No Way Forward, No Way Back:

China in the Era of Riots


Denim and its Discontents

 The story is now familiar: One morning in the spring of 2011, a migrant street vendor is harassed and beaten by police. That evening, rumors fly over the internet that the vendor has died. Hundreds of people gather in the streets, enraged by the apparent murder. They burn cars, loot ATMs and attack the riot police sent to disperse them. But they do not disperse. The riot spreads over several days, with participants growing into the thousands. Journalists who come to report on the events are held by security forces. Rumor of the uprising spreads over the internet even as the government uses all its resources to cut off access to the information.

Despite its striking similarity, this is not the story of Mohamed Bouazizi, the Tunisian street vendor, harassed by police, whose self-immolation sparked the Arab Spring. The man in the story above was instead Tang Xuecai (唐学才) a Sichuanese migrant in the city of Guangzhou. The riot[1] took place in Xintang, one of the Pearl River Delta’s many manufacturing districts, this one specializing in denim[2], with the majority of the rioters themselves migrant laborers in factories making jeans for export. And, unlike the riots and strikes that followed the death of Bouazizi in Tunisia, the Xintang riot was ultimately crushed as police took control of the district, made mass arrests, and forced the majority of migrants back to work

Aside from this stark comparison, there was nothing particularly special about the Xintang riot. In a strictly quantitative sense, cities like Guangzhou, Shenzhen and Dongguan in China’s Pearl River Delta (PRD) see more riots more regularly than even Athens. If one adds strikes, blockades and other such “mass incidents” to the list, Chinese protests regularly surpass global trends in scale and severity—especially since a lack (or exhaustion) of legal alternatives tends to transform what might be a benign picket or protest elsewhere into a multi-factory uprising that risks destroying millions of dollars of equipment. Yet we do not often see the avenues and alleys of Xintang as we see those of Athens, lined with burning cars as riot police advance and swarms of rioters scatter underneath the dim gold glow of a McDonalds sign. Instead, images of Athens burning are posed against the glowing skylines of China’s coastal cities, intercut with upward-trending graphs of productivity, profitability, progress.

Underneath the graphs, however, such “mass incidents” have been increasing over the last decade.[3] This rising unrest is, in fact, recognized by numerous official sources, such as the Annual Report on China’s Rule of Law (No 12). Other than attempting to tally and taxonomize the “incidents,” this report also noted that roughly 30% of them took place in Guangdong province, in which the PRD is located.[4] But many such reports, including this one, take only a small number of mass incidents reported in major media outlets and generalize from this subset. Others, such as the China Labor Bulletin’s strike map, mine reports from the Chinese internet in a much more systematic way, but the data stretches back only a few years.[5] Their map is also intentionally focused on strikes, rather than all “mass incidents,” and therefore often excludes forms of unrest that are initiated outside the workplace and do not take the form of labor grievances.

Discussions of Chinese unrest too often rely on partial sources or intuitively “obvious” trends, often paired with restrictive definitions. In order to discuss these phenomena, however, it is essential to expand the scale of our data. In recent years an unprecedented news-aggregator database, Global Data on Events, Language and Tone (GDELT), has become available, giving access to an enormous portion of the world’s news reports, in over 100 languages (using the Chinese state’s press agency Xinhua as one of its primary news pools) and coded for different types of news “events,” mostly diplomatic in nature but also including a variety of records for internal political strife.[6] This provides an alternative to the data gathered in official reports or mined from Chinese social media—not so much a replacement of these sources as a comparative supplement. Though not necessarily more authoritative or accurate in the details, it can provide a longitudinal context that the others cannot.[7] Querying data on riots using GDELT has shown a slight increase in riots worldwide since 1979, made more significant by a concurrent, and much more severe, global decrease in strikes.[8] Using the GDELT data, we are now able to see certain comparative patterns invisible in previous reports. Nonetheless, the GDELT data are also based on news reports, and therefore almost certainly underestimate the number of mass incidents in a country like China, with its regular media censorship.

Figure 1: Riots as a percent of total reported events (normalized by global total).

Figure 1: Riots as a percent of total reported events (normalized by global total).


Figure 2: Strikes as a percent of total reported events (normalized by global total)

Figure 2: Strikes as a percent of total reported events (normalized by global total)




Figure 3: Riots in China as a percent of total events (normalized by national total), compared to same data for world. This chart from Neel 2014.

Figure 3: Riots in China as a percent of total events (normalized by national total), compared to same data for world. This chart from Neel 2014.


Figure 4: Strikes in China as a percent of total reported events (normalized by national total), compared to same data for world. From Neel 2014

Figure 4: Strikes in China as a percent of total reported events (normalized by national total), compared to same data for world. From Neel 2014


The general trends in riots and strikes within China are visible in figures 1 and 2.[9] Both strikes and riots have spiked in recent years, though unevenly, and they are either outstripping (riots) or running counter to (strikes) the world trends, compared in figures 3 and 4.[10] It’s also clear that riots are far more common than strikes, as can be seen by comparing the event units on the y axes. Despite the fact that both are increasing, 2009 saw almost eight times as many riots as strikes. In 2012 the margin had narrowed only slightly, with five riots for every one strike, and by 2014 riots had dropped to a third of their 2009 magnitude while strikes peaked again, though remaining slightly lower than their 2010 high. It is also notable that, whereas riots in China surpass world averages, strikes (both reported and, likely, de facto) still seem to fall below the world average.

Despite these trends, uprisings in China have been substantially contained. There has been no millennial Tiananmen, and attempts to organize beyond a single factory or neighborhood have thus far been incapable of surviving in any substantial form. Maybe more importantly: riots and strikes in China are most often explicitly revindicative in nature—meaning that they often seem to make very specific, local demands of existing powers. Many such “incidents” thus operate well within currently accepted power structures and tend toward negotiation, particularly when demands take the form of appeals to the central government to ouster “corrupt” local officials, despite the fact that those local officials are often simply responding to material incentives designed by the central government itself.

This is a clear divergence from the tendency seen in Egypt, Greece, Spain or even the Occupy movement in the US, where “mass incidents” have increasingly abandoned their own demands for simple reforms or payoffs—such that even when they do formulate higher-order demands (“down with Mubarak,” “No to the austerity vote,” “everyone out of office”) these demands increasingly exceed accepted frameworks of power. This is not to say that these movements have become “demandless”—they usually articulate a variety of demands and exist alongside more traditional campaigns, of which SYRIZA and Podemos are today the foremost examples. It is more accurate to say that established frameworks for articulating demands and having them met have begun to break down in places like Greece and Spain. The severity of the crisis in these countries makes identifying possible reforms difficult and ensures that attempts to fulfill even minor demands meet herculean obstacles. In Greece, a simple “no” vote to austerity measures threatens the collapse of the Eurozone. In China, however, protestors’ demands have often been fulfilled quickly and with little fanfare.

In one sense, these mass incidents are simply the most recent—if more invigorated—oscillation of the “holding pattern” within which contemporary struggles remain muted. At the same time, it may also be the beginning of a return to conditions somewhat similar to those that gave birth to the revolutionary movements of the late nineteenth century in their earliest forms—a return that communist philosopher Alain Badiou has called the “rebirth of history,” and one that other contemporary communists refer to as the “era of riots.” As explained by Jason Smith:

Le temps des émeutes: this was the expression used in France after 1848 to refer to the early years of the workers’ movement, the two decades preceding the sudden eruption of revolt across Europe that year. This period was marked on one hand by a certain disconnection between the proliferation of socialist and utopian sects, with their alternately arcane or lucid schemes for treating the emergent so-called “social question,” and on the other by the immediate needs of workers themselves in their often violent responses to transformations of the production process occurring at the time.

[…]Over the past five or six years, probably beginning with the banlieue riots in France in November 2005 up to the London riots of August 2011, from the anti-CPE struggles in France in 2006 to the recent “movement of the squares,” from the anti-austerity general strikes in Greece over the past two years to the astonishing revolts in North Africa last year, we are awakening from the neoliberal dream of global progress and prosperity: after forty years of reaction, after four decades of defeat, we have re-entered the uncertain stream of history. We bear witness to a new cycle of struggles; ours is a time of riots.[11]


Certain facts are written on the surface of events. From Guangzhou to Cairo, it’s clear that something is awakening. But why do riots in China seem to take on such a different character than those seen elsewhere? According to the Anglophone communist collective Endnotes, global struggles are caught in a sort of “holding pattern” in which they are incapable of developing beyond the riot stage. One potential path out of the current “holding pattern” is “an intensification of the crisis, a global bottoming out, beginning with a deep downturn in India or China.”[12] If this is the case, these very dynamics may be key to understanding when and how this holding pattern might be broken. The most relevant question might then be: why have these strikes and riots, despite their size and frequency, been unable to pose a serious threat to power? Were conflicts in China simply a prelude to the Arab Spring? Or do they prefigure something larger still to come?


