Almost Nothing in Biology Makes Sense Except in the Light of Design

_{Gil Dodgen

August 14, 2010

Intelligent Design}

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

As we learn more, this is becoming increasingly, transparently obvious. Random errors can screw things up, and — in very rare circumstances — provide a survival advantage in a pathological environment.

The notion that random errors filtered by natural selection can account for all that is found in biological systems is a pathetically illogical, hopelessly improbable explanation for the information-processing machinery of the cell.

Chemistry is not the basis of life. Chemistry is the medium; information is the message.

On the subject of chemistry: The ID movement has thrown a satchel charge of trinitrotoluene into the Darwinian faux-edifice, and blown it to bits (pun intended).

Comments

If that’s really the worst thing about my paper, then I’m not sweating much at all.
No, the problem with the paper is its inadequate treatment of coordination. It would be better if it included some discussion of what actual biologists have said about the topic. I included quotes from Gould and Darwin. My observation is that you included a quote that would have led you to a better discussion of coordinated evolution, but you failed to follow up. This is not just a quibble about misrepresenting a quotation. It affects your earlier discussion of evolution.Petrushka_{August 18, 2010
August
08
Aug
18
18
2010
07:41 AM
7
07
41
AM
PDT}

his proposal embodies the theme of this essay the need to dissociate current utility from historical origin”. That’s the definition of exaptation, and, as Gould says, the theme of his essay.
Actually, the theme of Gould's essay is not exaptation, but "just so stories." The whole essay is devoted to discussing how stories about adaptive evolution can be invented without evidence. He notes that we don't know the history of giraffe evolution. Nor is the point of the quote about exaptation, but about gradualism and the appearance of coordination. He is discussion Darwin's conjecture, not his own. I provide links to Gould and to Darwin. There's no need to argue about this unless you care to cite the original references, as I have done. My point is that Darwin addressed your concerns about coordination in the 1870s, came to the same conclusion you did regarding the inability of evolution to produce coordination, and concluded that coordination did not occur and was not a necessary conjecture. The giraffe is a bad example for discussing coordination, because the fossil record is so sparce. If you want to discussed the coordination of features, it would be better to pick a lineage where we can see the intermediates. The mammalian middle ear bones might make a good case study. Oddly enough, the Piltdown hoax provides an object lession about what can happen when theorists are uncertain about the coordination of features.Petrushka_{August 18, 2010
August
08
Aug
18
18
2010
06:27 AM
6
06
27
AM
PDT}

Petrushka - This is simply exaptation. As Gould says in the extended quote - "his proposal embodies the theme of this essay the need to dissociate current utility from historical origin". That's the definition of exaptation, and, as Gould says, the theme of his essay. But, if you want to think that I am somehow misrepresenting Gould by quoting someone else who happened to be quoting him because I wanted to talk about Giraffe biology, then fine. If that's really the worst thing about my paper, then I'm not sweating much at all. The fact that someone looking for a "gotcha" in my paper managed to find something where they didn't think that I included enough context for a quote I used in a peripheral part of a paper as a guidance for how to use the other ideas of the paper in future studies, I think demonstrates the strength of my paper.johnnyb_{August 17, 2010
August
08
Aug
17
17
2010
10:04 PM
10
10
04
PM
PDT}

