Why Do Countries Become/Remain Democracies? Or Don’t? (23): The Great Leader Theory of Democratization

The death of Nelson Mandela is the perfect occasion to clear up some myths about the Great Leader Theory of democratization. Mandela has become almost the archetype of a great man and a great leader, and deservedly so. He was certainly instrumental and some would say essential in the democratization of South Africa. But what can we say in general about the role of leaders? That’s a lot harder and I’m afraid the answers will reveal some of our deep seated biases.

When countries turn to democracy, to what extent does it help to have a charismatic leader leading the process? Either an insider leading the old system to a new one, such as Gorbachev who in a sense started a top-down transformation. Or outsiders such as Wałęsa, Mandela and perhaps also Aung San Suu Kyi leading a popular opposition movement against those in power, overthrowing the old leaders, taking over from them and riding on a wave op popular support towards a new democracy. The opposite theory is also common: the Great Dictator who almost single handedly destroys democratic forces at home and wards off supposedly overwhelming international pressure. Castro comes to mind here.

Political science has shown that there are many causes of democratization, and that several causes are effective simultaneously, with different weights for each effective cause in each particular case. I hope my blog series on democratization has left no doubts about that. This reality is often obscured by the temptation to give undue weight to the “Great Leader” factor, because that is a cause that can be readily observed, as opposed to many other causes such as internal friction within elites, the economy, outside pressure etc. The appeal of the “Great Leader” theory hinges on the mysterious talent or faculty of charisma: no one knows what it means, what it does or how it does it, but it’s widely believed that charisma yields great power and influence. Combined with a view of the “populace” as dimwitted sheeple easily impressed – for good or for bad – by this mysterious faculty of charisma, it becomes almost self-evident that great leaders must have an overriding effect on regime change (or regime continuation as the case may be). It’s obvious that there are some biases at work here: we give more weight to what we can easily observe and ignore deeper causes that often need both detailed knowledge of specific circumstances and statistical or scientific analysis.

Note that I don’t claim leaders have no role to play. On the contrary. Mandela is a great historical figure – as are the others I’ve cited – precisely because he had an effect. All I’m saying is that we should be aware of our biases. The Great Leader Theory becomes all the more suspect when we look at the long term survival and flourishing of democracy – as opposed to the democratic revolution itself. It’s clear that leaders by themselves play only a limited role in the long term prospects of democracy: institutional development, economic prosperity, rule of law, the diminishing role of violence and of the military, bureaucratic development and other indispensable building bricks of democracy can never result from the actions of a single person. I’m afraid this is all too obvious when we look at Mandela’s country, or many other democracies for that matter.

Measuring Democracy (2): Polity IV, and Some of Its Problems

Polity IV is, like Freedom House and others, a project ranking countries according to their political regime type. It’s extensively used in comparative and causal analysis that require a distinction between democracies and non-democracies, partly because its time series start from the year 1800.


perspective envisions a spectrum of governing authority that spans from fully institutionalized autocracies through mixed, or incoherent, authority regimes (termed “anocracies”) to fully institutionalized democracies. The “Polity Score” captures this regime authority spectrum on a 21-point scale ranging from -10 (hereditary monarchy) to +10 (consolidated democracy). (source)

The Polity Score is the aggregate of 6 component measures that aim to record what are called key qualities of democracies: executive recruitment, constraints on executive authority, and political competition.

However, it seems that Polity IV doesn’t adequately measure what it claims to measure. Its concept of democracy is quite thin, resulting in a fair number of “perfect democracies”, whereas we all know that there is no such thing in the world we live in. And other countries, which are obviously dictatorial, are classified as fairly democratic. A quote from this paper (which is an attempt to improve Polity IV):

Polity’s 21-point democracy/autocracy scale, illustrated by the dashed line [in the figure below], tracks the major changes in British political history, but only roughly. The Reform Bill of 1832 revised a complicated system of determining the franchise by increasing the number of voters from 500,000 to 813,000. Despite the modesty of this expansion, changes in the Polity Score for Britain give a sense of greatly expanded democracy, moving from a -2 (democracy=4, autocracy=6) to a +3 (democracy=6, autocracy=3).

However, … only six percent of the adult population voted even after the reform.

While the male franchise had broadened considerably by 1884, suffrage still excluded agricultural workers and servants. Actual voter turnout reached 12% of the population only in the election of 1885 before falling, and didn’t return to that level again until 1918. All the while, Polity scores for executive recruitment and competition increased while institutionalized autocracy decreased. In 1880 the Polity democracy score stood at 7 (autocracy=0). By 1901 the democracy score rose to 8 and by 1922 Polity suggests that Britain was a “perfect 10” democracy, even though full male suffrage was not achieved until 1918 and full female suffrage until 1928.

Britain has received the highest democracy rating ever since, even though the voting rate has never exceeded 60% of the adult population.

The high scores that Britain receives from 1880 on are misleading and, with respect to changes in participation, mistimed. As Figure 1 illustrates, participation doubled during a period Polity records as unchanged and doubled again during a modest 2 point move in Polity.

The racial exclusion in South Africa also demonstrates the danger of conceiving democracy without taking account of the breadth of citizen participation. According to Polity, South Africa was a relatively stable democracy from 1910 until 1989. It was coded a 7 out of 10 on democracy and a 3 of 10 on autocracy, bringing its score to +4. A positive score is surprising because it ignores the exclusion of the 90 percent of the population that did not – most could not – vote.

Switzerland, our final example, has scored a perfect 10 out of 10 on democracy in the Polity dataset since 1848, even though women – roughly half the population – were not granted the right to vote until 1971, 123 years later. Furthermore, electoral turnout has hovered around 30% recently, despite virtually universal suffrage. One reason is that Switzerland’s collective executive is an organizational form that diminishes voter motivation by minimizing the significance of election outcomes. Surely such a system should be regarded as less democratic than one in which most citizens participate in elections that actually make a difference in the leadership and policies of the nation.