Field to Factory

In the most general sense, Chinese economic development since the end of the 1970s has been marked by two major class dynamics. Taken together, these dynamics signal China’s transition out of a chaotic, inconsistent socialism—which designates the absence of a coherent mode of production—and into global capitalism. The first of these two dynamics was the solidification of a “bureaucratic capitalist class,” beginning in earnest with the allegiance forged between the “red” (political) and “expert” (technical) elites in the reaction to the Cultural Revolution in the late 1970s.[13] Over the following decade, this allegiance would become an entrenched feature of the Chinese class hierarchy:

[…] many cadres and their kin and associates managed to amass enormous private wealth and turned themselves into the first generation of China’s cadre capitalist class, or bureaucratic capitalists, in a matter of a few years. Inflation, corruption, and class polarization reached a state of crisis in 1988, paving the way for the large-scale unrest in 1989.[14]

The events of 1989, however, were only the beginning of what soon became a tendency toward more or less continual unrest spanning demographics and emerging in almost every geographic niche. If anything, Tiananmen itself was the true inauguration of the restructured ruling class, through which the final resistant segments of the intelligentsia—the students themselves—were ultimately incorporated into the party.[15]

But while many of the rebellious students were offered lucrative careers, the workers were simply left to the tanks. Tiananmen, then, also inaugurates the second major dynamic of the period: the remaking of China’s working class in a process of “proletarianization”—i.e., the production of a population that has no ownership over means of production such as factories or large tracts of land, and who must therefore depend upon a monetary wage (often second or third-hand) in order to subsist on goods purchased through the market. In China, this process involved not only the gutting of the old state-owned industrial strongholds in the country’s rustbelt and the dissolution of the socialist-era working class, but also the birth of new industrial and consumer cores in the port cities of the sunbelt, staffed by a new generation of workers. [16] A significant segment of this new working class is made up of rural migrant laborers (农民工, nongmingong) who either can’t access or must pay more for state resources (education, unemployment insurance, etc.) in the areas where they work, becoming instead dependent on their employers’ state-mandated (but often un- or under-paid) contributions to insurance accounts. This process has sent ripple effects into other strata of Chinese society, as industrialization has driven urbanization and environmental degradation, leading to protests against land dispossession, overuse of natural resources and industrial pollution, all alongside skyrocketing labor unrest.

Despite the more or less complete industrial restructuring of the country, both legal labor disputes and extra-legal measures are still on the rise. According to a report on the 2010 strike wave,

The nation’s courts dealt with nearly 170,000 labor disputes in the first half of 2009, an increase of 30% over the same period the previous year, [a] survey revealed, without specifying how many of these disputes related to migrant workers and their employers.


In 2007, China had over 80,000 “mass incidents” (the official term for any collective protest or disturbance), up from over 60,000 in 2006, according to the Chinese Academy of Social Sciences, although many involved no more than dozens of participants protesting against local officials over complaints about corruption, abuse of power, pollution or poor wages.

[…]Strikes and protests at factories are becoming more common. Outlook Weekly, an official magazine, reported in December that labour disputes in Guangdong in the first quarter of 2009 had risen by nearly 42% over the same period in 2008. In Zhejiang, a province further north, the annualized increase was almost 160%.[17]

And since 2010, labor actions have taken a qualitative turn away from the simple “protests against discrimination” common among earlier generations of migrant workers:

Since [2010], there has been a change in the character of worker resistance, a development noted by many analysts. Most importantly, worker demands have become offensive. Workers have been asking for wage increases above and beyond those to which they are legally entitled, and in many strikes they have begun to demand that they elect their own union representatives. They have not called for independent unions outside of the official All-China Federation of Trade Unions (ACFTU), as this would surely incite violent state repression. But the insistence on elections represents the germination of political demands, even if the demand is only organized at the company level.[18]

The country’s continuing economic slowdown has since seen a turn back to defensive demands, but, again, the nature of the demands is not really the issue. More important is the continuing increase in both frequency of mass incidents and in the numbers of workers participating—with what may be China’s largest strike in modern history occurring in 2014, when forty thousand workers walked out of the Yue Yuen shoe factory in Dongguan.

Most of the workers engaging in these strikes are second or third generation migrants, as noted in the report on the 2010 strike wave: “the majority of Honda Lock’s employees are single women in their late teens or early 20s.”[19] And it is among these later generations of migrant workers that we see the greatest evidence of similarities with the class dynamics producing such strikes and riots elsewhere. These workers were born or raised in the reform period, entering the labor market in the late 1990s and early 2000s. In a series of the most extensive ethnographies available, sociologists Lu Huilin and Pun Ngai describe the general character of this second generation:

Although both the class structure and the process of an incomplete proletarianization of the new generation of dagongmei/zai [i.e., migrant workers] are similar to those of the previous generation, there are new life expectations and dispositions, new nuanced meanings of work, and heightened collective labor actions among those subjects who had grown up in the reform period. […] Characteristic of the second generation’s way of life is a greater disposition towards individualism, an increased proclivity for urban consumer culture, less constrained economic circumstances and greater pursuit of personal development and freedom, a higher rate of job turnover and less loyalty to their workplace. The second generation, born and raised in the reform period, is relatively better educated and better off materially but spiritually disoriented while having a cosmopolitan outlook.[20]

This generational aspect is key, argue Lu and Pun, since it is among the second and third generations that the proletarianization process “usually takes root.”[21] This process itself is defined by the authors in relatively simple terms, taken to characterize agricultural laborers “who come to work in industrial cities,”[22] effectively marking the transition from direct to indirect subsistence, now mediated by the wage:

This is the process of proletarianization, which turns agricultural laborers into industrial workers by depriving the former of their means of production and subsistence […] As a result, workers’ fate depends on the process of capital accumulation and the extent of the commodification of labor use. These workers neither own nor control the tools they use, the raw materials they process, or the products they produce.[23]

The authors go on to argue that, in the case of China, this process is made “peculiar” since “industrialization and urbanization are still two highly disconnected processes, as many peasant-workers have been deprived of the opportunity to live where they work.”[24] This has not only hindered the assimilation of workers into the urban sphere, but in fact created an internal separation integral to the Chinese model of accumulation—a separation that explicitly attempts to divide the process of reproduction of labor from commodity production and effectively externalize it. This creates “a spatial separation of production in urban areas and reproduction in the countryside.”[25] But, whereas Lu and Pun see this as a feature unique to China, the process bears significant similarity to basically every sequence of proletarianization witnessed over two centuries of capitalist history.[26]


The Most Recent Crisis

Already by 2007, the Chinese economy faced a slowdown due to rising costs of labor, fuel and materials, as well as currency appreciation and the implementation of new labor laws.[27] When the crisis hit, export centers such as the PRD saw an immense fall in GDP—with Guangdong alone plummeting from among the top GDP earners to the last-ranked in the seventeen provincial-level units with available information. This was accompanied by mass layoffs, wage arrears and factory closures. By the end of 2008, more than 62,000 factories in the province had shut their doors, with 50,000 of these closures occurring in the final quarter of the year, concurrent with the first stage of the global crisis.[28]

Unemployment among migrant workers reached record levels: “total unemployment for rural migrant labourers in early 2009 is estimated to have been 23 million, about 16.4% [of the total migrant labor force…] This rate of unemployment was disastrously high compared to previous years, as, quite contrary to common belief, rural migrant labour had had a very low unemployment rate (1-2%).”[29] Labor disputes surged in this period, but not as much as one might expect, given the severity of the crisis and the fact that certain regions, such as the PRD, suffered disproportionately from the downturn. The 23 million workers laid off at this time, though only 16.4% of the total nongmingong labor force in the country, would have been disproportionately concentrated in certain cities—leading to even higher unemployment in these areas. [30]

Comparing the situation of migrant labor in China to that of post-crisis Greece helps to put the severity of the downturn into perspective. Over five years, total unemployment in Greece climbed from a low of 7.3% (directly prior to the crisis) to a high of 27.7%, plateauing in 2013—a tripling of the total unemployed.[31] Similar patterns are observable for Spain,[32] and in both countries this immiseration, combined with a sovereign debt crisis, ultimately produced an explosion of popular unrest. The short-term increase of unemployment in China’s migrant-heavy regions, however, far outpaced that observed in either Greece or Spain, even if total unemployment never reached upwards of twenty percent. From a norm of 1-2% nongmingong unemployment, the country as a whole jumped to 16.4% in the space of six months (from late 2008 to early 2009). Even ignoring the certainty that real unemployment among migrant workers was actually higher in migrant-heavy cities such as Dongguan and Shenzhen,[33] this represents more than a tenfold increase (1093%) in unemployment—and not over five years, as in Greece and Spain, but instead over five months.

