An Information Test For Artificial vs. Natural Selection?

Yes, that is an implication of my argument.

1 Like

That would be a remarkable feat if it were possible. I cannot imagine how it could be accomplished without making use of cheat information (e.g. artifacts left by the humans, like camera footage).

To be Devil’s advocate, imagine if we were to land on Pluto and found a large tractor there. Would we be justified in inferring that it must have been put by humans or some intelligent life forms, instead of emerging from natural processes? (This is inspired by a case described by Plantinga.)


That is a different problem and you know it. Hehe. We are talking about something else here. It also falls into the category of “cheat” variables.

1 Like

It might be possible to take human breeding experiments as the extreme limit of what can be achieved by natural selection on existing genetic variation in a species…
Could this be used as a control to define what natural selection can do even with guidance…

Once that is done, perhaps we could define which phenotypes require novel genes/systems to emerge.

You would have to determine which mutations are part of the design and which are accidents, errors or mistakes. In “Not By Chance” Dr Spetner says that the only changes we can let the blind watchmaker call its own are point mutations, with all other changes in dispute.

With that in mind and understanding how information processing systems work, any loss of function that leads to an adaptation allowing for better survival we grant to the blind watchmaker. For example

I wrote: “If I gave you two organisms could you tell which one had traits or functions that were the product of natural evolution vs. one that had been through multiple rounds of human-managed selection?”

Eric’s claim is that human-managed selection adds a special type or amount of information to the genomes of animals that is not possible from natural selection.

Please give us something current, on the cutting edge of scientific knowledge. Scientific knowledge is advancing so rapidly, that 18 months is usually the life-expectancy of results. A paper a decade old, unless it is highly cited and high impact index, is usually not very remarkable. Realize 1000’s of papers are published in every field of science yearly. Picking one unremarkable on to discuss is kind of a waste of time. (unless there is something remarkable about some aspect of the paper.) Thanks.

We are just talking about human-managed selection v.s. natural selection, not mutagenesis or the origin of genetic variation. Dog breeders don’t irradiate the gonads of their animals with the intention of inducing mutations. They select from whatever variation appears.

1 Like

Natural selection could never produce the breeds of dogs that we observe. Natural selection could never produce the corn that we eat. And natural selection definitely could never produce eukaryotes.

Between natural and sexual selection the norm of a population is kept. Even genetic algorithms, which are alleged to mimic natural selection, use telic processes to solve the problems they were designed to solve.

Natural selection- The process by which in every generation individuals of lower fitness are removed from the population- Mayr "What Evolution Is"

The first step in selection, the production of genetic variation, is almost exclusively a chance phenomenon except that the nature of the changes at a given locus is strongly constrained. Chance also plays an important role even at the second step, the process of elimination of less fit individuals. Chance may be particularly important in the haphazard survival during periods of mass extinction. Ibid

I see that as a problem. How can we determine that all mutations/ genetic changes are chance events? If ID is right then there is at least some control over the genetic changes.

That assertion was not part of the original question. I’d appreciate keeping to the topic of the question I asked.

1 Like

That is always an enforceable request. Keep things on topic.

Interesting diversions will usually be split into a their own thread. It would be better, however, if a new thread was started from the outset.

There wasn’t any assertion

_ “If I gave you two organisms could you tell which one had traits or functions that were the product of natural evolution vs. one that had been through multiple rounds of human-managed selection ?”_

Well you first have to define what you mean by “natural evolution” and then you would have to demonstrate it has a mechanism capable of doing something of note. Something beyond keeping the norm and capable of producing organisms.

Yea, see the all wolves below:



1 Like

That was Argon’s question. I asked Argon to define the terms he/she is using.

That said, blind watchmaker evolution can’t even produce eukaryotes and it has to be given starting populations of prokaryotic-like organisms. If the pictures are of actual living organisms, given the fact that blind watchmaker evolution can’t produce eukaryotes, not one has traits or functions that were the product of blind watchmaker evolution.

Given two organisms, we can only list out their traits. That they arose by natural means is only an assumption. I could give you a Genetically modified plant or animal… and you would still tell me it arose by natural evolution unless proof to the contrary is provided.
There would be more than enough evidence of “common descent” :slight_smile:
Not being able to differentiate between purely natural processes vis a vis the intervention of an intelligence is a sword that cuts both ways…

As a non specialist, all I can say is that it should be important to be able to differentiate between natural processes vis a vis that involving agency/intelligence.


The problem with any breeding protocol is the almost inevitable requirement that the resulting population be reproductively compatible with the original/normal population.

The best example of rate of change… where the resultant forms are mutually incompatible is the marsupial diaspora in Australia.

Genetically quite related… but at least major phenotypes:

Marsupial vegetarian mole.
Marsupial carnivorous Tasmanian devil-like predators.
Marsupial omnivorous Coors.

These are your benchmark exemplars of change from one kind into 3 kinds!

Possible, but again not on topic to Eric’s claim.

I will mention that people have been successfully sued for planting seed with proprietary traits that had been introduced into the seeds’ forebearers by recombinant DNA technology. Those are pretty simple cases when one has a record of exactly what had been done previously (in this case, the sequence of the modified gene that was introduced, along with vector-specific flanking sequences). But again, this is probably distinct from Eric’s proposal.


Perhaps you should start a thread on this…
Do you know if the similarity in phenotype to regular mammals is just skin deep or is there similarity in genes… I read an interesting paper once on the saver tooth tiger version among marsupials being similar to true mammals in some aspects. I will try to dig out that paper.

I understand what you are getting at. I am just showing how it might be important to be able to distinguish between “natural processes” in evolution vis a vis where it has recieved help. I.e even if we can’t currently distinguish, a method is required. And if a method is impossible to develop, there is something too general about how we define “natural” processes. Improper and vague definitions are not good for any academic pursuit.
My comment was obviously not about Eric’s claim.