Your essay deals primarily with coordination, the simultaneous evolution of all parts of a complex system. Your misrepresentation of Gould and darwin lies in failing to discuss their response to the problem of coordination, which is not simply exaptation. My point is not that you are wrong, but that your discussion of Gould and Darwin misses the point they are making about coordination. Here's the Gould quote, in context:
In his subsequent and longest book, the two-volume Variation of Animals and Plants under Domestication (1868), Darwin does finally introduce the giraffe's neck in a discussion of natural selection. But again and ironically given the later codification of the case as our canonical just-so story in the speculative tradition, Darwin does not cite the neck of the giraffe to tell a tale about presumed adaptive advantages. Rather, he raises the example to discuss a more subtle issue central to the validity of natural selection as a general explanation of evolution. Even if we assume that the giraffe's neck evolved as an adaptation for eating high leaves, how could natural selection build such a structure by gradual increments? After all, the long neck must be associated with modifications in nearly every part of the body -- long legs to accentuate the effect and a variety of supporting structures (bones, muscles, and ligaments) to hold up the neck. How could natural selection simultaneously alter necks, legs, joints, muscles, and blood flows (think of the pressure needed to pump blood to the giraffe's brain)? In response to this problem, some critics had proposed that all relevant parts must be changed together in one fell swoop. Such suddenly coordinated modification would invalidate natural selection as a creative force because the desired adaptation would then arise all at once as a fortuitous consequence of internally generated variation. (Moreover, Darwin adds, we have no evidence for a deus ex machina of such complexly coordinated variation--and the whole proposal smacks of desperation and special pleading.) Darwin provides a cogent and subtle explanation (perhaps not thoroughly satisfactory by current views, but entirely logical and coherent). Interestingly, his proposal embodies the theme of this essay the need to dissociate current utility from historical origin. A giraffe's current functioning may require coordinated action of all parts that support the long neck, but these features need not have evolved in lock step. If the neck grows by ten Feet all at once, then every bit of supporting anatomy must be in place. But if the neck elongates by only an inch at a time, then the full panoply of supporting structures need not arise at every increment. The coordinated adaptation can be built piecemeal. Some animals may slightly elongate the neck, others the legs; still others may develop stronger neck muscles. Through sexual reproduction, the favorable features of different organisms may be combined in offspring.
Petrushka_{August 16, 2010
August
08
Aug
16
16
2010
05:15 PM
5
05
15
PM
PDT}

"By omitting this history of the problem, you imply that mainstream biologists have never confronted or addressed the problem of coordinated evolution." No, I addressed it at length in section 1.3. I didn't feel the need to re-argue what I had already pointed out in section 1.3, and demonstrated using computability for the next two sections. Why should I then repeat the paper at this point? The point of this section - if you bothered to do more than a cursory reading of my paper - was not to argue with Gould or Darwin or anyone else on the evo side - I had already done that. It was to show how to *apply* my ideas to biosystematics (as you should be able to see from the headings, it was in a section called "Application to Creationist Systematics". Since Gould is not a Creationist Systematist, I was only interested in his ideas on the *biology* of the Giraffe. My discussion was actually with Bergman, who thought the Okapi was a different created kind. My point, if you bothered to read it, was that if we determine that two organisms which have complex, coordinated structures, *have* in fact diverged, then we can use the information from that divergence to help us know what kind of purpose the organism served. If I had included the quote from Gould in section 1.3 where I discuss and respond to exaptation as an evolutionary process you would be right to criticize me for leaving out an important part of the argument. But here, I am assuming the previous part of my paper, and simply discussing the biology of the Giraffe, which, in fact, Gould supported the position quoted. Did he support my extrapolations from his position? Certainly not. But that wasn't what I quoted him for, nor did I imply such. As I said, I was mostly quoting Bergman, but Gould happens to be a better writer. Do you, in the papers you write, re-iterate every point you have previously made when you encounter a new example? Do you feel the need to re-argue everything again? Such would make for a very poorly written paper.johnnyb_{August 16, 2010
August
08
Aug
16
16
2010
02:45 PM
2
02
45
PM
PDT}

