Ethiopian dictator Meles Zenawi died last month. He hasn’t been succeeded by anyone yet. (Formally, officially, the Deputy Prime Minister has stepped up. But nobody seems to be taking him too seriously. He’s a mild-mannered guy, not a member of the revolutionary generation, and the wrong ethnicity to boot.) Zenawi was a “slash the tallest flowers” kind of guy; years of life on the run as a guerrilla in the bush seem to have rendered him permanently paranoid. Anyone who seemed too popular or too competent got demoted. At best demoted. So there’s no obvious successor.
I think the odds of Ethiopia staying a dictatorship are pretty good. It’s a country that has never known democratic rule or any sort of pluralism. Since the Emperor Haile Selassie consolidated his rule in the late 1920s, Ethiopia has had exactly three rulers: Selassie himself, Stalinist dictator Haile Meriam Mengistu, and the recently deceased Mr. Zenawi. That’s just three guys in almost 90 years, all of them absolute rulers with little or no tolerance for dissent. The country has almost no liberal traditions. Civil society does exist — there are NGOs and labor unions and even a few lawyers who are willing to sue the government — but it’s young and small and weak. It won’t be too surprising if a new strongman emerges. But it may take a while, and Ethiopia may go through a period of oligarchic rule and jockeying for power before things sort out.
Regimes with clear lines of succession are generally stronger regimes. Part of the reason is that they’re slightly harder targets — you not only have to take out the top guy, you have to get rid of his designated successors as well. But I think the bigger reason is that everyone knows what’s coming next after the dictator. Even if the successor is a dubious character, he’s likely to seem better than chaos and uncertainty.
Traditional monarchs have known this for millenia. But the lesson applies to dictatorships as well. We currently have at least five extant examples of non-royal dictatorships that have been successfully transferred to a son or brother of the dictator: Azerbaijan, Congo, Cuba, Syria, and North Korea (twice). Gabon and Togo are authoritarian regimes run by sons of former dictators; if we include them, the club expands to seven.
Examples of transfer to a non-relative are much much rarer. There have been a couple in the past. (Ataturk hand-picked his successor; Mexico in 1934-94 had an extremely powerful president who chose his successor from the party leadership, but they did not meet the definition of dictator. After all, they had to leave after six years in office.) But not a single extant dictatorship fits this pattern! (Burma’s dictator chose a successor who is still in power, but the successor seems to prefer being primus inter pares to being a true lone ruler.)
Now, it’s a long established principle that old age is a dangerous time for a dictator. All three of the dictators picked off by the Arab Spring were in their golden years. (Porfirio Díaz in Mexico and Juan Velasco in Peru also went down in their dotage.) More generally, an old dictator with no clear successor is pretty clearly putting himself at risk. That said, some dictators do run the course and die, more or less peacefully, in office.
So what happens when a dictator manages to pull this off — grows old and dies without first being overthrown or losing an election or whatever? Well, we can sketch out a tentative typology.
- A son or other close relative steps up. (Several extant examples, see above.)
- A hand-picked successor takes over. (Rare; no extant examples.)
- There is a party or other system independent of the dictator that chooses a new dictator. (Rare outside of the Communist world; Mexico in 1934-94 comes closest. No extant examples, though we might see one soon in Cuba.)
- After a period of uncertainty, brief or prolonged, a new dictator takes over. (A couple of extant examples — Turkmenistan, South Sudan.)
- There’s a shift to authoritarian but non-dictatorial rule, either by a single ruler, an oligarchy or junta, or a party system. (China, Vietnam, Ivory Coast.) This seems to be the single most common outcome in the last decade or two.
- There’s a relatively rapid shift to relatively liberal or democratic rule. (Portugal, Spain. In the former, it took six years and revolution to get democracy.) This happens, but it’s less common than we might like to think. A shift in the direction of liberalism is much more likely when the dictator has been overthrown, or has voluntarily stepped down (Indonesia, Tanzania, Singapore), or agreed to elections (Chile, Zambia, Senegal). A system that lets the dictator die in office seems less likely to produce democracy at first go.
Numbers 2 through 6 include the possibility of foreign intervention; the United States, France and the Soviet Union have all intervened in this manner at one time or another. We may call these 3a (party picks a new dictator, foreign approval makes the difference: the Eastern European model, 1945-1990), 4a (new dictator takes power subject to foreign approval) and so forth. There aren’t really any extant examples of these at the moment — the U.S. seems to have lost interest in installing dictators in Central America and the Caribbean — but I would not be surprised if Russia’s “near abroad” in Central Asia were to throw up a 4a or 5a in the next little while.
6a is damnably rare, actually nonexistant: Panama would be one of the rather few cases since WW2 in which foreign intervention replaced a dictatorship with a democracy. Iraq sort of maybe kinda if you squint really hard. The jury is out (to be charitable) in Libya. And all of those interventions happened when the dictator in question was very much still alive and in power.
Anyway. There are enough elderly dictators out there that we should expect to see some more examples in the next few years. I expect that (5) “a shift to authoritarian but non-dictatorial rule,” will continue be the most common outcome. Which means that some countries that are currently dictatorships will drop off the list! If the number of dictatorships is to remain constant, then new ones will have to emerge. How likely this is ... well, this post is long enough. In a bit.