If evolution is not a search, why is the term “evolutionary search” not an oxymoron?
Over at Uncommon Descent Elizabeth posted the following:
“…any “search” algorithm worthy of the name of “evolutionary search” comes with its own moderately smooth fitness landscape built in.”
So evolution is a search if it comes pre-built with its own moderately smooth fitness landscape built in?
Steve, to Piotr:
Which is why this baby’s head became dog food instead of being wasted. Praise the Lord! What an efficient design.
We have plenty of evidence the Earth has been through five major mass extinction events and another dozen or so minor ones in the last 500 million years.
Extinction events
Looks like Piotr has a very strong point.
Your ID explanation for the multiple mass extinction events is…?
Well Acartia,
It is very possible that a certain number of organisms exhibiting cannibalistic tendencies is a pre-requisite for a stable ecosystem.
Would that make the designer ‘bad’???
I guess if fish have feelings and such then you could make a case for a cruel designer.
But then you have humans that exhibit a peculiar trait to be able to grasp such a concept as cruelty and even make a choice to avoid that type of activity.
Just maybe that designer is not so cruel after all. Especially if the design is geared to awaken a particular segment of the design to knowledge of life.
No one said life should be pretty all around. Life as it is seems an awesome thing.
Steve I’d appreciate if you would give us the ID explanation for the five major mass extinction events the other dozen or so minor ones in the last 500 million years.
ID does have an explanation, doesn’t it?
Adapa,
If Life is designed to live in tandem with the environment, then the extinctions would represent cataclysmic environmental events.
The fact that Life survived 5 cataclysmic events proves the hardiness of design.
Evolution could not resurrect life 5 times. It could never assemble enough tools in time to get the job done.
The hardiness of life can only be explained by design.
After every single one of the major extinctions “the concentrationof life on Earth” was seriously diminished — that’s why they are called mass extinctions. The K-Pg event killed off some 75% of species, from microorganisms to the largest dinosaurs, breaking food chains, wiping out all land animals bigger than a dog and more than half plant species. The end-of-Permian extinction was still worse — the only one in which even insects were mass-killed.
Yep. To quote Mark Twain’s description of Mono Lake (in Roughing It):
There are no fish in Mono Lake—no frogs, no snakes, no pollywogs—nothing, in fact, that goes to make life desirable. Millions of wild ducks and sea-gulls swim about the surface, but no living thing exists under the surface, except a white feathery sort of worm, one half an inch long, which looks like a bit of white thread frayed out at the sides. If you dip up a gallon of water, you will get about fifteen thousand of these. They give to the water a sort of grayish-white appearance. Then there is a fly, which looks something like our house fly. These settle on the beach to eat the worms that wash ashore—and any time, you can see there a belt of flies an inch deep and six feet wide, and this belt extends clear around the lake—a belt of flies one hundred miles long. If you throw a stone among them, they swarm up so thick that they look dense, like a cloud. You can hold them under water as long as you please—they do not mind it—they are only proud of it. When you let them go, they pop up to the surface as dry as a patent office report, and walk off as unconcernedly as if they had been educated especially with a view to affording instructive entertainment to man in that particular way. Providence leaves nothing to go by chance. All things have their uses and their part and proper place in Nature’s economy: the ducks eat the flies—the flies eat the worms—the Indians eat all three—the wild cats eat the Indians—the white folks eat the wild cats—and thus all things are lovely.
Piotr,
The point being that only design could withstand mass extinctions and continue on. Unguided evolution, banking on fortuitous mutations in micro-step could never hope to regain what supposed success it had attained through a painfully slow process.
Not only once, but 5 are the times you are apparently claiming evolution was able to manage that feat.
That calls for some major green beer all around!!!
As for Mark Twain’s quote, i guess life is too uncomfortable for some. Where Twain sees the glass half empty, I take the half-full glass.
