Evolution, creation, and some code.

One of the perpetual, immortal errors creationists make is to say, e.g., “An airplane couldn’t arise by pure chance!” (The probability of that happening is such that the life of the universe wouldn’t be long enough for it to have any reasonable chance of occurring).

Of course, evolutionary theory doesn’t say that complicated structures arise all at once and by pure chance. Rather, they arise over time by cumulative selection, i.e., by the retention of beneficial random changes, and the discarding of deleterious ones. You don’t start with “chemical soup” and get to “human being” in one extremely lucky step. You start with “chemical soup” and get to “human being” in a gazillion cumulative steps.

In light of that... I wrote the following GAUSS code to implement the "typing monkeys" idea mentioned in Shermer's Why Darwin Matters, pages 82-83. The idea is to get a computer to randomly produce one of the most famous phrases in all of English literature, “To be or not to be.” The code starts with a random string, keeps the letters that match the target string, and tries again randomly for those that don’t. In other words, the process is not pure randomness; it’s randomness with selection.

Shermer reports that the researcher who originally thought of this had it take average of about 335 iterations, which on his machine at the time was 90 seconds. Here it takes an average that’s south of 100 iterations, and it's veritably instantaneous.

I have 28 characters included: 26 letters plus the underscore (used as a space) and the apostrophe (because it was not worth the trouble of excluding it). The target string is 18 characters long, including spaces, so there are 28^18 = 1.12 x 10^26 possible strings. Therefore, if I'm thinking about this correctly, the average number of tries it takes to produce the string by pure chance should be half that, or about 5.6 x 10^25 = 56 septillion.

With selection, as implemented in the following program, the mean number of iterations it takes to arrive at the string is 82.8. That's not 82.8 septillion. It's 82.8.

Anyone who might have Gauss is welcome to grab this code and run it. A note: Watch for the html tags. I.e., in order to get the “less than” and “greater than” signs to render correctly on the Net, I had to substitute their html codes & g t ; and & l t ; (otherwise LT would interpret them as an attempt to open/close an html tag). So if you want to run this code, you’ll have to find them and change them by hand before starting (there are only 3 “less-than”s and 1 “greater-than” in the code).

new;
output file=c:\Darwin.out reset;
cls;

@ Define the target string, in ASCII, from Hamlet's soliloquy: @
solil=116~111~95~98~101~95~111~114~95~110~111~116~95~116~111~95~98~101;

@ Generate a string of random numbers: @
x=rndu(1,18);
y=x*26+95; @ This puts the numbers into the useful range of ASCII codes. By the way, note this line should cause a non-conformability error, but it doesn't. Gauss assumes you want to add 95 to each element. @
z=trunc(y);
@ Apply the "chrs" command to turn the numbers into a text string and display it: @
chrs(z1,1~z1,2~z1,3~z1,4~z1,5~z1,6~z1,7~z1,8~z1,9~z1,10~z1,11~z1,12~z1,13~z1,14~z1,15~z1,16~z1,17~z1,18);
"";

j=1;
do while j<1000;

h=1;
do while h<19;
if abs(z1,h-solil1,h)>0.5;
monk=rndu(1,1); @ This random variable is dubbed "monk" because this is where the typing monkeys do their thing. @
z1,h=trunc(monk*26+95);
endif;
h=h+1;
endo;
"";

chrs(z);

if abs(z-solil)<0.1;
break;
endif;

j=j+1;
endo;

"";
"The monkeys produced the string in " j-1 "iterations.";

EDIT: Aargh! Brackets won’t render because LT thinks you’re trying to create a touchstone.

Interesting posts Carny. I don't have the skill set to use the code but the concept is entertaining!

I have no idea what this is all about. Wish I did! However I thought I would share a quote that I particularly like and that I use in the neurobiology class I teach. It came from a review on how the eye functions:

“Charles Darwin famously wrote that the eye caused him to
doubt that random selection could create the intricacies of
nature. Fortunately, Darwin did not know the structure of the
retina: if he had, his slowly gestating treatise on evolution might
never have been published at all.”

--- R. Masland (2012) The Neuronal Organization of the Retina. Neuron 76, 266-280.

k4k

cool!

>3 DugsBooks:, 4, 5 Thanks! A few month ago I read a quote, something like “If creationists sell you on the airplane thing and get you arguing that it can arise suddenly by pure chance, then they’ve already sold you on the straw man.” That quote came back to me when I read Shermer's passage...

Dear Carnophile, I suspect your example (#1) has also fallen into the "straw man".
You are talking about "Directed Evolution", a very useful tool, but still directed by an intelligence or purpose/design. I think of Dawkins pseudo insect shapes he used in one of his early books ( sorry can't remember the name).

These are all good examples of evolution but I suspect won't sway the "intelligence design" chap/chapess.

When that facial nerve (which in the Giraffe has a very long journey) travels around the heart to return to the face, is still seen as some sort of Divine purpose. I must confess I can't really be bothered talking to FUNDAMENTALISTS

ETA. A spelling mistake I couldn't stand after a week+

Guido.

“Debating creationists on the topic of evolution is rather like trying to play chess with a pigeon; it knocks the pieces over, craps on the board, and flies back to its flock to claim victory.”

― Scott D. Weitzenhoffer

Well the pigeon can't help it if it was designed that way.

>7 guido47: You are talking about "Directed Evolution", a very useful tool, but still directed by an intelligence or purpose/design.

To be sure, guido. But I think the 56 septillion vs. 83 makes the basic point.