Why did such a massive spike in unemployment, concentrated in a few core cities, not create the kind of popular threat to the existing order that accompanied the tripling of unemployment over five years in Greece and Spain? The first answer—and the one most avidly promoted by the CCP, as well as its Keynesian admirers in the west—is the Chinese state’s rapid and extensive fiscal response to the crisis. Not only was China, unlike much of Southern Europe, not on the brink of a sovereign debt crisis, it was actually an integral market for the debt of deficit-dependent countries elsewhere. Meanwhile, as the US congress was mired in bickering over whether or not to even provide government bailouts to the banks, the CCP rushed through a US $586 billion stimulus package that targeted public works, largely in China’s poorer inland provinces.[34] This quickly created millions of jobs for the rural migrants who had been ejected from the labor market in the first part of the year.[35]

These new jobs, located closer to migrants’ legally registered places of residence, also helped to secure a geographic fix that had already begun to ease the spike in unemployment. In normal years, the Spring Festival, occurring in January or February, is a time when migrants return home en masse. The phenomenon is called the “spring movement” or chunyun (春运), the largest recurring migration in the world. In the crisis year, however, the spring migration began more than three months prior to the Spring Festival itself, in late October 2008, when small but significant numbers of migrants started trickling home. Migration increased as the crisis hit the industrial cores, with up to 50% of workers returning to their home villages, compared to the norm of 40%. More importantly, a large portion of the returnees then stayed in their home villages for longer than usual (about 14 million of the total 70 million returnees, or 20% of the 50% who returned).[36] Many did not intend to come back to the cities they had left, as Chan notes:

Many migrants took home their appliances (such as TV sets), believing that they would not have the opportunity to come back to find a new job after the Spring Festival. What is even more indicative of the severity of the situation—and also ironic—is that in Dongguan, for instance, hundreds of workers lined up for hours to close their social security insurance accounts (mainly for pensions), their supposed bulwark against poverty and destitution. Migrants chose to cancel their accounts to cash in every last bit of money, as they had very little hope of coming back to the town.[37]

Even when employment was restored in part through the stimulus, the new projects (alongside newly funded industrial zones) were largely located in the interior provinces, solidifying the internal spatial fix. By comparison, cities such as Dongguan would see an emptying-out reminiscent of places like Detroit, with population density dropping and industry simultaneously mechanizing and fleeing to cheaper or more skilled locales, such as far-off Chongqing or neighboring Shenzhen, respectively.[38]

We thus see that the supposedly “incomplete” part of Lu and Pun’s “incomplete proletarianization” actually helps to ensure the easy administration of workers no longer needed for production by (temporarily) externalizing their own reproduction costs to the countryside and allowing for the de facto deportation of unemployed workers. Here the old socialist practice of rustification has been recuperated and marketized, with the rural interior used as a sop for the surplus labor that would otherwise be indigestible during a period of general crisis. But such a strategy (despite its well-tooled administrative character) hardly hints at a fundamentally different and basically “incomplete” style of proletarianization. In fact, the externalization of reproduction is both an historical facet of every proletarianization process as well as an increasingly necessary procedure for global capitalism after the crises of the late 20th century.

It is also, ultimately, a method for further industrialization and urbanization of China’s underdeveloped interior. The countryside has now been marketized and hollowed out to such an extent that staying in the village is untenable. During the crisis, the village instead became a temporary stop-off on the way to new employment in nearby cities. In future crises, even this may not be an option, as workers have already relocated closer to their hometowns, which themselves have now largely been pieced apart and sold off to real estate developers or large agricultural conglomerates. As reproduction becomes more troublesome, these external spaces for non-market subsistence grow more sparse.

Historically, proletarianization was always partially incomplete. The term itself designates a transition, by definition spanning both worlds of the “new working class” and those being siphoned into it. The incomplete character of the process has always taken on both racial and gendered characteristics, with the work of immigrants, black people, the colonized, the indigenous and women all deemed to be of less value than the “normal” work of those who were formally acknowledged as wage laborers, and also less likely to be remunerated with a wage at all. [39] Even where more explicit racial, national or gender divides may not exist, the same “incomplete” characteristics are produced by the uneven character of industrialization—as can be observed with the “Okies” in 1930s California or the southern “Terroni” working in the factories of northern Italy in the 1950s.

In every locale, as the proletarianizaton process was initiated, the reproduction of these new workers’ labor-power was externalized, with wages often too low or inconsistent to fully accommodate basic expenses, requiring complex networks of unpaid care work, foraging, squatting and other informal economies making up the difference—usually at the double expense of workers who were also women. When these “incomplete” proletarians became too troublesome, a wide array of responses were available depending on the situation, spanning from extermination to deportation[40] or assimilation. This is not an accidental side effect of proletarianization, but rather a necessarily disavowed component of the process:

This [process of racialization] was the flipside of what Marxists call “proletarianization”. Marked by ongoing histories of exclusion from the wage and violent subjugation to varieties of “unfree labor”, racialised populations were inserted into early capitalism in ways that continue to define contemporary surplus populations.[41]

The “flipside,” then, was always a constitutive element of proletarianization itself. The phenomenon that Lu and Pun describe in large part simply mimics these earlier processes of racialization—with one important exception.


Cracks in the Glass Floor

The determining difference today is the fact that capitalism is undergoing a general crisis of reproduction at an unprecedented global scale. This means that, underlying periodic financial crises or political upheavals, we can observe a secular tendency in which capital becomes increasingly difficult to reproduce through investment at a profitable rate and, at the same time, it becomes difficult to reproduce proletarians as productive workers (people working within the immediate process of production, where capital is combined with human labor to produce goods of added value). This results not only in financial crises and unemployment spikes caused by speculation and over-investment—when “safer” outlets of profitable investment cannot give adequate rates of return—but also in a general mechanization of production, such that the percentage of the population required to produce a given quantity of goods and extract a given quantity of natural resources diminishes over time. When workers lose their usefulness for the system (i.e., when individual workers cease to be important to the production of value), they are expelled into what Marx called the “surplus population.”

Much of this expulsion is currently accommodated by the rise of service industries, the majority of which are not directly productive of new value for the system as a whole (they can, of course, be profitable regardless). In some places—the “global cities” in particular—lucrative positions in the international division of labor accommodate the existence of many high-paying service jobs alongside vast state-funded, semi-speculative complexes of welfare and middle-income service work, most visible in the education, healthcare and “non-profit” industries. Some of this ultimately facilitates the creation of new value, helping producers to manage the bureaucratic complexity of the global market. But this “financialized” complexity is itself symptomatic of the secular crisis.

Such services, then, need to be understood not as magically productive (i.e., as cognitive/immaterial labor, a la Hardt and Negri or the marginal utility theorists), but as the baroque excess of a vast global wealth trapped in an incestuous stasis. As the rate of profitable returns shrinks in the value-producing industries, even these well-off economies find themselves constrained, such that unemployment rises and better-paying jobs in services, transportation or high-value-added manufacturing are replaced, if at all, by low-paying service work. The lavish becomes austere, beginning at the edges. This brings these economies closer to the global norm, in which most service work is somewhat informal, is often combined with various types of debt bondage (including outright slavery), and ultimately pays very little.

This means that the number of surplus proletarians is not simply increasing in absolute terms (though it may be doing this was well), but also expanding in general, meaning that traditional characteristics associated with the surplus population (informality, precarity, illegality) have again become relatively “normal” characteristics of the laboring population as a whole.

As the Greek communist group Blaumachen writes:

The crucial matter is not the production of a quantitative increase of the lumpen proletariat, but that of an increased lumpenisation of the proletariat—a lumpenisation that does not appear as external in relation to waged labour but as its defining element.[42]

This creates a “(non-)subject” at the heart of contemporary political unrest, rather than the traditional “revolutionary subject” of the leftist mythos, centered on workers, peasants, the lumpen, the colonized or some coalition of the above. This oscillating (non-)subject is defined by its “relation between integration and exclusion from the process of value production.”[43] And this ambiguous relation is the core class dynamic of capitalism, becoming more and more visible as the crisis of reproduction deepens.

This is not to say that the “relation between integration and exclusion” is unique to our era (as Blaumachen and others sometimes seem to imply). Such a tension has always marked the historical process of proletarianization, which has seen proletarians forced to combat one another along lines of ethnicity, geography, gender, etc., in order to secure themselves within the realm of the “included” via access to the wage—as well as formal recognition of this inclusion through citizenship, access to education, mortgages and other forms of credit. Similarly, the proletariat has seen relative “lumpenisations” before, through colonization as well as the simple immiseration of migrant workers from the countryside in the early stages of Europe’s industrialization. What has changed, then, is not so much the relations themselves (the relation between capital and labor, and between inclusion and exclusion), but the global context in which these integral antagonisms are playing out.

Formerly, colonized subjects and migrants staffing industrial zones still retained a substantial connection to histories that stretched (often within the space of a single generation) beyond the gambit of the geographically small capitalist economic core. This early capitalism was, moreover, surrounded by a diverse array of alternative modes of production. Some were undergoing their own crises, others were already partially or catastrophically tilted toward the gravity of the capitalist mass growing atop Western Europe, and still others lay wholly untouched by “the economy.” The new working classes frequently drew on folk histories of struggles waged, however incoherently, against the dispossessions and enclosures that had led to inclusion in the wage relation in the first place. Rather than being simple “programmatist” affirmations of workers’ own identity, all the early workers’ movements incorporated elements of these peasant or indigenous histories and folk traditions—and the vast majority of the insurrections and revolutions of the 18th through 20th centuries were staffed directly by peasants or a generation once-removed from rural life.

