Questions for Sean Pitman

Sean,

You are contradicting yourself now. You claim that language change is not a good analogy for biological evolution, yet you feel free to use it as one. You can’t have it both ways. Language either is or isn’t an informative analogy for evolutionary processes. Your comments about spelling and letters show that you don’t realize that language changes primarily via the verbal-auditory channel and not via orthography, or are allowing yourself to form sloppy arguments. Furthermore, you still don’t understand the significance of “batter.” Your “counterexample” of “cat” does not work because it does not fit the grammatical usage for the suffix “-er.” The hypothetical word “catter” makes no sense in English because “cat” does not refer to an action nor a tool or similar usable object. Because “bat” refers to a usable object and the action connected with the object then “batter” is grammatically understandable as “someone who bats.” Likewise, “catter” would means “someone who cats,” yet that makes no sense in English since cats is not a verb. Now, given that “cat” is a noun, “catter” could also refer to “someone who does something associated with cats;” however, that is more ambiguous unless more context clues are given. Yes appending the suffix “-er” to any word is not going to always make grammatical sense, but so what? Functional change only requires that it works for only a subset of all possibilities. It failing sometimes is a lot different than it failing everytime.

Perhaps if I use an example not found in real English you will be less likely to miss the point. Now “foo” is a new material that can be made from “mixt.” The verb “to foo” refers to the process of turning mixt into foo using. Now if I were to tell an English speaker that my brother is a fooer, he’d know to what I was referring given that he knew about fooing. Even if it had never dawned on him that someone could foo for a living he would instantly understand exactly what I was referring to. I suspect that most posters on this board are able to follow what I just said about the art of fooitry event if they had never studied foology or even read that chemical process of footion occurs when you put the prefoo in a foopot and add fooinators. That is what grammatical rules allow us to do. Although, this is the first time that you have either read or heard the above sentences you were able to follow along and understand them. Your knowledge of English grammar and vocabulary allowed me to introduce a novel concept with some related words and you were able to get the gist of what I am saying. Am I right? Of course, you’d understand even better if I took you to my brother’s foofactory when I told you this.

It’s ironic that you say that I am reaching when I describe language evolution, and then turn around and give a description of language change that is consistent with how I described it. Remember when I said that language change behaves similar to Ohta’s “nearly neutral” model for nucleotide/amino acid substitution. I suggest that you go read up on that subject before you contend to say that my comparisons are reaching.

I find it funny that you’d call genetic programming “bs.” The only reason you’d do so is because you are completely ignorant about the field. I suggest that you pick up a copy of the textbook, Genetic Programming, an Introduction by Banzhaf, Nordin, Keller, and Francone before you continue to assert that it is impossible to evolve computer programs via mutation and selection. In fact researches have even been able to use the same techniques to evolve hardware using the flexable architecture of FPGA. One experiment evolved a chip that could discriminate between 1kHz and 10kHz. It was more efficient than any one designed by engineers, but the researchers are still trying to figure out how it works. Another experiment sought to evolve a chip that produced an oscillating signal and the researchers got one. To their surprise an oscillator did not evolve; a radio did. That radio was “simply” stealing an oscillating signal from the air and outputting it. So for you to argue that evolutionary computing does not work shows your ignorance of decades of computer science research. Maybe next time you should do some research before you profess what is “bs” or not. Argumentum ex ignoratia is no way to operate in an honest, intellectual debate.
 
Upvote 0
RufusAtticus wrote: To you these examples and mechanisms of molecular evolution are not convincing because no one has observed them in the lab. Yet I do not see you applying this criterion to your own argument. Show me a real time experiment where a supernatural force was seen to guide a protein through a neutral gap. You have said that “naturalism” cannot explain the origin of new functional proteins. But this is simply not supported by your argument. All you have tried to show is that selection and drift can’t explain it. Basically you have assumed that there are only two sides to the issue: “nature as we currently understand it” or “the supernatural.” Sure you can try to disprove the first, but that will not prove the second. There is a third option, “nature as we don’t currently understand it.” Basically your entire argument is deus ex ignoratia dressed up in molecular biology clothing and wearing a crown of bad statistics. [/B]

In my opinion, your hypothetical examples of molecular evolution are not convincing because they are statistically unlikely if not impossible based on my current understanding of naturalistic processes.  If they were statistically likely and or possible, they should be able to be tested in real life. Yet, when they are put to the test, they fail. Even a convincing theoretical model is lacking. 