Maybe only peripheral to the topic but here is a recent article with the headline: Quarter of Americans Convinced Sun Revolves Around Earth, Survey Finds

Interesting; the press release says "only 74% understand that the earth moves around the sun", so it's nearly "26%", apparently.

Which again correlates well with the American Crazification Factor, which is pretty well established to be about 27%.

When you see a building, you know that there was a builder. Why is Creation any different?

An explosion in a print shop can't produce a dictionary. Why is the Earth any different?

Time. Deep time.

Next question?

And buildings don't replicate themselves and they don't compete with each other.

Carnophile wrote: "The idea is to get a computer to randomly produce one of the most famous phrases in all of English literature, “To be or not to be.” The code starts with a random string, keeps the letters that match the target string, and tries again randomly for those that don’t. In other words, the process is not pure randomness; it’s randomness with selection."

I would think to make this work, one can not have choices being made at either the individual level or the environmental (selection) level. When one makes a choice to keep the letters that work, one has the environment take on the characteristics of an agent which I don't think is permitted in evolution theory.

It should be easy to show if something can be generated by chance or not. Set up the simulated environment, dump in the chemicals and wait until you get the desired result. No agent making a choice to keep something or throw something away should be permitted.

>16 FrankHubeny: Actually environment is an agent in the evolutionary process - in fact it is the agent. If an organism is unable to survive its environment long enough to reproduce it fails/is discarded. There's no driving force to this, as the environment itself doesn't care if an organism survives or not.

So, the environment for the experiment is one in which combinations of letters that don't form English words don't reproduce - seems reasonable enough to me.

In the case of generating this phrase, the agent is making choices and is therefore conscious. I don't think the environment is considered to be conscious enough to make a choice in evolutionary theory. That would mean the environment is a driving force and one has a God of sorts doing the selecting. I don't think evolutionary theory allows for choice-making agency.

The environment provides a set of constraints on the changes in the organism. Those that fail to live with the constraints die. When the environment changes its constraints the previously surviving organisms have to keep up with the new constraints or die as well. That is all there is to the environment's agency.

>18 FrankHubeny: In the case of generating this phrase, the agent is making choices and is therefore conscious. I don't think the environment is considered to be conscious enough to make a choice in evolutionary theory.

In this case the "choice" that is analogous to "make an English phrase" is "living long enough to reproduce." That's not a "conscious choice" -- it's just the mechanism of evolution.

The environment provides a set of constraints on the changes in the organism. Those that fail to live with the constraints die.

In fact, the environment doesn't care whether something lives or dies. And the only thing that an organism "cares" about is whether or not it survives long enough to reproduce the next generation. So learning to reproduce quickly in a highly toxic environment is just as valuable a strategy as learning to live longer in that environment, and therefore have more time to reproduce.

ETA: >16 FrankHubeny: It should be easy to show if something can be generated by chance or not. Set up the simulated environment, dump in the chemicals and wait until you get the desired result.

The key part of that phrase is "wait until...."

It's a long wait, but it is not infinite.

I don't think the process of constructing the phrase mimics what happens in evolution enough to be useful.

You're right about living long enough to reproduce. That is what matters in the context.

You are also right that the time is not infinite. The universe (or earth) is not eternal, so there is a limit on the amount of time one has to get something done by chance along.

When you see a building, you know that there was a builder.

You know there is a builder because you've seen buildings being built. You know the building is a constructed thing because you have compared it to natural things that are not constructed. What are you comparing "creation" to in order to determine that it was constructed?

An explosion in a print shop can't produce a dictionary.

1. Millions upon millions of explosions might result in one that does. The universe is an enormous place with literally hundreds of billions of stars and hundreds of billions of planets orbiting those stars. Even incredibly unlikely events are certain to happen with incredible regularity.
2. Cosmology is not an "explosion". The universe does not operate by random chance. There is no guidance, but that doesn't mean there is not regularity.

>21 StormRaven: The universe doesn't occur by random chance.

What I love about science it that the more you understand, the more you realize there is still to learn, and the more amazing it all becomes.

On a personal level I find that "God made it" somehow takes all the beauty and wonder out of it all. :)

"God made it" is also so unsatisfactory.

When you're wondering about the beauty of a flower, "God made it that way" is even less informative than "It grew that way", and neither is a real answer.

>23 Amtep:"It grew that way"

Ah, but "It grew that way" is not a scientific answer. It's an "I don't want to get into the science of it" answer.

But I agree it doesn't have anything to say about the ultimate meaning of a flower, if that's the question you are asking of the flower.

I'm asking why are flowers beautiful :) David Deutsch wrote a nice essay about that, where he argued that flowers display objective beauty, which is distinct from our species-specific "what looks good to humans" beauty, and that they had to develop it because it was the only way to communicate with insects, which are such a different branch of life that there was no common language of beauty to appeal to. (He also, in the process, argues that there is such a thing as objective beauty, and that we have barely begun to discover it.)

SR at #21 : Even incredibly unlikely events are certain to happen with incredible regularity.

I've quoted it before, but it seems apropos here:

"One of the standard problems with the universe is that it's large enough that unlikely things happen pretty often."

- Nigel Sharp, program officer for extragalactic astronomy and cosmology at the NSF

Amtep at #23: When you're wondering about the beauty of a flower, "God made it that way" is even less informative than "It grew that way", and neither is a real answer.

I've long been unreasonably fond of D'Arcy Thompson's pseudo-answer in On Growth and Form:

"Everything is the way it is because it got that way."

>25 Amtep: he argued that flowers display objective beauty, which is distinct from our species-specific "what looks good to humans" beauty, and that they had to develop it because it was the only way to communicate with insects, which are such a different branch of life that there was no common language of beauty to appeal to.