But my reference to Gould was not about his position on the evolution of the Giraffe (in fact, I make no quotation from Gould whatsoever on the evolution of the Giraffe), but rather on its biology.
What you quoted from Gould was from an essay on the evolution of the giraffe. The particular quote was from a section of Gould's essay dealing with Darwin's speculation on the evolution of the giraffe. The whole point of the Gould quote is that it suggests that many structures must be modified simultaneously in order for the neck to lengthen. If you follow the Gould quote to its source, as I did, you find he is discussing Darwin's ideas on giraffe evolution, and in particular Darwin's ideas regarding the need for simultaneous evolution of several complex systems. Here's what Darwin said in 1875 (and found in Gould's essay immediately following your snippet):
I will conclude this chapter by some remarks on an important subject. With animals such as the giraffe, of which the whole structure is admirably co-ordinated for certain purposes, it has been supposed that all the parts must have been simultaneously modified; and it has been argued that, on the principle of natural selection, this is scarcely possible.
This is the problem Gould is addressing, and this apparently is what you are addressing with your statement:
The theoretical framework presented here would indicate that if complex, coordinated features such as these have evolved, that they would require either information from the outside (obtained through many possible mechanisms, including symbiosis, horizontal gene transfer, or by God’s direct involvement),
So the response to your simple questions is that Gould is addressing the evolution of the giraffe. You only have to read his essay, which is entirely devoted to the evolution of the giraffe. The next question is whether the Gould quote is consistent with the actual evolution of the giraffe, and at this point it becomes more complicated. The point of Gould's essay is that we don't know the history of the evolution of the giraffe, and that it is possible to make more than one plausible scenario. The length of the giraffe's neck could be due to sexual selection, much like an elk's antlers. But this is beside the point. One of the main points you derive from the giraffe is that the coordination of the evolution of structures is not possible via Darwinian mechanisms. The information required for this coordination must be supplied by some external source. Your point was made in 1875 by Darwin, quoted and discussed by Gould. By omitting this history of the problem, you imply that mainstream biologists have never confronted or addressed the problem of coordinated evolution.Petrushka_{August 16, 2010
August
08
Aug
16
16
2010
10:14 AM
10
10
14
AM
PDT}

Right - exaptation. Which, as I said, was dealt with in section 3.1. But my reference to Gould was not about his position on the evolution of the Giraffe (in fact, I make no quotation from Gould whatsoever on the evolution of the Giraffe), but rather on its biology. So, again, these are pretty simple questions: 1) is the quote INCONSISTENT with Gould’s view of the biology of the Giraffe? 2) more importantly, is the quote INCONSISTENT with the true biology of the Giraffe?johnnyb_{August 16, 2010
August
08
Aug
16
16
2010
05:49 AM
5
05
49
AM
PDT}

Peepul: Natural selection doesn't create any new information; it can only select new information created by purportedly random sources. The only thing natural selection does is kill things off - it eliminates information. Darwinists contend that the only information being actively generated is random or chance information, and that NS selection purportedly "kills off" that information which does not increase the rate/survivability of offspring (fecundity). IOW, the ONLY selective "algorithm" that sorts biological information is that which promotes successful progeny by killing off unsuccessful progeny. Now, ask yourself this question: how is it that through this sorting process, we have gone from very hardy creatures that could survive in a wide range of diverse environments, and which reproduced at very high rates throughout their lives (single-celled organisms), towards highly complex and thus less hardy organisms (because of the far greater capacity for system failure the more functionally complex and interdependent the systems are) that reproduce so slowly (in comparison to single-celled creatures) that we might as well not be reproducing at all? How exactly does a sorting process tuned only for "the most successful progeny" generate anything other than longer-living, more fecund organisms? The only natural sorting algorithm we have, would not, and could not, sort for humans, elephants, or sea turtles; all of the "higher" life forms are necessarily and definitionally "discrepancies" when it comes to explanation via natural selection, because they certainly do not have a greater fecundity than organisms which existed billions of years ago. Natural selection cannot explain the existence of a human being.William J. Murray_{August 16, 2010
August
08
Aug
16
16
2010
05:42 AM
5
05
42
AM
PDT}

