Xinjiang’s Rapidly Evolving Security State

Publication: China Brief Volume: 17 Issue: 4

The Chinese government has held several shows of force in the Xinjiang Uyghur Autonomous Region (XUAR), with machine-gun toting police and armored personnel carriers parading through the streets of the region’s major cities, foreground the ongoing instability in China’s far western region (Tianshan, February 28). Since the July 5, 2009 riots in the regional capital of Urumqi, thousands have died in violent clashes between the Muslim Uyghur minority and the Han-dominated Party-state. In response, the Chinese Communist Party (CCP) has built a multi-tiered security state with, among other components, the recruitment of nearly 90,000 new police officers and a 356 percent increase in the public security budget (Foreign Affairs, December 23, 2016). According to Chinese President Xi Jinping, Xinjiang is now the “frontline” in China’s battle against “terrorism,” and consequently a testing ground for new policing and surveillance methods (Xinhua, April 28, 2014).

Using data gleaned from public service postings, it is possible to map the development of this security state in Xinjiang. Beginning in 2006, XUAR public and civil service jobs have been publically advertised on the Chinese Internet in an effort to increase transparency. These recruitment adverts contain a range of useful information, such as the number and types of positions plus specific requirements for residency, ethnicity, and education, among other details. By aggregating recruitment adverts across hundreds of regional websites and then disaggregating and analyzing the resulting data, we have compiled a unique dataset that chronicles the ballooning security footprint in Xinjiang. [1]

Viewed in aggregate, the figures reveal a massive spike in police recruitment since 2012, with Xi Jinping personally praising local constabularies as the “fists and daggers” of China’s counter-terrorism effort (Xinhua, April 28, 2014). In fact, the number of advertised police jobs in 2016 exceeded the combined figures from 2008–2012 by 30,000. In part, this is an exercise in catching up with wealthier, coastal regions, but over time, the target sharpened to focus on sources of unrest in remote, chiefly Uyghur, parts of Southern Xinjiang. Last year, the outgoing Xinjiang Party Secretary Zhang Chunxian declared success, claiming that “the situation in Xinjiang is becoming ever more stable and the number of violent incidents has declined substantially as local authorities strengthen their ability to prevent and fight terrorist activities” (Renminwang, March 9, 2015). Yet the situation remains tense, with his successor Chen Quanguo implementing an even more comprehensive and intrusive policing strategy (China Brief, February 6, 2016).

Data collected from job postings reveals four stages in the Party-state’s incremental securitization strategy in Xinjiang:

Stage 1 (2009–2011): Responding to the 2009 Urumqi Riots

Despite claims that Wang Lequan ruled Xinjiang with an “iron-fist” during his fifteen-year tenure as Xinjiang Party Secretary (1994-2010), advertised police recruitment was relatively small prior to the 2009 violence. Yet the 7.5 Incident was a wake-up call, with top Party officials deeply embarrassed by the scale of the violence in Urumqi and their inability to quickly quell the unrest. The situation only stabilized after Beijing rushed 14,000 People’s Armed Police (PAP, 武警) forces as well as the relatively recently formed Special Police Units (SPUs, 特警) from 31 provinces to Xinjiang (China Digital Times, July 10, 2009; Sina xinwen, August 17, 2009). In response, the XUAR government initiated its first-ever recruitment of SPU officers by advertising 2,655 positions in December 2009.

While SPUs existed in Xinjiang prior to 2009, their numbers were insufficient to deal with large-scale security threats. SPU officers are heavily equipped with sub-machine guns and bulletproof vests. Physical requirements for admission are also very demanding, with new recruits subject to intense mental and physical training (Guojia gongwuyuan ju, April 27, 2011; Zhongguo Gansu wang, January 18). At that time, the state was intent on investing large sums for the creation of a highly trained and heavily equipped strike force.

The second part of the state’s response to the 7.5 Incident was an investment in new personnel across the public security agencies in the XUAR, which includes the regular police force (人民警察). Total security-related recruitment, across all agencies and job types, doubled, rising from 6,876 positions in 2006–2008 to 15,841 in 2009–2011. XUAR officials evidently sought to quickly boost the region’s undermanned security personal after Zhang Chunxian succeeded Wang Lequan as Party Secretary in April 2010.

Stage 2 (2012–13): Expanded Policing and Surveillance in the Rural South

In January 2012, the new secretary of the XUAR Political and Legislative Affairs Committee Xiong Xuanguo announced the recruitment of 8,000 new police officers in order to beef-up security ahead of the 18th CPC National Congress in autumn of that year (Tianshangwang news, February 1, 2012). This intake advertised 11,559 security-related positions, a 57 percent increase in adverts over 2009 and the highest figure yet for Xinjiang. The principal focus was Southern Xinjiang, due to XUAR officials’ conviction that the violence perpetrated in Urumqi come from Uyghur migrants from the south. [2]

For the first time in 2012, XUAR officials committed themselves to fully implementing the one village, one policeman (一村一警) scheme, which had been rolled out in various Eastern provinces from the early 2000s (China Brief, September 4, 2015). Under this policy, a single police officer leads up to three assistant police staff (协警 or 辅警) in each rural village or hamlet (Tianshangwang news, February 1, 2012). The latter are a highly informal police force. Assistant police are only supposed to assist regular police officers in their duties, and (in theory) do not possess any enforcement rights (行政执法权). These positions come with lower salaries because pay levels are determined by the local authorities at the county or city district levels. The combination of lower pay, contract-based employment, and lower recruitment requirements renders the assistant police force a highly strategic component of a multi-tiered policing strategy, placing a large garrison of low-skilled security staff under smaller numbers of more highly equipped and trained police.