Today, however, room for growth is scarce, the rural labor pool is shrinking and the industrial workforce is dwindling due to automation. These limits are most visible in the dire state of the planet’s non-human systems, but this is only one face of a crisis in which the basic reproduction of capital, labor and the relation between the two is becoming a problem in and of itself. Since the 1970s, “capital has been trying to free itself from maintaining the level of reproduction of the proletariat as labour power.”[44] This reproduction appears as “a mere cost” in global capital’s race to the bottom:

At the very core of restructured capitalism lies the disconnection of proletarian reproduction from the valorization of capital—within a dialectic of immediate integration (real subsumption) and disintegration of the circuits of capital and the proletariat—and the precarisation of this reproduction, which against the background of the rising organic composition of social capital and the global real subsumption of society to capital, has made the production of superfluous labour power an intrinsic element of the wage relation in this period.[45]

In this context, then, the Chinese case appears remarkable only to the extent that the state has been able to accommodate and administer this “intrinsic element.”

Yet the differences remain salient. Endnotes argues that, in a generalized crisis of reproduction,

the old projects of a programmatic workers’ movement become obsolete: their world was one of an expanding industrial workforce in which the wage appeared as the fundamental link in the chain of social reproduction, at the center of the double moulinet where capital and proletariat meet, and in which a certain mutuality of wage demands—an “if you want this of me, I demand this of you”—could dominate the horizon of class struggle. But with the growth of surplus populations, this very mutuality is put into question, and the wage form is thereby decentered as a locus of contestation.[46]

In an apparent contradiction to this thesis, however, wage demands have been precisely where the recent Chinese riots, strikes and blockades have tended to center themselves. And these demands have not only been won in marginal cases, but have in fact led to a general rise in manufacturing wages over the last decade, to the point that the stability of the “China price” is now in question.[47] The absolute number of manufacturing workers has also been increasing in the same period, rising from 85.9 million in 2002 to 99 million in 2009, which was an increase from 11% of the total labor force to nearly 12.8% within seven years.[48]

This is all the more significant when considering the ways that such a crisis of reproduction and the decentering of the wage that comes with it ultimately limit the possibilities for proletarians to attack the conditions that structure their own lives. The French group Théorie Communiste have argued that one of the most pressing limits of the 2008 riots in Greece was the rioters’ inability to break the “glass floor” between the reproductive and productive spheres:

But if the class struggle remains a movement at the level of reproduction, it will not have integrated in itself its own raison d’être: production. It is currently the recurrent limit of all the riots and “insurrections”, what defines them as “minority” events. The revolution will have to go into the sphere of production in order to abolish it as a specific moment of human relations and by doing so abolish labour by abolishing wage-labour. It is here the decisive role of productive labour and of those who, at a given moment, are the direct bearers of its contradiction, because they experience it in their existence for capital that is at the same time necessary and superfluous.[49]

In contrast to Greece, however, Chinese riots take place in extremely close proximity to, if not directly within, the “sphere of production,” with many literally starting in the factory cities themselves—spreading from shop floors to dormitories and canteens, and thereby jumping from workers at one factory to workers at the next.

This hints, then, that the inconsistency between the observed phenomena and the theories outlined above may simply be the product of different points of focus. Blaumachen, Endnotes and Théorie Communiste take Europe and North America as their starting point. But the particular way that the crisis of reproduction manifests itself in China will be distinct from that observed elsewhere. Théorie Communiste’s own diagnosis of the limits encountered by the Greek rioters signals the difference: whereas the Greeks encounter the hard limits of a “thick” glass floor separating them from the productive sphere, the position of regions such as the PRD within the global division of labor makes for an extremely thin “glass floor,” which requires increasing maintenance as cracks proliferate. The primary strategy for managing these conflicts, as noted above, is precisely to separate the volatile segments of the population (namely the unemployed) from the productive zone. Alongside costly state stimulus programs, the buffering of production becomes an absolute necessity, whether through quasi-deportation to newly-industrializing cities or through the transformation of the productive zone itself into a hub of total social control in the factory city—simultaneously workshop, leisure space and prison.[50]

This means that this central intrinsic limit of the given period of struggle is increasingly forced to manifest itself as an external constraint on proletarians, at least in China’s manufacturing regions. As an external constraint, it becomes embodied not only by the police (as elsewhere), but also by the individual’s very surroundings in the constructed environment—the new infrastructure constructed by the stimulus, the new urban doomscapes being inaugurated in the western interior, or the purposefully designed factory cities. This constraint, then, becomes manifest in an increasingly desperate, makeshift and intense mechanism to buffer the productive sphere by partially decoupling it from the reproductive (with the state, the family, or the criminal syndicate taking on the burden instead), while still facilitating their integration in the immediate process of production, by force if necessary.[51]


No Future

Our intuitive image of China as the “world’s factory” also tends to obscure actual trends in the composition of employment. When the data is examined in more detail, the Chinese economy as a whole appears to follow the same pattern of deindustrialization and informalization seen worldwide. While it is true, for example, that Chinese manufacturing absorbed ten million new workers between 2002 and 2009, accounting for an additional two percent of the total labor force, this growth occurred as a late spike following the massive deindustrialization of the country’s rust belt and the dismantling of the “iron rice bowl” in the 1990s. In absolute terms, Chinese manufacturing as a percentage of total employment has decreased sharply since the beginning of the reform era, falling from a high of 14.8% in 1985 to 11% in 2001, and only recently rising back to 12.8% in 2009. The net trend has clearly been downward (see Figure 5). This is despite the fact that agricultural employment has been on an historic decline as well, dropping from 63% in 1985 to 35% in 2011.[52] This means that in China, as elsewhere, the service sector has seen a net increase, and industries such as construction have become increasingly dependent upon state stimulus and financial speculation, rather than the expansion of industrial plants.


Figure 5: Manufacturing as a Percent of Total Employment in China

Figure 5: Manufacturing as a Percent of Total Employment in China


In addition to these trends, the character of Chinese manufacturing often goes unmentioned. It is frequently assumed that the immense factory complexes that hosted the protests at Foxconn or Honda are the norm, with their large size and quasi-Fordist models of labor discipline, but this is not the case. Most workers employed in Chinese manufacturing (64.4%) are actually employed in “Rural or Township and Village Enterprises (TVEs),” which are largely located outside the mega-urban cores, are often poorly counted by official Chinese statistics and include “workers outside of established enterprises who were self-employed or worked in household, neighborhood, or other small manufacturing groupings.”[53] Even the largest industrial hubs are dependent upon this mesh of small, networked and highly informal labor, which is most visible in industries such as recycling, small parts manufacturing, and labor-intensive resource extraction—the products of which then feed into larger industrial agglomerations such as Foxconn to be ultimately processed (alongside inputs produced in similarly informal settings in places such as India) into consumer products. There is also obviously a blurring of lines between manufacturing and service work here, as many industries also engage in transport, labor brokerage and local financing—whether through personal or family-network lending, or through the formation of increasingly large “shadow banks” that exist parallel to the formal banking system.[54]

The changing character of the country’s industrial structure has immense influence on the way that riots, strikes and other “mass incidents” are sparked, and how they are ultimately constrained. This structure also exerts a sort of gravitational pull on subjectivity that helps to shape the ways proletarians understand their own actions relative to the world around them. As such a strong factor in the formation of peoples’ everyday surroundings (including the rhythms of their activities and contact with others) work and the environments it creates are the terrain on which revolts operate, and against which they react. As a sequence of struggles evolves and adapts, this terrain is collectively (often intuitively) mapped, and as the conflict gains intensity there is an increasing awareness of the capacity to not only seize but also reshape these surroundings. In the past, large-scale industrial conglomerations in places like Detroit or northern Italy became hotbeds of traditional “workers’ movements” precisely because they concentrated enormous numbers of industrial workers in a few urban zones, these workers laboring alongside each other in enormous facilities and industrial districts housing thousands.

Given the prevalence of strikes in recent Chinese unrest and the simple proximity of these strikers to some of the world’s central factory zones, it is often assumed that the limits of present struggles in China will be overcome by a new union movement of some sort—one that is highly networked, autonomous from the government unions, and directly democratic. Though aided by the most recent digital technology (“autonomy + the internet” has become a sort of catch-all equation for the left in the past twenty years), this movement is conceived as more or less the rebirth of syndicalism in China—as exemplified by a recent report by the China Labor Bulletin, titled “Searching for the Union: The workers’ movement in China 2011-13.” The presumption here is clear. We already know the method of overcoming the current deadlock of the struggles in China: the union. This organizational model simply needs to be “found” by the “workers’ movement.” Rather than understanding organization to be the confrontation and overcoming of limits in a given sequence of conflicts, such an approach is purely formalistic.