The place where you depart from Gould and Darwin is in assuming coordinated evolution of all the features:
...saying (quoting Gould), “the long neck must be associated with modifications in nearly every part of the body—long legs to accentuate the effect, and a variety of supporting structures (bones, muscles, and ligaments) to hold up the neck.” Bergman further points out the further coordination with the giraffe’s circulatory system, including complex, unique valves and special tissue below the brain to regulate the sudden rush of blood to and from the brain. Such a necessary coordination of changes to achieve the result makes the giraffe’s neck likely RIC. ... The theoretical framework presented here would indicate that if complex, coordinated features such as these have evolved, that they would require either information from the outside (obtained through many possible mechanisms, including symbiosis, horizontal gene transfer, or by God’s direct involvement), or that their genomes are formatted in such a way as to facilitate development along these lines.
The links I provided are a direct response to the question of coordination.Petrushka_{August 15, 2010
August
08
Aug
15
15
2010
09:39 PM
9
09
39
PM
PDT}

Petrushka - My point was about Giraffes. I quoted Gould as saying, "the long neck must be associated with modifications in nearly every part of the body—long legs to accentuate the effect, and a variety of supporting structures (bones, muscles, and ligaments) to hold up the neck". Is this an inaccurate representation of Gould's thoughts on the Giraffe's neck? More importantly, is this an inaccurate representation of the biology of the giraffe? The point of this section was actually Bergman's description of the Giraffe, and only included Gould because Gould's writing (which Bergman quotes) is much better than Bergman's. The point of that section, interestingly, was actually pro-evolutionary, and, in fact, quite in keeping with a Gould-ish view of evolution generally. If you read my next paragraph, you will see that I am for evolution of the Giraffe from the Okapi, but point out that the nature of the structures likely means that it required information to do so. So, two questions for you: 1) is the quote INCONSISTENT with Gould's view of the biology of the Giraffe? 2) more importantly, is the quote INCONSISTENT with the true biology of the Giraffe? Since I was actually quoting Bergman rather than Gould, if either of these questions is "no", then I wasn't misrepresenting anything (and, in fact, the answer to both questions is no). Also, if you cared to read my paper, Gould's answer to how the giraffe's neck evolved (as was his answer to other Darwinian problems) was exaptation , (or, as Gould put it in this paper, "the dissociation of current utility from historical origin") which was dealt with at length in section 1.3.johnnyb_{August 15, 2010
August
08
Aug
15
15
2010
08:00 PM
8
08
00
PM
PDT}

johnnyb: Your paper on the limits of evolvability quotes from Gould on the giraffe's neck, but Gould is leading up to a rather lengthy quote and discussion from Darwin: http://darwin-online.org.uk/content/frameset?viewtype=side&itemID=F880.2&pageseq=223 Whether you are right or wrong about evolution, it would be better when quoting, to accurately represent the original authors argument and intention. Your snippet misrepresents the intentions of both Gould and Darwin. The Gould article is here: http://bill.srnr.arizona.edu/classes/182/Giraffe/Tallest%20Tale-Illus.pdfPetrushka_{August 15, 2010
August
08
Aug
15
15
2010
03:04 PM
3
03
04
PM
PDT}

JB: A classic. Gkairosfocus_{August 15, 2010
August
08
Aug
15
15
2010
02:29 PM
2
02
29
PM
PDT}

I don't follow the part about biological evolution not being a universal computing system. It seems quite clear from observation that fatal variations occur, neutral variations occur, and synonymous variations occur, as well as the occasional beneficial variation. Perhaps you are applying your hypothesis to the origin of life rather than to evolution. I've seen it argued here that truely random variation would kill the operating system. That's certainly a possibility, and in fact we see it happen in individuals. What's necessary to prevent the error catastrophe in populations is a mutation rate low enough to ensure that fatal mutations are compensated for by fecundity.Petrushka_{August 15, 2010
August
08
Aug
15
15
2010
01:44 PM
1
01
44
PM
PDT}

