Imagine the following: we wake up one morning to find that South Africa’s government has disappeared. (I see the obvious joke here, but I won’t take the bait.) There is no constitution and parliament, no defence force and police, no courts and prisons, no traffic officials, no Minister of Unknown Portfolio, no Home Affairs queues (because there is no Home Affairs department), no grant system, and, yes, no taxes.
Even more, our government-provided infrastructure, from bridges to ports, airports to waterpipes and, yes, even our power grid, has vanished, as have all firetrucks, public museums and border posts.
Here is the question: Waking up to such a situation, how would we organise a new government? Put differently: What would government look like if we could construct it from scratch?
Political philosophers have two theories of how governments emerge. Think of them as ‘demand side’ and ‘supply side’ theories.
The ‘Demand Side’ theory views the government as an organisation with a comparative advantage in providing public goods and resolving disputes. In other words, individuals might willingly give up certain freedoms and resources to a governing body, a concept encapsulated by the term ‘social contract’, coined by philosopher John Locke. In this social contract, citizens agree to surrender some of their autonomy and resources, like paying taxes, in exchange for the government’s promise to maintain order and protect their rights, such as life, liberty, and property. This voluntary surrender of freedoms is seen as a necessary step to establish a more orderly and fair society where rights are respected and protected.
This decision to establish a government is often motivated by problems that naturally arise in the absence of a governing body, such as externalities and coordination failures. An externality occurs when an individual’s actions have consequences on others that the actor does not consider. A coordination failure refers to a situation where individuals fail to collaborate towards a mutually beneficial goal. By forming a government, citizens aim to manage these issues better, ensuring more effective coordination and management of public goods and services, and reducing the impact of negative externalities on the community.
In the ‘supply side’ theory, governments are seen primarily as instruments of extraction by an elite group. This theory, often associated with Karl Marx and Friedrich Engels, posits that those in power form a governing body not necessarily for the public good but to consolidate their control and leverage over resources. In this view, the establishment of government allows for organised and systemic extraction of resources from the broader population, thereby benefiting the ruling class.
Despite the seemingly self-serving motives, the provision of public goods can still occur within this framework. This is often the result of bargaining between social groups. The elite, while seeking to maintain control and continue resource extraction, may provide public goods to keep the rest of society compliant or to avoid rebellion. Thus, the public goods provided are less about benevolence or mutual benefit, as in the ‘demand side’ theory, and more about maintaining the power dynamics that allow for resource extraction.
Which of these two theories would apply to us were we to organise a new government? Would we organise locally to provide the ‘public goods’ that the market would not provide? Or would we see elites contest control and establish a government to primarily extract resources for their benefit?
An experiment from almost 5000 years ago provides an answer to this intriguing question.
A new AER paper by three economic historians – Robert Allen, Leander Heldring and Mattia Bertazzini – uses data from ancient Mesopotamia (modern-day Iraq) to analyse the interplay between geographical changes and the demand for government intervention.
It works like this: Some of the first human settlements emerged in the floodplains between the Euphrates and Tigris rivers of Iraq. But over the centuries, these rivers changed course. The authors use one such shift around 2850 BCE as a case study to test their hypothesis. They propose that when a river shifted and moved away from a farming area, it could have created a ‘demand’ for government. This is because farming in this region relied on irrigation, and the sudden shift of the river would require cooperative efforts among distinct communities to build longer canals for re-irrigation. If a government emerges where the river shifted away, it suggests that the demand for coordination and public goods, like irrigation, is driving the formation of government (the ‘demand side’ theory).
However, the disappearance of the tax base due to the river shift could have weakened the incentives for elites to extract resources, possibly leading to the formation of governments where the river had shifted towards. If a government is more likely to form where the river shifted towards or remained stable, this might imply that elites are setting up governments primarily to maintain control over resources (the ‘supply side’ theory).
The authors test these two hypotheses by collecting archaeological data from various periods and locations, comparing areas directly affected by a river shift to those where access to water remained unchanged. The results indicate that a river shift led to an increase in the probability of that area becoming part of a state, an increase in public good provision (e.g., irrigation canals, defensive walls), and an increase in resource flow to the government (e.g., tribute payment). Although many historical states have been extractive, the authors suggest that governments initially emerged because of the need for public goods. It was only once they had some enforcement power that such power could be used for repression or extraction.
What are the lessons for South Africans five thousand years later?
There is no denying that many functions of government have become a way for the ruling elite to systematically extract resources from the broader population; the various state capture reports spell that out pretty emphatically. But what is also true is that many functions of government are indeed necessary to provide public goods or solve coordination failures.
One thing Ancient Mesopotamia might teach us is that it is best to attempt to solve these market failures at the local level: a ‘social contract’ only makes sense if everyone buys into it. The devolution of power from the national to the local level could be one way we shift the reason for government from extraction to construction.
An edited version of this article was first published on News24. The image was created with Midjourney v5.2.
It seems a bit odd to attribute theories of state origins as coercion to Marx (I know in this case it's the 'hydraulic' theory of state formation that's motivating that thought) when I'd point instead to anarchist social theory (they even cite Scott) or for that matter to right-libertarian thinking.
There's one part of the interpretation that I'm not sure I understand, which is how they infer the novelty of the state in the cities within their study data. It seems it's via the appearance of an administrative building, but it's entirely possible that this post-dates the appearance of a "state lineage"--that the ziggurat's appearance is not at the same moment that a kind of lineage-above-the-lineages comes into being. (Indeed, given what they want to argue about the contractual character of the emergence of the state lineage, it has to precede the building of an architecture to match--this can't be a chicken-and-egg problem, because you wouldn't build an administrative structure prior to the appearance of a stable administrative social formation. But this is important because their argument hinges so much on timing--that state emergence is tied to fairly precise shifts in the river.
I think there's also a kind of naive understanding of lineages as being contractual rather than coercive or extractive. Economic historians who study African societies are quite able to point to examples of lineage-based political economies where lineages are intensely hierarchical and extractive, where you would never think to call agreements between lineage heads in a multi-lineage community anything like a social contract. There's a big problem with taking the basically liberal conception of personhood embedded inside social contract and applying it to societies where most persons would not in any sense be consulted or agree to decisions to form some new kind of high-level coordination of multi-lineage decisions. More importantly, in a society fundamentally organized around lineage, there is no "outside" the lineage--e.g., persons dissenting from the choices of lineage heads cannot opt out because there is no alternative form of social being available to them, no "stranger" status. Which kind of loops right back to theories that suggest the extractive nature of these early states.
(In that regard, it's a bit odd that they take no note of Scott's argument that early Mesopotamian walls are not the provision of public goods but instead a sign of extractive coercion--that those walls served to keep people in, not to keep enemies out.)
It seems to me they've got a good argument for the endemic rather than external origin of small states in the region (e.g., that these are not the extension of larger states like Uruk into a hinterland) but not necessarily for "states were created to solve collective action problems rather than to enable extraction".