Kung Fu Tea

The Logic of Violence and its Relationship with the State

My academic background and doctorate is in political science where I specialize in a sub-field called “international political economy.”  That is where I have focused most of my teaching and writing over the years.  I have never really been an “area studies” person.  I actually became interested in the academic study of the martial arts precisely because China was such a great “case study” for many of the most important theoretical questions of my field.

For instance, if you want to understand how swings in global trade effect local society, or when revolutions are likely to happen, how the idea of “national identity” is spread or the specific relationship between a community’s  physical security and good governance, late Qing and Republic era China is a topic of endless interest.  It presents tons of well-preserved data that speaks to many really important questions and debates.  Better yet, relatively few people in the international relations literature use this era of Chinese history in their writings.  That further increases the chances of finding something new and interesting that might actually be helpful when trying to decipher other events around the globe.

While rereading a section of David Robinson’s volume, Bandits, Eunuchs and the Son of Heaven: Rebellion and the Economy of Violence in Mid-Ming China (2001, University of Hawaii Press) I was reminded of many of the core questions and concerns that drew me to Chinese political history in the first place.  If you are not familiar with Robinson’s work you should run, not walk, and get a copy of it.  It really is that important.

While not a history of the martial arts per se, Robinson provides readers with something equally important.  His brief study paints perhaps the richest picture of the violent world of soldiers, bandits, vagabonds, rebels and corrupt officials during the Ming dynasty that we currently have.  This is critical because it describes the milieu that the Ming (and later Qing) era martial arts actually arose in.  If the martial arts are viewed as a “solution,” a purposeful creation reacting to something in the social environment, Davidson does a wonderful job of describing many elements of the “problem” that generated them.

Of course that last sentence is actually subtly problematic.  It implied that the martial arts were a rational response to a discrete set of easily identifiable problems in the environment of late imperial China.  This characterization ascribes a certain degree of rationality to the martial artists, bandits, soldiers, performers and corrupt officials of the period.  As a political economist this is precisely how I am most comfortable looking at the world.  To oversimplify terribly, people face problems, so they create institutions to help them deal with those challenges, and some of these institutions work better than others.  Much of my professional work has to do with understanding why these institutions sometimes succeed, but more often fail.

Still, this is not how we usually talk about the martial arts, or any of the central institutions that control violence (the throne, the military, the clan structure or the imperial degree system) in medieval China.  Very often when discussing the martial arts we just assume that these are ancient traditions emerging out of the mists of time, carried on by a profound sense of cultural continuity.

If that is the case, whatever it was that the martial arts were a response to is long lost to the historical record.  It is something that existed hundreds of years ago and probably did not even have any direct bearing on life in the Ming.  As such there is no reason to think that the martial arts, or any of the key institutions of violence, are a rational response to anything.   They are simply artifacts of Chinese history and culture that Ming officials inherited and were forced to make the best of.

In current academic parlance we would say that the first view of the Chinese martial arts follows a “rational choice” approach, where the second does not.  It is a culturally driven, qualitative, understanding of the same phenomenon.  This distinction is actually pretty important, because it determines what sorts of skills and methodologies you will need to carry out the rest of your discussion.  It even tells you what the goals of that investigation should be, “causal inference” or “thick description.”

For instance, if we adopt a rational choice perspective what I really need to understand are the institutional constraints that actors, whether bandits or soldiers, in the Ming operated under.  If they worked under very tight constraints (for instance, a bandit might be executed if captured) I can rationally reconstruct for myself the strategies that he might employ not to be caught.  Even if it is impossible to guess what his exact moves are, I should at least be able to explain them in simple strategic terms.

If, on the other hand, these actors are responding primarily to cultural scripts and inherited identities, than strategic analysis is not likely to be all that helpful.  Instead I need to strive to attain a deep normative or cultural understanding of the meanings of their actions.  Maybe the martial arts novels of the period, or the developing ideas of “martial virtue,” will be more important than the seemingly universal dictum of strategists like Sun Tzu.  For instance, when discussing the Boxer Uprising Esherick notes that condemned men and women tended to adopt scripts and behaviors that were highly stylized and drawn from the executions of “heroes” in popular operas.  Throughout this work he wonders how much impact theatrical ideas and stories had on the actions and choices of other individuals in the region.

In broader terms this is really a discussion about the best way to understand the world, either through a “rational” or a “normative” lens.  It is probably the single most important debate in all of the social sciences today.  It is literally what keeps professors up at night (at least it does in political science).