Petrushka - "It is quite possibe –actually it is common — to program a fitness function without knowing the end form." Yes, and in these cases, it is not the fitness function which is guiding the evolution. The fitness function is not the only place to sneak in information, just a common one. The most common way to put information into evolutionary algorithms is in the way in which the problem is parameterized. That is, by carefully choosing what sorts of things are parameters and what sorts of things are not. If I evolve a TV antenna where the only parameter I evolve is the color, it's not going to evolve anything worthwhile. But by constraining the evolutionary algorithm to things that are likely to produce useful antennas, you can evolve a new TV antenna. You might be interested in my UD article on parameterized evolution, or my more technical paper on the limits of evolvability.johnnyb_{August 15, 2010
August
08
Aug
15
15
2010
01:14 PM
1
01
14
PM
PDT}

Peepul: that's more or less correct, but obviously I don't mean simply "information", but "complex specified information", or better still "digital functionally specified complex information" (dFSCI), which is the form which I prefer for discussion.gpuccio_{August 15, 2010
August
08
Aug
15
15
2010
12:44 PM
12
12
44
PM
PDT}

If natural selection guides because it has the end form in mind and is selecting because it is close to the end-form.
It is quite possibe --actually it is common -- to program a fitness function without knowing the end form. In fact, commercial versions of the travelling salesman algorithm are used specifically because the end form is not known. The "target" is irrelevant in such algoirithms. All one needs to know is which child or children in a population have more of some global attribute. In biology, the attribute is reproductive success -- not something that can be stored in a database.Petrushka_{August 15, 2010
August
08
Aug
15
15
2010
11:37 AM
11
11
37
AM
PDT}

- in biological evolution, the fitness function is irrelevant because NS does not create information.
I've never understood this line of reasoning. Of course NS does not "create" information. The "information" about which biochemical configurations are viable is a property of chemistry. The "information" about which configurations are better at replicating is a transient property of the ecosystem. Variation and selection simply "feels out" the landscape and records its findings. The learning algorithm involved is not programmed. It is a property of any replicating system. One can certainly argue -- considering our current state of ignorance about OOL -- about how the first replicator arose, but replicators will evolve.Petrushka_{August 15, 2010
August
08
Aug
15
15
2010
11:30 AM
11
11
30
AM
PDT}

Peepul - The problem (which is not spelled out well in the paper you reference) is that fitness functions often sneak in information. For example, in the WEASEL program, the fitness function used the ideal form (i.e. "Methinks it is a weasel" or whatever the phrase was) as the fitness. This is not Darwinian evolution, but evolution *with a goal in mind*. If, instead, the fitness function made sure that all unintelligible phrases were unselected, then the WEASEL program wouldn't get off the ground. When the fitness function is built *for the purpose of leading to the goal*, then all someone has done is moved the design elsewhere - they have not gotten rid of it. A similar trick was used in Avida to get to the EQU algorithm - they loaded up the environment to make sure that the specific lineage to EQU was selected for. If natural selection guides because it has the end form in mind and is selecting because it is close to the end-form (and not based on immediate need) then that is ID. If natural selection guides just because it gets rid of non-fit creatures, then there is no reason to think that the fitness function would match the needed evolutionary path as nicely as Darwinism needs. In fact, the No Free Lunch theorem gives us reason to think that it shouldn't.johnnyb_{August 15, 2010
August
08
Aug
15
15
2010
11:18 AM
11
11
18
AM
PDT}

gpuccio, Can I summarise what I think you're saying? Tell me if I've got it correct. - evolutionary algorithms developed by humans do not create information because their fitness function is derived by humans - in biological evolution, the fitness function is irrelevant because NS does not create information.Peepul_{August 15, 2010
August
08
Aug
15
15
2010
11:13 AM
11
11
13
AM
PDT}

The only fact is that the fitness function in NS needs no external input, because it really does not create any new complex information. That is perfectly in line with what Dembski and Marks say.
If you substitute "structures" for information, your statement becomes testable. Then you need to account for structures like the mammalian middle ear,which seems to have evolved incrementally from jaw bones.Petrushka_{August 15, 2010
August
08
Aug
15
15
2010
10:41 AM
10
10
41
AM
PDT}

