Is this even a thing?
In a Telugu film I watched recently, the protagonist ended up marrying his first cousin. I was somewhat surprised by this, since I’d always sort of assumed that this was a cultural taboo worldwide. After all, shared DNA of any form creates a higher incidence of recessive genetic disorders… right?
My gut reaction was based off of a few things. I knew that first cousin marriage is just straight up banned in many states in the USA (humans commonly, if unconsciously, associate laws with morality, even though the two are not always formally related). I was also familiar with examples that came up in my AP Biology class about the dangers of limited genetic pools (polydactyly among the Amish, for example. See: biology stack exchange). And of course, incest is something so strongly abhorred culturally that anything that invokes incest associations is going to be more than a little bit taboo.
But I decided to do some research, since my intercultural communications class in college taught me that assuming is dangerous. It’s not even a right or wrong thing: once you realize the hugely different values and perceptions that spring up among different cultures, you become rightly suspicious of social generalizations in general, even those made without malice or intent.
Frequency and incidence worldwide
I found the Wikipedia article on this subject fairly complete. Cousin marriage (and especially first cousin marriage) is something that is heavily culturally dependent, just as I had suspected.
One can come up with some pretty good reasons why cousin marriage would be economically desirable. As commonly practiced, cousin marriage allows for property to be kept within a single extended family group. Especially in vertical-collectivist societies without a lot of social mobility, power centralization very much revolves around family structures and strategic marriages. Dowry expectations would also play a role here, and would serve to explain why the practice is still in place in some parts of the world despite taboos and genetic factors. If dowry amounts exceed the means of a certain economic group, said group may turn to cousin marriage as a logical solution to their problem.
What is interesting is that in unilineal descent societies (i.e., societies that practice matrilineal or patrilineal descent), parallel cousins (see parallel and cross cousins) on one or both sides will belong to one’s own descent group, while cross cousins will not. If the society also practices descent group exogamy (an expectation to marry outside of one’s descent group), then some cousins are OK to marry but others are not. This is somewhat puzzling for us Westerners to think about, since it seems so arbitrary. If they aren’t any closer genetically speaking, why would it be “more taboo” to marry one cousin than another?
Genetic factors
In the genetics section, the Wikipedia article cites research that estimated the average risk of birth defects in a child born of first cousins at 1.1–2.0 percentage points over an average base risk for non-cousin couples of 3%, which is about the same increased risk that a child born of a woman over the age of 40 faces. This surprised me, since I thought the effects would be more pronounced. Based on this statistic alone, it seems difficult to uniformly condemn first cousin marriage from a genetics point of view. It is an additional risk factor, to be sure, but not one so overwhelming that it is easy to generalize on a moral level.
However, one cannot simply parrot this statistic without noting its context. Due to the founder effect and the fact that genetic drift has larger effects with smaller allele bases, any time one social group isolates itself, first cousins from this social group will actually be much closer to each other genetically than first cousins from the general population. For example, if first cousin marriage is repeated across several generations, the genetic risks go up significantly.
This is a crucial observation since some cultural groups that practice first-cousin marriage do it with enough frequency that it is not possible to assume roughly population-level base genetic diversity. For example, the British Pakistanis.
The odds for related cousins without shared DNA
I read an interesting article during the course of my Googling about exactly how much DNA close cousins share. The article bluntly asserts that no second cousins or closer have ever come forward that do not share a measurable amount of DNA. The standards to show a genealogical relationship without shared DNA (presented in the article) are quite rigorous. Essentially, you have to show that the cousins in questions do not share DNA with each other, but do share DNA with their respective parents at an expected level, and that their (related) parents also share DNA with each other at an expected level.
Given that there are somewhere around 7.5 billion people in the world at time of writing, it is not impossible that some people somewhere fit the bill. But in general, it is never safe to assume that related cousins are genetically different enough to produce offspring without any additional risk of genetic defects.