Quibble: What you want here is a Poisson distribution, where the sample mean estimates a rate. 
Behe calculates the rate at which resistance arises in a chloroquine- treated population. See the difference?
Of course no one is looking for this resistance if chloroquine is not involved, but that same mutation can still occur. Without the presence of chloroquine this mutation is unlikely to fix in a population, so these events are censored (we never see them).
For what purpose? The frequency with which chloroquine resistance evolves is informative only about the frequency with which chloroquine resistance evolves. It doesn’t tell you the frequency with which anything else evolves.
I do think the events are independent, are you saying that because of vertebrate eyes, cephalopod eyes are more or less likely?
So, a multitude of ways to accomplish the same function?
Yet I reiterate that the more the eye evolves, the less likely it is that evolution is doing it.
But a lens covering the eye opening? Just where it needs to be, and with a useful transparency and refractive index.
But you skipped my other points…
Why so?
Well, fine, but the events of interest are those that might meet a particular specification.
@lee_merrill Sorry to jump back so far, but I didn’t have time to respond to these earlier, then forgot for a while.
Evolution could be falsified in a simple experiment where selectable steps are available and favorable to fitness, therefore predicted to occur. If those steps are never selected we have a data point of evidence against evolution. No one seriously questions evolution in this way, we know it works.
The same is not true of design. In the same experiment the Designer could choose these same selectable steps, and would be indistinguishable from evolution. We can never falsify the action of the Designer.
Somewhat counter-intuitively, failure of this experiment might provide evidence for design, if we hypothesize that the Designer would prevent this sort of evolution from occurring. To my knowledge no one has ever suggested such an experiment, or posed any reason a Designer might do such a thing.
The Tiktaalik test is still valid. It need not be the first such example of such a creature to meet the prediction that such a creature should exist.
From their proposal that about 2000 steps were involved.
But the change has to be fixed, in order to have its effect on the population. At the end of the process, they population has an eye, correct?
[quote=“John_Harshman, post:177, topic:15043”]
Only if you assume that neural evolution happens in a way quite unlike the incremental steps of the eye evolution itself.[/quote]
No, I assume that neural evolution is complex, though, and is not 100% probable.
Some unique events are best explained by design, and if they are repeated, it is less likely that evolution did it, is the claim.
Because it’s a tautology. If the sequence has a particular specification then it meets that specification.
In Dembski’s design inference the specification is always chosen after the fact, by looking for a specification in the data.
I’m not sure what you are saying. I think all single point mutations were encountered in the 10^20 trials of the malarial parasite, and none of them seemed to confer chloroquine resistance.
But that does not mean someone cannot use their results in their calculations.
How does Behe ignore existing variation, when evolution at the start of chloroquine use had existing variation to work with?
Alternatively, the easier it just is to evolve eyes.
Yes. I conclude that you didn’t read the paper and you don’t understand how natural selection works. Are you expecting selection fo favor a lens just where it doesn’t need to be, and without a useful transparency and refractive index? Lens evolution, in the paper, results from a series of tiny changes, each favored by selection, just as the rest of eye evolution does.
Fine, so then design is invoked when such a pathway becomes inordinately improbable.
Again, Behe is observing what evolution actually did, where all paths were open to evolution. No possible paths were excluded!
You are assuming your conclusion. All you are observing is a pattern. It evolved is a hypothesis.
A competent ID proponent would not make this claim.
I did not ask about the likelihood. I asked what is the cause.
Spliceosome data.pdf (211.7 KB)
Other more probable adaptations are fine, and Behe is interested in the rare adaptations, which show where the limit of evolution might be. He then computes the probability of a new beneficial protein-protein binding site developing, and finds it comparable to 1 in 10^20. Two such binding sites, he calls the edge of evolution.
And again, you are confusing the probability of an event viewed before the fact (1 in 100 million) with the probability of that event viewed after the fact (some people won).
How so?
His results are applicable there, if a new human structure requires more than two new protein-protein binding sites, then it’s likely beyond the edge of evolution. But why is investigating the limits of what evolution can do, laughable?
I saw a news article announcing a claim that a boy had been born with cat eyes, and a scientist stated that that’s likely untrue. Is the edge of evolution of no interest to you?
No, but selective pressure and constrained solution can lead to similar outcomes. We observe a correlation because not all outcomes are equally likely.
A greater likelihood that a function for light sensitivity and interpretation (eyes) will evolve.
Do you also maintain that the more children a person has, the less likely they will have grandchildren?
You made an equivalent statement (will quote presently).
EDIT: That last bit isn’t accurate - my mistake.
I can see that it’s possible for evolution to reach physical limits, such that further improvement is not possible, or reaches a biological optimum where further improvement does not support increased biological cost. It does not follow that such a state is not evolved. ALL states of evolution could be designed, and that is not falsifiable.
Details, please?
One new binding site was produced, actually, which was beneficial.
Because it would convincingly demonstrate evolution in progress. And flagella re-evolving makes it even more unlikely that evolution is doing this, especially if these would likely be independent events.
The frequency of plasmodium replications with which chloroquine resistance evolves is not the probability of evolving a protein-protein binding site. That simply doesn’t make sense, and you can’t just pretend one is equivalent to the other.
Quibble: What you want here is a Poisson distribution, where the sample mean estimates a rate.
Behe calculates the rate at which resistance arises in a chloroquine- treated population. See the difference?
Of course no one is looking for this resistance if chloroquine is not involved, but that same mutation can still occur. Without the presence of chloroquine this mutation is unlikely to fix in a population, so these events are censored (we never see them).
I think you misunderstand him. My understanding is that he’s not talking about how good or effective the eye is, he’s talking about how many times it has independently arisen in the history of life. It sounds to me like he’s saying that the more independent origins we can find for eyes, the less does he believe evolution is responsible for each of those origins.