Peepul: The only fact is that the fitness function in NS needs no external input, because it really does not create any new complex information. That is perfectly inline with what Dembski and Marks say. If NS were really to create biological information, it would badly need an external input of information, otherwise it could not succeed in its task. That input of information is design. IOW, NS is not responsible for biological information. The input of information which makes biological information possible is a form of design. It's just as simple as that.gpuccio_{August 15, 2010
August
08
Aug
15
15
2010
10:25 AM
10
10
25
AM
PDT}

Thanks guys for your information, I appreciate it. I understand the scepticism you have that NS can create big enough changes to explain the diversity of body plans and signficant changes in form. For those changes that do occur, though, it seems that Dembski and Marks' statement that the fitness function contains information is not a problem for the evolutionary model. Whatever changes do occur, there need be no external input into the fitness function. Do you agree with that? If not, can you explain why? thanks againPeepul_{August 15, 2010
August
08
Aug
15
15
2010
09:41 AM
9
09
41
AM
PDT}

The point is, for NS to occur, the replicator must really be able to replicate better.
Actually, no. It must replicate. Not necessarily better. Neutral mutations and synonymous mutations have the same advantage as the unmutated allele.
The point is, in NS the environment is not in any way “aware” of the properties of the replicator. Nobody has any previous information about what kind of change will be able, in the short or in the long term, to give a reproductive advantage.
That limitation would apply to the designer also, unless the designer is omniscient, or capable of simulating protein folding more quickly and efficiently than chemistry, and also be capable of simulating entire ecosystems.Petrushka_{August 15, 2010
August
08
Aug
15
15
2010
09:37 AM
9
09
37
AM
PDT}

peepul: the way I see it is more or less as follows: 1) In GA, information is inputted in many different ways by the programmer. One of these is to set a fitness function which will deliver a very expected result in an easy way. The name for that is Intelligent Selection (IS). IOW, many GA work thorugh random variation (more or less targeted) and IS. And that king of GA can be very efficient to solve specific problems, provided that the appropriate fitness function be inputted. 2) What about Natural Selection (NS)? NS is in reality a tiny subset of IS: the subset where the only function which can be measured is reproductive success. Another difference between NS and IS is that in NS does not actively "reward" the replicator, while in GA it usually does. But the important difference is in the function which is measured by the system. The point is, for NS to occur, the replicator must really be able to replicate better. It must "self-reward" itself by its own merits, so to say, and only a true reproductive advantage will do the trick. That is an extremely limiting property of NS, because only changes which truly increase reproductive success will be "recognized" by it. On the contrary, In programming a GA, the programmer can choose which function or property will be measured and rewarded. That gives him immense potential power to obtain what he wants. Just to cite a very well known, and extreme, example, in a GA such as the famous Weasel, the previous knowledge of the solution makes it extremely easy to select it. The point is, in NS the environment is not in any way "aware" of the properties of the replicator. Nobody has any previous information about what kind of change will be able, in the short or in the long term, to give a reproductive advantage. Neither the replicator not the environment have any clue about that. Nor are the laws of physics or of biochemistry which cause the mutations in any way "targeted" in that sense, ot to obtain that result. IOW, any change can happen (they are, by definition, random). Now, if some changes can increase reproductive success in the short term, IOW if they are simple enough to be in the range of what random change can achieve, even in millions of years, then those simple changes can sooner or later occur and be selected, That's the case of microevolution, such as antibiotic resistance or similar models, where the "positive" change is in the range os one or two AA mutations. But when the change can be effective only "in the long range", IOW when tens or hundreds of AAs have to change in a coordinated a specific way for a new function to be expressed, and for that function to confer reproductive advantage, that will never happen, because all the intermediate steps of mutation cannot be "seen" by NS, and cannot be selected, fixed and expanded. To sum up, NS is a very blind and ineffective kind of selection, because it can recognized only a very minor set of variations, and cannot in any way build up new complex functions. Its only effect as positive selection is microevolution, in those few forms that we well know. Out of those models, it is totally ineffective. On the other hand, NS is very effective as a negative selection force. It can very effectively discard those mutations which can really hamper reproduction. So, NS is an important force in "defending" existing information, but it can scarcely modify it, and never in a complex way.gpuccio_{August 15, 2010
August
08
Aug
15
15
2010
08:51 AM
8
08
51
AM
PDT}

