Biology Needs to Work Harder Against the Claim of Random Chance!!

[Meadmaker]

I think the key here is to find out what they mean when they say “evolution is random”. Then, after finding out what they mean, you have to ask whether or not they are correct.

As we have seen in many threads on the subject, the phrase “evolution is random” or its sister-phrase, “evolution is not random” has practically no meaning at all. It could mean almost anything, depending on the mood, background, and intent of the speaker.

Having read a fair amount of what creationists write, I think I can safely say that they mean all sorts of different things when they say “evolution is random”, or some variation thereof. The most common variation is, “evolution says that life came about by random chance”.

So what do they mean by that. They mean that none of the events which gave rise to life were part of a plan, or had a purpose. They mean that the theory of evolution states that chance interactions of molecules are responsible for all life we see on Earth.

Some of them are more than happy to acknowledge the role natural selection plays in the theory of evolution. They are willing to do that because appeals to natural selection cannot address the main argument. Their argument is that the structures of life are too complex to have come about by random chance. The failed counterargument is to claim that natural selection isn’t random. It fails because natural selection doesn’t create. Natural selection throws out the losers, and makes room for the winners to multiply. Whatever natural selection selects had to be created as a result of random interactions.

That bears repeating, because it is so often misunderstood.

The creationist argument:

Random interactions cannot create complex entities.

The reply:

Anything not suited will be discarded through natural selection, which is not random.


Do you see why it fails to refute the argument? Their argument is that a complex, organism that is well matched to its environment cannot come into being at all by random events. The attempt to counter is to say that if an entity is not matched to its environment, it will not survive. The refutation ignores the original argument entirely.

So, is there a way to refute the argument? Of course there is, but lets go slow to see what’s wrong with it. First, we’ll have to see what’s right with it.

But, I think that will have to wait. For another time.

Brief summary: In order to refute the argument, you have to understand and address the argument. The argument is wrong, but you have to see why it is wrong in order to craft an argument that can win against it. And of course, it can’t really win in all cases, because the people who make it have a powerful will to believe it, despite all the evidence.

 

[Wowbagger]

No, "Biology" does not need to work harder at this; indeed it doesn't need to work at all on this.

If Biology wants to gain the respect of the people, it has the responsibility to work harder communicating its ideas to people. For too long it has allowed the "random chance" myth to fester.

I am willing to help. I am not a professional biologist, but if I can develop such a strategy, I could see myself becoming something like the "Mark Roberts of the Random Myth"!!

By first convincing tham that it takes no intelligence to "guide" something.

"Well, what about a railroad track, Mr. Wizard? It's already there, and an engineer doesn't steer anything!"

"Nice try, Tommy! But an intelligence laid down those tracks, so intelligence is still a guide."

It'll be a long, uphill battle.

Perhaps a tornado would be better. Its path is guided, ultimately, by physics. Most people I have debated accept that tornados are not driven by God. And, this has lead to some mild success. One guy, on DebateNation, no longer argues that Evolution is random chance, but now refutes the evidence that physics is behind the particular non-randomness of life. Not perfect, but at least it is an improvement.

Then scientists should stop using the term "random" when referring to changes in DNA due to things like cosmic ray strikes.

Absolutely!!

I suggest a "ban" on using the word random, at all, when describing Evolution to the public, because that has probably been the biggest cause of this mess.

You would have to explain random chance in a way that a Creationist would understand and accept. Understanding might be easyier than acceptance, however...

Right. Understanding comes first. We will deal with acceptance later.

Stop right there! To a Creationist, God is all you need to explain anything. As soon as you hit the word "or" the fundie has already cut you off. According to them, there is no "or" - only God, nothing more

Perhaps the first step, for those folks, might be to pry their imaginations open a bit, to the idea that other people disagree with that, and foster alternative explanations.

Fundie: "Of course science likes the 3rd option! Scientists set up the rules, scientists run the tests, scientists interpret the results, and scientists get paid to do it!"

So, let the Fundies say that! If we can get them to acknowledge Evolution is the 3rd option, and not the 2nd, at least it will be progress.
And then, we can eventually hit them with all the progress science has made, since then.

You might convince a Fundie/Creationist that the Theory of Evolution makes sense in principle, but he will always know what the Real Truth is.

And, that would be tremendous progress, at least!!