Well the possible existence of "objective beauty" is a topic unto itself, I agree. Does he explain why a flower's beauty is objective because it appeals to a bee as well as a person? What happens if it appeals to the bee but not the person? What if it doesn't appeal to a bee, but it looks like porn to a Luna moth? Does it have beauty just because it has a function (at least as far as the flower and the bee or moth is concerned)?

>27 AsYouKnow_Bob: D'Arcy Thompson's pseudo-answer in On Growth and Form

I love that book.

>16 FrankHubeny: When one makes a choice to keep the letters that work, one has the environment take on the characteristics of an agent which I don't think is permitted in evolution theory.

It doesn't matter whether there is some conscious agency conducting the selection. The key point is that there is some selection. In an arctic environment, white rabbits don't get picked off by foxes as often as black ones. So the next generation has more white rabbit genes (and more white rabbits) than the last generation.

You don't need a purposeful agency directing all this. You just need cause and effect.

It should be easy to show if something can be generated by chance or not.

Well, yes, actually, it is easy. It takes ~30 lines of code!

----------------------------------------------------------------

Edited to change "hawks" to "foxes."

>20 FrankHubeny: I don't think the process of constructing the phrase mimics what happens in evolution enough to be useful.

My intention here is not to propose a model of what goes on in the real biological world!

>29 Carnophile: It doesn't matter whether there is some conscious agency conducting the selection. The key point is that there is some selection.

If you have a conscious agent you are close to having a deity do the selection. However, those promoting a determinism plus chance approach to evolution, don't want an agent of any sort involved in either changes occurring to the individual or to the environment doing the selecting.

Now couple that with a modern knowledge of just how young the universe (or earth) is and these concerns of whether chance could have performed the change becomes important. Theists would claim this could not be done by chance and known deterministic laws.

For this code to counter those theistic arguments, it has to mimic more closely what evolution is supposedly doing. There has to be some interval of time between each random letter generation. The environment doing the selecting should not care what string is being generated. If it doesn't care, it will not be saving the letters that we expect to appear and rejecting the others.

For this code to counter those theistic arguments, it has to mimic more closely what evolution is supposedly doing. There has to be some interval of time between each random letter generation.

Why? Bacteria reproduce quite rapidly, and evolve quite rapidly as well.

>32 StormRaven: It depends on what one is trying to simulate with the letter picking selection.

What that code shows is that natural selection plus a super-human agent or deity keeping the desired letters after each random process would be able to complete the process within the time frame. But we already know that.

The code also shows that even sub-human agents could pull this off in the required time as long as they are making choices ("selecting") with an end in mind.

What it doesn't show is that random processes can complete anything within the time frame. In this scenario one cannot have any conscious agent "selecting" anything. Natural selection is just the environment providing a constraint on the changes a species randomly makes.

What that code shows is that natural selection plus a super-human agent or deity keeping the desired letters after each random process would be able to complete the process within the time frame.

No, just natural selection. That's why it is called "natural selection" - it is a process that keeps the most advantageous letters after each generation. What is and is not advantageous is determined by the environment, not a conscious selector.

For example – and maybe that is what StormRaven is thinking about in 32 – antibiotic resistance has shown up rather quickly on an evolutionary time scale (less than 100 years). That is random selection, unless one thinks that a sentient being is actually concerned about antibiotics.

>34 StormRaven: I normally assume what materialists mean by natural selection is the environment offering itself as a constraint. There is no conscious "selection" going on.

One could contrast that with an environment where agents exist. Then natural selection is non-human (non-artificial) selection done by various agents in the environment who offer constraints on other agents.

The agents involved could be bacteria.

The code example only works for an environment with agents since the selection process in the simulation involved the programmer's choices on what to keep and what not to keep.

I'll conjecture the following: If there were no agents of any sort, evolution would not happen. Since we are here and we know evolution has occurred, the environment is filled with agents of all sorts from microbes to humans who drive evolution.

The agents involved could be bacteria.

Bacteria are part of the environment. As are all of the other living organisms that are around you. But the environment also consists of the Sun, the Moon, the water, the weather, and everything else around you. You are trying to create a distinction where none exists.

You also seem to think that evolution has some sort of "end goal" in mind. It doesn't. Evolution doesn't "aim" for bigger, faster, stronger, or smarter creatures. Evolution merely results in the creatures best able to adapt to the current environment surviving.

>31 FrankHubeny: those promoting a determinism plus chance approach to evolution, don't want an agent of any sort involved in either changes occurring to the individual or to the environment doing the selecting.

Arctic, rabbits, foxes, black, white, cause, effect, etc.

>37 StormRaven: Bacteria are part of the environment.

Of course they are. Every species is both part of the environment doing the selecting and a species being selected by the environment. I don't think we disagree about the process of evolution. It has no purpose in itself. From the perspective of the species it looks like Eldredge and Gould's punctuated equilibrium. From the perspective of the environment it looks like a constraint that naturally rather than artificially selects the species who will successfully reproduce.

Regarding the code: Since the code contains an agent making choices, it only works if one believes the environment has agents making choices. I think that is the only kind of environment in which evolution works anyway.

Let's look at alternate environments that people actually believe in:

1) You could have a super-agent, some God, creating the universe in 6 days and generating an illusion that evolution occurred.

2) You could have an environment totally determined by quantum uncertainty and laws at the chemical or physical level. In this world evolution and agents making choices are illusions.