You see things differently, and put your faith in this "third option".  Perhaps you do this because you see and understand things more clearly than I do?  Or, perhaps you do this because you are so devoted to the doctrine of Naturalism that you refuse to even consider the possibility of intelligent design?  For you, there seems to be only one possibility... Naturalism.  There is no need to consider other possibilities because Naturalism must answer all questions.  This has been predetermined in your mind so that there is no need to consider other options.  Other options are wrong by default.  And yet, how is this scientific? It seems rather religious to me.  You have faith that even if naturalistic mechanisms are not understood now, that they will be found at some future time.  This sounds an aweful lot like the arguments I hear from religous people concerning their faith.  Even though they don't know how or why, they believe by "faith" without a need for evidence.

Certainly naturalism may be "true", but there is no way of completely verifying it this side of eternity.  Science is all about the weight of evidence, not the finding of absolute evidence.  Therefore, for scientists to only consider naturalistic theories as potential explanations for various observations is not really being as objective as possible.  Any assumption that is placed upon the scientific method biases the result in favor of the preconceived notions that are brought to the table.  The test is therefore weakened and may even be invalidated completely by such biases.  Often biases are subconscious, but they still affect the validity of the test.   So, to deliberately decide that all outcomes of all tests will support a particular hypothesis ahead of time, is no less than scientific heresy. 

Most scientists seem uneasy with any theory that does not have its basis in naturalism for fear that the only alternative to naturalism, intelligent design, might bring back the darkness of superstition.  Many in fact propose that the scientific method cannot, by its very nature, detect anything but the workings of a mindless nature.  The detection of intelligent design is therefore beyond the realm of science and the scientific method.  However, many of these same scientists hope to find evidence of intelligent life in the universe beyond our own world.  Even within our own world, entire scientific disciplines, such as forensic science, are based on discovering the workings of purpose and intelligence without having ever met the author of the work they are investigating.  Clearly then, scientists seem quite confident in their abilities to detect intelligent activity as long as it has nothing to do with the origin of life or the workings of the universe.  

Is this not a contradiction?  How can purposeful intelligence be detected on the one hand and yet be beyond the realm of science on the other?  How are scientists so sure that if purposeful intelligence was involved in the origin of life on this planet that it could not be detected?  Upon what is this hypothesis based?  Is there evidence to support this claim?  Or, is it simply a philosophical position?

You say that questioning naturalism is not the same thing as supporting the design hypothesis, but I think that it is.  Inference of design can only be done when all other possibilities have been excluded.  But, you say that it is impossible to exclude the "third option" (ie: Even though we don't know how and even though it seems unlikely, according to our current understanding, some naturalistic process still might have done it). 

Please, you call yourself a scientist?  Upon what basis is this third option an automatic default?  Why do you make the a priori assumption that naturalism must be the only answer?  To say that natural causes could be the answer does not explain how they could be the answer when, for now, it seems as if they are not.  It might be fine to keep trying to find a naturalistic explanation for certain phenomenon, but until such an explanation is found, the assumption of naturalism is not an automatic default. 

For example, if I see a broken window as I walk by a house in the morning, what should I think when I see that the window is fixed when I walk by the house in the afternoon?  Should I automatically assume a naturalistic explanation?  Perhaps there is some law of nature that I am not aware of that fixes windows?  Certainly I must at least consider that perhaps there is such a force in nature.  But, until I am aware of it, the only other option I have is a "supernatural" option where some intelligent force fixed the window by deliberate design. 

You ask me to show you where neutral genetic gaps have ever been crossed by supernatural/intelligent agencies.  This is easy to demonstrate. It happens all the time.  Human scientists are in fact able to cross these genetic gaps.  Using higher intelligence and reasoning capabilities they are now starting to create novel genes with novel functions... so called, "designer genes."  :cool:  These designer genes were in fact created with supernatural intelligent design.  Yes, the creations of human beings are "supernatural" creations since they are above purely naturalistic processes.  The same thing is true for computer programming.  Your examples of computer evolution are nothing more than the evolution of functions with in a very small area of flexibility.  However, when it comes to functions such as calculator programs, word processing programs, photo editors . . .  comptures will never be able to evolve such functions using mutations and function based selection.  Yet, humans are able to cross the neutral gaps and create such software functions in short order via the "supernatural" power of intelligent design.

The coded nature of living things is very similar to information systems that have been designed with the use of higher intelligence beyond naturalistic mechanisms.  There is a known method for the creation of such systems of function and it involves higher intelligence.  There is no known naturalistic mechanism to explain such systems that has ever been demonstrated in real time.  So, just because such a naturalistic mechanism might be out there somewhere undiscovered, we must believe in it?  Please.  Science doesn't work that way.  Science is supposed to put value on the weight of current evidence and not on the hope of some future evidence.

