Anything HGT does, Darwinian evolution did not do

_{Denyse O'Leary

August 13, 2015

horizontal gene transfer}

Share: Facebook; Twitter; LinkedIn; Flipboard; Print; Email

Talk to the Fossils.jpg From Horizontal gene transfer: Sorry, Darwin, it’s not your evolution any more:

Richard Dawkins: For over a century, Darwinism was the “must be” explanation, the only “scientific one.” As Dawkins put it (p. 287, Blind Watchmaker, 1986):

My argument will be that Darwinism is the only known theory that is in principle capable of explaining certain aspects of life. If I am right it means that, even if there were no actual evidence in favour of the Darwinian theory (there is, of course) we should still be justified in preferring it over all rival theories.

But Darwinism is not “the only known theory that is in principle capable of explaining certain aspects of life.” Claims that were formerly merely preferred must be tested against HGT. True, some of the example findings given above may need revision or replacement. But many more will likely turn up, as research uncovers HGT in many genomes.

Anything HGT does, Darwinian evolution did not do. As more and more pieces are carved out of Darwin’s territory, just think of the impact on the vast project of “Darwinizing the culture.” More.

See also: Links to the rest of the series at Talk to the fossils: Let’s see what they say back

Follow UD News at Twitter!

Comments

Carpathian, You need help.
So why don’t you break the following down for me.
I did. They are both examples of artificial selection. That means it is a strawman as it doesn't represent the reality of selection and elimination. Mayr explained it. What part of Mayr's explanation didn't you understand?
What Darwin called natural selection is actually a process of elimination. pg 117

Do selection and elimination differ in their evolutionary consequences? This question never seems to have been raised in the evolutionary literature. A process of selection would have a concrete objective, the determination of the “best” or “fittest” phenotype. Only a relatively few individuals in a given generation would qualify and survive the selection procedure. That small sample would be only to be able to preserve only a small amount of the whole variance of the parent population. Such survival selection would be highly restrained. By contrast, mere elimination of the less fit might permit the survival of a rather large number of individuals because they have no obvious deficiencies in fitness. Such a large sample would provide, for instance, the needed material for the exercise of sexual selection. This also explains why survival is so uneven from season to season. The percentage of the less fit would depend on the severity of each year’s environmental conditions. pg 118
Eliminating the bottom 3% is much different from selecting the top 3%. That is basic. Instead of responding to that you erected a strawman in which selection and elimination are artificially the same. Clearly Mayr states they are not the same nor do they produce the same result.Virgil Cain_{August 22, 2015
August
08
Aug
22
22
2015
08:49 AM
8
08
49
AM
PDT}

Virgil Cain, kairosfocus, Virgil Cain needs help in understanding "metaphors". Also, please explain to him how negative logic is used on a motherboard. It may help him understand the concept of a "concept".
As I said you are too stupid to grasp what Mayr said.
Typical response when you don't have a good argument. So why don't you break the following down for me.
Carpathian: I go to the supermarket and I select the three best tomatoes on the shelf. The other seven I ignore because they’re not good enough for me. OR…… I go to the supermarket and take the seven worst tomatoes and eliminate them. I take the remaining three and buy them. In both cases, I end up with the same three tomatoes.
Carpathian_{August 22, 2015
August
08
Aug
22
22
2015
07:54 AM
7
07
54
AM
PDT}

Carpathian:
And nature performs natural “selection” by “eliminating” those that can’t “survive”.
Yes, those 3% that are eliminated and those 97% that survive. In contrast selection would choose 3% and eliminate the 97%. As I said you are too stupid to grasp what Mayr said.Virgil Cain_{August 21, 2015
August
08
Aug
21
21
2015
12:09 PM
12
12
09
PM
PDT}

Virgil Cain:
Carpathian, You are clueless. I am saying that you have erected yet another strawman. Obviously you are too stupid to grasp what Mayr said. BOTH of your scenarios are artificial selection.
And nature performs natural "selection" by "eliminating" those that can't "survive". Again, the idea of a "concept" or "metaphor" is lost on IDists. Ask kairosfocus for help on understanding what a metaphor is.Carpathian_{August 21, 2015
August
08
Aug
21
21
2015
11:03 AM
11
11
03
AM
PDT}

Carpathian, You are clueless. I am saying that you have erected yet another strawman. Obviously you are too stupid to grasp what Mayr said. BOTH of your scenarios are artificial selection.Virgil Cain_{August 21, 2015
August
08
Aug
21
21
2015
10:10 AM
10
10
10
AM
PDT}

