Monday, 2 June 2014

The 150 year failure to replicate natural selection with artificial selection


Charles Darwin's Origin of Species has as perhaps its main argument that because artificial selection by animal breeders has made large changes over just a few generations, therefore the natural environment should, over many, many generations, be able to make even larger changes.

In other words, the same process by which animal breeders produce change - by breeding more from stock with the desired traits than from stock with undesired traits - is also happening in nature - but without conscious direction.

This natural selection is just like artificial selection, except that it happens as a consequence of the natural increase of living things leading to competition and differential reproduction linked to traits.


In sum, natural selection is argued to do exactly the same thing as artificial selection - but because NS can operate over many, many generations the amount of change can be much greater; indeed, Darwin suggests that natural selection can - in principle - lead to all types of life: all species.

In sum, natural selection - working by the same mechanism as artificial selection - is suggested tobe able to lead to things such as:

1. The origins of life itself.

2. All forms of life: each and every species, genus, family, order, class, phylum and kingdom.


My point is simple: in 150 years, artificial selection has failed to replicate these two assertions:

1. Artificial selection has failed to originate life.

2.  Artificial selection has failed to do more than originate very low level new forms - new sub-species and species can be made, and new traits; but that had already happened when Darwin was writing - but artificial selection has created no new high level and fundamental forms such as classes, phyla or kingdoms - or to effect transitions between these fundamental forms.


This is, I think, both surprising and a significant failure.

I suspect nearly everybody would have assumed that, long before now, biologists would have been able to make life by artificial selection, and to shape life into new fundamental forms by artificial selection.

But they cannot and have not.

Therefore, we are no further forward than in Darwin's day.


Note: This failure to replicate Natural Selection by artificial selection is not a refutation of the theory of NS - because it is not possible to refute NS - because it is not the kind of theory that can be refuted.

Natural selection is a metaphysical theory that stands outside of biology, and structures and interprets biology - NS cannot be refuted by biology.

Likewise, if artificial selection has been able to create life and new fundamental forms of life, this would not have proven Natural Selection as a whole; but it would have confirmed that the mechanism of selection was capable in principle and in these specific circumstances of doing what Darwin asserted it was capable of doing.

As things stand, artificial selection provides no empirical, scientific evidence that the mechanism of selection in general or Natural Selection in particular really is capable of creating life or changing the fundamental forms of life - not even in principle or in specific situations.

150 years of failure to replicate.

It is astonishing that this big, big negative attracts so little notice!


Second note: There is no doubt that Natural Selection leads to adaptation within species and other low level categories.


Bruce Charlton said...

From an e-mail correspondent:

The basic idea here seems correct and important. I have considered NS for some time to be more of a truism than anything else - there are certain forces that will cause differential reproduction. The claims of 'fitness' usually just come down to 'those who reproduce more given certain selective forces will reproduce more'.

In this sense it's not really a metaphysical assumption - just a truism. Of course there are forces in nature that can cause certain kinds of animals to reproduce more than others, and given these traits are to some extent heritable, that will lead to change of form on a population-level over generations.

Rather, I have thought the interesting part of the neo-Darwinian
synthesis has been the idea that step-wise, random genetic mutations are the main driver of genetic novelty. This, on the other hand, is not only not a truism but has very little evidence behind it, relative to the sweeping claims it makes.

I think this aspect is more the metaphysical assumption - that is, if you assume there is only randomness + natural laws, then you need one or both to get you to the forms we see today.

Therefore, random mutation (in some sense - perhaps also random shuffling of genetic material and so on) as the basis of novelty becomes the only assumption that seems to really work.


My comment - I stick by my assertion that NS is the metaphysical framework for modern biology - insofar as it has one. In fact there is very little real biology in the modern world, as I will be discussing in a forthcoming post - since the mid-twentieth century biology has fragmented into mathematics, physics, chemistry and medicine.

Wm Jas Tychonievich said...

The failure is not at all surprising. Why should we expect to be able to accomplish in 150 years what took natural selection millions of years?

Bruce Charlton said...

@WmJas - There have been over two million generations of some bacteria such as E coli (reproduce about fifty times per day) - and artificial selection - being deliberate and directed - is potentially much, *much* faster than natural selection - e.g. dogs.

Wm Jas Tychonievich said...

The selection aspect of evolution is much faster when it is artificial, but the real source of novelty is mutation, which is not sped up at all by artificial selection.

jgress said...

Actually, we can also speed up mutation through e.g. radiation. Haven't they tried this with fruit-flies? A cornerstone of creationism is the argument that researchers have failed to generate truly beneficial mutations, i.e. mutations that increase fitness and spread through the population, over thousands of generations of fruit-flies.

Bruce Charlton said...

@jgress - yes - but it is not just creationists. The mainstream standard biological position is that Natural selection (plus neutral drift) is the *only* process responsible for the origin of life and species.

Anyone who posits another process - including the various chaos/ complexity or field (Sheldrake) theorists - ideal-type biologists (of the Goethe, D'Arcy Thomson, Brian Goodwin, Stuart Kauffman lineage), those who believe in the reality of forms (Aristotelian types), even dear old SJ Gould (!)...

there are quite a lot of people who regard Natural Selection as obviously real (of course it is real), but not always applicable, and in particular only a *partial* (and probably not the most important) explanation for the emergence and diversity of life.