Here are my reactions to last week’s Book Club comments, starting with a fine exchange between John Alcorn and KevinDC.


1) In previous posts, you argue that totalitarian regimes can maintain power indefinitely — or at least much longer than they do — if successors would practice ruthless repression like the founders. For example, loss of nerve among rulers after Stalin, culminating in Gorbachev, explains the collapse of communism.

In your post about war, you argue that war is an efficacious means to the end of justifying ruthless domestic repression, and that war also spontaneously occurs among power-hungry dictators.

Why, then, did successors often lose their nerve in 20th-century totalitarian regimes? (We’re back to sideward glances at western prosperity, and tensions between totalitarian empire and national sentiments in smaller, satellite States.)


I suspect this has a lot to do with the nature of power struggles in dictatorships. Initially, they are won by the most ruthless and cold blooded people – the ones who will do absolutely anything to get power. But almost by definition, in the process of gaining power they also push aside or eliminate everyone who was almost but not quite as ruthless as they were. And during their reign, they keep a firm eye out for and move swiftly against anyone who might be ruthless enough to challenge them. As a result, when the first dictator passes, there’s nobody left who has that same level of brutality and brutal competence, so their successor is inevitably less brutal and more moderate. This may also explain what’s different in the case of North Korea – being an explicitly familial dynasty, you could select for equally brutal successors in a way that wasn’t true in the Soviet Union.

Alcorn again:

2) Re: North Korea.

Are you sure that dynastic succession (kin lineage) facilitates selection for efficacious brutality? As you point out, trust might allow the founder to inculcate brutality in the son. However, natural endowments, too, matter. Brutality genes might skip a generation! Regression to the mean is probable. Kin lineage greatly reduces the scope of eligible pool of talent in efficacious brutality.

Blaise Pascal argued that kin lineage reduces both competence and strife.

Both John and Kevin make good points.  My reconciliation, to channel Gordon Tullock:

1. Revolutionary dictatorships are the worst of the worst, because revolutions select for bloodthirsty risk-taking true believers.  After a successful revolution, prospects are bleak until the whole founding generation dies off.  When Mao finally croaked, China was amazingly lucky to get a crusty pragmatist like Deng Xiaoping instead of a second Maoist fanatic.  Perhaps the Red Guards who crippled his son shattered Deng’s youthful revolutionary faith – and rescued a billion Chinese from the Maoist hell-state.

2. Subsequent generations of dictators are generally a big improvement.  Sure, the upper echelons struggle eagerly for power.  But stable regimes attract slightly squeamish risk-averse opportunists.  After two generations, these opportunists come to vastly outnumber bloodthirsty risk-taking true believers.

3. Strictly hereditary dictatorship, per Pascal, has the lowest selection pressure for bloodthirsty power-hunger.  While plenty of hereditary dictators are still awful tyrants, hereditary dictators are the most likely to peacefully relinquish power, or at least “go with the flow.”  The main worry is just that weak hereditary leaders will be reduced to figurehead status by whoever wins the tournament to “advise” them.

So what happened with Gorbachev?  He mostly fits my profile of a “slightly squeamish risk-averse opportunist.”  You could object that a risk-averse leader would never have embarked on glasnost and perestroika, but I say Gorbachev didn’t realize he was playing with fire until it was too late to retain power without a swift reversion to mass murder.  And to his credit, Gorbachev was too squeamish for that.

David Henderson:

I was expecting, when I saw that you had a link to the statement “It takes an outsider to see the ideological landscape as it really is,” that you would reference Hayek’s The Road to Serfdom. I’m disappointed that you didn’t. You might argue that, coming from Austria, which was so close to Germany, Hayek was not clearly an outsider. But that makes his accomplishment all the more impressive.

Fair point.  Though I’m not a fan of Hayek, I agree that he deserves credit for popularizing the totalitarian model in The Road to Serfdom.

Henri Hein:

I agree that the Thought Police is efficient in fictional Oceania, but I have often found this to be one of the less plausible constructs in the novel. If government is so inefficient at everything, why should it be able to run an efficient Thought Police? I understand that the Russian equivalent was frightening, and somewhat effective, but given the powers and resources they were given, I don’t see any reason to accept they were efficient.

I agree that Orwell’s depiction of the efficiency of the Thought Police is implausibly high.  Once Winston and Julia get arrested, we learn that the Thought Police was on to them for years; they were sitting on piles of redundant evidence the whole time.  And the only clear “false positive” in 1984‘s system of repression is the character of Parsons, who was plainly a loyal Party member falsely denounced by his own children.  Real totalitarian regimes, in contrast, heavily persecute even their loyal followers.  Still, we should not underestimate the ability of totalitarian regimes to excel in tasks they prioritize.  As I’ve said before:

Communist regimes did provide poor incentives to produce consumer goods for ordinary citizens.  But they provided solid to excellent incentives in the sectors they really cared about: the military, secret police, border guarding, athletics, space programs, and so on.

Performance in these sectors was often (though hardly always) world-class.

I’ll post my final thought on Orwell’s book-within-a-book next week, along with replies to any general comments participants care to offer.