What?!! "Provisional Truth"? To a fundie, you've just broken the 9th commandment. There is only Absolute Truth or lies - no middle ground allowed. Nope! Not gonna hear it! I can't hear you anymore! LA-LA-LA-LA-LA-LA-LA-LA-LA-LA-LA-LA- (et cetera).

We could argue that, since science is only provisional models, it does NOT replace the Absolute Truth they are used to; but can be used to supplement their understanding of their Universe.

"Hey, don't be frightened of Evolution. It's only a model!"
"Oh yeah?! Then why push it on us, if it might not be true?!"
"We like models. You can accept this one, or not. But, now you know it does not defeat your beliefs."

To a fundie, a model is something you play with when you can't obtain the real thing.

And this is different from science, how?

No one has ever obtained perfect knowledge about the real world. Science only plays with models: very good, high quality models, of course. But, they are only models.

Look, if you want to win over a fundie/creationist, you have to learn how they think ... neo-cons, with GOD instead of GWB at the top. Anything that even hints at contradicting the word of God sets off alarm bells in the fundie's mind. Try reading Scripture (Book of Genesis) to them and deconstructing it in scientific terms. Then maybe - maybe - they'll let you off with only a harsh look and a snort of derision.

Not all creationists are fundies. I agree the fundies are much more difficult. But, my aim is much broader. Virtually all creationists, of every level, seem to suffer this particular distortion.
For example, I do not think Ben Stein is really a fundie, but he seems to suffer from it.

 

[Wowbagger]

What biology needs to understand is that evolution by natural selection works even if both mutation and natural selection are random and that no amount of creationist hemming and hawing or biologists' insistence that the probability theoretic definition of "random" "makes everything random" will ever change that.

What you need to understand is that, in the context of this thread, Random is defined, quite specifically, as "completely unpredictable", which is what most Creationists seem to think Evolution is. They use words like "random chance", "blind chance", "happy accident", or "dice roll", etc. We would be extremely lucky to be here!!!!!!!!

I am sure even you, mijo, agree that Evolution is not really random in that way. It might be random according to other definitions, and contexts, but not the one I am attacking in this thread.

(And, in case you really do think Evolution is a theory about random chance: You would be wrong, and your view of it is distorted. Those who study Evolution aim to remove chance from the picture, as much as possible.)

I think the key here is to find out what they mean when they say “evolution is random”. Then, after finding out what they mean, you have to ask whether or not they are correct.

I have asked. And, virtually all of the Creationists give one of those phrases I mentioned above: "random chance", "happy accident", etc.

Of course, Evolutionists give a different answer: They might say "mutations are random with respect to the life form's survival", or "random is just a quick way of saying 'we need a more powerful computer'". Both of which do not conflict with what the Theory says.

So what do they mean by that. They mean that none of the events which gave rise to life were part of a plan, or had a purpose.

that part would be correct, if that is all they meant, however...

They mean that the theory of evolution states that chance interactions of molecules are responsible for all life we see on Earth.

...this part would conflict with the goals of those studying the theory! "Chance interactions" does not help us describe reality to any precise detail. Biologists might use that phrase, temporarily, until they can learn more details. (Perhaps if they had a much more powerful computer to calculate the interactions, they would not use it at all). But, the ultimate aim is to remove chance from the picture, as much as possible.

Their argument is that the structures of life are too complex to have come about by random chance. The failed counterargument is to claim that natural selection isn’t random. It fails because natural selection doesn’t create.

First of all, it does not matter if it fails to convince them, yet. The first step is to have them acknowledge that any claim of random chance goes against what the theory says, and they should stop using those words, unless they only wish to make Strawman arguments all the time.

Second of all, you can bring on examples to show how natural selection does "create" complexity, over time. They still might not accept it, but at least they will know what they are up against. And, that would be progress.

Natural selection throws out the losers, and makes room for the winners to multiply. Whatever natural selection selects had to be created as a result of random interactions.

If that is what the Creationists claim, then tell them they got the theory wrong. "random interactions" offers nothing of value to science, so scientists seek to avoid relying on those arguments.

The refutation ignores the original argument entirely.

So, we must find ways of telling them "that's because your original argument does not really describe the theory. Physics ultimately dictates selection, not random interactions."

And of course, it can’t really win in all cases, because the people who make it have a powerful will to believe it, despite all the evidence.

I will have to accept whatever progress we can make, one step at a time.

