Intelligent Evolution?

I love it, Apathia-- is it your own?

Father William and his ideas are, indeed, old.

The way it's written up is mine. The theme is from Southwind and others in this thread.

"The Headless Watchmaker" is my turn of phrase.
"You are old, Father William" is a literary allusion to this poem by Lewis Carroll:
http://www.web-books.com/classics/poetry/anthology/Carroll/YouAre.htm

Willaim Paley, of course. is the author of the Watchmaker arugument for God and creation.
 
You mean how long would human 'randomness' take? How long are you prepared to wait - as long as it took biological evolution to perfect VTOL? You see, yet again, it all comes down to what you mean by 'intelligence'.

Why have you ignored the "working definition" of "intelligence" that I have provided for you several times?

The working definition that I have used is intelligent is the ability to perceive causal relationships and understand them well enough to obtain a specific outcome by inputting specific initial conditions.
 
Last edited:
The way it's written up is mine. The theme is from Southwind and others in this thread.

"The Headless Watchmaker" is my turn of phrase.
"You are old, Father William" is a literary allusion to this poem by Lewis Carroll:
http://www.web-books.com/classics/poetry/anthology/Carroll/YouAre.htm

Willaim Paley, of course. is the author of the Watchmaker arugument for God and creation.

I recognized the Father William poem... and the reference to William Paley and the summary of the argument. I think it was a clever parody...

I had to memorize that damn poem in middle school... it has the word "suet" in it. I remember having to look up that word.
 
It's not that I'm not up with the debate (because I am); it's that I haven't found a single argument about why we should completely disregard the involvement of an intelligent agent or the existence of intelligence itself put forward thus far convincing. Basically, the possession of intelligence (which is a completely natural and evolved trait) changes the way the agent can effect the causality of the system, because the agent can recognize mistakes and and try to prevent them in the future.


The best reason I see for disregarding a "designer" is that, based on the best evidence, a designer isn't needed to explain the way things are. Why introduce another element? If things are explicable without I.D. why propose it? Asking science to disprove I.D. is like asking an atheist to disprove the existence of God. The burden of proof should be on the one who has faith but no evidence.

Specific to your post, the existence of intelligence in animals or humans says absolutely nothing about the likelyhood of an intelligent designer. Non sequitor.
 
The best reason I see for disregarding a "designer" is that, based on the best evidence, a designer isn't needed to explain the way things are. Why introduce another element? If things are explicable without I.D. why propose it? Asking science to disprove I.D. is like asking an atheist to disprove the existence of God. The burden of proof should be on the one who has faith but no evidence.

Specific to your post, the existence of intelligence in animals or humans says absolutely nothing about the likelyhood of an intelligent designer. Non sequitor.

Whit all due respect, you completely missed the point of my post. I was saying that the presence of intelligence is precisely what distinguishes technological development from biological. I understand that there is absolutely no intelligence at work in evolution, but the contention of those who support the analogy is that there is no intelligence active in technological development or that intelligence itself does not exist. That is what I am responding to in my post, not some notion that the existence of an intelligent designer explains the genesis and diversity of life on Earth.
 
Last edited:
Why have you ignored the "working definition" of "intelligence" that I have provided for you several times?

Because this intelligence can so easily be removed from the technological development process and substituted with randomness with absolutely no impact on the outcome, other than timescale. That's why.
 
Originally Posted by jimbob
Isn't the actual reproductive "strategy" and process something that evolves?
Do you think so? In what sense do you see the 'strategy' and process having 'evolved'? Do you consider this relevant to the validity of the analogy? If so, how?
It isn't covered in the analogy, but how an organism reproduces is one of the main traits that evolves.

Organisms are selected on their success in reproduction.

Originally Posted by jimbob
In your analogy how is the number oc fopies determined?
Do you think that matters? So long as at least one copy emerges evolution will occur.
Yes it does, with your analogy, you wil only get a single member of a more successfull variant. If an organism's descendents are successfull, there will be many, all slightly different and all subject to evolution. The "optimisation process" works far better in such a situation.
Originally Posted by jimbob
If you sell one, do you make one replacement, or do you make many?
Do you think that matters? The number of copies simply determines the rate of evolution, not whether evolution occurs per se.
Yes it does, it helps differentiate between marginal success and vast success, see above.
Originally Posted by jimbob
How do you cull off the "failed" designs. Or do you just leave them wasting showroom resource?
They're culled off by virtue of the fact they haven't sold. It doesn't matter how long you give them to sell. Once you've decided they've been 'culled off' you disregard them. It's the ones that survive that matter. Evolution doesn't arise through failures, you should know that by now! Where the 'culled' ones remain or end up is completely irrelevant.
Once you've decided they've been 'culled off' you disregard them.

