The comment thread below got me thinking. About 3 years ago I told Randall Parker that I didn't really want to talk too definitively about Islam because I wanted to know more and offer a more thorough and nuanced opinion. Over the years I realized that I was being naive, I'll be dead before I am totally confident about talking about some topics. In regards to Islam my atheism means that I take an instrumental attitude toward knowledge acquisition. I don't think Muslim thinkers are really talking about anything sensible or coherent when they spend 50 pages on tawhid, anymore than I think the Athanasian formula exhibits genuine sense, but I understand that the quasi-concepts that are encapsulated by these words are significant to the people who devote their lives to them. The more I read the more I notice that my ability to discern bullshit is increasing at a far greater rate than my genuine original positive insight. I know what I do not believe more than what I believe.

Consider the following passage from The Coming Anarchy by Robert D. Kaplan:

A recent visit to Azerbaijan made clear to me that Azeri Turks, the world's most secular Shi'ite Muslims, see their cultural identity in terms not of religion but of their Turkic race. The Armenians, likewise, fight the Azeris not because the latter are Muslims but because they are Turks, related to the same Turks who massacred Armenians in 1915. Turkic culture (secular and based on languages employing a Latin script) is battling Iranian culture (religiously militant as defined by Tehran, and wedded to an Arabic script) across the whole swath of Central Asia and the Caucasus. The Armenians are, therefore, natural allies of their fellow Indo-Europeans the Iranians.

When I first read this in 2000 I was dubious about some of the details. A "Turkic vs. Iranian" culture war across Inner Asia? Armenians aligned with their fellow Indo-Europeans, the Iranians? There is obviously a lot of truth here.

1) It is probably true that Azerbaijani Turks are the most secular Shi'ite Muslims in the world. 70 years of Soviet rule will do that to you.

2) I know now that the Azerbaijani government does cultivate relations with Turkey based on common pan-Turkism. Azerbaijan looks to Kemalist Turkey as its ideal.


But there are some serious problems with Kaplan's analysis, some of which I saw then, and some of which I know now. First, there is no "Indo-European" bond between Armenians and Iranians, the common linguistic affinity goes back thousands of years, at least 4-5 thousand years. Iranian languages are closer to the Indo-Aryan tongues of northern India than they are to Armenian. On the other hand, there is a pan-Turkic ethnic identity sealed by a mutually intelligible swath of Turkish languages/dialects stretching from Western China to Turkey. Kaplan is attempting to establish a false dyad between pan-Turkism and non-pan-Turkism. Another major issue is that there are over twice as many Azeris who live in Iran as who live in Azerbaijan. Azerbaijan itself is simply the portion of Iran populated by Azeris which was annexed by the Russian state, the heart of Azeri culture is northwest Iran, in particular the city of Tabriz. So when Kaplan says that Azeri Turks are the most secular Shia in the world, he's wrong. That is, he is simply not telling the truth. I simply can't believe that he doesn't know there are many religious Azeri Turks in Iran, so he was simply trying to pass this paragraph off to an ignorant readership. One of those Iranian Azeri Turks is the current Grand Ayatollah Ali Khamenei! Here Kaplan is trying to emphasize the secularity of the Azeri Turks, but he neglects to point out that the operational leader of the Iranian cultural bloc is an Azeri Turk! Khamenei was Supreme Leader when Kaplan was writing, so he is either deceiving or ignorant. Additionally, if you do the math you see that over 1/4 of the Iranian population is Turkic. In Sons of the Conquerors by Hugh Pope the author notes that Azeri Turks continue to maintain their traditional roles in the military of the Iranian state, and are well represented in the elites from the Supreme Leader on down to the business class. The Iranian nation may at its core be Persian, but Persian speakers form only 60% of the population of Iran.

On this particular topic, the role of Azeris within Iran, I am especially motivated because it seems to me that people are playing shell games with an ignorant public due to ideological motives. I am particularly skeptical of the idea promoted by a few neoconservatives before the 2006 Congressional elections that the ethnic divisions in Iran could be used to fracture the nation. I criticized GNXP reader Jeff Boulier in sharp terms for keeping open the possibility that Azeris could be used to foment divisions in Iran. In our last round Jeff responded:

That doesn't, of course, mean that my opinions are not warped by my priors. I think my opinion is less likely shaped by tribal identification with my political group but, well, my beliefs about the willingness of people to engage in tribal behavior.