"... An intelligible communication via radio signal from some distant galaxy would be widely hailed as evidence of an intelligent source.  Why then doesn't the message sequence on the DNA molecule also constitute prima facie evidence for an intelligent source?  After all, DNA information is not just analogous to a message sequence such as Morse code, it is such a message sequence." (Thaxton, Charles B., Walter L. Bradley and Robert L. Olsen: The Mystery of Life's Origin, Reassessing Current Theories, New York Philosophical Library, 1984, p. 211-212.) .


Sean

http://naturalselection.0catch.com/Files/Methinks%20it%20is%20Like%20a%20Weasel.html


http://naturalselection.0catch.com/index.html
 
Upvote 0
Hmm how would one define intelligent design? Human intelligence? Or do you think that you can actually understand the intelligence of God to call his design an intelligent one? I think using a naturalistic approaching is a very "physical" way to explain our world instead of abstract metaphysics, which comes in handy when we have to "subdue" our world.
 
Upvote 0
21st February 2003 at 05:20 PM RufusAtticus said this in Post #61

Sean,

You are contradicting yourself now. You claim that language change is not a good analogy for biological evolution, yet you feel free to use it as one. You can’t have it both ways. Language either is or isn’t an informative analogy for evolutionary processes.


I'm not contradicting myself.  I said and I still say that language is a great analogy for the genetic code and the coded system of living cells.  Where we disagree is how codes evolve.  Human language is a coded system just like the information contained in DNA.  However, humans are gifted with higher intelligence.  Humans can manipulate not only their own language codes, but also genetic codes as well.  They do not need to rely on random mutation and natural selection to do the changing for them. 

On the other hand, individual cells are not creative or intelligent. The are like computers running on algorithms.  They are not intuitively creative.  They can only do what they are programmed to do.  They can evolve certain limited functions over time, but this evolution is quite limited as previously detailed. 

So you see, there is a huge difference between human language evolution and genetic evolution since intelligence is involved on the one hand but not on the other.  Human language evolution is not a purely naturalistic process, but genetic evolution must be.  For example, by themselves, computers cannot evolve their own software beyond very limited restrictions.  They are not creative or intelligent.  They can only do what they are programmed to do.  They cannot increase their own programming to any significant degree without the help of outside intelligence and creativity from humans.  Humans can evolve computer code, but computers cannot evolve their own codes very much at all. 

So you see, wherever humans are involved, a purely naturalistic process is not being demonstrated.  Human language evolution does occur because it involves humans who are intelligent.  Computer evolution does not occur without humans because computers are not creative or intelligent.  Living systems are no more intelligent than computers are.  So, their evolution must rely on purely naturalistic processes without the aid of outside intelligence be it human or other sources of intelligence and creativity.  The codes are the same, but the processes of evolution are different.  Genetic language evolution must be purely naturalistic while human language evolution involves intelligence.

Furthermore, you still don’t understand the significance of “batter.” Your “counterexample” of “cat” does not work because it does not fit the grammatical usage for the suffix “-er.”

Do you not see that you are simply restating my own argument here?  Only those letter sequences that are defined in a particular way will be recognized by an English speaking person.  The word "batter" is recognized because of the way in which it is defined by the English dictionary.  The word "bat" is a noun.  Likewise, the word "cat" is a noun.  Someone not familiar with the English language might think that the suffix "er" could equally be attached to both cat and bat to give new meaning.  However, it simply does not work this way because there is no "catter" defined in the English dictionary. Of course, you argue that the rules of the English language would still give meaning to the word "catter" as someone who does something with cats even though this word is not found in the English dictionary.  Ok, this is a reach, but I'll give it to you.  Even given this of sort leway for giving words function, it doesn't help you much.  It adds a few more potential functions to the list.  However, the list of non-functional sequences is so much larger that it really doesn't matter. 

For example, lets take a sentence like, "It really does not matter."  How many different place would the addition of the letters "er" be meaningful in this sentence?  Is "iter" a word?  Is "reallyer" a word?  Is "doeser" a word?  "Noter" might be a word according to your definition of words, but does it have meaningful function in our sentence?  No.  Also, who is to say that the letters "er" will get pasted at the end of words?  They could get mutated anywhere in the sentence.  The sentence could read, "It reerally does not matter."  Now, the function of this sentence is really messed up. 