That is an intelligent intervention.

Without it, your system wouldn't work, as they would use up showroom resource.
Originally Posted by jimbob
Maybe they aren't failed, just "slow breeders", maybe they might sell less frequently, but at a higher price?
Doesn't matter. How fast they sell simply determines the rate of evolution; not whether evolution happens per se.
Rabbits and elephants had a common ancestor, but they breed at different rates. The breeding rate is an evolutionary trait. Elephants couldn't succeed in their niche breeding several times a year, whilst rabbits couldn't succeed if they waited to breed untill after they had been killed. Because they are subject to heavy predation and mortality, rabbits need to breed young. Because elephants aren't, they can "invest" more resources in eachoffspring.

<derail>There is a similar argument about aging: There is no benefit in an organism evolving to have a longer lifespan than that allowed for by (random) predation etc. Indeed there is an evolutionary cost involved, as the organism's resources would be devoted to "keeping youthful" instead of "investing" these resources in the success of their offspring.

</derail>
In your analogy, suppose one variant was akin to a Ford Fiesta, and another akin to an Aston-Martin.

Both are viable "solutions", but they would have different "lifetimes". This is something that evolves in biology.

Originally Posted by jimbob
There is room in your syystem for a self-replicating version to arise: Suppose a variant subverted the pressure sensor to say that it had been sold, when it hadn't
.
You are joking?!

This line of questioning by you indicates to me either that you've turned into a troll, or that you not only misunderstand but have absolutely no perception of the purpose of the analogy. These question, and the answers thereto, are all completely ancilliary to the main issue, and have absolutely no relevance to it.
I was under the impression that you were arguing that your analogy was a good analogy for evolution and that self-replication was an unimportant factor in evolution. If not I apologise.

Evolution needs variation and selection.

The variation comes from the "imperfect copying" process; selection acts on the "variant" within the environment. For the copying instructions to be copied, (i.e. selected), they need to be associated with the variant and in the environment.

What do you think the main issue is?
 
It isn't covered in the analogy, but how an organism reproduces is one of the main traits that evolves.

Organisms are selected on their success in reproduction.

Has it crossed your mind that the reason it's not covered in the analogy is because it's irrelevant to the analogy?! I'm not sure you've ever really understood the purpose the analogy sets out to serve, but allow me to summarise it as it now stands:

The purpose of the analogy is to show that seemingly irreducible complexity would arise in technological development even if intelligence, i.e. intent and forethought, were removed from the process and replaced with random, incremental changes.

As such, the 'process' by which 'reproduction' occurs in technological development is immaterial, as is any comparison with biological evolution. I suggest you go right back to the AA or the Sam & Ollie story if you've forgotten that this is so.

Yes it does, with your analogy, you wil only get a single member of a more successfull variant. If an organism's descendents are successfull, there will be many, all slightly different and all subject to evolution. The "optimisation process" works far better in such a situation.

After the 'device' is released into the environment there are only two possible outcomes. Either the device will sell and it will then be reproduced with a random variation, or it will not sell and it becomes extinct. In the first case the new random variant will either be detrimental, indifferent or beneficial. Detrimental variants will fail to sell, and the AA will then go back a step to the version immediately prior and 'try again' with a different variation. This is no different from biological evolution. All of the replications that don't carry a particular mutation, i.e. the vast majority, can be considered to be a 'step back', some of which will carry a new variation. Indifferent variants can be ignored, by definition, and beneficial variants will probably sell. I think you're of the mistaken belief that only beneficial variants get fed back into the reproductive system whereas all variants send the requisite message back. This is pretty simple and logical to me, but I can see how you might, and probably will, have trouble getting your head around this comparison. In the second case the fact that the device failed to sell is a strong indication that it carried a detrimental variant, but not necessarily so. It could be that a competing device just happened to have acquired a more beneficial variation.

Yes it does, it helps differentiate between marginal success and vast success, see above.

The difference between 'marginal success' and 'vast success' as you put it, is simply time. If you prefer, visualize an army of AAs or Sam & Ollies instead of just one, just like in biological evolution.

