Defining Civil War Again, FFS!

CNN has a post today pondering whether Syria is or is not a civil war.  It almost gets things right and then cites a bunch of people with strange ideas.  I have posted about this before, but because there seems to be much confusion and it is so easy to crap all over the CNN post, here we go.

The post does mention folks reporting the classic definition: two or more sides (that each has killed at least 100 so that it is not just a mass killing [genocide/politicide may co-exist with civil wars but are not the same thing]) of some severity.  The 1000 battle death total is arbitrary--it could be 100 or 500 or 10000.  1k was borrowed from the inter-state war scholarship because it made for less measurement error.  Using a lower threshold, like 25, might mean missing heaps of conflicts and maybe mis-coding things that are not really civil wars.

But the point is that a civil war is a violent set of events involving two or more sides, contesting the governance of a country.  It can be separatist--the contest is over who governs a spot of territory within the country.  It can be about regime change.  It can be revolutionary--who gets to re-shape the social and political order.  The goals are political--not just an outburst of violence aimed at another ethnic group--that would be a riot.  Two or more sides, violence, who governs. 

A popular uprising may or may not be enough to be a civil war--depends on whether there is two-sided violence that is more than just a single outburst.  So, this line is not entirely right:

"Syria did not start out as a case of civil war because the opposition to the government mainly took the form of a popular uprising in March 2011," she said. [Anuradha Chakravarty, assistant professor of political science at the University of South Carolina,]

One could say that it started out as repression and then become a civil war once folks started fighting back.   Which is what Chakravarty then says.  So she is very much on target as are Stephen Biddle and James Fearon who are also cited in this article.

Other folks?  Not so much.
Assad?  Well, we don't expect him to get it right.  A sectarian conflict can be a civil war if the state is the target.  Two groups fighting with each other is often called communal conflict and may not be a civil war, like the battles between Hindus and Muslims in India are violent, but not aimed at the state.  But plenty of civil wars are "sectarian" if that means ethnic or religious--that groups with different identities combating to control the government is very much a civil war.  Sectarian conflict and civil wars are not mutually exclusive.

Michel Weiss is simply wrong:

"Civil war suggests the previous state that exists all but failed and collapsed," he said. In some regions, the government lacks control and there is a "growing equalization" of forces, he said.

The article had just cited Fearon's example of the US Civil War, where the US was hardly a failed state in the north.  Indeed, most civil wars are between various groups and a government that still exists.  The contest is often uneven and "equalization" is hardly inevitable.  Indeed, many contenders to the throne government lose.  Biafra lost its battle to secede from Nigeria--it did not become equal, it got crushed.  Many civil wars can be sustained for a very long time, especially if one side chooses to engage in an insurgency.  Not all insurgencies are civil wars, not all civil wars are fought via insurgency. 

Steven Heydermann suggests there is some threshold that has not yet been crossed.  Where is that threshold?  For most scholars of civil war, a civil war exists if it is in one part of a country--it does not have to be all over the place.  If Libya and Lebanon are the "models" for civil war, it is not clear why Syria is not, as he insists, given the size of the opposition in Libya last year vs Syria this year.  Now I am really confused.

Jeff White, a fellow at an institute, at least makes it clear that this is his definition (which is a good thing since it is not one shared by many):

"My definition of civil war is a situation which is characterized by conflict between two, or more, segments of society. I still see the situation in Syria as one of fundamentally an armed and unarmed insurrection against the government. That is, the people are fighting the state, not each other."

Um, huh.  So, the American Civil War is not a civil war, since the South fought the US government?  One of the problems with fixations about failed states and the ethnic security dilemma (sorry about that) is that governments do not disappear once the firing starts--they are combatants.  Not just any combatant but the one with the most guns and most capability much of the time.  What would Jeff White call this thing where people fight the state, if not a civil war?  What was Libya last year?  Oh, and by the way, when people are fighting the state, the state is made up of people who fight back.   White's entire definition is not of civil war but of ethnic conflict.  Not all civil wars have ethnic cleansing.

This is all reminiscent of Rwanda where folks avoided calling it a genocide since that seemed to require an international response.  BUT there is a convention against genocide that requires a response.  Responsibility to Protect is not nearly as established, but perhaps may be responsible for some of this hedging about the Syrian civil war.

I read this piece because someone complained via twitter--did we really need such a long piece defining civil war?  No, we didn't.  The article could have stopped after Biddle's quote, Chakravart's or Fearon's.  Oy.