You see the problem don't you?  Do you understand now how "foo" is not going to help you here?  Certain mutations may in fact have meaning, but the vast majority will not have meaning.  You are talking about a small minority of words that can use the suffix "er", but I am talking about the vast majority of sequences that cannot use "er" anywhere.  It is the ocean of non-function that is the problem, not the small puddle of function that you are talking about.


I find it funny that you’d call genetic programming “bs.”

I did not call genetic programming "bs".  I called computer software evolution "bs".  Certainly computers can mutate and evolve novel functions as long as these functions are within a limited range of complexity.  The examples of computer evolution that you described are very much guided by human intervention and are limited to a relatively small range of options.  In other words, their "evolution" is directed.  Without human input, a computer could never and will never evolve its own software programs via mutation and function based selection. 

Give me the answer to this question, yes or no:  Can anyone program a computer to evolve a software program, like a photoediting program, using only mutation and function based selection?  If you answer "yes" to this question, then this begs the follow-up question, "Then why do we still need computer programmers?"  Computers are so much faster and can "remember" so much more information than humans can remember, and yet they still need our help to come up with new functions?  Why is this?  Because really truly, computers are dumb.  They are not intelligent or creative.  They cannot evolve their own software without human help... Period.  If you ever come up with a computer program that will allow computer software evolution outside of human guidance, then you will be richer than Bill Gates.

So for you to argue that evolutionary computing does work shows your ignorance of decades of computer science research. Maybe next time you should do some research before you say that computers can evolve in the same way the life forms supposedly evolved the fantastic functions that we see all around us.  Clearly, if computers could evolve in the same Darwinian fashion, their lightening speed would have allowed them to evolve at a very rapid rate, and yet they are still relatively helpless without human input?  You say that anything that can undergo random mutation can evolve new functions.  If so, then all you would have to do is make a computer program to produce random changes in computer code and it would come up with all kinds of brilliant new functions all by itself.  Computers really wouldn't need human programmers.  All a person would have to do is go to the store and by a computer without having to worry about software to run the computer.  The computer would only have one program... the mutation program.  The person would just have to tell the computer, "Evolve me a word processor" and the computer would do it in short order.  Hey, why not?  What's the problem with this idea? 

Please... your argument here is exceptionally weak.  Argumentum ex ignoratia is no way to operate in an honest, intellectual debate. :)


Sean
 
Upvote 0
Sean,

My point about the “third option” was completely lost on you. I am not putting stock in it, because I don’t have to. The known natural mechanisms for evolutionary change are enough to account for the diversity of life on this planet. As I’ve demonstrated, your hand-waving and badly formed statistical arguments are hardly proof against evolution. The importance of the “third option” is that your argument rests on the disproof of one thing (selection, drift) being evidence for another (divine input). The problem is that it argument is not logically consistent. Disproving “A” is only proof of “B,” if “B” is the converse of “A;” otherwise, there is a possibility that both could be wrong. (That’s simple set theory.) The problem with your argument is that you haven’t proved that “god” is the converse of “selection+drift.” As such, your argument utterly flawed from the beginning. Deus ex ignoratia didn’t work for lightening, planetary motion, or disease; it wont work for biology.

I find it interesting that in order to preserve your argument for supernatural intervention you try to claim that human actions are “supernatural,” which is an obviously flawed argument is humans are natural beings not supernatural beings. You obviously aren’t familiar with much research into the nature of information systems if you want to claim there is no known “natural” mechanism for developing a code, without intelligent intervention. If you have a collection of possible codes, a collection of “things” that they can refer to, a population of code sets that reproduces, and a fitness function that rewards individuals with similar linkage of codes to things, then the population will converge to a consensus code set. These code sets are rarely optimal and often have multiple codes that refer to the same thing and often don’t have any code that refers to some things. With the genetic code we have the pre-requisites: nucleotides that act as codes, amino-acids that act as things, reproduction, and fitnesses being determined by similarity in peptide sequences produced. So it is abundantly clear that unintelligent processes can produce an information system, like the genetic code.

