I come both to bury the Gulen movement and to praise it.
Since the failed coup attempt in July 2016, the Gulen movement seems to be fading into the background in the story of Turkey. If you hear about it much at all, it’s mostly relating to whether or not Gulen himself will be extradited to Turkey. The dogs bark, but the caravan moves on.
As a political force in Turkey, the Gulen cult seems to be mostly destroyed. Then again, when you’re talking about the sort of entity that the Gulenist movement is, it’s always difficult to separate fact from fiction, or to declare it totally dead.
But let us assume that it is in fact finished and see what lessons can be drawn from its fascinating rise and fall.
The Gulen movement is difficult to map to modern American institutions. When you attempt to describe it concisely, to western ears it sounds literally incredible, like some sort of practical joke, but it goes roughly as follows: Imagine that the Kaplan test prep center was running a conspiracy of regime change by creating a society like the Freemasons, comprised of its graduates that it placed in important jobs around the country. Meanwhile, the doctrine of the organization itself was a cross between an Islamic religious movement, and a cult of personality based around its leader who lives in exile in the Poconos in Pennsylvania.
Notwithstanding the absurdity and implausibility of the prior description, the more you read of it, the more the picture resembles the strange caricature above. The Steve Sailer version above roughly matches the New Yorker, which is about as close to political consensus as you can get.
If my article has one central thesis, it is this – nearly all of these strange aspects served some useful adaptive function for the group’s success.
In terms of its history, this group was slowly building its power and influence over a period of maybe 40 years. This coup was a long time in the making.
The group’s first rise to power came when it allied with the Islamists, including current Prime Minister Recep Erdogan, to undermine and overthrow the secular, Kemalist military which had dominated the country for a long time. Turkey has had a number of coups before – Wikipedia lists 1960 and 1980 as coups, along with 1971 and 1997 as involving “military memoranda” (which I think means “jump or you’ll be pushed”), and one fascinating entry labeled “1993 alleged Turkish military coup”. But they all involved the military deciding that the government had become too Islamic, which was threatening the secular nature of the established order.
In conjunction with the Islamists, the Gulenists helped to destroy the military establishment in the mid 2000s through a series of prosecutions, jailings, smear campaigns, and firings against top military leaders, often in the context of allegations of planning coup attempts. The military, unlike the Islamists, has essentially an all-or-nothing choice – boot out the government and implement a coup, or let the situation stand. Whether they didn’t or couldn’t pull the trigger at that time is now largely moot.
But the game theory of three-sided wars is inherently messy and unstable. Nearly always, two sides begin by ganging up on the third – often, the two smaller ones fighting to overthrow the larger, more powerful one. But once they start succeeding, at some point, the current allies realize that they’re soon going to be fighting each other, and so someone defects and attacks the other. Like in a repeated prisoner’s dilemma, there are incentives to do this sooner, to get the upper hand. But what this often means is that it’s not uncommon for the defection to happen before the original common enemy is fully defeated. Hence the instability in three-sided conflicts.
So it seems to be here. The Gulenists and Islamists teamed up to destroy the secularist military, but by late 2013, it became apparent that that job was mostly done, and that the two sides began to come into conflict with each other. The Gulenists began a campaign of publicly exposing the corruption of Erdogan and his allies in December 2013. Erdogan retaliated with firings when he could, and was apparently planning a big push to extirpate the Gulenists from public life. Faced with this impending purge, the Gulenists went ahead with the coup attempt last July.
So much for the generally agreed-upon history. From here, the story gets more speculative, but more interesting.
You’ll notice that in the above description, and in the discussion of the undermining of the military, I said that it had become too late for the secularist military to pull off a coup. In doing so, I’m implicitly claiming that the coup in July wasn’t the work of the secular parts of the military, but only the Gulenists. This is both a shorthand and an assumption. I’ve asked this question of every single Turk I’ve known or met since last July – how much do you think the coup attempt was the work of the Gulenists, and how much do you think it was the work of the remaining secular parts of the military, at least in terms of the relative manpower and organization? I’ve heard a wide range of answers, from 50/50, to nearly entirely Gulenist, though all believed in significant Gulenist involvement. It’s something very hard to determine at this stage, since nobody has any incentive to confess publicly to their role or motivation. Presumably Erdogan and his men know, and they talked almost entirely about the Gulenists, but their motives are probably to blame the more powerful enemy group, regardless of their actual involvement.