Virgil Cain, kairosfocus would not get this wrong.
Virgil Cain: No, moron. You get to select 3 OR eliminate 3.
He's explained things to Mung before, so maybe he can give you a hand now.
Carpathian: I go to the supermarket and I select the three best tomatoes on the shelf. The other seven I ignore because they’re not good enough for me. OR…… I go to the supermarket and take the seven worst tomatoes and eliminate them. I take the remaining three and buy them. In both cases, I end up with the same three tomatoes.
Carpathian_{August 21, 2015
August
08
Aug
21
21
2015
09:54 AM
9
09
54
AM
PDT}

Natural selection is a subset of selection.
Perhaps in your limited mind.
Weasel demonstrates selection, but does not model natural selection.
Selection is telic.Virgil Cain_{August 21, 2015
August
08
Aug
21
21
2015
09:49 AM
9
09
49
AM
PDT}

Carpathian:
How could you have missed that I have said to you the very same thing you’re saying to me.
Who says that I missed it?
This is why Dawkins’ Weasel algorithm is so difficult to understand for IDists.
It is you who doesn't understand it.
“Selection” is a “concept” which Weasel does a good job of simulating.
Selection is only an intelligent design process.
I go to the supermarket and I select the three best tomatoes on the shelf. The other seven I ignore because they’re not good enough for me. OR…… I go to the supermarket and take the seven worst tomatoes and eliminate them. I take the remaining three and buy them. In both cases, I end up with the same three tomatoes.
No, moron. You get to select 3 OR eliminate 3.Virgil Cain_{August 21, 2015
August
08
Aug
21
21
2015
09:47 AM
9
09
47
AM
PDT}

Virgil Cain:
Carpathian, Mayr explained the difference. The top 97% is very different from the top 3%. Selection gets you the top 3% and elimination gets you the top 97%.
How could you have missed that I have said to you the very same thing you're saying to me. I go to the supermarket and I select the three best tomatoes on the shelf. The other seven I ignore because they're not good enough for me. OR...... I go to the supermarket and take the seven worst tomatoes and eliminate them. I take the remaining three and buy them. In both cases, I end up with the same three tomatoes. We use the term "selection" metaphorically , not literally. This is the problem I see between the two sides. IDists tend to take a "concept" and analyze it literally while evos look at that same "concept" from a virtual/metaphorical point-of-view. This is why Dawkins' Weasel algorithm is so difficult to understand for IDists. "Selection" is a "concept" which Weasel does a good job of simulating. Weasel is not actually demonstrating every aspect of evolution.Carpathian_{August 21, 2015
August
08
Aug
21
21
2015
09:29 AM
9
09
29
AM
PDT}

Mung: It allegedly demonstrates “cumulative selection” (whatever that means). Are you saying that natural selection is not cumulative selection? Natural selection is a subset of selection. Weasel demonstrates selection, but does not model natural selection.Zachriel_{August 20, 2015
August
08
Aug
20
20
2015
05:05 AM
5
05
05
AM
PDT}

I could full a bucket with sand. Then I could allow people to select a handful of sand from the bucket. That demonstrates selection.Mung_{August 19, 2015
August
08
Aug
19
19
2015
09:28 PM
9
09
28
PM
PDT}

Carpathian- The main problem is that reproduction is the very thing that requires an explanation and reproduction is granted in the algorithms.Virgil Cain_{August 19, 2015
August
08
Aug
19
19
2015
04:54 PM
4
04
54
PM
PDT}

Carpathian, Mayr explained the difference. The top 97% is very different from the top 3%. Selection gets you the top 3% and elimination gets you the top 97%.Virgil Cain_{August 19, 2015
August
08
Aug
19
19
2015
04:46 PM
4
04
46
PM
PDT}

SA,
You introduced it, wd. Then you asked several pointed questions about Weasel. Then tried to defend it, amazingly.
I sometimes really wonder if you even read these posts.wd400_{August 19, 2015
August
08
Aug
19
19
2015
03:58 PM
3
03
58
PM
PDT}

Virgil Cain, What point? If you eliminate the bottom, you select the top. So the bottom of the fitness scale gets eliminated which means the top has been selected. Just like "Darwinism" says.Carpathian_{August 19, 2015
August
08
Aug
19
19
2015
03:41 PM
3
03
41
PM
PDT}

Mung:
Yet the Weasel algorithm is a search algorithm. What’s wrong with this picture then? Oh, it’s not apt for biology.
The selection function could display pixels on the screen corresponding to the bits in the string and accept input from viewers as selection criteria. The algorithm would then use the most pleasant display as voted on by viewers as the new parent. The string that eventually evolves might not be readable or predicted by anyone. This demonstrates selection . Since there is now no target with this new selection function, can we agree that selection is what is being demonstrated?Carpathian_{August 19, 2015
August
08
Aug
19
19
2015
03:38 PM
3
03
38
PM
PDT}