I think neither of those environments fit the reality that we see which contains agents of all sorts making choices in an environment which punctuates their equilibrium with survival challenges that they overcome. They make change happen over a reasonably long period of time (more than 6 days) and yet do this faster than chance would be able to complete the job (if chance ever could complete the job). Such a theory does not require illusions for the theory to work.

>39 FrankHubeny: Since the code contains an agent making choices, it only works if one believes the environment has agents making choices. I think that is the only kind of environment in which evolution works anyway.

I think you are muddying the pond by talking about agents. You seem to mean an entity that "wants," for lack of a better term, a change to happen, and sets about to make it happen.

But what you call "an agent making choices" is, if I understand the program Caronphile wrote, simple a criteria for selection. That's not a "choice." Just like evolving to breathe oxygen is not a "choice" of some agent. Being able to deal with oxygen just became a selective criteria at some point for planet Earth.

I think neither of those environments fit the reality that we see which contains agents of all sorts making choices in an environment which punctuates their equilibrium with survival challenges that they overcome.

You may think that, but there is no evidence of it. What you call "agents" are simply other species, other organisms, also adapting to their environment. Now you may resist this idea because it excludes free will, but it's likely that at a biological level, free will is just not applicable. And what you and I experience as free will -- such as my choice to make another pot of coffee -- is only relevant within the narrow constraints of my behavior and from a wider view this illusion of choice is simply one more genetic trait that may or may not turn out to be a successful adaptive strategy.

Your idea of the time required for chance to evince an adaptive change in a species is also somewhat off base. Mutation and adaptation is, on the genetic level, something that is happening constantly.

39> Since the code contains an agent making choices, it only works if one believes the environment has agents making choices

Wrong. The code may have been written by an agent making choices but the code itself is not making choices. Unless what you mean by choices is just a mechanistic if then sort of choice that a computer algorithm can make, but that's the same sort of 'choice' that the environment makes on an organism.

>33 FrankHubeny: It just occurred to me that there's an outside chance that you're merely trying to say that evolution is not goal-directed. If this is your point, then well and good; high-fives all around. However, I'm having to do a bit of stretching to read it that way.

As to the rest of it, I can only repeat: The point here is a specific, narrow one: to show that order does not have to appear in one fell swoop by mammoth luck. It can be gradual and cumulative. I'm not trying to put forth any model of biological evolution, let alone a complete one.

Here is more arithmetic:
The mean number of iterations to produce the TBONTB string without gradualism should be 56 septillion, or
56,000,000,000,000,000,000,000,000.

My computer does about 16,000 iterations per second, so it should take
56,000,000,000,000,000,000,000,000 / 16,000 seconds, or 110,984,271,943,176 years, on average.

That's 111 trillion years.

It actually takes between 0.02 and 0.03 seconds, typically. This is the power of retention of partial progress, accumulating over time.

By the way, I've whomped up a probabilistic version of this. In other words, instead of keeping the matching characters with certainty, it discards them with a small probability. (In other words, even white rabbits do get picked off sometimes; it's just a lot less likely than if they're black.) So it has to start over for the character in that position.

It still works; just takes longer to converge.

>41 jjwilson61: The agent is the programmer in the case of the code. The running program doesn't make any choice on its own. It is not conscious. The question is whether that code actually simulates what happens in the real world or not. To the extent that it shows that agents are required to get anything done, then I would say it does. However, I don't think that was the intent of the programmer.

>40 southernbooklady: I think you are muddying the pond by talking about agents.

In a way I am deliberately muddying the waters. I want to get agents and choice back into the picture. Organisms (agents) make choices in response to their environments. That is what moves evolution along.

In both of the extremes I mentioned, there are no typical agents. In the super-agent environment, where God creates everything in 6 days, or in the chance-deterministic environment, where nothing likely gets done, the normal agents (everything from microbes to humans) are irrelevant. They are explained away by God or determinism. In neither of these environments is evolution possible--at least as I see it.

What you call "agents" are simply other species, other organisms, also adapting to their environment.

Yes. Other species are the agents.

illusion of choice

That would be a theory and it would probably put your views in line with determinism. One of the major cracks in determinism came with quantum uncertainty.

It is possible that this "illusion of choice" is itself the illusion.

Organisms (agents) make choices in response to their environments. That is what moves evolution along.

No, it isn't.

>40 southernbooklady: But what you call "an agent making choices" is, if I understand the program Caronphile wrote, simple a criteria for selection.

>41 jjwilson61: The code may have been written by an agent making choices but the code itself is not making choices.

Yes. Really, it just comes down to some characters having a higher probability of staying around than others. (In the posted version the probability is 100%, but that's not actually required.) In this very loose metaphor, characters are like genes.

>44 FrankHubeny: In a way I am deliberately muddying the waters. I want to get agents and choice back into the picture. Organisms (agents) make choices in response to their environments. That is what moves evolution along.

As Storm Raven pointed out, the only thing that "moves evolution along" is reproduction.

You also seem fixed on the notion that what an individual does matters on an evolutionary scale. That evolution proceeds by creating an endless series of Adam and Eves.

But from an evolutionary perspective the selective pressure can be thought of as on the gene, not the individual, and it is a pressure that is applied across populations. Statistically, one individual is irrelevant. Only their genes are relevant. In that sense we are all Adams and Eves, because we are all slightly different combinations of genes that across the species may or may not turn out to be successful in adapting to the next pressure it meets in the environment.

Regardless, there is no evidence in any of this for your "ultimate agent" -- that computer programmer that supposedly started the universe running and gives it a kick every now and then to help it along. It's all ticking along just fine.

>47 southernbooklady: I am not promoting an "ultimate agent". Rather, the organisms that one finds in nature are the agents that move evolution along through their choices one of which is to reproduce.

