Homosexuality in giraffes and other non-human animals

>Suppose it is beneficial to carry the "gay gene", because it means that more of your descendants don't die, even if some of them won't procreate

That doesn't work because even though less of your gay descendants will starve, the straight organisms in the population will also reap the benefits and they'll still be at an evolutionary advantage.

Yeah, but most social sciences are nowhere near the standards that physics imposes on what it holds to be true.

That's not a flack. It's inherent to the nature of the social sciences. Since every individual is part of a complex social system, it's incredibly hard to make repeatable experiments to test assumptions in the same way that physicists do every day for breakfast.

The comment I responded to contrasts "innate" phenomena with "a choice" or "something you can catch". I think the argument is very strong that homosexuality (as an inclination) is not "a choice" -- people don't decide on what they want. And 30% concordance for identical twins is much, much higher than the concordance for random strangers drawn from the same broad demographic pool, so it's clear that there is genetic influence in a sense.

But it's also clear that there is not genetic influence in the ordinary sense that most people will imagine if you say a trait is genetic. Compare the question "is this person gay?" to "does this person speak French?" --

Can we predict whether a person speaks French just by looking at their DNA? Yes.

How good will the prediction be? Decent. French is mostly spoken by people in countries where French is the language of daily life. There is noise from a few sources: some people are naturally attracted to learning languages; some countries are more recent than others and the genetics of their residents are less informative; some countries have an imperfect layer of French on top of some other language(s). (You can easily identify a Frenchman by looking at his genes. It's harder to identify a French-speaking Canadian in Quebec that way. And it's much harder to identify a Vietnamese collaborator family that way.)

There is a genetic influence (in the intuitive sense) on whether you speak French. I mentioned it above -- some people are inclined towards learning languages. Within any given non-Francophone population, those people are more likely to speak French than everyone else is. But the very weak genetic signal amounts to nearly nothing because it is overwhelmed by very powerful environmental influence. If your parents speak only French, and everyone you ever meet speaks French, you will speak French too. And if none of the above is true, you're unlikely to speak French even if you might like to do so in the abstract.

So the fact that you can predict French-speaking from a person's genetics is mostly spurious -- it comes from the fact that genetics and French-speaking coincide without much direct influence of one on the other. The concordance of French-speaking in identical twins is much higher than 30% -- it is nearly 100%, despite the fact that the trait is not much influenced by genetics, because identical twins almost always grow up in the same country. And this is for an entirely non-genetic trait! Where a trait is genetically determined, you would see similar concordance levels -- since identical twins have the same genetics, their concordance on genetically-determined traits must be close to 100%. Studies of identical twins raised apart always find that a twin pair coincides on a creepily large number of apparently inconsequential behaviors, like habitually wearing seven rings. They should coincide more on very consequential behaviors such as refusal to breed.

30% concordance is so low as to preclude the idea that genetics is causing the one twin to be homosexual. It suggests instead that genetics is giving both twins a weakness, but a minor enough weakness that even when one twin succumbs the other twin usually stays healthy. There are two models that could match this:

- Homosexuality might strike at random, like radioactive decay. Genetics set the odds of it happening to you, and every day God rolls the dice to see whether you suddenly become gay.

- Homosexuality might be transmissible ("something you can catch"). Genetics set the odds of your successfully resisting infection.

The first option could reasonably be described as "innate". It cannot, however, be described as "born that way", which as I understand it is the argument that people are making when they say that homosexuality is innate. 30% concordance in identical twins conclusively demonstrates that homosexuals are not born that way. An adult's homosexuality is not specified by their genes -- it is a failure state to which their genes left them more or less vulnerable, and which needs some kind of trigger (possibly chance).

Summing up:

We were given three options: that homosexuality is "a choice", "innate", or "something you can catch".

As you note, "innate" is not particularly well defined. Neither is "a choice". But we can still reject the "choice" option pretty easily. From a common sense perspective, it is clear that people don't choose their wants. And the MZ twin concordance rate is too low -- where identical twins get to make choices, they generally make the same choices.

While "innate" can mean several things, it is very often used to mean an "intentional" effect of the genes, a body that faithfully executes its genetic plan (regardless of whether that plan is good or bad). (This isn't well-defined either, sadly.) The MZ twin concordance rate is much too low to support this for homosexuality. It is also used to mean an effect which will manifest regardless of circumstance. This is easy to reject, as there are no such phenomena. But with a more liberal interpretation of "regardless of circumstance", we can rule this out too.

I don't think anything discussed here is evidence for or against the idea that homosexuality is "something you can catch".

I will also note that this trichotomy fails to include some obvious types of causes. If, say, a rock falls on your arm and you lose the arm, I think that's better characterized as an "environmental shock" than as a disease, a genetic effect, an "innate" effect, or a choice, even though there is genetic and "choice" influence on the event. I tend to agree that 30% MZ twin concordance is too high for the environmental shock model to make sense in the US context.

But as I said above, the evidence in favor of "innate" as an explanation is not very good.

Haplodiploidy. Specifically, the relatedness section on this Wikipedia article:

https://en.wikipedia.org/wiki/Haplodiploidy

> It's a gene that, in the long term, slightly increases the chance of every other individual reproducing while immediately greatly decreasing its own chance. That just doesn't work.

Don't forget that all those other individuals in vicinity carry some of the same genes - the closer, the more genetically related they are, in general. You'd need to compute how much that small but broad effect can add up to the total percentage of genes propagated indirectly.

I get that it doesn't sound very "common sense", but I don't think it can be rejected outright without doing the numbers and modeling first. Which is exactly what people in the field do, and if they say that it's plausible (i.e. that they can come up with models that are consistent with observations, and that demonstrate stable propagation), then it can't be dismissed out of hand. Look up "group selection" for the kind of models that they use.

Also, that Wikipedia article gives some far more extreme examples of biological altruism, like self-sacrificing ants - noting that this is still evolutionary viable because of how much shared genetic material is there in the rest of the colony (indeed, one could say that the very social structure of ant colonies with non-reproducing workers is an extreme example of biological altruism taken to 11 by evolution). And then we have our own social mores, which have included literal self-sacrifice for the sake of one's "tribe" as a virtue for pretty much as long as we can trace human culture back - and it's not on its way out, either. Clearly, we aren't ants, but in this case, is it really a qualitative difference, or simply the result of the same process applied to different starting conditions and different evolutionary pressures, producing different points on the same scale?

Side note: when I was digging into modern evolutionary theory, one surprising (to me) thing that came up is that there have been substantial revisions in the estimated speed of evolution as a process since Darwin's days. Sexual selection in particular is a mechanism that turned out to work much faster than originally anticipated, and can in fact push other traits that wouldn't be viable otherwise over the viability threshold. One theory is that it developed in the first place because faster selection - i.e. adaptability - is itself a selected-for trait, so it's kinda getting very meta. And there's a fringe but intriguing argument that our human culture (and its building blocks, like language) is, essentially, a further development along these lines to reduce the lag even further - memes can change faster than genes, so population groups that rely less on genetically hardcoded behavior and more on transmitted social mores, are more adaptable. But of course the genes are still there, so those transmitted behaviors that can be implemented on top of hardwired stuff are going to be stronger, even if the original effect was very subtle - e.g. modern concepts of fairness and justice (presumably evolved from hardwired behaviors like parochial altruism).