Saturday 25 July 2015

The meaning of heredity

I generally use the term "heredity" to refer to the transmission of traits from one generation to the next.

However, I notice that some sources differ - confining the idea of "heredity" further. The Encyclopedia Britannica is explicitly DNA-gene centric:here:

heredity: the sum of all biological processes by which particular characteristics are transmitted from parents to their offspring. The concept of heredity encompasses two seemingly paradoxical observations about organisms: the constancy of a species from generation to generation and the variation among individuals within a species. Constancy and variation are actually two sides of the same coin, as becomes clear in the study of genetics. Both aspects of heredity can be explained by genes, the functional units of heritable material that are found within all living cells.
Reference.com confines the idea of heredity even more - to organisms that experience meiosis - with this:

heredity:the transmission of genetic characters from parents to offspring: it is dependent upon the segregation and recombination of genes during meiosis and fertilization and results in the genesis of a new individual similar to others of its kind but exhibiting certain variations resulting from the particular mix of genes and their interactions with the environment.

These sources are simply wrong. However, their mistake is widespread and leads to confusion about cultural evolution. For example, Larry Moran defines evolution using the term 'heredity' - saying: "Evolution is a process that results in heritable changes in a population spread over many generations." - and then goes on to argue that:

I've already alluded to one of the classic questions that a proper definition can answer — the increased height of Europeans over the past five centuries. Armed with a good definition of biological evolution we can focus on one of the key requirements; namely, heritable change. It turns out that the increase in height is due to a better diet and not to genetic changes. Therefore, this is not evolution according to the scientific definition.

This is a serious conceptual mistake. European diet has improved (largely) through memetic evolution. Memes are passed from one generation to the next in cultural evolution - and rather obviously this should meet any sensible scientific definition of evolution - as is attested to by the now-massive literature on cultural evolution. Yet Moran dismisses it - apparently due to his conception of the definition of the term 'heredity'. This seems like a spectacular mess to me.

Larry is not alone in this evolutionary denialism. Here's Mark Ridley on why cultural evolution doesn't qualify as being "evolution":

Changes that take place in human politics, economics, history, technology and even scientific theories are sometimes loosely described as evolutionary. In this sense "evolutionary" means mainly that there has been change over time - and perhaps not in a preordained direction. [...] human ideas and institutions can sometimes spit during their history - but their history does not have such a clear-cut branching tree-like structure as does the history of life. Change and splitting provide two of the main themes in evolutionary theory.

His complaint appears to be that cultural evolution is too reticulated - one of the most dopy objections to cultural evolution ever - once you consider how reticulated bacterial evolution is.

Anyway, these days, almost everyone agrees that cultural evolution is a type of evolution. The remaining debate in the area is mostly over whether cultural evolution is Darwinian. However we obviously have a mopping-up operation to do - as some folk don't even regard cultural evolution as being a type of evolution! That position seems indefensible to me.

Sunday 5 July 2015

In light of cultural evolution

There's a recent trollish article doing the rounds titled:

The Theory Of Evolution Does Not Apply To Modern Human Beings.

Basically it argues that humans don't behave as the theory of evolution dictates. For example, the more resources you give a human, the fewer children they have.

I think this article nicely illustrates the confusion associated with a lack of understanding of cultural evolution. Almost everything in the article makes me think: "yes: but only if you ignore cultural symbionts".

It's well-known that parasites can reduce host reproduction - and even drive hosts extinct. The demographic transition is driven by cultural symbionts that reduce host reproduction. This has been extensively modeled by cultural evolution enthusiasts. This isn't contrary to the theory of evolution - you just have to include the evolution of memes in order to understand it.

If you add resources to a human population the memes absorb the resource faster than the human hosts do - and more memes often means less host reproduction.

Something very similar happens in the organic realm - if you add sugar to a human population. A little sugar might help with reproduction - but beyond a certain point, fertility begins to decline. Instead of making more human genes, the sugar fuels the reproduction of gut microbes at the expense of the genes of the human host. The host's belly swells up to accommodate them all. Eventually the host is effectively sterilized.

This all illustrates the dictum: Nothing in human evolution makes sense - except in the light of cultural evolution.

References

Terence Mckenna - on memes in 1990

Terence Mckenna - on memes in 1990. From a lecture titled: 'opening the doors of creativity'.

Here are some other meme-themed videos by Terence.

Wednesday 1 July 2015

Recombination as a meme repair mechanism

In the organic realm, the idea that recombination can repair bad genes by replacing them with good ones is one of the two main theories that accounts for the origin and maintenance of sexual recombination (the other main theory is the Red Queen hypothesis).

The idea suggests that sexual recombination results in an uneven distribution of deleterious mutations in the offspring - with some having many bad genes and others having few. Those offspring with many bad genes are culled by selection - while those with few get to try again in the next round.

Even random assortment among mating partners would produce this effect. However in practice, the best quality individuals can often seek each other out - and so have offspring with an especially-low mutational load.

The "gene repair" theory of sex has been championed by Richard Michod, among others. He presents the theory in a stimulating popular book, titled: Eros and Evolution.

MIT is in the news today with a souped-up implementation of this idea applied to computer software. They are using recombination to fix crashing programs by using code from working ones. They call their software CodePhage. Here's the associated MIT press release. Their paper is called Automatic Error Elimination by Horizontal Code Transfer across Multiple Applications. It's an interesting case of bio-inspired computing.

I think recombination is a common meme repair technique. If you realize you have a bad meme, it makes sense to find someone with a functioning copy and acquire it. Much the same applies at the level of organizations and institutions. The need to repair dysfunctional memes probably drives a significant quantity of the memetic recombination that we see in the ideosphere.