PS: Remember, break-point no 1 for biodiversity [cf also here] is the origin of dozens of phyla and subphyla at or about the time of the Cambrian revolution on the usual timeline. That is, on the conventional timeline [a widely accepted model framework of the past on earth], you have to account for the dFSCI etc for dozens of top level body plans, in a short window relatively speaking, on earth.kairosfocus_{August 15, 2010
August
08
Aug
15
15
2010
06:56 AM
6
06
56
AM
PDT}

PP: I see you appreciate the force of the OOL challenge, especially in light of the need for a metabolic sub system AND an information system based self-replicating facility. And, that is actually enough to ground a strong inference to design of life. Which then changes the picture for everything beyond that. Now, for MINOR changes, with less than say 250 - 500 DNA bases [out of 100+ k or 1 mn up], there is no in-principle problem on moving about within an existing island of function, which may well have one or more mountain ranges of optimal niches. (Though, observe Behe on observed limits of RV + NS in a particularly favourable case.) That gets you to some form of microevo, which is observed to some extent. And so far as I am -- and many others are -- concerned, I have no problems up to say roughly Families of animals and plant equivalents, on a case by case basis. (Note too how morphological stasis is a characteristic of fossil records.) Body plan origins is a second level of challenge. The problem is that life forms are on separate islands set by major body plans that require early embryological expression, and to get to a new island, you have to innovate ~ 10 mn bases of embryologically early bio-info, of the life form is not feasible. That is, you have to get to shores of function on archipelagos scattered across genome space. Islands separated by seas of non-functional configs of DNA etc. And with a config space challenge that has exploded vastly beyond the one to get to first life. This leads to the expectation that life forms -- fossil and living -- SHOULD not form a smoothly branching and grading tree of life pattern, opposite to Darwin's expectations. (He had protested that the fossil record was scanty and poorly investigated. That is no longer a credible claim. The gaps are real and dominant [despite the headlined "links" -- that tend to fade away after a time -- and mosaics -- that, e.g. the Platypus with its genes from all over the animal kingdom, actually undercut the claim and all homology/similarity and differential taxonomy arguments], hence expedients like Punctuated Equilibria.) So, body-plan level darwinian macroevolution is not only an inference beyond observations, but it is questionable on information issues, fossil record and mosaics as well as diverse molecular classification trees. The information requisites to get to islands of function, joined with those to get to first life, point to the design of the body plans. GEM of TKIkairosfocus_{August 15, 2010
August
08
Aug
15
15
2010
06:40 AM
6
06
40
AM
PDT}

Kairosfocus, thanks for your detailed reply. I understand the points you are making. I accept that the origin of the first replicator is a question that has not been resolved. But given a replicator and an environment, where exactly is the problem with the environment providing the information on the fitness function? Do you agree that this information is provided by the environment? If not, where does it come from? if it is provided by the enviroment, what is the relevance of Dembski and Marks' work to Darwinian evolution? it's this specific question I'm looking for an answer to.Peepul_{August 15, 2010
August
08
Aug
15
15
2010
06:02 AM
6
06
02
AM
PDT}

