Intelligent Evolution?

[cyborg]

Conception.

You're thinking of the trigger in the wrong way: the trigger to reproduce something is given when you've missed a sufficient number of bullets that could have caused your failure.

The thing is that in the analogy by the time the trigger is activated reproduction is an inevitable consequence but in your analogy it is not.

 

[articulett]

The information contains "something" which encourages that information to get replicated-- it doesn't self replicate, dammit, Jimbob You don't self replicate... viruses don't self replicate. The information contains something that encourages it's own replication.

When instructions (information) added to a shampoo bottle contained the word "repeat"-- as in "wash rinse repeat"-- more shampoo was used... and more shampoo bottles were made and more directions were placed on the new bottles-- so the information coded in the word "repeat" was responsible for that information being put on more shampoo bottles. Sure, humans or machines or whatever wrote it-- but the information ensured it's own replication... in a very similar way that the butterfly mutation that conferred parasite resistance ensured it's own replication.

The mutation did NOT replicate itself. It needed to get into a butterfly that produced gametes that successfully formed zygotes that successfully grew up to reproduce. Jimbob, since you do not understand what "self replication" means in terms of evolution or the analogy-- just stick to the word replication.

Why is some information copied more than others in given environments? Could it be, that the environment selects based on something about that information? Why is the word "repeat" on so many shampoo bottles? Why do all the male butterflies in the article now have the beneficial mutation? Why would humans preferentially copy the designs of airplanes shown to fly in their environment rather than ones that didn't? Why would they copy more efficient designs as they appeared and not less efficient ones?

(Damn... I'm caught in the jimbob loop again... it's like being Charlie Brown with Lucy threatening she won't pull the ball away this time...)

 

[jimbob]

I have been arguing the wrong thing with you jimbob - that which is "natural".

You are arguing that a natural consequence of self-imperfect-reproduction and selection is evolution.

I am arguing a natural consequence of imperfect-reproduction and selection is evolution.

You are arguing that a natural consequence of self-imperfect-reproduction and selection is evolution.

Nearly: I am also arguing that natural selection follows from any self-replication (perfect or otherwise), as any resulting population would be limited by malthusian factors.

I look on it as two sides of a coin.

Variation comes from imperfectly copying "the blueprint" (I think we all agree on this).

I am arguing that selection doesn't act on the blueprint, but on the product of the "copying". (I am not sure if we agreee on this)

For brevity: in the rest of this post I will use "copying" to mean imperfect copying, unless otherwise stated, as I think we all agree on the need for this.

If selection acts on the product of the blueprint, and if the copying is by an external agent, then the results from the selection need to feed back to the copying system to be acted upon.

I am arguing that in this case, intent is needed to set this system up. Once running, the system needn't require any further intelligent input, but I would argue that the selection criteria are somewhat arbitary:- one could aim to select for "torches" or "radios" or "what sells", or "what doesn't sell", and the evolutionary algorithm would work towards that arbitary goal.

Finite resources will add some of the malthusian factors of natural selection, but "culling" the "product" (by destroying it) needn't affect the selection of the blueprint associated with that product. Indeed, one could select for a "short-lifetime" product.

If there is self-replication (with variation), then once we ge beyond abiogenesis* and have a self-replicating system; anything that culls the "product" of the copying also culls the particular "blueprint" associated with that product. Malthusian factors alone will ensure that there is selection.

Maybe it is interesting to explore what would happen if the evolutionary algorithm was selecting towards self-replication...

I am a megalomanic who has set up a lab to create artificial life using evolutionary algorithms.

Initially, the variants that were closest to self-replicating would get selected. After a while, one variant does actually self-replicate. It is now competing against its "siblings" that are almost self-replicating.

Copies of the self-replicating variant are now being made by my lab, and by the descendants of the replicator(s).

The descendants of the replicators are independent of my arbitary selection criteria. I now decide to alter the selection criteria for my copier to produce systems that write "Jimbob" (I am a megalomaniac after all).