 

[Fredrik]

I think the real problem is to get them to start thinking at all. I have spent enough time debating various issues with people who think the way creationists do to know that this is what's going to happen almost every time we try to debate them:

1: The creationist says that "life obviously didn't happen by random chance".
2: The science nerd explains what's wrong with the creationist argument.
3: The creationist ignores the refutation and moves on to talk about "gaps in the fossil record" or just quotes someone with a fancy title who once said something negative about the theory of evolution.
4: The science nerd explains what's wrong with that argument.
...
n: The creationist points out another imaginary flaw with the theory of evolution.
n+1: The science nerd refutes it.
n+2: The creationist leaves the debate.
n+3: The creationist starts talking to someone else, and says that "life obviously didn't happen by random chance".
...
(...and so it continues until the creationist dies).

I've seen this type of behavior so many times the past 6 years that it has significantly lowered my opinion of our entire species. I don't have a solution to it either. Maybe we could kidnap them one at a time, and keep them locked in a room where they only get food those days when they have debated some issue with us using logical arguments.

 

[Wowbagger]

I think the real problem is to get them to start thinking at all.

Many prominent creationists are capable of thinking. If we could get them to represent the Theory properly, (even if they still do not accept it), it could eventually trickle down the other folks.

Maybe we could kidnap them one at a time, and keep them locked in a room where they only get food those days when they have debated some issue with us using logical arguments.

Well, aside from the kidnapping, denial of food, and other crimes and inhumanities (which I assume you were joking about).....

Discussing these issues with each of them, on a one-on-one basis, has been shown to work, with some people. Though, hitting a broader audience is a different, and much more challenging, beast.

 

[Meadmaker]

What you need to understand is that, in the context of this thread, Random is defined, quite specifically, as "completely unpredictable", which is what most Creationists seem to think Evolution is.

I don't think this is an accurate presentation of creationist views. Of course, "creationist views" aren't uniform anyway. There is no one "creationist view" of evolution, or of creation.

I have asked. And, virtually all of the Creationists give one of those phrases I mentioned above: "random chance", "happy accident", etc.

Certainly that is true. It also happens to be accurate. Name any example of an event which increased the diversity of life on Earth that was not accidental.

"random is just a quick way of saying 'we need a more powerful computer'". Both of which do not conflict with what the Theory says.

I'm afraid they do. See the "Evolution not random" thread for a recap of why it doesn't matter how powerful your computer is.

Second of all, you can bring on examples to show how natural selection does "create" complexity, over time.

I certainly can't, because I don't know any. Indeed, I'm fairly certain there aren't any. Natural selection eliminates and chooses from the available organisms, but it doesn't create them. They are created by random variation.

If that is what the Creationists claim, then tell them they got the theory wrong. "random interactions" offers nothing of value to science, so scientists seek to avoid relying on those arguments.

You had best tell the physicists about this. They are under the impression that they are quite significant.

 

[sol invictus]

Those who study Evolution aim to remove chance from the picture, as much as possible.)

I really have to take issue with the way you're talking about this. The people that study evolution are scientists, and the aim of science is to understand the world.

If chance plays a role (and I think it does - a central and important one) then those who study evolution aim to understand and characterize it, not "remove" it (whatever that even means).

If that is what the Creationists claim, then tell them they got the theory wrong. "random interactions" offers nothing of value to science, so scientists seek to avoid relying on those arguments.

There are certainly very well-understood cases in physics where the results of an experiment are totally unpredictable (what I would term "random"), and will never be predictable no matter how precisely you know the initial state of the system or how powerful of a computer you have. The world is fundamentally random in that sense, like it or not - and those processes are crucial for evolution (because they can cause mutations, for example).

 

[jimbob]

How about games of chance?

Roulette is random, but in the long run the house is going to win. Natural selection is "random" in a similar way.

I read many creationist arguments as this:

"Scientists say that evolution is random, and that makes no sense because how could something haphazard create all the wonderful animals that you see in the world?"

There often seems a (deliberate?) confusion of what "random" means, and a pretense that in this case it is synonomous with "haphazard".

Try it, you can almost always replace "random" with "haphazard" in the creationist arguments about evolution, which is inappropriate in the context of evolution.

ETA:

Most people, if they think about it do understand that "random" can include events of different probabilities. Our pet creationist at work simply refers to evolution as "heresy", . But he does talk a lot about it... Yes I am in the UK, so virtually everyone else thinks this is a demonstration of a very narrow mindset, and most people probably think of him as a great advertisment for atheism...