Jeff alludes to the fact that I accused him of accepting particular beliefs because his own political tribe espoused them. This is no sin that any of us are free from of course. But Jeff's point about the willingness of people to engage in tribal behavior is a fair one. I agree with the rough theoretical model:

* Humans are socially conformist and are easily induced into tribal behavior based on trivial differences

It follows that tribal behavior based on non-trivial differences (language, religion, etc.) is also common. Tribal behavior can also erupt rather quickly from a serene milieu. Yugoslavia is perhaps a perfect example of this. So why do I object to Jeff's high rating of the plausibility of fomenting division within Iran? Because though I accept his theoretical premise, in the context of international relations or historical events these general truths have enormous error bars. Given enough iterations these truths will beat random expectation, but in any given situation they are very likely to be wrong. And, when you add in specific contextualizing facts a very different inference can emerge. Consider the following....

a) a person's grandparent is Polish

...now, consider the following variable

b) their name is Levy

You are probably savvy enough to know that there is a big difference between Polish Jews and Polish non-Jews. Just saying that someone's ancestor is from Poland might give you a general theoretical model of how they would behave and view the world, but specifying the name makes crisp specific inferences and reduces your error.

In the case of the Azeris of Iran, they are an ethnic minority in a nation where the dominant culture is that of the Persian majority. From this one might conclude that they are resentful of their place. The model of tribalism as a natural property of humans which can be exploited in international relations suggests that inducing division might bear fruit. But, add some more information into the model. Azeris are dominant within the military. The most powerful individual in the nation is an ethnic Azeri. The previous president was of part-Azeri heritage, and Azeris are a prosperous community. Differences of language and ethnic identity still remain, but when you look closer you see that those differences do not take succor from a ground of resentment, segregation and marginalization. As I have pointed out before Turks have ruled Iran for most of the past 500 years (and much of the previous 500). Azeri Turks founded the modern Iranian Shia state during 16th century. I believe these details render the plausibility of a short term expectation of ethnic conflict in Iran to be nearly zero. This does not mean that long term chances of conflict are trivial, but the key is that proximately for public policy considerations this should be taken off the table. Now, the question is whether those who propose this solution actually believed that such division was ever possible, or whether they were employing a coarse theoretical framework because they knew no better. I suspect the former because it seems to me that there were ideological considerations at play, and the polemicists knew that no one was going to check their facts and call bullshit on them.

So back to giving a "nuanced" picture of how the world works. I think that all things are possible under heaven sociologically. Just because today homosexuality is not considered acceptable within Islam does not mean I believe in the future it won't be. Nevertheless, we shouldn't expect a 5 year transformation. Similarly, just because ethnic conflict in Iran is not likely within the next few years does not mean that in a generation such a conflict won't emerge. Just because today the Salafi terrorists call takfir (i.e., non-Muslim) on the Shia does not mean that in the future differences will be set aside because of a "more accurate" interpretation of the Islamic religion. And finally, though evangelical Protestants might naturally be averse to the Mormon Mitt Romney purely on theological grounds in this election cycle, this does not mean that within a generation they might not be as accepting of Mormons as they are of conservative Catholics in the present. It takes time. The key is to not assume that people have changed their minds, it is to remember that with enough time passed they will have forgotten that they ever believed otherwise. You see, white Americans in the southern United States were never racist in the 1960s, it was only their neighbors. The red-hot evangelical fervor for the pro-life movement is simply a fundamental part of their Christianity, forget the fact that Roman Catholic organizations were frustrated at the relative lack of interest from Protestant groups in the anti-abortion in the wake of Roe vs. Wade. Suicide bombers are not suicides, but martyrs. Citizens of a hostile state are not civilians but soldiers by virtue of voluntary participation in the polity. And so on. Self-delusion takes time. Until time can work we must always keep in mind the pattern of delusion.