A more reasonable starting point should be the opposite. There is no reason to assume that a “workers’ movement” exists in the traditional sense simply due to the agglomeration of strikes, nor that “the union” is the organizational form that guarantees a method of overcoming the failures of these struggles simply because it has (allegedly) played this role in history. These may have been the conditions that most readily birthed the “mass worker” in the West, but these conditions are largely absent in a deindustrializing China, just as they are absent in the deindustrialized US and EU of today.

It would be a mistake, however, to assume that the de facto absence of conditions for a “workers’ movement” means the doom of any attempt to overturn the present system. This assumption arbitrarily selects one potential high point from the US and Europe out of many in a diverse history of struggles against capitalism, generalizing this point into an absolute condition for a new cycle of revolt to unfold. In fact, the opposite may be true. It was precisely under the large, Fordist-style factory regimes that the existence of a healthy “workers’ movement,” whether promoted by socialist parties (as in Europe) or Great Society liberals (as in the US), extinguished the last embers of revolution still burning from the insurrections of the hundred years prior. Meanwhile, the syndicates, communist parties and revolutionary armies of that last century were hardly the product of “revolutionary consciousness” being engendered in a largely industrial workforce through the collectivizing processes of capitalism itself. Obviously, industrialization and the demographic transition played an integral role in sparking revolts against this immiseration. But these earlier insurrectionary movements that emerged out of it were just as much the artifact of peasant and indigenous traditions of resistance to capitalism from outside, as well as simple contingencies of culture, history and tactical accident.

Though the prominence of wage and workplace demands in Chinese unrest seems at first glance to signal the rise of a new workers’ movement, something very different is going on beneath the surface. Similar tactics, placed in different circumstances, can signal very different political potentials. Despite the prominence of wage demands in Chinese strikes (whether for raises or arrears), there is little evidence that such demands accurately represent the workers’ complex of desires. To take one glance at such wage demands and conclude that workers simply want higher wages states the obvious, but it misses the point. This would be the same as observing people loot stores in American anti-police riots and concluding that people really want the stuff they looted—true enough, but barely scratching the surface. Like riotous looting, wage demands in China have a “get all you can take” character, in which the very single-mindedness of the motive is itself the signal that there is an unacknowledged excess behind it.

In China, then, there is no workers’ movement brewing: and this is a good thing. There is, for example, hardly any momentum toward the organization of traditional unions that would step in to affirm the identities of migrants as “normal” workers, help broker the price of their labor and thereby facilitate their full incorporation into the wage relation. Despite the attempt of some leftist NGOs in this direction, it appears that the state itself, by attempting to revive the more active role of the All-China Federation of Trade Unions[55] and speaking out against the corruption of low-level officials, is the only significant force pushing for such an incorporation.

The persistent absence of such a “movement” is not unique to China. The crisis of reproduction is also a crisis of the wage relation, in which the wage demand itself becomes “illegitimate,” in the words of Theorie Communiste. Here “illegitimate” signals a systemic impossibility, in which demands for higher wages become harder to fulfill even as inflation and labor shortages make them increasingly necessary. This is accompanied by a crisis in the creation of money, as the system faces proliferating limits to capital liquidity—in other words, a general crisis of value arises:

The current crisis [the one that began in 2008] broke out because proletarians could no longer repay their loans. It broke out on the very basis of the wage relation which gave rise to the financialization of the capitalist economy: wage cuts as a requirement for ‘value creation’ and global competition within the work force. The exploitation of the proletariat on a global scale is the hidden face and the condition for the valorization and reproduction of this capital, which tends toward an absolute degree of abstraction. What has changed in the current period is the scale of the field within which this pressure was exerted: the benchmark price for all commodities, including labour-power, has become the minimum world price. This implies a drastic reduction or even disappearance of the admissible profit rate differentials, through the discipline imposed by financial capital which conditions productive capital.[56]

The “China price” has, for the past two decades, acted as this “benchmark price for all commodities,” and the crisis of the wage relation immanent to China takes the form of general currency turbulence (though suppressed through monetary policy), rising wages in established industrial zones such as the PRD, the relocation of labor to cheaper productive hubs in interior cities, and a massive expansion of speculative investment, particularly in real estate, but also visible in the ballooning size of informal finance alongside the increasing necessity of state-led investment through stimulus and overseas FDI. All of these phenomena confirm the “illegitimacy” of the wage demand, since wages need to and do rise for workers (due to inflation, marketization of the countryside, etc.) precisely when profit rates are already narrowing. Industries relocate, economic growth slows, currencies destabilize, and conditions are established for new strike waves and mass riots.

It is this “illegitimacy” that makes the rise of an effective workers’ movement impossible, leaving the state in an “insurgency trap.”[57] Unable to profitably reform its labor institutions at the national level, China is caught between falling rates of profit and rising waves of strikes and riots. Conceding to one triggers an opposite reaction in the other. The wage demand is made illegitimate by the sharp margins in which it operates. But this illegitimacy doesn’t just preclude the possibility of a workers’ movement, it also creates conditions where attacks on the wage such as those seen recently in China are, in fact, striking at a far more volatile faultline than has been the case for most labor unrest over the past half-century. The previous gains of social democracy in Europe and post-war liberalism in the United States were possible because of the marginal benefits accrued to these regions during brief boom periods following decades of depression and war. Wage demands in that era led to unionization, public works programs, the incorporation of new workers into the more privileged strata of proletarians, and thereby the soft suppression of any revolutionary impulses that remained from the previous era—all because these things were affordable. Today they are not. Workers cannot be affirmed as such, because they cannot be afforded.

The subjective attitude of the workers themselves illustrates this conundrum. The proletarianization process has thus far failed to generate any movement that tends toward the affirmation of workers’ identities as workers. Instead, the subjectivity that is generated takes an abjectly negative form: “there is no future as a laborer; returning to the village has no meaning.”[58] Lu and Pun echo this:

The reform embodies a contradiction: As new labor was needed for the use of capital, Chinese peasants were asked to transform themselves into laboring bodies, willing to spend their days in the workplace. […] Yet, as disposable labor, when they were not needed, they were asked to go back to the villages that they had been induced to forsake and to which they had failed to remain loyal. […] If transience was a dominant characteristic of the first generation of migrant workers, rupture characterizes the second generation, who now spend much more of their lives in urban areas. Transience suggests transitions, and so encourages hopes and dreams of transformations. Rupture, however, creates closure: there is no hope of either transforming oneself into an urban worker or of returning to the rural community to take up life as a peasant.[59]

And this sentiment even begins to override the economic imperatives of migration itself. Describing one of the workers they interviewed, Lu and Pun write:

If the pursuit of material rewards is the shared ambition overriding the internal differences among the working class, the pursuit had lost its meaning for Xin. The concept of work was blighted for him, creating a rupture in his life. “Wherever I work, I don’t feel happy. My soul is never at peace. I always feel that I should do some big thing.”[60]

This keeps the migrants in constant oscillation, generating what Pun and Lu call the “Quasi-Identity” of rural migrants: “one of the women workers we met in Dongguan noted, ‘I missed my home while I was out to [work in the cities]. When I returned home, I thought of going out again.’”[61]

This dynamic has characterized much of the labor unrest in recent years. For these workers, “[a] vicious circle has been created: the reform and the rural-urban dichotomy foster a desire to escape the countryside; escape leads only to the hardship of factory life; the frustration of factory life induces the desire to return.”[62] Xin goes on to lead his co-workers in a strike at his plastics factory: “Caught in the limbo of no return and no progress, they were ready to take radical action.”[63] It is precisely because Xin cannot be fully incorporated into either the “working class” or the dwindling peasantry that he and those like him are driven to attack the conditions that surround them. The absence of the workers’ movement is not a weakness, then, but in fact an opening. When it becomes too expensive to sustain and affirm the lives of workers as workers, this signals that the mutually reinforcing cycle between labor and capital has begun to decay, and the possibility of breaking that cycle altogether emerges.

The result is that the rural migrants’ “quasi-identity” probably has far more in common with the complex, contradictory subjectivity of rioting youth from London council estates than with the toiling, class conscious workers of the leftist historical imaginary. It is notable that Blaumachen describe their “(non-)subject” in terms similar to those used in Lu and Pun’s ethnography:

Precarity, the constant ‘in-and-out’, produces a (non-)subject of the (non-)excluded, since inclusion increasingly tends to be by exclusion, especially for those who are young. […] We are not only referring to the radical exclusion from the labour market, but mainly to the exclusion from whatever is regarded as ‘normal’ work, a ‘normal’ wage, ‘normal’ living.