That is an intelligent intervention.

Without it, your system wouldn't work, as they would use up showroom resource.

jimbob, do you not realize that the 'showroom' is simply a metaphor for the environment. If it helps, consider un-sold cars simply being despatched to the scrap yard, or left in a field to rust away. Again, this comment of yours seems to show a complete inability to grasp the thrust of the analogy, or analogies per se.

Rabbits and elephants had a common ancestor, but they breed at different rates. The breeding rate is an evolutionary trait. Elephants couldn't succeed in their niche breeding several times a year, whilst rabbits couldn't succeed if they waited to breed untill after they had been killed. Because they are subject to heavy predation and mortality, rabbits need to breed young. Because elephants aren't, they can "invest" more resources in eachoffspring.

I really don't see how you can believe that this concept of differential breeding rates has any relevance to whether the removal of intent and forethought from the technological development process precludes the possibility or likelihood of seemingly irreducible complexity. It seems to me that you simply have trouble constraining your thought process within applicable parameters or a particular context. You seem to allow your thought process to wander down blind alleys but still pay attention to what crops up enroute!

<derail>There is a similar argument about aging: There is no benefit in an organism evolving to have a longer lifespan than that allowed for by (random) predation etc. Indeed there is an evolutionary cost involved, as the organism's resources would be devoted to "keeping youthful" instead of "investing" these resources in the success of their offspring.</derail>

Blind alley!

In your analogy, suppose one variant was akin to a Ford Fiesta, and another akin to an Aston-Martin.

Both are viable "solutions", but they would have different "lifetimes". This is something that evolves in biology.

The fact that we've actually arrived at the position where we're producing either Ford Fiestas or Aston Martins is testament to the validity of the analogy, and I thank you for acknowleging that fact! Regardless, remember that, for the purpose of the analogy, 'lifetime' is arbitrary. The AA can wait as long as it likes; forever, if necessary, or it can reproduce after a pre-set timespan. Just remember that if it reproduces before the last model has sold then that's taken as a failure (even if that model would sell given more time), and the AA would go back to the preceding model as the basis for the next. As I wrote above, that's no different from the multitude of natural births that occur evey minute of every day in nature; each one can be considered to be a product of the preceding 'proven' version.

I was under the impression that you were arguing that your analogy was a good analogy for evolution and that self-replication was an unimportant factor in evolution. If not I apologise.

That's right; I am.

Evolution needs variation and selection.

And my analogy has both.

The variation comes from the "imperfect copying" process; selection acts on the "variant" within the environment. For the copying instructions to be copied, (i.e. selected), they need to be associated with the variant and in the environment.

That's right. You're describing my analogy to a T!

What do you think the main issue is?

It seems that the main issue is your apparent inability to constrain your thought process to the points you yourself have essentially summarised in your description immediately above. You summarise the key points nicely, but then when you enter into an analysis and debate you introduce many other irrelevancies.
 
Last edited:
Does your analogy work without your decision that thay haven't sold?

They're culled off by virtue of the fact they haven't sold. It doesn't matter how long you give them to sell. Once you've decided they've been 'culled off' you disregard them. It's the ones that survive that matter. Evolution doesn't arise through failures, you should know that by now! Where the 'culled' ones remain or end up is completely irrelevant.

Your analogy imposes an arbitary, fixed lifetime, which is not what happens in biological evolution.

Your analogy is moving further away from what acually happens in technological development, yet it still retains a dependence on intelligence. Biological evolution doesn't.

O
riginally Posted by jimbob
The variation comes from the "imperfect copying" process; selection acts on the "variant" within the environment. For the copying instructions to be copied, (i.e. selected), they need to be associated with the variant and in the environment.
That's right. You're describing my analogy to a T!

How are the coying instructions associated with the variant in the environment? They are associated with the copier, and have been fiexd: "If it fails to seel within time x, replace, and if it does sell then make a certain number of copies." The copying instructions are not associated with your electronic variants, and are not subjected to selection, as the variants do not self-replicate.

ETA:

Evolution requires Natural Selection which is what this analogy lacks
 
Last edited:
As your analogy doesn't need any intelligent input, how it would work without any descision as to when a failed variant should be "culled"?
 
Because this intelligence can so easily be removed from the technological development process and substituted with randomness with absolutely no impact on the outcome, other than timescale. That's why.