 

[Complexity]

Biology needs to continue to work hard at understanding life.

Don't task biology or any other field of endeavor for the existance of cretins.

 

[Cardelitre]

I've seen this type of behavior so many times the past 6 years that it has significantly lowered my opinion of our entire species.

I have made the same observation and I agree with your feeling about humanity, sadly.

I don't have a solution to it either. Maybe we could kidnap them one at a time, and keep them locked in a room where they only get food those days when they have debated some issue with us using logical arguments.

I like this idea.
We should just make sure Amnesty International doesn't hear of it...

 

[Wowbagger]

Okay, I will acknowledge that chance might play a role in Evolution. However, if it is an important or central role, it is only so in a provisional manner. The more we learn about the process, and its initial conditions, the more that role will shrink. There is also "smearing" that occurs at larger scales of observation.

In physics, there is the Heisenberg Uncertainty Principal. You probably already know the drill: Can't precisely measure the position and speed of an atom at the same time, etc., etc.
This fundamental uncertainty leads to heavy usage of probability in quantum mechanics. However, those probabilities "smear out", when you examine the world at a large scale, so their influence is minimized, if not completely removed.

I do not think the same kind of chance has been demonstrated in life forms, as the Uncertainty Principal has been at the atomic scale.

Right now, there is a chance that an atom in a pen, on your desk, is "spontaneously" leaping a centimeter to the left. It is only a very small chance, but the chance is still less than zero. And if it does, it would go unnoticed, being only a single atom.

However, we do not expect the entire pen to "spontaneously" leap a centimeter to the left. The chances of all the atoms in the pen doing so are... well... practically nil as far as we are concerned.
Hence the uncertainty of the quantum level is smeared out at the large scale we live in.

There are those who might try to say: "Okay, if you admit chance plays a small role, then that means Evolution is a theory about chance after all!" That claim misses the point I make in the first paragraph of the post.
Scientists might use chance to model biology, but they do so only as a provisional model. Their understanding is that the reality of the process does not rely on chance. They would rather see as much stuff fit into functions of physics and other aspects of the fitness landscape, as possible.

If Evolution was supposed to be about random chance, you would see scientists ripping stuff out of their self-organizing patterns, and their cumulative adaptive measurements, and physics; and they would just stuff everything into the box marked "random chance". Instead, we see the opposite trend.


It also depends on how you define "chance". Physics will inevitably lead life forms down certain paths of Evolution. Which ones are lead, could be up to chance, as far as we understand it, now. That would be like the lottery example: It is inevitable that someone will eventually win the big jackpot. But, who specifically, is up to "chance".

 

[Wowbagger]

I don't think this is an accurate presentation of creationist views. Of course, "creationist views" aren't uniform anyway. There is no one "creationist view" of evolution, or of creation.

The phrase "completely unpredictable" is functionally equivalent to "random chance", "accident", etc. All those words imply fundamental unpredictability.

I agree that there is no one-single "creationist view" of evolution. However there is one inaccuracy of such views that I find too common amongst MOST creationists, of any sort. And, that is the aspect I am battling, here.

Certainly that is true. It also happens to be accurate. Name any example of an event which increased the diversity of life on Earth that was not accidental.

Oh, not you, too!

Name such an event that was accidental.

I'm afraid they do. See the "Evolution not random" thread for a recap of why it doesn't matter how powerful your computer is.

You misunderstand. Using the word "random" as a shortcut for "we need a more powerful computer" does not conflict with the theory. However, that does not imply that there are certain aspects of the process we may never be able to calculate, anyway.

It is like predicting the weather. No one (worth mentioning) thinks anything in the weather is random chance. We understand that it is all fundamentally driven by physics and stuff. We just can not build a powerful-enough computer to track all the variables; and our understanding of the "initial conditions" is limited.

Back to life: The more we understand about the process of Evolution, and its initial conditions, the less of it will be in the box marked "chance". Although, there is a good chance we will never have perfect knowledge to remove the box, completely.

I certainly can't, because I don't know any. Indeed, I'm fairly certain there aren't any. Natural selection eliminates and chooses from the available organisms, but it doesn't create them. They are created by random variation.

Read these, for starters:
http://www.talkorigins.org/indexcc/CB/CB940.html
http://www.talkorigins.org/indexcc/CB/CB200.html
http://www.talkorigins.org/indexcc/CB/CB200_2.html

You had best tell the physicists about this. They are under the impression that they are quite significant.