[…]For the moment, within the crisis of restructured capitalism, the (non-)subject is by now becoming an active force. It continually reappears, and its practices tend to coexist ‘antagonistically’ with revindicative practices, while revindicative practices tend to ‘emulate’ the practices of riot, which unavoidably magnetise them, since ‘social dialogue’ has been abolished.[64]

And the absence of ‘social dialogue’ in China is increasingly apparent. When Xin and his co-workers take their grievances through legal channels, they are finally disregarded at the highest level (petitioning the central government in Beijing). Ultimately, “their reception filled them with despair” and “they realized that they were on their own.”[65] As Xin’s co-worker, Chen explained: “We have to rely on ourselves. We can’t trust the government; we can’t trust management.”[66] In such a situation of “no progress or retreat,”[67] the migrants are forced to “face their trauma and turn their anger outward.”[68] The intrinsic limits of the class dynamic in China increasingly take the form of such external constraints. Repression, administration and social control all become transparent in their brutality, and, backed into a corner, there is little choice but to fight.


No Past

Previous high-points of communist activity have been staffed largely by peasants or by a generation once removed from rural life who still had familiarity with heterodox folk traditions. This heterodoxy had itself emerged from early resistance to landowners and other agents of capitalist subsumption, since these partisans were operating within a globalized but hardly total global accumulation regime that still had ample room to expand its territory. There remained significantly large regions of the globe where this system exerted only a light gravitational pull.

In China, as elsewhere, this created conditions whereby global markets combined with the colonial activities of the core nations to destabilize indigenous power structures and ignite chaotic, inchoate forms of resistance to both new and old regimes. This partial subsumption took the form of a deeply uneven economic geography, in which most large industrial activity took place in a handful of port cities, employing only a small portion of the Chinese population. The majority lived in the countryside, working in agriculture, handicrafts or small workshops distributed between intensive garden-plots, all well beyond the din of the metropolis.

In the more developed port cities, worker resistance initially took the form of anarchist labor syndicates modeled after the French variety, as well as secret societies, sometimes apolitical, sometimes openly aligned with various leftist or nationalist projects. Here, again, the workers joining these organizations were often peasant migrants or their children, and these early forms of coordination were tied together as much by simple anti-foreign sentiment and nationalism as by the loftier universalist goals laid out in the numerous newspapers printed by the Chinese left. All these organizations operated with some degree of secrecy, many were at least partially armed (often staffed by martial arts adepts), and their tactics ranged from simple strikes and boycotts to mass riots and the assassination of industrialists and bureaucrats complicit with foreign interests. Ultimately, however, these early forms of urban organization were unable to overcome their inherent limits. Many secret societies were absorbed into the rising nationalist party (Guomindang) backed by the US, while the anarchist projects collapsed and their members were split between nationalist and communist forces. The communists themselves quickly found their urban network of armed gangs and labor unions crushed by the nationalist military, forcing them to flee to the countryside.

And it was here that the initial limits to the revolutionary project would be overcome. Rural resistance took the form of bandit gangs, the religious cults, and finally the peasant associations founded by revolutionaries. Two decades of war and chaos stretching from 1920 to 1940 had a sort of pressure-cooker effect, melting down and combining all these methods of resistance into the peasant army, marking a general “militarization” of the revolutionary project. In China, as well as in Vietnam, Korea, and elsewhere, it was the peasant army, rather than the union or workers’ movement that proved to be the most successful vehicle for revolution. This was not due to any programmatic or ideological purity, nor to simple questions of strength, weakness or moral appeal, but instead to the simple fact that, given a particular complex of material conditions, the peasant army proved the most adaptive and resilient form of coordination capable of attacking both the encroaching capitalist system and the old order simultaneously, while also providing the infrastructural means to ensure a degree of stability and prosperity in liberated areas. Given the limits of its era, the peasant army was at least able to overcome them in purely tactical terms.

But limits here need to be understood in two senses. First, they are tactical and strategic limits to a specific fight. These are the things that prevent a relatively enclosed conflict (over wage arrears at a single factory, for example), from obtaining its immediate goals or spreading to other factories or neighborhoods. Tactical limits can be relatively straightforward, such as the inability to match and defeat the force of militarized riot police. But they can also be strategic limits to the toppling of the present order, such as the inability to coherently challenge the Chinese state, and the difficulty of any coordinated action surviving censorship, appeasement and outright repression. Historically, such a strategic limit was evident in the incapacity of urban labor syndicates and armed leftist groups to mount sufficient resistance to the nationalist military—the limit finally overcome by the peasant army.

Second, limits must also be understood as limits to the struggle taking on a communist character. Tactical and strategic limits can be overcome in many ways, none of which are in and of themselves communist—the peasant army has historically failed in precisely this respect. A number of measures taken in a given struggle may appear to be consistent with a “left-wing” politics and nonetheless set the trajectory elsewhere. These limits, then, are not ideological limits (problems of “false consciousness”) so much as material limits structured into the conflict. The path-of-least-resistance for a conflict is rarely communist in character, and “consciousness-raising” alone (if at all) cannot force a conflict off this path. No amount of cultural agitation, then, could have pushed the society created by the peasant army’s victory onto a communist trajectory. Instead, such agitation became nothing more than the grotesque embellishment of that society’s slow collapse into capitalism.

But today the peasant army and the conditions that spawned it are gone. Both the potentials and limits of a struggle waged from outside the capitalist system are now absent. There is no way forward, and no way back. So, in a present as grim as ours, what are the current limits of the conflict within the so-called world’s factory? There are the obvious tactical and strategic limits, to begin with: Riots and strikes have simply been unable to survive repression. Some of the larger fights, such as the recent strike at Yue Yuen,[69] have been sustained slightly longer than normal only through the tacit endorsement of the central government. In other cases, demands are won after the strike itself has been crushed and its most active leaders blacklisted or imprisoned.

Often, however, the riots have no concrete demands that could easily be met. They take on the character of an inchoate violence haphazardly targeted at immediate figures of repression and authority. In Wenzhou, a massive crowd nearly beat several chengguan (城管— special civil police) to death after the chengguan harassed a shopkeeper and attacked a journalist who took pictures.[70] In instances such as this, the tactical and strategic limits are less about how to win given demands on the shopfloor and more about how to sustain and focus the power of “the rabble” itself. Nonetheless, the intertwining of direct repression and lucrative concessions has ensured that these mass riots have been effectively prevented from becoming mass occupations of neighborhoods, factories and public squares, a la Gwangju, [71] Tiananmen or Tahrir. As both repressive and re-distributive wings of the state become more constrained by the demands of profitability, however, these preventative measures will begin to falter.

Beyond this, there are the material limits preventing these conflicts from being pushed onto a communist trajectory. The most salient of these at first appears to be the “composition problem.” As described by Endnotes, “the ‘composition problem’ names the problem of composing, coordinating, or unifying proletarian fractions, in the course of their struggle.”[72] This problem arises when “there is no pre-defined revolutionary subject” or, in other words, “no ‘for-itself’ class-consciousness, as the consciousness of a general interest, shared among all workers.”[73] In China, the clearest intra-class divide is the apartheid division between urbanites and ruralites, based on hukou status. But there are plenty of other significant and visible divisions, whether based on gender, race, education, or degree of incorporation into the state’s privilege structure. These divisions proliferate at almost every level, with substantial separations between industries, regions, cities, and even among departments within large factories themselves. No current political project (aside from nationalism, maybe) seems capable of fusing these groups into some sort of for-itself subjectivity.

In the context of urban strikes and riots, the composition problem also appears in the relatively delimited character of each “type” of mass incident. Environmental protests usually remain distinct from labor struggles and forced demolitions or land grabs—even when many of the same social strata are involved in each. These types also have their own forms of discourse, usually adapted for specific varieties of negotiation. Each may exceed this negotiation to some degree, but so far they have not linked up in any substantial way.

All of these struggles, insofar as they remain within the framework of a given form of negotiation, point somewhere other than communism. Even if these conflicts were to gain intensity, they would probably remain negotiations for rights, a better price for land or labor, or slightly more participation in a system over which the participants have no real control. If an overarching profitability can be maintained, even an unprecedented explosion of strikes and riots would be unlikely break out of the cycle of negotiation. It is only when such social dialogue fully breaks down—as the crisis of reproduction deepens—that the possibility of fusing these conflicts into a communist project can arise.

This does not mean that the “composition problem” is solved simply by an intensification of the crisis, but rather that the present “composition” of the class is not really the problem. Composition can be understood as a noun, with the existing composition of the class either containing or not containing some sort of “predefined revolutionary subject,” or as a verb, in which revolutionary subjectivity is composed via action.  By equating the noun with the verb, Endnotes’ hypothesis becomes inadvertently ambiguous on the difference between historical givens (some “predefined” subject) and historical acts. In their fourth issue, Endnotes eliminates this ambiguity via an historical analysis in line with our own. Here, we echo this, positing that the absence of a “predefined revolutionary subject” has nothing to do with the “composition problem.” Instead, it is only the possibility of “social dialogue” between the vicars of capital and certain fractions of the proletariat that makes the activity of composition a problem. Divisions among the proletariat will persist, but as the capacity for social dialogue breaks down, these divisions will flatten, becoming easier to bridge. The idea of a “for-itself class consciousness” or a revolutionary project based on the “general interest, shared among all workers” has always been a myth, trumpeted by the deluded and the powerful—especially those clutching to the helm of dying revolutions. The “unified” revolutionary subject is something that does not precede revolutionary momentum. It is made, rather than given.