Furthermore, you are still making the same mistakes about the evolution of language. You are stuck on the idea that languages change happens by the changing of vocabulary, i.e. it involves conscious decisions about signal-meaning associations. But that is wrong because a language is not its vocabulary set but rather the collection of phonemes, grammar, syntax, and other rules that specify the structure and nature of the language itself. Changes in vocabulary are a minor factor in the evolution of language. Two languages can have the same vocabulary but be mutually unintelligible if any of the other features vary greatly. It is clear from linguistic research that errors occur when a child acquires a language. (Mutations in reproduction.) These are not intelligent choices to change the language, but simply errors. It is also clear that whether such errors go to fixation is governed by chance. (Substitution caused by drift.) Claiming that a brain is involved doesn’t make it an intelligent design process. If you want to do so, you will have to show how on toddler acquisition of language, which you’ve ignored so far, is a conscious design process.

Yes, the suffix “-er” would not produce a “functional” word on the majority of English words. But so what? Evolution doesn’t require it be functional most of the time, but only functional some of the time. Selection can promote even the rarest of mutations. Thus in arguing your numbers game you have completely forgotten what evolution is about.

In fact, you did call genetic programming “bs” because genetic programming is computer software evolution. Look it up if you don’t believe me. None of my examples involved human intervention to produce the final results. If you don’t believe me, read the papers for yourself. You have also argued that it is limited in what it can do making appeals to word processors, image editors, and similar GUI oriented software packages. But this ignores the fact that the majority of software programs and the majority of algorithms used in computer programming have nothing to do with human interaction. (Ever heard of UNIX? Ever programmed?) Genetic programming requires large populations of software programs, a fitness function, and many generations. Now to evolve a computer program that includes human interaction, then human evaluation needs to be included in the fitness function. Of course if you are doing a population of thousands of programs over thousands of generations, that amounts to millions of man hours involved. You can’t just simply ask for a word process and expect to be given one that suits your needs at the drop of a hat. Remember it took 3.5 billion years of evolution to produce something as complex as humans. Yet you think the success of genetic programming is limited simply because it hasn’t given you a word processor in 10 years. Please. . . .
 
Upvote 0
Part 1 of 2 Reply

Yesterday at 10:15 PM RufusAtticus said this in Post #65

My point about the “third option” was completely lost on you. I am not putting stock in it, because I don’t have to. The known natural mechanisms for evolutionary change are enough to account for the diversity of life on this planet.


What "mechanisms" are these?  You have failed to explain the mechanism by which neutral gaps in genetic function are crossed.

As I’ve demonstrated, your hand-waving and badly formed statistical arguments are hardly proof against evolution.

You haven't demonstrated anything.  You have tried to say that the destruction of functions or the changing of allelic frequencies in a population are examples of evolution in action.  Certainly these are changes, but these examples do not explain now novel functions evolve.  The few examples of the evolution of novel functions that there are do not cross functional gaps any wider than one or two point mutations.  You are basically explaining how to break a window but have failed to explain how to fix a window via naturalistic processes.

The importance of the “third option” is that your argument rests on the disproof of one thing (selection, drift) being evidence for another (divine input). The problem is that it argument is not logically consistent. Disproving “A” is only proof of “B,” if “B” is the converse of “A;” otherwise, there is a possibility that both could be wrong. (That’s simple set theory.) The problem with your argument is that you haven’t proved that “god” is the converse of “selection+drift.” As such, your argument utterly flawed from the beginning. Deus ex ignoratia didn’t work for lightening, planetary motion, or disease; it wont work for biology.

Your problem with this argument is that you really haven't come up with a "third option" at all.  Your third option is that a naturalistic explanation might still be found even if it might not be known currently.  Really, there are only two options.  Either there is a naturalistic cause, or an intelligent cause.  There is no viable third option. 

Your argument here is also interesting in that you are in fact capable of infering design all the time.  If you see a broken window one day and then see that it is fixed the next day, you automatically infer design. Why?  Because there is no known naturalistic mechanism for fixing windows.  The same methods would be used in detecting intelligent life elsewhere in the universe, which scientists are trying to do all the time.  If these methods of detecting design can be used, theoretically, to detect alien forms of intelligence, then the inference of intelligent design is not an "impossibility" as your argument seems to be indicating. 

Why is it then that when the topic concerns the origin of life itself that the option of intelligent design cannot be considered?  Why is this devotion to the doctrine of Naturalism required?  I mean, even the very act of questioning the idea of naturalism cannot be done, according to you, because that wouldn't rule out naturalism or support the other option of design.  Well, according to the scientific method, if a hypothesis is not falsifiable, then it is not scientific.  The hypothesis of design certainly is falsifiable.  All one has to do is demonstrate a naturalistic cause.  However, it is a lot harder to falsify naturalism because even if a naturalistic cause cannot be found now, one could always argue that someday such a cause will be found.  Well, this is not science, but faith. 

