Sunday, 31 March 2013

On directed mutations

One argument used against cultural evolution is that culture features "directed mutations" - while in the organic realm, the "direction" of mutations is not associated with positive fitness outcomes.

Here's Massimo making that case:

a cardinal tenet of the Modern Synthesis (and indeed of the original Darwinism) is that mutations — the ultimate source of novelty in evolution — are random with respect to fitness outcomes. It is important to understand this point.

This is indeed an idea associated with the modern synthesis. However:

  • It wasn't part of Darwin's thinking;
  • It is not correct.

Darwin may have called variation "spontaneous," "accidental" and "chance" - but he didn't reach the level of dogmatic assertions on this issue later authors did. His descriptions were generally OK.

Mutations are normally deleterious, and are heavily constrained by natural selection. Selection builds error correction and detection mechanisms. If influences mutational hotspots. It acts on chemical concentrations of intracellular mutagens. It acts on mutagenic DNA sequences - such as LINEs and SINEs. Organic mutations prefer junk DNA to coding regions. They also prefer recently-mutated regions. Both patterns bias mutations in the direction of being less maladaptive than "random" mutations would be. Selection acts on mutation rates in many ways - controlling its rate on a species-by-species basis - and a tissue-by-tissue basis. Mutations hit the old more than the young. They hit somatic tissues more than germ-line cells. They hit redundant organs more than unique ones.

The result of all this is that the most deleterious mutations are eliminated, or fixed - and the remaining ones are closer to neutral. This leads to mutations being less deleterious than they would otherwise have been.

In other words, actual mutations are less deleterious than random mutations would be. Mutations are thus directed in favour of positive outcomes.

Steven Pinker says:

Genetic mutations and recombinations are strictly typographical, twiddling the As, Cs, Ts, and Gs with no foreknowledge of their effects on the organism's interactions with the world.

However, this is just wrong. We know that mutations occur in a manner shaped by natural selection to avoid their most negative consequences, while promoting adaptive outcomes - for example via mutations in the immune system.

What is the response to all of this by supporters of the modern synthesis? It turns out that that they radically shift the goalposts:

A genetic mutation is a matter of chance from the evolutionary perspective – or is a matter of "evolutionary chance" – if and only if it is not specifically caused in an (exclusively) adaptive way by a physico-chemical process in response to environmental conditions.
- Evolutionary Chance Mutation: A Defense of the Modern Synthesis’ Consensus View - Francesca Merlin

So: apparently, these days, mutations are a matter of "evolutionary chance" iff they are adaptive!

I think it is pretty clear that the idea of mutations being random with respect to fitness has been lost. That is too ridiculous an idea for anyone to seriously entertain these days. It is contradicted by too much evidence. Instead the goalposts have been dramatically shifted to mutations not being adaptive - on average. That's a pretty high bar.

I should add that I think that the idea of "evolutionary chance" is a farcical idea. Let's leave the concept of "chance" to statisticians - who are clear what they mean by it.

Jerry Coyne offers a quite different defense:

What we mean by “random” is that mutations occur irregardless of whether they would be good for the organism. That is, the chances of an adaptive mutation occurring is not increased if the environment changes in a way that would favor that mutation. The word “random” does not, to evolutionists, mean that every gene has the same chance of mutating, nor that mutation rates can’t be affected by other things. What it means is that mutation is not somehow adjusted so that good mutations crop up just when they would be advantageous
Here mutations must not only occur "non-randomly" - and even being beneficial is not enough - instead, the chance of beneficial mutations arising must be correlated with environmental fluctuations. Again, this is surely a raising of the bar for what it takes to qualify as a "non-random" mutation.

Where does this leave cultural evolution? The claim that mutations are directed in cultural evolution and are undirected in organic evolution is false. Mutations in both cultural and organic evolution are both directed. Differences in the degree of directedness are best regarded as being quantitative - not qualitative.

The other issue that critics raise is that mutations in cultural evolution are much more strongly directed in cultural evolution compared to organic evolution. However, it turns out that most of these critics are comparing memes with genes - using between-host transmission times to calculate the generation time of the memes in question. They ignore - or maybe just don't understand - the idea that memes often have a multi-stage lifespan - which includes multiple rounds of reproduction, recombination and selection inside the brains of their human hosts.

This isn't a fair comparison. If you are compare mutations with multiple rounds of mutation, recombintion and selection, the no wonder the latter process looks like intelligent design coming out out of nowhere. However, that's a topic for another day.

No comments:

Post a Comment