Besides, exceptions never do well as substitute teachers. They do better as consultants on specific issues
Like for instance, Mono Lake could very well be like the spoon of yoghurt starter i keep in a small tub in case my bulk yoghurt goes bad. I have had the same Russian yoghurt strain for 15 years, given to me by an American priest living in the mountains surrounding Taipei.
Maybe I dont need that safety net so ensure the endless supply of homemade yoghurt I enjoy, but doesnt hurt to have a stash at the ready.
So from a design perspective, Mono Lake makes a lot of sense.
Glass half-full!
Adapa,
To answer your questions specifically, my position is that the earth was designed interactively where the environment and life were developed simultaneously, early life contributing to the development of the biosphere and in turn the developing environment contributing to the characteristics and varieties of life.
We do know that life started early in the earth’s history. With all the turbulence most likely present at that time, early life could not go 5 rounds without a starter-kit. It would get knocked out in the first 5 seconds.
So its clear design is the better explanation for life, not unguided evolution.
Its a question of how it was designed, not if.
keiths,
I fully agree. The owners of that dog did a piss poor job of shaping that dog’s temperament.
They ‘designed’ it to attack intruders but ended up designing a temperament that attacked anything.
So maybe if the dog owners had, you know, taken a course in how to train a dog properly, if wouldn’t have turned out eating baby heads.
Poor designs do indeed turn out ugly.
Back to the drawing board, dog owners. That’s after you finish your stint in jail for negligent homicide that is.
Ouch!!!
OK, how was it designed, and by whom?
So you have no ID explanation for the extinctions and subsequent re-radiations of vastly different species. All you can do is toss out a hand waving non-sequitur. What a surprise.
Evidence-free wild speculation noted. If the environment was designed too then we’re back to the mass extinction question. Did the Designer purposely smack the Earth with the Chicxulub asteroid? And the Deccan Traps?
If you’re going to make up crap on the fly at least try to be consistent with your story.
More evidence-free assertions.
Your conclusion is unsupported. You might as well claim the blue sky was designed because the Designer liked the color blue.
Then how was it designed? How did we get to the 15 million or so extant species and the 100x that number which have gone extinct in the last 3.5 billion years?
Just once, I’d like to see a proof of concept from an ID advocate.
Tell us by what process you design biology.
Steve,
No, you don’t. You’re trying to get your God off the hook for designing every predator that ever attacked and ate a human.
Creation was obviously designed to be indistinguishable from evolution.
The original post in this thread was by Mung. He had asserted at UD that I was wrong in my characterization of Dembski, Ewert, and Marks’s concept of “search”. I had said their “searches” included all sorts of crazy and dysfunctional ones. They are free to define “search” that way, but it is important to understand how they are defining it. Because if “search” is defined that broadly, then just from having organisms that have fitnesses, one has substantial amounts of their “Active Information”.
I have in fact replied to Mung here. I have seen no response by Mung, who seems to be commenting elsewhere these days. When Mung gets a chance to come back here and comment, he should acknowledge that he was the one who was wrong about what DEM’s “searches” include.
After that is settled, we can discuss whether or not evolution “is a search”.
I’ll remind him when I see him 😉
Joe Felsenstein,
I’ve informed him. He didn’t seem too pleased.
Joe F:
Joe, if you thought I disputed that, I’d like to know why.
Richardthughes on May 8, 2015 at 3:49 am said:
Joe Felsenstein,
I’ve informed him. He didn’t seem too pleased.
Hang on to Joe’s coattails hard. Don’t breathe in the fumes. Don’t contribute anything original. Don’t think for yourself. God forbid.
Mung,
Waaaaaaambulance, for Mung.
Pleased you’ve re-engaged Mung. You will no doubt be making a strong positive case for design at some point. I look forward to your contributions.
Here is what Joe F. wrote that I objected to:
Here’s Joe F in his latest:
Some of you will note the difference. Others won’t.
Richardthughes:
Feel free to start your own thread. Supposing you’ve been hard-wired to do so.
idiots
Nah, this is the point where Mung scurries back to UD and brags about how he embarrassed all those ignorant evos at TSZ with his razor-sharp wit. He’s not very bright when it comes to the evolutionary sciences but there’s nothing wrong with his mouth.