The code simulation does have an ultimate agent, the programmer.

>48 FrankHubeny: the organisms that one finds in nature are the agents that move evolution along through their choices one of which is to reproduce.

Then your use of the word "choice" sounds anthropomorphic.

Rather, the organisms that one finds in nature are the agents that move evolution along through their choices one of which is to reproduce.

No, they aren't.

Rather, the organisms that one finds in nature are the agents that move evolution along through their choices one of which is to reproduce.

It should be "the organisms that one finds in nature are the agents that move evolution along through their ability to reproduce most efficiently".

>51 krazy4katz: I don't think one can replace choosing with having an ability, but having an ability assumes one will exercise that ability and that involves some choice.

>50 StormRaven: I disagree.

>49 southernbooklady: Our species is not the only one able to make a choice. Watch any pets you might have.

I don't see how one can rationally claim that the code example which required a programmer in any way simulates an environment where there is no programmer.

That being said, we exist. We got here in some way. It took some time (so we do not need to involve a deity), but not an infinite amount of time (so we can eliminate chance as well as the deity). Evolution looks like it describes the mechanism in which this occurred. We only need to acknowledge the existence of agents at all levels to make this work. We can see them right in front of us, so acknowledging their existence should be a no-brainer.

52: I disagree.

You're still wrong. No agents are required for evolution. Your agency is irrelevant to evolution. Every other organism's agency is irrelevant to evolution.

>52 FrankHubeny: That being said, we exist. We got here in some way.

Evolution does not speculate on the existence or nonexistence of the programmer. It is concerned with describing the mechanisms of life that is already here. For which no programmer is necessary.

Our species is not the only one able to make a choice. Watch any pets you might have.

But those "choices" are still not various gradations of free will. I have had many many pets. I never make the mistake of thinking of them as furry people. I let them be what they are.

>52 FrankHubeny:: choice is also programmed in by natural selection/evolution. Those who do not "choose" to reproduce will die out, leaving the others to survive. Hence (I suppose) the massive up regulation of interest in sex (at least in humans) at puberty when fertility will be highest. The brain is very susceptible to hormonal stimulation. For animals who mate only when the female is in heat, their "interest" also peaks at that time. This is not choice in the sense of choosing to watch a movie or read a book.

k4k

>52 FrankHubeny: I don't see how one can rationally claim that the code example which required a programmer in any way simulates an environment where there is no programmer.

For the Xth time, where X is a large number, my intention was not to simulate a biological environment. I was making on specific, narrow point about the gradual and cumulative, versus sudden, appearance of complexity.

But since you bring it up, actually there are scads of software models that do just exactly that. A lot of them are agent-based, which I grok means that they're set up from the get-go to track the properties of various agents in an environment in which they interact. There is an entire chapter (or maybe more) on this in Kevin Kelly's Out of Control, as well as Holland's Hidden Order, an entire book on nothing but that.

By the way, Out of Control is quite an interesting book in general. I found the Holland dreary and had to force myself to finish it.

You can create a bunch of agents in a software environment and have them reproduce - both sexual and asexual versions exist - subject to mutation. Guess what? Evolution occurs spontaneously.

Kelly mentions an amusing case in which an artificial agent not only did something unanticipated, but that the programmers didn't want it to do: It "lied" to the software about how much memory space it was entitled to. This let it acquire resources by deception, facilitating its reproduction. Really interesting stuff is going on in this area.

Lists of agent-based software:

http://www2.econ.iastate.edu/tesfatsi/acecode.htm

http://www.openabm.org/page/modeling-platforms

A summary:
http://www.scholarpedia.org/article/Agent_based_modeling

Whoa! It automatically adds the poster's name now! I thought y'all were doing it by hand!

>57 Carnophile: It automatically adds the poster's name now!

Intelligent Design, a la Tim.

Creationist!
:)

>52 FrankHubeny: We got here in some way. It took some time... but not an infinite amount of time (so we can eliminate chance...)

Sigh.

>54 southernbooklady: "But those "choices" are still not various gradations of free will. I have had many many pets. I never make the mistake of thinking of them as furry people. I let them be what they are."

As far as I'm concerned if something can make a change, I assume it had enough free will to do so unless there is evidence that it is a machine. I would agree that a machine does not have adequate free will to make a choice, but a pet would.

>56 Carnophile: "For the Xth time, where X is a large number, my intention was not to simulate a biological environment. I was making on specific, narrow point about the gradual and cumulative, versus sudden, appearance of complexity."

It seemed to me that you were trying to argue that complexity could arise by chance. If you were not, then we are in agreement.

Your code shows that if you have agents you can have change occurring that introduces complexity. It does not show that this could ever have occurred by chance.

I looked at the links. It seems they are considering agency the way I am, in particular the one about agent based market simulation models.

>55 krazy4katz: "choice is also programmed in by natural selection/evolution. Those who do not "choose" to reproduce will die out, leaving the others to survive. Hence (I suppose) the massive up regulation of interest in sex (at least in humans) at puberty when fertility will be highest. The brain is very susceptible to hormonal stimulation. For animals who mate only when the female is in heat, their "interest" also peaks at that time. This is not choice in the sense of choosing to watch a movie or read a book."

There are a lot of deceptive metaphors used to describe what happens with evolution. It doesn't make sense to say that "choice is also programmed in". Similarly it doesn't make sense to talk about a "selfish" gene, as Dawkins does, unless one allows that gene to be conscious enough to be selfish.