It is therefore perfectly reasonble to argue that if problems with a naturalistic explanation are found that intelligent design might have been involved.

I find it interesting that in order to preserve your argument for supernatural intervention you try to claim that human actions are “supernatural,” which is an obviously flawed argument is humans are natural beings not supernatural beings.

What is your definition of "supernatural."  My argument is one of intelligent design.  Humans are intelligent designers.  You say that humans are "natural beings."  What the heck does that mean?  Humans are clearly capable of manipulating their environment in an intelligent and creative manner.  This creativity goes beyond natural laws and mechanisms that supposedly govern molecular evolution.  However, molecules are not intelligent.  Neither are genetic systems or computers.  They are "mindless" entities.  They are dependent for change on completely random mutations combined with the selection of a mindless nature without forsight. Human intelligence is above such mindless methods.  Human thinking is cearly "superior" to such mindless natural processes.  As such, human intelligence can be thought of as "supernatural."

You see my point don't you?  I am arguing for the idea of intelligent design regardless of where that intelligence came from.  I am arguing for the idea that there is evidence of an intelligent orgin for living things... evidence of a thinking, directed, intelligent mind not unlike our own except in degree. 

If you have a collection of possible codes, a collection of “things” that they can refer to, a population of code sets that reproduces, and a fitness function that rewards individuals with similar linkage of codes to things, then the population will converge to a consensus code set.

If you have a collection of symbols and a collection of "things" that they can refer to, how do you decide which symbols refer to which things?  For example, DNA is made up of four different chemical symbols.  How the heck were these symbols attached to various amino acids when they could just as easily been attached in meaning to a host of other things, such as sugars, or even English language functions?

You see the problem don't you?  Just because you have a bunch of symbols doesn't mean that you have made a code.  A code must attach symbols to idea or objects that are unrelated in any way other than the fact that they have been arbitrarily attached to each other by an outside entity who has choses such connections in order to communicate ideas.  Codes do not self assemble.  The letters in the English alphabet are completely meaningless without humans giving them arbitrary function.  By themselves, they have no language or code function whatsoever.

With the genetic code we have the pre-requisites: nucleotides that act as codes, amino-acids that act as things, reproduction, and fitnesses being determined by similarity in peptide sequences produced. So it is abundantly clear that unintelligent processes can produce an information system, like the genetic code.

Come again?  How did this little paragraph make it "abundantly clear" how the genetic code came to be via naturalistic mechanisms?  It sounds to me like you are arguing that, "Given the symbols and the functions, all in the same place, the code will self assemble."  Oh really?  Then do it.  According to you, it sounds so simple.

The fact of the matter is, language codes never spontaneously assemble themselves. When they are created, their formation is always directed by an outside intelligent agency.  A self assembly of a functional language code has never been demonstrated... period. 

The reason for this is that codes are, by nature, arbitrary constructions. The genetic code is an arbitrary construction.  There is nothing to say that DNA must code for amino acids since the same molecules could have been set up to code for absolutely anything else.  It all depends upon who or what is "reading" the code.  Living cells just so happen to be set up to understand DNA in a certain way and the DNA just so happens to be set up so that the reading system of the cell can understand the message sequence in the DNA.  However, this sequence can be changed so that the cell cannot read it or understand what it is saying.  There is no reason why DNA must read in the way that it does. Also, there is nothing to say that codons must be 3 base pairs in length.  The code could have been set up to use a binary system like computers use.  Or, it could have been set up to use 4 or 5 or 6 or any number of other sequence groupings. 

Really, the genetic code is quite arbitrary, and that is its problem.  Since it could have been set up in any number of various ways, the fact that it is like it is, is quite amazing.  There is no known naturalistic method to explain its existence.  It has never been reproduced in the lab and I am betting that it never will be.  Why?  Because, there are so many more non-functional ways in which the molecules could have arranged themselves that the spontaneous assembly of a functional code is, for all practical purposes, impossible.  Sir Frederick Hoyle (Big Bang Theory) came to a similar conclusion.  In fact, he thought that the odds against the self assembly of the genetic code was so unlikely that he proposed that life could not have originated on this earth, but must have come from "outer space."