Mung,
Will you be participating Mung? Should I call it “Mung’s positive case for a designer”? You have author privileges, are you sure you don’t want to lead the charge? I don’t want to steal your thunder, only bathe in your tears.
I see Steve has gotten cold feet and run off too. Scientific questions about their inane claims always seem to have that effect on the IDiot Creationists.
I’m sure we’ll see him at UD soon bragging about his “victories” along with Mung, phoodoo and WJM.
There’s a rumor that all of aurelio smith’s posts at UD have disappeared. Big win for Winston Ewert.
In answer to Mung:
Let’s go back to the comment you made at UD. You said
I replied that it is not false, because the “unless” part of your statement is true. Every process that results in a probability distribution of outcomes in the search space is one of DEM’s “searches”. Even crazy ones such as those that end up preferentially finding less fit genotypes. Or find random ones. (And of course “a probability distribution” can be a distribution that puts probability 1.0 on one point in the search space, so the “searches” include the results of all deterministic and all stochastic processes.
So the part of your statement that I was wrong was itself incorrect, because the other part of your statement holds. Every process on the search space that leads to a probability distribution of outcomes is one of their “searches”.
So we are agreed about that?
Important to be on the same page, because that page turns out to be one on which simple presence of reproducing genotypes that have fitnesses is enough to have a lot of “Active Information”.
I think you underestimate Mung’s ability to quibble.
“every process” and “every process in the search space” are not exactly the same, and Mung is under no obligation to ask for clarification before running his mouth.
In fact, Mung is under a compulsion not to engage in adult dialog or try to reach consensus on the terms of the discussion.
I will retract that if anyone can point to a discussion where Mung engaged in an attempt to clarify terms.
If this thread was an attempt by its original poster (Mung) to discuss the issue of evolutionary search, and of DEM’s definition of “search”, then it would be reasonable to expect him to know that “every process” means every process that operates on a search space, and that comes up with a result.
Asking about other processes, such as me doing my laundry (which I am in fact in the process of doing right now) would of course be a silly diversion. I would not suggest that Mung would engage in such a diversion.
Joe Felsenstein,
Give a recent exchange:
I think perhaps he has withdrawn to learn the basics?
I think you are following Elizabeth’s rules religiously.
I think you are giving Mung the credit he merits under elizabeth’s rules.
I am trying to make sure that if someone descends into name-calling, it won’t be me. At least this time.
Joe Felsenstein,
Rodger that. I hope he returns and continues.
Joe F: So we are agreed about that?
No.
I already know how they define search. I don’t know how you define “every process.” If you define “every process” as every process that DEM define as a search then I have no issue with you. That wasn’t clear in your original post.
petrushka: “every process” and “every process in the search space” are not exactly the same.
Indeed. I thought I made that clear.
petrushka: and Mung is under no obligation to ask for clarification before running
his mouth.
two for two.
petrushka:
I attempted to clarify specifically what I was referring to in a recent conversation with Elizabeth.
So there you go.
There’s a rumor that all of aurelio smith’s posts at UD have disappeared. Big win for Winston Ewert.
Appears to be some substance to that. All I can say is wow. Just wow. I need some time to process that.
And TSZ is, as per usual, bipolar:
keiths:
I request definitions, I lose.
I don’t request definitions, I lose.
I guess I just lose. 🙂
Fitness surfaces and searches: Dembski, Ewert, and Marks’s search for design
It seems a bit peculiar to me, but Joe’s statement that “DEM call every process a ‘search'” is mysteriously absent from that article.
I am not simply saying “I’ll have what he has”. I am trying to understand what DEM call a “search”. They designate as a “search” any probability measure on their search space. In effect, any distribution of points on Ω.
