Monday, January 20, 2014

Hobbes and Somalia: Is Anarchy Better Than a Predatory Government?

Is it true, as Hobbes declares in chapter 13 of Leviathan, that the state of nature, where there is no government, must be a state of war where human life is "solitary, poor, nasty, brutish, and short"?

Hobbes presents this as an empirical claim based on five kinds of evidence.  First, it's an inference from what we know about the human passions that lead to conflict.  Second, it's confirmed by our personal experience when we lock the doors of our houses at night.  Third, it's supported by what we know about the savage life of the American Indians living without government.  Fourth, we can predict that when any people lose their government, they will fall into civil war.  Finally, international politics shows the governments of the world to be either at war or in a posture of war because there is no common power over them to keep peace.

Much of the history of modern political philosophy has turned on the debate over this Hobbesian teaching about the state of nature.  And it's an illustration of how debates in political philosophy ultimately depend on empirical claims that can be verified or falsified by the scientific study of political psychology, political anthropology, and political history.  A biopolitical science would embrace all of this.

So, for example, an evolutionary political anthropology would deny Hobbes's assertion that human life in a state of nature is "solitary."  Human beings have never lived as asocial animals.  For most of our evolutionary history, our ancestors lived as hunting-gathering foragers organized as small bands of people bound together by kinship ties and with some informal and episodic leadership.  This suggests that Hobbes was wrong in denying Aristotle's claim that human beings are by nature political animals.

But there is some ambiguity about this in Hobbes's writing.  For instance, Hobbes says that "the savage people in many places of America, except the government of small Families, the concord whereof dependeth on natural lust, have no government at all; and live at this day in that brutish manner, as I said before" (89).  If families have some kind of government, then how can he say that they have "no government at all"?  In Hobbes's Latin translation of Leviathan, he writes: "Are there not many places where people live so today?  The people of America live so, except that they are subject to paternal laws in small families; and the concord of those families is sustained only by the similarity of their desires."  So here he clearly recognizes "paternal laws" as constituting a kind of government based on kinship.  He even acknowledges that a large family that is not part of a commonwealth might be considered "a little Monarchy" (ch. 20, 142; ch. 22, 163; ch. 30, 235).

Hobbes also suggests that even where there is no "Commonwealth" to set a "common Rule of Good and Evil," such a rule can be set by "an Arbitrator or Judge, whom men disagreeing shall by consent set up, and make his sentence the Rule thereof" (ch. 6, 39).  And one of the laws of nature specified by Hobbes is to submit controversies to the judgment of an arbitrator (ch. 15, 105, 109).  Indeed, in bands of foragers, it is common for some influential individuals to act as arbitrators in mediating disputes, which shows an informal kind of governmental leadership.  If "anarchy" literally means "no ruler," then human beings have never lived in anarchy, although they have often lived without a centralized state.

But even if Hobbes concedes that human beings can live in ordered societies with customary laws but without a centralized state to enforce a formal system of law, his argument seems to be that societies without a centralized sovereign state tend to be unstable and inclined to fall into civil war.  Even if life in a stateless society is not solitary, it is "poor, nasty, brutish, and short."

Hobbes's argument seems to be supported by political scientists today who talk about the problems that come with weak or failed states--like Somalia, for example.  Any yet, Peter Leeson, an economist at George Mason University, has argued that the recent history of Somalia actually shows how life in a stateless anarchy can be better than life under a predatory government.

In 1960, the Republic of Somalia was formed as a independent country uniting what had been British Somaliland and Italian Somalia.  Then, in 1969, Major-General Mohamed Siad Barre led a military coup that overthrew the democratic government of Somalia and established a military dictatorship that became a socialist dictatorship.  All land and most of the industrial and financial property was nationalized.  The government was repressive in denying all civil and political liberty, in suppressing all opposition to the government, and in promoting the power of Barre and those he favored.  Somalia is divided into clans, and Barre used the government to advance the interests of his clan (the Marehan) over other clans.