Carpathian:
If you select the top 3%, you have eliminated the bottom 97%.
Really? Wow
If you eliminate the bottom 3%, you have selected the top 97%.
And that is much different than selecting the top 3%. Thank you for proving my point.Virgil Cain_{August 19, 2015
August
08
Aug
19
19
2015
03:34 PM
3
03
34
PM
PDT}

Virgil Cain:
Eliminating the bottom 3% is much different from selecting the top 3%. That is basic.
If you select the top 3%, you have eliminated the bottom 97%. If you eliminate the bottom 3%, you have selected the top 97%.Carpathian_{August 19, 2015
August
08
Aug
19
19
2015
03:30 PM
3
03
30
PM
PDT}

eigenstate: I agree it’s not a search — “search” implies a kind of teleology that isn’t apt for biology. Yet the Weasel algorithm is a search algorithm. What's wrong with this picture then? Oh, it's not apt for biology. Then why is it constantly coming up (or other programs like it) in discussions of biological evolution? Oh, I know, blame the creationists!Mung_{August 19, 2015
August
08
Aug
19
19
2015
03:27 PM
3
03
27
PM
PDT}

eigenstate: Weasel doesn’t implement natural selection, nor does it purport to. Don't be a fool. It allegedly demonstrates "cumulative selection" (whatever that means). Are you saying that natural selection is not cumulative selection? Or perhaps I should say don't take us for fools.Mung_{August 19, 2015
August
08
Aug
19
19
2015
03:24 PM
3
03
24
PM
PDT}

Eigenstate, So soldiers were selected during the Battle of Wounded Knee? Is that the proper basic way to look at things?Box_{August 19, 2015
August
08
Aug
19
19
2015
03:19 PM
3
03
19
PM
PDT}

eigenstate, Ernst Mayr explains the difference in "What Evolution Is":
What Darwin called natural selection is actually a process of elimination. pg 117

Do selection and elimination differ in their evolutionary consequences? This question never seems to have been raised in the evolutionary literature. A process of selection would have a concrete objective, the determination of the “best” or “fittest” phenotype. Only a relatively few individuals in a given generation would qualify and survive the selection procedure. That small sample would be only to be able to preserve only a small amount of the whole variance of the parent population. Such survival selection would be highly restrained. By contrast, mere elimination of the less fit might permit the survival of a rather large number of individuals because they have no obvious deficiencies in fitness. Such a large sample would provide, for instance, the needed material for the exercise of sexual selection. This also explains why survival is so uneven from season to season. The percentage of the less fit would depend on the severity of each year’s environmental conditions. pg 118
Eliminating the bottom 3% is much different from selecting the top 3%. That is basic.Virgil Cain_{August 19, 2015
August
08
Aug
19
19
2015
03:07 PM
3
03
07
PM
PDT}

@Joe
eigenstate- the only selection in biology is artificial selection. Natural selection is eliminative, not selective. Natural selection does not optimize. Whatever is good enough gets the chance at reproduction.
Removal is selection -- the unremoved remain, the removed are... removed. If you deal 5 cards to you and tell you to discard three, by removing three cards from your hand -- any three -- you are necessarily selecting the remaining two. It can't be otherwise, Joe, and the is basic.eigenstate_{August 19, 2015
August
08
Aug
19
19
2015
02:28 PM
2
02
28
PM
PDT}

wd400
I should have guessed that brining up the Weasel algorithm would lead to a side-track circus.
When you ask a clown (Dawkins) to demonstrate some tricks on a toy model, and the carnival barkers try to sell it with double-talk ("Step right up! Take a look at something that's a kind of selection, and watch it create Shakespeare! With no intelligent design involved at all!") Yes, I'd say you have the beginnings of a circus right there. You introduced it, wd. Then you asked several pointed questions about Weasel. Then tried to defend it, amazingly.Silver Asiatic_{August 19, 2015
August
08
Aug
19
19
2015
01:45 PM
1
01
45
PM
PDT}

eigenstate- the only selection in biology is artificial selection. Natural selection is eliminative, not selective. Natural selection does not optimize. Whatever is good enough gets the chance at reproduction.Virgil Cain_{August 19, 2015
August
08
Aug
19
19
2015
01:34 PM
1
01
34
PM
PDT}

eigenstate, Thank you for proving “weasel” has nothing to do with natural selection.
Weasel doesn't implement natural selection, nor does it purport to. Selection in biology isa vastly more complex, dynamic and non-linear process. Weasel just shows how a cumulative process -- simple like Weasel's or mind-crushingly complex like selection in biology -- vastly accelerates convergence on local optima.
Evolutionism isn’t a search. And its “cumulative process” will give you an albino dwarf with sickle-cell anemia.
I agree it's not a search -- "search" implies a kind of teleology that isn't apt for biology. But the process is one that optimizes. And that optimization is radically improved by cumulative feedback loops. Weasel is a nice window into the power of cumulative feedback loops in a converging processes.eigenstate_{August 19, 2015
August
08
Aug
19
19
2015
11:56 AM
11
11
56
AM
PDT}