Name one, outside those talking about the quantum level. I accept the quantum level as driven by chance. But, that chance is minimized at larger scales.

If chance plays a role (and I think it does - a central and important one) then those who study evolution aim to understand and characterize it, not "remove" it (whatever that even means).

I mean removing unpredictability. The more they understand and characterize it, the more predictable things become.

There are certainly very well-understood cases in physics where the results of an experiment are totally unpredictable (what I would term "random"), and will never be predictable no matter how precisely you know the initial state of the system or how powerful of a computer you have. The world is fundamentally random in that sense, like it or not - and those processes are crucial for evolution (because they can cause mutations, for example).

I cover this in the last post. I accept the quantum level as driven by chance. But, that chance is minimized at larger scales.

I read many creationist arguments as this:

"Scientists say that evolution is random, and that makes no sense because how could something haphazard create all the wonderful animals that you see in the world?"

There often seems a (deliberate?) confusion of what "random" means, and a pretense that in this case it is synonomous with "haphazard".

Try it, you can almost always replace "random" with "haphazard" in the creationist arguments about evolution, which is inappropriate in the context of evolution.

Yep. "haphazard" would be another synonym for the creationist distortion of Evolution.

Biology needs to continue to work hard at understanding life.

Don't task biology or any other field of endeavor for the existance of cretins.

Right now, disrespect of biology is causing real damage to the science. Biology is weakened by the poor way in which it is communicated and taught to (most) students and the general public.

Biologists should do more, if they wish to garner more respect from the people. This is important for the sake of the health and welfare of all life forms on Earth.

Biology should continue working hard to understand life. However, I think it would a good idea to work a bit harder on the P.R. The better informed the people, the more of them will help out, in the endeavor to understand life.

 

[tsg]

I think the whole argument is kind of pointless unless you acknowledge the misunderstanding of the role of chance in evolution is a result of their desire to not believe in evolution rather than the cause of it.

 

[Meadmaker]

Okay, I will acknowledge that chance might play a role in Evolution. However, if it is an important or central role, it is only so in a provisional manner. The more we learn about the process, and its initial conditions, the more that role will shrink. There is also "smearing" that occurs at larger scales of observation.

The discussion in the "Evolution not Random" thread was actually quite good on this subject. The section where some of us were discussing with Belz the role of quantum mechanics and chaos theory went pretty well. If you're interested, I suggest reading that. Here's the executive summary:

Quantum events really are truly random, and no amount of observation or knowledge can ever reveal anything which will eliminate the randomness.

Quantum events can have macroscopic consequences, which can be magnified to genuinely global scale consequences.

As applied to evolution, this means that extinction events, mutations, births, deaths and all the other events that occur and influence evolution can be a consequence of quantum events, which means they are also truly random. We cannot predict the end results of evolutionary process, and this inability is not just a consequence of incomplete measurements or inability to analyze data. It truly hasn't been determined yet.

 

[Meadmaker]

The phrase "completely unpredictable" is functionally equivalent to "random chance", "accident", etc. All those words imply fundamental unpredictability.

When I roll a die, I say that the outcome is due to random chance. Despite this, I know that I will not roll a 7. The result is not completely unpredictable.

Oh, not you, too!

Name such an event that was accidental.

I'm afraid so.

All right, I will. Chimpanzees and humans have different numbers of chromosomes. I can't recall the details of who has more, but it is known that at some point in the gene line of the common ancestor of the two, two chromosomes combined. (Or maybe one split. I can't recall the details.) Scientists can say which two chromosomes in one of the genomes are equivalent to the single chromosome in the other genome.

That chromosome changing event was random. It was an accident. It just happened for no particular reason.

(Apply similar reason to every other variaion of life. These things just happen.)

It is like predicting the weather. No one (worth mentioning) thinks anything in the weather is random chance.

I'm afraid this is incorrect. See the other discussion, although we can rehash it here if you are interested.

200 years ago, Laplace famously said that an intellect who could measure and analyze every part of the universe in sufficient detail could know all the past and the future. 150 years later, modern science decided Laplace was mistaken.

 

[Meadmaker]

Read these, for starters:
http://www.talkorigins.org/indexcc/CB/CB940.html
http://www.talkorigins.org/indexcc/CB/CB200.html
http://www.talkorigins.org/indexcc/CB/CB200_2.html

Only the first deals with something that has been the subject of this thread, so I;ll discuss it.