Most of these assistant police officers were recruited to man small, community-based police sub-stations (警务室) across the XUAR (Xinjiang Daily, December 29, 2012). [3] These comprise both fixed structures and bus-like units that can be transported with trucks, which facilitates a highly mobile form of policing. Correspondingly, 71 percent of the region’s 2012 police sub-station recruitment targeted regions with a Uyghur population share of 40 percent or higher, far more than in 2009, when only 40 percent of all security-related recruitment targeted such Uyghur regions. [4] Moreover, 78 percent of advertised police sub-station recruitment positions in 2012 were designated for rural regions, compared to only 42 percent in 2009.

The establishment of a police sub-station network was by no means a Xinjiang innovation. Rather, by the end of 2006, police sub-stations were commonplace across most urban regions of China, while more developed provinces, like Zhejiang, extended their reach into rural communities as well (Xinhuanet, February 5, 2007). Their accelerated implementation in Xinjiang was, at least partially, a response to a string of violent attacks on state targets in Uyghur regions beginning in 2012, as well as a possible factor underlying retaliator attacks.

Stage 3 (2014–2015): Grid-Style Community Policing and Big Data Surveillance

Following a series of high-profile terror attacks, including a suicide car bombing in Beijing (October 2013), train station stabbing in Kunming (March 2014) and market bombing in Urumqi (April 2014), Party officials announced a nation-wide counter-terrorism campaign, with Zhang Chunxian declaring a “people’s war on terror” in Xinjiang and Chinese president Xi Jinping calling for “walls made of copper and steel” and “nets spread from the earth to the sky” to capture these “terrorists” ((People’s Daily, May 26, 2014; Xinhua, May 29, 2014).

In response, XUAR security-related recruitment again surpassed the 10,000 position mark. The 2014 intake continued the 2012 trend away from a well-equipped, expensive policing force toward a more cost-efficient yet surveillance-intense posture, with most of the new positions based on casual employment contracts.

The 2014 recruitment drive introduced several new employment categories, signaling the regime’s efforts to extend the reach of what officials were now calling grid-style social management (社会网格化管理). Grid management employs CCTV cameras, mobile Internet technologies and big data analytics to monitor all suspicious activities within a discrete geometric zone. This approach was first trialed in Beijing and Shanghai during the early 2000s and gradually rolled out in the frontier regions of Xinjiang and Tibet after the 2008-2009 unrest (China Tibet News, November 3, 2014; Yaxinwang, January 24, 2013).

In 2014, security recruitment included, for the first time, video surveillance (视频监看) staff. Following the 7-5 Incident, XUAR authorities installed millions of new security cameras, initially in major urban areas like Urumqi but increasingly across rural and remote communities. To maximize their surveillance capabilities, dedicated police technicians were now employed. Other new recruitment categories included patrol and prevention (巡逻防抗) as well as grid patrol and prevention (网格化巡控) staff. Most of these positions were poorly paid and hired on a short-term contract basis. Unlike the previous focus on the rural south, patrol and prevention staff were recruited equally across the XUAR.

The evolution toward new surveillance-oriented, technology-focused security jobs continued in 2015 with the introduction of internet surveillance and prevention (网络监看) positions on top of the existing internet security (网络安全) job category first introduced in 2009. At the same time, 2015 also witnessed a resurgence of more formal recruitment, with 2,502 new positions for police sub-station officers and 3,478 public security or SPU officers. The 2015 intake evidently aimed to shore up staff numbers across all levels of this evolving multi-tiered security apparatus. With a total of 9,314 security-related adverts, its size remained slightly below that of the 2012 and 2014 intakes.

Stage 4 (2016-): “Convenience Police Stations” and the Massive Expansion of Surveillance Manpower

However, the largest boost in policing capabilities took place in 2016. A total of 31,687 security-related positions were advertised, more than a three-fold increase over the previous year. This unprecedented recruitment drive sought to boost the Party-state’s surveillance capabilities across all regions of Xinjiang, as only 35 percent of advertised positions were designated for regions with a Uyghur population of 40 percent or higher.

Notably, 89 percent of these new hires were associated with so-called convenience police stations (便民警务站), which are currently being built across the XUAR in the tens of thousands. Chen Quanguo had first introduced these stations in the Tibetan Autonomous Region in 2011, where media reports praised them as Lhasa’s “unquenchable lights” (Sohu news, April 3, 2015). [5] After being transferred to Xinjiang in August 2016, Chen ordered their construction across Xinjiang. In comparison to the more mobile police sub-stations, convenience police stations are (in most cases) sophisticated concrete and bulletproof installations (Xinhua, October 27, 2016). They house basic medical equipment, umbrellas, charging stations for mobile phones, and many other “convenient” community services, and are festooned with decorative elements or even ethnic colors and styles. Japan’s kōban (交番) or “police box” system is one source of inspiration for Chinese security official (Renmin Anquan, March 3, 2009), although so-called community-based policing is now an international normal.