One of the things that convinces me of the existence of free will (to some minimal extent) is the research in sexual desire. I'm specifically thinking of the research summarized in Young and Alexander's The Chemistry Between Us. What they describe are reward and punishment systems that overall motivate the way the individuals in a specific species will either bond or not. These are not deterministic systems, but reward systems and hence the support for minimal free will.

I actually think the mating process you describe is a choice. I agree it is different than our choice to watch a movie, but it is still a choice. The fact that an individual in the species chooses to do what is pleasurable or what will avoid pain does not mean that individual did not make a choice. A choice was made.

>63 FrankHubeny: There are a lot of deceptive metaphors used to describe what happens with evolution. It doesn't make sense to say that "choice is also programmed in". Similarly it doesn't make sense to talk about a "selfish" gene, as Dawkins does, unless one allows that gene to be conscious enough to be selfish.

Biology and evolution is littered with "deceptive" language because it is hard to talk about choice without also implying a kind of intent. But as a rule science is careful to clarify that words like "choice," "adapt," etc are used to describe processes, not "will" or conscious intention.

The selfishness of Dawkins's gene is a good case in point. In the 30th anniversary edition of his book he discusses how he was advised at one point to use a different, more "up" (ei, "positive") word for the title -- Such as "The Immortal Gene" -- which, given the theory's premise, would have been an accurate title.

But "selfish" was chosen because Dawkins was presenting a theory of the evolutionary advantages of altruism -- a characteristic that is often cited as de facto proof of free will. In placing the selective pressure on the gene or set of genes, instead of the individual organism, Dawkins was able to show how altruism could be a successful evolutionary strategy at the level of the gene, even if it meant an individual organism (what Dawkins calls "the vehicle of reproduction") did not reproduce. But in no sense is a gene "selfish" as in "lacking consideration for others; concerned chiefly with one's own personal profit or pleasure."

Personification is a good teaching device, but science uses it at its peril and indulges in it with caution. A bird does not choose to avoid eating the red berries because they are poisonous. It doesn't eat the red berries because it has been genetically programmed not to. A monkey, who might have no specific programming regarding red berries, is however programmed to watch and learn from others what is or isn't safe to eat. If his brother eats the red berries and doesn't die, then he will as well. Is that monkey's choice to eat the red berries an example of agency? of monkey free will? A kind of carpe diem attitude towards red berries? No. He's just hungry, he wants to eat, and he is learned that red berries are safe to eat.

>64 southernbooklady: I have heard similar accounts where Dawkins' word "selfish" would have been replaced by something more positive. I don't think it would matter whether the intent was selfish or altruistic. Both imply an agency that I am aware Dawkins wanted to avoid.

I agree with you that the intent is to exclude will in these metaphors, but it underlies the problem that evolution just might not work without intent. Using the word "process" doesn't get rid of the intention that allows the process to happen.

I assume that putting the metaphorical intent at the level of the gene rather than the cell was also a way to eliminate the primacy of the organism in evolutionary theory. However, that just makes the theory appear at its core irrational. The gene cannot exercise any selfishness without the cell and yet it is just using the cell as if it were a machine that it intentionally programs.

This brings up the question of whether evolution could actually work for machines whose genetic changes come about by chance. I don't think it can, but I don't have a good proof against that except that there was not enough time (4.5 billion years for earth) for chance to complete the process.

>65 FrankHubeny: I don't think it would matter whether the intent was selfish or altruistic.

The "intent" is nonexistent.

but it underlies the problem that evolution just might not work without intent.

So you keep saying, but observational evidence indicates otherwise.

Using the word "process" doesn't get rid of the intention that allows the process to happen.

This sentence makes no sense. A process does not imply an intention.

I assume that putting the metaphorical intent at the level of the gene rather than the cell was also a way to eliminate the primacy of the organism in evolutionary theory.

Nope. But perhaps you should familiarize yourself with the theory. Genes are "the thing-that-replicates." An organism is "the-vehicle-of-replication." It certainly changed the way we view the processes of species development.

that except that there was not enough time (4.5 billion years for earth) for chance to complete the process.

And yet, the fact that we are here is observational evidence that there was and is plenty of time for the processes to occur.

But then I think part of the problem I have understanding you is in your conception of "chance" vs. "agency." You seem to regard the latter as any action anything takes in response to its environment. Things react to their environment constantly, but this does not imply the kind of directed "intentional" action implied by the word agency. It's really just all the biological equivalent of the chemical reactions between reagents and reactants.

I don't think it can, but I don't have a good proof against that except that there was not enough time (4.5 billion years for earth) for chance to complete the process.

And yet again you make an argument from personal incredulity. And once again, you are simply wrong, because you haven't bothered to educate yourself on the topic that you are cluelessly expounding upon.

>66 southernbooklady: "And yet, the fact that we are here is observational evidence that there was and is plenty of time for the processes to occur."

We agree on that piece of evidence. We are here. It took 4.5 billion years for life to evolve on earth.

The question is how did that happen? Did some God do it in 6 days? Probably not, since it took longer than that. Did chance do it on machine-like objects? Probably not, since chance gets nothing done. It would be a miracle for chance to reverse the flow of entropy. However, it wouldn't be a miracle for an organism to do such a thing.

One reason many worlds and other multi-verse ideas are presented is to resolve this problem with chance not being able to do anything. Otherwise, relying on chance in a 4.5 billion year time frame falsifies the machine model of evolution.

I want to make it clear, that I don't believe in many worlds, but I do think there are other universes out there. If ours started 13.7 billion years ago, many others started as well. However, I don't think these other universes can save chance.

As I think about it, I wonder if anything occurs by "chance", which would be an event that occurred "randomly" under a uniform distribution of possibilities.