End: Part 1 of 2 Reply
 
Upvote 0
Part 2 of 2 Reply to RufusAtticus, Post #65

Furthermore, you are still making the same mistakes about the evolution of language. You are stuck on the idea that languages change happens by the changing of vocabulary, i.e. it involves conscious decisions about signal-meaning associations. But that is wrong because a language is not its vocabulary set but rather the collection of phonemes, grammar, syntax, and other rules that specify the structure and nature of the language itself. Changes in vocabulary are a minor factor in the evolution of language. Two languages can have the same vocabulary but be mutually unintelligible if any of the other features vary greatly. It is clear from linguistic research that errors occur when a child acquires a language. (Mutations in reproduction.) These are not intelligent choices to change the language, but simply errors. It is also clear that whether such errors go to fixation is governed by chance. (Substitution caused by drift.) Claiming that a brain is involved doesn’t make it an intelligent design process. If you want to do so, you will have to show how on toddler acquisition of language, which you’ve ignored so far, is a conscious design process.

Again, you seem not to understand the involvement of human intelligence in human language evolution.  I clearly understand that mistakes can be made, accidently, unknowingly, in the reproduction of a word.  However, such accidental mistakes or mutations will remain meaningless as far as interhuman communication unless someone else makes the connection as to what this person is trying to get across with this mutated word.  This connection or determination between a mutated word and an idea requires intelligence and intuition.  Since a perticular mutated sequence is not represented in the language code or at least is used in a non-coded manner, the ability to determine the intent of the speaker and attach new meaning to the mutated word, requires higher intelligence. 

For example, a child might point to a dog and say, "doy doy".  Of course, there is no coded meaning to "doy doy" in the English langauge.  However, the parents of this child, being creative in their ability to bring a great many ideas and connections together in a creative manner, most likely would be able to understand what the child is trying to say.  If they agree to use this particular word with this child, they create their own language code for that word.  Doy doy did not simply evolve its own function by naturalistic processes.  Intelligence was required, both on the part of the child, to connect this word with an idea, as well as on the part of the parents to understand this connection.  Human language always "evolves" in this manner.  Intelligence is always involved in human language evolution.

Yes, the suffix “-er” would not produce a “functional” word on the majority of English words. But so what? Evolution doesn’t require it be functional most of the time, but only functional some of the time. Selection can promote even the rarest of mutations. Thus in arguing your numbers game you have completely forgotten what evolution is about.

After all this time you still don't get it do you?  This is the whole problem in a nutshell.  If you understand that the addition of "er" would not produce a "functional" word in the majority of English words or phrases, then you should understand the problem of functional gaps.  You say that the theory of evolution only requires a mutation to be functional some of the time.  Yes, this is true, but the rarer the function, the more time is required for its evolution.  Evolution can only promote something if it is functionally different from what came before.  It cannot promote a rare function until that function evolves.  Even if that function is very close to being functional, nature will not see it until it actually is functional.  Therefore, with each neutral mutation that is required before the function is realized, the time needed to cross this neutral gap increases dramatically.

For example, consider the word, "dirmzpt."  Changing any one of the letters would still result in a non-fuctional word.  Compared with the parent word, both words would be equally non-functional.  Therefore, the mutational change would be neutral since both words would have the same non-functional function.  Nature would not be able to select for one over the other.  Further drift is required to achieve some sort of function in the English langauge.  Human intuition is not allowed.  Only the mindless detection of function by a mindless non-intuitive English language code. 

The same thing happens with genetics.  Depending on the rarity of a particular functional code in "Mendel's Library" of functional codes, a non-functional sequence might have to drift around a very very long time before comming across such a sequence.  Until this time, there is no selective advantage given by nature to "guide" it in its drift since nature only selects for functional changes.

Now to evolve a computer program that includes human interaction, then human evaluation needs to be included in the fitness function. Of course if you are doing a population of thousands of programs over thousands of generations, that amounts to millions of man hours involved. You can’t just simply ask for a word process and expect to be given one that suits your needs at the drop of a hat. Remember it took 3.5 billion years of evolution to produce something as complex as humans. Yet you think the success of genetic programming is limited simply because it hasn’t given you a word processor in 10 years. Please. . . .

Even if you had trillions of years, you still couldn't evolve a relatively simple computer program, such as a four function calculator program, without human guidance.  Even with human guidance it couldn't be done if the human guidance was limited to mutational selection based only on function.  For example, take a computer scientist and place him/her in front of a computer and have the scientist select every mutation that is closer to a perticular desired function, but allow the scientist to only see changes in actual program function, not program code.  Since computer code is arbitrarily attached to computer function, the scientist would not be able to see mutations in the code that did not result in some change in computer function.  Only those mutations in the code that resulted in something the scientist could see in the monitor would be selectable.  Starting with a screen saver program, the scientist would never ever, even given trillions upon trillions of years, be able to get the computer to evolve a relatively simple four function calculator program. 