Ideally. To what extent does it actually figure into the behavior of human tech progress? Why do people think it's so prevelant in the process?

Perhaps this thinking of causes and effects is a kind of worry that saps energy from healthy development. Could we put into place structures and policies that would encourage the natural evolution of tech products without this unnecessary attempt at cognition?

Planning, intent, cognition, consciousness, these are all abstract illusions that never figured into the process of human technological development anyway. :spongebob
 
Steven Jones nozzle example was provided as exactly what you mention apathia. Also evolutionary algorithms and biomimicry are examples.

Aging is similar to planned obsolescence, Jimbob... Computers age out too... we get newer cheaper, and better ones instead of endlessly fixing the old. The information doesn't need to care about the products it builds and how long they last... only that they keep moving "forward"-- that their designs become more efficient and easier to reproduce, recombine, and tweak, by their replicators (humans). Genes are doing the same in organisms over time. The things they "build" interact with the environment to determine which information sticks around to be built upon. It's the same.

Those hung up on the "hows" and "intent" and "intelligence" are missing the forest for the trees. Those are the human words with human centered meanings... the overall process is the same... information that is successful at getting itself copied via the environment it finds itself in, drives evolution. It doesn't matter if the information codes for widget designs or bird species. The information amasses, gets copied, gets tweaked, gets added to, specializes, recombines, and then is tested via vectors in the environment which then selects which information goes on.
 
Those hung up on the "hows" and "intent" and "intelligence" are missing the forest for the trees. Those are the human words with human centered meanings... the overall process is the same... information that is successful at getting itself copied via the environment it finds itself in, drives evolution. It doesn't matter if the information codes for widget designs or bird species. The information amasses, gets copied, gets tweaked, gets added to, specializes, recombines, and then is tested via vectors in the environment which then selects which information goes on.

Thanks Articulett. That frames it very well. There's no need then to get bogged down in questions of inteligent behavior or intervention.
The Copernican Revolutiion continues. Once again subjective and antropocentric concepts such as intelligence, intent, and consciousness are bleached clean from the cooking surface.
 
Show me an example of VTOL that wasn't designed by human intelligence. :)
You mean how long would human 'randomness' take? How long are you prepared to wait - as long as it took biological evolution to perfect VTOL?


No, I meant show me an example of VTOL that wasn't designed by human intelligence.

To claim that it will happen - that it is true biological evolution will produce an example of VTOL only because it hasn't been proved false biological evolution will produce an example of VTOL - commits an argument to ignorance.

Besides, don't you know that to say biological evolution perfected VTOL is the same as what Mickey Mouse told the divorce court Minnie was doing?


You see, yet again, it all comes down to what you mean by 'intelligence'.


I think intelligence can be defined as the ability to reason, plan, think abstractly, solve problems, comprehend complex ideas, to learn from experience, and to do this all quickly.

I don't think intelligence can be defined as change in the inherited traits of a population from one generation to the next.
 
Does your analogy work without your decision that thay haven't sold?

Of course it does. All that's needed for replication is a trigger; a trigger, that is, that's pulled only when the entity in question has proven its survival in its environment, thereby validating its features and characteristics. As I've written many times now, increasing complexity only arises from replication, i.e. through variants that are beneficial, and hence survive, to be built upon and added to. Biological evolution has an essentially defined timescale between each cycle of reproduction for any one species, so why not the analogy? Your problem, I think, is determining what that cycle should be in the absence of intelligence to 'decide'. Well, if the entity survives, i.e. it's sold, well that's self-evident. The signal, and sale proceeds, are sent back to the factory with the instruction to produce another one. The question is: how long does one wait for the signal? Well it doesn't really matter. Imagine the automaton, instead of producing one entity each time, produces 1,000, or 10,000, or more, just like any particular natural species. It can then afford to wait indefinitely, if necessary, because the vast majority which are superior compared to the competition will reveal themselves very quicky, and the automaton can proceed to replicate those only. If one particular variant takes considerably longer to reveal its success, then for all intents and purposes it can be considered to have failed, as it will probably get usurped by those that reveal themselves immediately anyhow.

Your analogy imposes an arbitary, fixed lifetime, which is not what happens in biological evolution.

No it doesn't. As explained above, the automaton can wait indefinitely for the success signal to arrive, but that does, of course, require multiple copies to be made such that at least some will sell. As I wrote before, though, that's no different from the natural world.