>67 StormRaven: I just want to say, StormRaven, that I am not ignoring you. I do read your posts, I just have no comment to make.

It would be a miracle for chance to reverse the flow of entropy.

Where do you get the idea that evolution reverses the role of entropy?

69: Your main problem is that you don't actually have much of anything to say to anyone. You're floundering in ignorance and you don't even know it.

>70 StormRaven: An example of reversing entropy would be for life to start from non-life or for evolution to go from single celled organisms to multi-celled organisms.

This occurs through evolution. This process increased complexity or locally reduced entropy. Did chance do this? I don't think so. Was this the result of some agency, perhaps Maxwell's demon? That could happen. The theory just needs to include agency or intention or will to allow us to get to the state we are in today.

As I think about it, to view evolution as evolution of machines is to replace evidence with a theory. It is to "over-think" the problem. The evidence shows that we are organisms not machines. Organisms can easily evolve, but if one includes agency at all in evolution one opens up the door to any agent, including deities. If one doesn't include agency, then evidence has to be thrown out simply because it doesn't work with someone's fantasy land.

An example of reversing entropy would be for life to start from non-life or for evolution to go from single celled organisms to multi-celled organisms.

Except that isn't an example of reversed entropy at all. The Earth is continually bombarded by energy from the Sun. You seriously need to brush up on your basic science before you try to opine upon what is or is not possible through an unguided process, because you're making basic, elementary school level mistakes in your evaluations.

Did chance do this? I don't think so.

And you are engaging in logical fallacies. This one is called the argument from personal incredulity. It makes you look foolish every time you use it.

Chance is only one component of evolution, the other is natural selection. That second is what allows complexity to happen.

>72 FrankHubeny: I agree with StormRaven in 73. Your hypothesis involves a misconception of entropy. The web is full of explanations regarding how the Second Law of Thermodynamics is misused to dismiss evolution, so I won't repeat the arguments here. Fortunately we don't have to worry about the Sun running out of energy in the near future. The Earth is not a closed system.

>75 krazy4katz:

The question is whether we are deterministic machines or organisms that can make choices. Can evolution operate on machines? I don't think it can. My position is that evolution requires agents. With agents evolution makes sense.

One of the ways to test if evolution could operate on machines is to ask if chance operating on machines can increase complexity in a system that is constrained by an environment that is also a set of machines. We know how much time was spent to get the job done. Was there enough time (4.5 billion years) for chance to increase complexity to what we see today?

Not only do I not believe that chance could complete this in 4.5 billion years, I don't think much occurs by chance anyway. What looks like chance is what appears as uncertainty, perhaps the result of choices at various levels. However, "chance" is the materialists' god of the gaps. If the materialist can't come up with a deterministic law, chance is brought in to fill the gap.

Again, your focusing on the chance and ignoring the actual mechanism. Do you really understand what the modern theory of evolution is?

Not only do I not believe that chance could complete this in 4.5 billion years, I don't think much occurs by chance anyway.

These are assertions, not evidence. Do you have any actual evidence that this is the case, or are you ignorantly talking out of your ass again like you were when you brought up "reverse entropy"?

>76 FrankHubeny:

OK, I don't know if this helps but "chance" is admittedly an imprecise word. Random mutations in genes of living organisms would be more accurate. Some of these are lethal, as we know from observing the tragedy of congenital diseases. Populations with those mutations die out. Some of them are locally beneficial. For example, the lightening of skin that occurred in some populations of humans that moved north out of Africa. The lighter skin presumably allowed them them to make more Vitamin D (yes, I know this is a theory). Now, since many of those lighter skinned people have migrated south, it makes them more susceptible to melanoma than dark-skinned people. So environment is a selective pressure on survival and darker skin is beneficial in sunnier places.

With modern technology, we have managed to partly alter the influence of the environment on ourselves and other species - both for better and for worse. Evolution is not over. A million years from now, perhaps there will be dramatic differences depending on the environment, our ability to control genetics etc. One example that I mentioned before is the invention of antibiotics. We have cured many previously lethal bacterial diseases. However we now find -- not by our own intention! -- that antibiotic resistance among bacteria is increasing, so these diseases may re-emerge. This is selective pressure (i.e. evolution) in the bacterial world that is happening right in front of us very quickly, considering that antibiotics were developed in the 20th century.

Does this make sense or are we talking past each other?

If you are talking about how life began in the first place, from the "primordial soup," that is a more complex question that has been the study of many scientists. There are various theories and experiments where people have put basic chemicals together under "ancient earth" conditions that have caused the formation of amino acids etc. I don't remember many of the details of those experiments.

Best wishes,

k4k

>79 krazy4katz: OK, I don't know if this helps but chance is admittedly an imprecise word.

In the context of this conversation, so is the word "machine."

>80 southernbooklady:

Yes, I have been confused about that too. In what context is the word machine being used?

I've been using it to refer to the biological organism itself....the "thing" that is coded by DNA.

Ah, I see. Thanks. k4k

>62 FrankHubeny: It seemed to me that you were trying to argue that complexity could arise by chance. If you were not, then we are in agreement.

I have argued that complexity could arise by chance in the past and no doubt I will argue it again in the future. But for present purposes, you can think of my point in posts 1 & 2 as being this:

IF someone were to argue that complexity could arise by chance, they would not have to argue that it can arise in a single step. That is not necessary. They only need argue that it can arise in a series of many smaller steps.

>62 FrankHubeny: I looked at the links. It seems they are considering agency the way I am, in particular the one about agent based market simulation models.

The first one is mostly econ, you're right. Try the second one.

