Brendan Smith, Jeremy Brecher and Tim Costello have an interesting piece on Global Labor Strategies entitled "China's Emerging Labor Movement." In it they argue that China is in the early stages of a full-fledged labor movement. Like workers in the United States before the introduction of the Wagner Act, Chinese workers must organize outside the law and without the representation of proper unions--for now.
Smith, Brecher and Costello approve of China Labour Bulletin's "CC-2005 Campaign" of exploiting the loophole in Chinese labor law that allows workers to collectively bargain on their own if there are no ACFTU representatives on their shop floor. In campaigns like CC-2005 the authors see the possibility of transforming China's thousands of protests (87,000 last year alone) into something more solid.
I find their historical analogy compelling. However, there are two things worth noting.
First, many of China's protests are in the countryside, not factories. This is not important because of some Marxist dogmatic distinction between farmers and proletarians. It matters because China is huge and protests are so spread out and so cut off from each other (due to government censorship and a documented penchant on the part of protesters to AVOID linking up with others for their own safety) that they do not necessarily pose any threat to the state or capital. Many countries--India, across Latin America, etc.--have done just fine with massive unrest as long as it is scattered.
Second, the Chinese state's capacity for repression exceeds that of the U.S. at the beginning of the twentieth century (when China too had a powerful labor movement). Nor is there a separate party within the state to take advantage of / moderate / channel popular discontent, as Roosevelt did.
This is not to say that a real, fairly unified labor movement can't be built. And China Labour Bulletin's plan is the clearest on the table yet. But we must keep innovating, based on what makes the Chinese case different.