In 1991, Somalia's government collapsed, the country fell into civil war, and there was no central government at all.  After twenty years of being stateless, the leaders of Somalia's clans finally agreed to a new constitution in 2012, which established a parliamentary government.  And yet it's unclear whether the new government can maintain its power in the face of continuing factional conflict among the powerful warlords and clans.

Although this might seem to confirm Hobbes's warning about the propensity of stateless societies to fall into a war of all against all, Leeson claims that "Somalis are better off under anarchy than they were under government."  His evidence for this is that many economic and social indicators--such as life expectancy, health care, extreme poverty, and economic production--show that life in Somalia is better than it was under Barre's predatory government.

Leeson is mistaken, however, in identifying this stateless condition of Somalia as a situation of anarchy with no government.  As he indicates, Somalia has had a legal system based on private, customary law enforced by clan leaders, with security provided by clan militias.  Thus, Somalia has had a kind of government--the government that comes from the rule of clans, which has prevailed in many societies throughout human history. 

Leeson refuses to recognize this as government because there is no "monopoly on the law or its legitimate enforcement" (700).  He simply assumes that government must be identified as a Hobbesian Leviathan state with a centralized, monopolistic claim on the legitimate use of force.  He thus ignores the fact that most human societies have been governed by very decentralized forms of government, including "the government of small families" in foraging bands and the government of extended families in clan societies.

Strictly speaking, anarchy is impossible, if we define it as a total absence of any form of governance.  Those who identify themselves as anarchist theorists implicitly recognize this, because what they defend as anarchy is actually a society that is stateless but governed by voluntary associations.  Similarly, Leeson seems to defend anarchy broadly defined as stateless self-governance.  In Peter Marshall's grand history of anarchist thinkers and movements, he indicates that anarchists distinguish society and the State and defend society as a self-regulating order. He explains: "Pure anarchy in the sense of a society with no concentration of force and no social controls has probably never existed.  Stateless societies and peasant societies employ sanctions of approval and disapproval, the offer of reciprocity and the threat of its withdrawal, as instruments of social control" (12-13).

Leeson rightly identifies Somalia as a possible example of how a stateless society might be better for its members than a predatory government that oppresses them.  But he also recognizes that it would be even better for Somalia to have a liberal state with limited powers for protecting life and property and providing public goods for the common welfare.

Would it be possible and desirable for a country like Somalia to move from a clan society to a liberal state?  Or would the move to liberal individualism require a costly loss of the moral solidarity that comes from the communal life of a clan society?  Would this require, as Friedrich Hayek argued, a painful suppression of those natural human instincts that favor the social life of clans or tribes based on ties of extended kinship?

Those questions will be the subject of my next post.

Thomas Hobbes, Leviathan, ed. Richard Tuck (Cambridge: Cambridge University Press, 1991).

Peter T. Leeson, "Better Off Stateless: Somalia Before and After Government Collapse," Journal of Comparative Economics 35 (2007): 689-710.

Peter Marshall, Demanding the Impossible: A History of Anarchism (Oakland, CA: PM Press, 2010).

Some related posts can be found here, here, here, and here.


Roger Sweeny said...

I'm pretty sure it was in
Power And Prosperity: Outgrowing Communist And Capitalist Dictatorships (2000) that Mancur Olson made a distinction between roving bandits and stationary bandits. Roving bandits come and take and don't really think about the future. But if you're a stationary bandit who takes from the same people all the time, you have an incentive to make your victims prosperous. Thus, there will be more to steal in perpetuity. So, argues Olson, even a government that desires nothing more than exploitation can care about the well-being of the people it exploits.

David Henderson links to a 1993 APSR paper by Olson that was an early presentation of the idea.

Gene Callahan said...

"And it's an illustration of how debates in political philosophy ultimately depend on empirical claims that can be verified or falsified by the scientific study of political psychology, political anthropology, and political history. "

No, you've shown that this one claim is connected to some empirical facts, not that all debates in political philosophy "ultimately" depend on empirical claims.

clifford said...

Have you seem John Scot's The art of not being governed (Yale)