It's amazing to me how often people get tripped up on the behavior of colonial insects like bees and ants. Because all reproduction is done by a queen, it makes little sense in most circumstances to think about the sterile workers as individuals.
A better way is to think of the the workers as body parts. Some of the parallels are pretty direct. The movement of ants is strongly influenced by chemical scent gradients, much like our muscles respond to ion gradients. This puts insects' seemingly-extreme altruism into perspective. Do you consider your skin cells altruistic when they are burned by excessive sun?
If workers in a bee colony are body parts, then why not the queen and drones also? The queen is like a brain in the body. Following on, then why not males/females in an animal tribe? People communicate with each other using subliminal smells indicating sex, anger, fear, etc. Just because animals use 50-50 gene splicing instead of 67-33 doesn't mean the group is any less an organism in itself.
My personal observation suggests some humans (and presumably other animals) are the natural born leaders, most others are like the cytoplasm in a cell, and some others like the outer membrane, showing the most altruistic behavior. Pinker has a valid point that all this shows is some individuals are better at exploiting others tendency to altruism, but my observation is some humans in a group are less motivated to survive and mate than others are. Then there's those who do survive but are hard-wired to mate with their own gender, not producing offspring. How do these two groups ("Gays" and "Losers") consistently show up in animal societies?
The difference between worker bees and cells in a body is that the cells in a body (call them "somatic cells") are 100% genetically identical to the cells that actually do the animal's reproduction - i.e. the cells that produce sperm or eggs. There's an automatic alignment of interests.
Now, in bee colonies there is not this perfect alignment between worker and queen, since the workers are only the queen's daughters, related to her by 50%. There's the potential for various types of conflict, between worker and queen, or between worker and worker...
For example, in many species the workers can in fact reproduce, favouring their own interests, but there's conflict between the workers to disallow this. [1] Naturally this conflict does have costs for the colony as a whole.
It's tempting to think that since a group is bigger than an individual, they should have some sort of power to select beyond the individual's procreation - but while groups are more powerful than individuals, groups are not more powerful than procreation.
It's easy to expect, and believe, and accept group selection, because the method of transmission is so obvious, that behaviours are taught to the next generation! This obscure 'years later, gene expression' stuff just seems so unlikely - humans aren't wired to expect causality over years or decades.
If you find this interesting you might also want to look up Richard Dawkins' demolition[1] of Wilson’s book The Social Conquest
of Earth, in which Wilson (according to Dawkins) promotes a
discredited and confused theory of group selection.
(My interest in evolutionary biology is quite lax, so bear with me if I say something particularly wrong.)
I was under the impression that, in recent years, there had been a small but significant surge of interest in the possibility of epigenetic inheritance, evidence for which had been found in a few studies on chickens, rats and plants. This article seems to explicitly states that no such thing is currently recognized.
Have I been deviated by fringe neo-Lamarckist press releases, are they simply too recent and in need of consolidation, or did Pinker avoid mentioning them for the sake of clarity and fidelity to the most safe, "gcd" theory?
I'm not sure how it's relevant to the issue of group selection and altruism; but anyway such inheritance still seems to be relatively rare, as far as I know.
It's interesting that you call it Lamarckian - the epigenetic inheritance system itself would have to evolve in a normal Darwinian way... e.g. if genes get methylated (turned off semi-permanently) and passed down to offspring in this state, one could call that Lamarckian, but the methylation system itself is in need of a Darwinian explanation.
I assume it is not relevant, indeed: I was merely asking for comments over an unrelated point that was made in the essay, which is that there are no working Lamarckian mechanisms in biology.
I certainly agree over your second point, which I had considered myself, but that did not seem to be particularly relevant to my doubt.
Interesting that Dawkins and Pinkers rebuttals of group selection have gotten so much more press than Wilson's affirmative proposals. Has anyone actually seen a serious proposal of the narrow notion of group-level selection that Pinker is refuting?
The original 2010 Nature paper coauthored by Wilson (http://www.nature.com/nature/journal/v466/n7310/full/nature0...) received a lot of press when it was initially published, and Wilson's back-and-forth with Dawkins and Pinker got just as much press back then. What Pinker is arguing against is not the initial paper (which he does address in this article briefly by repeating the kin fitness argument) but the claims that Wilson is reaching toward in his new book, The Social Conquest of Earth. Wilson's "press" is essentially this book. The "serious proposal" Pinker is refuting is this book, so this article is essentially a preemptive criticism.
A better way is to think of the the workers as body parts. Some of the parallels are pretty direct. The movement of ants is strongly influenced by chemical scent gradients, much like our muscles respond to ion gradients. This puts insects' seemingly-extreme altruism into perspective. Do you consider your skin cells altruistic when they are burned by excessive sun?