Sure enough I find that the variants made by my copier get closer and closer to writing "Jimbob", and some still self-replicate and others don't, (I no longer care about self-replication). The descendents of the original self-replicators, however have no such trend towards writing my name.

My assistant is a bit miffed, and being cleverer than me, decides to alter the environment so that there is a selection pressure for replicators that write "Igor". This will affect the descendants of any products that replicate in the environment, including any of mine that still self-replicate.


​

*(Von Neumann cellular automata are surprisingly simple, and given enough appropriate random chemical reactions, it is not surprising that a self-replicating collection of molecules arose, with or without panspermia).

 

[jimbob]

You're thinking of the trigger in the wrong way: the trigger to reproduce something is given when you've missed a sufficient number of bullets that could have caused your failure.

The thing is that in the analogy by the time the trigger is activated reproduction is an inevitable consequence but in your analogy it is not.

Yes, I prefer to think that in self-replicating systems, "self replication" is what is "supposed" to happen except for the darn universe getting in the way. In a system without self-replication the copying needs to started as opposed to stopped.

If there is not self-replication, then I am arguing that you do indeed need a "start signal" (there are always external stop signals due to finite resources).

If there is self-replication I am arguing that the "start signal" is at the system's inception: timescales, lifetimes and methods etc. are just traits that evolve to optimise reproductive success. There are plenty of stop signals.

 

[articulett]

In your robot analogy--what is it that you are changing to have the robots make more robots? Their "program"? That's information. What is selecting the next generation to mate? The environment? That's natural selection. It works the same way in biology. Information recombines and changes at random... the environment selects the products of that information for participation in the next "round" (or not).

It's the same.

 

[Belz...]

Photocopiers evolve.

Actually, I'd say that the information for building photocopiers evolved-- giving the appearance of photocopiers evolving...

Just as genomes evolve over time giving the appearance of species evolving-- but every animal is born and dies the same species it was born... it is only the recombination and honing of that information in the environment over time that gives rise to the impression of animals morphing into new species-- just as photocopiers morph into more efficient and "evolved" copying machines.

And it's, of course the same with the information to build airplanes or chairs. The information in genomes and in designs don't "self replicate"-- they get themselves replicated... DNA needs to get in a gamete and form a zygote to become part of an evolving system. A virus needs a cell to replicate. The information for building planes needs to successfully and predictably produce planes that fly in order for humans to copy (replicate) the design and the design so that the design can evolve. Airplanes don't technically evolve-- the information that makes them evolves-- based on how the product coded for performs it it's environment. It appears via snapshots in time that airplanes have evolved-- it's the information for making airplanes that has actually done the evolving.

I'm just clarifying because Jim keeps confusing the evolution of information with the the thing the information codes for. He confuses the genome for the butterfly it codes for... the blueprint for the airplane itself... Parts of the former (the "directions") can live on indefinitely and evolve-- based on how the latter (the product they code for) performs in the environment of replicators. When life forms do as they are programmed to do via their genes-- they can't help but be replicators as well as "environmental selection inputs" honing the other evolving systems in their environment.

Jim may or may not know why he types what he types... the reasons he gives may or may not be correct. But he is assimilating information in the form of words and replicating it by posting it. Others read that information and use it, replicate it, tweak it, ignore it, etc. I see myself as using his information to understand what he's missing in his understanding of natural selection... I want people to be able to understand the analogy and share it with others... I like understanding the best way to convey information about evolution... So his "information" can't help but effect my evolving explanation as well as my understanding of his blind spot. His information also has a chance to evolve in itself--if it was useful, furthered understanding, or had some sort something that inspired other readers to want to copy it or borrow pieces of it. From my perspective, his explanation is just the right kind of pedantic muddledness to be of use to creationists trying to obfuscate understanding of evolution. I can't imagine any other reason to pass it on (replicate it). It doesn't clarify anything; it isn't useful; it' not crisp or memorable. It doesn't further any goals or understanding. It's one of the new combinations and/or mutations of information that isn't likely to be selected in this environment. But humans generate information constantly and some such information is selected and replicated and honed by other humans. I can try to come up with ways to get "my information" replicated or with reasons why I think my words should be replicated-- or posit what I imagine I'm accomplishing by posting her and why I do so... but as far as evolving information systems are concerned, it's irrelevant. The stories that humans tell themselves about intent and reasons they replicate, recombine, and tweak information is irrelevant. The fact that some information is preferentially copied over others is all that matters to the analogy... it's the very essence of how natural selection gives the appearance of design...

