  **p  <  .5  ***p  <  .01.     All  entries  are  absolute  values	
      	
  	
  	
   	
   	
   	
  111	
   	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
  	
   	
   0 20 40 60 80 10 0 0 5 10 15 Combined Freedom House Measure 2(demo) to 14(non) Internet and Mobile use per 100 pop average Fitted values Note: This graph appears to show a positive correlation between democracy level and the level of ICT penetration. But as chapter 4 investigates this view is misleading. It shows that more democratic states tend to have higher ICT levels, but this does not show how democracy level varies over time given variance in ICT level. 	
   Figure	
  A.1	
  -­‐	
  ICT	
  versus	
  Democracy	
  Level	
  in	
  Asia	
  (all	
  years	
  all	
  states)	
   	
   	
  112	
   	
   Figure	
  A.2	
  –	
  Schematic	
  of	
  Chinese	
  ICT/DSC	
  Contestation	
  (SARS	
  2003	
  &	
  Wenzhou	
  2011)	
   	
   	
   Note: Please see chapter 6 for details. In brief, this diagram represents how forces driven by societal actors empowered by ICT (DC) contest with those of the state through digital state capacity (DSC) application. This is demonstrated with net changes in transparency, accountability, and resulting regulations stemming from each case. 	
  	
  

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0073175/manifest

Comment

Related Items