PP: Part of the problem is that the F/F and GA search approach is already within an island of function. So, you have a restricted use of randomness to see which way gives an uphill path to better performance, within an island of function. But the real problem being implicitly assumed away -- as the old joke about the economist on a desert island trying to get a can open puts it "assume a can opener" --is to get to the shores of such an island of function. If you aggregate the information-carrying capacity for the GA and the F/F, you will see that it is well beyond the dFSCI threshold, 500 - 1,000 bits. By far and away most arbitrary configs of that many bits will not land you on an island of function, and the config space is so vastly more than the search capacity of the observed cosmos across its thermodynamic lifespan, that you cannot carry out a credible search of he space as a whole to land you on initial function to get climbing. What Marks, Dembski and others have been pointing out, is that GA designers -- and GA's are known designs -- are actively inputting a LOT of information tracing to their foresight, intent and insight. Take that factor out, and on average you will not over-perform random search, which is in this case utterly hopeless. Now, of course, in Darwin's warm little pond full of salts and struck by lightning, or the equivalent, there is no intelligent designer to constrain things to be on islands of organised function. All the fantasy chemistry about RNA worlds, autocatalysis, etc simply does not get you from extremely dilute organic solutions with a lot of water around to hyrdolyse condensed, endothermic polymers, and it does not get you to organised systems of codes, algorithms and molecular machinery to effect it, to have a metabolising entity with a self-replicating facility. So, there is no base for differential reproductive success to get you to hill-climb. there is no credible root for the Darwinian tree of life. Which his convenient truncation of his theory at that point allowed him to evade addressing. Then, to branch the tree to make major body plans requires embryologically feasible sets of major mutations expressed early in development, dozens of times over. And requiring not 1,000+ bits worth of capacity, but credibly 10+ mn. Each. Codes, on our empirical observation, don't write themselves out of lucky noise. Algorithms, similarly, don't write themselves out of lucky noise. Data structures, too, don't set up conventions and protocols out of lucky noise. Nor --as thermodynamics (both classical and statistical) will tell us if we will but listen -- does implementing machinery come together in precisely coordinated organised configurations to effect the codes, algorithms and data structures out of lucky molecular motions. There is no free lunch. And, there is but one observed causal pattern for such active information and integrated, complex functional organisation. Directed contingency. AKA, Design. So, on inference to best explanation in light of observed causal patterns . . . GEM of TKIkairosfocus_{August 15, 2010
August
08
Aug
15
15
2010
05:29 AM
5
05
29
AM
PDT}

Hi all, can I ask a general question about ID? Not strictly related to this thread. I have been reading this paper by Dembski and Marks http://evoinfo.org/papers/ConsInfo_NoN.pdf They argue that the fitness function provides the information required to make evolutionary algorithms work, and that therefore evolutionary algorithms do not create new information. 'Searches that operate by Darwinian selection, for instance, often significantly outperform blind search. But when they do, it is because they exploit information supplied by a fitness function— information that is unavailable to blind search' I have a question about the fitness function. in Darwinian evolution, the fitness function is 'evaluate the reproductive success of this the organism in its environment'. If there is information in the fitness function here, it is not provided by a designer, but by the natural environment. Unless I assert that all aspects of the natural environment experienced by an organism were designed to the most detailed level (which is definitely untrue), there seems to be no problem here. Information required in the fitness function is provided by the environment. I don't see how this is a valid objection, then, to Darwinian evolution. What's the ID perspective on this question?Peepul_{August 15, 2010
August
08
Aug
15
15
2010
04:45 AM
4
04
45
AM
PDT}

Gil: I always like your posts, which get to the core of issues with admirable simplicity and efficacy. Keep on. Bruce David: There will come a time when people will shake their heads in wonder that any intelligent person in possession of the facts of biology could ever have believed that life could have arisen and evolved by natural causes. True. I hope I will be there to see at least a part of that.gpuccio_{August 15, 2010
August
08
Aug
15
15
2010
12:40 AM
12
12
40
AM
PDT}

1 2 Next

You must be logged in to post a comment.

Leave a Reply