Why was this information chosen over all others... why was this built upon... why are most chairs about 2 feet off the ground? Did someone plan it? Does it have meaning? Why was this general direction about chair height chosen preferentially to be the dominant model over all potential chair heights? The same tools we use to understand how complicated creatures such as ourselves came to be-- are the same tools we can use to understand How the other objects in our world came to be. After all, the atoms that make them up have existed for eons. But why no chairs until now? Follow the information and the replication of that information backwards. All complexity comes about from the bottom up. When we use human inventions as examples of "intelligent design"-- we are ignoring the information that needed to be assimilated over time for that product to emerge... there IS no top down design--no design in a vacuum. We have no actual model for "intelligent design" on our planet, since even our most intelligent design didn't poof into existence. Anything attributed to intelligent design that did not go through the iterative-information-selected-by-the-environment-over-time process would HAVE to be supernatural. We have no natural explanations for "assembled" matter springing into existence without an evolving information source. Even non complex things like rocks are matter formed by input from their environment over time.

It's probably too esoteric for most. But I feel like the explanation in the analogy is a good tool for understanding all sorts of complex evolving systems... and it also makes an "intelligent designer" so unlikely and unnecessary. Just as technology continues to increase in efficiency and not go backwards (the behemoth computers of yesteryear are firmly in the past)--understanding of our world goes forward to.

... It's humbling and exciting to be a part of that process.

Now THAT is what I call expanding upon an idea!

 

[cyborg]

I am arguing that selection doesn't act on the blueprint, but on the product of the "copying". (I am not sure if we agreee on this)

We do - the phenotypical expression of the genotype is the thing itself. Sometimes these are one and the same but they may not be depending on the context.

I am arguing that in this case, intent is needed to set this system up. Once running, the system needn't require any further intelligent input,

No jimbob - it isn't. This is why you are arguing the ID case - THEY will see the exact sort of requirement for "feedback" into the system and say their IDer put it there.

The point is that IF feedback occurs by some natural mechanism by chance AND the feedback mechanism is positive THEN mechanisms that provide positive feedback are naturally occurrent - i.e. without the need for "intent".

It doesn't matter to the system whether or not the "instigator" of the process was some object that considered itself intelligent or not - what matters is that the system has come to be what it is.

We might identify particular systems as being instigated by one or the other but when talking about entire classes one has to be far more careful about making blanket statements.

but I would argue that the selection criteria are somewhat arbitary:- one could aim to select for "torches" or "radios" or "what sells", or "what doesn't sell", and the evolutionary algorithm would work towards that arbitary goal.

It has changed what is "natural" to the system.

Finite resources will add some of the malthusian factors of natural selection, but "culling" the "product" (by destroying it) needn't affect the selection of the blueprint associated with that product. Indeed, one could select for a "short-lifetime" product.

Yes you could - the question is whether or not the "short-lifetime" product producers are more successful or not. That will dictate whether or not the system "chooses" that strategy.

Maybe it is interesting to explore what would happen if the evolutionary algorithm was selecting towards self-replication...

Well this is the point I was trying to make to you earlier - the "goal" of an evolutionary system is simply whatever is "natural" to it. Therefore if it is "natural" for the system to produce self-replicators it will (and in the absence of any other factors this could be called the "most" natural system). If we "set" what is "natural" to the system we haven't changed how the system will operate, just how it will progress.

We say that it is "artificial" but that is only a useful way for us to label actions we have and have not taken by "intent".

Now this is why I have a problem when you say one is "different" to the other - they are not different in type, but in configuration.