It is implicit in their definition that this is the result of some process, as they also define multiple steps to such a process, but in the end a “search” is any probability measure. One possible “search” is the set of points tied for lowest fitness. Another possible “search” is a distribution that places equal probability on all points in the space.
So, in addition to probability measures that put higher probabilties on points that have higher fitness, their “searches” include all sorts of crazy ones. And that becomes important when we start discussing whether a model of reproducing organisms with fitnesses already has a substantial amount of “Active Information”.
So, are we agreed about DEM’s definition of “search”?
If not, do you find some other definition of “search” in their paper?
Joe F.,
Your original statement:
You began by saying:
What you have said, in effect, is that there is no process that is not included in their universe of searches.
If you are continuing to argue that their definition of search includes “every process” then no, I do not agree.
If you are saying that by “every process” you mean every process they define as a search then that is tautological. How could I disagree.
What process can’t be represented as a probability distribution, even if the distribution is 1.0 or 0.0?
Please provide some examples.
Adapa, I’m sorry, but you’re plagiarizing a question I already asked Joe.
But at least you understand what Joe’s claim entailed, which is what I was pointing out.
But it seems to me that Joe is saying that all he meant by “every process” was every process that DEM define as a search.
It would be nice to get a clear unambiguous answer from Joe as to whether he thinks every conceivable process meets the definition of a search as defined by DEM.
That’s pretty much what this entire exchange has been about.
Still up for some math learnin’, Mong?
I asked you the question Mung, not Joe or Ewert. You. If you don’t understand the discussion well enough to answer that’s fine. It will only confirm what most of us already knew.
You can now run back to UD and put up more insulting posts about Dr. Liddle and the other posters here like you normally do.
I’ve been trying to make this understandable.
When I say “every process” I do not mean every process that they define as a search. Of course that would be tautological and of no interest here.
There is a search space (say, all 1000-base DNA sequences). Imagine all processes that end up choosing a point (a DNA sequence) from that space. Any such process ends up defining a probability distribution, the distribution of points that result. If the process is deterministic, it would give a distribution with probability 1.0 on one point, 0.0 on all the others. If it is stochastic, one would have other probabilities.
Each process (each one that chooses a point in that space, either deterministically or stochastically) is one of DEM’s “searches”. That is how they define a “search”, and I am in this discussion using their definition of a “search”. Which is appropriate because the intent is to discuss their results.
I am, I repeat. not saying I am just defining “search” as “whatever they define as search”, without really knowing what their definition is. I am understanding how they define it. For the purposes of this discussion I am understanding their definition and using it.
That definition then includes, together with more sensible ones, all kinds of crazy dysfunctional searches.
So are we agreed?
Honestly Joe, I am scratching my head.
You say:
When I say “every process” I do not mean every process that they define as a search.
ok, to me that’s progress. That was my original question and you’ve answered it. But how is it that not every process is a search, as search is defined by DEM?
So here is how I read your latest:
When I say “every process” I do not mean every process that they define as a search, but DEM call every process a “search”, even crazy ones.
I’m just having difficulty wrapping my mind around that.
As soon as you start talking about search spaces and probability distributions I think you’re talking about DEM
Then you say:
That definition then includes, together with more sensible ones, all kinds of crazy dysfunctional searches.
I honestly don’t care. What DEM define as a search is only an issue if they define “every process” as a search.
You say:
When I say “every process” I do not mean every process that they define as a search.
Yet you said:
DEM call every process a “search”
Let me try to be even clearer. You said that if I was just defining a “search” as “whatever DEM call a search”, then that is tautological. I agree. It is incumbent on me (and you) to figure out what DEM’s definition of a “search” is, and use that so we can discuss their arguments.
So I have tried to understand what their criteria for a “search” are. And what I come up with is, that they call something a “search” if it is a probability distribution of points in their search space Ω
OK? Agreed?
Hi Joe,
In Winston Ewert’s own words:
Yes, I think we agree.
For example, tossing a six-sided die [or a two-sided coin] is a process that can be represented as a probability distribution and would qualify as a search.