Like Vichy France, when it was in power, it had many supporters, but once the Allies invaded, the size of De Gaulle’s forces miraculously multiplied in the ex-post narrative. And few things are less popular than failed coups, so everyone has huge incentives to downplay their involvement ex post. One story that I heard from a few places, which I would believe, is that not all of the soldiers even knew they were taking part in a coup. Their officers told them that they were undertaking a training exercise. This makes sense, for a number of reasons. The Gulenists had infiltrated the military, but probably not enough to overthrow the government if they were relying only on their loyal supporters. So instead, they issue an order and use the general military command structure to get obedience in the early stages, when an overwhelming display of force is necessary. Some of the secular elements of the military would then have to make a quick decision whether to support it or not, without necessarily a ton of co-ordination in advance.
But the pictures of hapless soldiers being dragged out of tanks and lynched fits cleanly with the idea that not all the soldiers knew this was a coup. They started out following orders in what they thought was an exercise, up until the orders were to drive over and shoot the civilian protesters, at which point they balked. However, the protesters thought they were active coup supporters, and so lynched them anyway. Sometimes, there are no good choices, and you just get caught in the middle. Of course, every coup solider is going to claim afterwards that they were just following orders and they thought it was an exercise. I think any honest observer has to conclude that we’ll never really know who made what decision.
But the other hallmark of the coup being Gulenist is that it wasn’t clear at the time who its actual leaders were. This looks a lot like the Gulen movement. When you’re organizing a standard military coup, the leader typically wants to rely on the military chain of command aspect to ensure his support, and also to shore up his position in the subsequent post-coup environment. In other words, you have strong incentives for at least the military plotters, if not the public at large, to know that this is the work of Colonel Gaddafi or whoever, since you’re probably going to be killed anyway if it fails. But if even the coup executors are working for Gulen, whose ultimate authority over the group is relatively secure, it makes more sense to keep the personnel relatively anonymous until after the fact.
So let us assume for the purposes of argument, as I think is probably the case, that this coup was mostly the work of the Gulenists.
This leads to the first astonishing but largely unremarked conclusion, from which everything else must proceed.
This was not, in a conventional sense, a military coup. That is, the main power structures and organizational loyalties used to implement it did not come from the conventional military personnel and chain of command, and the coup plotters did not plan to rule subsequently on behalf of the military (or even segments of it). This was a coup by an entirely different third party, who had infiltrated the military. This is a much rarer phenomenon. Think of all the recent coup attempts, successful and unsuccessful. The vast majority of them had the military as their essential command structure, as evidenced by the fact that the military took over afterwards. Egypt in 2013, Thailand in 2014, Fiji in 2006. Turkey itself, in 1960, 1971, 1980 and 1997. The list goes on and on. Now, how many serious non-military attempts do you know? ISIS, for sure, though this is in the context of a warzone. Islamists in Trinidad in 1990, for instance. But there aren’t nearly as many. Check Wikipedia if you like, and see how many entries explicitly list either the military or someone with a military rank (though these criteria don’t give an exhaustive list of military-run coups).
And not being primarily a military coup is a huge difference. Just think about the relative degree of difficulty involved. In a conventional military coup, the main plotters already have a lot of guns at their disposal, a chain of command that gives them personal authority, and a group of men with ties of personal loyalty to each other and a willingness to fight and sacrifice their lives, even if not for the present cause. Admittedly, there are large difficulties of coordination, getting everyone to agree to move at once. But the military itself starts with a lot of advantages when it comes to overthrowing the government.
Gulen, by contrast, built up the entire organization from scratch, over 30 years. Sure, you can say the Gulen movement failed. But this seems like the equivalent of saying the glass is 5% empty. While true, the astonishing thing is that they came as close as they did. I feel much the same thing about the Silk Road, the illegal online drug marketplace. The surprising thing isn’t that it eventually got shut down. The surprising thing is that it worked as well as it did, for as long as it did.