What you have to ask yourself is what is really different in your scenario if you got a bunch of toddlers in instead to mess around with the system in an "unintelligent" way?

 

[cyborg]

Yes, I prefer to think that in self-replicating systems, "self replication" is what is "supposed" to happen except for the darn universe getting in the way. In a system without self-replication the copying needs to started as opposed to stopped.

Jimbob - the thing you do not seem to appreciate with this argument is that before a "self-system" existed on Earth there must have been a "non-self" system. Non-self systems require "starting".

If starting requires "intent" then now you're making the ID argument.
If starting doesn't require "intent" then there's really no difference between the "starting/stopping" dichotomy - either the system continues to self-sustain in the same manner as it was before or it dies out. How it came to be in a configuration that could self-sustain doesn't really tell us anything about the self-sustaining system itself.

 

[jimbob]

Cyborg, this is my argument why intent is not needed...

*(Von Neumann cellular automata are surprisingly simple, and given enough appropriate random chemical reactions, it is not surprising that a self-replicating collection of molecules arose, with or without panspermia).

 

[Southwind17]

Jimbob: putting aside any fundamental differences between self-replication and external-replication, in the natural environment, what do you see as the trigger for replication? In other words, from the moment a cheetah, for example, is born, what then determines whether it replicates or not? Let's take this step by step to identify the point in the evolutionary process where you're missing the point. Please try to be concise with your answer. Thanks

Conception.

I will expand.

I think of it as a bit like the duck-shooting game at the fair. At conception the cheetah is set of on the path to reproduce (reach the other end of the booth) but on the way the environment is shooting at the same cheetah. The closer to the end, the better the chance of reaching the end, and what "natural selection" is doing, is stopping the reproduction.

For the analogy to convey the chances of a cheetah zygote reproducing, the universe should probably "use" a machine gun (whether "evolved" or not).

You're thinking of the trigger in the wrong way: the trigger to reproduce something is given when you've missed a sufficient number of bullets that could have caused your failure.

The thing is that in the analogy by the time the trigger is activated reproduction is an inevitable consequence but in your analogy it is not.

Yes, I prefer to think that in self-replicating systems, "self replication" is what is "supposed" to happen except for the darn universe getting in the way. In a system without self-replication the copying needs to started as opposed to stopped.

If there is not self-replication, then I am arguing that you do indeed need a "start signal" (there are always external stop signals due to finite resources).

If there is self-replication I am arguing that the "start signal" is at the system's inception: timescales, lifetimes and methods etc. are just traits that evolve to optimise reproductive success. There are plenty of stop signals.

Jimbob, let's backtrack a little please. In my post above I asked what you see as the trigger for replication after the cheetah is born. I asked it like this because I'm tying to get you to realize that replication is dependent upon survival (obviously - a cheetah cannot breed if it's dead!), and survival is dependent upon how well equipped the cheetah is in the context of its environment, as cyborg clearly acknowledges above. I am puzzled, therefore, when you respond: 'conception', because at that stage it is too early to measure the effects of any mutations that the embryonic cheetah might undergo. It's critical to the analogy that we reconcile this fundamental difference in understanding.

Your duck-shooting analogy, to my mind, contradicts your 'conception' response, as, by your own admission in using the duck-shoot analogy, the replication trigger isn't pulled unless and until the duck makes it to the other end of the booth, i.e. the trigger is pulled at the end of the booth and not at the beginning, the fundamental difference being that only by making it to the end of the booth can we say that the duck was evidently suitably equipped to survive.

Could you please reconsider this and clarify your position jimbob. Thanks.

 

[articulett]

Yes Jim... the analogy isn't about what got the snowball started... the analogy is about how it came to be so big. A force such as a hill would be an environmental input shaping the final product. A snowball is an organized form of matter that takes time and environmental input. All organized matter requires at least that much... and it can become increasingly organized and "efficient" once there is a coding system in place that can be replicated in some manner producing products that can select in some manner.