Mung:
First, I can’t examine the source code, be cause it no longer exists. So you are wrong about that.
I don't understand why you insist on looking at source code all the time. How is it you don't simply understand the underlying concept? If I gave you a function name and it's parameters you should be able to understand what's going on. Secondly, whether or not one implementation of algorithm X contains a certain functionality, doesn't mean that other implementations do. You should be able to write a Weasel program yourself from scratch that does not require the code to be modified with different string lengths or content. If you can't, then that's a limitation of yours, not the Weasel algorithm's.Carpathian_{August 19, 2015
August
08
Aug
19
19
2015
11:39 AM
11
11
39
AM
PDT}

Mung:
eigenstate: More importantly, the part of the program that is generating mutations and new candidates does not know, and cannot access the target value. The mutation loop calls the fitness function, and supplies the candidate string, just getting back a score in return indicating the fitness, or relative distance to the target, but not the target itself:
Mung, this is a key point. Different parts of the process do not have an indication of what the other parts are doing. For instance, the selection function might be changed in the middle of a run and the other functions would not know that the selection criteria had changed. There is no "conspiracy" between the functions that result in the output. I am amazed that someone who likes to talk as much about software as you do, has such a misguided understanding of how software works.Carpathian_{August 19, 2015
August
08
Aug
19
19
2015
11:25 AM
11
11
25
AM
PDT}

eigenstate, Thank you for proving "weasel" has nothing to do with natural selection.
The pedagogy here is to show that for any given search space, large or small, cumulative processes with positive feedback loops will converge much faster on any given target (and this true whether the target is static or dynamic across the running time of the program) than a simple incremental search will.
Evolutionism isn't a search. And its "cumulative process" will give you an albino dwarf with sickle-cell anemia.Virgil Cain_{August 19, 2015
August
08
Aug
19
19
2015
11:00 AM
11
11
00
AM
PDT}

@Mung,
First, I can’t examine the source code, be cause it no longer exists. So you are wrong about that.
There are weasel implementations available in a large number of languages see here for example, a C++ implementation.
Second, Yes it does “peek” and I provided two clear instances of how it does, and you even go on later to admit that this is in fact the case.
There's no peeking. The implementation linked above uses, as did Dawkins, a fixed string length, but this is define in the structure of the code at compile time. The program doesn't "look" anywhere to discover anything that will help it in either case. More importantly, the part of the program that is generating mutations and new candidates does not know, and cannot access the target value. The mutation loop calls the fitness function, and supplies the candidate string, just getting back a score in return indicating the fitness, or relative distance to the target, but not the target itself: // this function returns 0 for the destination string, and a // negative fitness for a non-matching string. The fitness is // calculated as the negated sum of the circular distances of the // string letters with the destination letters. static int fitness(std::string candidate) { assert(target.length() == candidate.length()); int fitness_so_far = 0; for (int i = 0; i < target.length(); ++i) { int target_pos = allowed_chars.find(target[i]); int candidate_pos = allowed_chars.find(candidate[i]); int diff = std::abs(target_pos - candidate_pos); fitness_so_far -= std::min(diff, int(allowed_chars.length()) - diff); } return fitness_so_far; } This function is the "oracle" I mentioned earlier. It has access to the target value, which it must in order to assess fitness. But the caller of this function does not get the target value returned, or have access to it. There's no "peeking" at the answer, just a score returned that enables prioritizing of candidates as some score better than others. As this runs in a cumulative fashion, the repeated variation and re-scoring will converge on the target, which remains unknown until the actual target is matched exactly (fitness score == 0). Adding variable string length, or Unicode support is not hard to do, but adds nothing to the demonstration, except make it run longer to show what it is designed to show. There's nothing special about 28 characters -- it could have been a longer string (will take more time on average), or shorter (less time). This is just the length of the string Dawkins fancied from Shakespeare to use for the example. The size of the search space is not a factor in this demonstration. It can be arbitrarily small or large, doesn't matter. The pedagogy here is to show that for any given search space, large or small, cumulative processes with positive feedback loops will converge much faster on any given target (and this true whether the target is static or dynamic across the running time of the program) than a simple incremental search will.eigenstate_{August 19, 2015
August
08
Aug
19
19
2015
10:31 AM
10
10
31
AM
PDT}

Prev 1 2 3 4 … 7 Next

You must be logged in to post a comment.

Leave a Reply