Where something approximating such a class consciousness did exist historically, it was not at all an inevitable outcome of any given regime of industrial geography or labor deployment. Instead, such “consciousness” was forged from a messy amalgamation of peasants, artisans, manual laborers, hordes of unemployed, gangs of feral children, angry housewives, starry-eyed millenarians, minor state functionaries, students, soldiers, sailors and bandits, all thrown together in the alembic of the early capitalist city and drawing on diverse traditions of resistance. “Consciousness” was not an idea in people’s heads, but simply the designator for their combined activity.[74]

Similarly, it is clear that any attempt at overcoming the present limits of struggle in China must take the factory city as its central terrain, and operate across divided strata of proletarians unified more by geographic proximity than any innate consciousness of themselves as a class. Today, however, the earlier folk traditions of resistance have grown dim. The revolutionary tradition itself often acts as a substitute, with early Chinese revolutionaries having themselves incorporated and transformed many of these older practices into the mythos of the socialist state. Today, symbols and practices from the socialist era are frequently invoked to justify attacks on those in power. Probably the most prominent symbol of this is the current popularity of Mao worship, practiced by roughly twelve percent of the population (particularly the rural poor) in the traditional style of Chinese folk religion.[75] When combined with popular unrest, these traditions have ultimately tended both to bolster the CCP’s left wing (exemplified in Bo Xilai’s “Chongqing experiment” and Wen Tiejun’s “New Rural Reconstruction”) and to veil the actual potential of insurrection with the mystifying effect of socialist nostalgia.

The second key difference is the changed scale and composition of the capitalist city. Enormous numbers of Chinese proletarians live and work immediately adjacent to large concentrations of productive infrastructure. The Xintang riot took place in an industrial suburb producing one-third of the world’s denim. Several years of unrest at Foxconn plants across China have, similarly, raised the specter of a shutdown in the global supply of iPhones. At first glance, this appears to resemble the situation of industrializing Europe, where the myth of a unified proletarian subject could take hold precisely because such a significant portion of the proletariat was employed in the immediate process of production. But this not the case in today’s China: changes in the technical composition of production have ensured the tendency toward deindustrialization. The fusing of a new revolutionary subject, then, cannot be undertaken through the affirmation of “worker culture” (gongren wenhua – 工人文化), even if the propagation of such myths proved helpful in the past.

Despite this relative deindustrialization, large numbers of Chinese workers are still located at vital positions in the global economy. Riots in Athens, Barcelona, London and Baltimore, for all that they signify, have little chance of breaking the “glass floor” into production. Even if they did, the result would be people filling simple logistics spaces—ports, big box stores, railyards, universities, hospitals, and skyscrapers, all quickly reduced to deserts of empty rooms and shipping containers after the good stuff is looted—or at most a handful of hi-tech factories making specialized goods, with no access to the raw materials or knowledge required to run them. In China, however, engineering knowledge and basic technical acumen is widespread, supply chains are tightly-knit and redundant within industrial agglomerations, and the blockage of a single factory complex’s output can prevent significant portions of global production from going to market.

Meanwhile, the “global factory” constituted by logistics infrastructure is itself built largely in China, where 82% of the world supply of shipping containers are manufactured:

China boasts the world’s largest container and crane manufacturers, is now the third-largest ship-owning country after Germany and the second-largest shipbuilding country after Japan, and has surpassed India as the largest ship-recycling country. [76]

The ability of Western blockades to strangle the accumulation cycle at its consumer end is inherently limited by these factors. Even though production is spread out over global networks and extends fully into the social sphere (the so-called “social factory”), intervention into these networks is not weighted equally everywhere. Even massive obstructions in countries like Greece and Spain can simply be circumvented—troublesome markets can be abandoned, since most are dying anyways as proletarians run out of easy credit. Autonomous zones and workers’ states can be constructed in any of capital’s wastelands without posing any real threat—at most offering a degree of life-support for marginal populations until some future subsumption during a new cycle of expansion.

The point is that there are simply some proletarians who are closer to the levers of global production than others. The goal of a communist project is not to seize these levers and run the system for the benefit of all—because the system is built for immiseration as much as for production. The goal is to terminally disrupt this system, dismantle it and repurpose what can be repurposed, but in order to do this, its key fulcrum must be broken—the immediate process of production, where workers meet capital and things are made. And in order to dismantle and repurpose its components, it is necessary to have knowledge of how the mechanism works, and the technical ability to make sure everyone doesn’t starve in the meantime.

This knowledge is not some abstract object of contemplation, but is instead the embodied product of training and experience within the sphere of production itself. The Chinese workforce was a lucrative source of labor for global capitalism precisely because of this embodied knowledge—the socialist education system had produced a highly literate nation with a glut of mid-level engineers. Today, despite the high turnover of migrant labor, the Chinese proletariat retains a greater and more widely distributed competency in the “technical knowledge of the organization of this world”[77] than is generally the case among proletarians in places like Greece, Spain or the United States. The problem is a practical one. Without a future or a past, we are left only with what lies at hand.



[1] For a mainstream news overview of the events, see Yoon, Eunice, “China’s riot town: ‘No one else is listening’”, CNN, June 17, 2011 9:35 pm EDT. <http://www.cnn.com/2011/WORLD/asiapcf/06/17/china.riot.town.yoon/>

[2] Xintang produces roughly one third of the world’s denim, earning it the moniker: “denim capital of the world.” See:

Li Guang, Jiang Mingzhuo, Lu Guang, “The denim capital of the world: so polluted you can’t give the houses away.” Chinadialogue, 13 August, 2013. < https://fashiontech.wordpress.com/2014/08/05/the-denim-capital-of-the-world-so-polluted-you-cant-give-the-houses-away/>

And: Malcolm Moore, “The end of China’s cheap denim dream,” The Telegraph, 3:02PM GMT, 26 Feb 2011. < http://www.telegraph.co.uk/news/worldnews/asia/china/8349425/The-end-of-Chinas-cheap-denim-dream.html>

[3] See Figures 1 through 4 for details. Note, however, that these figures calculate data for China as a whole. The number of riots, strikes or blockades in a populous and industry-intensive region such as the Pearl River Delta exceeds the national average, as is apparent when the same data is mapped.

[4] The report itself is available here: < http://www.ssapchina.com/ssapzx/c_00000009000200010006/d_0907.htm>
And a summary of its contents in English is available here: < http://www.chinadaily.com.cn/china/2014-04/09/content_17415767.htm>

[5] The China Labor Bulletin strike map is available here: <http://strikemap.clb.org.hk/strikes/en>

[6] Leetaru, Kalev and Schrodt, Philip (2013). GDELT: Global Data on Events, Language and Tone, 1979-2012. International Studies Association Annual Conference, April 2013. San Diego, CA.

[7] It’s also notable that several much lesser-known peaks of unrest are, in fact, picked up by GDELT. The peasant unrest of the late 1990s—particularly 1997—and the early 2000s, discussed elsewhere in this issue, appear clearly in Figure 1.

[8] See: Neel, Phil A. (2014) “Counting Riots,” Ultra. <www.ultra-com.org/projects/counting-riots>

[9] These figures are original calculations from the full GDELT database, normalized by total world events (versus total events within the country) and transformed by a simple multiplier to ensure that the numbers on the y axis do not appear in scientific notation, with a Loess smoother applied—note that such a smoother is not a stand-in for a regression line or curve. These figures are meant to demonstrate characteristics of the observed data in a descriptive manner, not to imply a regression model of the data or other inferential methods. The original data can be found at <http://gdeltproject.org>

[10] Reproduced from Phil A. Neel “Counting Riots,” Ultra, May 22, 2014. <http://www.ultra-com.org/project/counting-riots/> These figures use trend lines rather than a Loess smoother, but, again, the intent is to show characteristics in the observed data, not perform linear regression or similar procedures. Also note that these are normalized by national event totals, rather than global event totals. The pattern is the same but the relative heights over time change slightly. Neither method is incorrect, they simply emphasize different aspects of the total data. In Figures 1 and 2 we normalized by global totals to scale the phenomena within global trends, whereas Figures 3 and 4 include the global trend lines themselves.

[11] Jason E. Smith “Occupy, the Time of Riots, and the Real Movement of History,” Scapegoat Journal, Issue 03, 2012.

[12] “The Holding Pattern,” Endnotes, Issue 3, 2014. < http://endnotes.org.uk/en/endnotes-the-holding-pattern>

[13] See: Joel Andreas. Rise of the Red Engineers: The Cultural Revolution and the Origins of China’s New Class. Stanford University Press, 2009.

[14] Ho-Fung Hung. “Labor Politics under Three Stages of Chinese Capitalism,” The South Atlantic Quarterly, 112(1), Winter 2013. pp. 203-212.