Your analogy is moving further away from what acually happens in technological development, yet it still retains a dependence on intelligence. Biological evolution doesn't.

No it isn't. The analogy can be envisaged for any technological development situation you can think of (except the exceptions to the 'rule', which are out of place in this discussion). All you need do is remove the intelligence and replace with cycles of random variations, then sit back and wait, a long time, for something positive to emerge, then wait again, etc.!

How are the coying instructions associated with the variant in the environment?

Are you able to clarify what you mean here by 'variant in the environment'?

They are associated with the copier, and have been fiexd: "If it fails to seel within time x, replace, and if it does sell then make a certain number of copies."

All copying instructions are 'associated with the copier'; it's within the 'copier' that the copying instructions reside. What's important is the 'trigger' for implementing those instructions. The copying 'trigger' in the analogy is no different from natural triggers, the message being: As soon as the device/organism has proven itself to have survived its environment, then make another one. If, after a period of time, a particular variant hasn't proven itself, consider that a failure and start over with the previous version. How long would give a cheetah not to breed before you would consider it to have 'failed'?

The copying instructions are not associated with your electronic variants, and are not subjected to selection, as the variants do not self-replicate.

I wondered how long it would be before the erroneous 'self-replicate' chestnut would try to sneak back in by the backdoor!

ETA:

Evolution requires Natural Selection which is what this analogy lacks

Natural selection is simply the process by which an organism's ability to survive its environment, including out-performing competing organisms, as measured by its success in reproducing, is judged. It's no different from winning or losing a beauty pageant, or any other competitive sport, for example, except that the 'prize' is slightly different! 'Winning' in the analogy is signified by being bought by somebody who had a free choice between all competing devices. I have clearly shown previously how the analogy contains all of the ingedients of natural selection. Which ingredient(s) do you think are still missing?
 
So Southwind, what is Dawkins saying if he isn't saying that he thinks imperfect self-replication is nescessary and sufficient for evolution?

Or as Richard Dawkins said in an interview for Mark Ridley's "Evolution" textbook:

"Natural selection will operate wherever there is heredity, variation and
competition. I suspect that the conditions for natural selection to work are
very minimal indeed: namely, the existence of the phenomenon of heredity."


http://www.blackwellpublishing.com/ridley/video_gallery/RD_What_are_the_conditions_for.asp

What else does he mean by heredity?


Southwind, is your "exception to the rule" argument meaniong that most engineering actually is analogous to evolution?

Almost every case of a diosruptive technology has been developed because it showed potential to superceed a previous type of technology, but in its prototype stage it was generally poorer than the "mature" technology. It would have been "outcompeted" at this stage, except for intelligent intervention.

Jet aircraft have been mentioned, the early prototypes were inferior to piston-engined aircraft, but they still had potential.
 
Last edited:
Superficially attractive, but not really a good approximation at all.

What "evolution" changes a cathode ray tube TV into an LCD or a plasma? It is not a matter of gradual change of the CRT to effect this but a major saltation involving new technology not included in the previous design.
Simple: recombinant technology.

My word, 47 pages. Well, in case my post appears bizarre, it is from page one. I'll try to catch up.
 
Last edited:
...

Southwind, is your "exception to the rule" argument meaniong that most engineering actually is analogous to evolution?

Almost every case of a diosruptive technology has been developed because it showed potential to superceed a previous type of technology, but in its prototype stage it was generally poorer than the "mature" technology. It would have been "outcompeted" at this stage, except for intelligent intervention.

Jet aircraft have been mentioned, the early prototypes were inferior to piston-engined aircraft, but they still had potential.
I didn't read Southwind's posts yet, but in keeping with the analogy theme, in evolution of higher organisms there is a range of survivable organisms and while the 'fittest', to use the term loosely, might emerge as the most prolific, it really depends on a few other factors. A change in the environment may favor a less fit organism for some reason, allowing it to emerge and evolve, sickle cell trait and malaria for instance. Keep in mind, the sickle cell trait gene is not done evolving as every other gene is not done evolving until it dies out.

Also, every prototype is not disadvantageous, more expensive, maybe but even that, maybe not. Nor does every prototype go on to become an improvement. I'm sure the patent office can attest to that. Variation, OTOH is advantageous. So the more prototypes the more likely one will emerge as an evolved technology.
 
Last edited:

Back
Top Bottom