Not surprisingly, this debate reoccurs throughout the theoretical literature.  For instance, Benedict Anderson sees nations as “imagined communities.”  While the first batch of nations (in the Americas and Western Europe) may have evolved organically and liberally, during the post-colonial period state leaders (such as those in China) callously crafted definitions of “the nation” to include some groups and exclude others in ways that benefited themselves politically.  Sadly this often involved great violence.  While Anderson is interested in how values and identities are created, he sees them as emerging from a fundamentally rational and strategic set of processes.

Ernest Gellner and Anthony Marx (an old professor of mine, now the President of the New York City Library) have instead argued that leaders are not free to create nations any way they would like.  Rather the seeds of this identity are inherited from the distance past and they constrain what sets of options are open to leaders at any given time and place.  As such “the nation” tends to reduce down to some other preexisting category like language, ethnicity or religion.  These irreducible categories also cause conflict both within and between nations.

David Robinson’s discussion brings up many of these same issues.  While written as a purely historical text about the political economy of violence during the Ming dynasty, his book suggests some pretty relevant questions about how we should view patronage, power, coercion and community violence in the current era.

Of course care must be taken when extrapolating arguments about the past into the present.  At minimum we need to have a clear conception of what our theory is, and what the driving forces behind it actually are.  Only then can we safely ask whether this is similar to, or different from, what we see today.

Unfortunately there is some ambiguity in Robinson’s discussion on exactly these points. He never directly addresses the question of rational vs. normative theories, and seems to draw freely from both sets of distinctions when it suits his purposes.  On one level there is nothing wrong with that.  Real life is messy.  Sometimes we respond to rational inducements (like when we go to a job we hate because we get paid to do so) and sometimes cultural considerations and identities take over (like when we gender certain sports, such as wrestling or boxing).  In that sense Robinson’s story likely reflects an ambiguity on these points that his subjects also felt.  If our goal is thick description and descriptive inferences, that is fine.

If our goal is to figure out how much of this argument applies to the social regulation of the martial arts in Hong Kong in the 1960s and 1970s (which is my actual area of focus) then I need something else.  It is no longer enough to say “the world is complicated, everything matters,” rather I must decide which specific variables and historical pathways matter the most.  Yes it is difficult to tease these arguments apart, and many normative theorists doubt that this sort of “causal explanation” is even possible.  But that is why Robinson is getting paid the big bucks.  So what specifically does he have to say on the subject?

The Martial Arts and Violence: Two Theories of Social Control.

Robinson describes two different models of how social elites sought to manipulate independent bases of coercive power in traditional Chinese society.  One of these strategies of explaining the existence of violence and its subjugation to the state is fundamentally normative in nature.  The other one ignores Confucian or cultural concepts and instead looks at the material power dynamic between core and peripheral elements in society.  In a nutshell, the core tolerated and even attempted to co-opt independent bases of violence because it lacked the strength to do anything about them, even at the supposed “height” of the Ming dynasty.

Let’s start by assuming that there is a special logic of violence in Chinese society that exists independently of any material conditions or “facts on the ground.”  What would a cultural or normative theory of this situation look like?

There seems to be a paradoxical dialectic in traditional Confucian thought that might be a good place to start.  Orthodox Confucian thinkers assert that by their very nature all men are good.  This means that with the right guidance, education and institutional oversight, almost anyone can be made a productive member of society.  When this process derailed good Confucian scholars were not above actually looking at a situation in detail to attempt to find the place where local institutions had failed, letting all parties down.

This faith in humanity is actually an important point to keep in mind when thinking about the social regulation of crime and violence.  In the modern west we tend to remember traditional Chinese justice only for its violence or seeming brutality.  Torture and execution were used much more liberally at the time than would be acceptable today.  Nor were all people equal before the law in China.  Social relationships and status were often the defining facts in how a crime or lawsuit was resolved.

Still, when you read the accounts of criminal trials in Robinson’s book, it is clear that there was genuine mercy and actual attempts at rehabilitation in the Chinese justice system.  Over and over criminals I expected to simply be executed (because that is how they would have been dealt with in Europe at the time) were given relatively short prison terms or relocated to another community through exile.  It would have been easy to execute all of these criminals when they were in state custody, but the belief that they could be made “better” had a material effect on the way that justice was carried out.

