Doesn't Natural Selection lead to Specified Complexity?

quote:

---

By 'drift' do you mean 'neutral genetic change'?

​

In order to determine the dererminism (or not) of natural selection one has to 'fix' the inputs ... so drift becomes irrelevant.

---

Except that it is not an input, so you can't fix it. It happens and affects the outputs so it isn't irrelevant. So where does it fit into your view ?

quote:

---

If we narrow the environment to a petry dish, and the gene set to a single (at first) bacterium I'm pretty sure we could 'rig' the environment to get a particular genetic change to fix in out petrie dish population.

---

Assuming that the gene was in the population, and you could rig the environment to give the particular allele you wanted a strong selective advantage then you could do that. But being able to rig things is not enough, especially if you need special conditions to do it.
In a more hostile environment, even that might not be enough.

The way I look at it is:Inputs { Genetic Mutation, Environmental Conditions} -> Process {Natural Selection} -> Output{Genome Distribution Shift}If not all of the genetic mutations are 'used' by natural selection, then those are not relevant to natural selection -- but that is, of course, a function of the environment. I could put my view as:Genome Distribution Within a Population = Natural_Selection(Genetic Mutation, Environment) + Drift(Genetic Mutation)Making drift independent of natural selection as a process/function.As for the bacterium thing, the assumption is not necessarily that the gene is there, but that there is a finite probablility of that gene entering the genome (i.e the original bacterial cell doesn't need the resistance provided one of it's successors does -- assuming it lives long enough to reproduce of course).

quote:

---

Doesn't that imply that specified complexity just means 'intended outcome' ... and backs round to not helping at all in detecting 'intelligence'?

---

There's a bit more than that. It's something that seems intended even after you consider all the other possible explanations. And by "consider" I mean calculating the probability that the explanation would give that outcome, and find out that it is mind-bogglingly improbable.

quote:

---

If we observe a pattern after-the-fact in which natural selection 'formed' the pattern or an 'intelligence' formed the pattern what difference would there be from Dembski's view point -- nothing I would suggest.

---

That would depend on the outcome of the probability calculations. Using Dembski's method you are supposed to eliminate natural selection as a possible explanation by showing that it is way too unlikely to produce the observed pattern.

quote:

---

Genome Distribution Within a Population = Natural_Selection(Genetic Mutation, Environment) + Drift(Genetic Mutation)

​

Making drift independent of natural selection as a process/function.

---

But ignoring the facts that drift is also a product of the environment (as you have defined it) and also that the genetic distribution is part of the input to both natural selection and drift. This is an iterated process and the results feed back to influence events.

quote:

---

As for the bacterium thing, the assumption is not necessarily that the gene is there, but that there is a finite probablility of that gene entering the genome (i.e the original bacterial cell doesn't need the resistance provided one of it's successors does -- assuming it lives long enough to reproduce of course).

---

In which case the result is not guaranteed. If the mutation does not occur, it cannot be selected.

Drift is not a function of the environment (if I understand the term correctly) since the prevalence of that 'trait' is not dependent on environment (if it were it would be bound up in the natural selection).Yes there is a time component such that:GDiP[k] = NS(Env, Mut)[k-1] + D(Mut)[k-1]k being a 'snapshot' in time.Drift affects evolution, but is not a factor in natural selection ... and it's the natural selection that generates 'specificity'.If a natural process can produce a pattern, then specified complexity is not a marker for design ... no matter what you consider complexity to be.As for the bacterium: the result is not guaranteed becasue one of the inputs to natural selection cannot be controlled, not because natural selection is non-deterministic.The idea was to narrow the vast array of variables that are inputs to the process to a manageable sub-set.Given a mutation that we know can occur, and enough pretie dishes we will see the result.And could repeat the experiment and get the same result.... but then if that were not the case the whole concept of natural selection would fall by the way-side.

So .... this whole thing is just an argument from incredulity then.One consequence of evolution as a natural process is that no particular outcome is actually intended. In fact the whole idea of 'intent' is equivalent to 'intelligent design' ... so doesn't that mean that this whole idea is attempting to say 'If it looks intelligently designed, it was.'Rather than being meaningful in any way about detecting intelligence in a design.

Yes and no. There are some mutations that make no changes in function or phenotype (e.g. synonymous mutations). There are also mutations that produce changes in phenotype but such changes do not result in a change in fitness. In the latter case the environment does play a role in determining selective pressure and the prevalence of a trait. OR . . . . evolution is a designer. That always confuses the ID supporters.

I think you mean that drift is not a function of the environment as it is usually defined. Unfortunately, that is not the case when using your definition.

How so?

Yes, meant 'not a marker for INTELLIGENT design'.In the latter case that you mentioned, where a phenotype change does not alter fitness -- but selective pressure still applies ... that seems a contradiction.If selective pressures are placed upon the new phenotype, then natural selection is at play, but does not favor the new phenotype over the pre-existining one (relevant to the mutation).Bu in respect of THAT genetic mutation selection remains a factor.

Because you define the environment as everything that happens. Thus any event which affects reproductive success is a part of the environment as you define it. Even if it is a freak accident -which would be excluded under the more usual understanding.

But how does that make drift a part of the environment?Drift isn't an event.Drift is a kind-of hap-hazard fixing of a genetic change within a population for no specific reason.Changes in behaviour may have an impact on subsequent changes ... so it may result in different types of event in the subsequent 'round', but that doesn't make it an event.If we were to view natural selection (as I do) as a process which acts iteratively on a set of environmental effects & a set of genetic mutations (within a population of interest) then that process must be deterministic, else it is not a process at all.It's like a computer function that takes a rand() as one of it's arguments ... seed your random number system the same each step and all your results will be the same -- change the seed ro something else (like date) and your results will be different.The variability in the inputs doesn't make the function non-deterministic.

What you mean to ask is "how does the fact that the events that produce drift are part of the environment mean that the environment is a factor in causing drift". Which pretty much answers itself.

quote:

---

If we were to view natural selection (as I do) as a process which acts iteratively on a set of environmental effects & a set of genetic mutations (within a population of interest) then that process must be deterministic, else it is not a process at all.

---

Then, it seems the problem is in your view of what a process must be. Perhaps dropping the arbitrary demand of determinism would be better than adopting non-standard definitions that bring more problems.

You need to carefully parse what I said. I stated that the environment determines which changes are under selective pressure. The environment determines which mutations are under negative, positive, or neutral selection. I am relating all of this back to a previous post where you stated:"Drift is not a function of the environment (if I understand the term correctly) since the prevalence of that 'trait' is not dependent on environment (if it were it would be bound up in the natural selection)."The environment does determine which mutations are neutral, and therefore determines the probability of that trait being passed on to the next generation.

Drift is a mathematical model which describes the spread of mutations that do not affect fitness in a given environment. Which mutations are neutral is a function of the environment just like the environment determines which mutations are beneficial and which are detrimental. Natural selection is never deterministic. It is always probabilistic. The game of craps is a good analogy. You have the highest probability of rolling a 7 on any given roll, but this doesn't mean that you will roll a 2 or 12 occasionally. The same for selection. Sometimes beneficial mutations are not passed on, and other times detrimental mutations are passed on. For beneficial mutations the odds are in their favor while the odds are against detrimental mutations. For neutral mutations there are even odds. All mutations fall somewhere in this spectrum.