The first notable attribute, which the Gulenists share with both ISIS and the Trinidadian coup plotters, is religion. In the case of the Gulenists, religion provides a strong advantage. Given the significant risk of its coup plans, of getting killed during its execution, of getting captured and horribly punished afterwards, they needed a very large amount of espirit de corp to inspire a significant number of people to fight for a chance at a new government. Western militaries do without, but for the Gulenists, religion presents a cheap and dirty way of getting people to not only inflict violence, but to obey a chain of command while doing so. It’s not that Islam is the only vector for this kind of mindset. In centuries past, the Catholicism worked just fine for the Gunpowder plot, but it was a very different type of Catholicism than the Pope Francis variety.
But it shouldn’t surprise observers that the Gulenists made use of Islam. They needed some way of getting lots of people to believe strongly enough in an idea such that they were willing to risk everything to make it happen. However, the Gulenists even had an extra advantage, aside from the promise of future payout–this is where the unusual brilliance of the test prep center comes into play. By promising access to jobs, better universities, and influence, the Gulenists could give people a concrete incentive to join now, not just based on the promise of what would come once they implemented regime change.
Admittedly, signing up to be a part of a test prep course doesn’t guarantee that potential Gulenists then sign up to a program of regime change, but with two years of indoctrination, building of group loyalty, and a willingness to follow orders, it ensures a steady supply of new members who are ready and willing to become dedicated Gulenists, without them first having to sign on to the entire ideological program.
The test prep center also served another vital aspect which I haven’t seen remarked on very much, but which I think is quite important. Namely, the organization should be profitable. Not just profitable in expectation of future outcomes, but profitable as an enterprise in the meantime because profitable operations are self-sustaining, and can thus bootstrap themselves to grow with the money they create.
Since the Gulenists ostensibly intended a wholesale replacement of the Turkish superstructure, they needed a steady supply of new members. And they needed them in positions of influence and importance, spread over lots of important power centers – journalists, judges, generals, teachers, the whole lot. Gulen had them, but perhaps not enough, at least ex post (though truthfully the decisive factor was probably coordination, not numbers). But a reasonable minimum here is the number of people purged in the aftermath of the failed coup, which was around 50,000 within a week of the coup, though this was the tail end of the number. I’ve seen estimates of the group’s total membership is in the hundreds of thousands or millions. Recruiting that many people is a challenge.
When thinking about recruitment, there’s a tradeoff to be made between mass appeal and organizational discipline. Specifically, one could motivate lots of people to sign on to a mass ideology – communism, tribal allegiance, Black Lives Matter, whatever – but the very strength of this approach is its weakness, namely decentralization. The more a movement lets anyone sign on by just professing allegiance to an idea, the more its message will spread faster and wider, but the harder it will be to coordinate members and get them to do anything useful.
Viewed from this angle, the Gulen structure combines both a religious aspect and a cult of personality centered around Fethullah Gulen himself. The former ensures a general mass emotional appeal, but the latter ensures a reasonably strong chain of command. It also makes it very difficult for other people within the organization to subvert it for its own purposes, because supporting Gulen himself becomes a strong Schelling point. In military coups in particular, this is always an obstacle – there are typically a number of generals involved in the execution of the coup, and the savvier among them know that most likely only one will end up as the actual ruler, and the others risk getting sidelined or killed even if the coup succeeds. The phrase ‘cult of personality’ has an obvious modern pejorative aspect to it. A simpler version would be to say that it helps if you’re trying to install a king, and the movement is united in advance on who the king is.
The alternative failure mode is the French revolution, where nobody could agree who was in charge, and thus everyone who is currently in charge gets killed by some or other usurper in the organization. With a “cult of personality”, everyone agrees in advance who will be in charge if the coup is successful.
Of course, the big tradeoff of having a king as a coordinating device is that this provides a huge single point of failure. Capture or kill the putative king, and the organization suffers a serious blow, potentially a fatal one. Admittedly, there are costs to a government of killing a populist opposition leader, since it might ignite widespread protests and chaos. But Gulen wasn’t that – there weren’t enough Gulenists to cause carnage in the streets, so killing or capturing Gulen would have been a big problem for the organization, which is why Erdogan kept taunting (or begging rather) Gulen to come back to Turkey, rather than staying the U.S.–hence the strange importance of the leader living in, of all places, Pennsylvania. He is thoroughly out of reach of the government he’s trying to replace.