Local media have praised these convenience police stations as “bringing zero-distance service” (让服务零距离) to the people of Xinjiang (Urumqi city government, October 28, 2016). Yet, their real purpose is surveillance, cleverly designed to make Orwellian levels of securitization more palatable, while bringing 24-hour “zero-distance” policing to an ever-increasing number of neighborhoods. During a recent inspection tour, Chen Quanguo called on these new police officers to response to any signs of trouble in under a minute’s time (Tianshan, February 21). The combination of low-skilled foot-soldiers stationed in and around convenience police stations and high-tech equipment connected to extensive information processing systems has dramatically increased the Party-state’s surveillance capabilities, providing what local media claims is “complete coverage without any chinks, blind spots, or blank spaces” (Boertala bao, November 3, 2016).

Conclusion: A State of Insecurity?

The unprecedented security-related recruitment in 2016 completes what we can tentatively call a three-tier policing strategy in Xinjiang. The top tier consists of regular police officers and the SPU police force that are recruited through the civil service and employed on stable, long-term contracts. The middle tier is composed of sub-station police officers who form localized hubs of policing, often in rural and remote regions with low levels of previous securitization. Finally, the third and lowest tier consists of various types of assistant police staff, mostly hired through informal, contract-based recruitment channels. The third-tier is designed as a cost-effective way to leverage the capabilities of the upper two tiers and provides the staffing capacity for grid-style and community-based policing. Between 2012 and 2016, 52 percent of all security-related recruitment in Xinjiang fell into this bottom tier, vastly increasing the panoptic gaze of the Party-state across China’s largest territorial unit. This surge in recruitment is continuing in 2017. In January and February alone, the region advertised just under 10,000 new police jobs, nearly all of them third tier positions (Weikouwang, February 28, 2017;, March 6, 2017).

The implications for Xinjiang and China are complex. On the one hand, this more intrusive form of policing has dramatically increased the ability of the Party-state to patrol and control anti-state activities at a local level. The result has been fewer reported terror attacks in Xinjiang from mid-2013 onward, although reporting on such incidents are frequently suppressed. Yet, on the other hand, this heavy-handed approach also creates an atmosphere of insecurity by legitimizing increasingly intrusive surveillance measures that are chiefly focused on the Uyghur minority. In recent months, Southern Xinjiang has experienced at least three violent attacks, as some Uyghurs first buckle and then lash out against this now all-encompassing state pressure (RFA, September 19, 2016; Reuters, January 5; RFA, February 15). This exacerbates the Uyghurs’ sense of cultural and physical insecurity as they feel increasingly under siege in their own homeland. By widening the gap of distrust and misinformation between the Uyghur minority and the Han-dominated Party-state, securitization directly undermines long-term social stability —ironically the very aim that it seeks to achieve.

Adrian Zenz is lecturer in social research methods at the European School of Culture and Theology, Korntal, Germany. His research focus is on China’s ethnic policy and public recruitment in Tibet and Xinjiang. He is author of “Tibetanness under Threat” and co-editor of the “Mapping Amdo” series of the Amdo Tibetan Research Network.

James Leibold is an Associate Professor in Politics and Asian Studies at La Trobe University in Melbourne Australia, and an expert on ethnic policy and ethnic conflict in contemporary China. He is the author and co-editor of four books and over twenty peer-reviewed articles and book chapters, and a frequent contributor to the international media on these topics.



  1. These figures are limited to public security agencies (公共安全) and do not include recruitment of the People’s Armed Police (人民武装警察) nor the Ministry of State Security (国家安全部). Also, recruitment for the Xinjiang Production and Construction Corps (XPCC, 兵团) is not included. Available data for both Xinjiang and other provinces, such as Qinghai and the Tibet Autonomous Region (TAR), indicate that on average, approximately 80 to 90 percent of advertised positions are actually awarded. See Fischer and Zenz (2016) for a detailed account of the methodology.
  2. Xu Ping and Wang Ping, “Investigative research into the Uyghur floating population and the 7.5 incident” [7.5事件与维吾尔族流动人口调查研究], 调查与研究Investigation and Research, 2009 (internal, restricted distribution): pp. 28–40.
  3. The informal police staff stationed in rural police sub-stations are referred to as either “police sub-station policemen” (警务室民警) or “village policemen” (村警).
  4. Of all advertised positions within the 2012 police sub-station intake, 6,738 were for regular officers and 955 for assistant police officers. The latter marked the first time that Xinjiang recruited security-related staff through temporary job contracts.
  5. Very similar installations with slightly different names had previously been innovated in other regions of China, for example “police service stations” (警务综合服务站) in Wuhan since 2012 (Sina News, August 1, 2012). However, the latter do not seem to offer such a wide range of “convenient” citizen services, and are hence more akin to the Japanese kōban (discussed below).