The "human needs" of the scientist, included in the "fitness function", would not be enough to overcome the problem. Why? Because, like nature, the scientist would be blind to neutral mutations.  When you're blind, it doesn't matter what your needs are, you probably won't find them because you must search through so much worthless stuff without direction that the few helpful things you might come across and "see" are extremely rare.

That is the problem.  That is what you haven't explained.  Your math doesn't seem so clear to me, but it seems clear enough for you to be able to understand this problem (I do have further questions for your about your statistical calculations which I will type out shortly).  Until then, please, try and explain the crossing of functionally neutral gaps using a process that only recognizes functional change.

Sean

 
 
Upvote 0

Morat

Untitled One
Jun 6, 2002
2,725
4
48
Visit site
✟12,690.00
Faith
Atheist
Even if you had trillions of years, you still couldn't evolve a relatively simple computer program, such as a four function calculator program, without human guidance. Even with human guidance it couldn't be done if the human guidance was limited to mutational selection based only on function. For example, take a computer scientist and place him/her in front of a computer and have the scientist select every mutation that is closer to a perticular desired function, but allow the scientist to only see changes in actual program function, not program code. Since computer code is arbitrarily attached to computer function, the scientist would not be able to see mutations in the code that did not result in some change in computer function. Only those mutations in the code that resulted in something the scientist could see in the monitor would be selectable. Starting with a screen saver program, the scientist would never ever, even given trillions upon trillions of years, be able to get the computer to evolve a relatively simple four function calculator program.

Speaking as a computer scientist: BULL. You want to do it? Easy. First, get rid of the human. You don't need him as a fitness function. Computers can do it easier. All you have to do, as a fitness function, is a check to see if random inputs, with random operations, equal the right answer.

Heck, it's not even a good example, because of the nature of programming! Mathematical operations are pretty common, so an evolving code would rather easily stumble on the pluses and minuses and equals needed.

Sean, I've writtend GA programs that evolved out answers quicker than I could have solved them. From simple "Weasel" programs (evolving out specific strings through simple mutation and reproduction) to more complex algorithms (finding the approximate solution to certain math problems).

Yet I all I needed to get my answer was a fitness function, reproduction, and mutation. That's it.

Copying the answers from generation to generation, the ability to decide which are closest to the "solution" and the ability to introduce random errors.

Of course, since I happen to understand why sex is so advantages, I did allow sexual reproduction, not asexual.

I've watched completely random strings evolve into plain English, Sean. That **** about "no new information" doesn't work on someone who not only understands information theory (you know, it's sort of important in computers) but has actually watched a program accumulate new information.

There was one truly ingenious sort algorithm I developed that way.....I looked it up later and saw that I'd evolved a salt-shaker sort. Which, according to the book I looked it up in, was the maximal solution to the type of sort I was doing.

I didn't know the algorithm to a salt-shaker sort. I certainly hadn't coded it. But there it was.....if messy. (There was a lot of extraneous code. That's evolution for you).
 
Upvote 0

Smilin

Spirit of the Wolf
Jun 18, 2002
5,650
244
57
Appalachia, The Trail of Tears
Visit site
✟15,906.00
Faith
Atheist
Marital Status
Married
21st February 2003 at 08:20 PM RufusAtticus said this in Post #61

I find it funny that you’d call genetic programming “bs.” The only reason you’d do so is because you are completely ignorant about the field. I suggest that you pick up a copy of the textbook, Genetic Programming, an Introduction by Banzhaf, Nordin, Keller, and Francone before you continue to assert that it is impossible to evolve computer programs via mutation and selection. In fact researches have even been able to use the same techniques to evolve hardware using the flexable architecture of FPGA. One experiment evolved a chip that could discriminate between 1kHz and 10kHz. It was more efficient than any one designed by engineers, but the researchers are still trying to figure out how it works. Another experiment sought to evolve a chip that produced an oscillating signal and the researchers got one. To their surprise an oscillator did not evolve; a radio did. That radio was “simply” stealing an oscillating signal from the air and outputting it. So for you to argue that evolutionary computing does not work shows your ignorance of decades of computer science research. Maybe next time you should do some research before you profess what is “bs” or not. Argumentum ex ignoratia is no way to operate in an honest, intellectual debate.

Hey Rufus,
I caught wind of the oscillator experiment, do you have a link to the research? as well as the frequency analyzer?
(intrigued electrical engineer here)
 
Upvote 0