Balancing this out, late imperial society also included a health degree of class based suspicion and even hatred.  While it was theoretically possible to make anyone a useful member of society through education, in actual fact officials tended to regard peasants as hopeless objects.  Refugees, soldiers, bandits, merchants, wandering monks, vagabonds and “urban toughs” were viewed even more harshly.  These individuals were seen as irredeemably based or coarse.  They existed in a world dominated by violence, and in the views of their social betters, violence was the only language that they truly understood.  They were beyond any real hope of redemption.

This view even found its way into the civil service exam tradition.  The children of traveling performers, opera singers, prostitutes and boat people were all prohibited from even taking the exam.  Thus the only official path for social mobility was closed to individuals from the lowest social caste.

And as far as good society was concerned, this was just fine.  Rather than rescuing these individuals from their benighted existence, the real goal of a statesman was to use their “unique skill-set” to execute and carry out government policy.  After all, a talented bureaucrat might be able to come up with a plan to reinforce the border, keep bandits off the road, or dredge a canal, but they certainly could not carry out such a scheme without a very large number of “rough men” who knew about labor and violence.

The classic illustration of this line of thought is the character Monkey from Journey to the West.  At the start of the story Monkey (a natural martial artist) is more concerned with wreaking havoc on the world than doing anything good or wholesome.  In fact, Monkey seems to have an actual aversion to “good and wholesome.”  As Robinson points out, he is a shockingly violent character.

And that is exactly what Heaven needs.  A religious work is about to be undertaken and Heaven needs an “enforcer” to make sure the task can be seen through to completion.  Heaven needs a hero that can act as the left hand of God (or the Buddha in this particular case).  Rather than destroying Monkey he is set to this particular task.  A deity fits an iron band around his head that causes him indescribable pain when he does anything “evil” and makes him the apprentice/servant/guardian of a wandering monk intent on retrieving ancient scriptures from India.

Throughout the rest of the story Monkey is forced to save his master from all kinds of monsters, and he usually does this in the most violent ways possible.  Yet his actions are now “good” because they have been subordinated to Heaven’s cause.  It is not clear that Monkey is made any better through this process, and he is clearly just as violent at the end of the story as he was at the start.  Yet that is precisely what makes him useful.  As a specialist in violence Monkey literally exists to be exploited.

When looking at the formation of patronage networks and the social regulation of violence in Ming China it is very possible to see exactly this sort of thing going on.  Elites never really tried to cure the root causes of violence because they didn’t see it as intrinsically bad.  In all reality they needed a supply of hardened individuals to carry out their plan, and so they accepted a situation in which such people would always be in ready supply.  Rather than remaking Ming society into something much more strongly centralized and institutionalized, they were content to refine their own “iron band” of social control over the martial artists, thugs and bandits that they accepted into their households.

There are some problems with this narrative.  Robinson goes a long way towards normalizing the reality of violence and social coercion in daily Ming life.  I think that this is a valuable corrective for modern readers who are generally exposed only to elite accounts of how life was supposed to be (i.e., everyone-including the Emperor-listens to the advice of the Confucian elite and there is harmony in the realm).  The reality of life during this period was actually so far from the ideal that one wonders how the educated class managed to keep up the charade.

Part of the Confucian meta-myth is that the martial forces of destruction and disorder (wu) can be institutionalized and controlled by the powers of civil statecraft and education (wen).  This is basically the same pattern that we saw in Journey to the West, but now on a much broader scale.  Of course it is hard to “subordinate” these forces without first acknowledging them, and then giving them something in return.  So even at the best of times we might be subjected to the jarring sight of high officials cultivating relationships with local toughs, bandits and (heaven forbid) common soldiers.  But that was fine because in Confucian thought there was no balance between Wen and Wu; the former clearly dominated the later.

Except that in real life that almost never seems to have happened.  Robinson’s entire book is one protracted case study in elites being out-maneuvered, inconvenienced and generally held to account for the ill-considered self-enrichment schemes of their informal retainers.  No level of society seemed entirely safe from this.  Village elders, brothel owners, landlords and even high court officials were constantly being bitten by their own guard dogs.  Economists call this sort of thing a “principal-agent dilemma” meaning the desires of the employee (the military retainer) are not properly aligned their employer (the prince or landlord who sponsors them).  In other words, social control of these “independent bases of violence” tended to exist much more in theory than in fact.

Nor does it seem that any of these individuals really have any idea how to address these problems.  Military retainers and thugs were essential because the roads around the capital were filled with bandits.  They were necessary to protect your estate and to intimidate your enemies.  But they were never fully under control and inevitably contributed to the very problems they were supposed to solve.  Time and again the court would order a crackdown on banditry and the assigned officer, with the supposed weight of the kingdom behind him, would fail (at which point they were usually demoted and sent to the frontier).