[15] See Sikander, “Twenty-Five Years since the Tiananmen Protests: Legacies of the Student-Worker Divide,” Chuang (originally posted on Nao), June 4, 2014. <https://chuangcn.org/2014/06/twenty-five-years-since-the-tiananmen-protests-legacies-of-the-student-worker-divide/>

[16] For an overview of this process, see: Ching Kwan Lee, Against the Law: Labor Protests in China’s Rustbelt and Sunbelt. University of California Press, 2007. pp. ix-xii

[17] Movement Communiste and Kolektivne Proti Kapitalu, “Worker’s Autonomy: Strikes in China,” 2011, p. 28 <https://libcom.org/library/workers-autonomy-strikes-china-mouvement-communiste>

[18] Eli Friedman, “China in Revolt,” Jacobin 2012 Issue 7-8, 2012. <https://www.jacobinmag.com/2012/08/china-in-revolt/>

[19] Movement Communiste and Kolektivne Proti Kapitalu 2011, p. 32

[20] Lu Huilin and Pun Ngai, “Unfinished Proletarianization: Self, Anger and Class Action among the Second Generation of Peasant-Workers in Present-Day China,” Modern China, 36(5), 2010, p.495

[21] ibid. p. 497

[22] ibid.

[23] ibid.

[24] ibid.

[25] ibid.

[26] For an overview of this exact same migrant-centric proletarianization, as it occurred in California agriculture from the 19th century onward, for example, see: Carrie McWilliams Factories in the Field: The Story of Migratory Farm Labour in California. University of California Press 1935

[27] Kam Wing Chan, “The Global Financial Crisis and Migrant Workers in China: ‘There is no Future as a Labourer; Returning to the Village has No Meaning,” International Journal of Urban and Regional Research. 34(3), September 2010. p.665

[28] ibid.

[29] ibid, p.667

[30] In Dongguan, 70-80% of the labor force is composed of migrants and migrants make up more than three quarters of the total 6.5 million population. In neighboring Shenzhen 7 million of the city’s de facto 8 million urban residents do not have a Shenzhen hukou. (Chan September 2010, pp. 663-665).

[31] See: “Greece Unemployment Rate,” Trading Economics, 2013. <http://www.tradingeconomics.com/greece/unemployment-rate>

[32] See: “Spain Unemployment Rate,” Trading Economics, 2013. <http://www.tradingeconomics.com/spain/unemployment-rate>

[33] This number, taken from Chan (September 2010), also potentially underestimates the absolute number of unemployed, since it assumes that all migrants who remained in the cities retained some form of employment. (See Chan September 2010, p. 667, Figure 2).

[34] The stimulus was equivalent to one-eighth of the entire output of the Chinese economy.

[35] Note also that the Chinese state was capable of this fiscal intervention even while fears of internal debt crisis were beginning to rise, with bubbles originating at the level of enterprises and local governments. Actions like this have further secured an inflated faith among investors in the CCP’s regulatory capacity.

[36] Chan 2010, pp. 666-667

[37] ibid. p.666

[38] See Gordon Orr, “What’s Next for Guangdong,” Harvard Business Review. < https://hbr.org/2012/02/what-next-for-guangdong>

This also signifies a state-led attempt to completely shift the economic geography of the country post-crisis. Xi Jinping’s “New-Type Urbanization Plan,” announced in 2014, imagines a “country of cities,” secured by capping growth in established (mostly coastal) mega-cities and instead building up small and medium scale regional centers (many of which will be in the interior) into large urban zones in their own right. Meanwhile, the existing urban cores are being encouraged to invent an entirely new economic base founded on services and hi-tech production.

[39] The white male worker was, historically, the prototypical wage-holder in the American context, but other countries have seen other wage hierarchies based on different complexes of history, gender, ethnicity and simple geography.

[40] Again, see McWilliams 1935 for a history of such labor management in California.

[41] Chen 2013, p. 203

[42] Woland/Blaumachen and friends. “The Rise of the (Non-)Subject,” Sic: International Journal for Communisation, Issue 02, 2014, Bell & Bain Ltd., Glasgow. p. 66

[43] ibid, italics in original

[44] Rocamadur/Blaumachen. “The Feral Underclass Hits the Streets,” Sic: International Journal for Communisation, Issue 02, 2014, Bell & Bain Ltd., Glasgow. p. 99

[45] ibid

[46] Endnotes 2, p. 17

[47] Banister, Judith. “China’s manufacturing employment and hourly labor compensation, 2002-2009,” Bureau of Labor Statistics, June 7, 2013. <http://www.bls.gov/fls/china_method.pdf>

[48] see Banister 2013, “Manufacturing in China” and Figure 8

[49] Theorie Communiste. “The Glass Floor,” in Les Émeutes En Grèce. Sononevero, 2009. <http://libcom.org/files/glass-floor.pdf> pp.41-42, emphasis in original

[50] See: Al, Stefan. Factory Town of South China: An Illustrated Guidebook. Hong Kong University Press, 2012.

[51] This mechanism, though tending toward totality in the abstract, is by no means truly “total,” in the sense of its actual effectiveness. The existence of the riots themselves are clear evidence of this, as are the persistence of slum zones such as the 城中村 (chengzhongcun – “villages inside the city”), and the state’s inability to reign in corruption or even have its basic directives followed by local governments.

[52] See: “Employment in Agriculture,” International Labor Organization, World Bank. Accessed on: April 8, 2014. <http://data.worldbank.org/indicator/SL.AGR.EMPL.ZS>

For a broader explanation of the same dynamics across the BRICs, see: Joshua Clover and Aaron Benanav, “Can Dialectics Break BRICs?” The South Atlantic Quarterly, 2014. <http://krieger.jhu.edu/arrighi/wp-content/uploads/sites/29/2014/03/Can-Dialectics-Break-BRICS_JHU.pdf>

[53] Banister 2013, p.2

[54] Gwynn Guilford. “Five charts to explain China’s shadow banking system, and how it could make a slowdown even uglier,” Quartz, February 20, 2014. <http://qz.com/175590/five-charts-to-explain-chinas-shadow-banking-system-and-how-it-could-make-a-slowdown-even-uglier/>

[55] See: Eli Friedman, Insurgency Trap: Labor Politics in Postsocialist China. Cornell University Press, 2014

[56] Roland Simon, (R.S.). “The Present Moment,” Sic: International Journal for Communisation, Issue 01, 2011, Bell & Bain Ltd., Glasgow. p. 104.

[57] See: Eli Friedman, Insurgency Trap: Labor Politics in Postsocialist China. Cornell University Press, 2014.

[58] Chan 2010

[59] Lu and Pun, p. 503

[60] Ibid, p. 507

[61] ibid.

[62] ibid., p. 513

[63] ibid. p. 511

[64] Woland/Blaumachen and friends. “The Rise of the (Non-)Subject,” Sic: International Journal for Communisation, Issue 02, 2014, Bell & Bain Ltd., Glasgow. pp. 66, 67

[65] Lu and Pun2010, p. 512

[66] qtd. in ibid.

[67] ibid., p. 514

[68] paraphrased from ibid. 513

[69] See: Friends of Gongchao, “The New Strikes in China,” 2014 < http://www.gongchao.org/en/texts/2014/new-strikes-in-china>

[70] For the story in English, see: Alex Stevens, “Rioting crowd severely beats 5 chengguan for killing civilian,” Shanghaiist, April 21, 2014. <http://shanghaiist.com/2014/04/21/rioting-crowd-beats-5-chengguan-for-killing-civillian.php>

[71] The Gwangju uprising in May, 1980 in South Korea. See: <http://libcom.org/history/1980-the-kwangju-uprising>

[72] “The Holding Pattern,” Endnotes 2, pg. 47. < http://endnotes.org.uk/en/endnotes-the-holding-pattern>

[73] ibid. pg. 48

[74] In short: “Subjectivity” has always been “(Non-)-Subjectivity,” in Blaumachen’s terms, only now with less peasants. For a more detailed version of this argument, see Endnotes, Issue 4.

[75] Li Xiangping, “Xinyang, quanli, shichang–Mao Zedong xinyang de jingjixue xianxiang” [李向平:信仰·权力·市场——毛泽东信仰的经济学现象, Faith, power and the market–The economics of Mao worship], January 28,2011, Zhongguo zongjiao xuexhu wang [Academic website on Chinese religion], http://iwr.cass.cn/zjyjj/201101/t20110128_6030.htm

[76] Cowen, Deborah. The Deadly Life of Logistics, University of Minnesota Press, 2014. p.67

[77] See The Invisible Committee, To Our Friends, Semiotext(e), 2015. p.95

[78] Liu Shifu, whose anarchist organization founded the first modern labor unions in China, previously belonged to the underground China Assassination Corps. An Elizabeth Perry, in Shanghai on Strike (Stanford University Press, 1995) discusses the historical role played by gangs on both sides of the political spectrum in Shanghai’s labor movement.

[79] Eli Friedman, Insurgency Trap: Labor Politics in Postsocialist China. Cornell University Press, 2014, p.1

[80] Ibid, p.2


Chuǎng, issue #1 (2016)