Monday, 29 September 2014

Criticisms of cultural kin selection

Cultural kin selection has faced some criticism. I'll try and track some of the most common objections on this page:

  1. Humans are too good at tracking relationships

    In 2008, Boyd and Richerson expressed frank incredulity at the idea that humans are being fooled into thinking that non-relative are relatives - and so behaving nicely towards them. They wrote:

    Living primates are very good at discriminating between relatives and non-relatives and behave very differently toward each. It is hard to see why early hominids should have been less discriminating in their behavior.

    I don't think this is a particularly challenging puzzle. Most primates don't live with rapidly-evolving cultural symbionts who need to manipulate them into coming into peaceful contact with other members of their own species in order to allow the symbionts to reproduce.

    Additionally, memes don't just regard other humans as potential homes of their own future offspring. Other humans are often existing containers for their own offspring, parents and siblings. Memes use the kin-detection mechanisms in human hosts to preserve copies of themselves in other bodies via host manipulation.

    It is these cultural symbionts - which humans have and most other animals don't - that means that the human kin-recognition mechanisms are so frequently the target of successful manipulative attacks. Essentially: the cultural symbionts have short generation times, evolve rapidly and actively seek out the holes in the host kin-recognition psychology.

    Anthropologists have long recognized that kin categories are indeed influenced by culture. For example, they have distinguished between "biological kinship" and "social kinship" (Hawkes, 1983) and between "natural kin" and "nurtural kin" (Watson, 1983). That kinship relationships can be significantly influenced by culture is really a commonplace fact these days.

    Also, other species are not completely immune from this sort of kinship-based manipulation. For example, cuckoo hosts are regularly fooled into thinking that cuckoo chicks are their kin - and into providing resources for them. Notice that active manipulation by a symbiont is also involved in this case. In the case of cuckoos, succeess comes to them not because their genes evolve rapidly compared to their host - but because they are relatively rare - and so are not worth defending better against.

    Lastly, I think that there's a bit of a straw man in the framing of this objection. In cultural kin selection, people are not normally literally fooled into thinking that non-relative are really relatives. Instead they are fed sensory sitmulii that act as a superstimulus to kin detection routines in their unconscious minds. If you ask brother Mark whether brother John is a blood relative, he will probably give the correct answer. However this doesn't mean that their shared memes and shared monastic robes aren't relevant to the extent of their cooperation. Manipulation can take place unconsciously. Also: relatedness isn't a binary quantity; there are degrees of relatedness. Memes can and do massively increase levels of perceived relatedness between their human hosts - and it makes sense that they do this partly to increase their own inclusive fitness.

  2. Relatedness is hard to estimate in cultural evolution

    Here is Peter Richerson in 2010:

    In the case of culture, the analog of kinship is very hard to estimate. Having two parents with equal genetic contribution makes the calculation of relatedness easy. In cultural transmission, one, two, a few, or many people in your social network are possible sources of culture. People may use different parts of their network for different cultural domains. No one has proposed a way to estimate cultural relatedness in the face of such problems.

    I have previously worked through this objection in my article on cultural kin selection - in the section titled "memetic relatedness". To recap: it is not true that no one has proposed a way to measure cultural relatedness. Also, relatedness between two humans is one problem, and relatedness between two artifacts or two messages are different problems. The latter problems are significantly more tractable. Cultural information spends some of its time inside brains and some of its time moving between brains - and during these "external transmission" phases, it is often much easier to quantify it.

    It is often harder to measure relatedness in cultural evolution - due to the lack of meiosis. However, this is not a show-stopping problem. Not all creatures in the organic realm feature meiosis in the first place - yet they still have relatives and must allocate resources between themselves and their offspring and parents. The theories associated with kin selection still apply in these cases.

    Cultural relatedness is often easy to calculate. For example, Alice's dollar bill is related by around 100% to Bob's dollar bill, and around 0% to Charlie's Japanese yen. Also, having some relatednesses that are difficult to quantify is not a problem unique to cultural evolution. For example, in organic evolution, it isn't easy to measure the relatedness to two Portugese man o 'war individuals. The fact that they are a symbiotic conglomerate is a complication - but not a show stopping problem.

  3. Maximization of inclusive fitness may not apply "commonly"

    In 2014, Dan Sperber (and coauthors) wrote:

    How deep is the analogy between biological and cultural evolution? Memetics assumes that it is deep indeed; that the main relevant details of the biological case have direct equivalents in the cultural case, such that there is, for example, a cultural phenotype, which achieves a certain level of (inclusive) fitness, which will in turn determine the phenotype’s relative success in the population.
    This is good so far, though I would identify this as a conclusion from many decades of observations - rather than an "assumption". The authors go on to say:

    Darwinian selection leads to the maximization of inclusive fitness, and this explains the appearance of design in the natural world. Is there an analogous result for cultural attraction? As selection is a special case of attraction, design is possible and in some cases explicable in standard Darwinian terms. Having said that, such explanations will not apply generally, and may not even apply commonly.