As for selection acting on the "product" and not the blueprint... how exactly is that different than in nature? When the environment preferentially selects the butterflies with the parasite resistant mutation-- they are acting on the "product" (the butterfly) of the information (genome) that builds the butterfly with the mutation. It is only acting on the information indirectly. The same as with the blueprint.

Once the butterfly is selected because it has a mutation that confers a survival advantage, then the mutation has a chance to get itself copied by being part of a gamete that becomes a zygote that codes for a butterfly that successfully reproduces. It doesn't copy itself--it gets itself copied! Recombinations through sex and random mutations make for a great variety of product that the environment can then select from.

When the environment achieves a product that can fly through recombination and mutation of information (not unlike the process that achieved the butterfly mutation) then the environment acts upon that product. The environment of humans says, "hey us humans have evolved a design that codes for a machine that can reliably fly"-- and they they replicate the design and begin to evolve and hone that design. Info. that makes an airplane fly is info. that gets replicated in various manners by those who want to fly.

The basic directions for a "machine that flies" is on par with the mutation that codes for "butterfly that survives parasite infestation". It's info. that has a means of getting itself copied into the future to become a part of evolving information systems-- things that are increasingly "complex", "efficient", "amazing", and "seemingly part of some big plan" or "design".

Info. that makes butterflies survive to reproduce gets replicated in various manners by butterflies who "want" to have sex and do the things that butterflies are programmed to do.

In both cases-- the information had a "trick" for getting itself copied via the product it coded for in the environment it found itself in. That is ALL self-replication means. The environment always acts on a "product". Replicators act on information--they recombine it, copy it, assimilate it, and tweak it. A computer program is information. It can't be selected until or unless it makes a program or parts of a program so that replicators (humans or computers) can be aware of the info. and replicate it.

 

[President Bush]

Once the butterfly is selected because it has a mutation that confers a survival advantage, then the mutation has a chance to get itself copied by being part of a gamete that becomes a zygote that codes for a butterfly that successfully reproduces. It doesn't copy itself--it gets itself copied! Recombinations through sex and random mutations make for a great variety of product that the environment can then select from.

In your example... is the consequence of the mutation its cause?

When the environment achieves a product that can fly through recombination and mutation of information (not unlike the process that achieved the butterfly mutation) then the environment acts upon that product. The environment of humans says, "hey us humans have evolved a design that codes for a machine that can reliably fly"-- and they they replicate the design and begin to evolve and hone that design. Info. that makes an airplane fly is info. that gets replicated in various manners by those who want to fly.

In your example... is the consequence of the design (change) its cause?

 

[Southwind17]

In your example... is the consequence of the mutation its cause?

In your example... is the consequence of the design (change) its cause?

How can the consequence of something also be the cause of it, unless it's a circular process, like an internal combustion engine? Is that what you mean?

 

[articulett]

How can the consequence of something also be the cause of it, unless it's a circular process, like an internal combustion engine? Is that what you mean?

A cause is a change in information (recombination, imperfect replication, design tweak, etc.)

The consequence is what happens when that change is preferentially selected or culled via the action of the thing the information codes for in the environment.

Changes in information that are preferentially selected by the environment drive evolution. Information that has a "trick" for getting itself copied, gets copied preferentially and survives to evolve further...This makes the "things" that the information codes for appear to evolve (animals, technological items, scientific theories, etc.)

Matter doesn't evolve. Information that codes for the way matter is assembled can and does.

(This is in answer to Prez. B since I have in on ignore, and only know what he posts when the smart people quote him )

 

[Southwind17]

A cause is a change in information (recombination, imperfect replication, design tweak, etc.)

The consequence is what happens when that change is preferentially selected or culled via the action of the thing the information codes for in the environment.

Changes in information that are preferentially selected by the environment drive evolution. Information that has a "trick" for getting itself copies, gets copied preferentially and survives to evolve further...This makes the "things" that the information codes for appear to evolve (animals, technological items, scientific theories, etc.)

Matter doesn't evolve. Information that codes for the way matter is assembled can and does.