There are lots of fascinating passages about simulated evolution in Kelly's Out of Control. I'll quote some if you like. (And possibly even if you don't, heh.)

>72 FrankHubeny: Organisms can easily evolve...

I thought I understood you to be denying this. Now I am uncertain as to what are you saying.

>79 krazy4katz: "However we now find -- not by our own intention! -- that antibiotic resistance among bacteria is increasing, so these diseases may re-emerge. This is selective pressure (i.e. evolution) in the bacterial world that is happening right in front of us very quickly, considering that antibiotics were developed in the 20th century."

I acknowledge this is happening and this is part of evolution. Those bacteria are agents, in their own way to the extent they are able, making choices to survive. We are the environment they must respond to. And we are agents making our choices to survive.

>86 Carnophile: "I thought I understood you to be denying this. Now I am uncertain as to what are you saying."

So you thought I did not believe that evolution occurred at all? Of course I do. It is pretty obvious that the process works and it could be generalized to other changes that we experience from responding to personal crises in our own lives to changes in cultural acceptance of new scientific paradigms. It does not only apply to changing species.

What I don't think is true is that evolution happens to machines. There must be agents involved making the right choices to survive. This is different from a theory that says nothing makes a choice to survive and change occurs through random events.

>84 Carnophile: "IF someone were to argue that complexity could arise by chance, they would not have to argue that it can arise in a single step. That is not necessary. They only need argue that it can arise in a series of many smaller steps."

We are also in agreement that changes occurs in small steps over a rather lengthy period of time. We disagree on the chance part, but it is worthwhile to understand the points where we agree.

I think the Eldredge-Gould theory of punctuated equilibrium is the core idea behind evolutionary change. They seem to have the mechanism down in terms of how the different sandstone strata are formed. There is a long period of time with equilibrium when the fossil record remains the same and then a briefer period when it gets punctuated and the surviving and changed set of species become part of the new stratum and fossil record.

>80 southernbooklady: "In the context of this conversation, so is the word "machine.""

I agree. I don't have a good definition of what separates and organism from a machine. As a short-cut, I usually refer to the "choice" an organism can make, but then one wants to know if all organisms can make choices. I would say: yes, they can. But that requires more evidence.

As a short-cut, I usually refer to the "choice" an organism can make, but then one wants to know if all organisms can make choices.

What choices do ferns make?

>87 FrankHubeny:

Bacteria don't make a choice (like ferns as SR says above) to survive antibiotics. It is random mutagenesis. Some survive and some don't. This has been tested in the laboratory.

We are clearly talking past each other. The same words have different meanings to you than the rest of us.

k4k

>92 krazy4katz: "Bacteria don't make a choice (like ferns as SR says above) to survive antibiotics. It is random mutagenesis."

This is a point where we differ. Plants respond to stimuli. Specifically, how they respond is their choice. It doesn't look like the choices we make, and it isn't, but some people would say we don't make any choices either with everything being determined.

The random genetic mutation position would be in the category of theories that try to make sure no choices are made. What appears as obvious choices (at least when we do them) is replaced by the materialist's god of the gaps chance. Even the materialists don't like chance, but quantum theory has frustrated total determinism.

The question centers around what drives evolution: chance or choice.

This is a point where we differ.

That's because we are dealing with reality, and you're engaging in a process called "pulling things out of your ass".

Plants respond to stimuli. Specifically, how they respond is their choice.

I think you need to go back and study definitions, because that's not choice at all. Did you choose the makeup of your DNA? Because that's the primary determiner as to whether a creature will survive or not to pass their genes on to succeeding generations, and their "choices" only make a very slight difference.

Suppose there was a field of wild wheat that was attacked by wheat rust. Some of the wheat will die. A small portion of the wheat will live. Where is the choice here? Does some of the wheat "choose" to die? Do other stalks "choose" to live? Wheat can't feel pain, and doesn't have any ability to choose anything. Wheat rust is a fungus with just as little ability to "choose" to live, die, or anything else. Where exactly is the choice being made here?

What part of the wheat is doing the "choosing"? What part of the wheat rust is doing the "choosing"? You keep asserting "choice", but you haven't said what choice is other than to say it isn't "chance". There are a lot of things that aren't "chance" that are also not "choice", so that's not any kind of definition at all. The Earth's orbit around the Sun isn't chance, but there's no choice there either.

>93 FrankHubeny:

"Random genetic mutation" is not a theory. It is an observation made in millions of laboratories for many, many decades. We can read the sequence of DNA. We can see spontaneous mutations in bacteria, yeast and in cells created from mammalian tissues that can live outside the body. We can increase mutation rates in the laboratory with chemicals. We can watch organisms make random repairs to their DNA in the laboratory. We can alter their DNA by providing different environments in which some will mutate and survive whereas others will mutate and die and some will not mutate and also die. These are scientific facts. With the ability to examine DNA from specimens of organisms that lived millions of years ago, we now have the ability to trace these random mutations, see how they branch etc. What dies out, what is carried forward to modern species. These are not choices in the sense that people normally use that word.

I still think we are talking past each other but I don't understand why.

>96 Jesse_wiedinmyer:

That was a neat concept.

Yes, very interesting. Thanks!

Re.#94: Gee, I think I'd like to get a PhD in "Pulling Things Out Of Your Ass" ... I know that LOT of universities specialize in this, but are any granting advanced degrees in the field?

I know that LOT of universities specialize in this

Is that a fact that you just pulled out of your ass?

Yep!

Well then, BTRIPP, consider that you have just been awarded an honorary degree.

Academically yours,

k4k, PhD