It is one thing to say that this situation was allowed to exist because of some cultural norm, but it is quite another to realize that even when the government tried to address fairly minor manifestations of this lawlessness (such as banditry directed against tax payments) they discovered that they actually lacked the troops and institutional strength to do anything about it.  A fair percentage of the time the bandits turned out to be the government’s own soldiers, and even the elite Palace Guard themselves!

By modern standards, what Robinson describes is clearly a failed state.  Not only that, it’s a state that is stuck in a poverty trap.  The institutions of the day do not give anyone an incentive to solve the really pressing economic and political problems.  Further, the government itself seems paralyzed and unable to radically rethink its own institutions because deep down inside they don’t really believe they can ever dig their way out of the hole. Of course to be totally fair one would be hard pressed to find any well-functioning states in the 1400s.  At least the Chinese leadership should receive credit for making a serious go of it.

This is a fascinating conversation precisely because this interplay between corruption, violence and institutional failure is not unique to medieval China.  In fact, it has been seen in most areas of the world at one time or another.  Granted it never takes quite the same shape in any two places (so “standard operating procedures” are a bad idea), but certain basic principles seem to be universal.  If you want to get out of this trap you need some set of institutions that will bring individual incentives in line with social incentives.  Ming era China lacked what development economists call “efficient institutions.”  That is what Robinson illustrates so aptly.  It doesn’t matter who you look at, court officials, military officers, common soldiers, gang members, or large landholders, none of them have incentives that are actually aligned with the good of society as a whole.  All of these individuals are embedded in a corrupting environment that force them to seek their own self-interest in ways that are destructive to the greater community.

Note that we are now having a very different sort of conversation than was the case with the previous theory.  We still need to know enough about Chinese society to understand how its institutions function on a fairly detailed level, but at the end of the day we are discussing rational actors who are motivated by their own self-interest.  Why do they choose violent strategies?  Because in this environment violence pays, and sometimes it pays extraordinarily well (at least in the short-run).  Do we need to know anything about ancient Chinese culture or literature to understand this?  No.  Do we even need to stop and define the “national interest?”  Not really.  That rarely comes up in a rent-seeking state.

Testing our Theory with the Opera Rebellion: Guangdong, 1854-1855.

I personally suspect that the toleration of high levels of violence, including extensive militarized patronage networks, likely had more to do with the general weakness of the state than any sort of cultural toleration of martial virtues.  One cannot help but notice that when the pendulum swings in the other direction, and the state is genuinely strong, it tends to be much less tolerant of crime, disorder and anyone who would use force to challenge the legitimacy of the government.  For instance, there is generally less disorder at the start of a dynasty when the state is strongest and can still exert considerable control over (recently pacified) local society.

Another good example of this would be the Red Turban Revolt which ripped through Guangdong in 1854-1855.  It makes a useful test case as it occurs later and in a different region of the country.  As such it is totally independent of Robinson’s research.  For any graduate students out there it is critical that you remember to always test your theories on a different body of data than you used to create them.  Yet many of the same basic variables and social conditions that he was interested in can be seen here as well.

Sometimes referred to as the “Opera Revolt” because of the participation of some Cantonese opera companies, this uprising had fewer connections to the Taiping Rebellion than is commonly assumed.  The event actually started as a violent tax revolt along the East Branch of the Pearl River.  Because of the strained nature of Guangdong’s economy (which was being forced to pay for the entire war effort against the Taipings) it quickly spread throughout the province.

While the actual uprising may have had its roots in economic (rather than religious or ideological) grievances, it posed a very serious threat to the existence of the imperial order in Southern China.  For months it was touch and go as to whether Guangzhou would be able to defend itself from the rebels.  Foshan (the location of important cannon foundries) and other cities in the region fell relatively quickly in the first burst of revolutionary fervor.

Eventually the government forces triumphed and were able to counterattack throughout the Pearl River delta driving the remaining rebel forces north and west.  Usually that is where the telling of the story ends, but for the purposes of this post what happened next is critical.  Previously the government had lacked the ability to adequately control banditry and piracy in the region for a number of reasons.  It lacked the troops, it lacked the cooperation of the major clans (who played a much greater role in local politics in the South than in other parts of China) and it was too infiltrated with corrupt clerks and allies of the various secret societies.