He insists that evolutionst are in unanimous in their agreement that complex structures did not arise by chance. So, if I can find an evolutionists who doesn't agree, the statement would be disproved, would it not? Do you really want me to go there? You know that I can find evolutionists who disagree with the statement. However, rather than citing an expert, let's just look at what is stated in the next paragraph.

Selection of randomly introduced variation is known to be able to produce complex formations, including functional circuits (Davidson 1997; Thompson 1996) and robots (Lipson and Pollack 2000). Creationists have never proposed a reason to explain why the same processes would not produce the same results in nature.

How does "selection" produce anything? It doesn't. As a graduate student, I worked with various genetic algorithms and related subjects, and the process was quite routine. Generate variations. Test them (somehow) for fitness. Throw out the unfit ones. Repeat as necessary. The "test them for fitness" part was analogous to natural selection. What happened in that step? Things were removed. Nothing was created.

If I had had a hugely fast processor and a huge memory store, I could have skipped the deletion process, and I would have evolved things faster. Selection didn't help, it just was part of a recognition that there were finite resources available.

 

[godless dave]

First, de-personify Evolution. It is a process, not a person. It has no goal, which would imply a plan, which in turn would inply intellect.

In other words, Mother Nature doesn't like to be anthropomorphized.

 

[mijopaalmc]

Okay, I will acknowledge that chance might play a role in Evolution. However, if it is an important or central role, it is only so in a provisional manner. The more we learn about the process, and its initial conditions, the more that role will shrink. There is also "smearing" that occurs at larger scales of observation.

In the mathematics that describes random processes, such as most of quantum mechanics, the "smearing out" is called convergence of random variables and arguably occur because the system is described by random variables rather than in spite of it. Convergence of random variables is observed quite frequently in simple probabilistic experiments such as coin tossing or dice rolling in the fact that in repeated trials the proportion of a certain event happening out of all the other events approaches a fixed value.

 

[sol invictus]

It is like predicting the weather. No one (worth mentioning) thinks anything in the weather is random chance. We understand that it is all fundamentally driven by physics and stuff. We just can not build a powerful-enough computer to track all the variables; and our understanding of the "initial conditions" is limited.

That statement is false in two important ways:

1) the weather is ultimately determined by quantum mechanics, which is truly random, and

2) the weather is chaotic, which means tiny variations in the initial conditions (for example, small errors in measurement or small quantum fluctuations) get magnified enormously over time.

The result is that certain aspects of the weather are fundamentally unpredictable, regardless of the size of the computer used. For example, the exact temperature a week from this moment at the tip of the Eiffel tower cannot be predicted, period.

However certain other aspects of the weather are predictable - that it will be warmer on average in July than January in the northern hemisphere, for example.

The same is true for evolution - certain general features are predictable, but the specific details are not.

 

[jimbob]

How does "selection" produce anything? It doesn't. As a graduate student, I worked with various genetic algorithms and related subjects, and the process was quite routine. Generate variations. Test them (somehow) for fitness. Throw out the unfit ones. Repeat as necessary. The "test them for fitness" part was analogous to natural selection. What happened in that step? Things were removed. Nothing was created.

If I had had a hugely fast processor and a huge memory store, I could have skipped the deletion process, and I would have evolved things faster. Selection didn't help, it just was part of a recognition that there were finite resources available.

Of course all the crud is removed, as well as a lot of "good stuff". And finite is what is real. I know you understand this, but feel it could do with reiterating.

Without random variation, you would either run out of variation, or you would need some nonrandom source of variation. Without random variation, you would be limited, and probably would require an intelligence able to understand the most complex organisms tio develop them. With random variation you need no intelligence.

An analogy from electronics, (actually quoted earlier, in passing by Meadmaker).

When designing electronic circuits, the classical approach is to analyse every subsystem, and predict how the subsystems work, then analyse how these subsystems interact. This way, surprisingly complex systems can be comprehended and analysed and designed.

However evolutionary algorithms have also been used with FPGAs ^WP, and one interesting feature is that the "evolved" circuits are often not amenable to analysis by our current techniques, because they rely on complex interactions with "parasitic components" and non-ideal behaviours. The evolved circuits rely on characteristics that human design can't analyse easily, so can't design with, and have to work around.

A simple "dumb" process producing a solution that we currently couldn't have "intelligently designed". And all due to (pseudo)random variation and a selection algorithm.​