So while it’s not clear that Fethullah Gulen had to live externally to Turkey, you can see why this has its advantages. And it highlights another aspect to the movement, which I think is essential to the Gulenist project. While Gulen himself was ubiquitous, the identity of the entire rest of the organization was a secret. This seems to have been one of the enormous strengths of the movement. Gulen was channeling Michael Collins with the IRA – if you put out the eyes of the British, by killing all their spies, then they become like a blind giant. This is how a smaller group can overcome a bigger one – the bigger one doesn’t know where to strike. And when the movement has plausibly infiltrated the entire establishment, who do you even lash out at?
Perhaps the most speculative aspect of the coup, which I’ve pieced together from discussions with various Turks over the past few months, involves this aspect. But it fits in with some of the details we can observe, and it’s consistent with at least this account in the New Yorker.
The story goes that the Gulenists had a means of communicating which the government wasn’t able to crack. And as part of this, the government for a long time didn’t know who was and wasn’t a member of the organization. This explains why you read stories about the government being in a conflict with the Gulen movement for several years, while the campaign against them kept grinding on slowly with periodic firings of those known to be supporters. They were in conflict, but it wasn’t always clear who to strike – a lot of the targeted early members seemed to center on police and prosecutors involved in going after the secular military establishment, since the Gulenists had already infiltrated the judicial and law enforcement system. The story goes that Gulen had members even within the Prime Minister’s close staff. This meant that the Gulenists had a large advantage in their ability to stay abreast of Erdogan’s plans.
The key turning point (again, allegedly), came when the government was able to crack the communication system of the Gulenists, around Christmas 2015. This meant that they suddenly had much more of an ability to expose and attack the organization. And as result, the Gulenists knew that the net was constricting around them. The rumors were that Erdogan was planning a big purge of Gulen members from civic society and the military, scheduled to start within a few months of the July coup.
This is all speculation, of course. But there are two big facts consistent with it.
Most importantly, within days of the failure of the coup, the government had fired or suspended 50,000 people, including teachers, judges, prosecutors, and university deans (later rising to 81,000. This is over and above the 35,000 people detained in the aftermath of the coup.
If Erdogan could announce this within days, given the sheer level of confusion and chaos going on in Turkey, and the ambiguity to the outside world as to what exactly was going on, I can only interpret this to mean one thing – he knew in advance whom he wanted to purge. Somehow, he had obtained a significant membership list before the coup took place, and this fits exactly the story above.
And this would also explain a second aspect of the coup, which was commented on somewhat, though the significance was often missed – namely, the coup didn’t seem particularly well executed.
There are a number of ways to show this, but let’s focus on the most obvious – nobody with a brain in their head deliberately begins a coup at 9 p.m. on a Friday, when everybody is already out in the street. This is especially true if the movement you’re leading is not based on mass support, because in that case it’s probably more likely to just incite mass opposition, which is the last thing you want. At a bare minimum, you’d start it at 4 a.m. when the surprise advantage is the largest, the ability to respond quickly is the smallest, and civilians wake up the next morning and find out that the coup is a fait accompli.
From this, it seems pretty obvious that the actual coup attempt we observed was not Plan A, and may not have been Plan B or C either. It strikes me as almost inconceivable, given everything else about the movement. You simply do not slowly plan and build a movement for 30 years and then deliberately kick things off at 9 p.m. on a Friday just because you hadn’t thought about it very much. Yet faced with the likelihood of jail and organizational annihilation anyway, I don’t blame them for thinking it was worth rolling the dice.
Indeed, there is a broader point here too – it’s not even clear that Plan A involved a conventional coup at all, or indeed a violent takeover. What if the plan was just to take over enough important levers of power and then have them all demand that Gulen peacefully take power? Or to come in after Erdogan was out of the way on corruption charges and rig an election to make Gulen win, and them transfer powers to himself? In other words, it’s not clear that the regime change in the originally planned scenario even involved violence. That’s something that’s moot at this point, and we’ll never know. But it’s a point especially important to reactionaries, for whom violent uprisings in the street, and the associated risk of the military having to gun people down, is the epitome of chaos, disorder and bad government that we seek to undo. It would be highly unimaginative to think that the only potential use of an organization like the Gulenists was just to set up and end game where a faction of the military attempted a violent overthrow.
This also suggests that we shouldn’t judge the organizational brilliance of the Gulen movement (from a Machiavellian standpoint, notwithstanding the highly dubious worth of the cause itself), just based on the failure of the coup.