    The concept of "inclusive fitness" is a simplified model of kin selection which no-one believes applies generally. However, kin selection is a very generally-applicable idea - it is a consequence of natural selection itself in structured populations.

    As for the generality of Darwinism itself - it all depends on what you mean by the term. I think most accept that evolutionary theory has moved on a bit since Darwin's era - with the incorporation of symbiosis, game theory and an understanding of self-organizing systems. However, many still use the term "Darwinism" for the resulting evolutionary theory - as a way of giving Darwin credit for coming up with the basic idea in the first place. Ultimately, this is a terminological issue.

    Personally, I don't like Sperber's various terminology proposals. In fact, I think that they suck. "Attraction" is a basic concept in dynamical systems theory. If you want to redefine it, you had better have a good case. Sperber doesn't present such a case; he doesn't have one. Overloading the term with multiple similar meanings is not an attractive option.

  4. Criticsms of group selection - and cultural group selection

    Kin selection and group selection are broadly equivalent - so many criticisms of group selection also apply to kin selection - and many criticisms of cultural group selection also apply to cultural kin selection.

    For example, Max M. Krasnow & Andrew W. Delton argue that there is no evidence for group selection in humans - in: "Is there evidence for special design of a group-selected psychology". They say:

    The debate is not about selfishness versus generosity or individualism versus groupishness. The debate is about whether generosity, cooperation, altruism, etc. are instantiated by a psychology designed by individual or group selection. If the former, then this psychology should have design features that, on average and under conditions that match ancestral conditions, eventually lead to net benefits for the individuals or their kin.

    However, this is a bad framing of the problem. Modern versions of kin selection and group selection are equivalent - and make the same predictions. If you argue against group selection on the grounds that kin selection is responsible, you haven't understood this - and so can't really usefully contribute to the discussion.

    Another example is Steven Pinker. He criticizes group selection, cultural group selection and cultural evolution in The False Allure of Group Selection. Pinker's arguments vary from being reasonable to being wrong. Overall, he has no coherent case against group selection, cultural group selection and cultural evolution - assuming the group selection variants are interpreted as being equivalent to the corresponding kin selection theories.

    Pinker's article is remarkable in including a lot of commentary - which illustrates how much confusion surrounds these topics on all sides.

  5. It all boils down to genes

    Wilson-style sociobiology - and most evolutionary psychology - has long held the opinion that in biology, it all boils down to DNA genes. That DNA genes are the only medium of inheritance in evolution that matters. Memes are ignored - and all advantage is ascribed and attributed to DNA genes. From the perspective of memetics, this position seems ridiculous - but it seems to be a remarkably mainstream position.

    In the context of cultural kin selection, uniforms become a way for humans to manipulate other humans. Holy fathers and mother superiors become ways for church leaders to manipulate their flock. Rather than memes manipulating humans for their own benefit, humans are seen as using memes to manipulate other humans - for the benefit of their own DNA genes.

    There's obviously some truth to this perspective. Human genes do often benefit from the manipulations described by cultural kin selection. Military leaders do benefit if their army behaves as though it is a big family.

    The problem arises when advantage to memes is completely neglected. This tends to result in an impoverished perspective on the evolutionary process, which loses much of its explanatory power. Those who think that only genes matter are missing half of the picture.

Recent criticisms of kin selection itself deserve a mention here. Group selection advocates have long been critical of kin selection - arguing that kinship is only one of many ways in which organisms can come to share features, that group selection is more general, that inclusive fitness is a awkward, artificial concept, that kin selection is too hard to apply - and so on.

In some cases widespread criticism from the scientific community pointing out that some of the more extreme critics are plainly off their rockers has helped with some of this.

Widespread recognition that kin selection and group selection are broadly equivalent concepts and represent different perspectives on the same kind of phenomena should help to damp down some more of these criticisms.

However, kin selection has had its flaws. The widespread failure to apply kin selection to cultural phenomena is one of these. This failure is explicitly mentioned by some critics - e.g. here. However, that is not so much a problem with kin selection, as it is a problem with the scientific failure to get to grips with cultural evolution. Anthropologists - whose job largely involves studying human culture - are mostly-living in a pre-Darwinism timewarp - in which evolutionary theory is not applied to their subject matter. The way to deal with this is not to use kin selection less, but to use cultural kin selection more.

While kin selection is not without issues, these seem minor into comparison with the problems with group selection. Group selection seems to be a fountain of junk science. I think it should come with clear health warnings - and, generally speaking, it does do so.

Anyway, this probably isn't the best place to review the whole kin selection vs group selection debate. I do have another blog where that is one of the main topics.

Kin selection has been a tremendously useful and productive tool in the biological sciences, and there is absolutely no reason why it can't be similarly useful and productive in cultural evolution - and in the social sciences generally. At this stage, scientists just need to pull their fingers out and start applying it. Hopefully, my reviews of the topic have indicated the abundance of low-hanging fruit in this area.

No comments:

Post a Comment