I see. The slight difference in the relationship between the words 'cause' and 'consequence' from the norm threw me off slightly.

So, I'd say the answer to both of Mr President's questions is probably 'yes'. Would that be correct?:

In your example... is the consequence of the mutation its cause?

In your example... is the consequence of the design (change) its cause?

 

[articulett]

Nah... his question was bad. Your answer is right. He's confusing the information and the thing it codes for like Jim.

While humans can intentionally tweak (or recombine)the first part ("cause") with the hope of generating the preferred "consequence"... they are just doing what they "evolved" to do (whether there is intent or not), and humans in the envrironment are just doing what they evolved to do when they "select the info." that sticks around (whether that is their intent or not.)

Humans imagine themselves acting with much more intent than they actually do, and unintended consequences and serendipity play a role as to what human generated information evolves... Those who don't seem to understand the analogy seem to place anything with humans outside the system and label it "intelligent"-- but it's just part of a continuum of information evolving better information processors over time.

Humans don't "invent" the information in their genomes or environment--they just play a role in its evolution and in selecting other information in other evolving systems (or culling that information).

Your analogy works on a very basic level-- but it's also correct on a much deeper level... and, in fact, is just further along the confinuum

In the tree of life...humans would be at the top--and the information they generate would be sprouts off the tops of that twig (representing humans) that evolve and branch off further (or make leaves or fruit or whatever else you want to imagine at the end of the evolving information that represents the evolution of species over time.)

The algorithm is just continued... information begets better information processors which then evolve that information.

 

[President Bush]

The slight difference in the relationship between the words 'cause' and 'consequence' from the norm threw me off slightly.

Definition of cause.

Definition of consequence.

So, I'd say the answer to both of Mr President's questions is probably 'yes'. Would that be correct?

No.

 

[President Bush]

A cause is a change in information (recombination, imperfect replication, design tweak, etc.)

The consequence is what happens when that change is preferentially selected or culled via the action of the thing the information codes for in the environment.

Changes in information that are preferentially selected by the environment drive evolution. Information that has a "trick" for getting itself copied, gets copied preferentially and survives to evolve further...

Is it your intent to anthropomorphize the idea of causality?

 

[articulett]

Cause sets evolution in action. Remember the analogy? Consequence is based on what is selected from the new info that was "caused".

Lots of things cause changes in info... sex, recombination, insertions, random mutations, purposeful tweaking, selection pressures, external forces... Only some are selected. The information stupid people generate and assimilate and mutate isn't usually selected by others because it's useless. Not every bit of information is a winner. But having tons of it out there means there will always be a pool of information to cull from.

 

[articulett]

The beneficial change in the butterfly was caused by butterfly sex and imperfect copying of genome information... pretty standard way for genome changes to take place in evolution...

The beneficial change in design is due to accidental or purposeful or some other means of changing the design or recombining information... pretty standard way for information to change in the world of humans and the information that utilize.

The consequence of the mutation in the butterfly like many such mutations involved in evolution-- is that it allows the butterflies carrying the mutation to preferentially survive so that the mutations have a chance to be copied into more butterflies and the mutation can thus be part of an evolving information system (which code for each live butterfly that contains the mutation in it's genome and its mutation containing descendants.)

The consequence of a the information that caused (coded for) a plane that flies also preferentially survives and thus that "mutation version" has a chance to be copied preferentially and become part of an evolving information system we call "aviation evolution". http://www.hq.nasa.gov/pao/History/SP-468/contents.htm

It's not anthropomorphic any more than the Selfish Gene is. It's a way of showing how information that codes for it's own replication drives the system-- not intent or foresight. The latter are just things that can tweak the info and cull the info. as it evolves-- humans evolved these "traits" for doing so just as bees evolve traits that help their genomes (information) survive in an enviroment of other bees, flowers, hives, etc. It's a feedback loop that can't help but generate increasing "complexity" and the appearance of design (a human term for certain types of pattern detection regarding the organization of matter, sounds, images, etc.)-- stuff that just seems to fit together so damn well!