The Red Turban Revolt changed all of that.  It was destructive enough that it actually forced officials (many of whom were literally besieged and cut off from outside communication) to make a choice as to whether they were going to support central government control, or to throw their lot in with the many local patronage networks that were turning against the established social order.  Likewise clan elders, degree holders and local landlords were all forced to decide whether they favored state regulation or violent revolution.

It was a stark choice, and revolution likely meant economic ruin for all of the existing social elites in southern China.  These elites cut their ties with networks that aligned themselves with the revolutionaries, organized their own households into government registered militia units and threw their full economic support behind the Governor.  It was this social realignment more than anything that happened on the battlefield that assured the eventual defeat of the uprising.

At the end of the formal fighting the local government found itself in a unique position.  Typically in southern Chinese politics the Governor General was forced to balance the competing economic interests of the literati/landlords with the masses.  Either side could cause a public disturbance if they were unhappy with the course of events or how they were being treated.  Still, the Governor could control the situation with relatively few troops precisely because he held the balance of power between these two competing blocks.

Now the situation was different.  The Governor had entered into a firm alliance with the local landlords against the masses.  Further, the gangsters, drifters, martial artists and bandits who had been the “muscle” of the poor residents of the area had been vanquished.  In short, the local government of southern China was suddenly stronger, and had more coercive influence over society, than it had possessed at any time since the end of the “clearances” at the start of the dynasty.

And what did the government do with its new found power?  It started hunting down socially undesirable people and killing them, often for no reason, in massive quantities.  Government troops and gentry led militia members rounded and killed not just former rebels but also bandits, secret society members, traveling performers, homeless individuals, wandering monks and priests, and anyone else who they thought could be “trouble.”  Clan leaders took the opportunity to have bothersome community members eliminated and scores were settled by the tens of thousands.  Noted historian Fredric Wakeman estimates that in the attempt to restore “good social order” follow the defeat of the rebels, the government and its allies may have killed up to one million people in the Pearl River Delta region alone.  The loss of life and social disruption in this purge clearly dwarfs the death and destruction of the Red Turban Revolt.  The ban on the public performance of Cantonese opera that is of such interest to modern martial artists was actually just one small part of a much larger campaign of terror directed at the lower levels of society.

So why did the government suddenly decide to kill all of the socially undesirables?  Why not work them back into renewed and strengthened patronage networks?  I suspect that the only explanation one can really give is that the government took this course of action because it could.  While there had been previous campaigns against pirates, gangsters and bandits in the region, the State had never undertaken an operation on this scale because it lacked the material strength and elite support to do so.  However, the moment the government had sufficient strength the Confucian values of “moderation” and “education” were replaced with an extreme love of (equally Confucian) “justice” and “social order.” And to some extent it worked.  While crime can never totally be eliminated (especially with the British importing huge quantities of opium into the region), the purges of the 1850s seem to be responsible for the generally conservative nature of region up through the 1920s.

The Red Turban Revolt is an interesting case for a number of reasons.  While historians often neglect it (situated as it is between the first and second Opium Wars and overshadowed by the Taiping Rebellion) it is a critical episode in the history of the southern Chinese martial arts.  Many modern schools love to associate themselves with the colorful gamblers, gangsters and opera performers of the rebellion in their creation myths.  Of course these myths themselves mostly date to the 1920s or 1930s when the actual events were far enough in the past that they were safe to re-imagine.  In the 1850s being “a rebel” was not socially popular; it was a literal death sentence.

The government did an extremely good job of figuring out who the rebel leaders actually were and hunting them all down.  We know this because the British seized all records relating to the incident and its aftermath when they occupied Guangzhou a few years later.  The original accounts of the Red Turban Uprising and the execution of its leadership can still be seen in London today.  I suspect that many of the region’s martial arts actually owe more to the 19^th century gentry militia movement that put the rebellion down than they do secret societies or revolutionary groups that promoted it.

The “White Terror” following the end of the rebellion is an interesting case study in what elites, both in government and society, thought the ideal community should look like in the late imperial period.  Given unlimited power they did not move to “educate” or “reform” the bandits, vagabonds, rebels and floating population.  Instead they moved decisively to wipe them out.  While cultural narratives may be a useful tool for discovering how people understood and rationalized the world around them, they don’t seem to offer much of an explanation for the choices that are made in actual times of crises and civil war.  That behavior instead appears to be both materialistic, opportunistic and highly strategic.  Rather than being caught up in circular cultural narratives, these are the sorts of variables that we need to consider when looking at the overall political economy of violence in late imperial China.